Industrial Furnaces and Methods of Contr

This is a reproduction of a library book that was digitized
by Google as part of an ongoing effort to preserve the

information in books and make it universally accessible.
https://books.google.com
APR 30 1907
Mining dept.
LIBRARY
OF THE
UNIVERSITY OF CALIFORNIA.
Class
着
INDUSTRIAL FURNACES
AND
METHODS OF CONTROL
BY
EMILIO DAMOUR
Ingenieur Civil des Mines
Authorized Translation with Additions

by
A. L. J. QUENEAU , B.S. ( Paris) , E.M. , A.M. (Columbia)

Consulting Engineer, The New Jersey Zinc Company
FIRST EDITION
LIBHARY
OF THE
UNIVERSITY
OF
CALIFORNIA
New York and London

THE ENGINEERING AND MINING JOURNAL
1906
TNI
T2
AtINING UUR
Mee
COPYRIGHT 1906
BY
THE ENGINEERING AND MINING JOURNAL
Also Entered at
STATIONERS ' HALL COURT, LONDON , ENGLAND
AU Rights Reserved
LIST OF ILLUSTRATIONS.
Fig. 1. Graphic Method for the Determination of the Temper

ature of Combustion .
Fig. 2 . Temperature of Combustion of Hydrogen in Air.
Fig. 3. Temperature of Combustion of Carbon Monoxide in Air .
Fig. 4. Temperature of Combustion of Water Gas ( CO + H2) in
Air.
Fig. 5. Temperature of Combustion of Carbon in Air .
Fig. 6. Temperature of Combustion of Siemens Gas (CO + 2N ).
Fig. 7. Temperature of Combustion of Pittsburg Natural Gas.
(a) Combustion with no Excess of Air.
( b) Combustion with 25 % Excess of Air.
Fig. 8. Temperature of Combustion of Iola Natural Gas, with no
Excess of Air.
Fig. 9. Temperature of Combustion of Iola Natural Gas, with
25% Excess of Air.
Fig. 10. Temperature of Combustion of Iola Natural Gas, with
5% of Gas Unburned .
Fig. 11. Curve of the Dissociation of Carbon Dioxide, CO ,.
Fig. 12. Curve of the Dissociation of Carbon Monoxide, Co.
Fig. 13. Siemens Gas Producer with Siphon .
Fig. 14. Taylor Gas Producer with Mechanical Feed .
Fig. 15. Morgan Gas Producer with Mechanical Feed .
Fig . 16. Ebelmen Gas Producer.
Fig. 17. New Siemens Furnace with Horseshoe Hearth .
Fig. 18. Water - Sealed Taylor Gas- Producer with Bild Automatic
Feed .
Fig. 19 . Dellwick - Fleischer Water Gas - Producer .
Fig. 20. Swindell Gas-Producer with Double Face .
Fig. 21. Duff Gas - Producer.
Fig. 22 . Boëtius Furnace .
Fig. 23. Curves Showing the Percentages of the Heat Utilized
and Lost in Direct-Fired Furnaces.
fi
IV LIST OF ILLUSTRATIONS .
Fig. 24. Puddling Furnace of the Forges d'Eurville.

Fig. 25. Section through a Belleville Boiler and Economizer.
Fig. 26. Longitudinal Section and Front Elevation of a Belleville
Boiler and Economizer.
Fig. 27. Section of the Niclausse Multitubular Boiler.
Fig. 28. Front Elevation of a Niclausse Multitubular Boiler.
Fig. 29. Curves Showing the Percentages of the Heat Lost and
Utilized in Multitubular Boilers .
Fig. .
30. Curve Showing the Theoretical Temperatures of Forma
tion of Siemens Gas, the Primary Air Being Injected
at 0°, 600 °, 800 °, 1000º, 1500 ° C.
Fig. 31. Illuminating Gas Muffle Furnace of the Compagnie Par
isienne du Gas .
Fig. 32. Illuminating Gas Muffle Furnace of the Compagnie Par
isienne du Gaz .
Fig. 33. Glass Tank Furnace Fired with Siemens Gas, with Re
cuperators of the Lencauchez Parallel Counter -Current
Type.
Fig. 34. Graphic Determination of the Temperature of Combus
tion of Siemens Gas ( CO + 2N ), Heated to 1290 ° C. ,
the Secondary Air Being Pre-heated to 1000 ° C.
Fig. 35. Lencauchez Brass Furnace, Fired with Mixed Gas, with
Simple Recuperation of the Parallel Counter -Current
System.
Fig. 36. Morgan Gravity -Discharge Billet -Heating Furnace .
Fig. 37. Morgan Suspended -Roof Billet -Heating Furnace.
Fig. 38. Siemens Glass Tank Furnace .
Fig. 39. 50 -Ton Campbell Basic Furnace , Steelton , Pa.
Fig. 40. Kloenne Retort Furnace .
Fig. 41. Queneau System of Triple Recuperation of the Primary
Air, Secondary Air and Gas.
Fig. 42. Curves Showing, at Different Temperatures , the Theo
retical Economies Accruing from the Pre -heating of
the Primary Air, Secondary Air and Gas.
Fig. 43. E. M. F. Curves of Different Thermo-Couples.
Fig. 44. Pyrometer Cane for the Protection of the Wires of
Le Chatelier Thermo -Electric Pyrometer.
LIST OF ILLUSTRATIONS . V
Fig. 45. Apparatus for the Protection of the Thermo- Couple in

Roasting Kilns.
Fig. 46. Apparatus for the Determination of Boiling Points.
Fig. 47. Apparatus for the Determination of the Boiling Point
of Zinc.
Fig. 48. Fletcher Furnace.
Fig. 49. Simple Apparatus for the Determination of the Melting
and Freezing Points of Metals.
Fig. 50. Queneau Electric Resistance Crucible Furnace.
Fig. 51. Electric Resistance Tube Furnace.
Fig. 52. Siemens-Halske Direct-Reading Galvanometer for Le
Chatelier Thermo -Electric Pyrometer.
Fig. 53. Pellin Registering Galvanometer for Le Chatelier
Thermo- Electric Pyrometer.
Fig. 54. Siemens -Halske Recording Galvanometer for Le Cha
telier Thermo-Electric Pyrometer.
Fig. 55. Optical Pyrometer of Le Chatelier.
Fig. 56. Féry Heat-Radiation Pyrometer.
Fig . 57. Féry Heat-Radiation Pyrometer. Lens Type.
Fig. 58. Féry Heat-Radiation Pyrometer, Mirror Type.
Fig. 59. Focusing Device of the Féry Heat-Radiation Pyrometer.
Mirror Type.
Fig. 60. Cambridge Scientific Instrument Company Direct -Read
ing Galvanometer, for the Féry Heat-Radiation Py
rometer.
Fig. 61. Apparatus for the Determination of the Water Vapor

Contained in Furnace Gases.
Fig. 62. Elliott Gas Apparatus.
Fig. 63. Showing the Various Positions of 3-way Cock F.
Fig. 64. Showing the Various Positions of 3 -way Cock G.
Fig. 65. Showing the Various Positions of 3 -way Cock H.
Fig. 66. Elliott Gas Apparatus with Dreshschmidt Platinum Tube
for the Determination of Methane and Hydrogen.
Fig. 67. Section of the “ Ados ” Automatic Gas Analyzer.
Fig. 68. Card of the Ados Automatic Gas Analyzer.
Fig. 69. Cross - Section Through the Mahler Calorimeter (Atwater
Modification ).
VI LIST OF ILLUSTRATIONS .
Fig. 70. Berthelot-Mahler Apparatus Ready for Operation, in the

Metallurgical Laboratory of Columbia University.
Fig. 71. Various Parts of the Berthelot-Mahler Calorimeter..
Fig. 72. Parr Calorimeter.
Fig. 73. Junker Gas Calorimeter.
Fig. 74. Train of Apparatus for the Elementary Analysis of Fuels.
Fig. 75. Incorrect Flue Connection for a Horizontal Regenera
tive Chamber.
Fig. 76. Incorrect Flue Connection for a Vertical Regenerative
Furnace.
Fig. 77. Correct Flue Connection for a Horizontal Regenerative
Chamber.
Fig. 78. Section through a Perforated Block for Recuperative
Chamber .
Fig. 79. Section through the Equivalent Cylindrical Tube.
Fig. 80. Perforated Firebrick Blocks for the Parallel Counter
Current System of Heat Recuperation .
Fig. 81. Wailes Water-Sealed Valve .
Fig. 82. Reversing Valves.
Fig. 83. Forter Water- Sealed Valve.
Fig. 84. Regenerative Chambers of a By - Product Coke Oven .
Fig. 85. Various Types of Ports.
Fig. 86. Wellman Tilting Furnace.
Fig. 87. Diagram giving the Thermal Capacities of Gases per Molec
ular Volumes for any Temperature between 0 ° and
3000 ° C.
Fig. 88. Diagram giving the Thermal Capacities of Gases per Kilo
gram for any Temperature between 0° and 3000 ° C.
LIST OF TABLES.
1. Table of Thermo -Chemical Reactions .

2. Table of the Volumes and Weights of the Air Required for
the Combustion of One Kilogram of Chemically Pure Fuels
and of the Volumes and Weights of the Resulting Products
of Combustion .
3. Thermal Capacities of Gases, in Absolute Temperatures, per
Molecular Volumes.
4. Thermal Capacities of Gases, in Degrees Centigrade. per Molec
ular Volumes.
5. Thermal Capacities of Gases per Kilogram .
6. Thermal Capacities of Gases, per Molecular Volumes, for Each
200 ° C.
7. Thermal Capacities of Gases for Each 200° C. per Kilogram .
8. Dissociation of Carbon Dioxide .
9. Dissociation of Carbon Monoxide .
10. Analyses of Various Industrial Gases.
11. Comparative Table of All Systems of Furnaces at Ruling Tem
peratures of 10000 and 1500 °.
12. Table of Goutal Factors.
13. Composition in Weight of the Atmospheric Air.
14. Composition in Volumes of the Atmospheric Air.
15. Weights and Specific Gravities of Gases Found in Furnaces.
16. Specific Heats of Gases per Molecular Volumes .
17. Specific Heats of Gases per Kilogram .
18. Mean Specific Heats Between t . and ti per Kilogram .
19. Values of Factors at and ( 1 + at) for Temperatures Between
0° and 1800° C.
20. International Atomic Weights for 1905, of the More Usual
Chemical Elements.
21. Conversion Tables .
22. Elementary Analyses and Calorific Powers of Fuels .
23. Prechtl Table of Stack Measurements for Regenerative Furnaces.
24. Coefficients of Heat Conductivity in Calories.
CONTENTS.
PART I.
CHAPTER I.
PAGE
Heat of combustion and calorific power–Tables 1 , 2

' -Thermal capacities and specific heats of gases
Tables 3 to 7 — Temperatures of combustion of various
substances under given conditions - Dissociation
Tables 8, 9..... 7-36
CHAPTER II .
Origin and rapid development of gas furnaces - Descrip

tion and functions of the essential organs of gas
furnaces — Modifications and improvements in gas
furnaces --Table 10 37—52
CHAPTER III .
Influence of the temperature on the heat utilization in a

furnace - Definitions : Heat utilized , heat lost , effi
ciency, equation of the utilization of heat - Funda
mental principles of the utilization of heat in furnaces
- Hypotheses required for the mathematical theory
of combustion in the recuperative gas -fired furnaces ;
selection of the method used for the calculation of
the efficiency . 53—70
CHAPTER IV .
Characteristics of direct - fired furnaces and their principal
industrial applications - Heat lost for all temperatures
-Furnaces followed by steam boilers - Heat lost in
them when combustion is effected by the use of a
large excess of air.. 71-85
CONTENTS . IX
CHAPTER V.
PAGE
First group of gas - fired furnaces — Siemens gas-recuper

ative furnaces, fired with Siemens gas - — Details
of cases of recuperation numbered one , two, three,
four . 86-96
CHAPTER VI.
Second group of gas-fired furnaces — Reactions of com

bustion - Details of cases of recuperation numbered
one , two , three, four 97—115
CHAPTER VII .
Furnaces making use of the so -called regeneration of the

products of combustion - New Siemens furnace -Use
of pure carbon dioxide in gas-producers- Details of
case-- Injection of hot products of combustion in the
gas-producer 116-121
CHAPTER VIII .
Classification of the various types of furnaces — Table 11. . 122-128
CHAFTER IX .
Experimental study of the heat utilization in furnaces

Methods and data-Ultimate analysis and calorific
power of the coal - Measure of the temperatures
Calculation of the efficiency of the furnace 129-139
CHAPTER X.
Discussion of the theory of heat recuperation in the light

of experimental data --General conclusions and classi
fications 140—144
х CONTENTS .
PART II .
CHAPTER XI .
PAGE
Pyrometric practice and instruments — Forms of couples 1
-Standardization - Crucible methodElectric fur

naces - Wire method --Pyrometric galvanometers
Forms of pyrometers.. 147—173
CHAPTER XII .
Gas analysis — Sampling — Sampling tubes—Water-vapor

determination – Tests - Manipulation - Calculations
Automatic self-registering gas-analyzing machine .....174–191
CHAPTER XIII .
Calorimetry — Types of calorimeters—Method of using

apparatus - Observations required - Calculation of re
sults - Accuracy of calorimetric method - Approximate
determination of the calorific power of a fuel by Mahler
and Goutal formulas — Table 12 .. 192-222
CHAPTER XIV .
Elementary fuel analysis - Determination of elements .... 223—227
PART III .
CHAPTER XV .
Chimneys — Consideration and details of construction
Height - Gas analysis .... 231-237
CHAPTER XVI .
Regenerators - Velocities and sojourn of gas currents in

regenerative furnaces- Details .. 238-251
CONTENTS . XI
CHAPTER XVII.
PAGE
Recuperators -- Details... 252-261
CHAPTER XVIII .
Reversing Valves - Hearth - Ports ... 262-271
CHAPTER XIX .
Powdered fuel - Coal-dust stokers - Details ... .2724282
CHAPTER XX .
List of United States patents covering the stoking of

powdered fuel ..... 283–286
CHAPTER XXI .
Tables 13 to 21 -- Conversion tables - Tables 22 to 24 .... 287–299
BIBLIOGRAPHY .
Fuels and calorimetry - Combustion - Gas analysis — Ther

mal chemistry - Chimneys - Pyrometry - Furnaces
Metallurgy ... ... 300_305
TRANSLATOR'S NOTE.
The knowledge of the lack of any book in English covering

the field of a modern study of furnaces led to the preparation
of this work. The first part of the volume is a translation of
Emilio Damour's "Le Chauffage Industriel et les Fours à Gas." I
have supplemented the work with diagrams, plates, tables and nu
merical problems. The second part of the bookhas been written to
cover the field of Pyrometry , Gas Analysis, Calorimetry, and Fuel
>
Analysis, which have only been touched upon in the work of

Damour. Chapters have also been added to give typical examples of
the methods to be followed in designing the various parts of gas
recuperative furnaces.
It has been my aim to present a book which would be helpful
alike to the works manager and to the theoretician . I have had
the opportunity of applying the methods given in tests made upon
the furnaces of the New Jersey Zinc Company.
I wish to express my warmest thanks to Myrick N. Bolles, Ph.
D. , for his contribution of the chapters on “ Calorimetry ” and
the " Ultimate Analysis of Fuels, " as well as for his many valuable
suggestions in reviewing the manuscript.
A. L. QUENEAU.
South Bethlehem , Pa . , September 21 , 1905.
PREFACE.
Mr. Queneau has here laid the industrial world, and especially
the metallurgical world , under great obligation , by bringing to
gether his own admirable translation of Damour's admirable
work and welcome chapters by himself and Dr. Bolles, both of
whom I am happy to count among my past students, on the con
trol and efficiency of heating operations through pyrometry and
chemical analysis, and on the design of chimneys and of regen
erative gas furnaces. To the practicing metallurgist as well as
to the advanced student, this work should be very valuable, giving
as it does, with well combined clearness and condensation , the re
sults of scientific experiment and of practical experience, and
the explanations needed for a firm grasp of the subject. The
writers are to be congratulated on having reduced within moderate
limits the use of mathematics, probably recognizing, as so many
experienced teachers do, that to the minds of nearly all men ideas
are brought by far most easily by means of their own native
and every -day language, and of visible pictures. If we are to
teach , let us teach not only with all our force , but with all our skill,
along the line of least resistance to the introduction of new thought,
that with the strength and time allotted us we may work the great
est aid to our fellows.To all but a very small group of men
mathematical formulæ remain not only a foreign but a repellant
language, to be used for giving perfect precision and definition to
ideas already expressed in the vernacular, or in extremities when
verbal language fails, somewhat as when in despair we turn to
Latin or Greek to drive a thought home deeper and surer than
we can with English , sometimes perhaps because of its defects,
but oftener because of our own clumsiness in its use ; or as when
we turn to medicine because diet and temperance fail .
By his translation , in which his skill in a foreign language, our
XIV PREFACE .
puzzling English, rouses our applause, Mr. Queneau gives us easier

access to Damour's valuable work ; but in their original chapters
he and Dr. Bolles add the fruits of their own keen study, and of
their labors at once faithful, strenuous, and efficient.
Let us, their beneficiaries, wish them the fullest enjoyment of
the second greatest of human pleasures, the consciousness of work
valuable to our kind, well and faithfully done .
HENRY M. Howe .
Columbia University in the

City of New York ,
March 23 , 1906 .
INTRODUCTION .
1
The problem of industrial heating is very complex and often

more difficult of treatment than questions of a purely scientific
nature, since in addition to the attainment of a high temperature,
the proper heating of a furnace of large dimensions, and other
purely technical questions, comes the economical factor, the touch
stone of success .
The problem may be stated as follows : An industrial furnace
is an apparatus, applicable to any industry, so designed that it
will economically produce a high temperature, uniformly distributed
in a hearth of large dimensions.
In the general case there are four conditions to be satisfied ,
namely : Economy, high temperature, all dimensions and all in
dustries. Sometimes the problem becomes more complicated .
Thus, a furnace may be required to give alternately high and low
temperatures ( re-heating furnaces ), or to give simultaneously
different temperatures in different parts of the hearth (glass fur
naces) , or, still further, different temperatures in different com
partments (Hoffman furnace for bricks and potteries ).
For other adaptations, on the contrary, the problem is sim
plified ; for instance, when a low temperature will suffice, a hot
flame is useless and even harmful (steam boilers ); again , where
there is limited hearth area, necessitated by hand working ( puddling
furnaces ). Even the question of fuel economy, which would appear
all important in industrial furnaces, may become secondary when
it conflicts with factors of greater importance financially, as the
efficiency of the men , etc. The engineer should be able to solve
not only the general problem , but further to modify his design to
suit special cases.
The selection of the proper type of furnace is made difficult
by the considerable number of apparatus extant of widely different
2 INTRODUCTION .
designs, all ingenious, the advantages of which are often more

apparent than real.
The four general conditions covering the heating problem have
often been confused . Siemens' invention of recuperation offered
a solution so perfect that it at once realized a saving of from one
to two- thirds of the fuel. Temperatures higher than could be
previously obtained permitted the heating of furnaces of dimen
sions unknown before and , furthermore, the invention was suited
to most varied applications , as the fusion of steel and glass --sub
stances differing as radically as possible, from a thermal point of
view .
The rapid advance created by one form of apparatus caused
some confusion . Engineers and builders came to think that the
different phases of the heating problem were all intimately con
nected . High temperature and recuperation appeared to be
functions of each other, and gas firing, with recuperation , seemed
to be suited only to large units and ill adapted to small ones; even
the character of the substances to be heated was lost sight of.
These ideas are not only most inaccurate, but also detrimental
to the correct study of furnaces. To be convinced of this, it is
sufficient to recall that, disregarding economy, it is possible to ob
tain very high temperatures without recuperation by an indepen
dent heating of the air of combustion, as was done for many years
in blast -furnace practice ; likewise, in ceramics, it was possible
to heat very large kilns without recourse to gas firing ; there are
instances even in which fuel economy is not to be sought at all
by recuperation. We will show this to be the case for low tem
peratures.
In order to be able to solve rationally the many problems met
with in industrial heating, the various phases of heat utilization
must be taken up, and the several questions which have been
united in the happy solution offered in the gas-fired recuperative
furnaces must be separated .
A complete study of the art of heating comprises, then, the
following chapters :
1. Heat utilization in furnaces; fuel economy ; recuperation of
lost heat .
INTRODUCTION . 3
2. Production of high temperatures; influence of pre-heating

the gases ; radiation losses.
3. Furnace dimensions; large furnaces with long flames; small
furnaces with short flames; influence of the velocity of the gases
and of the shape of the issuing ports.
4. Applications of gas-fired furnaces to varied industries — metal
Turgy, ceramics, etc.; design of the furnace as influenced by the
character of the body to be heated .
5. Finally, fuel gasification calls for special study, to ascertain
the most suitable type of producer and the best methods of gasi
fication , which should vary according as the gas is used hot or
cold , and whether with or without recuperation .
Of all these questions the one of heat utilization is doubtless
the most interesting, now that the production of high temper
atures is no longer a serious problem . Economy is the constant
aim of modern industry. It is a problem which is and always
will be of necessity interesting, and therefore it is the one which
will be treated first.
It is the purpose of the present work to study the means of im
proving the utilization of heat and of decreasing, through recuper
ation, the loss of heat in the waste products. To accomplish this
study, the supposition will be made that the furnaces are governed
by the physical laws of combustion and exchange of calories, each
phenomenon taking place completely . At first the practical con
siderations, as that of time, which limit most of these phenomena,
will not be ken into account . In a word , a Theory of Recuper
ation will be built, which will allow the classification of furnaces
according to their maximum efficiency. This will constitute
the first part of this book .
In the second part will be introduced into the discussion the data
given by the experimental study of furnaces. It will then become
possible to make a thermic balance sheet which will serve for
reference, and also to give the ultimate classification of the various
systems of heating in actual use .
This will then fully cover the first problem : Heat Utilization and
Recuperation .
EMILIO DAMOUR .
1
2
PART I
INDUSTRIAL FURNACES
EXPERIMENTAL DETERMINATION OF
THEIR EFFICIENCIES
BY
EMILIO DAMOUR
Ingenieur Civil des Mines
Chef des Travaux Chimiques a l'Ecole Nationale Superieure des Mine
Authorized Translation and Additions

by
A. L. J. QUENEAU , B.S. ( Paris) , E.M. , A.M. ( Columbia)
Consulting Engineer, The New Jersey Zinc Company
1
$
>
LIBRARY
OF THE
UNIVERSITY
OF
CALIFORNIA
PART 1.
THEORETICAL STUDY OF RECUPERATION .
CHAPTER I.
SCIENTIFIC Data NECESSARY FOR THE STUDY OF COMBUSTION

PHENOMENA .
The basis for all thermic study is the knowledge of the heats of
combustion , for the determination of which science is indebted to
Vieille and Berthelot, who used for their work the calorimetric
bomb. But these data were insufficient to solve the problem we
have undertaken , until Mallard and Le Chatelier had determined
the specific heats and thermal capacities of gases. It is easily seen,
then, that it is only recently that it has been possible to solve the
theoretical question of the industrial production and utilization of
heat .
Heat of Combustion and Calorific Power . - By heat of combustion
under constant pressure, or more simply heat of combustion of a
chemical element or of a well defined chemical compound , is meant
the number of calories? liberated by the combination with oxy
gen of a number of grams of this substance equal to its molecular
3
weight. By definition, when the body is a gas, the molecular
weight corresponds to a volume, alike for all gases , of 22.32 liters .
This unity of volume, adopted by Berthelot in his thermochemical
>
work , is what we will call the Molecular Volume.

1Mallard and Le Chatelier madethese determinations jointly , in This
well known work as members of the " Commission du Grisou .
connection with their
explains without
doubt why these data , all important to science and industry, have been promulgated so
slowly .
We make use of the large calory throughout this work .
The molecular weights of elements are generally double their atomic weights ; the
molecular weights of compounds are equal to the sum of the molecular weights of the
constituent elements.
8
1.
TABLE
REACTIONS
-CHEMICAL
THERMO
OF
TABLE
Kilograms
Available
Calories
Required
per Weight
Kilogram of
Substance Combustion
of
Reaction Remarks Products
of of
Sub-
Oxy
of
of Combustion
Air
Oxygen
gen
stance
perMol.Vol.
Molecular
Calories per
Formation
Heat of
Volume
Mol . Vol .
Molecular
Formula
Molecular
Liberated
. it.
LGrs Water
69.0
Liquid
.as } 4223
34500 .Kgs
Hydrogen
. 20
.H 22.32
H2+H2
02- 34.664
8.000 9.000
58.2
Water
as
Vapor 29100
3637
Carbon
. +10
,C 29.4 1931
2602 2.333
5.777
1.333
Carbon
Monoxide 22.32
0
$CO
+282 68.2 29.4 4263
2436 1.571
2.472
.571
.Carbon C2 C
42
+0
=O2 97.6 3030
8080 3.667
11.555
2.667
195.2
Water
Vapor
as 3050
12200 =2.750
CO2
Methane
. CH C
22.
16
,=.+2
20
..0H
O232CH 18.8 4.000
17.332
Water
213.5
Liquid
.as 3266
13063 =2.250
H20
០បុវa \ ។ ពី ដី ។
304.9
Water
Vapor
as
.. 3812
11727 3.385
CO2-
22.32
26
C2H2
2
,-
210
22+
.!+H
CAcetylene
..0H002 51.5 13.256
3.076
Water
315.7
Liquid
.as 3947
12142 .H20
=691
339.9
Water
Vapor
as 113303036 =2.933
CO2
Ethane
. 201
C2H6
22.32
C2H30
!+3=2H102
C02 29.9 3.733
16.203
Liquid
as
Water
.372.3 3325
12410 H20-1.800
Water
319.6
Vapor
as 3328
11414 3.143
CO2-
Ethylene
. 20.
+3
C.H.
22.32
28
C2H4
2=
2
1+0
002
H 8.0 14.848
3.429
.
Liquid
as
Water
341.2 3458
11858 H20-1.286
Gas
As
69.2 2170
2170
Sulphur
. +
2-
.S
03S2O2 4.350
1.000 2.000
In
8
.76
Solution 2400
2400
.
Solid
103.6
As 2170
3220
Sulph
. ur S 32 ,-
$0
+
OO2
.S3. Solution
..
In
141.0 2940
4400 6.525
1.500 2.500
M
.
Gas
As
91.8 1920
2800
INDUSTRIAL FURNACES AND METHODS OF CONTROL .
TABLE
.2
)
PURE
0
AND
FUELS
CHEMICALLY
KILOGRAM
OF
ONE
COMBUSTION
THE
REQUIRED
FOR
AIR
WEIGHTS
AND
VOLUMES
THE
COMBUSTION
.
RESULTING
OF
PRODUCTS
WEIGHTS
THE
AND
VOLUMES
REQUIRES COMBUSTION
PRODUCTS
AS
PRODUCES
AND
Cubic
Meters
Kilograms
in Meters
Cubic
in in
One
Kilogram
of to
Burning Kilograms
Oxy Nitro Oxy Nitro CO2 2H0 N Total
Air Air CO
gen gen gen gen
2.33 4.46 6.79

CO 1.33 4.46 5.79 0.93 3.55 4.48
1.86 3.55 5.41
1.57 1.91 3.48

CO CO2 0.57 1.91 2.48 0.39 1.53 1.92
0.80 1.52 2.32
3.67 8.93 12.60

C. CO2 2.67 8.93 11.60 1.87 7.15 9.02
1.87 7.12 8.99
9.00 26.72 35.72
H H20 8.00 26.72 34.72 5.59 21.29 26.88
11.19 21.29 32.48
2.75 2.25 13.36 18.36
CH 2H
+
CO2
20 4.00 13.36 17.30 2.80 10.64 13.44
1.40 2.79 10.65 14.84
3.14 1.29 11.49 15.92
C2H 2H20
,+
2C0 3.43 11.49 14.92 2.40 9.12 11.52 1.60 1.60 9.12 12.32
2.93 1.80 12.42 17.15
C2H 3H
+
2002
20 3.73 12.42 16.15 2.07 9.90 11.97 SO2
.
0 1.49 2.23 9.90 13.62
THEORETICAL STUDY OF RECUPERATION .
2.00 3.33 5.33

S. SO2 1.00 3.33 4.33 .69 2.65 3.34
0.70 2.65 3.35
Gas
50 Toldt
egenerativ
,“ fen
age
R.,"p-oFriedrich
6
10 INDUSTRIAL FURNACES AND METHODS OF CONTROL .
When the body under combustion is a fuel, coal, coke, etc. , at

the ordinary temperature, the heat of combustion is designated
more specifically by the name of Calorific Power. This experi
mental constant, easily determined by means of the Mahler bomb,
represents, then , the available calorific energy - that is, the quantity
of heat the complete utilization of which is to be sought, since all
industrial combustions take place without compression , at a pres
sure very nearly equal to that of the atmosphere. These two
data - the heat of combustion and the calorific power under con
stant pressure - are those that perform the fundamental function
in the theoretical and practical questions occurring in industrial
heating
The heat of combustion of chemical elements , as well as the
heats of combustion produced by the combination of these elements,
have been determined by Berthelot and are published in the “ An
nuaire du Bureau des Longitudes ." From the 1897 edition we
have taken the figures required for our study, and have formed
from them a table which will be sufficient for all problems of com
bustion in furnaces.
Thermal Capacities and Specific Heats of Gases. - When a gas
maintained at a constant pressure is brought from an initial tem
perature T ,0 to a higher temperature T, the number of calories ab
1
sorbed by the gas in this operation is called the Thermal Capacity

between the temperatures T. and T.
Inversely, the Thermal Capacities will give the number of cal
ories set free by the gas when cooling from T, to T. These con
.
stants are the basis for all calculations of exchange, in calories, in

the phenomena of heating and combustion.
The Specific Heat of a gas at any temperature, which we do not
need to limit , is easily derived from its Thermal Capacity. It
is the first derivative of the function expressing the Thermal 1
Capacity. It is not possible, inversely, to derive from the specific

heat of a gas at a given temperature, or even from the mean specific
heat between the temperatures 0° and 100 ° C., the Thermal Capacity
of the gas at a temperature above 100° C.
The fact is that all Specific Heats are variable and increase in
general with the temperature. If the law of increase is unknown
THEORETICAL STUDY OF RECUPERATION . 11
it is evidently impossible to go from the derivative to the primitive

function , that is, from the Specific Heat to the Thermal Capacity .
It follows that the Specific Heats of gases under constant pressure
between 0° and 100 ° C. , determined by Regnault, are not sufficient
to calculate the quantity of heat absorbed by a gas in heating or
radiated in cooling, and hence all calculations based on these figures
are subject to a more or less grave error, the importance of which
could not formerly be estimated . In short, the problem of combus
tion and exchange of calories cannot be taken up with the single
datum of the specific heats between 0° and 100 ° C.
The formulas expressing the law of heating of gases under con
stant pressure have been given by Mallard and Le Chatelier as a
function of absolute temperature .
The Thermal Capacities of Gases from absolute 0° (-273° C. ) to
a temperature T = 273° + t may be expressed by a parabolic
formula of two parameters :
T T?
= a + b
1000 10002
in which " a " is a constant, common to all gases, equal to 6.5, and
“ b ” is another constant, variable for the different gases ; its values
are :
Perfect Gases Water Vapor Carbon Dioxide Methane

O 2, N , H 2, CÓ H2O CO , 2 CH ,
0.6 2.9 3.7 6.0
It follows that the total Thermal Capacities of a Molecular Vol

ume (22.32 liters ) of any gas between
to = T. — 273° and t T- 273 °
will be expressed by the formulas given in Table 3.

1 A large number of calculations of heat balance-sheets have been published , based on
these insufficient data .
TABLE 3 .
THERMAT. CAPACITIES OF GASES PER MOLECULAR VOLUMES
IN ABSOLUTE TEMPERATURES
Gas Under Constant Pressure Under Constant Volume
T - TO0 T2 – T.2 T - T T2 - T.2

Perfect Gases ( O2, N 2 , H 2, CO ) 6.5
103
+ .6
106
4.5
103
+.6
106
Water Vapor, H 20 . 6.5

T— To+ 2.9
T2 — T.2
4.5
T - To + 2.9 T2 - To 2?
103 106 103 106
T - TO+ T2 — T , T- To T2 -T02
Carbon Dioxide, CO 2 6.5 3.7 4.5 + 3.7
103 106 103 106
Methane , CH , 6.5
T— To+ 6.0
T2 – To ? 4.5
T - To+ 6.0
T2 – T.2
103 106 103 106
TABLE 4 .
THERMAL CAPACITIES OF GASES PER MOLECULAR VOLUMES
IN CENTIGRADE DEGREES
Gases Under Constant Pressure Under Constant Volume
t t2
.6 t 2
t
Perfect Gases ( 02, N 2 , H2,
, CO) 6.83 + .6 4.83 +
103 106 103 106
t t2
Water Vapor, H 20 .. 8.08
103
+ 2.9
106
6.08 + 2.9
103 108
t2 t
Carbon Dioxide, CO 2 8.52
103
+ 3.7
106
6.52
103
+ 3.7
106
t2 t2
Methane, CH . 9.78
103
+ 6.0
106
7.78
103
+ 6.0
106
TABLE 5 .
THERMAL CAPACITIES OF GASES PER KILOGRAM IN CENTIGRADE DEGREES
Oxygen ... .213t + 19 X 10 -et ? .150t + 19 X 10 -67 2

Nitrogen and Carbon Monoxide . .243t + 21 X 10 - 072 .1710 + 21 X 10 -6t 2
Hydrogen .. 3.400t + 300 X 10 -ft 2 2.400 + 300 X 10 -67 2
Water Vapor .417t + 162 X 10 -642 .335t + 162 X 10 -67 2
Carbon Dioxide , 193t + 84 X 10 -6t 2 .150 + 84 X 10 -67 2
Methane . .608t + 374 X 10 -6t ? 491t + 374 X 10 -6 2
TABLE 6.
THERMAL CAPACITIES OF GASES PER MOLECULAR VOLUME
Temperatures 02, N 2 , H2, CO H20 CO2 CH ,4 0.002t
0 C. 0 0 0 0 0
200 1.39 1.73 1.85 2.19 0.4
400 2.82 3.69 3.99 4.85 0.8
600 4.31 5.87 6.44 8.02 1.2
800 5.82 8.23 9.18 11.46 1.6
1000 7.43 10.98 12.22 15.77 2.0
1200 9.05 13.87 15.55 20.37 2.4
1400 10.73 17.00 19.18 25.44 2.8
1600 12.46 20.35 23.10 30.99 3.2
1800 14.21 23.86 27.21 36.86 3.6
2000 16.05 27.76 31.84 43.55 4.0
2200 17.91 31.82 36.65 50.54 4.4
2400 19.84 36.10 41.76 58.02 4.8
2600 21.81 40.62 47.16 66.04 5.2
2800 23.82 45.64 52.84 74.42 5.6
3000 . 25.89 50.64 58.86 83.34 6.0
TABLE 7 .
THERMAL CAPACITIES OF GASES PER KILOGRAM .
Temperatures 02 N 2, CO H2 H 20 CO2 CH je .096
OC. 0 0 0 0 00
200 .. 47.3 50 700 100 43.1 136.6 18
400 . 88.0 100 1400 203 91.0 303.0 36
600 . 134.0 154 2150 326 145.0 499.0 54
800 . 181.0 207 2900 461 208.0 726.0 72
1000 .. 232.0 264 3700 609 277.0 982.0 90
1200 . 284.0 325 4550 770 354.0 1269.0 108
1400 . 334.0 383 5350 943 435.0 1584.0 126
1600 . 391.0 445 6250 1130 523.0 1931.0 144
1800 . 444.0 508 7100 1330 618.0 2307.0 162
2000 . 503.0 575 8050 1542 728.0 2712.0 180
2200 558.0 637 8950 1751 840.0 3148.0 198
2400 620.0 708 9900 1985 950.0 3614.0 216
2600 . 681.0 777 10900 2241 1070.0 4109.0 234
2800 735.0 850 11900 2520 1200.0 4635.0 252
3000 810.0 921 12950 2799 1355.0 5190.0 270
These formulas make it possible to find the Thermal Capa

cities between any two temperatures.
The Specific Heats at any temperatures are to be obtained by
taking the first derivative of the formulas for the Thermal Ca
pacities. They are given in Table 2. This, however, is without
interest to our study. As we will constantly be obliged to compute
the Thermal Capacities, it has appeared advisable to reproduce
them in a table, for each 200°, between the limits of temperature

found in industrial furnaces .
By plotting the expression for Thermal Capacity in rectangular
coördinates, we obtain a curve from which it is possible to find,
with sufficient accuracy, the values of the Thermal Capacity for
any temperature.
Temperatures of Combustion . — The knowledge of the Heats of
Combustion and of the Thermal Capacities makes it possible to
solve a fundamental problem in our study — the problem of the
Temperatures of Combustion .
By “ temperature of combustion ” is meant the temperature to
which the gaseous products are brought, after combustion under
constant pressure. It is supposed that this action takes place so
rapidly that radiation is negligible and that the enclosure is ather
man . These conditions are nearly similar to those met in prac
tice ; the pressure in the furnace is that of the atmosphere. The
instantaneousness of action and the athermancy of the furnace are
not entirely fulfilled; this explains why the temperature of a flame
in a furnace is inferior to the temperature of combustion . How
ever, other things being equal, the temperatures of combustion
allow an estimate of the practical value of a fuel which the calorific
power and the specific heats alone cannot do. Formerly many
computations, based only on the calorific power of the gases, were
made in order to estimate the value of a special furnace ; these
data are always insufficient.
The computation of the exact temperature of combustion is
logically derived from the knowledge of the Thermal Capacity.
Soon after the completion of the work of the Commission du Grisou ,
Professor Le Chatelier showed how the former theoretical com
putations of temperatures were inaccurate.
General Equation of the Temperature of Combustion . — During
combustion the heat liberated, under constant pressure, is used i
only to raise the temperature of the final gaseous mixture; and
since by hypothesis the phenomenon is atherman , there is equality
between the latent heat of combustion and the sensible heat con
tained in the gaseous mixture after combustion has taken place ,
hence the equation :
Σ 2a (T— T. ) + Eb (T? —T,? ) ( 1 )

L := Ea
This is a corollary of Joule’s principle. Since no work is done
under constant pressure , the sum of the quantities of heat in play
in a closed cycle is equal to zero .
ot
L
-- cdt (2 )
The equation ( 1 ) is of the second degree, and gives the value

of T as a function of T. The solution of the equation may be
0.
tedious in case of a complex mixture of gases. It is much quicker

to use the graphic method , if a table of Thermal Capacities is at
Calories
hand .
EL
rt te
Ea - To ) + [ or TOP
X
Х
1400 1600 1900 ° Temperatures
Fig . 1. - GRAPHIC METHOD FOR THE DETERMINATION OF THE TEMPERATURES
OF COMBUSTION .
Graphic Method for the Computation of the Temperatures of Com

bustion . — The intersection of the line EL with the parabola
za (T - T. ) + εb ( T ? -- T ,?)
is to be found. The parabola is traced by points corresponding
to temperatures for which the Thermal Capacities are given in the
table ( see preceding paragraphs), and thus the solution of the equa
tion is obtained at once.. The graphic method is used entirely in

these pages; it cannot be commended too highly for rapid computa
tions ; it has the further advantage that, though not so accurate as
the mathematical method, it is much less liable to error.
Unless otherwise stated, all the computations given in the present
work are made with the supposition that a molecular weight of
the fuel is burned. This molecular weight will be taken as the
basic unit. It is evident that , were any other weight taken , the
same temperature would be obtained, because in the general equa
Calories
the masses do not enter as a factor.
yo
70
60]
0 = 58.2
50 Temp .
1800 20000 2200 X
X,
Fig. 2. — TEMPERATURE OF COMBUSTION OF HYDROGEN IN Air.
Temperature of Combustion of Hydrogen in Cold Air . - In order

to burn the molecular weight of hydrogen , 2 grams (22.32 liters ),
to water vapor, a half molecular volume of oxygen ( 10, ) is re
quired ; this oxygen will be accompanied by about a quadruple
volume of nitrogen ' (2N2 ) , that is to say, two molecular volumes.
From the table of Thermal Capacities may be found the number of
' In volume air is composed of 20.8 parts of oxygen , and of 79.2 parts of nitrogen ,
or of one part of oxygen for 3.8 parts of nitrogen , or about 4 parts. To simplify our
computations we will always take the ratio 1 : 4. For more accurate ratios see Tables 13
and 14 .
calories required to raise this gaseous mixture to different tem

peratures.
Products of Combustion 1800 ° 2000 ° 2200 °
a b с
H20 . 23.86 27.76 31.82
2 N2 28.42 32.10 35.82
Total 52.28 59.86 67.64
The total heat liberated by 2 grams of hydrogen burning to

water vapor is 58.2 calories, which is the value of SL in the general
equation. From the above table it is seen at a glance that the
unknown temperature of combustion is included between 1800 °
and 2000 ° C. If the curve characterized by the three points
a, b , c is drawn , its intersection with the line Q = 58.2 will give,
graphically, the temperature of combustion of hydrogen : 1970° C.
Temperature of Combustion of Carbon . - Burned with cold air
and without any excess of air, the temperature of combustion of
carbon is 2040° C. ( see page 21 ) ; but if we suppose that the products
of combustion carry 5% of free oxygen , or 25% excess of air, the
temperature is lowered to 1650°.
These two results are interesting, as showing the maximum
temperatures that can be obtained in a furnace. These tempera
tures represent the limit of temperatures of direct- fired furnaces.
In such furnaces it is practically impossible to secure complete
combustion without a large excess of air, rarely less than 5% .
This temperature of 1650 ° C., which does not allow for radiation
and all other exterior losses of heat, shows conclusively the im
possibility of obtaining high temperatures by direct firing.
If there is an excess of 5% of carbon monoxide in the products
of combustion , the temperature reaches a higher figure - 1930 ° C.
Such a combustion is to be obtained only in gas firing ; this indi
cates that, in order to obtain high temperatures, it is advisable,
in gas furnaces, to run with excess of carbon monoxide in the waste
products, though evidently it is less economical, the coal consump
tion being higher. If economy and temperature are both taken
into account, each particular case must be taken into considera
tion , but in general it.may be said that, for furnaces requiring high
temperatures, the running with excess of carbon monoxide is to
be preferred . Excess of air, on the other hand, is preferable for
low -temperature furnaces, as steam boilers, where it is of prime
importance that the combustion be complete.
In both cases, the running without excess of either air or carbon
monoxide is more advantageous in respect to the amount of coal
burned .
General Case. Find the temperature of combustion of a pro
ducer gas, the volumetric analysis of which is as follows:
Carbon Dioxide. 5.0 %
Carbon Monoxide . 20.0
Hydrogen . 12.0
Methane . 3.0
Water Vapor . 2.0
Nitrogen . 58.0
100.0 %
The gas is pre-heated to 1000 ° C.; the air of combustion is also
pre- heated to 1000 ° C.
The volume of air required for the complete combustion of the
gas is obtained as follows :
20 volumes of CO require 10 volumes of oxygen
66 66 C6
12 H, 6
66 66 66
3 CH , 6
Or, for 100 volumes of the gas , 22 volumes of oxygen are required
or 110 volumes of air.
The heat brought by the air and gas at 1000 ° C. is as follows:
Composition in Calories as Calories as
Gas
Mol . Vol. Sensible Heat Latent Heat
CO2 05 .6
CO 20 13.6
H2 12 6.6 6.9
N .58
CÁ, .03 .5 5.6
H20 .02 .2
1.00
Air required.... 1.10 8.1
Total. 16.0 26.1
EL 42.1 Calories
The following table gives the Thermal Capacities for the volumes
of the products of combustion, for different temperatures.
Thermal Capacities
Products of Combustion Volumes
2000 ° 2200 ° 2400 °
CO , 0.28 8.5 10.3 11.7

N , 1.46 23.4 26.1 28.9
H20 .20 5.5 6.3 7.2
37.4 42.7 478
It is seen at a glance that the temperature of combustion is

nearly 2200 ° C. This very high temperature shows the practical
value of pre-heating the air and the gas for high temperatures.
Combustion of Carbon Monoxide in Cold Air.-The weight of
carbon monoxide to be burned is 28 grams, corresponding to the
molecular volume, 22.32 liters; the heat liberated is 68.2 calories.
Calorie
70
*
P = 682
600
X
sal T
0012
18000 2000 ° 2200 Temperature

FIG. 3. - TEMPERATURE OF COMBUSTION OF CARBON MONOXIDE IN AIR .
20 INDUSTRIAL FURNACES AND METHODS OF CONTROL ,
By plotting the curve characterized by the three points a, b, c,

and the straight line Q 68.2 , we find their intersection to cor
respond to a temperature of 2100 ° C.
The following table gives the thermal capacities:
Thermal Capacities
Waste Products
1800 ° 2000 ° 2200 °
CO2. 27.2 31.8 36.6

N2 28.4 32.1 35.8
a = 55.6 b 63.9 C = 72.4
Temperature of Combustion of the Theoretical Water Gas (CO

+ H , ) in Cold Air . - Its combustion takes place according to the
following reaction :
CO + H2 + 0 , CO 2, + H2O + 68.2 + 58.2 126.4 calories .
Hence the following table and diagram ( Fig. 4 ) .

Calorie
140
130
02 /26.4
6
120
no
X,
20301
100
20000
1800 "
2200 ° Temperature
Fig . 4. - TEMPERATURE OF COMBUSTION OF WATER Gas (CO + H , IN AIR
Thermal Capacities
Waste Products
1800 ° 2000 o° 2200 o°
CO,. 27.2 31.8 36.6

H20 . 23.8 27.7 31.8
4N2 56.8 64.2 71.6
Total 107.8 123.7 140.0
The intersection of the two lines gives 2030° C. as the tempera

ture of Combustion .
The preceding results show that if carbon monoxide, CO , is free
from nitrogen, its temperature of combustion is higher than that
of hydrogen ; also , that that of the theoretical water gas is inter
mediate between the two.
Temperature of Combustion of Carbon in Cold Air . — The weight
of amorphous carbon burned is 12 grams. This corresponds to
the molecular volume 22.32 liters of carbon dioxide, CO2. Heat
liberated , 97.6 calories. The following table and diagram give a
temperature of combustion of 2040° C.:
1104
100
Q - 976 -
Calories
90
X ; = 20400
80
1800 " 2000 " 2200 ° C
Fig. 5. - TEMPERATURE OF COMBUSTION OF CARBON IN AIR.

Thermal Capacities
Products of Combustion 2200 °
1800 ° 2000 °
CO2 . 27.2 31.8 36.6

4N2 56.8 64.2 71.6
84.0 96.0 108.2
X - 2040 ° C.
Temperature of Combustion of Bituminous Coal in Cold Air.

Analysis of the bituminous coal:
Total Carbon .. 75.2 %
Oxygen 02.. 8.2
Nitrogen N2 . 1.0
Hydrogen H , 5.2
Hygroscopic Water H20 . 3.4
Ash . 7.0
100.0 %
The Calorific Power determined in a Mahler bomb calorimeter

was found to be 7423 calories.
The combustion of this coal will yield waste products containing
the following :
Carbon Dioxide,
Water of Combination ,
Hygroscopic Water,
Nitrogen coming from the coal,
Nitrogen coming from the air.
The composition of the products of combustion in Molecular
Volumes is obtained as follows :
Twelve grams of carbon give, in burning, one molecular volume
of carbon dioxide ; 75.2 grams give 6.27 vol.
Two grams of hydrogen give, in burning, one molecular volume
of water vapor ; 5.2 grams give 2.6 in molecular vol .
Eighteen grams of water give one molecular volume of water
vapor ; 3.4 grams give 0.19 vol., a total of 2.79 vol . of water vapor.
The oxygen required for combustion is :
6.27
1.30
volumes for the formation of CO2
H2O
Molecula r volume corresponding to 8.2 grams of Oxygen in the fuel, 8.2 + 32 – .25
r
molecula volumes .
Oxygen required from the air , 7.57 — .25 7.320 Vol .
Nitrogen required " 7.32 X 4 29.280
from the coal , 1.0/ 28 = 0.036
M
Total volume of Nitrogen . 29.32
The volumetric composition of the waste products is, then :

Carbon Dioxide . 6.27
Water Vapor 2.79
Nitrogen ... 29.32
The combustion of 100 grams of coal liberated 742.3 calories.

Referring to the table of Thermal Capacities per M. V .:
Thermal Capacities
Products of Combustion Mol. Vol .
1600 ° 1800 ° 2000 °
CO2.. 6.27 144.8 170.6 199.6

H2O.. 2.79 56.7 66.5 77.4
N2 29.32 365.3 416.5 4710
Total. 566.8 653.6 748.0
It is seen at once that the maximum temperature to be realized

with this coal in a direct - fired furnace is nearly 2000 ° C.
Temperature of Combustion of Siemens Producer -Gas Burned
with Pre- heated Air . - The composition of the theoretical Siemens
producer -gas (no steam being injected under the grate ) is CO + 2N ,.
Both the air and gas are pre -heated to 1000 ° C. One-half mole
cule of oxygen is needed for the molecule of carbon monoxide :
this oxygen carries with itself two molecules of nitrogen.
Calories liberated by the combustion 68.2
brought by one vol of CO . 7.43
3.72
4vol O N2.
vol.. of
.
29.72
Total. 109.07
The thermal capacities of the combustion products are :
Gas 1800 ° 2000 ° 2200 ° 2400 °
CO2 ... 27.2 31.8 36.6 41.8

4 N2 . 56.8 64.2 71.6 79.4
Total . 84.0 96.0 108.2 121.2
By plotting the curves of Q and of the thermal capacities, we

find the temperature of combustion to be 2215° C. !
In the same manner we find that , if both the air and the gas
are pre-heated to 500° C. , the temperature of combustion is 1865° C.
110,0
109=Q Air
and Gas anot
Tooo
-
Fig. 6 — TEMPERATURES OF COMBUSTION OF SIEMENS Gas (CO + 2N ,).

Calories
100 c
у
90.
500°
at
and
Gas
Air
87.8Q=
1
3
80
X, 22159
a
180501
Xa= =
2
68.2Q= 70
X15000= 1 Temperatures
1500 1600 1800 2000 2200 X
If the air and gas are cold, the temperature of combustion falls
to 1500 °.
Combustion Temperatures of the Pittsburg Natural Gas.. - We -
next seek to ascertain the temperature of combustion of the Pitts

burg natural gas when burned with the theoretically requisite air,
the air being burned cold ; also, the temperature of combustion if
>
the air is pre -heated to 1000 ° C .; further, the temperature of com

bustion of the same gas if 25% excess of air is injected in the
furnace, the air being given, first cold and secondly pre -heated to
1000 °.
The gas has the following composition , by volume :
CH , 67.00 %
H2 . 22.00
CH , 1.00
C2H6 . 5.00
CO . .60
CO2 .60
02 . .80
N2 . 3.00
Total. 100.00 %
Volume of oxygen required for combustion, and the resulting

volumetric composition of the products of combustion :
Volume of Oxygen required Volume of the Products of Combustion
Gas 02 CO2 H2O N2
CH , .67 X 2 1.34 .67 1.34

H 22 X .5 .22
CH , .01 X 3 .03 .02 .02
CH .05 X 3.5 .175 .10 .15
006 X .5 .003 .006
CO2 .006
In gas .03
Broughtby ( 2 6.28
Total 1.658 .802 1.73 6.31

Oxygen in gas . .008
Oxygen from the air . 1.65

Nitrogen from the air 6.28
Total volume of air . 7.93
If we introduce 25 % excess of air , the
volume of air becomes 7.93 + 1.98 9.91
26 INDUSTRIAL FURNACES AND METHODS OF CONTROL.
The composition of the products of combustion in this latter

case becomes :
Molecular Volumes.
CO 8.02
1.73
H02., .41
7.85
Thermal capacities of the products of combustion, in the case

of no excess of air :
Gas Vol. 1800 ° 2000 2200 ° 2400 3000 °
N2 .. 6.310 89.7 101.3 113.0 125.1 163.4

HO . 1.730 41.3 48.0 55.0 62.4 87.5
CO2. .802 21.8 24.5 29.3 33.4 47.1
Total. 8.842 152.8 173.8 197.3 220.9 298.0
Calories brought by the air at 1000 ° C. (no excess of air):

7.43 X 7.93 58.9 calories.
Calorific power of the gas:

CH , .67 X 195.2 130.78 Calories.
H2 22 X 58.2 . 12.84 88
CH .01 X 341.9 3.42 60

CH . .05 X 321.77 16.09
CO .006 X 68.2 .41
Total 163.54 Calories.
Total calories, sensible and latent, in the case of no excess of air:
58.9 + 163.54 = 222.44 calories .
The intersections of the curve and straight lines give , as the tem
peratures of combustion , 2415° C. when the air is heated to 1000 ° C.
and 1900 ° when the air is cold.
Pillsburg Natural Gas

combustion with No Excess of Air
./
250 25 %
Q = 237.14
Air ar 1000
230
Q = 222./4
Air at 1000
-A
210
190
1
I
I
1 1
1701
1
1
Q = 163.54
1
7 1
24151
1.099/
1
2235
1900
150
1600 1800 2000 2200 24005
Fig . 7.
Thermal capacities of the waste products with 25% excess of air :
Gas Mol . Vol . 1600 ° 1800 ° 2000 ° 2200 °
N2 .. 7.85 8.26 103.0 117.4 132.5

.41
148.0
02 ...
H20 . 1.73 35.2 41.3 48.0 55.0
CO2 .802 18.5 27.8 24.5 29.3
Total 156.7 180.5 205.0 232.3
Calories brought by the air at 1000 ° C.

7.43 X 9.91 = 73.60 Calories.
Latent heat of gas . 163.54
Total . 237.14 Calories.
From the diagram we find that the temperatures of combustion

are respectively 2235° C. and 1660 ° C.
To what temperature should the theoretically required air be
pre -heated in order that the temperature of combustion be 3000 ° C. ?
The thermal capacities of the products of combustion at 3000 °
are 298.0 calories ; the gas has a latent heat of 163.54 calories;
therefore, the air must bring an extra number of calories equal to
298.0—163.54 134.46 calories.
This is for the total air, which has a molecular volume equal
to 7.93 ; the calories per molecular volume must be equal to
134.46 7.93 16.97 calories.
From the diagram of the thermal capacities we find at once

that the air should be pre -heated to 2100 ° C.
The Natural Gas of Iola, Kansas, contains by analysis:
Methane, CH4 89.66 % by volume.
Carbon Monoxide, CO 1.23
40
Carbon Dioxide , CO2 .90

Nitrogen , N 7.76 00
Oxygen , 0, . 45
Required :
( 1 ) The temperature of combustion, if the gas is burned cold ,
with the theoretical amount of cold, dry air necessary.
(2 ) The temperature of combustion, if burned cold , with the
requisite air pre-heated to 1000 °. Also if pre-heated to 500 °.
(3 ) The temperature of combustion , the gas being burned cold
with 25% more air than theoretically necessary, the air being cold ,
pre -heated to 500 ° and to 1000 °.
( 4 ) The temperature of combustion, the gas being burned cold
with a quantity of air smaller than theoretically necessary, so that
5 % of the gas remains unburned . The air being cold, then pre
heated to 500° and finally pre -heated to 1000°.
Calorific power of the gas per molecular volume (22.32 liters )

Calories per M. V.
CH 8966 X 195.2 = 175.0 Calories .
CO .0123 X 68.2 = .84
Total .. 175.84 Calories.
Molecular volumes of oxygen required for combustion :

CH , 8966 X 2 1.793
CO .0123 X 0.5 .0062
Total. 1.7992
Oxygen in gas .... 0045
Oxygen required from atmosphere . 1.7947
Nitrogen carried by the oxygen :

1.7947 X 3.808 – 6.836
Composition of the products of combustion :

CO .8966 + .009 9056
Ho 1.79 1.79
N2 6.836 + .0776 6.914
Calories brought by air at 1000 ° :

6.914 X 7.43 51.37
Latent calories of gas . 175.84
Total . 227.21

6.914 X 3.57 24.68
Latent calories of gas = 175.84
Total . 200.52
Thermal capacities of the products of combustion :

Gas Mol . Vol . 20000 2200 ° 2400 ° 2600 °
N2 . 6.914 110.95 123.85 137.20 150.08

H20 1.74 49.70 57.07 64.77 72.71
CÓ, 9056 28.83 33. 20 37.82 42.71
Total. 189.48 214.12 239.79 265.50
a b с d
The parabolic curve corresponding to the points a, b, c, d has

been plotted in Fig. 8. Its intersections with the straight lines
Q = 175.84, Q = 227.21 and Q = 200.52 give the required tem
lola Natural Gas .

Combishon
with no Excess of Air
260
Q- 22721
1000
220
c
1
Q = 200.52
Air ar 500
200 1
x - 2400 °
180
Q =__17584 cold, air

213.5 "
1
1885
60
1800 2000 " 2200 2 400 2600
FIG. S.
peratures of combustion , namely , 2400° , 2135° and 1885 ° re

spectively.
Combustion with 25 % excess of air:
Free Oxygen , 1.79 X .25 .45
Excess of Nitrogen . 1.709
Excess of air .. 2.16 Molecular Volumes.

Composition of the products of combustion : 8.623

N2 6.914 + 1.709 -
02 . 45
H20 1.79
CŐ , .91

8.63 X 3.57 30.81
Latent heat of gas : 175.84
Total . 206.65

10.79 X 7.43 = 80.2
Total. 256.04
lola Natural Gas . Combustion

200
with 25 % Excess or Air
Q- 256.04
-
ar 1000
240
X , - 2220 °
220
-
Q = 206.65
ar 500
200
- 1930 % 1
Xa
180
Q = '25 .8 / Cold Aint

.
1
1648
160
1500 1600 1800 2000 2200
Fig . 9

Gas Mol . Vol. 18000 2000 ° 2200 ° 2400 °
N2 9.073 129.0 145.6 162.55 180.00

02.
H202. 1.79 42.72 49.70 57.07 64.77
CŐ2 91 24.65 28.83 33.20 37.82
Total . 196.37 224.13 252.82 282.59

a b с d
The intersections of the straight lines of which the equations are

Q 175.84 , Q -
206.65 and Q 256.04 with the curve charac
terized by the points a, b, c, d give the required temperatures,
namely 1640º, 1930 °, 2220°.
Combustion with 5% of unburned gas :

Calorific power 175.84 X.95 167.1 Calories .
Oxygen required , 1.79 x 95 1.7
Nitrogen required, 1.7 X 3.808 = 6.48
Volume of air . 8.18
Composition of the products of combustion :

CO2 .8966 X .95 + 009 .86
H2O 1.792 X .95 1.703
N 6.48 + .0776 6.55
CẢ . .8966 x .05 044

8.18 X 3.57 29.2 Calories.
Latent heat of gas == 167.1
Total. 196.3 Calories.

8.18 X 7.43 60.8
Total .. 227.9

Gas Mol. Vol . 1800 ° 2000 ° 2200 ° 2400 °
N2 6.55 93.18 105 . 117.3 130 .

HO 1.70 40.56 47.2 54.1 61.4
CO2 .86 23.40 27.38 31.5 35.9
CH 044 1.62 1.91 2.2 2.6
Total... 158.76 181.49 205.1 229.9

a 6 d
In the usual manner, as shown in Fig. 10, we find the required

>
temperatures of combustion to be as follows: 1875°, 2130°, 2385° C.

DISSOCIATION .
Until recently the thermal capacities were known to 100 ° C.
only ; these values were applied by an impermissible interpolation
lola Natural Gas. Combustion

250
with 5 % of Gas unburned
1
230 P = 2279
1
Air at 10000 1
210
Q= 796.3 |
Air at 500 °
190 .
1
170
Q = / 167 / 1
Cold Tair I
1 21300
a 1
!
18751 2385°)
150
1800 2000 22009 2400 200g
Fig. 10
to the calculation of the temperatures of combustion, and, natu
rally, very much higher temperatures were obtained .
In order to explain the discrepancy between theoretical figures
and practice, physicists claimed that dissociation played an im
portant role ; that, by preventing the complete combination of the

elements, the number of calories liberated was decreased , thus
7
bringing the computed figures nearer the values obtained in prac

tice. On the authority of Mallard and Le Chatelier, the phenom
enon of dissociation , even at 2000 °, is of little importance, as will
be seen from the tables giving the results of their research . '
Dissociation of Carbon Dioxide.
TABLE 8 .
DISSOCIATION OF CARBON DIOXIDE .
Coefficients of Dissociation
Temperatures
t = T - 273
p =- 1 Atmosphere p = 1o Atmosphere
1000 ° 0.0006 0.001

1500 ° 0.0008 0017
Temperatur
2000 ° 0.04 0.08

2500 ° 0.19 0.33
saan
L. Babu, “ Traité de Metallurgie Générale," p . 220.
2500 eure
srs e
k e
e s ur
re
p r
moP es
he
Yio At Pr
sp
mo
At
-
2000 " -
-
e
_on
1500"
1000L 2 .3 .4
Dissociation coefficients
Fig. 11. - CURVE OF THE DISSOCIATION OF CARBON DIOXIDE, CO,.
1 Berthelot was the first to reject the importance of dissociation in combustion phe
nomena , He contested the probable value of the coefficients of dissociation when com
puted from the temperatures of combustion and the specific heats of gases between 0 °
and 100 ° C. He asserted thelaw of increase of the specific heat of gases, a law which
the work of Mallard and Le Chatelier made evident.
Dissociation of Water Vapor.—The quantitative values of the

factors of dissociation of water vapor have been determined by
Le Chatelier. They are very close to those given for the dissocia
tion of carbon dioxide.
It is, then , entirely permissible, both in our preceding calcula
tions and the theory of furnaces which is to follow , to overlook
the dissociation factor of which, however, account might be taken ,
since the products of combustion are formed of carbon dioxide
and water vapor, and the industrial temperatures are lower than
2000 ° C.
Dissociation of Carbon Monoxide . - The dissociation of this gas
represents remarkable peculiarities. Its decomposition into Carbon
and Carbon Dioxide, according to the reaction
C + CO, ,
2 CO = C
sounovodworong
is a maximum and almost complete at 450° C. , but decreases to
almost nothing at 1000 ° C.
Iso
900
850
800
750
700
650
600
550
500
450
.8 .9
V
20
arico
Dissociation Coerricients
Fig. 12.CURVE OF THE DissociATION OF CARBON MONOXIDE, CO
TABLE 9 .
DissOCIATION OF CARBON MONOXIDE .
Condensation
Coefficients of Dis
Temperatures sociation at
Atmos. Pressure
of CO = C , CO2 – C2
450 ° 0.02 2.98 1.00

500 ° 0.05 0.95 0.97
600 ° 0.23 0.77 1 0.87
700 ° 0.58 0.42 0.59
800 ° 0.90 0.10 0.182
900 ° 0.965 0.035 0.067
1000 °. 0.993 0.007 0.013
L. Babu, " Traité de Metallurgie Générale , " p. 227.
This phenomenon plays an important rôle in the running of gas

producers. It explains why it is impossible, especially at low tem
peratures, to obtain carbon monoxide free from carbon dioxide,
which is the rule in lean industrial gases. It explains, also, the
deposit of soot in the pipes and exits of producers. To it will be
due, in our general study of furnaces, the most important source
of variation between theory and practice.
1 This has a very important bearing on the practical operation of gas-producers. -
Translator.
CHAPTER II .
PROGRESS IN INDUSTRIAL HEATING CLASSIFICATION OF THE
DIFFERENT SYSTEMS OF Gas FURNACES — THEIR

ORIGIN AND DEVELOPMENT,
The number of gas furnaces in practical use today is so large

that it is necessary to classify them before we can take up their
general theory. We will take advantage of this to trace briefly
the progress realized from the origin of gaseous fuel, or rather
since the introduction of the method for the gasification of the fuel.
It is this method that has transformed the problem, by opening the
way to improvements not possible with direct fire furnaces.
Origin and Rapid Development of Gas Furnaces. - Gas furnaces,
which are most essential to modern industry, are of recent invention.
Their origin dates back to the patents granted Frederick and
William Siemens in 1856 and 1861. Doubtless, in the older in
dustries it is possible to find approaches to gas firing. In Wales,
gas-producers were used from the time of coal mining. The me
tallurgists of that country used to carry a bed of coal of 60 to 80
cm . on their grates, in order to lengthen the short anthracite
flame, for heating their rather large reverberatory furnaces.
Even the inversion, the basis of the Siemens regenerators, had
been tried in the first hot-air motors. The idea of pre-heating the
air necessary for combustion had been realized in some blast fur
naces and in Ponsard and Bicheroux furnaces.
But all these scattered ideas gave only insignificant results, until
Sir William Siemens ( who united an admirable perseverance to
his engineering genius) succeeded , after twenty years of study,
>
in giving the problem of furnaces a solution so perfect that thirty

years have added practically nothing to it, certainly no new
principles.
The original Siemens Furnace is too well known to require de
scription. Let it suffice to recall that, designed for the fusion of

steel on an open hearth , it included a reverberatory with a very
low arch . Its chambers were four in number, allowing the pre
heating of the gas and partial heating of the air ; the inversion was
effected by means of valves; the gas-producers, located at a certain
distance from the furnace, were connected with it by a sheet
iron flue shaped like an inverted siphon . The draught of the
producer was induced by the inequality in the length of the ver
tical legs.
This type of furnace was introduced first in metallurgical works;
its progress was impeded for a time , between 1867 and 1871 , by
ME
Fig . 13 - SIEMENS Gas -PRODUCER with Siphon .

( From D. K. Clark's " Fuel. " )
the appearance of the Bessemer process ; thereafter it grew very

rapidly. It was first tried successfully in 1867 in plate-glass
works at St. Gobain . However, it was not adopted by the plate
glass industry until 1876, after the tank furnace had given good
results.
Today the gas furnace is in general use in metallurgy , in the
glass, gas, and by -products coke industries ; also, though rarely,
in ceramics. It is the general tendency to introduce it in all in
dustries requiring high temperatures, consuming large quantities
of coal.
PROGRESS IN INDUSTRIAL HEATING . 39
Its success is partly due to the advantages inherent in gas firing,

easy regulation of combustion, which can be made perfect with
out excess of air, the possibility of a long flame and therefore
of large furnaces, and last, but not least, the use of fuel of infer
ior quality . Its success is largely due also to the original perfection
of the furnace, and the fact that regeneration permits high tem
perature as well as economy.
It is readily admitted, after reviewing the successive trans
formations which have made perfect the first solution and seeing all
the parts of the primitive furnace, one after the other, improved,
that one of the greatest merits of the gas furnace is, that it has
made a distinction between the various functions required for
industrial heating, and has applied to each of them an especial organ .
The well defined function of each facilitated their regulation .
This advantage of gas-fired furnaces has often been lost sight of.
In the older methods of heating everything was confused ; the
old glass-pot furnace gives a striking example. It was formed of
a hearth, containing the pots, placed above the fire-grate, without
a stack ; the draught was susceptible of regulation only by var
iation in the depth of the fuel on the grate. Under such con
ditions, firing was truly an art . Today, in a gas-fired furnace
the grate , the hearth, the chimney, and the new organ , the recuper
ator, are distinct. Each may be controlled in a methodical way.
Thus it is often possible, even in the absence of skilled workmen ,
to avoid accidents and to ascertain the cause of those that may take
place. The art of the skilled furnaceman , often so difficult to ob
tain, has given way to the science of the engineer, the acquisition
of which is always possible.
The distinction made in the various furnace organs has had
still further the advantage of opening wide the door to progress.
While in the old industry the lines of furnaces were transmitted
generally by tradition , without attempt at alteration, even in the
least particular, for fear of disaster, today inventors have tried
to improve each and all of the organs, and they have been able
to do so with the utmost satisfaction. Thus, without any single
great improvement, gas- fired furnaces have gradually become
perfected.
These improvements will be reviewed in the case of each organ .

Their enumeration is the only classification that can be undertaken
until a theory of combustion has been formulated which allows
the only rational classification — that based on the heat utilization .
DESCRIPTION AND FUNCTIONS OF THE ESSENTIAL PARTS OF

Gas FURNACES.
A gas furnace has in general the following parts :

( 1 ) The gas-producer, in which combustible gases are generated
by the incomplete combustion of the fuel, by means of a first in
troduction of air—the PRIMARY AIR . This organ may have
complete autonomy, its own draught apparatus ( Siemens siphon ,
etc.), or it may form a close part of the furnace and depend upon
its chimney.
(2) The inversion valves, which consist of butterfly or bell valves;
their function is to direct to one side or the other the products
of combustion and the gases required for the firing of the furnaces.
These valves do not always exist - for example, in furnaces where
the heat of the products of combustion is recuperated by means of
the counter parallel current system ; when they do exist (as in the
Siemens furnace) , they are the exact division line between the re
generative chambers and the chimney on the one hand, and the
recuperative chambers and the gas -producer on the other. This
exact limitation of the chambers and of the gas-producer, which
we give as a definition, appears the most rational, both from the
theoretical point of view and for the practical control of the fur
naces . In the study of a furnace , the gas-producer should not be
limited to the space where the fuel is gasified, but should be made
to include also all pipes and flues, whatever their length , leading
to the entrance of the chambers. The gas-producer is made re
sponsible up to the moment that it delivers the gas to the recuper
ative chambers. It is with this extended but well defined meaning
that the word " gas -producer ” is to be understood here .
(3) The chambers of recuperation, whose function it is to utilize
or to bring back, so to speak , to the furnace those calories which
otherwise would be carried away by the products of combustion .
PROGRESS IN INDUSTRIAL HEATING , 41
The chambers begin at the distributing valve, where the producer

ends, and terminate exactly at the entrance to the furnace, by the
ports where the combustion begins ; they include, therefore, the
ports of the gas and of the air.
(4) The laboratory or furnace proper or hearth , where the in
dustrial operation is effected , and where the combustion of the
fuel is completed by a new influx of air — the SECONDARY AIR.
The laboratory is limited by the ports of admission and of escape.
( 5) The chimney, or the draught apparatus taking its place,
for the purpose of discharging into the atmosphere the products
of combustion , and of insuring at the same time the circulation
of the various gas currents, either simply from the hearth to the
stack or onward from the gas -producer. The stack begins where
the chambers end , that is, at the distributing valves. The chimney
is completed , as far as the draught is concerned , by a certain num
ber of dampers, or their equivalent, at the various flues of ad
mission or exit .
These regulating devices, which are distributed throughout
the various parts of the furnace, are of the greatest importance,
and in general, the greater their number, the more easily will
the furnace be controlled .. In order to show their usefulness, we
will classify the various currents of gases which circulate in a com
plete gas furnace.
( a ) In the gas-producer a certain quantity of air is admitted
the primary air, which must be regulated by means of a dam
per or an injector.
(6) In its journey through the producer the primary air gas
ifies the fuel, transforming it into a mixture of CO , CO2,, CH 4 ,, H 2,
H ,O, etc.; this is the producer- gas.
(c) At the entrance to the hearth the gas meets a new afflux
of air, the secondary air.
( d ) The products of combustion are divided into two parts,
one going to the gas recuperators, the other to the air- chamber.
It is of prime importance to regulate this division, since it has a
most important bearing on the ruling temperature of the hearth .
(e) Finally, the two currents of the waste products meet at the
base of the stack , to be consigned to the atmosphere. It is also
Fia. 11. - TAYLOR Gas-PRODUCENT WITH MECHANICAL FEED .
useful to be able to regulate the draught of the stack, since the

pressure in the hearth is a function of the chimney draught.
The importance of the multiplicity of dampers is easily seen
from the foregoing. The first Siemens furnace had only two or
three, while the more recent furnaces have not less than four
two at the admission , and two at the exit .
MODIFICATIONS AND IMPROVEMENTS EFFECTED IN

Gas FURNACES.
( 1 ) Gas- Producers.-- The question of gas-producers has been

the most studied. The Siemens inclined grate has given place
almost everywhere to producers with horizontal grate, or even
without a grate at all . The inclined grate had the advantages
of distributing the coal along a talus parallel to the face of the
producer and of facilitating its equal distribution in a deep bed ,
of closing up at each feeding the blow-holes that the combustion
had produced , and of requiring less skilled labor. Such, at least,
was the idea advanced by its inventor.
The horizontal grate, on which is placed a bed of 80 to 120
cm . or more of fuel , requires the close attention of the workman ; he
must also distribute the coal evenly at each feeding, as well as break
up any blow -holes formed. It has, however, a much higher ef
ficiency than the inclined -grate type.
Let us here also note the trials of grateless producers. This type
is very interesting, since it allows the thorough pre-heating of the
primary air — the last desideratum of the furnace problem . The
St. Chamond steel works operate gas-producers without any grate
in which the coal places itself along its angle of repose ; but they
do not seem to be preferred to producers provided with grates.
The best form of the grateless producer is the Taylor, with
its rotating hearth . Here, however, the injected steam and air
cannot be pre- heated , or have not been pre -heated , above 400°.
Lastly may be mentioned the Ebelmen fusion producer, in
which the ashes are fluxed and fused . This idea, though very
attractive, has been practically applied to no large extent.
It is chiefly in perfecting methods of gasification that inventors
have tried to obtain a gas richer than the Siemens gas. It may be
said in general that the Siemens gas carries 5% or more of carbon
dioxide, a rather high proportion, which lowers greatly the calorific
power of the gas , due partly to the dissociation of carbon monoxide
Fig . 15. — MORGAN GAS-PRODUCER , WITH MECHANICAL FEED .
(see page 36 ). This appears difficult to overcome; nevertheless,

many inventors have attempted it.
The Siemens gas, generated by means of the atmospheric air,
carries a weight of nitrogen which is harmful in all combustion
reactions. Many have tried to eliminate this nitrogen by gas
ifying the fuel with a gas other than the air ; for instance, by steam
or even carbon dioxide, which reacts at a red heat on the carbon

according to the well -known reactions
CO , + C = 2 CO
H
H2O CO + H ,
+ C = CÓ
It will be seen further on that the industrial production of pure
1
water gas, in practical and simple apparatus , is not possible , '
S
Fig . 16. -EBELMEN GAS -PRODUCER.

(From A. H. Sexton's " Fuel.' ')
1 Great progress has been made in the last few years in the manufacture of water gas,
through the introduction of the Dellwick process and apparatus ( Fig .19 ). In the old
process, airisblownthrough the fuel bed, asintheusualproducer,untilthe fuel reaches
a bright redheat, the coal being oxidized to Co , and producer -gas formed. At this
point,
the air is shut off and steam injected. Water gas is generated at the expense of the sen
sible heat of the fuel . When the lowest temperature limit is reached the steam is shut
off and air admitted, the cycle of operations being repeated. In the first period , only 29.4
calories are liberated per moleculeofcarbon gasified to CO,andin the second 28.8 calories
are absorbed . It follows that the times of the two periods must practically be equal,
usually 10 minutes. Dellwick devised to oxidize the coal not to carbon monoxide, but
at once to carbon dioxide, the heat liberated, in this case, being 97.6 calories per molecule
of carbon. The calories are then as 18: 9 or about 3.5 to 1 . In practice , the air is injected
for 14 minutes and the steam for 8 to 10 minutes , a great difference in time. The ap
paratus is also much simplified , the gas on the air phase being allowed to escape to the
stack ; one gas -holder is then sufficient. The thermic efficiency is also much improved.
The combustion of the coal to CO , is obtained by means of the injection of air under high
pressure - 300 mm. or more of water. - Translator.
and that the so-called regeneration of the carbon dioxide of the

products of combustion , according to the ideas of Biedermann,
Harvey, and many others, is always unnecessary and often harm
ful. Nevertheless, furnaces have been constructed and are operated
on these principles , and they will be described further on .
TABLE 10 .
ANALYSIS OF VARIOUS INDUSTRIAL Gases.
„Bituminous
Bituminous
B0ituminous
ituminouంs
Anthracite
19 racite
Anthracite
Anthracite
పిండంల
Mond Gas
Recovery
Re ecovery
Mixed Gas
Without
Anth
NH3
With
Riché
B|లో
Gas
NH3
8
C:: oke
( Wood )
1. 2. 3. 4. 6. 7. 8. A
7 9. 10 . 11 . 12 . 13 .
H. 17.3118.31 20.75 20.0 10.0 18.73 14.0 14.0 10.0 45.0 12.2 27.5 24.3
CH 3.0 0.31 0.4 15.0 3.4 2.0 2.0
C2H4 . 0.5 0.31 1.0
CÔ .. 22.39 21.54 20.63 25.0 23.025.07 26.0 27.20 29.0 20.0 11.6 11.0 13.8
CO 2 .. 8.36 9.50 10.41 5.0 5.0 6.57 6.9 5.50 4.5 20.0 9.3 16.5 13.9
N2 51.94 50.65 48.22 49.5 58.0 46.98 52.8 53.30 56.0 62.5 43.0 46.0
N2 + H20
H20 H20 H20
02..... 14.04 3.37 5.02 0.5 0.5 ' 0.03 .. 0.5
1 , 2, 3 , 8 , 17, 18 , 21 , 30 . A. L. Queneau .
4, 5, 9, 12, 13, 19 , 22, 23 , 24, 25, 28. W. L. Case. J. S. C. I. , June 15 , 1905.
6. Dowson, The Engineering and MiningJournal, XXXIX , page418.
7, Fisher -Gautier, . Essai des Combustibles "; gas from a Klonne Gas -Producer blown
with air and steam .
10, 11 , C. Vigreux and L. Bardolle , " Le Gas Riché , ses Applications Industrielles ."
If the pure water gas is not practical, the mixed gas - a mixture of
Siemens and of water gas - is widely used,1 and has, in some systems >
of recuperation, some very real advantages.

The mixed gas is obtained either by blowing steam in the pro
ducer provided with a closed ash -pit, or by keeping the ash - pit
filled with water. In the latter case, the fall of the clinkers and
the radiated heat are sufficient to produce a large vaporization ,
giving as high as 12% of hydrogen in the producer-gas. Of all
systems of gasification, the one of the mixed gas is by far the most
used .
A recent improvement made in the construction of gas-pro
ducers is the bettering of the draught. When steam is injected
by means of a Koerting steam injector, this apparatus furnishes
at the same time the required air. Thus, the Siemens siphon
may be dispensed with . This doing away with the siphon takes
place even now in the case of producers provided with natural

draught. For this purpose it is sufficient to place the producer
at a lower level than the furnace, in such a way that its grate
will be at a certain distance below the level of the ports. The
rising column of hot gases acts as a chimney. This is an important
TABLE 10. ( Concluded .)
Hydrogen
ANALYSIS OF VARIOUS INDUSTRIAL ( Ases .
Carbureted
Oven
Coke
Gas
Gas
Water
Gas rt
Water
Methane
Dellwick
Reto
Gas
65
Cas
Blast Furnace
Gas
Siemens Gas Water Gas
as
Oil
Gas
14. | 15 . 16 . 17 . 18 . 19. 20 . 21 . 22 . 23 . 24 . 25 . 23. 27 . 23 . 29 . 30.

1
6.5 1.9 1 48.0 48.5 48.5 46.0 32.0 40.0
2 50.058.7 0.80 1.0 3.6 3.68
1.9 2.0 2.0 0.5 0.540.0 48.0 25.0 36.6 10.0 0.2 3.62
5.0 16.5 8.5 4.0 13 0.0 .
23.7 29.4 30.9'34.1 33.8 38.0 40.4 43.6 6.0 19.0 6.0 20.0 39.62 27.5 26.0 27.50
5.3 2.3 3.3 8 1.3 6.0 5.2 3.4 0.5 3.0 1.5 10.0 4.65 11.5 9.0 6.15
62.6 66.7 66.7 64.9 64.8 5.5
5.6 4.0 2.0 3.0 4.0 2.0 3.83 60.0 51.3 59.27
H20
0.5 ... 0 0.5 0.5 0.5 0.20 10.0
14 , 15 , Fisher-Gautier , loc. cit ., is from a Klonne Gas- Producer blown with air alone.
18, Shabel-Gautier, " Traité de Métallurgie Générale."
20, Leybold , Journal jür Gas Beleuchtungz, 1898 , page 1828.
26, JournalGas Lighting, page 90, 1905.
27 , Fisher -Gautier , loc. cit .
29, Lurman , Stahl und Eisen , page 245 , 1898 .
advantage , especially in the case of glass furnaces where, on ac

count of the constant opening of the working doors, the chimney
has no effect on the producer.
It is seen that we have to discuss furnaces fired with three different
gases — Siemens gas, mixed gas, and gas produced by the so - called
regeneration of the carbon dioxide of the products of combustion .
(2) Regenerators or Recuperators. — The least progress has been
made in the question of heat recuperation . Two systems are
still in practice — the Siemens recuperator with inversion, and
recuperation without inversion , by parallel counter currents.
In the first system , the heat of the products of combustion is
accumulated in a network of firebricks, through which the waste
products are made to circulate . These bricks will give back their ac
cumulated heat to the cold air and gas, which will flow through them
in a reverse direction at the succeeding inversion . In the second
system the cold air travels always through the same passages with
out interruption ; it receives by transmission through the walls of
the passages the heat of the waste products which are passing in
.
a parallel and opposite direction, surrounding at the same time
the air - ducts.
For furnaces in which the ruling temperature is not higher
Opening
than 1000 ° it may be possible to use cast-iron recuperators, and

mm
60
Brick
rack
Eng Firet
1,00
1.15 06 35 1.23
B.
1
P
3Concrete
-3.08 1.68
Fig . 17. New SIEMENS FURNACE, With HORSESHOE HEARTH AND INJFC
TION OF SOME WASTE PRODUCTS UNDER THE GAS-PRODUCER GRAIE .
Two RECUPERATIVE CHAMBERS. SECTION A , B , C, D.
( From F. Toldt, “ Regenerativ Gas-ofen . ” )
Over each chamber is placed , at I. a Koerting steam injector, by means of
which it is possible to inject under theash-pit either hot air (primary air) from
the recuperator, or hot waste products, or a mixture of both . A third steam
injector serves to inject cold primary air directly from the atmosphere
under the ash-pit .
for such cases the second system ( parallel counter current) is to be

preferred, on account of its simplicity and the lack of reversing
valve. For higher temperatures its use is not to be recommended.
The Siemens recuperative system has not been changed in
its fundamental principles, though in its many applications its
? This question is treated more fully later in the book .
proportions have been greatly modified ; thus, the gas chambers

have been reduced to one- half of their original volume, and the
latest tendency is to dispense with them altogether. From the
point of view afforded by theoretical study, the method of recu
peration is less important than the way in which it is applied .
TW
10
NAM
J
00
385
dva
Refractory
Material
Sena
Usua
NA
NIN
ry Material
LEN
Quality
RefractoHave
Il
LETTEN
Ru
SC
23
Fig . 170. - SECTION EF. FIG . 177 .-- SECTION G H.
Four different ways present themselves :

(a) Pre -heating of the secondary air alone.
(6) Pre -heating of the secondary air and of the gas.
(c) Pre-heating of the total air - primary and secondary.
( d) Pre- heating of the secondary air, primary air and gas.
(3) Furnace proper, or Hearth.—
The form of gas -fired reverber
atory furnaces has undergone two important modifications ,
first, as regards the roof, it being made very high (in some new
radiation furnaces there are 250 cm . between roof and the hearth );
and next, as to the disposition of the ports, in the so -called horse
shoe furnace. The radiation recommended by Frederick Siemens

has given excellent results in glass furnaces, where the temperature is
much lower than in steel furnaces, and where it is necessary to
keep the flame away from the surface of the melted glass. The
inventor, however, was mistaken in trying to make of his dis
covery a general theory to be applied indiscriminately to all furnaces,
without taking into account the material to be heated . Furnaces
with very high roofs have given but poor results in metallurgy.
-1.44
70
SIUVETTAVA
27
1.63 10 435
Fig . 17c. SECTION J K. Fig . 17e . - SECTION O P.
In the horseshoe furnace ( Fig. 17 ) the flame, instead of traversing

the hearth from end to end along its longer axis , enters on one side
parallel to its plane of symmetry, then doubles, escaping by a port
placed by the side of the port of entrance . The furnace thus gives
free approaches to three sides, often a great advantage.
This system has also been applied to gas furnaces. How
ever, the furnace appears to be more difficult to operate, as the
more distant parts of the furnace are exposed to undue cooling .
This style of furnace may meet with favor, bụt in general , for double
recuperation , we believe it to be better to use the older style of

hearth with the flames crossing from end to end.
Such are the main transformations that the Siemens furnaces
and gas firing have undergone. From the point of view of the
II
TUTUL
3.17
Di
D
d 354 1.85 70 65
IT
Fig 17d . - Section L M.
general theory, which is taken up in the succeeding chapters,

we must remember that the four methods of heat regeneration
may be applied to the following three different producer-gases:
US
25mm
WINAN
Fig. 17f.
(a ) Siemens gas, mixture of carbon monoxide and nitrogen.

(6) Mixed gas, mixture of carbon monoxide, hydrogen and
nitrogen .
(c) The so -called regenerated products of combustion , of about
the same chemical composition as the mixed gas. There would thus
1
be twelve cases, if, as will be seen later, one case was not impossible .
There will be eleven cases to be examined, and it will be seen
that all of them have received either full or partial application.
To illustrate the modifications of the original Siemens furnace
we may give as a type the New Siemens furnace, which unites a
majority of the improvements. This furnace is fired with mixed
gas, the producer being blown with steam ; it has double regen
eration — that of the total air, primary and secondary , with a
single pair of recuperative chambers, without siphon ; the producer
is built right against the furnace, and the hearth is horseshoe in
shape.
We do not wish to say that this furnace is the best ; still less do
we desire to defend the so -called regeneration of the carbon dioxide
contained in the combustion products, because theory will give us
its exact value; but from the above enumeration it is easy to see
that this new furnace has many of the aforementioned modifications
which , applied individually, have given good results.
1 The case of the triple recuperation with the Siemens gas.
CHAPTER III.
PURPOSE OF TIIE THEORY OF RECUPERATION'FUNDAMENTAL

EQUATIONS OF HEAT UTILIZATION - DATA , HYPOTHESES,
AND GENERAL PRINCIPLES .
Influence of the Temperature on the Heat Utilization in a Fur

nace.The study of the progress made in industrial heating, which
was considered in the preceding chapter, led us to a classifica ion
of eleven systems of gas-fired furnaces-furnaces with single,
double (itself subdivided into two classes ), and triple recuper
ation . The furnaces may be fired with Siemens gas, mixed gas,,
and gas resulting from blowing the producer with air, steam and
products of combustion . Direct firing should also be included ;
though rarely used for high temperatures, it is common for mod
erate ones, but is used almost wholly for low temperatures, as, for
instance, boilers. This forms a twelfth case to be treated .
This classification may appear quite complex when applied
to forms of apparatus which , with the exception of direct - fired
furnaces, are called indiscriminately gas furnaces, but which are
often in reality Siemens furnaces. This is, however, almost the
exact state of industrial conditions, from which will be taken
some ten cases as illustrations of the different classes .
The multiplicity of expedients, tried and remaining in use, for
solving a problem apparently capable of a single solution is
partly due to the present difficulty in establishing thc relative
value of the various systems. It is also due to the high cost of
construction of gas furnaces and to the difficulties attending their
successful introduction . Hence, the hesitation exhibited by
managers of works in substituting a perfected and more eco
nomical furnace in place of an older type in good working condition .
1 Throughout this treatise we will take the word “ recuperation " in its general sense,
without discriminating between Siemens and parallel counter -current recuperation,
unless otherwise specified. - Translator.
It is not uncommon to see in works several types of furnaces in

operation.
The problem is complex and permits of several solutions ac
cording to the nature of the work to be accomplished, the length
of the operation, the temperature, the condition of the draught, the
nature of the fuel, its cost, etc. The general theory, which we
will present, may be used as a guide, though its requisites may be
overruled by economical considerations. All disturbing factors
1
cannot be touched upon here. There is one, however, more
mport than any other, which cannot be omitted in a theory
of heat recuperation ; it is the ruling temperature of the hearth .
The ruling temperature is that temperature best suited to the
operation to be performed in the hearth . In a general theory this
temperature is arbitrary, as the same type of furnace may be used
in various industries and at widely different temperatures. In
actual practice , however, this ruling temperature is selected to give
the best results . It should not vary in continuous furnaces, and
if it is subjected to fluctuations in intermittent work (as puddling
furnaces, zinc furnaces, etc. ) , it should be regulated entirely with
regard to the work to be done. It is in any case a condition to be
established first in the experimental study of a furnace.
The ruling temperature has a considerable influence on heat
utilization . It is evident that the waste products escaping from
a furnace at, say , 1500º carry away, for a given weight of consumed
fuel , more calories than when leaving at 300 °, as in the case of
boilers. Recuperation applied to these two cases will doubtless
recover the calories, but only in part, allowing the remainder to
escape to the stack . The absolute value of the latter will be the
greater the higher the original number of calories in the waste
gases , that is , the higher their original temperature. It follows
that the lost heat and, by difference, the utilized heat, are func
tions of the temperature. It is also clear without any calculations
that the law of efficiency, for furnaces operated under similar con
ditions, is a continuous function of the temperature, so that it
would be possible to plot for each of the twelve furnaces its curve
1 In order to make more interesting a theory rendered rather long by the complexity
of the industrial cases, we will endeavor, when the discussion leads to it , to derive prac
tical conclusions from figures, and thereby to elucidate a number of these particular cases
PURPOSE OF THE THEORY OF RECUPERATION . 55
of heat utilization, the examination of which , at any temperature ,

would permit us to judge the relative value of each of them at a
given temperature .
In order to avoid undue complexity, we have deemed it ad
visable to select two fixed temperatures - 1000 ° and 1500 °. The
first corresponds to the temperatures of coke ovens, which may
be taken as the type of industry using medium temperatures; the
second is lower than the ruling temperatures of steel furnaces,
while it is higher than that of the glass, ceramic and zinc indus
tries. It will be easy by interpolation or extrapolation to find
the efficiency at any other temperature. Thus limited, our general
theory has a threefold object :
( 1 ) To study the utilization of heat in the twelve systems in
order to classify them at the two temperatures of 10000 and 1500 ° .
(2) To show the influence of the temperature on the efficiency,
by comparing the heat utilization of the different systems, at
the arbitrarily selected temperatures.
(3) To indicate the method of calculation which permits of
ascertaining, for any given temperature, the comparative efficiency
of the various furnaces, and to make the best selection .
Definitions: Heat Utilized , Heat Lost, Efficiency, Equation of
the Utilization of Heat. - We must still define closely the meaning
of " efficiency," " heat lost ," and " heat utilized , " and justify these
definitions just as we were obliged to in the case of the various parts
of the furnaces, and to justify the precise limits given them .
We understand, by “ heat utilized ” in a furnace, the whole
of the heat consumed in the hearth, either in the performance
of the industrial work, by exterior radiation (often necessary to
preserve the brickwork ), or heat completely lost by conductivity
or any other cause. So defined, the utilized heat is exactly the
difference between the total heat (sensible or latent) brought into
the hearth by the secondary air and gas and the total heat carried
away by the waste products.
By “ heat lost " we understand all heat spent outside the hearth ;
that is , liberated into the atmosphere by the waste products,
radiated or conducted by all organs other than the hearth (by
the producer, the chambers, the flues, etc. ) , or consumed in the
vaporization of the water in the ash-pit , or used in producing the

draught, etc.
Under these conditions it is evident that the sum of the lost
heat and of the utilized heat represents the total heat absorbed
in the furnace; that is, the total heat liberated by the fuel , or, in
other words, its calorific power.
Our definition is then expressed in the equation :
Q = P+ U
in which
Q The calorific power of the fuel.

U utilized heat ; that is , heat consumed in the hearth .
P Lost heat; that is, heat carried away to the stack or
lost by all parts other than the hearth .
Finally , by recuperative efficiency we understand the ratio ý
of the heat spent in the hearth to the calorific power of the fuel.
These definitions give to heat utilized and heat lost a more
comprehensive meaning than their usual one. Generally the
only lost heat considered is that carried away by the waste pro
ducts. By utilized heat is understood the heat necessary to the
chemical reaction or to the physical operation to be effected --
melting of the glass, or reheating of the ingots.
Our definitions are justified by the purpose of our study, which
is recuperation . The object of recuperation being to collect all
the heat not made use of in the hearth , the recuperative organs
are responsible for the heat which is turned over to them , and they
should restore it as completely as possible. If the restitution is
done imperfectly, it does not matter whether the losses are to
be charged to the liberation of the waste products into the at
mosphere, or to the chambers of recuperation , or even to the radi
ation of the producer flues , for all these organs are agents of re
cuperation . The total loss alone is of interest in determining
the efficiency of a furnace.
1 This remark has a practical interest; it often happens that the system of recuperation
is estimated by the temperature of the stack. This datum is of minor importance, how
ever, for the waste products might have lost a large quantity of heat by radiation , or by
becoming mixed with cold air, without any advantage resulting to the hearth .
That which Ledebur calls " the useful effect , " and which is the
heat really utilized, is not, in a theory of industrial heating, of
as great an interest as it might appear. In most cases, the masses
to be heated ( crucibles, retorts, muffles, furnace walls, etc.) have
a specific heat much higher than the specific heats and the heats
of reaction of the bodies under treatment. Often the heat of
reaction is but a small fraction of the heat consumed in the hearth .
It happens even, as an extreme case , that an exothermic reaction
can take place only at a high temperature, so that in reality the
heat utilized is nil,, however perfect the furnace. In such a case,
the calculation of the absolute efficiency or useful effect would
be without any interest.
Furthermore, the recuperation we are studying does not take
into account the manner in which the calories are spent and dis
tributed in the hearth ; it concerns itself only with the calories
escaping from the hearth, and which it should bring back
there.
To justify further our definitions, let us notice that, in the com
parative study of the different types of furnaces, it is our perfect
right to suppose that the hearth conditions are the same for all .
The ruling temperature being invariable, it follows that the ratio
between the utilized heat and the useful effect is invariable , and
that, knowing one of the terms, the other may be ascertained .
The calculation of the recuperative efficiency 7 will give, therefore ,
the measure of the relative value of the furnace, considered from
the point of view of the utilization of heat as well as of the useful
effect .
Evidently all the preceding definitions apply as well to experi
mental tests of furnaces as to the theoretical determinations. In
the case of experimental tests, the idea of the utilized heat is of
great importance, since it can be reckoned by means of temper
atures and of gas analyses ; while, in general, the heat used for
the industrial operation — the useful effect-can be computed
only by the help of the specific heats and of the heats of reaction of
the bodies present .
FUNDAMENTAL PRINCIPLES OF THE UTILIZATION OF HEAT IN
FURNACES. - The study of the utilization of heat in furnaces is
controlled by a few fundamental principles so evident that it would

be superfluous to recall them if some of their most immediate
deductions had not been either misunderstood or forgotten. It
is only by keeping in mind these self - evident truths, comparable
to the principles of thermo-dynamics, that the errors of calcula
tions or of reasoning, so easily made, will be avoided in the question
of the exchange of calories, sensible or latent, and of recuperation
in which the available heat is several times transformed .
( A ) Principles Relating to Combustion and to the Total Available
Heat .-( 1 ) In a gas furnace the source of heat is the fuel burned
on the grate. All the reactions of combustion, whatever modi
fications are wrought by recuperation in order to retain the cal
ories, or by the steam , carbon dioxide, etc. , in order to enrich the
gas, are always reduced to this single resultant: A liberation of
calories equal to the number of calories set free by the fuel com
bining with cold oxygen , borrowed from the atmosphere."
This heat of combustion , which is easily ascertained by con
sulting tables of the calorific power of chemical compounds or
elements, or by experiment in the case of coals, etc. , is then the
invariable term of comparison for all exchanges of calories, for
all heats - lost, recuperated or utilized in the furnace .
(2) Burned or oxidized gases, as water or carbon dioxide , can
never furnish any calories. If by any reactions they are broken
up in a portion of the furnace, in order to obtain either carbon
monoxide, hydrogen or oxygen , this reaction absorbs heat and
cools the gas and the furnace at that point. The later combi
nation of the gases set free has only the effect to restore to another
part of the furnace an equal quantity of heat, without the total heat
liberated by the fuel and made available being in any way modi
fied . In this case there is transport of heat from one point to
another, without any gain in calories.
(3) The only gas supporting combustion in a furnace is the
atmospheric air. If steam or carbon dioxide used in a furnace
1 This is understood evidently of a self-contained gas -producer. If the necessary steam
is vaporized in an independent boiler , or if the air necessary to the combustion is pre
heated in an independent furnace, there will be a further contribution of calories. If,
on the contrary, the water is introduced as liquid in the ash-pit of the producer, and is
liberated with the waste products in the atmosphere as steam , there will be from this fact
an additionalexpenditure of calories for the vaporization, to be subtracted from the avail
able heat.
BIFR ARE
OF THE
UNIVERSITY
ΟΣ
CA! IFORNIA
is decomposed in one portion of the furnace, liberating oxygen ,

they recombine in another part without having given up any oxygen
to the carbon or combustible gases from the fuel .
Axiom : The total quantity of air necessary for the fuel com
bustion , escaping through the stack as water vapor , carbon dioxide
and nitrogen , is always the same for a given weight of the burned fuel.
Two consequences of this principle have not been recognized
by several inventors. The first one is that in the products of
combustion the ratio of the volume of carbon dioxide to that of
nitrogen is always the same for a given fuel , whatever the method
of gasification , whether the producer is blown with air alone, air
and steam , or air, steam and carbon dioxide from the waste pro
ducts. People have erroneously sought to decrease this ratio
by blowing steam into the producer. This course is wrong. 1
The second fact not recognized is that the mass of the products
of combustion sent through the stack cannot be decreased for a
given weight of the fuel .
In the New Siemens furnace, in sending to the gas-producer part
of the waste products, the idea may arise that some of the waste
products with their calories have been diverted from the stack .
This reasoning, which may give a wrong impression of the value of
the New Siemens furnace, is also inaccurate . ?
1 The theoretical water gas CO + H 2 does not carry, it is true, any nitrogen as the
2
Siemens gas , CO + 2N .2, but the combustion of the two gases gives CO2 + 4 N2 + H20
and in the other CO2 + 4 N2 , and the waste products have in both cases the same ratio
of CO2 to N 2,2 therefore the quantity of air going through the furnace is the same in both
cases .
2 We believe it useful to refute this error , in regard to the New Siemens furnace, by
explaining the circulation of the various gas currents in that furnace. Upon leaving
the hearth the waste products are effectively divided into two parts; let us suppose that
the two parts are equal, which case is the most favorable to the system . This allows
the gasification of the fuel without any primary air being required ( see page 116) . A mass
of products of combustion expressed by the formula CO , + 4 N 2 goes to the stack , while
an equal mass goes to the gas producer. Upon contact with the incandescent carbon
the following reaction takes place:
CO2 + 4N , + C - 2 CO + 4 N 2 .
In order to burn this gas , which has the same amount of nitrogen as the ordinary Siemens
gas, a mass of air 0 , + 4N2 is required. After combustion we have
2 CO + 4 N2 +0 , + 4 N2 – 2 CO , + 8 N 2,
or a quantity exactly double that sent to the stack or to the producer, and also double
that resulting from the combustion of the fuel with air . The mass of the products of
combustion passing through the furnace is then double that of an ordinary Siemens furnace.
It is possible to see that there are two gas currents - one acting as in the ordinary furnace,
going from the producer to the stack ; the other traveling in a closed circuit, from the pro
ducer to the hearth and from the hearth to the producer,without any modification in the
heat -balance sheet , once the working is established (and without any very evident ad
vantage ).
( B) Steam and other Auxiliaries in Gas- Producers. Since the use

of steam does not modify the total available heat, nor the total
necessary air for combustion , nor the proportion of nitrogen cir
33000
Fig.18. -WATER -SEALED Taylor Gas - PRODUCER, WITH BID AUTOMATIC FEED .
PURPOSE OF THE THEORY OF RECUPERATION , 61 ,
culating through the hearth , the function fulfilled by steam , the

usefulness of which is well established and approved by general
practice, should be given with precision. It is closely related to
recuperation .
In a producer blown with steam , the reactions giving rise to
gas are the following:
C + 0 = CO + 29.4 Calories
C+H , 0 = CO + H 2, + 29.4 – 58.2 Calories = CO + H2-28.8 Cal.
H2O
-
А A
B
10
89
HIE
Fig . 19. — DELLWICK -FLEISHER WATER-GAS PRODUCER .
The first equation is exothermic, and tends to bring the producer

to 1300 ° C.; the second is endothermic, and would tend to ex
tinguish the producer if the steam were not regulated, so that
the temperature is not brought below 600 °, the lowest temperature
limit of the ignition of carbon . The first effect of steam is the

lowering of the temperature of the gas in the producer. To this
lowered sensible heat there corresponds an increased latent heat
in the gas.
These results are of importance, but the resulting advantages
vary greatly, according to the different cases . When the gas is
burned in the hearth as it leaves the producer, the increase in
10
-11-6
Water Seal Line Nators
Fig . 20. - SWINDELL GAS - PRODUCER with DOUBLE FACE . VERTICAL SEC
TIONS ON LINES AT Right ANGLES with Each OTHER.
( From Ingalls' " Metallurgy of Zinc,” p. 316. )
latent calories is purely fictitious, since in the hearth the latent

heat and the sensible heat are equivalent .
On the contrary, when the gas is burned only after having suffered
a change in its temperature, either becoming cool as in furnaces
using cold gas, or becoming heated as is the case when the gas is
recuperated , the increase in the latent heat is then very useful.
In furnaces provided with double recuperation the two effects
supplement each other. The lowering of the temperature allows

better recuperation, and owing to the use made of calories, other
wise lost, the injection of steam gives a very real gain of calories,
and therefore increases the efficiency.
A final result of the use of steam is the modification it brings
in the ratio of the primary air to the secondary air. It is well
known that, in gas- fired furnaces, the combustion of the fuel takes
place in two steps, partly in the producer by the injection of air
mixed with steam , and partly in the hearth by the admittance of
the secondary air. In most furnaces the primary air is not re
cuperated, as the question of the preservation of the grate when
using hot blast is yet unsolved ; the secondary air, however, is
always recuperated .
The decrease in the amount of the primary air has as a result
a corresponding increase of the secondary air, and thereby allows
a further saving of calories and in most instances affords a greater
efficiency.
It will be seen , then , that steam acts as an auxiliary of recuper
ation . Doubtless in some instances it will react on the tempera
ture of the hearth, but only in the same way as the recuperation
itself, tending to raise the temperature of combustion. Its power
to increase the latent calories in the gas is of importance only
when the gas is used cold ; this case, however, does not enter into
our study. 1
Other gases than steam can be used as auxiliaries of the fuel
gasification . Thus carbon dioxide acts precisely in the same
manner :
C + CO , = 2 CO + 29.4 Cal . 68.2 2 CO 38.8 Calories .
The breaking up of carbon dioxide in the producer absorbs heat

and lowers the temperature of the gas ; its use decreases the amount.
1 When producer -gas or lean gas is used for explosion gas engines , the increase of the
latent heat is the most beneficial:
with steam are much used .
for such
Remark purpose the Buire -Lencauchez
: A furtheradvantage producers
ofthe use of blown
steam ,foreign
to recuperation , is the help it gives in the operation of gas-producers, whenever the fuel
used has a fusible ash . By lowering the temperature of the producer it not only prevents,
to a large extent, the formation of clinkers, but also disintegrates the clinkers formed .
The decrease in the amount of clinkers will result in a cleaner ash , the carbon contents
being lower. - Translator.
IOB
Ppo
Blast
Oo
Holes
Poke
a
a
13 D
O
Door
Hopper
-154
100
191 IT
106
86
CDI
Fig . 21. -DUFF GAS-PRODUCER .

Cuts, from top, left : Section A B. Section CD . Elevation . Section E F.
Top view of grate and brick bearers. Plan .
(From Ingalls' " Metallurgy of Zinc,” p. 306.)
of the primary air and increases that of the secondary . If the

carbon dioxide is borrowed from the waste products, it has been
seen that the volume of the gases traveling through the hearth is
doubled . If the carbon dioxide is borrowed from other sources ,
as inventors have proposed doing, it would act exactly as the
steam , with the additional advantage of not possessing any latent
heat of vaporization. It would be more efficient, theoretically
speaking, if it could be obtained in the gaseous state without any
cost.
Among other bodies acting in the gas -producer in the same man
ner as carbon dioxide and steam may be cited heavy oils, tars,
asphaltum , and even the hydrocarbons of the coal. We find in
this the partial explanation of the better efficiency of some coals
when used in the producers. When these bodies are thrown into
the producer, they are volatilized and highly heated at the ex
penditure of the sensible heat of the producer. Therefore, they
cool the gas and prepare it for a good recuperation ; and, further,
their gasification does not acquire any primary air.
The use of tar in the producers of some steel works, in order to
increase the efficiency of the recuperation and thereby the tem
perature of the hearth, is strictly in line with the above deduction . 1
It indicates also the advantage which would follow the use of
petroleum , hydrocarbons, natural gas, ?2 etc., in the Siemens
furnace .
( C) Fundamental Principles of Recuperation . The utilization of
the calories carried away by the waste products as they leave the
hearth at a high temperature necessitates the use of solid bodies,
in order to avoid the mixing of the reacting gases. These solid
bodies are either networks of refractory bricks, or ducts of metal
or refractory materials. In the first case they accumulate the cal
ories for a certain period of time, and during the succeeding period
transfer them to the gases to be recuperated ; in the second case
they are permeable to heat to a greater or lesser degree, allowing
1 Another contributing cause in this case is that the heavy hydrocarbons generated
by the use of tar in the producer burn with a very luminous flame. It is well known that
the luminosity of a flame has a very useful effect in the furnace hearth , by increasing the
radiation . - Tranelator,
2 Natural gas cannot be efficiently pre-heated on account of the heavy deposit of soot
in the brick network , resulting from the breaking up of the hydrocarbons. — Translator.
the transfer of the calories from the hot gases to the cold ones (air
or producer- gas). In both cases they act only as heat- carriers;
therefore it is permissible, in a theoretical study, to consider only
the reacting gas masses, one mass losing its calories, the other
gaining them . Under such conditions, two self-evident laws
govern this exchange. Both are derived from the fact that an
exchange of calories can take place only from a hot body to a cooler
one .
( 1) In order that the total recuperation of the calories held in

the products of combustion be possible, it is necessary that the
gases that are to absorb the calories be at the temperature of the
outside air into which the products of combustion are discharged .
This condition is fulfilled only by the air, as the producer-gases
are always at a higher temperature than the atmosphere. It fol
lows that, in a furnace with double recuperation, secondary air
and gas, the total recuperation is always impossible, because a
part of the waste products cannot give up all the calories
it holds.
(2) In order that the exchange of calories between two gaseous
masses be perfect (that is, that after recuperation the final tem
perature of one mass be equal to the initial temperature of the
other), it is necessary that the total thermal capacities of the
masses present be equal between the temperatures under consider
ation . If the specific heats are unequal-- or, what is more fre
quently the case , if the masses are unequal — the hot gas which is
to give up its calories will escape at a temperature different from
that of the gas to be pre -heated .
When the mass of the first gas is too large the recuperation
will be incomplete, and the waste gases will escape at a temper
ature higher than the one of the atmosphere. Inversely, when
the gas to be pre-heated is in excess, these gases will absorb all
the calories, but their temperature will not reach the initial tem
perature of the waste products.
The first case - excess of waste products -- is the one generally
met with in furnaces. The thermal capacity of the air necessary
for combustion is always lower than the thermal capacity of the
products of combustion ; therefore, even heating the total air
would not secure complete recuperation . The double recuperation

by the air and the gas does not secure complete recuperation ,
since the thermal capacity of carbon monoxide is lower than that
of carbon dioxide.
Triple recuperation - primary air, secondary air and gas - alone
may give complete recovery of the calories of the products of
combustion . On the point of masses the limit is reached and even
exceeded , since the primary air is pre-heated twice; it passes once
>
through the chambers before its injection into the producer, and it
is subjected to a second recuperation with the producer- gas on
is way to the hearth. In the second case, on account of the in
equality of the masses , the air and the gas will leave the recuper
ators less hot than the products of combustion at their entrance
through the same, however perfect the apparatus.
HYPOTHESES REQUIRED FOR THE MATHEMATICAL THEORY OF

COMBUSTION IN THE RECUPERATIVE GAS -FIRED FURNACES;
SELECTION OF THE METHOD USED FOR THE
CALCULATION OF THE EFFICIENCY.
In order to eliminate some variables from our general equation

of heat utilization and to reach the general theory of recuperation,
it is necessary to assume certain hypotheses.
( 1 ) We will admit in our calculations that the complete gas
ification of carbon into carbon monoxide, without any admixture
of carbon dioxide, is possible.
This hypothesis does not conform with practice, since producer
gas always carries a certain percentage of carbon dioxide. Lede
bur, in his treatise on metallurgy ', gives examples of gas analyses
where the Co , is below 1 % , which would justify our hypothesis;
however, since such results are very rare, it will be necessary for
us to discuss the influence of the premature combustion of the
gas on the heat utilization .
(2) We will suppose that the exchange of calories is perfect
between the two gas masses, the products of combustion giving
up their calories, and the gas to be pre-heated absorbing them .
1 " Genie Civil," Tome XXIX , No. 4, p. 59.
In other words, we will suppose that the temperature of the re

cuperated gas will be equal to the temperature of the products of
combustion at the moment of their leaving the hearth .
This hypothesis is quite far from reality, since the rate of ex
change of calories between a gas and a solid body in contact with
it is greater the larger the gap between the irrespective tempera
tures; as the gap becomes small the rate of exchange becomes
very small. Since recuperation necessitates a double transfer
of calories between a solid and a gas , it is found in practice that
the exchange from gas to gas is limited to a difference in tem
perature generally greater than 300°. Here, also, we will be obliged
to discuss the influence of this disturbing factor, since we can
not include it in our calculations.
(3) We will admit that the furnace has an invariable ruling
temperature in all its parts.
This condition is fulfilled in continuous furnaces using the par
allel counter- current system of recuperation , but it is not exactly
true of furnaces using inversion . Let us assume, then, that the
chambers are large enough and the inversion frequent enough
to maintain the constancy of temperature between two inversions.
Our hypothesis will, therefore, conform closely to practical condi
tions. Hence it will not be necessary for us to discriminate between
the two systems of recuperation ; the results found will apply
as well to the Siemens system as to the parallel counter - current
system .
(4) The heat radiation from all organs of recuperation, chambers,
flues, and gas producers is supposed to be small enough to be
negligible.
This hypothesis is evidently inaccurate ; radiation is essentially
1
a function of the form and of the method of construction of the
furnaces . It would be impossible to enter such variables in our
calculations. The radiation of the hearth, however, is part of
what we have defined as “ utilized heat. "
(5) We will suppose that our fuel is carbon, the equivalent
of an ashless coke .
It results from all the above hypotheses that our gas will be
formed of carbon monoxide, or carbon monoxide and hydrogen ,
without admixture of carbon dioxide, water vapor or hydrocarbons;

that the gas brings to the furnace the whole of its sensible heat ;
that the gases subjected to recuperation enter the hearth at its
ruling temperature; and, finally, that the products of combustion
enter the recuperators at that same temperature, arbitrarily se
lected . In other words, if the ruling temperature of a furnace
with double recuperation is, say, 1500 °, the secondary air and the
gas enter the hearth at a temperature of 1500 °. If the furnace
has single recuperation , the air enters the hearth at 1500 °, the
gas enters at the temperature of combustion of the carbon burned
to carbon monoxide in air, mixed or not mixed with steam, and
the products of combustion escape from the hearth at 1500 ° and
enter the chambers at the same temperature.
We may remark that these hypotheses favor the perfect utili
zation of the heat in the furnace. They mean simply that, from
the point of view of the utilization of heat, the organs of recupera
tion have reached perfection . It is toward this perfection that
all improvements tend. Our hypotheses are therefore legitimate.
The theoretical calculations made by their use will give us, in an
accurate manner , only the maximum heat utilization or the min
imum losses in the furnaces. The results thus obtained are not only
interesting but they may also serve as a sure guide in an experi
mental study of a given furnace, as they allow appreciation of its
good and bad points.
However, from the point of view we have taken, the compari
son and classification of the various systems of furnaces are not
sufficient. Nothing warrants us to say , a priori, that the classi
fication of the furnaces, based on their maximum efficiency, will
still hold true when the furnaces are working under the conditions
met in practice.
We will then have to review our various hypotheses in order to
see what modifications in the absolute values of our figures will
bring the practical conditions, and thus ascertain the change in
the order of the classification . We will make use , therefore, of
our hypotheses, with the reservation of a subsequent examination .
Calculation of the Efficiency of a Furnace.-- In our general equation
Q = P + U
we will give to the terms P and U the specific meanings given them
in our hypotheses:
P = heat escaping recuperation, or heat escaping, with the
waste products.
U Maximum of the heat utilized in the furnace .
And since Q is always known either by the direct measurement

of the calorific power, or indirectly by means of the heats of
combustion of chemical elements or compounds, it is possible to
compute U or P and to obtain by difference P or U.
Of the two methods we have selected the direct calculation
of P, with the idea that the knowledge of the heat lost is the essen
tial point in a study of furnaces. Furthermore, the computation
of P allows us to easily find the temperatures of the gases leaving
the chamber, and to obtain an insight of what takes place in the
recuperative apparatus.
The direct calculation of U has its special interest in the case
of the study of the phenomena taking place in the hearth . Its cal
culation will be the object of a later study.
CHAPTER IV .
DIRECT - FIRED FURNACES-BOILERS.
Characteristics of Direct-Fired Furnaces, and their Principal

Industrial Applications .-- The combustion of the fuel in direct
fired furnaces takes place in one operation ; the carbon is at once
burned to carbon dioxide, the reaction being the following:
C + 0 2, + 4N 2, = CO2 + 4N ,
1 4 1 4 ( Molecular Volumes)
97.6 calories are liberated by the burning of one molecule
( 12 grams) of burned carbon .
In this case no attention need be paid to the steam sometimes
injected under the grate, as is the case with boilers operated under
forced draught with a steam injector. Sometimes, also, in furnaces
operated for high temperatures, the workmen are accustomed
either to wet their coal or to throw water on the grate, with the
idea of bettering their fires. Whatever be the rationale of this
practice, water is not broken into its constituents in a direct- fired
furnace, and therefore does not take an active part in the com
bustion .
In order to better utilize the heat, the body to be heated is placed
directly over the flame, and often in contact with it. All the
waste products go to the stack, carrying away all the available
calories not made use of in the furnace proper.
In order to have perfect combustion, it is necessary to have
an excess of air. This excess is often considerable, as the air gen
erally has free access to the grate, and can be regulated only as
far as the fire can be regulated ; nevertheless, it is seldom that
the combustion is complete, as is indicated by black smoke, show
ing that abundant coal-dust and soot have been subtracted
from the combustion .
With the above conditions - loss through smoke, excess of air

and incomplete combustion -- a low efficiency results. In spite
of this, direct-fired furnaces are largely in use on account of their
simplicity and the economy of their construction .
Section on Line A-B
E
Air
F
Air
Air
Fig . 22. - BOËTIUS FURNACE.

( From A , H. Sexton's " Fuels and Refractory Materials." )
The old glass-pot furnaces and most reverberatory furnaces for

metals other than iron , kiln furnaces for porcelain and earthen
wear, muffle furnaces, annealing furnaces for glass and brass ,
DIRECT- FIRED FURNACES . 73
etc. , are without recuperation. Certain furnaces with a deep

firebed constituting a true producer, and where high temperatures
are obtained by pre-heating the air, as the Bicheroux furnaces
used in metallurgy, and the Boëtius furnaces in glassworks, are still
furnaces with direct firing without recuperation. Their products
of combustion are discharged directly into the atmosphere.
Puddling and reheating furnaces are also direct - fired furnaces;
in most instances, however, they are connected with steam boilers
-excellent organs of recuperation . They are worthy of a special
study. Boilers form by far the largest class of this group.
It should be said , however, that if all these furnaces are not
provided with gas recuperators, they are not wholly without means
for making use of some of the heat escaping the main hearth .
Without mentioning furnaces followed by boilers, many reverber
atory furnaces have several hearths of decreasing temperatures.
The ceramic furnaces have a hearth where the first firing given the
potteries is at a temperature lower than that of the main hearth ,
where the enameling is done.
Through these devices the efficiency is often greatly improved .
From this remark it follows that the ruling temperature of the
first hearth is without any great importance to the heat utilization.
The two items which , in direct- fired furnaces, allow the estimation
of the value of a furnace, are the temperature and the chemical
composition of the products of combustion at the stack . " From
this data the heat lost can be computed, and by difference the util
ized heat is found.
In our study of direct- fired furnaces we will calculate the heat
losses for all temperatures and for the composition of waste pro
ducts usually met in practice, assuming the fuel to be ashless
coke .
Heat lost, P, for all Temperatures, the Combustion Taking Place
)
with the Theoretical Amount of Air. - Let us take a weight of pure

coke (carbon ), say 12 grams, giving by its combustion a molecular
volume of carbon dioxide, and liberating 97.6 calories. The
necessary oxygen is accompanied by a volume of nitrogen 79.2
20.8
1 Let us recall that in non -recuperative furnaces the stack begins at the hearth exit,
and the temperature should be taken at that place.
which for the sake of simplicity we will call quadruple the volume
of oxygen . We have seen that the equation of combustion is
C +0 2, + 4N 2, = CO , + 4 N 2.
The products of combustion will be composed of 4 molecular vol
umes of nitrogen and 1 volume of carbon dioxide. The calories
carried away by these products are easily found by using the table
of thermal capacities (Tables 6, 7) . The calculations are shown
in the following table :
CALCULATION OF Lost HEAT, P, in DIRECT -FIRED FURNACES AT 300 °, 1000 °, 1500 ° C.
Reaction -C + O2 + 4 N2 - CO2 + 4 N 2
Composition of the Waste Products = CO , + 4 N 2
Calories set free Q 97.6 . 300 ° C. 1000 ° C. 1500 ° C.
Calories carried away by į by Molec. Vol. of CO 2 3.1 12.2 21.1

the Waste Products ... by 4 Mol . Vol . of N , 8.4 29.7 46.4
Total P 11.5 41.9 67.5
P + Q - Ratio of Heat Lost to total Available

Heat, P = 97.6 . 11.8 43.0 69.5
Percentage of Utilized Heat U 88.2 57.0 30.5
The utilization at 0° is evidently perfect, the efficiency being

100 % . This is practically the case in domestic furnaces, where
the apparatus is so built as to send the products of combustion
to the stack at a very low temperature.
The efficiency is nil, the heat being totally lost with the products
of combustion at the temperature of combustion of carbon in cold
air, which we know to be 2040° . We are able to plot the curve
of the heat lost, P, by connecting the five points found. The
equation of this parabola, deduced from the thermal capacities, is
the following :
T T 0 T2 - T .
Q ( P ) = 5 X 6.5 ° + [ 4 X .6 + 3.7 ]
1000 1000 ?2
The examination of this curve ( Fig. 23) gives useful information

on recuperation. We find the summit of the curve to be 2040 °.,
DIRECT - FIRED FURNACES . 75
the limit of temperature at which direct- fired furnaces can be used ;

beyond that temperature the use of grate furnaces is evidently
impossible, unless the air is first pre-heated by recuperation or by
some other device, such as is used in the Bicheroux furnaces.
Keeping in mind that an excess of 5% oxygen in the products
of combustion is difficult to prevent, and that it corresponds to an
excess of 25% of air, the temperature of combustion is lowered
to 1650 ° ( see page 17 ) . The study of the curve of heat utilization
DO
Ut
Lood
H il
is
% eat ed
75
Iso
at st
% He Lo
25 25
300 1000 500 20+ mature

Temperatures
Fig . 23. -Curves ShowinG THE PERCENTAGE OF THE HEAT UTILIZED AND
LOST IN DIRECT-FIRED FURNACES.
shows that, for a heat efficiency of 10% , there should be a difference

of about 150° between the temperature of the hearth and the
temperature of combustion ; it follows that 1500 ° is about the
maximum temperature to be realized under industrial conditions
when burning cold air, and then the efficiency is far from being
satisfactory. The only furnaces in use under such conditions are
blacksmith forges, where, with forced draught, heating is done by
bringing the iron pieces into close contact with the fuel . Here
radiation is reduced to a minimum , and the excess of fuel allows us
to always have perfect combustion in a portion of the firebed cor

responding to 2040 ° C. The temperature of 1500 ° explains the
necessity of recuperation for the metallurgy of steel. It may be
added that, without recuperation , the melting of steel in open
hearth furnaces would not be possible, since the ruling temperature
in these is 1600 ° and over.
In the glass industry, on the contrary, the ruling temperature
is only 1450º, or even less ; hence, direct- fired furnaces may still
be used , and there are quite a number in operation at present.
When recuperative-furnace builders guaranteed to the glass in
dustry a saving of fuel of 60%, that guaranty was not excessive ,
since the maximum heat utilization at 1450º is only 30% .
If the ruling temperature is lowered to 1200°, the efficiency
rises to 50% . At 1000 ° it becomes equal to 60% , and this rela
tively high efficiency explains why, in industries where use is
made of this temperature, it is so difficult to effect a substitution
of the recuperative furnaces for the direct- fired furnaces. The
manufacture of coal gas in retort furnaces, metallurgical operations
with the Bicheroux furnaces, glass-making with the Boëtius fur
naces, and kiln - firing of porcelain and earthenware, are effected
in the neighborhood of 1000 ° . There is still a possible economy
of 40% . Many such furnaces have been provided with recupe
rators ; there are, however, many places yet where the introduction
of recuperative gas -fired furnaces might result in aa marked economy.
Now that the problem of high temperatures has been satisfactor
ily solved, this class of furnaces presents today the most promising
field to engineers and furnace builders. From 500° down the
efficiency becomes good, and the benefits to be derived from the
use of recuperators are doubtful . Without condemning the at
tempts made in this line, let us recall that the introduction of the
Siemens system to boiler firing has not been successful. In this
case, from 500° down, steam boilers are , in fact, the best agent
for the utilization of the heat of the fuel. 1
Furnaces Followed by Steam Boilers. Calculation of the Lost
Heat, P, in a Puddling Furnace. In the discussion of direct-fired
1 This opinion has often been advanced , and Lencauchez has maintained that the steam
boiler is the best recuperator ; our figures corroborate this opinion .
furnaces, puddling and re - heating furnaces have been purposely

omitted . In these, direct firing is, and doubtless will be for a long
>
time to come, the general rule. The large amount of motive

power required by iron works has led users to place steam boilers .
in connection with puddling and re-heating furnaces, thereby
reducing the waste of heat. These boilers generate steam in general
for low -pressure condensing engines. It is possible to give the
boilers the design and heating surface best suited to high efficiency ."
Thus, there is a series of beneficial conditions particular to metal
lurgical furnaces followed by boilers which have a very good heat
utilization . It is interesting to find their exact efficiency .
027
SL
320
K2202
00-:
009
099
091
750
O 1000
O-300
1600
1050
850
220 * 220
300
300
>
Fig . 24. -PUDDLING FURNACE OF THE FORGES D'EURVILLE. SECTION A B.
In Fig. 24 is shown an example of a puddling furnace with water

cooled hearth ; it is followed by a boiler, not shown in the drawing.
The boiler is horizontal, and gives a steam evaporation of 4 to 4.5
kg. of water per kg. of coal burned on the grate. The waste pro
ducts have a temperature of 200 to 250 ° at the stack . Supposing,
by analogy with similar puddling furnaces, that the combustion
takes place with an excess of 5% oxygen in the waste products,
corresponding to the formula
CO 2, + 4N , + } (0 , +4 N 2)
However , they usually cannot be crowded , as their heating depends entirely on the
puddling furnace , and it may happen that, at the time when most steam is required, the
boilersfurnishthe least. - Translator .
with a temperature of 200 ° at the stack, we find that only 9.8

calories out of 97.6 have been lost, corresponding to an efficiency
of 90%-an efficiency higher than that of most systems of re
cuperation .
Therefore, there is little incentive, from the point of view of
fuel economy, to substitute recuperative furnaces for the above
type of puddling furnaces.
Steam Boilers. Heat Lost When Combustion is Effected by the
Use of a Large Excess of Air.— The most interesting case to take
up is that of the multitubular boiler used in marine engineering.
There, fuel economy is most highly valued, and there, also, are met
1050 580
1850
530 1000 -520 360

B
1700
1000
-1600 1180
1300
900
400
OST
150
400
-300 -500
Fig. 240. — PUDDLING FURNACE OF THE FORGES D'EURVILLE. Section, C D.
the most unfavorable conditions ( small volume and the necessity of

quick steaming) . This problem has also been the most thoroughly
studied . Experimental data have been given us by the builders
of the best known of this class of boilers. We give as examples the
drawings (Fig. 25, 26 and Fig. 27, 28) of two types of marine
1
boilers — the Belleville and the Niclausse. Both are multitubular,

but the water circulation is different.
In the Belleville the tubes are placed in series ; in the Niclausse
they are in multiple on vertical collectors, and the circulation is
established variously, according to the rate of steaming at which
the boiler is operated. Such are the main differences between the
two systems . In both boilers and , in general, in all boilers with

intensive vaporization , the difficulty encountered is the keeping
bi
B1
Bi
bi
K
h
1
B
Du
Fig . 25. - SECTION THROUGH A BELLEVILLE BOILER AND ECONOMIZER
of a low temperature in the waste products when the rate of com

bustion is increased .
The economizer introduced by M. Belleville is designed to re
cuperate some of the heat, otherwise lost, by heating the feed -water
and thereby bettering the fuel efficiency.
,
B
Aj C
B
с
B
D ?
Fig. 26. — LONGITUDINAL SECTION AND FRONT ELEVATION OF A BELLEVILLE

BOILER AND ECONOMIZER .
The Niclausse boiler has no economizer, but its system of cir

culation is such that the tubes furthest from the grate act as feed
water heaters.
The distinguishing feature of boilers, as compared with furnaces,

is the lack of a ruling temperature. The temperature varies in
a decreasing manner continuously from the grate to the stack .
qu
o
0597
Loso
00000 0 ooo
OOC
Fig . 27. - SECTION ALONG A B OF THE NICLAUSSE MULTITUBULAR BOILER .
In a theoretical study it is not even permissible to take as a ruling

temperature the temperature of water vaporization at the steam
working pressure , since the extreme tubes in the Niclausse and
the economizer in the Belleville have a lower temperature. The

only data allowing the calculation of the lost heat are experimental
2.100
या
o
-2,170
2.100
Fig. 28.-FRONT ELEVATION OF A NICLAUSSE MULTITUBULAR BOILER .
-the temperature of the products of combustion at the stack and

the analysis of the gases.
The following data have been given us by M. Niclausse from

a series of trials made of boilers built for torpedo boats. The tem
perature at the smoke -box varies widely , according to the rate of
combustion. It is as low as 140 ° for a vaporization of 20 kg. per
square meter of heating surface, and for a consumption of 100 kg.
per square meter of grate -hour. It reaches as high as 370-400 °
for a vaporization of 55 kg. and a consumption of 400 kg. of coal.
The Belleville boilers fitted with economizers have a temper
ature in the smoke- box varying from 175° for a coal consumption
of 70 kg. per square meter of grate area to 375° for a consumption
of 170 kg. of coal. The composition of the gases - the average of
three tests on Niclausse boilers - is the following:
CO , 0, CO N,
8.2 10 . .80 81 .
It is seen that it corresponds to a large excess of air : 10% of

free oxygen corresponds to an excess of 50 % of air. The water
vapor present was not given .
To estimate the heat utilization in these boilers, we must make
the calculation of P within the limits of the temperatures 100 °
and 500 °, with the supposition that the excess of air in the waste
products is equal to the volume of the products of combustion.
With this hypothesis the composition of the waste products is
CO , + 0 2, + 8 N2 .
2
We have 97.6 calories liberated per molecular weight ( 12 grams)

of burned carbon .
Thermal capacities of the waste products:
Gas 200 ° 400 ° 600 °
CO2 1.8 4.8 6.4

0,+8 N , 12.6 25.4 38.8
Total. 14.4 30.2 45.2
% of loss . 14.4 31.0 46.3
The curve of the heat utilization is traced from this data (Fig .
14) . It is seen that, as soon as the boilers are crowded, the ef
ficiency falls rapidly. For a temperature of 500 ° in the smoke
box-a temperature that may be reached at times — the utilization
of the heat is only 65% .
It is, then, permissible to remark that there is a field for im
provement in steam -boiler practice. The methods of gas re
cuperation have small efficiency at 500 °. The recuperation by
the heating of the feed -water has also a limit, and it would appear
100 %
है & so
80 ,
70
60
Pa
50
Ye
40
30
20
'00 200 300 900 500 600 ° c Temperatures

Fig . 29. - CURVES SHOWING THE PERCENTAGE OF THE HEAT LOST AND
UTILIZED IN MULTITUBULAR BOILERS.
that the economizers have already given all that may be expected
from them . The only remedy seems to be in the direction of per
fect combustion without excess of air. The solution of the problem
is, however, far from having the simplicity that might at first
appear.
The use of gaseous fuel, which at high temperatures allows a
closely neutral combustion, would probably be of poor efficiency
in boilers, since the rate of combustion of gases is slower the lower
the temperature. The combustion would most likely be incomplete,
DIRECT- FIRED FURNACES . 85
and sooty deposition would take place. The very hot flame re
sulting from combustion without excess of air would injure the
boiler shell, and other obstacles might be encountered. It re
mains true, however, that it is in the line of complete combustion,
without excess of air, that the most gain is to be expected .
CHAPTER V.
FIRST CLASS OF Gas - FIRED FURNACES — SIEMENS Gas- Recu

PERATIVE FURNACES, FIRED WITH CARBON MONOXIDE .
The following reaction takes place in the gas producer:
C + 10 , + 2 N , = CO + 2N ,
2
S
primary air
1 2 1 2 Molecular Volumes,
with a liberation of 29.4 calories. The primary air being intro

duced cold, the temperature of the gas issuing from the producer
is 1290°. If the primary air has been pre-heated , to the heat of
combustion should be added the sensible heat brought from the
producer. The temperatures of combustion are given in Fig. 34
for the temperatures 0°, 600 °, 800 °, 1000 °, 1500 ° C.
The combustion is completed in the hearth according to the
reaction :
CO + 2 N , + 10 2, + 2 N , - CO , + 4N , + 68.2 Calories
2 2
gas secondary air

The total liberated heat-29.4 + 68.2—is 97.6 calories, and the
composition of the waste products is the same as in the case of
direct- fired furnaces — CO 2, + 4 N 2.
First Case. Recuperation by Means of the Secondary Air Alone. -
The furnaces coming under this heading are the retort gas-furnaces
of the Compagnie Parisienne du Gaz, Siemens system , or Lencau
chez system , where the draught is secured through natural means
without the injection of any steam in the ash -pit ; also, some
glass furnaces provided with counter -current recuperation, as the
Radot furnaces, Göbbe system large glass tank furnaces with
Siemens recuperators , etc
FIRST CLASS OF GAS - FIRED FURNACES . 87
60
Aire
40
1200
20
9 Calories
Primary
30
1
1 1
A - 1290
600
Air
ar
1400
8000
Air
ar
vo si
0091
.00
O : 1705
0001
0.7845
2000
6-1985 °
2200
2. 2380 °
Tem
240p0.
Fig . 30. — CURVE SHOWING THEORETICAL TEMPERATURE OF FORMATION OF

SIEMENS Gas, THE PRIMARY Air BEING INJECTED AT 0°, 600 °, 800 °, 1000 °,
1500 ° C.
It is apparent that, notwithstanding the large use of mixed gas

the number of furnaces of this type is still large.
We give two examples of these furnaces. Fig. 31 , 32, from
drawings furnished by the Compagnie Parisienne du Gaz, rep
88 INDUSTRIAL FURNACES AND METHODS OF CONTROL
2.070 -2.070
LA
6.000
-2.195
3.900
Ateller Floor
NANZA
4.100
ON
-800-160 F1.090– -1.370 4604800+

1.100
Generator
-2.200
D Basemen Floor
Futies
Draft
ehimsey
Fig . 31. - ILLUMINATING -GAS MUFFLE FURNACE OF THE COMPAGNIE PARISI

ENNE DU GAS. SECTION A B.
G
1.395
Gonciator Air
882
1425
3504354-882
1.300
OS
A
Oxyde Carbon
*1R .165
85
-1.395
008'S:
Generator B
1:10
Air
Fig . 31a . - SECTION C D

43.40
2011
CO .40.00 !
-3 :800
-1.166-07
680-735-135750
-1.900 50-950
Generator
36.9
Painel Air
FIG . 32. - ILLUMINATING -Gas MUFFLE FURNACE OF THE COMPAGNIE

PARISIENNE DU Gaz. SECTION EF, G H
WANA
1.165
1.165
Fig. 32a . - SECTION LJ.

resent its most approved type of Siemens furnace for gas retorts.
From a practical point of view its construction is very interesting,
for by the happy grouping of the recuperators the radiation of the
chambers is reduced to a minimum . From the point of view of
theory it comes very close to our hypothesis, since that furnace
is fired with gas coke. The ruling temperature is about 1000 °.
Fig. 33, from drawings given by M. Lencauchez, illustrates &
furnace with only secondary air recuperation , the recuperators
being of the parallel counter -current type , Lencauchez system .
Its ruling temperature is higher than 1400 °. This recuperator,
as is well known, is formed by hollow firebricks.
CalcuLATION OF THE LOST AND UTILIZED Calories in FURNACES FIRED WITH SIEMENS
Gas , with SINGLE RECUPERATION BY THE SECONDARY AIR .
Available ( Calories = 97.6 . 10000 1500 °
I CO2 12.2 21.1

Thermal Capacities of the Combustion Products ... 29.7 46.4
14 N2
F Total . 41.9 67.5
Calories Recuperated by the Secondary Air . 202 3.7 5.8
2 N2 14.9 23.2
R Total... 18.6 29.0
Lost Calories - F-R 23.3 38.5

% of Lost Calories P 24.0 39.5
% of Utilized Calories, Efficiency = U 76.0 60.5
The calculations for the heat efficiencies are given above.

As was the case with direct firing, the efficiency increases with
the lowering of the ruling temperature, but at a much lower rate.
There are no upper limits to the temperature to be attained .
If we have the secondary air pre-heated to 1000 °, the gas being
at its theoretical temperature 1290°, we find by the graphic method
( Fig. 34) a temperature of combustion of 3000°.
We may, therefore, conclude that furnaces with single recu
peration , fired with Siemens gas, can give all the usual industrial
temperatures, and that within these temperatures the efficiency
should always be 50% . For lower temperatures, and especially
for the distillation of gas from coal , the efficiency reaches 75% .
D
C -.- .
12
Hot TATT
Cold Air
2 L
FIG . 33. - Glass TANK FURNACE , FIRED WITH SIEMENS GAS, with RECUPER
ATORS OF THE LENCAU CHEZ PARALLEL COUNTER - CURRENT TYPE. (SEC
TION ALONG G H.
Cold
Air
Fig . 33a. - SECTION ALONG KL.

This efficiency explains why the Compagnie Parisienne du Gaz

has kept in operation a system of furnaces in which the recu
peration is not perfect, but where the further gain of calories to be
attained would not justify the cost and complexity of furnaces
with four recuperative chambers fired with mixed gas. There
fore, we reach a priori, by theory, results that have been found in
practice.
IRO
Q = 116.17
- K
ene
nt
587
Po
of W. co
100
ity
ac
Bolle
p
80 Ca
al
926
Therm
Ziai
gooi
60
1000 2000 2 400 2800 32000
Fig. 34. - GRAPHIC DETERMINATION OF THE TEMPERATURE OF COMBUSTION

OF SIEMENS Gas (CO + 2N , ) HEATED TO 1290° C. , THE SECONDARY AIR
BEING PRE -HEATED TO 1000 °.
Second Case. Furnaces with Double Recuperation for the Second

ary Air and Gas .-All the first Siemens furnaces, and especially
the first Martin-Siemens furnaces, come under this heading. They
were generally constructed with the inverted siphon ; its function
has been explained . This organ has been condemned, since it has
become possible to build furnaces without it.
This much debated question will be discussed and our calculations
will give its true value.. It is one of practical interest, since there
are furnaces of this type in existence.
FIRST CLASS OF GAS- FIRED FURNACES . 93
In order to establish the heat balance-sheet of a furnace with

double recuperation, it is necessary to add to the calories absorbed
by the secondary air those absorbed by the gas. The gas is not
cold , as it leaves the producer ( according to our hypotheses) at
the temperature of combustion of carbon burning to carbon monox
ide in cold air - that is, 1290°. Therefore, for any ruling tem
perature below 1290 °, gas recuperation is impossible, furnaces with
four recuperative chambers are unnecessary , and the theoretical
cfficiency of such furnaces would be equal to that of a furnace
with simple recuperation . We have to make our calculations only
for 1500° .
CALCULATION OF THE LOST AND UTILIZED HEAT IN A FURNACE FIRED WITH SIEMENS
Gas , WITH DOUBLE RECUPERATION , SECONDARY AIR, Gas .
Available Calories = 97.6 1500 °
F - Thermal Capacities of the Products of Combustion 67.5
by the Secondary Air from 0 ° to 1500 ° 29.0

R - Calories Recuperated... by the Gas from 1280 ° to 1500 °. 4.7
Total... 33.7
F--- R -Lost Calories.. 33.8

% of Lost Calories - P. 34.6
% of Utilized Calories - U 65.4
Remark . — The calculation at 1000 ° is impossible ; the double recuperation at that tem
perature would give the same result as the single recuperation.
The lost heat, P, is 34.6% , thus giving a bettering of ef

ficiency of 4.9% over that of furnaces with single recuperation .
Still a third of the available calories is swept away to the atmos
phere by the products of combustion - more than enough to pre -heat
the gas to 1290 ° (29.4 calories). Thus, with good recuperation
it is possible to pre- heat the gas if it has been cooled by the siphon
or by a long flue before entering the recuperator. In the case of
Siemens gas there is, therefore , no waste of calories through the
use of the siphon. It follows that it is perfectly rational to place
the producer at any convenient location , since such design is not
wasteful either for proper efficiency or for the production of high

temperatures.
The Siemens siphon and very long flues evidently have their
drawbacks, since they cause deposition of tar and soot and there
by impoverish the gas. They should not be condemned on ac
count of the radiation of the sensible heat of the gas, which would
be wasted in case gas is recuperated . In that respect they are
very rational. They have rendered good service in the beginning
of the art, and may still be useful when it is not desirable to use
>
mixed gas and when natural draught is a necessity.

As an example of a furnace with double recuperation using
air and Siemens gas, there exists a glass furnace, Göbbe system ,
built in Russia, using wood for fuel . The gas, as it leaves the
producer, encounters a water spray in order to condense any steam
>
produced in the wood distillation . The gas enters the recu

perator cold . We have here an extreme case in which gas recu
peration is necessary .
Third Case. Double Recuperation : Primary and Secondary
Air. — The complete recuperation of the total air is impossible
under present industrial conditions. The ruling temperature of
a producer blown with primary air, heated to the temperature of
the hearth, would be so high that no known type of producer co’ld
be used ; we have seen that it would be over 2000 °. If the Ebel
men producer (Fig. 16) , with ash fusion , could stand such a tem
perature the exterior radiation would be very high , thus com
pensating to a great extent for the advantage resulting from the
pre-heating of the primary air.
The question of the preservation of the grate prevents any appli
cation of the system . It is difficult to reach a temperature of 500°
in the ash -pit without burning the grates, so that this temperature
is almost the limit of the heating of the primary air, and its appli
cation is even then very limited.
With the exception of the Klönne furnace at Dartmund , fired
with mixed gas, and of trials with pre-heated air with Taylor,
Lencauchez and Siemens producers , there is only the new Siemens
furnace that realizes a partial recuperation of the primary air. The
design of that furnace is well suited to it. It is only necessary to
operate it in a way the reverse of that employed by Biedermann

and Harvey (in placing the ash -pit in communication with the air
chamber and not with the waste -product chambers ) to have a
simple pre-heating of the primary air. If necessary , the pre -heated
air should be diluted with cold air, in order to lower its temper
ature below 500 °.
Some of the new Siemens furnaces are today operated in this
manner ; the greater part, however, inject some products of com
bustion , steam , hot and cold air under the ash -pit. In any case,
it is necessary to study the double recuperation by the total air in
order to appreciate its value.
The calculation is very simple ; the primary air being equal to the
secondary air, it is sufficient, in the case of single recuperation, to
double the figure of the recuperated heat. The results are given
below .
CALCULATION OF THE LOST AND UTILIZED CALORIES IN FURNACES FIRED WITH SIEMENS
Gas, WITH DOUBLE RECUPERATION BY THE TOTAL AIR .
Available Calories 97.6 .. 1000 ° 1500 °
Thermal Capacities of the Combustion Products .. 41.9 67.5

Calories recuperated by the total Air .. 37.1 58.0
F- R Lost Calories... 4.8 9.5
% of Lost Calories - P 5.0 9.7

% of Utilized Calories , Efficiency U 95.0 3
90.3
These results show the very high value of double recuperation

by the total air. At 1500° there is still a loss of 10% , but at 1000 °
the efficiency may be considered as perfect, since the 6% left is
1
required to secure the proper draught in the stack. "
Theory gives us , then, the direction in which progress is to be
sought. It lies with medium recuperative furnaces, in which re
cuperation is today far from giving the fuel efficiency that might be
attained .
! The case of the triple recuperation with primary, secondary air and gas is impossible
with Siemens gas, since the temperature of the gas generated in a producer blown with
recuperated air is too high to allow any recuperation .
The advantages of high-temperature producers were foreseen by

Ebelmen , who advocated a grateless producer much like a cupola
furnace. This idea, after having been a long time dormant, has
been recently revived, and the Saillers producer illustrated in Lede
bur' is practically an Ebelmen producer. The displacement of
steam injectors by positive-pressure blowers in gas-producer prac
tice gives a new lease of life to the problem of Siemens gas, and forms
in a way a step towards the solution of the problem of hot- blast pro
ducers.
It must further be remembered that Ledebur showed that the in
crease of temperature in the producer tends to decrease the pre
mature combustion of the gas. It is easy to conclude that the
questions of recuperation by the primary air and the introduction
of hot-blast producers are vital.
1 Ledebur: " Manuel theorique et pratique de la Metallurgie du Fer. ” Paris. Baudry
et Cie .
CHAPTER VI .
SECOND GROUP OF GAS -FIRED FURNACES-MIXED GAS.
Reactions of Combustion .The continuous production of water

gas is impossible without the use of an external source of heat in
connection with the gas -producer. The reactions of gasification
and of combustion in a furnace fired with mixed gas are as follows :
We burn our carbon unit, the molecule ( 12 grams), and a fraction
" m " of the fuel is gasified by steam :
m (C + H2O) = m (CO + H2) (1)
with the absorption of
(29.4 58.2) m = - 28.8 m calories .
The remainder of the fuel is burned by air.

( 1 – m) (C + 2N , +102) = ( 1 –- m ) (CO + 2N )
2 (2)
with the liberation of ( 1 -- m ) 29.4 calories.
The mixture of the two gases has for its composition
CO + mH , + ( 1 – m) 2N 2,
2 ( 3)
The combustion of this gas with the secondary air will give rise to
products of combustion of the following composition :
CO , + 2N 2, + m (H ,0 + 2N 2) + ( 1 - m ) 2N , = CO ,
2
+ 4N , + mH ,0 (4)
The factor " m " controlling the reactions ( 1 ) and (2) is arbitrary
between limits, and is a function only of the quantity of steam in
jected under the grate. The temperature of the generated gas
varies inversely as the factor " m " is smaller or larger. This tem
perature must not be allowed to fall below a certain limit, other
wise the producer will be extinguished .
Experience has shown that below 600° the combustion of carbon

in air ceases . It follows that the temperature of the gas will always 1
1
be between 600 ° and 1290°, the temperature of gasification with
dry air. As the amount of steam varies, the temperature of the
gas also varies. In practice, when it is advantageous to produce
mixed gas, the policy to follow is one of maximum production,
with, therefore, the gas near 600 °. This is the temperature limit
used in our calculation . It is possible to find the numerical value
of the factor “ m ," and therefore the chemical composition of the 1
gases . 1
Let us suppose that the primary air is injected cold. To solve

the problem it will be sufficient to know that the calories liberated
by the combustion of ( 1 – m ) grams of carbon in air, less the
calories absorbed by the dissociation of “ m ” grams of water, will
be in such amount as to bring the temperature of the gas to 600° .
600
– 28.8 m + 29.4 ( 1 – m )
0
m ( CO + H ) + (1ẩm)( CO + 4N 2) }
We find that for m = } this condition is more than fulfilled , so
that for a gas formed of } water gas and fair gas its temperature
will be a trifle over 600 °, and the producer will work satisfactorily.
This theoretical result is in accordance with the established
practice of blowing the producer with a quantity of steam corre
sponding to a gasification of about one-third of the fuel.
If the primary air is pre -heated, the latent heat of the air is
added to the 29.4 calories liberated by the partial carbon com
bustion , and the value of " m " increases. At the temperatures
of 1000° and 1500 ° the pre
1
-heating of the primary air would allow
respectively 48 and 54% of the fuel to be gasified , providing the
temperature of the producer is maintained at 600 °. In the same
1 Results have been obtained as follows:
1000 600
29.4 — 58.2m + ( 1 - m ) C ( 4 02 + 2N 2) = C (CO + mH2 + ( 1 - m) 2N ,]

0 0
29.4 + 18.2 -- m ( 58.2 + 18.2) 12.6 -4.2m

35
* .48
72.2
SECOND GROUP OF GAS- FIRED FURNACES . 99
6
manner, if the steam were superheated , “ m ” would increase in
1
value.
First Case. Recuperation by the Secondary Air.- This case
occurs in the furnaces of the Compagnie Parisienne du Gaz ( Fig.
31 , 32) , when they are operated with water in the ash -pit and
by natural draught, or with closed ash -pit and injection of air by
Fig . 35. — LENCAUCHEZ Brass FURNACE , FIRED WITH MIXED GAS, with Sim
PLE RECUPERATION OF THE PARALLEL COUNTER-CURRENT SYSTEM .
means of a Koerting steam injector, The Siemens glass tank fur

naces and the Göbbe and Radot -Lencauchez furnaces are also
operated with mixed gas , as well as many furnaces in use in the
metallurgy of copper, as those of the Société des Métaux .
We give as an example (Fig. 35) a furnace for brass melting,
with simple recuperation . It is of the Lencauchez parallel counter
current system. The producers of this furnace generate a gas
1 Throughout the discussion of the cases arising from the use of mixed gas, we pre
suppose that the water is introduced in the form of steam ; that is, we neglect the latent
heat of vaporization - 10.8calories per molecular volume of steam .
LIBRARY
OF THE
UNIVERSITY
OF
CALIFORNIA
100
M FIG
36.
ORGAN
GRAVITY
D ISCHARGE
BILLET
H EATING
FURNACE
.--
SECOND GROUP OF GAS-FIRED FURNACES. 101
containing 12% of hydrogen . The ruling temperature is slightly

below 10000
Let us calculate the heat utilization , giving to " m " the value :
$ ( C + H ,0 ) = $ ( CO + H2)
3 ( C + 100 , +2 N2) = f (CO + 2 N2)
2
The composition of the gas is

CÓ ++ H 2 + N. 2
The required quantity of secondary air is

10 , + 2N , and 2 $0, + f N 2
for CO for 1 H ,
It is to be noticed that this quantity is greater by one-third
than for the Siemens gas . The composition of the waste products
is CO , + 4N2 + } H20, and the thermal capacities of these
2
waste products is also greater than that of the Siemens gas by

the entire thermal capacity of the water vapor. This heat is in
no wise negligible, since to heat one molecular volume of water
vapor from 0° to 1000º and from 0° to 1500° there are absorbed
respectively 11 and 18 calories.
CalculaTION OF THE HEAT Lost and UtilizED IN A FURNACE FIRED WITH MIXED
Gas , with RECUPERATION BY THE SECONDARY AIR ALONE .
Available Calories 97.6 . 1000 ° 1500 °
12.4 21.4
Calories carried away by the Products of SCO?
Combustion .. 4 N2 29.7 46.0
( * 120 3.7 6.3
Total. 45.8 73.7
Calories recuperated by the Secondary Air ! } 02 + 2 N2 18.5 28.9

from 0 ° to 1000 ° or 1500 ° Ti 02 + IN2 6.1 9.6
Total . 24.6 38.5
Lost Calories . 21.2 35.2

% of Lost Calories * P 22.0 36.0
% of Utilized Calories U. 78.0 64.1
In comparing these results with those above, we find that there

is very little advantage in using mixed gas in the case of single
recuperation. The bettering of the recuperation coming from

the increase of the secondary air is offset by the greater loss in
calories in the waste products, due to the steam , the thermal ca
pacity of which is very high. Thus, at 1500° the gain in calories
is 3.3 , and at 1000 ° only 2.1 . This last figure is very interesting
in connection with the furnace shown in Fig. 31, 32. It is seen
that there is very little to gain by the use of steam for single re
cuperation . In the above deduction is found the explanation and
confirmation of a practical fact already established .
Fig . 37. — MORGAN SUSPENDED -RooF BILLET -HEATING FURNACE.
Second Case. Double Recuperation by Secondary Air and

Gas.- This case is the most important one in the theory of gas firing
with mixed gas. Most Siemens furnaces for steel and glass use
a gas enriched through the use of steam . The inclination to do
away with the siphon accentuates further the tendency to use
mixed gas, since by the use of steam the gas is cooled as well as
enriched, in contradistinction to the effect of the siphon.
The interest of this case is further increased if we remember
that it is always easy to pass from Siemens gas to mixed gas by
slight modifications in the gas-producers.
In Fig. 38 is given an illustration of a tank furnace for glass
103
LIT
Y TELE
QUARIAL 2
TE
MA L
RY RIA QUALITY
1
TO TE
AC MA
FR RY
RE TO
AC
FR
SECOND GROUP OF GAS-FIRED FURNACES .
H
-9:
90
17.3333
SUPPLY
GAS
8.55 es.
1.15
INLET
FEED
1:38
CONCRETE
SECTION
FURNACE
TANK
GLASS
SIEMENS
38.A.-
B.
Fig
17136115
Refractory
Material
1Quality
ipality
110
-96
。
FIG . 38a. — SIEMENS GLASS TANK FURNACE .
Cement 030000
SECTION CDEF
F
能。
00
B
ne:(
114
AM
ore
23
|||||
曾曾曾曾曾
C
FIG . 38B - SIEMENS GLASS TANK FURNACE . SECTION GH.
SECOND GROUP OF GAS-FIRED FURNACES . 105
HD
18
70 1920
Fig . 38c. - SIEMENS Glass Tank FURNACE . SECTION JK .
of 150 tons capacity. The producer is of the Siemens type, with

water seal, and the producer-gas carries from 8 to 10 % of hydrogen.
The ruling temperature of the hearth is 1450 °.
In order to find the value of the utilized heat, it is only neces
sary to add to the figures of the preceding case the number of
calories taken up by the gas from 600° to 1000° or to 1500 °.
CalculATION OF THE HEAT Lost AND UTILIZED IN Gas FURNACES FIRED WITH Mixed
Gas, with DOUBLE RECUPERATION BY THE SECONDARY AIR AND BY
THE Gas, COMING AT 600 ° FROM THE PRODUCER.
Total available Calories 97.6 . 1000 ° 1500 °
Calories carried by the Combustion Products F. 45.8 73.7
Calories recuperated by the Secondary Air. 24.6 38.5

Calories recuperated by the Gas = CO + H2 + IN2 . 6.3 19.7
Total Recuperated Heat R. 30.9 58.2
Calories lost F-R 14.9 15.5

F-R P.
% of Lost Calories 97.6
15.2 15.8
% of Utilized Calories Efficiency U 84.8 84.2

Air Flue
29'0 "
6'6
+9
30
H30
OOO
0227
9314
.Ko
-15'14
224
Machinery
Tilting
KE
Line
6: ,65
Cen.
-lo'iva 10'14 30+ -120 -H'a 1309)

SER
26'o -26'ox
-12'ot 22'6
%0,00
Xt
20'3 "
-16'6 ‫ – لایه‬18 ay
Fig . 39. — 50 - TON CAMPBELL Basic FURNACE, STEELTON, PA .
( From H. H. Campbell's " Metallurgy of Steel," Fig . VIII -C .)
13"
16 15'0
of
Furnace
Line
Center
2'354483-30
18*
Lorel
Floor
Level
Ground
Air
-18'0 "
7104
TFlue
-12'7
%
MTV
FACERE
Y-
X
SECTION
PA.—.
STEELTON,
FURNACE,
Basic
CAMPBELL
50-1
39a
Fig.
of",“(-C.)
VIII
Fig.
Steel
Metallurgy
Campbell's
FromH.
The economy resulting from the double recuperation is apparent;

for the temperature of 1000 ° the gain is 84.8 - 75.7 = 9.1 % , and
for the temperature of 1500 ° it is 84.3 - 66.3 = 18% , thus giving
an economy of nearly one- fifth of the fuel . If we carry the tem
perature to 1800°, as in steel open-hearth furnaces, the economy
will be over 25% .
Such figures explain the favor with which the mixed gas was
received from the first. We have seen that the higher efficiency
is due wholly to the better recuperation. One-third more air is re
cuperated , and the gas has been recuperated from 600° to 1290 ° ;
to this double recuperative gain is due the fuel economy, which
we have seen may reach 25% .
Third Case. Double Recuperation by the Primary Air and the
Secondary Air; that is, Pre-heating of the Total Air.—No complete
application of this case is to be found in practice, since the pro
ducer grates would fail under the injection of the recuperated air.
Some partial applications are embodied in the Klänne retort fur
nace as well as in many of the New Siemens furnaces, where aa mix
ture of hot air and steam is injected together with some cold air,
so that the temperature in the ash -pit shall not rise above 500 °.
We give in Fig. 40 an outline drawing taken from the English
patent of Henry ; it is the Klönne system , applied to a gas
retort furnace. The recuperation takes place in two chambers by
the parallel counter- current system , one chamber being used for the
primary air, the other for the secondary. We have seen that
by pre -heating the primary air to 1000° and to 1500° , the amount
of water gas can be raised respectively to 48 and to 54 % .
CALCULATION OF Lost and Utilized Heat In Gas FURNACES FIRED with Mixed Gas
WITH DOUBLE RECUPERATION : TOTAL AIR . THE GAS PRODUCER
IS MAINTAINED AT 600° C.
Available Calories - 97.6 .. 1000 ° 1500 °
Calories carried away by the Combustion Products = F . 47 4 77.5

Calories recuperated by the Total Air R ... 37.0 57.8
Calories lost = F - R.. 10.4 19.7
Loss, % F - R Р. 10.6 20.2

97.6
Utilized Heat, Efficiency U. 89.4 79.8
4% -15'14
'
-16'44" 32'0
Furnace
Line
of
Center
6130
Line
Floor
150'-
48'3
151
II
ILISI
12
01
IBH TES TYG' 3:09
TLI
LIশ
-6'0"
3
821"
-3'0 "
ER
-200x -26'0x
SECTION
X1
Y-, SECTION
Y
X-,:
STEELTON
PA.
FURNACE50,-.
Basic
CAMPBELL
Ton
Fig.
39b
VIIIC.-,”()
Fig.
Steel
of
Metallurgy
H.
Campbell's"
From
· The heat utilization is very good for all temperatures, and it

.
compares favorably with the double recuperation by air and gas .

These results show that there is much to be expected from the
application of this system .
The recuperation by the total air has the great advantage of
reducing the number of the chambers to two, from which the
primary and the secondary air may be drawn. The better ef
ficiencies shown by the total recuperation of the air in the case of the
Siemens gas can be foreseen , since the steam used in the case of the
mixed gas carries to the stack, without any possible return , all
the calories necessary for its heating to the ruling temperature
of the hearth.
It should not be concluded, however, that the mixed gas is without
interest in the case of recuperation by the total air, since without
steam the temperature of the producer would become prohibitory ;
the dissociation of the steam used lowers the temperature to a
point where it seems that practical application should not be im
possible . We are brought to the dilemma of either finding means
of preserving the grate or using the Ebelmen type of producer.
The New Siemens furnace is well suited to the adoption of the pre
heating of the total air. 11
Fourth Case. Triple Recuperation by the Primary Air, Secondary
Air and Gas. — This case to our knowledge has not received any
industrial applications. It would necessitate three recuperators ,
the third recuperator to collect the heat left in the products of com
bustion after pre-heating the secondary air and gas . We have seen
( page 105) that there is left of the heat 15.2 and 15.8 % , at 1000°
and 1500 ° respectively. In spite of its complexity, this system is
worthy of serious consideration , especially for very high tempera
tures, on account of the high efficiencies it gives. In fact, they
>
have reached the very limit of the utilization of the heat of the fuel,
the remaining 6% being required for the draught in the cham
bers. From the point of view of recuperation alone, the limit is
? Many trials have been made in this direction, especially by F. Siemens, who
designed producer without a grate in which the coal took its own angle of repose . The
problem is, however, far from solution, since we do not know of any producer in which the
temperature at the ash -pit is ever as high as 450° , and beyond this temperature the pro
ducer does not work satisfactorily. There is still a large gap to be crossed ; the recuper
ation cannot be considered as complete below 1000 ° C. under the ash -pit.
Fig. 40. - KLÖNNE RETORT FURNACE .
# 0 0 \ C C E /Q
@ Oঃ
BZ
WE ( @ @ ) এ C ত টু (
do 00000000000
20 000000 00 00
হট চড়ে ৫ ক্রে ি হ কাজেগঞ্জছো
কাজঙ্গি হাজাহাঙ্গ ত্রিত
even exceeded in a sense , since the mass of the recuperating gases

is larger than the 'mass of the waste products. The primary air
passes twice through the chambers — once in the primary air re
cuperator and a second time in the gas recuperator. The products
of combustion , on the contrary, travel but once through the recu
perators, and furthermore their thermal capacity is less than the
combined thermal capacities of the total air and of the gas.
CALCULATION OF THE Lost AND UTILIZED HEAT IN GAS FURNACES FIRED with MIXED
Gas with TRIPLE RECUPERATION .
1000 ° 1500 °
48 % Water Gas 54 % Water Gas

52 % Siemens Gas 46 % Siemens Gas
Gas : Gas :
Available Calories 97.6 CO + .48H 2 + 1.04 N , CO + .54 H2 +.92 N.
Waste Products : Waste Products :
CO2 + 4N , +.48 H 20'CO2 + 4N2 + .54 H 20
2
Recuperation from 1000 ° to 600 ° to 1500 ° to 1600 ° to

600 ° 0° 600 ° 0°
Thermal Capacities of the Waste Pro

ducts F. 21.7 26.5 50.0 26.8
Available Calories for Recuperation by the

Total Air .. 15.6 21.5 36.4 21.5
Available Calories for Recuperation by the
Gas at 603 7.8 0.0 17.9 0.0
Total R .. 23.4 21.5 54.3 21.5
Excess of the Thermal Caparities of the

Recuperating Gases over that of the
Products of Combustion 1.71 4.32
Excess of the Thermal Capacities of the
Products of Combustion over that of
the Recuperating Gases = F - R 5.0 5.3
% of Loss = P. 5.1 5.4
% of Utilized Heat U. 94.9 94.6
1 The recuperation limit being exceeded by 1.7 Calories between 1000 ° and 600 ° , the re
cuperation is perfect.
2 The recuperation limit being exceeded between 1500 ° and 600 °, the recuperationis
perfect.
Under such conditions, it might appear that perfect recupera

tion-an efficiency of 100% -should result. This, however, is
evidently not the case , as it is easy to understand when it is re
SECOND GROUP OF GAS-FIRED FURNACES . 113
membered that only from 0° to 600° the total air takes part in the
recuperation , and that its thermal capacity is less than that of
the waste products. Therefore, between 0° and 600 ° there is a
loss, which is absolute, not being compensated in the next step
from 600 ° to the ruling temperature 1000 ° or 1500 ° -in which the
thermal capacity of the recuperating gas is larger than that of the
products of combustion .
It follows that, in order to be able to calculate the heat utiliza
tion, it is necessary to operate in two stages — one from 0 ° to 600 °,
the other from 600° to the ruling temperature, 1000 ° or 1500 °.
As an application of the triple recuperation of the gas, secondary
air and primary air, the system of A. L. Queneau may be mentioned .
B
MANA
FIG . 41.-QUENEAU SYSTEM OF TRIPLE RECUPERATION OF THE PRIMARY AIR,

SECONDARY AIR AND Gas.
It has the ordinary Siemens chambers for the gas and secondary
air, with a single chamber of the parallel counter -current type for
the primary air. Usually the waste products leave the Siemens
chambers on their way to the stack at a temperature which allows
the use of cast -iron pipes for the recuperator. In case of high
temperatures a firebrick recuperator is used .
The primary air recuperator is designed so that the waste pro
ducts leave it at a temperature of about 200 C. , a temperature
necessary for an efficient draught in the stack. The primary air

is forced through the recuperator by means of a positive blower,
while the heated air is led to the producer through a brick -lined
flue . In order to utilize the calories of the primary air to the best
advantage, without endangering the producer, the primary air
meets a system of water sprays (the steam injector being entirely
dispensed with) . The vaporization of the water injected is ob
tained wholly at the expense of the recuperated waste heat (doing
away with the boiler plant). By injecting the water in liquid form
in the producer and obtaining its vaporization thereby, the fire
zone of the producer is cooled more efficiently than by steam in
jection . The amount of injected air and water can be varied in
dependently at will, since they are not interdependent, as in the
case of the steam injector. The use of the parallel counter-current
system for the primary air does away with the complications of a
third set of valves . The regulation of the temperature of the pri
mary air recuperator is automatically obtained by the regulation
of the temperatures in the Siemens chambers. Fig. 41 shows the
application of the system to an open -hearth furnace. The drawing
is taken from the patent specifications . The calculations are shown
above.
From an examination of the table two conclusions may be made .
( 1 ) The very high efficiency of furnaces with triple recuperation .
(2) The very small influence of the ruling temperature on the
heat utilization .
This system of recuperation is , then , particularly suited to high

temperatures ; its use would result in a fuel economy of 10% over
that of the Siemens furnace .
There is a last case where the use of triple recuperation would
give an economy even greater than 10% ; it is in its application
to industrial operations, in which the waste products consist of
the products of combustion of the fuel, and of gases liberated by
the materials under treatment in the hearth - that is , water vapor,
3
carbon dioxide, sulphurous dioxide, etc. In the various systems

already reviewed , the calories carried by these gases would be
utterly lost, since the products of combustion of the fuel have higher
A
SECOND GROUP OF GAS - FIRED FURNACES . 115
thermal capacities than the recuperating gases. In the case of

triple recuperation the contrary is true , and therefore these extra
calories can be brought back to the hearth .
Glass furnaces present the typical example of this supplementary
recuperation. The materials charged in the furnace carry as much
as 45 % of volatile products; the coal required for the fusion of the
glass weighs about 60 % of the weight of the fused glass. The mass
of the volatile products is , therefore, mathematically speaking, a
quantity of the same order as that of the products of combustion
of the fuel . The ratio of the masses may be as high as zo, corre
sponding to a loss of 20 of the available calories. The recupera
tion of these lost calories, added to the increased economy resulting
from triple recuperation proper, would bring an increase of 15%
in the fuel efficiency by the application of this system to glass
furnaces.
In summing up the problem of the application of mixed gas to
furnace firing, it may be said that single recuperation is suited
only to high -temperature furnaces ; it is very advantageous to all
furnaces with double recuperation, and in the case of triple recu
peration it offers the maximum efficiency to be expected , irre
spective of the ruling temperatures.
CHAPTER VII .
FURNACES MAKING USE OF THE SO-CALLED REGENERATION OF

THE PRODUCTS OF COMBUSTION—New SIEMENS
FURNACE - USE OF PURE CARBON
DIOXIDE IN Gas- PRODUCERS.
Reactions of Combustion . — General attention has been called

to the regeneration of the products of combustion owing to the
introduction of the New Siemens furnace, in which this principle
receives a partial application. The furnace is so designed that
the injection of the waste products in the gas- producers can be
accomplished to any extent desired .
A number of years before Biedermann and Harvey secured
their patent, E. Göbbe had received a patent covering the in
troduction of products of combustion in the ash -pit of gas - pro
ducers. He was not able, however, to make a practical application.
It is only since 1893 that furnaces, based on the principle of the
regeneration of the products of combustion, have been constructed.
It has been proposed also to inject pure carbon dioxide in gas
producers, the carbon dioxide being taken from lime-kilns, for
example. No applications of this idea, which is , however, a rational
one, are known ; the difficulty, evidently, is to obtain carbon dioxide
free from nitrogen without undue cost .
Thus, theoretically, we have two cases to examine:
( 1 ) Injection of the production of combustion.
(2) Injection of pure carbon d'oxide.
The results of the use of carbon dioxide have been studied ( page
63) ; we have seen that it acts precisely as steam :
( 1 ) By decreasing the amount of the primary air.
(2) By lowering the temperature of the gas, on account of the
heat absorbed in its decomposition .
REGENERATIVE FURNACES . 117
C + CO , = 2 CO with the absorption of 38.8 calories per

molecular volume; the cooling action of the steam is thus less
efficient by a difference of 38.8 – 28.8 = 10. calories . II, how
ever, the vaporization of the steam is effected in the producer
itself, at the expense of its sensible heat, the efficiency of steam is
equal to that of carbon dioxide. Like steam , carbon dioxide cannot
furnish any calories, and its use does not increase the number of
available calories.
First Case. Injection of Pure Hot Carbon Dioxide in Gas-Pro
ducers. - This case does not properly belong to the study of heat
recuperation , since the injection of hot carbon dioxide constitutes
an addition to the available calories of the fuel. We shall examine
it, however, as it represents the only true case of carbon regen
eration . Let us suppose that , close by the gas -producer, we
have a supply of pure carbon dioxide at 1000° C. To calculate
the maximum economy to be realized , we have only to consider
that the whole of the sensible heat of the carbon dioxide is made
use of in the regeneration of the carbon . It suffices to calculate
the amount of carbon dioxide that can be injected at 1000° in
the producer without lowering its temperature below 600 ° C.
The calculation is precisely the same as in the case of steam .
Let " m " be the proportion of carbon gasified by the carbon
dioxide in the total amount ( 12 grams) of the burned carbon.
m (C + CO2) 2mCO
liberating m ( - 38.8 + 12.7) calories, 12.7 being the thermal

capacity of a molecular volume of carbon dioxide from 0 ° to
1000 ° .
( 1 – m) (C + 10 , + 2N 2) ( 1 – m) (CO + 2 N2)
liberating ( 1 – m ) 29.4 calories .
The chemical composition of the gas will be
(1
1 + m)
m CO + ( 1 – m) 2N ,
of which the thermal capacity from 0° to 600° is ( 3 – m) 4.3 cal
ories. Hence the equation :
- 26.1 m + 29.4 ( 1 – m) ( 3 – m) 4.3
m .33
One-third of the fuel will, therefore, be gasified by the carbon

monoxide, the remaining two -thirds by the air. The resulting
economy will be, per molecule of burned carbon, one-third the
12.7
thermal capacity of carbon dioxide from 0° to 1000 °, or 3
;
12.7 X 100
in percentage, the economy is 3 X 97.6 4.3 %
Evidently to this economy is to be added that resulting from

a better recuperation , as in the similar case of the mixed gas group,
and that due to the lack of latent heat of vaporization . Neverthe
less, the advantages accruing from the use of pure carbon mon
oxide are not sufficiently large to warrant many trials of applica
tion .
Second Case. In¡ection of Hot Products of Combustion in the Gas
Producer.— The injection of the waste products in the gas-producers ,
like the mixed gas, may be applied to the four methods of recu
peration .
As the applications of this system are limited , and further since
two of the cases -- namely, that of the triple recuperation and
that of double recuperation by secondary air and gas - have never
been attempted , it will suffice to condense in one table the results
of all the calculations .
The carbon dioxide taken from the waste products is accompanied
by a volume of nitrogen , which is the same for all furnaces in
which carbon is the fuel. This very large volume of nitrogen
increases to a marked degree the amount of sensible heat going
to the producer. It follows that it is possible to send to the pro
ducer a much larger volume of carbon dioxide than could be done
with steam or pure carbon dioxide without lowering the tem
perature of the producer below 500°. The other conditions being
the same, the influence on the recuperation will be correspond
ingly increased. A simple calculation shows that, when the waste
products are injected at 1450º, no more primary air is required ;
all the air will be secondary. At 1000 ° it will be possible to send
44% of carbon dioxide.
Under the above condition its use will result in a great improve
ment of the efficiency . It must be remembered , however, that this

latter is due wholly to better recuperation , by the increase in the
secondary air and the cooling of the gas , and that no gain whatsoever
comes from the so -called carbon regeneration , and furthermore
that there is no increase in the number of available calories.
CALCULATIONS OF THE EFFICIENCIES OF FURNACES FIRED WITH GAS OBTAINED FROM
PRODUCERS INJECTED WITH PART OF THE PRODUCTS OF COMBUSTION .
Recuperation Double Recu- Double Recu- Triple

by the peration peration i Recuperation
Available Calories 97.6 -Q Secondary by Air and by Total
Air Gas Air
1000 ° 1500 ° 1000 ° 1500 ° 1000 ° 1500 ° 10000 1500 °
Thermal Capacities of the Pro- ' 42.1 67.5 42.1 67.5 42.1 | 67.5 42.1 67.5
ducts of Combustion F
Calories Re 26.5 57.81 26.5 57.81

Secondary 37.0 57.81 37.0 57.81
cuperated Primary Air
by Pre 7.52 7.32 5.11 - 7.32
heating the | Gas
Total P.
26.5 57.8 34.0 65.1 | 37.0 57.8 42.1 65.1
F -R Calories Lost .. 15.6 9.7 8.1 2.4 1 5.1 9.7 2.4 2.4
F-R P ! 16.0
% of Lost Calories 10.0 8.5 2.5 5.6 10.0 2.5 2.5
Q
% of Utilized Heat, Effi
ciency = U . 84.0 90.0 91.5 97.5 94.4 1 90.0 97.5 97.5
1
1 Total Air .
1 Total Air : 1 Total Air . 1 Total Air . 1 2 The Recu
see page 116 . 2 The Recu peration limit
peration limit is exceeded ;
is exceeded ; the the gas could
gas could ab absorb at
sorb at 1000 1000 13.4
Remarks. 15.12 Cal . ; at Cal. and at
1500 °, 14.4 Cal . 1 500 ° 44.4
if the equiva Cal. if the
lence of the equivalence of
Masses did not Masses did
limit the Re not limit the
cuperation . Recuperation .
The efficiencies given above are remarkable ; they are better

than most of the corresponding cases of the Siemens and of mixed
gas firing. This is to be expected, since the hot waste products
have a greater effect on the recuperation per molecular volume, and

the mass of the waste products is not increased by their use, causing
a useless waste of calories to be discharged into the atmosphere.
However, it must be remembered that this class of furnace is,
on the whole, entirely theoretical. Only single recuperation and
that of the total air have been attempted, 1' and that only to a
small extent, by using but a trifling part of the available waste
products. The question of the preservation of the grate prevents
any more complete application.
We have greatly simplified the treatment of the problem by
assuming the absence of the large amount of steam required for
the injection of the waste products into the producer. If we take
the case of the New Siemens furnace we see ( Fig. 17) that it is pro
vided with three Koerting injectors, one in communication with
the waste products flue , the second with the air chamber, the third
with the atmosphere. The large amount of steam used reduces
very largely the amount of waste products that could be used ; it
therefore brings this furnace very close to that of a furnace fired
with mixed gas with secondary air recuperation and partial pri
mary air recuperation . The saving in fuel that this furnace has
given in many instances is probably due to the very happy grouping
of all the organs, the radiation losses being reduced in a large
degree.
If we take the case of a New Siemens furnace with a ruling tem
perature of 1000 °, the heating of the primary air gives a maximum
efficiency of 75.7% with Siemens gas , of 78% with mixed gas, and
of 84% with the injection of the waste products. The recuperation of
the secondary air brings the efficiencies respectively to 94.8, 89.4,
91 % . The efficiency of the New Siemens furnace should lie in an
intermediate position with regard to these various figures. It
may be safely said that this furnace will have a slightly better
efficiency than either of the other two cases with simple recu
peration.
In the case of the Siemens gas , when the total air is pre-heated
the same efficiency - namely, 94%-is attained as with the in
1 F. Siemens has since patented a four-chamber regenerative furnace with injection of
waste products corresponding closely to the total recuperation by primary air, secondary
air and gas .
jection of the waste products ; therefore it is rational to inject

into the producer an equal volume of waste products and heated
air. The superiority of the New Siemens furnace is not so much
due to the so-called regeneration of the carbon dioxide as to the
total heating of the air. In any event, for medium temperatures
this furnace is better than any other type of furnace using single
1
recuperation ."
' At 1500 ° single recuperating furnaces have efficiencies of 60.50
and 64.11 % when fired with Siemens and mixed gas respectively.
Double recuperating furnaces fired with mixed gas give 84.3% .
The difference between these efficiencies is too large to be over
come by the partial application of the so -called regeneration of
the products of combustion . At this ruling temperature the
New Siemens furnace may give better results than the first
Siemens furnace with air siphon when fired with Siemens gas; but
it will not equal that of furnaces with double recuperation fired
with mixed gas.
The New Siemens furnace is of a construction more economical
than furnaces with double recuperation ; it is better adapted than
any other to furnaces of small capacities ; and its regulation is
easier than that of furnaces with natural draught. These are ad
vantages of enough importance under industrial conditions with
out a priori giving to the furnace a better fuel utilization , which
theory disproves and which practice has not as yet shown .
1 When , as in present practice, only one - fifth of the available waste products is injected
into the gas producer, the accrued economy is very low, from 1 to 3 % This may be
verified by a simple calculation. If, besides, the waste products are cooled, the resulting
economy is practically nil .- Translator.
CHAPTER VIII .
CLASSIFICATION OF THE VARIOUS TYPES OF FURNACES.
In Table 11 we give a résumé of the preceding chapters. This

table gives the absolute value for the maximum heat utilization
in the furnaces, from the purely theoretical point of view , without
taking into account their degree of practicality. Therefore, from
such tables we should not deduct the classification of the furnaces
by merely taking into consideration the value of the utilized heat
U. The classification so made would be exact only for furnaces
capable of complete industrial application; for all the other cases
there is room for discussion .
On this basis, we will review briefly the foregoing study.
For a temperature of 1500° and higher, the best furnace in use
today, from the point of view of the utilization of the heat of the fuel,
is the Siemens furnace, with double recuperation by the secondary
air and gas when fired with mixed gas , the producer not having a
siphon. Its recuperative efficiency is 85% .
By pre-heating the primary air it would still be possible to
gain a further 10 % , since the limit of the triple recuperative fur
nace is 95% . The pre-heating of the primary air for such fur
naces would be less difficult than for furnaces with simple recu
peration , as heating the air to 800° would be sufficient to secure
the 10% increased efficiency .
This improvement, to be realized by the triple recuperation
with mixed gas without siphon , would give a perfect utilization of
the available calories of the fuel and be a solution of the problem
which might be considered as final. 1
1 Let us recall that the triple recuperation furnace presents a special interest in the
case where to the mass of the products of combustion comes an additional gaseous mass,
issuing from the material worked on the hearth , as in glass furnaces. We have seen that
the supplementary efficiency which would result is 3 to 4 % . In this special industry a
saving of 10 to 14 % could be secured . This is a very appreciable advantage, since the
heating is continuous and the cost of the fuel is a very important item in the cost of pro
duction .
TYPES OF FURNACES . 123
With marked inferiority follow the furnaces with double recu

peration fired with Siemens gas, with an efficiency of 66% , closely
followed in turn by the furnaces with single recuperation of mixed
gas with 64 % , and lastly the furnaces with single recuperation of
Siemens gas with 60 % .
The furnaces with injection of the products of combustion into
the producer have not been included in the preceding classifica
tion , since their industrial applications are very limited as yet ;
all that can be said of them is that they are superior to the last
two types mentioned . In their present condition that is , with
a temperature in the ash -pit of 400 ° to 500º --they can give a better
efficiency than the original Siemens furnace, but they are inferior
to the Siemens furnace fired with mixed gas.
The classification may then be condensed as follows :
()
1 ) Mixed gas furnaces with triple recuperation .
(2) Mixed gas furnaces with double recuperation by secondary
air and gas .
( 3) (a) Siemens gas furnace with double recuperation by air
and gas.
(6) Furnaces with injection of waste products into the pro
ducer ( New Siemens furnace ) .
(4) Mixed gas furnace with single recuperation.
(5) Siemens gas furnace with single recuperation .
( 6 ) Direct -fired furnaces.
In the case of medium temperature furnaces, say 1000°, the best

furnace is also the double recuperative furnace fired with mixed
gas , but it does not appear that the double recuperation has here
a very great advantage. The difference in efficiency between a
furnace with one pair of chambers ( 78% ) and that of a furnace
with two pairs ( 85% ) is too small to warrant the increased cost
and complexity of construction ; the additional radiation from the
extra pair of chambers may offset the difference. For such tem
peratures furnaces with one pair of chambers are to be preferred .
TABLE 11 .
COMPARATIVE TABLE OF ALL SYSTEMS OF FURNACES AT RULING TEMPERATURES OF
1000 ° AND OF 1500 °
Maximum Utilized Heat U
10000
Classification Diagrams of the Percentage of the Heat Industrial
Lost P and Utilized - U Applications
Copper furnaces, Boë .

Direct firing tius, Bicheroux fur
naces .
Retort furnaces for il

luminating gas.
Siemens gas No possible applica

tions.
Partial application in
some retort furnaces.
Retort furnaces, Sie

mens , Lencauchez
furnaces for brass,
etc.
Mixed gas (Siemens Siemens furnace.

with the permissible
maximum water gas) Partial application in
Klönne retort fur
naco .
No applications.
the New Siemens.
Siemens gas with the No applications.

permissible maxi
mum of regenerated
products of combus Partial application in
tion .. the NewSiemens,
No applications.
Remark.- Direct, Siemens gas and mixed gas firing have received complete industrial
applications.
TABLE 11 (Concluded)
Minimum Lost Heat - P
1500° Industrial
Method of Diagrams of the Percentage of the Heat Applications
Recuperation Lost land of the Heat Utilized -U
Puddling furnaces,
JIO
glass and ceramic

No recuperation . furnaces.
Single recuperation by Glass open -hearth fur

the secondary air. naces,
Martin .- Siemens fur

Double recuperation naces for steel, glass ,
secondary air, gas. etc.
006
Double recuperation No applications.

secondary and prim
ary air.
Single recuperation by Glass open -hearth fur

the secondary air. paces .
Double recuperation
secondary air, gas. Siemens furnaces.
Double recuperation Partial application in

secondary and pri the New Siemens.
mary air.
Triple recuperation No applications.
secondary and pri
mary air, gas.
Single recuperation by the New Siemens.
the secondary air.
Double recuperation No applications.

secondary air , gas.
Double recuperation Partial application in
the New Siemens.
secondary and pri
mary air .
Triple recuperation No applications.
secondary, primary
air ana gas.
ਸ
The applications of the various cases of the so -called regenerated products of combus
tion are either very imperfect or wanting ; their classification offers, therefore , only a
theoretical interest .
Today most of the furnaces with the ruling temperature of 1000 °

have single recuperation by secondary air alone, and are fired either
with Siemens or mixed gas. We have seen that theoretically the
two systems are equivalent at the temperature under considera
tion , and the selection to be made depends very much on the char
acter of the fuel.
ency
Increased
Effici
50
40
So
y
mar
Pri Air
20
y
ondar
Sec Air
10
Gas
Temperatures orRecuperation
200 400 600 ° 900 1000 1200 500
Fig 42. - CURVES SHOWING , FOR DIFFERENT TEMPERATURES , THE THEORET

ICAL ECONOMIES OCCURRING FROM THE PRE -HEATING OF THE PRIMARY
AIR , SECONDARY AIR AND GAS .'
It would be advantageous, however, if practical obstacles do

not block the way, to introduce a system allowing the pre-heating
of the primary air without increasing
2
the number of the chambers,
as in the Klönne furnace, or to send back under the grate
part of the waste products, or to do both , as is done in the New
1 Evidently the various economies are cumulative for a given temperature. Thus, if at
a ruling temperature of 1,000° all three gas-currents are pre -heated, the total resulting
economy is 17 + 8 + 30 = 55 % of the fuel which would be used if no recuperation was
obtained .
2 See page 111 .
Siemens furnace. For our purpose these various improvements

tend toward the same end -- a maximum efficiency of 94 % ; in the
case of a partial application the various efficiencies would be about
equal .
All these applications are barred at present by the unsolved
problem of the preservation of the grates. In supposing that the
primary air is heated only to 400°, the economy to be realized
would be 6% ; this is not negligible in the case of continuous fur
naces, as those for illuminating gas.
The classification of furnaces for temperatures of about 1000 ° is,
then , as follows:
( 1 ) Furnaces with double recuperation by primary air and

secondary air, or with injection of waste products in the producer
( New Siemens furnace ).
( 2) Mixed gas furnace with single recuperation .
(3) Siemens gas furnace with single recuperation .
(4) Direct- fired furnaces.
We may finally remark that, for the industries using furnaces
with the ruling temperature of 1000°, with direct firing, it would be
a decided advantage to replace them by gas- fired furnaces, for
there is thus an economy of 30% certain of realization . There
remain still today many such industries in which the introduction
of gas firing would result in increased fuel efficiencies. It is hard
to understand because of such certainty why direct- fired furnaces
are still in operation , unless it be that it is more difficult to find
competent men to operate gas furnaces than men skilled in the
running of direct -fired furnaces.
For low temperatures, especially for steam boilers, there is little
to expect from gas recuperation ; the progress to be made lies in
the more complete combustion of the fuel, without a great excess
of air.
The progress to be realized in industrial heating theoretically is
not enormous. It means a saving of from 10 to 15 % of the fuel
in increased efficiency over the present best practice. This, how
ever, is worthy of serious cinsideration under modern industrial
conditions.
The successful introduction of gas firing depends largely on the

appreciation from the industrial world of the technical problems
involved in the design of gas furnaces and in the knowledge of
the science of their practical operation .
The preceding theoretical study has shown the maximum efficien
cies that could be expected from the various systems. The experi
mental testing of these efficiencies will be treated in the following
chapters.
CHAPTER IX.
EXPERIMENTAL STUDY OF THE HEAT UTILIZATION

IN FURNACES .
The experimental determination of the heat efficiency of a

furnace should be obtained by following the method pursued
in the theoretical study which gave the general classification
of all systems of recuperative heating. The starting point of any
such investigation will evidently be the exact determination of the
available calorific power of the fuel. This will serve as a term
of comparison for all the quantities of heat lost and utilized in the
various organs of the furnaces. In respect to these thermal quan
tities, we will be able to make their determination when we know
the chemical composition and the temperature of the gases at the
various points of division of the furnaces.
If the scientific data which have enabled us to conduct our theore
tical examination are recent, still more so is the apparatus that may
be used for the accurate experimental determination . The Mahler
calorimeter which has rendered practical the Berthelot calorimetric
method was not brought out before 1892, and the thermo-electric
Le Chatelier pyrometer dates from 1889.
Heat Balance Sheets of Furnaces. The experimental determina
tion of the heat balance -sheet of a furnace has for its purpose
the estimation of the ratio of the heat utilized in the hearth to
the total available heat ; also the values of the various heat losses;
and, finally, the quantity of heat recuperated.
The value of a furnace can be determined with accuracy only
through such tests, and if the efficiency is found wanting it will
be possible to find the reason therefor.. They are, for furnaces,
the equivalent of the complete tests which are made so frequently
and with such good results on steam engines. Nevertheless,
very few have been the number of complete tests made on furnaces,
hence it is difficult to make a well based comparison between the

various systems of heat recuperation and the different types of
furnaces suited to a given industry. This lack of experimental
data is detrimental to progress . The example set by steam engin
eers should be followed by the men who have charge of gas -fired
furnaces .
It will be seen in Part II that the apparatus needed for the
experimental determination of heat balance -sheets is simple as well
as practical; we will show , by a typical example, that the work
presents no special difficulties. The selected case is one of the
more complex examples offered by works — that of a Siemens fur
nace fired with mixed gas, with double recuperation by air and gas.
Experimental Methods. - The determination of a heat-balance
sheet requires the following:
( 1 ) Heat utilized .
(2 ) Heat carried away by the waste products .
(3) Heat lost in the gas- producer by conductivity or radiation .
( 4) Heat lost by the radiation of the chambers.
The sum of these four quantities of heat , taking our definition

of the limits of the various organs, must be equal to the total
available heat - that is , to the calorific power of the fuel . It
follows that one of these four quantities can be determined by
difference when the other three quantities are known; we will thus
be able to make an accurate determination of the heat lost by the
radiation of the recuperative chambers . This heat lost escapes
any direct determination on account of the double action suffered
by the traversing gases , heating or cooling by the checker work and
cooling by the enclosing walls. A quantity of heat useful to de
termine is the amount of the recuperated heat, though it does not
enter directly into the heat balance -sheet. It can be measured
with accuracy in function of the temperatures and composition
of the gases. This will give us the only exact means of comparison
between the various systems of recuperation.
If it is desired to obtain an absolutely exact test, it is necessary
to weigh exactly the amount of coal, steam and water used in the
FURNACE TESTS . 131
producers, beside making the usual gas analyses and heat de

terminations. It will then be possible to control synthetically
the gas analyses . The test in this case must be made to cover
a sufficient length of time , so that the weights of the coal and
water may be accurate ; the samples of gases are to be taken often
enough to represent a fair average during the total length of the
experimental period of time. 1 The samples of the coal used
for the ultimate analysis and for the determination of the cal
orific power are to be taken from the fuel as it is charged in the
producers. Twenty -four hours at least are required, and better
36 in the case of large producers .
The direct measure of the steam and water used is, however, not
necessary for most tests . The determinations based on the temper
ature measurements , gas and coal analyses and determination
of the calorific power of the coal are sufficient, if careful weights are
taken of the fuel burned for 36 hours, and of the ashes produced
during a corresponding period.
Experimental Data Required for the Calculation of the Heat Bal
ance - Sheet of a Siemens Furnace with Double Recuperation, Fired
with Mixed Gas.—The experimental data required are the
following:
( 1 ) Calorific power of the fuel .

(2) Elementary analysis of the fuel and ash .
(3) Analysis of the gaseous currents.
(4) Temperature determinations at the various portions of the
furnace .
Ultimate Analysis and Calorific Power of the Coal. These de

terminations are made on representative samples of the fuel , the
sampling to be done with great care, as very often the large pieces
of coal have a different composition from the fine.
1 It is by this method that we have made the test on a glass tank furnace , the results
of which have been published in the Annales des Mines. The test was carried out for
36 hours, during which the weights of the coal and water used were taken ; we were able
to arrive directly at a figure which had been omitted in the chemical analyses. The exact
measurement of the water corresponding to the amount of steam injected in the producer
may prove difficult . The exact determination of the coal used is also almost impossible
for a short period on account of the variations in the level of the producers. It is for these
various reasons that a 24- ( or better , a 36-) hour test is required.
The coal burned in our test had the following composition :

Carbon .. 82.50 %
Hydrogen . 5.00
Nitrogen . 1. 20
Oxygen .. 6.00
Ash.. 4.10
Hygroscopic water 1.20
Total. 100.00 %
Calorific Power ... 8210 calories.
As the combustion of the coal is always incomplete, it is necessary

to weigh the amount of ash produced, to sample it and to make
carbon and water determinations . It must not be forgotten
that the ashes to be collected must come from the coal the sample
of which has been taken . Thus, before the collecting of the ash
is begun , sufficient time must be allowed for the fuel charged in
the producer to have come down in the ash -pit as ashes. The
time to be allowed depends evidently on the size of the producer
and the method of running it.
In our test the figures for the coal consumption and ash pro
duction were the following :
Coal consumption (36 hours) . 14500 kg.
Ashes 3500
Analysis of the Ashes :

Water .. 28 %
Carbon . 40
Ash . 32
Total. 100 %
The coal contained 14500 X 82.5 = 11960 kg. of carbon .
The ashes contained 3500 x 40 1400 kg. of carbon .
140
The carbon of the coal escaping combustion is then of the
1196
total carbon of the coal.
We may write the composition of the coal as follows:
Burned carbon . , 728
Unburned carbon . 97
Hydrogen .. 50
Nitrogen . 12
Oxygen . 60
Ash . 41
Hygroscopic water . 12
Total . 1000
1 The figures here used are taken from experimental studies, however ; they do not
all come from one such test. We intend to give here a numerical example to be used as
a reference for future tests. It is not an actual furnace test.
FURNACE TESTS . 133
The calorific power is to be decreased by the heat of combustion

of the carbon unburned , which is equal to
97 X 97.6
= 788 calories .
12
Calorific power of coal = 8210 - 788 = 7422 calories; cal

orific power per burned molecule ( 12 grams) of carbon of the coal:
7422 X 12
122.35 calories .
728
Thus, the combustion of one molecule ( 12 grams) of carbon of

the fuel will correspond to a liberation of 122.35 calories, coming
from both the carbon and the hydrogen of the fuel. It is useful
to thus compare to the burned molecule of carbon not only the
total available heat, but also all the quantities of heat reacting
in the furnace and all the gas analyses. The carbon is introduced
in the furnace only by the combustion of the fuel (if we neglect
the atmospheric carbon dioxide), so it is this element which can
best serve as a unit of comparison. 2 The determination of the
carbon in a gas will allow the estimation of the weight of the fuel
corresponding to a given volume of the gas, and thereby the cal
culation of the ratio of the calorific power of the gas to the total
available calorific power of the fuel ; also , since we have used the
1
thermal capacities per molecular volume, it is , then , the molecule

which is most convenient to use .
Gas Analyses. — Two analyses are to be made—that of the pro
ducer- gas and of the waste products.
The samples taken must represent a fair average, and their
multiplication can but be beneficial to accuracy. The waste pro
ducts are to be collected at corresponding times of two periods
of inversion; if the inversions are made every half-hour, the sam
pling will be made hourly. The sampling of the producer -gas will
depend on the charging, as the composition of the gas depends
788 X 100
1 A loss representing 9.60% of the total available calories.
8210
: If carbonaceous matter is charged in the hearth , as in glass or cement furnaces, it is

evidently necessary to make allowance for this extra supply of carbon by following the
method used the case of the ashes of the producers.
on the time elapsed since the last charge, and the hydrocarbons
are in larger amount soon after the charging. The sampling
should then cover the time of a full cycle of a charging of the pro
ducers. The collecting tube should evidently be placed on the
flue common to all the producers carrying the gas to the furnace,
and not located on the flue of a single producer.
The analyses thus made in our test are as follows:
VOLUMETRIC ANALYSIS OF THE PRODUCER -GAS.
CO2 6.10 %
CO 20.30
н. 13.20
CH, 2.70
N2 57.70
Total 100.00 %
Water vapor .. 2.50 %
The gas composition is to be compared with the molecule of

carbon , the molecular volume is taken as the unit of volume
in all the calculations that follow .
The figures for the carbon dioxide, carbon monoxide and meth
ane correspond to a number of carbon molecules equal to their
volume. Therefore the 100 volumes of dry gas or the 102.5 volumes
of wet gas correspond to a number of burned carbon molecules:
6.1 + 20.3 + 2.7 29.1 molecules

CO , CO CH ,
each molecule having liberated 122.35 calories .

VOLUMETRIC ANALYSIS OF THE PRODUCTS OF COMBUSTION .
CO2 14.30 %
N2 84.00
02 1.70
Total 100.00 %
Water vapor . 8.70% 108.70%
In this case , the composition of the products of combustion

corresponds to 14.3 molecules of burned carbon , each having
also liberated 122.35 calories.
Measure of the Temperatures. The temperatures should be
vaken at the following points: At the entrance of the producer
FURNACE TESTS . 135
gas in the recuperative chambers, and as near the inversion valves

as possible.
We assume this temperature to be 600° at the admission ports
in the furnace of both the air and gas. Several determinations
must be made, since the temperature varies from the beginning
of an inversion to its end. Even in furnaces with parallel counter
currents , where there are a number of ports (as is the case in muffle
furnaces for zinc distillation or for the manufacture of illuminating
gas, and also in glass tank furnaces) the temperatures are not always
uniform , and it is well to take readings at several ports. In this
measurement it is absolutely necessary to introduce the pyrometric
cane well inside the ports, so that the thermo-couple junction
is well in the outgoing gas current. It must be far enough inside
to be protected against the heat of the hearth . The cane must be
kept in position until the needle of the galvanometer has come
to rest , indicating that an equilibrium of temperature has been
reached .
We have assumed these temperatures to be 1200° for both the
air and gas.
In the hearth , it is well to make several determinations, as far
as possible during the time allotted between two inversions.
We assume this temperature to be 1500°.
The temperatures at the base of the stack must be taken at fre
quent intervals and as close to the inversion valves as practicable.
The cane is placed in the flue collecting the waste products issuing
from both chambers, and an average temperature of the two cur
rents is thus obtained . Great care is to be taken in this determina
tion ; to avoid error resulting from cold air rushing along the sides
of the pipe on account of the stack draught, the cane should be
thoroughly luted in .
It is seen that often a large number of temperatures are to be
taken in a very limited time, the usual length of an inversion period
being but 20 to 30 minutes. It has been found useful to place in
position one thermo-couple at the gas flue, one at the stack , and a
third one for the temperature determinations in the ports and
hearth . The three couples are placed in electrical connection with
the galvanometer by means of a switch -board, and it is thus an easy
matter to make a large number of determinations in a compara

tively short time.
The same method is to be followed in taking gas samples ; as
many samples as possible should be taken in order to obtain a
representative average.
The man in charge of the test has all he can do to attend to the
pyrometric measurements, so he should have helpers on whom he
can rely to take the necessary gas samples.
We assume the temperature of the products of combustion to
be 400 °.
It cannot be insisted upon too strongly that great care must be
exercised if accurate results are to be obtained . '
CALCULATION OF THE EFFICIENCY OF THE FURNACE .
We have 122.3 calories as the total available number of calories
per burned molecule of carbon . We will find the total heat carried
by the gas as it leaves the producer at 600 °, for a volume of gas cor
responding to one molecule ( 12 grams) of burned carbon .
The percentage of the calories lost in the gas producer can be
obtained at once :
CALCULATION OF THE CALORIFIC POWER OF PRODUCER Gas at 600 °
Sensible Heat at 600 ° Latent Heat
Gas %
:
Per Per
Molecular In the Gas Molecular in the Gas
Volume Volume
CO2 6.10 6.44 39.28

CO 20.30 4.31 87.50 68.2 1384.5
H 13.20 4.31 56.90 58.2 768.3
CH 2.70 8.02 21.65 195.2 527.0
57.70 4.31 248.69
H20 2.50 5.87 14.67
468.69 2679.8
Total heat 3148.5 calories

3148.5
Per molecule of burned carbon 29.1
= 108.2 calories.
In the case of furnaces with parallel counter-current recuperation it is well to test

the leakageofair through cracks and breaks in therecuperators by taking samples of
the waste products as they leave the hearth , before they enter the recuperator, and to do
the same thing in the fue leading to the stack. It has come within the experience of the
writer that a counter-current recuperative furnace appeared to give a good efficiency
owing to a low stack temperature, due merely to considerable air leaks, which lowered
the temperature of the waste products on their way to the stack . - Translator .
FURNACE TESTS . 137
122.35 108.2 14.15

11.50 %
122.35 122.35
This represents the calories lost by radiation, convection and by

the premature combustion of the carbon to carbon dioxide. 11
We will now calculate the heat utilized in the hearth , by finding
first the total calories brought by the air and gas at the temperature
of 1200 °, and then the total heat carried away at 1500 ° by the
products of combustion.
The calorific power of the gas at 1200 ° is calculated exactly in
the same manner as has been shown for the gas at 600° . It was
found to be 3689.41 calories, or, per molecule of burned carbon :
3689.4 1
= 126.79 calories .
29.1
Volume of Air Injected in the Chambers.-- This volume, taking into

account the excess of air represented by the free oxygen, is derived
from the analysis of the gas and that of the waste products.
20.3
13.2
volumes of H
CO2 require 10.15
6.60
volumes 0 2, ++ 25.13
38.65 of N ,
2.7
00 00
CẢ , 10
5.40
..
+ 20.56
Excess of air . 3.46 13.17
Total volume of air . 25.61 97.51
123.12
The sensible heat brought by the air at 1200° is 1114.2 calories,

1114.2
or, per molecule of burned carbon , 38.29 calories.
29.1
Total calorific power of the gas and air :

3689.41 + 1114.2 4803.61 calories.
4803.61
Per molecule of burned carbon 165.08 ,
29.1
a number greatly superior to that of the fuel; the difference be

1 The true loss in the producer ( which , however , has no bearing on the efficiency of the
other furnace organs) is : 9.60 + 11.50 = 21.10 % . To this should be added the coal
burned on the boiler grate to generate the steam required for the injector when used. A
total loss of25 to 30 % can be considered a fair figure as the total absolute loss in con
verting the coal into gas.- Translator.
tween this total calorific power and that of the gas at its leaving
the producer is due to recuperation.
165.08 108.2 56.88
Recuperative Effect =
46.50 %
122.35 122.35
The 56.88 calories are recovered from the products of combustion

that otherwise would have carried them to the atmosphere.
Calories Contained in the Products of Combustion at 1500 °.-- This
quantity of heat is found as follows :
Gas
Thermal Capacities per Total
% Molecular Volume at 1500 °
CO2 14.3 21.0 300.3

02 1.7
84.0
11.50 985.55
N2
H20 . 8.7 18.50 160.95
Total 1446.8 calories .
1446.8
Per molecule of burned carbon 101.18 calories .
14.3
According to our definition of Heat Utilized, the difference be

tween the heat brought in the hearth and the heat carried away
by the waste products gives the amount of the heat utilized :
165.08 – 101.18 63.9 calories.
63.9
Percentage of the Heat Utilized 51.23 %
122.35
Calories carried away by the waste products into the atmosphere:

Gas % Thermal Capacities per Total
Molecular Volume at 400 °
CO2 11.3 3.99 57 06

171
84.01
2.82 241.67
HO 8.7 3.69 32.10
330.83
FURNACE TESTS . 139
330.83
Per molecule of burned carbon 23.13 calories.
14.3
23.13
Percentage of heat lost 18.91 %
122.35
We have, then, the following heat balance -sheet:

Heat lost in gas-producer . 11 50 %
Heat utilized ( * Ü * ) . 52.23
Heat lost in the waste products. 18.91
Heat lost by cooiing of the recuperative chambers (by
difference ).. 17.36
Total .. 100.00 %
CHAPTER X.
DISCUSSION OF THE THEORY OF HEAT RECUPERATION IN THE LIGHT
OF EXPERIMENTAL Data — GENERAL CONCLUSIONS
AND CLASSIFICATION .
The general study we have given shows that there may be a

marked difference between the theoretical and the actual effi
ciencies. Thus our double recuperative furnaces fired with mixed
gas had an efficiency of 52.23% against a theoretical efficiency
of 85 %
The calories lost by radiation in the producer and in the chambers ;
the premature combustion of the carbon in the producer, and in
general all the practical conditions in contradiction to our theo
retical study) , had a marked influence on the efficiency.
In order to preserve the correctness and exactness of the theory,
as well as of the conclusions made therefrom , we must again take
up our hypotheses and show that, while modifying the absolute
values of the figures of the calculated efficiencies, they have little
bearing on the relative figures; at least, they cannot change the
Order of the classification .
Before taking up the general discussion, let us remark that
the maximum efficiencies given in Table 11 are limits toward which
the efficiencies of the furnaces will tend with the advance in furnace
construction .
If the classification is not mathematically correct today, it will
tend to become more and more so in the future ; on this account,
alone, the classification is worthy of interest.
We will show, however, that the order of the classification can
not be changed, with the understanding that the furnaces be built
under similar conditions as regards the grouping of the various
organs of the apparatus and of the cooling surfaces . '
1 It is evident that two furnaces huilt along widely different lines may from this very
reason show anomalous efficiencies. Thus, a well -grouped furnace with single recuperation
may have a better efficiency than a double recuperative furnace with its cbamberg id
THE THEORY OF HEAT RECUPERATION . 141
Our first hypothesis ?-complete fuel gasification in carbon monox

ide — is far from being realized . There is almost always about
—
5% of CO , with 20 to 22% of CO. This premature combustion of
the carbon lowers the efficiency by the increase it brings in the
primary air and in the gas temperature; both results are harmful to
good recuperation . Neglecting the slightly increased radiation in
the producer resulting from the higher gas temperature, it is easily
seen that this premature combustion cannot modify the classifica
tion .
It is permissible , in fact, to consider furnaces fired with gases
containing 5% of carbon dioxide as one- fifth direct - fired furnaces,
and the remaining four- fifths as gas-fired furnaces with the efficiencies
we have given for the two systems. The differences between the
efficiencies of the various systemswill be modified in the ratio 4 : 5
without altering their classified order.
The second hypothesis concerns the invariableness of the ruling
temperature throughout the furnace — in the producer, the hearth, etc.
In furnaces working continuously, with Siemens inversion , the
temperature of the chambers will oscillate about 100 ° between the
beginning and the end of an inversion , according to the time occur
ring between two consecutive inversions. These differences affect
only the sensible heat of the gases, and are small enough to be
negligible. With parallel counter -current furnaces the invariable
ness is maintained .
The third hypothesis, in which we omit the cooling of the gas
producer, has a greater importance. The heat loss has been found
to be 11.50 %
For the same method of gasification and the same type of gas
producers, and operated in a like manner, this thermal loss will
be a constant which will not affect our classification . If the change
is made from Siemens gas to mixed gas, the cooling will be less in
the second case than in the first, and it will give a further advan
tage to the mixed gas. This will serve only to increase the differ
sufficiently protected and not properly grouped for the prevention of radiation losses.
It will also be possible to have a better efficiency with a New Siemens furnace than with
a furnace with double recuperation with mixed gas, if the cooling surfaces of the latter
are considerable . This would not , in any case , invalidate our theory or modify our con
clusions.
2 See page 67 for the hypotheses, which have been taken as the basis of the general
theory .
en ce indicated by the theory in the better recuperation and in the

accentuation of the classification .
The two hypotheses, which are also the least in accord with
practice, remain - that of the neglect of the cooling of the recuperative
chambers, and that which admits as possible a perfect exchange of
calories between the cold and the hot gases .
It has been found in our experimental test that the cooling of
the chambers may reach 20% , and the gap between the tempera
tures of the waste products and the recuperated gases may be 300 °
for a temperature of 1500 °. Finally, in spite of this difference,
which reduces more than one- fifth the efficiency of the recupera
tion, the recuperated heat may be from 40 to 50% .
These figures have weight and should be considered . In order
to appreciate their influence on the efficiency we may note that
both result from the nature of the recuperative organs, and if we
show besides that they are functions of recuperation itself - that
is, of the calories won from the waste products—we will have shown
that their effect on the efficiency is proportional to the various
efficiencies themselves, and therefore cannot modify the clas
sification .
The volume, or more accurately the useful surfaces, of the re
cuperative chambers should be calculated to suit the exchange
of the calories to be effected , according to the specific heats of the
recuperating refractory bricks . These chambers will be the larger,
and thereby their radiation will be the greater, the more efficient
the recuperation . There will then be a certain relation between
the loss by the cooling of the chambers and the recuperated heat.
The difference between the final temperatures of the recuper
ating gases and the initial temperature of the waste products will
have the same relative importance. This loss is incurred only by
the gases taking an active part in the recuperation, and the loss of
calories resulting therefrom will be the greater the larger the mass
of gases flowing through the recuperators , and therefore it increases
with the bettering of the efficiency.
This proportionality is exact if air is the only recuperating gas.
In the case of producer -gas recuperation , owing to its high initial
temperature, the radiation loss and the recuperative loss will in
THE THFORY OF HEAT RECUPERATION , 143
crease with the initial temperature of the gas. It follows that the
mixed gas will again be more efficient than the gas of any other
system . Our classification is thus made only more marked in its
divisions .
We have finally to appreciate the change that may be brought
in the classification by the nature of the fuel . We made the hy
pothesis that the fuel was an ashless coke, containing nothing
but carbon. It is evident that we might have repeated our var
ious calculations with different fuels, hydrogen , methane, etc .; but
these substances exist only in small quantities in the coals, and
such calculations would present only an academic interest. It will
be sufficient to note that the presence of these bodies , which en
rich in so marked a degree the various fuels, will counteract in a
way the premature combustion in the producer, and therefore
will tend to bring closer the actual and the theoretical classification
without changing the relative positions.
It is permissible now to affirm that the classification of the
various systems of furnaces,, based on the calculated theoretical
efficiencies, is in accordance with practice , and can serve as a
sure guide in the study and selection of the type of furnace to be
used for a given set of conditions.
Doubtless , furnaces may be found in practice that form an
apparent exception to these laws ; for example, a poorly designed
furnace with double recuperation may consume more coal for a
given work than a good single recuperative furnace; but from
such a fact it is impossible, and certainly not legitimate, to draw
any conclusion asserting the superiority of the second system
over the first.
The only objections worthy of consideration are those based on a
complete heat halance-sheet , which alone can allow a logical criti
cism of a heating system . We do not believe that up to date there
has been a sufficient number of such tests made to control ex
perimentally the theory of heat recuperation. To multiply such
tests is vital to the progress of the art . They would constitute
an ensemble of experimental facts , which , taken together with
the undeniable theoretical figures , would make a proper judgment
of the furnace question possible.
In the absence of experimental data, the safer guide will be

to rely on the theoretical efficiencies so much the more since the
discussion has only confirmed their value.
We believe that we may safely conclude :
( 1 ) In maintaining a new practical value of the table of ef
ficiencies, taken with the reservation we have given.
(2) In asserting once more the two results that appear to us of
the greatest interest: the actual superiority of furnaces with double
recuperation with mixed gas in the case of high temperatures;
the importance of improvement of gas -producers for the injection
of highly heated primary air, which alone will allow of a simpler
recuperating apparatus with an efficiency higher than any in
practical use today .
The first of these conclusions concerns the present condition of
the art. The second is an indication for the future which we
hope may serve to show the desiderata yet to be realized . Though
the theory of the maximum efficiency is established in a definite
and permanent manner , the industrial efficiency is still subject to
such variations that the classification given in our study may be
modified in the future according to the never -ceasing progress in the
science of heating. 1
1 We insist, forcibly , on the necessity of complete heat balance -sheets, with records
of gas analyses and temperatures. Furnace builders seldom pay any attention to them.
The only datum that interests them is the ratioof the weight offuel burned to that of
the metal treated. This , evidently , is of great importance to works managers , but we
maintain that it is insufficient to estimate the value of a furnace, since it is not given to
gether with the length of the operation and the dimensions of the furnaces.
PART II
METHODS OF CONTROL
PYROMETRY . GAS ANALYSIS
BY
A. L. J. QUENEAU
CALORIMETRY . ELEMENTARY
FUEL ANALYSIS
BY
MYRICK N BOLLES
1
1
LIB
RAR
UNIV
OFER
THE
Y
SITY
CAIFO
RNIA
PART II .
CONSIDERATION OF PRACTICAL PROBLEMS.

CHAPTER XI .
PYROMETRY .
From the preceding chapters we have seen that it is necessary ,

in making furnace tests , to be able to measure rapidly and ac
curately the temperatures in the various parts of the furnaces. The
instrument to be selected for the work should therefore be portable,
accurate, easy of handling, and should allow the reading of tem
peratures at widely distant points .
Of the many types of instruments the Le Chatelier thermo
electric pyrometer offers all of the mentioned desiderata. The
thermo -electric instrument being small can be introduced into
any part of the furnaces, and since a few seconds are sufficient
for the establishing of an equilibrium of temperatures, it can serve
to measure temperatures close to 1780° C. , the melting point of
pure platinum , without endangering the life of the wires. On
account of the electric current generated, any number of thermo
couples can be connected with a single galvanometer by means of
a switch -board; for the same reason the instrument may be placed
at any convenient distance from the furnace .
The Le Chatelier pyrometer is thoroughly reliable, if ordinary care
is taken of the wires. However, in order to be successful with the
instrument, its limitations should always be borne in mind ; these
are due entirely to the nature of the metals forming the couple.
Practically the same care that is to be taken with platinum ware
is required with the wires of the thermo- couples. Platinum is
readily attacked by the vapors of volatile metals; silver, zinc , an
1
timony and copper are especially to be dreaded. In a reducing

atmosphere, silicates and phosphates are also objectionable.
In spite of the most thorough protection, platinum and its alloys
suffer physical alterations, resulting in great brittleness, by mere
continuous heating to temperatures above 1000 °. Several types
of galvanometers are made, giving good satisfaction , especially
Millirolu 100 Roo
8. Ho
Iro
Co
ns C
ntons
n ta
Co an ta
pp nta
er , n
Si
lv
Ag
pr
er
25,
75
%
%
Pl
8P S
at
in
um
90%
1,0
%
800
80
1000
Fig . 43.-E. M. F. CURVES OF DIFFERENT THERMO -COUPLES.
the Siemens and Halske, the Pellin, and that of the Cambridge
Scientific Instrument Company.
The Le Chatelier pyrometer, though very widely used, has
not yet received the universal approval that it deserves, probably
from lack of knowledge as to its practicality.
For the same reason great stress has been unduly placed on the
importance of the composition of the thermo-couples. The truth
PYROMETRY . 149
is that there exists a large number of thermo- couples, each with

its own field, that give good satisfaction . With platinum couples
>
the question of purity is important, especially with the platinum

wire, as a slight amount of impurity will cut down the electro
motive force much more quickly than a corresponding amount
in an alloy wire.
Among the couples used by the writer are the following:
( 1 ) Pure platinum and an alloy of 10% rhodium and 90 %

platinum.
(2) Pure platinum and an alloy of 10% iridium and 90 % platinum .
( 3) Pure silver and an alloy of 25% nickel and 75% copper
( constantan) .
(4) Pure copper and constantan .
( 5) Pure silver and an alloy of 25% platinum and 75% silver.
(6) Pure iridium and an alloy of 10% ruthenium and 90 %
iridium .
Iridium and rhodium give equally good results, with the material
advantage for the iridium alloy that it is much cheaper, that it
can be readily obtained from any platinum refiner in sufficient
purity for industrial work, and that it has a higher E. M. F. at
temperatures above 1000 ° C. than the rhodium alloy.
The ruthenium -iridium couple is useful only for high temperature
measurements , those above the melting point of platinum . It is
extremely brittle, but can be used up to 2100 ° C.
The silver and platinum -silver couple is very convenient for
temperatures up to the melting point of silver ( 962 ° C.) . This couple
costs much less than the platinum element, and has nearly double
its E. M. F.
The constantan couples are very useful for low temperatures
from 0° C. to 300° C. They have very high E. M. F. , with the
further advantage of being cheap.
It follows that we have at hand a number of couples from which
we may select the one most suited to the work to be performed .
The main point to be remembered is that, owing to the necessary
standardization , we may use any couple giving concordant results.
It is only required to know the law of the E. M. F's. to be able

to use the couple with accuracy.
Principle of the Thermo- Electric Pyrometer.- Whenever two
metals in contact are heated , an electric current is generated which
is a function of the temperature . In a circuit composed of a
number of metals, the resulting current is equal to the algebraic
sum of the various component currents .
The E. M. F. of a couple is not altered when one or several metals
are inserted between the metals of the couple. It follows that
any metal can be used as solder; from this principle is derived the
wire method of standardization of Holborn and Day (see
p . 159) .
On account of the importance a minute quantity of impurity
plays in the pure metal used in the thermo- couple, it follows that
no two couples made from different ingots have exactly the same
E. M. F. It is thus a useless refinement to provide the galva
nometer with a temperature scale ; the use of that scale is limited
to the use of the couple sold with the instrument, and does not
entirely suffice for that, since the couple will change its E. M. F.
after hard usage, requiring restandardization .
When couples are to be used in large number, it is good policy
to order a length of wire made from a single ingot, long enough
for one year's supply. The wire lot should be sampled at both ends
and at the middle, and all samples standardized . If all the sam
ples give the same results, the wire lot may be considered as homo
geneous, and no further standardization is necessary.
For testing furnaces of temperatures ranging to 1600 ° C. and
above, the iridio -platinum is to be used. Before making the
couples, the wires should be annealed to a temperature equal to ,
or even higher than, that at which they will be exposed . The
annealing is best done by an electric current, though it can also
be done with the oxygen lamp.
The junction should be made by twisting the wires together and
melting the twist carefully with an oxygen blast until a globule is
formed. Care should be taken to avoid twisting and kinking the
wires after the annealing. Even in perfectly homogeneous metals ,
a distorted section will , when heated, give rise to an electric current ,
PYROMETRY . 151
which of course will affect, in greater or lesser measure, the read

ings of the galvanometer.
It must be borne in mind that the wires will not stand heating
over 1000 ° C. for more than a few hours without serious change in
their physical conditions , even with the best of protection against
the contact of silicious material and surrounded by an oxidizing
atmosphere . The writer has had wires in spelter furnaces for 48
hours continuously, the temperature reaching 1500 ° C. , but they
gave correct readings as long as they were left undisturbed ; the least
jar was sufficient, on account of the induced extreme brittleness,
to make them fall to pieces . The effect seems in this instance to
have been purely physical, since the couples gave correct deflection
to the last. This was ascertained by having a fresh standard
couple inserted at the time of the breaking down . The brittle
ness is specially marked with the platinum alloys; the pure pla
tinum wire will still be in fair condition when the platinum -iridium
wire has completely failed.
With proper care the Le Chatelier thermo -couple can be used
for a large number of tests. With the body of the couple properly
protected, the length of the couple exposed need be only 3 cm.
In such a case , the thermo- couple introduced into a furnace, say
at 1600 ° C. , will reach a condition of equilibrium in about five
seconds, often in three ; it is then possible to measure very high
temperatures without danger of losing the wires , or of injuring
them .
A common false economy, in the case of the Le Chatelier thermo
couples, is to make use of wires of too small a diameter; usually
the wires sold are 0.6 mm . in diameter. One millimeter is a better
diameter, the increased cost being fully offset by the much greater
resistance of the wires to corrosion and mechanical rupture.
The cost of the Le Chatelier unstandardized iridio-platinum
couples is about 75c . a gram ; the scrap wire can be sold for about 60c '.
If constantan or silver wires are used , the cost is nominal. The
writer knows of over one hundred couples put in furnaces, several
years ago, for continuous record of temperatures below 1000 ° C.,
which are as good today as the first day they were put in place.
1 Varying with the metal market.
PYROME
ETHERMO
CHATEL
LE
OF
-WIRES
THE TER
C
. LECTRIIER
FOR
CANE
P
– YROMETRIC
4FIG
. 4. PROTEC TION
PYROMETRY . 153
Outside of the advantages that may result from the knowledge

of the temperatures, the use of registering pyrometers always tends
to insure regularity of operations. The men know very soon
that there is in operation an instrument that records automatically
the charging, tapping, reversing of valves, etc. , and the moral effect
is conducive to good work ; very soon they will recognize in the
instrument a valuable help in the correct operation of the fur
naces. Once the standard basis of good work has been established
in relation to some automatically recorded factor, it may be good
policy in some cases to give a premium for increased efficiency.
In blast furnaces, it is good practice to have recording pyrometers
on the bustle-pipe and on the down -comer (and on the equalizer if
one is used ), besides having a direct-reading instrument connected
with the various ovens or stoves . The couple on the down-comer
gives valuable information as to the running of the furnace, the
position of the stock line, the time of charging, etc.
The continuous record of the temperature of the waste gases at
their leaving the chambers may also give interesting information
regarding the regularity of operation of regenerative furnaces.
Another use of the pyrometer, outside of determining exact
furnace temperature, is as a means of control of operations in the
furnace, especially in connection with a recording galvanometer. If
a fixed temperature relation exists between the walls of the fur
nace and those of the hearth , the thermo-couple will give just as
useful information , placed in the wall, flush with the inner face,
as if it were in the center of the hearth . The relation between the
two points is easily ascertained by inserting a thermo-couple in
the selected part of the hearth and comparing the readings given
by the galvanometer when connected with either of the couples .
As an instance of this method the writer installed thermo-couples
in the walls of a number of blende -roasting kilns, the end of the
firebrick protecting tube being flush with the wall of the kiln
( Fig. 51) .
If care is taken to have a complete spare thermo-couple always
at hand , it is a small matter to replace an injured instrument; all
there is to do is to make the lead connections. The thermo-couple
wires should always be soldered to the copper leads, and never
154
E H
o
Soome
CKils
ROASTING
INOUPLE
THERMO
OF
.FOR
PROTECTION
THE
--
45.
Fig
A PPARATUS
PYROMETRY. 155
connected by merely twisting them together. The galvanometer

may be located at any convenient place—a mile removed if neces
sary . Care must always be taken, in order to insure accuracy,
to standardize the couples with the same length of wire as that
used for the work . The necessary length of the copper wires may
be actually used, or an equivalent resistance.
It is to be remembered that the thermo-electric current gen
erated is proportional, not to the actual temperature of the thermo
junction, but to the difference in temperature between the hot and
the cold junctions. Care, therefore, should be taken to protect
the connections of the thermo-çouples with the copper leads ( the
cold junction ) from undue heating; in case this protection is not
possible, it becomes imperative to use a water -cooled end. This ,
however, is seldom necessary , as it is usually sufficient to make
correction for the temperature of the ends, which is about that of
the surrounding atmosphere."
STANDARDIZATION .
In order to find the temperature corresponding to a given de

flection in millivolts of the galvanometer, it is necessary to find
the number of millivolts corresponding to known temperatures.
Such temperatures are given by the melting and boiling points
of chemical elements or definite chemical compounds. Mathe
matically speaking, with the pure platinum and iridio- or rhodio
platinum couples, it is sufficient to determine two points; by
means of a logarithmic formula it becomes possible to find any
other temperature corresponding to a given deflection. How
ever, both for rapidity and safety it is better in industrial work
to use the graphic method . If, on a sheet of coördinate paper, the
temperatures are plotted on one axis , the millivolts on the other,
the intersections of the ordinates and abscissa will give a series
of points lying on a curve characteristic of the couple.
The fixed points to be selected are the following:
I The mercury rheostat of William Henry Bristol affords an ingenious and effective
method of compensating this source of error: it is formed of a coil of wire sealed in a ther
mometer, the mercury of which cuts off a greater or lesser length of the resistance with the
change in the ambient temperature .
2 S. W. Holman , Proc. Am . Academy Arts and Sciences, XXXI ( N. S. XXIII ) , p. 234,
Boiling points of Water, H, 100 ° C.

Naphthaline, CiH g . 218°
Sulphur, S .. 445 °
Freezing points of Zinc, Zn . 419°

Aluminum , 655°
Silver , Ag. 962 °
Gold , Au . 1065 °
Platinum , Pt . 1780 °
For determining the boiling points of water, use naphthaline

and sulphur, the apparatus of Barus (illustrated in Fig. 52) .
The bulb test -tube is inserted in a muff of plaster of paris, after
having been filled about one-third its lower length with the sub
stance to be used ; heat is carefully applied until ebullition has
begun . Care should be taken , especially with the sulphur and the
naphthaline, to keep the level of the vapors somewhat below the
mouth of the tube, otherwise the vapors will catch fire. For these
determinations it is not usually necessary to have any insulation
around the wires in the test-tube ; it is sufficient to spread them
apart so that they will not touch each other and thereby short
circuit the current. From the test-tube to the galvanometer it
may or may not be necessary to protect them . The wires of the
thermo -couple may be led directly to the galvanometer or they may
be connected to copper leads. The junction of the copper wires
and of the couple wires may be inserted in a large -necked bottle
containing water, in order to keep the cold junction at a uniform
temperature ; in most instances, however, it is a useless re
finement, since there are quite a number of other junctions, name
ly , that of the copper leads with the binding -posts of the galva
nometers, and the different junctions in the body of the galvanom
eter, made in most instances of different metals. It is as
well for all practical work to connect the leads directly to the
binding -posts of the galvanometers and to note the temperature
of the room . The complete standardization of a couple can be
done in about two hours, and in that interval of time it is seldom
that the difference of temperature will be more than a few degrees;
the reading of the thermometer will give the means of making the
necessary connections. Care is to be taken to have the galvanom
eter sufficiently removed from the source of heat.
With the boiling points, the needle of the galvanometer will
PYROMETRY. 157
80
Og/
ZO
‫لور‬
joo
Fig . 46. - APPARATUS FOR THE DETERMINATION OF BOILING POINTS .
reach a stationary position when the thermo-couple junction

has reached an equilibrium of temperature with the vapors . The
junction should be placed well within the plaster muff, but not
inside of the liquid . It is always well to allow the liquid to cool,
and to re -heat it to the boiling point several times, in order to be
certain of the results.
When couples other than that of the platinum group are stand
ardized, it is well to protect the wires from the contact of the
sulphur vapors; in such a case , a covering can be made of hard glass,
as shown in Fig . 48.
MEASUREMENTS . )
f
Fig . 47. - APPARATUS FOR THE DETERMINATION OF THE BOILING POINT OF
Zinc. ( From LE CHATELIER - BOUDOUARD - BURGESS' High TEMPERATURE
For the determination of the melting points or freezing points

the wires should always be protected ; the best covering is afforded
by porcelain tubes, as shown in Fig. 49. About 35 to 50 grams
of the substance to be used is taken and placed in a small crucible ;
the No. 00 graphite crucibles are well suited for that work . For
zinc, aluminum, silver, and copper, the crucible method is to be
preferred, while for the precious metals, gold and platinum , the
wire method is the best ; it is also well adapted to silver and
aluminum
PYROMETRY . 159
Crucible Method . — The furnace to be used whenever possible

is the electric-resistance furnace, the resister being either nickel
or platinum wire or platinum foil , according to Heraus' method.
In Fig. 50 is illustrated such a furnace; the wires are either plati
num or nickel. With platinum wires , pure nickel has been melted
in a little over 30 minutes, starting with the furnace cold. The
great advantage of the resistance furnace is that a perfectly uniform
temperature can be obtained and maintained for any length of
time, and that no reducing flames are present. For the wire
method a tube furnace can be very well used. In Fig. 51 is given
the design of a furnace that can be built at almost any works ;
it is comparatively cheap and highly efficient. The resister can
ΛΙΑ .
GAS
Fig . 48. -FLETCHER FURNACE .
be made of nickel, platinum wire or ribbon . If an electric furnace

is not to be had , use can be made of the Fletcher furnace, using
either air or oxygen blast. If a Fletcher furnace cannot be pro
cured , a simple and effective furnace is made, as shown in Fig. 49,
by taking two crucibles fitting snugly one into the other. One
or two tuyere holes are cut in the outside crucible and the blast is
applied through the openings ; the apparatus answers well for
the melting of gold or copper.
When making determinations it is well to insert the porcelain
tubes at the beginning of the operation before the flame is applied,
so as to avoid the cracking of the tubes by too sudden a heating.
The heating should be conducted slowly and continued until the
-
+
Fig . 49. - SIMPLE APPARATUS FOR THE DETERMINATION OF THE MELTING AND
FREEZING PointS OF METALS .
PYROMETRY . 161
metal is melted ; this can be ascertained either by direct observa

tion by feeling the solid metal ingot with the porcelain tubes, or
by the deflection given by the galvanometer. It is not necessary
in the crucible method to watch for the melting point; once certain
that the metal is fused, the heating is stopped, the furnace is allowed
to cool slowly and regularly, and at the instant the mass has reached
the freezing temperature, a well -marked stop is observed on the
galvanometer needle. With proper care, and with gradual cooling,
it is very easy to have the needle stationary for 30 seconds with a
mass of 35 gr. It is always well to repeat the operation.
Wire Method .The wire method allows the determination of the
melting points of the precious metals with the use of only a very
small weight of the same . A 20 -mm . length of this wire is twisted
and soldered at the end of the wires of the thermo- couple, in such
a way as to form the thermo-junction. (See Fig. 51. ) Care is to
be taken that the wire is not under tension, otherwise the thermo
couple wires would break the softened wire some degrees before
the true melting point. The thermo-couple is inserted in a suit
able furnace, preferably an electric furnace, which is slowly heated
to the required temperature. The needle of the galvanometer
is abruptly released when the melting point is reached ; the max
imum reading corresponding to the sought temperature.
The determination of the melting point of platinum is con
veniently made at the time of the fusing of the thermo-junction.
A twist some 10 mm. in length is formed from the thermo-couple
wires, and the flame of the oxygen blowpipe is made to play at the
very end of the twist. The flame follows the bead of molten plat
inum in its travel toward the free wires; the fusion is stopped
when the latter are bathed in the bead. The melting point cor
responds to the steady maximum deflection. Care is to be taken
that the flame does not come in contact with either wire their
thermal capacities being less than that of the bead, they would be
quickly cut.
ELECTRIC FURNACES .
The electric furnaces suited to pyrometric work are limited to
resistance furnaces or induction furnaces . Direct current being
m
m
Fig. 50. — QUENEAU ELECTRIC RESISTANCE CRUCIBLE FURNACE .

PYROMETRY . 163
most easily obtained, we will limit ourselves to the description of

resistance furnaces .
The resistance for high -temperature work may be formed of car
bonaceous material, metals and their alloys, and, according to the
method of Nernst, of oxides of the rare earths. When the facility
of construction, the cheapness of apparatus, the constancy of the
resistance and the ease of getting the necessary materials are con
sidered, it is apparent that the choice will rest with the metallic
resisters. Not that they give in themselves perfect satisfaction ,
but if the furnaces are operated within the safe temperature limits,
the metallic resisters will prove satisfactory . We are evidently
limited to metals or alloys with high melting points and those not
readily oxidized. Thus we cannot go outside of the platinum group
of metals, although for temperatures below 500° C. it is possible
to use nickel for considerable periods of time.
According to the discovery of Dr. Heraus, of Hanau, the most
efficient form of metallic resistance is in the shape of thin foils;
they afford a very large radiating surface with a small weight, an
important consideration with a metal as expensive as platinum .
Broadly speaking, furnaces may be divided into crucibles and
tube furnaces. A type of crucible furnace which has given good
satisfaction is illustrated in Fig. 50. The furnace is enclosed in a
cast-iron casing, inside of which is a first layer of plates enclosing
in their turn an inner layer which form the hexagonal recess . On
top and on the bottom the cavity is similarly completed with a
double set of plates. The top , bottom and outside plates are made
of good fireclay, mixed with the proper amount of firebrick sand ,
while the thin inner plates are made of magnesite, thoroughly cal
cined at a temperature higher than the highest temperature to be
reached in the furnace. The calcined magnesite is mixed with a
binder formed of a thick solution of dextrine, sodium silicate and
some raw fireclay. The mold is provided with the necessary num
ber of wires to furnish the passages for the metallic resister. The
type illustrated has plates with an inner face of 51 mm . by 110 mm .,
each plate having 10 holes. The plates are wired separately and
mounted in series .
By selecting the proper cross-section of wire , knowing the re
164
sistance of the metal selected, it is possible to utilize the current

at hand. It is always well to try the length of the wire selected in
the open air; by sliding a movable binding -post it is possible to
check accurately the calculations. The walls and bottom of the
crucible chamber are made smooth by filling up all inequalities in
Fig. 51. - ELECTRIC RESISTANCE TUBE FURNACE , ILLUSTRATING THE WIRE

METHOD.
the surfaces by what is known as asbestos cement (a mixture of

shredded asbestos fibers and water - glass ).
With such a furnace the fusion of pure nickel is readily obtained ;
however, it is better adapted for temperatures up to 1300 ° C. as a
maximum . With a rheostat it is possible to maintain a temperature
Fig . 51a . - CROSS -SECTION OF THE TUBE FURNACE .
constant to within a few degrees. The furnace was designed for

the study of Stephan's law ; it gave the radiation of a black body.
The same type of refractory bricks , with perforations close to
one of the faces , forms a material well adapted to the making of
hot plates for laboratory works. By modifying the design to suit
PYROMETRY . 165
special conditions it is possible to use the plates for the tot-air

bath , sand plates, stills , combustion furnaces, etc.
A tube furnace is illustrated in Fig. 51. It is cheaply and easily
built , the only expense being the cost of the metallic resister. The
porcelain tubes which are commonly used as insulators in elec
trical wiring answer very well up to the highest temperatures
obtainable with platinum wires. The wire, preferably rolled flat,
or the Heraus foil, is wrapped along a helix of small pitch ( 7 in.)
as uniformly as possible. ( By tracing a helix first in pencil on the
white porcelain tubing, it is an easy matter to make an almost
perfect laying.) The ends being fastened by means of strings. a
HA
SIE S & LSKE
Fig . 52. SIEMENS-HALSKE DIRECT- READING GALVANOMETER FOR LE CHA

TELIER THERMO - ELECTRIC PYROMETER .
jacket of asbestos cement is placed by hand over the wires and

the cement allowed to dry. This drying operation may be effected
more rapidly by connecting the ends of the wires with the poles
of the source of electricity by means of a rheostat. The heating
should be done very gradually until the jacket is sufficiently hard
ened. The tube is placed in the center of the sheet - iron body, as
shown in Fig. 57, the space between the jacket and the sheet-iron
walls being packed tight with aa mixture of quicklime and charcoal ,
broken to hazelnut size . The lime will prevent the oxidation of
the charcoal, and the mixture forms a good heat insulator. The tube
is plugged by flat cylinders of graphite, cut from graphite brushes.
PYROMETRIC GALVANOMETERS.
The instrument should be " deadbeat " ; that is, the oscillations
of the needle should stop dead after a few vibrations. If the instru
ment is to be located permanently, the support should be made of
a concrete or brick pier with independent foundations, so as to
have as little vibration as possible. If vibrations cannot be alto
LE
PU . PECUN , PARIS
FiG 53. -PELLIN REGISTERING GALVANOMETER FOR LE CHATELIER THER

MO -ELECTRIC PYROMETER.
gether eliminated , it is well to place a deadening pad of rubber

between the coping and the pier. In order that the instrument give
good readings, it must be perfectly level. It is well to check the
level with a new pier, especially if there is any chance of settlement.
PYROMETRY . 167
The coping of the pier can be made of a cast -iron plate about
14 in. thick , with side lugs to be fastened to the cover -box of the
instrument. The plate is purposely made thick to avoid placing
any anchor- bolts. The cover -box is made with a glass top , and
should be dustproof.
The needle of a galvanometer in good order swings freely, and
any retardation in its free motion shows that some obstruction
is in contact with the moving coil . The instrument cannot then
be relied upon, and should be overhauled . If the galvanometer is
kept in a dust-tight box the trouble is not likely to arise, as such
010
P
SH H
M
Fig . 54. - SIEMENS -HALSKE RECORDING GALVANOMETER FOR LE CHATELIER

THERMO -ELECTRIC PYROMETER.
occurrence is due to careless construction; for this reason the in

strument should be always thoroughly tested before use . Before
moving instruments that are used at intervals for checking other
instruments or for testing furnaces, great care should always
be taken to fasten the needle very carefully, and it is also well to
have for them a portable box especially built and well lined .
For blast -furnace work the insulation of the couples is effect
ively done by the use of hard glass tubing. The couples being
permanently installed there is very little danger of breakage,
since all that is required is the insulation of the two wires. For
the couple located in the down- comer, where it may be affect

ed by slips, the best protection to the couple is given by placing
the wires in the usual double -bore clay tubes and then wrapping
them with asbestos cord . As has been said before, by using a
couple made of silver or copper and constantan it is possible to use
the Le Chatelier pyrometer for temperatures as low as 100 ° C. ,
and even lower.
For blast -furnace work, where the temperatures of the gas
currents in the bustle and down - comer are always below 1000 ° C. ,
there is no possibility whatever of alteration of the couples. The
writer knows of couples placed in the bustle-pipe of a furnace
three years ago that are as good as when put in place. A com
P
PELLIN
.PARIS
1 BALABAZE
FIG. 55. — OPTICAL PYROMETER OF LE CHATELIER.
plete installation , with two recording galvanometers - one for the

bustle-pipe, the other for the down - comer — with couples at six
stoves with a direct-reading galvanometer, can be installed at
considerably less expense than any other type of recording in
strument; the indications, besides, will be found more accurate.
As we have seen , the Le Chatelier pyrometer, in spite of its many
advantages, has its limitations. It can be used for intermittent
temperature determinations up to the melting point of platinum ,
say, 1780 ° C. Above this temperature it is necessary to use py
rometers based on different principles. The field is left at this
PYROMETRY . 169
range to the instrument making use of some optical properties

or employing heat radiation . Several very good optical instru
ments have been brought out.
Le Chatelier Optical Pyrometer.—
The Le Chatelier optical py
rometer and its modification , the Féry optical pyrometer, measure
the intensity of the red radiations emitted by heated bodies. The
field of vision in a telescope is divided in two halves, as in pho
tometers ; one half is lighted by a standard lamp, the other by the
heated object under observation . The instrument having been
calibrated by comparison with heated objects at known temper
atures, it is possible to know the temperatures of an object having a
given emissive intensity. The instrument gives remarkably accu
rate and concordant results, even in the hands of the inexperienced.
E
Leben DI
M
C B Jo &
P
C
FIG . 55a .
By the nature of the instrument the observer is required to keep

the eye at the objective, which causes fatigue if the observations
are carried on for any length of time . With the Le Chatelier it is
necessary to focus, while with the Féry the size of the object and
its distance have no effect on the illumination .
The oil lamp is not convenient for use in places where draughts
exist; it is better, then , to replace the oil lamp by a low - voltage
incandescent lamp , the current being produced by a battery.
Holborn and Kurlbaum , and Morse Pyrometers. — These two
instruments, built on precisely the same principle, have an in
candescent lamp, the brightness of the filament of which is directly
compared with that of the heated object. The current flowing
through the filament is varied by a rheostat until the filament
disappears from view ; the corresponding voltage is read on a

voltmeter. By calibration against known temperatures the
temperatures corresponding to the voltages have been established .
The shape of the filament — a horseshoe in the Holborn and
Kurlbaum , a spiral in the Morse - constitutes practically the
difference between the two instruments . The lower limit of the
instrument is 600 ° C.
Wanner Pyrometer. — The light from a low constant voltage elec
tric incandescent lamp and that from the heated object under
observation are polarized. The instrument is a modification of
König's spectrophotometer. The observer matches the illumi
bb
PHPELLOP
ALADIEZ
Fig . 56.–FÉRY HEAT-RADIATION PYROMETER .
nation of the standard half of the field by turning an analyzer.

The angle through which the analyzer is rotated is a measure of the
temperature. The instrument gives very concordant and accurate
results. It cannot be used below 900 ° C.
Remarks. — These optical instruments in the hands of trained
men can give accurate results; however, they all present the same
difficulty — that of matching two illuminated surfaces, thereby
requiring the constant attention of the operator.. Care has to be
taken in regard to the age of the lamp and the constancy of the
PYROMETRY . 171
current. The indications of the instruments are not reliable if

the heated object is strongly lighted by sunlight.
Heat-Radiation Pyrometer.The heat radiated from a heated
object varies according to Stephan's law (mathematically demon
strated correct on thermodynamic considerations), as the fourth
power of the absolute temperature. In the Féry radiation pyrom
eter the heat radiated is measured by means of a very sensitive
thermo- couple, formed of constantan and copper ( constantan being
an alloy of 75% copper and 25% nickel). The thermo-junction is
heated and an E. M. F. is generated, the measure of which is given
by a d'Arsonval galvanometer. The instrument is calibrated by
comparison with the Le Chatelier thermo -electric couple up to the
melting point of platinum ; above that by extrapolation.
需
E
D C
AL
Fig. 57.–FÉRY HEAT-RADIATION PYROMETER, LENS TYPE .
In Fig. 57 is given a section of the instrument. The thermo

electric couple is formed by two extremely thin wires supported
by D and C , electrically connected to the binding -posts b, 6'. In
front of the couple is placed a cross-shaped metallic screen , C, which
leaves only the thermo-junction exposed. Once the image of the
body under observation covers the thermo- junction, the instru
ment becomes independent of the dimensions of the object.
The minimum distance is 1 m. , for which distance the minimum
diameter of the object must be 3 cm . With a larger diameter the
distance of observation may be largely increased ; for example,
with a diameter of 60 cm. the instrument may be placed at 20 m .
The instrument has to be focused for each distance ; however,
the indications are rendered independent of the distance by means

of the diaphragm E, the cone of heat rays falling on the thermo
junction having thereby a constant angle, whatever the distance.
富。
FIG. 58.—FÉRY HEAT -RADIATION PYROMETER, MIRROR TYPE.
This type of apparatus is well suited to the observation of tem

peratures above 900 ° C. Below this, and in fact up to 1100 ° C.,
the absorption of the heat radiations by the lenses is unduly high .
The instrument illustrated in Fig. 58, of the telescopic type, has a
m
m'
Fig. 59. - FOCUSING DEVICE OF THE FÉRY HEAT- RADIATION PYROMETER ,

MIRROR TYPE ,
中(
sensitiveness ten times that of the other for the lower range of tem
perature. It will give correct readings down to 300 °. The instru
ment has the same construction as to the means of eliminating
the factors of distance and dimension of the objects under obser
vation .
The correct focusing is assisted by two small mirrors, m, m',
PYROMETRY . 173
through the use of which the image of the object appears distorted
when the focus is taken too short or too long (as shown in Fig. 65) .
As the thermo-couple is never exposed to temperatures of above
80 ° C. as a maximum , it preserves absolutely the constancy of its
E. M. F. The instrument once set up, sighted and focused, no
25- CFERY
s
la
Co
Z.
Fig . 60. — CAMBRIDGE SCIENTIFIC INSTRUMENT COMPANY DIRECT -READING

GALVANOMETER FOR THE FÉRY HEAT- RADIATION PYROMETER .
further adjustment is required. The temperature of the body is

automatically shown on the dial of a dead -beat galvanometer,
giving, therefore, practically instantaneous readings.
The instrument being absolutely free of the personal equation,
any workman with ordinary intelligence, without previous training,
can make accurate observations. The indications of the instru
ments are not affected by the extraneous lighting of the observed
body, which may be an ingot under treatment through rolls or
under presses. The instrument is also made self-registering.
CHAPTER XII .
Gas Analysis.
The methods here given apply only to technical analysis in
which results correct to within 0.5% are amply sufficient. The
gases to be analyzed may contain the following: Oxygen, carbon
dioxide, carbon monoxide , hydrogen , methane, ethane, ethylene,
nitrogen , water vapor, etc.
In producer- gases, with which we have mainly to deal, the heavy
hydrocarbons seldom occur over 0.5% , and it is usually sufficient to
reckon them as heavy hydrocarbons without further differentiation .
It is of prime importance that the analyses be made in the shortest
time possible.
In large works it is often impossible to have aa special laboratory
located conveniently to the various furnaces ; in such cases it is
necessary to utilize the first handy location . It follows that, for
technical analyses, preference should be given to methods requir
ing simple apparatus and a minimum time. All methods in which
mercury is used should be discarded . Determinations made with
explosion burettes and eudiometers filled with water cannot be
regarded as accurate, as the pressure exerted during the explosion
is sufficient to dissolve a large amount of the carbon dioxide formed .
It is also often necessary, for weak mixtures, to add oxyhydrogen
gas, which requires a generating apparatus . The method of
burning the gases by means of a spiral of platinum wire necessitates
the use of an electric current.
The method here outlined has been found to give sufficiently
accurate and concordant results with a minimum expenditure of
time and apparatus. The water vapor is absorbed by a Liebig
bulb filled with concentrated sulphuric acid. This is done very
conveniently while the sample is being taken . The carbon dioxide
is absorbed by a 33.3% solution of potassium hydrate, the heavy
GAS ANALYSIS . 175
hydrocarbons by strong bromine water, and the oxygen by an

alkaline solution of pyrogallic acid ; carbon monoxide is absorbed
by an acid cuprous chloride solution. The hydrogen and methane
remaining are burned by means of a Dreschmidt capillary platinum
tube, the carbon dioxide formed being absorbed in a potash solu
tion . A complete gas analysis can thus be performed in an hour, or
even less, the time varying according to the skill of the operator.
Sampling. This can be done by introducing a piece of gas
pipe into the flue, the pipe being of such length that its opening
will reach about the middle of the flue. The pipe should be
large enough to allow the placing of a loose wad of asbestos fiber
to collect any dust, soot and tar that may be present in the gas. If
the flue is built of bricks, the work is quite simple ; if the flue is
of sheet iron , in order to have the pipe reach inside it is convenient
to use the shape of pipe illustrated in Fig. 61. The required length
of pipe to be introduced in the flue is reduced in diameter so that
it will easily pass through the hole drilled for the main portion of
the pipe; thus the hole is tapped and the conical shoulder threaded.
It is convenient to reduce the diameter of the pipe at the other
end to about one- quarter inch in order to screw on a quarter-inch
gas -cock .
Sample Tubes. — The tubes have a capacity of about 300 cc . ,
with capillary ends provided with three-way stop-cocks, as shown
in Fig. 61.. These cocks are very convenient, as they allow the
removal of air, always present in the open end . The tubes, gen
erally a set of twelve, are handily carried in a box made especially
for them .
It is better not to fill the tubes with water before the taking of
the samples, as in all instances it is well to pass a rather large
amount of gas through the tube, the last trace of enclosed air being
thus removed. In very cold weather, if the flue is exposed to
the outside temperature, it is almost impossible, if the tubes have
been filled with water , to prevent the freezing of the water in the
capillaries. In all technical analyses the determination of the
water vapor should always be made. In the making of producer
gas, it is usual to consume fully 10% , and sometimes 15 % , of the
total fuel burned on the steam -boiler grate; the water-vapor con
KTM
Joan U
Fig. 61. – APPARATUS FOR THE DETERMINATION OF WATER VAPOR IN GASES .

The Sample Tube is shown inserted between the Mariotte Flask and
the Safety Sulphuric-acid U - Tube.
tents will give in all cases important indications; the determination
is, as we have seen, absolutely necessary in making furnace tests.
Water -Vapor Determination .This can be done either by the
GAS ANALYSIS . 177
determination of the dew point, as shown by Mahler ', or preferably,

in industrial practice, by actual weighing. A fixed volume of gas,
at least three liters, is passed through the Liebig tube on its way
to the sample tube ( Fig. 61). The gas is drawn under a constant
head of water by means of a Mariotte flask , of 5 liters or more
capacity. The weight of the Liebig bulb being determined before
and after the operation, the difference in weight will give the weight
of the water vapor contained in the volume of gas taken ; knowing
the molecular weight of the water vapor, 18 grams for the molecular
volume 22.32 liters, it is an easy problem to find the corresponding
volumetric percentage of the condensed water weighed .
The Liebig bulb should be connected to the gas -cock with as
short a piece of tubing as possible, the length of the iron sample
tube being such that the temperature of the cock will never be so
high as to cause the rubber to soften and distillate; on the other
hand, it should not be so long that the temperature of the issuing
gases will fall below 100 ° C. This is very important in order to
avoid the premature condensation of the water vapor. The red
or white tubing is better suited for this connection than the
black , as it will stand a higher temperature.
The quantity of gas passed through the Liebig bulb is ascer
tained by catching the water escaping from the Mariotte flask , in
a vessel of known capacity. If the volume of the recipient is 3 liters
or over, it will be sufficiently accurate to measure the water used ,
without having recourse each time to weighing.
The Liebig bulb is followed by a U -tube filled with glass beads,
moistened with concentrated sulphuric acid, in order to prevent
the contamination of the Liebig bulb by the water vapor from the
Mariotte flask . One Liebig bulb is always sufficient for the total
absorption of the water vapor ; trials have been made with two
bulbs, and the second one never increased in weight. It is con
venient to make a frame to carry the train of apparatus.
With strong stack draught it may be found that the gas will
not flow ; this is easily remedied by increasing the length of the
discharge tubing from the Mariotte flask, thus increasing the head .
As in all other chemical work, for accurate results it is of prime
" P. Mahler. “ Etude sur les Combustibles, Solides, Liquides, Gazeux,” p 51 .
importance to perform all operations, as far as practicable, under

the same conditions.
The room selected for the analytical work should be kept, as far
as possible, at a uniform temperature. A large supply of water
necessary for the various determinations should be kept in a suit
able reservoir, so that the water used in the analytical process shall
be at the prevailing temperature of the room.
For practical purposes, it is better to use pure water rather than
a saturated solution of sodium or magnesium chloride, as the salt
solutions, by capillary action, cause a deposit of the salt in all
the joints and cocks of the apparatus used .
It is well, to insure accuracy , to saturate the water used with
1
the gas in use. The composition of the producer-gas in a plant

using the same grade of coal and in good running condition is very
nearly constant ; it is then sufficient to pass a stream of the gas
through the bottle ; this will insure a constancy of condition better
than the using of salt solutions. Since the method selected is
rapid, requiring in the hands of a trained man less than an hour's
time for a complete analysis, it follows that it is an unnecessary
refinement to make corrections for temperatures, vapor tensions
and barometric pressure . If this is thought desirable, however,
it is a simple matter. The errors resulting from this neglect, when
the work is otherwise properly conducted , are usually of a smaller
order than the other errors introduced by the use of water, un
saturated reagents, errors of reading, etc.
Several very good forms of apparatus can be used for technical
gas analyses, all giving satisfactory results. In order of accuracy ,
we may mention the Hempel, the Elliott and the Orsat. This
order is changed if the rapidity of the determination is considered ;
it then becomes the Elliott, the Hempel and the Orsat. For
technical work when the analyses have to be made in rooms not
especially adapted to the purpose, the writer gives preference to the
Elliott apparatus, which has only one absorption tube . The ap
paratus is placed in a fixed position, and there is very little chance
of breakage. With the Hempel apparatus a number of absorption
1 A new type of Orsat apparatus has lately been put on the market. The gas under
analysis is made to bubble through the reagent's burettes: the time of absorption is thus
much reduced .
GAS ANALYSIS . 179
R
с
Fig . 62. - ELLIOTT APPARATUS FOR GAS ANALYSIS.
pipettes are required, and unless a troublesome record of the quan

tity of gases absorbed by each pipette is made errors may be
introduced by the use of too old reagents. The manipulations
are also much more complicated than with the Elliott modification
of the Bunte burette, as fresh reagents are introduced every time;
this, naturally, means an extra waste of materials.
We will confine ourselves to the description of the method we
have followed in the Elliott gas apparatus. For the description

of the Hempel apparatus, we will refer to the excellent work of
Dr. Hempel on the subject. It is very important that the deter
minations be conducted under similar conditions.
The filled sample tubes are left to stand in the room long enough
to reach the surrounding temperature. The burette readings are
to be made always after the same lapse of time. This is to
give the water the same time to run down the walls of the
tube; for this purpose a small sand - glass is very convenient. The
same volume of gas, to within a few cubic centimeters, should be
taken for analysis each time. It is useless to try to measure an
exact number of cubic centimeters — say one hundred - since,
with methane present, calculations are always necessary. The
calculations are, furthermore, very simple, especially with the use
of a slide- rule '. The same quantity of wash -water should be
used in each case, care being taken to avoid handling the tubes,
as the heat conducted suffices to vitiate the results. If all these
precautions are taken , even with unsaturated reagents, an ac
curacy of 0.5% is attainable with the Elliott apparatus.
Potash Solution . - A solution of 33% strength is made by dis
solving one part of commercial potash sticks in two parts of water .
It is well to make enough at one time to last for a large number of
analyses, as it cannot be used immediately after preparation on
account of the exothermic reaction, the solution becoming very hot.
Bromine Water. - A saturated solution is obtained by placing
100 cc. of bromine in 900 cc . of water ; it is not necessary to in
troduce any of the bromine in the absorption tube, the saturated
water solution being sufficient.
Alkaline Solution of Pyrogallol.- Five grams of pyrogallol are
dissolved in 15 cc . of water ; in another beaker 80 grams of com
mercial potash are dissolved in 200 cc. of water; the two solutions
are then mixed . The solution must be kept in a well -corked bottle,
to prevent the access of air.
Cuprous Chloride Solution . The choice lies between the ammonia
calsolution of cuprous chloride, or an acid solution . The acid solution
has a more energetic absorbing power and is more easily prepared .
1 The 7-in. Sexton omnimeter, a circular slide -rule, is particularly serviceable
GAS ANALYSIS . 181
Many tests have satisfied the writer that one absorption with
about 25 cc. of the fresh solution will completely remove the car
bon monoxide. The solution is prepared as follows: 1000 grams
of cuprous chloride are placed in a 2500 cc. bottle ; 1000 cc. of
water are added , then 250 cc. of concentrated hydrochloric acid ;
as many lengths of copper wire are introduced as is convenient,
and the bottle is then filled with water to the neck and a good rub
ber stopper inserted . The bottle is thoroughly shaken . The
solution will be turbid and brown -colored at first, but after 24
hours will become thoroughly limpid ; it is better to have the
Position No. 1 . Position No. 2.
Position No. 3 . Position No. 4.

Fig . 63.SHOWING THE VARIOUS POSITIONS OF THREE-WAY Cock , F.
solution in a dark - colored bottle, as the light affects it . The

measuring -tube is protected from sudden changes of temperature
by means of an air -jacket; water may be used, but is not necessary.
Manipulation . — The sample-tube, having reached an equilibrium
of temperature , is connected with the water-bottle by means of
rubber tubing on which a pinch-cock is inserted . The air in the
tube and in the capillaries is forced out by water before the three
way cock is opened. The upper capillary of the sample-tube

is now connected with the measuring burette by means of a cap
illary ell, and the air in the connections is removed by forcing
water through the three -way cock of the burette, by bringing
the water -bottle A to the upper level. The bottle C is placed on
the table, and the cocks of the sample-tube as well as cock F of
the burette are opened to position No. 2 ; the gas will now flow
in the burette. When the proper amount has collected, F is changed
>
to position No. 3. The upper three -way cock of the sample-tube

is also closed, and the capillary connection between the burette and
the sample-tube removed . The volume of gas in the burette is
read by raising bottle A until the levels of the water in the bottle
and in the burette are on the same horizontal plane. The reading
PosiTION No. 1 . Position No. 2.
1
POSITION No. 3 .
Fig . 64. SHOWING Various PosiTiONS OF THREE-WAY Cock , G.
is noted , the bottle A raised to its upper position, the bottle B

lowered to the table , and the cock F is opened to position No. 4,
and cock G to position No. 1. The gas is now confined in the
absorption tube. Cock G is brought to position No. 2, and F
to position No. 3. The gas is then under a slight depression.
The reagent cup R is filled half-way with the potash solution, and
cock H is turned to position No. 2 , so that the solution will flow
uniformly in small ripples down the walls; when the solution
ceases to flow , or merely flows more slowly, cock G is turned to
position No. 3, very slowly ; special care is to be taken here to
2
No. 3.
No. 2.
No 1 .
Fig . 65. - SHOWING THE Various PositioNS OF THREE-WAY Cock, H.

avoid an inrush of air. Cock H is turned to position No. 1 as soon

as the potash solution nears the capillary. F is opened for a few
seconds to wash the connection , after it has been brough tto po
sition No. 3. H is brought to position No.3, and aa small amount of
wash -water is allowed to flow ; H is then turned to position No. 1 .
The gas, which is now free of carbon dioxide, is to be brought
back to the measuring burette; to effect this operation the bottle
B is brought to a higher level, G is turned to position No. 1 , and
F to position No. 4 ; the bottle A is lowered to the table ; the
gas will now flow into the burette; the water is allowed to flow in
the capillary down to the zero mark , F is turned to position No.
3, and , as before, the volume of the gas is measured with the usual
precautions.
If the gas contains ethylene, C ,H ,, the next absorption will be
4
by means of bromine water; the potash solution left in the ab

sorption tube must first be thoroughly removed . If no ethylene
is present, the next absorption is by means of an alkaline solution
of pyrogallic acid ; the potash need not be washed out in this case .
To wash the absorption tube, G is turned to position No. 3, and
H to No. 2 ; the tube is thereby emptied , and wash -water is in
troduced into the cup, a little at a time; the flow must be regulated
so that the water will run entirely along the walls ; finally H is
brought to No. 3, and some more wash-water allowed to flow .
H is then turned to position No. 2 , G to position No. 1 , and the
absorption tube is filled until the water overflows into the cup R.
Bottle B is then lowered to the table, A is brought to the upper
level and F is turned to position No. 4, the gas being transferred
to the absorption tube. The cycle of operations is now the same.
The bromine water is introduced into the tube; to remove the
vapors of bromine which have a high tension , some potash solution
is introduced after absorption. The gas is then removed to the
burette and the contraction measured as before.
Oxygen . — The next gas to be determined is oxygen. The ab
sorption is effected by means of the alkaline pyrogallol solution .
When no ethylene is present, this absorption is effected im
mediately after that of the carbon dioxide, the potash solution
being left in the absorption tube. Care should always be taken
U
K
S T E F
H
E-Fig
.66.
LLIOTT
APPARATUS
FITTED
DREHSCHMIDT
WITH
PLATINUM
TUBE
DETERMINATI
THE
FOR
HYDROGEN
OF ON
METHANE
.AND
to remove the reagent solutions from the upper part of the absorp
tion tube and in the capillary connection, in order not to con
taminate the water contained in the measuring burette. The
manipulation given under the determination of carbon dioxide
should be adhered to . The pyrogallol solution is to be thoroughly
removed before the next absorption, that of carbon monoxide.
Carbon Monoxide. — The acid cuprous chloride solution is in
troduced into the tube, about 25 cc. being used . The solution is
allowed to flow freely along the walls, this being easily accomplished
by having the tubes of the apparatus in a vertical position. The
tubes, however, should be thoroughly cleaned before erecting
them , to remove any greasy film that would prevent the solutions
from wetting the walls .
Though the vapor tension of the cuprous chloride is small, it
is better to introduce some potash solution before passing the
gas into the burette in order to absorb the vapors of hydrochloric
acid . The volume of gas remaining is measured, and the con
traction is equal to the amount of carbon monoxide present in the
gas. The remainder of the gas consists now of hydrogen, methane
and nitrogen .
Hydrogen . — The gas can be burned by passing it over palladium
asbestos; below 100° the methane is not affected . In any event,
combustion has to be resorted to for methane. It appears, there
fore, quicker to subject both gases at the same time to combustion .
We have seen that, with water as the confining liquid, explosion
is not to be recommended. The combustion is effected by using
the Drehschmidt tube or the Coquillon platinum spiral, heated
by an electric current. For practical purposes preference is to
be given to the Drehschmidt tube, which requires no electric cur
rent , but simply a gas-jet or an alcohol lamp.
The Drehschmidt Tube consists of a capillary platinum tube
100 mm . long with a diameter of about 0.7 mm . internal, and 2.5
mm . external. The tube is soldered to the copper tubes at each
extremity, as shown in Fig. 72. The ends of the platinum tube
should extend for about 10 mm . inside the copper tubes , so that
the soldering may be thoroughly cooled . The copper tubes are
about 2.5 mm . inside diameter, 3.5 mm . outside, and are U -shaped.
Sih
AID 10
mm
F, Mobile Bottle . T , Plunger.

H, Path of Bottle F. R , Registering Lever.
G , Gas Burette. U , Registering Drum .
A , Absorption Bottle. F1, Registering Pen .
S6, Safety Vessel. Q , Regulating Valve.
G , Glycerine. A , A , Air Chambers.
K , Caustic Potash .
Fig . 67. - SECTION OF THE " ADOS " AUTOMATIC GAS ANALYZER.
Four or five platinum wires are inserted throughout the length

of both the platinum and copper tubes; the wires may be 0.008 mm .
in diameter.
The two copper tubes are to be thoroughly cooled ; an easy way
is to insert the U portions of the tubes in deep beakers filled with
water, the beakers being protected from the radiation of the flame
by asbestos cardboard. Cooling by means of a stream of water
is to be avoided , as it requires rather complicated apparatus.
The combustion is effected in the capillary tube through the
catalytic action of the platinum at a good cherry heat . The
wires assist in giving additional points of contact, and they also
decrease the free volume inside of the tube ; in case of rich mixtures
there is a possibility of an explosion. There is little likelihood ,
however, of its being carried back into the burette, on account
of the low temperature existing in the copper tubes, which would
bring the gas below the explosion temperature .
The apparatus required for the use of the Drehschmidt tube
is shown in Fig. 66. The tube is connected with the two burettes
by means of rubber tubing. The gas is transferred to burette K,
and enough air, with about 10% excess , is then introduced to
insure the complete combustion of the gases. The total volume
of the gases is measured . The tube is thoroughly shaken to pro
mote a good mixture. The burette is again connected with the
copper tube; meanwhile the heating apparatus is placed under the
platinum tube. When the tube has been brought to a cherry
heat the gases are passed slowly through the tube from one burette
to the other. Care is to be taken to avoid introducing any water
into the copper tubes . The gases are passed four or five times,
the last pass bringing them in the measuring burette. F is turned
to position 3, the Drehschmidt tube is removed and the contraction
measured. The gas is then removed to the absorption burette
for the absorption of the carbon dioxide produced . By a simple
calculation the percentage of hydrogen and methane is obtained .
The nitrogen is found by difference . The temperature in the
Drehschmidt tube is too low to result in the partial combustion
of the nitrogen .
GAS ANALYSIS . 189
Calculations. - As all the operations can be performed in con

siderably less than an hour, no corrections need be made for tem
perature and pressure . The percentages of carbon dioxide, ethy
lene, oxygen and carbon monoxide are found by dividing the
corresponding measured contractions by the original number of
cubic centimeters .
Hydrogen , Methane and Nitrogen.

Let H , = X CH , = y N, = 2
if V = original volume ,
C = Contraction after explosion .
D = Volume of carbon dioxide produced.
We have, then (since one volume of hydrogen combines with
one- half volume of oxygen to form water) , for one volume of hydro
gen a contraction equal to one and a half times that volume.
Also, since one volume of methane combines with two volumes
of oxygen to form one volume of carbon dioxide, there results a
contraction equal to twice its volume :
2C 4D ( 2C - 4D) 100
X = in % x =
3 3A
100D
y = D у
A
Water Vapor. - Let the volume of gas passed through the Liebig
bulb be 3450 cc. , the increase in weight being 67 mg.
The molecular weight of water vapor is 18 grams for the molecular
18
volume 22.32 liters; or, per liter, grams .
22.32
18 X 3450
For 3450 cc . we have 2782 mg .
22.32
The true percentage of water vapor for 67 mg. is then

67 x 100
- 2.40 % .
2782
%..IISateau
181
1904
1cavril
....
..._ele
eoMoyenoe
,b.... 2.13.cas.vil
904
445002.eo
Moyeno
....
XI ..Щү
x|мүX
!jour.D I
V7!VIV
I'I
VWXX N
Staufflet
Chauffeur
1
...
....Chaudière Cha
......uffeuf
famosta
ithusse o
.Bcharbow.
Observations
S
C"THE
.68.
Fig
-C
AT
"A
,GAS
AOF
STACK
THE UTOMATIC
ARD
ANALYZER
HOWING
DOS
RECORD
CO ONTENTS
FROM
GASES
F
HIRED
.-A AND
BOILER
GAS ANALYSIS . 191
Automatic Self-Registering Gas-Analyzing Machine. It is very

important to be able to keep a record and at the same time a con
trol of the composition of gases, such as producer-gas or products
of combustion. Up to date it has not been possible to make prac
tically more than one determination, that of carbon dioxide, CO .
This is, however, sufficient in most instances, as generally the
same coal or fuel will be used at a given plant.
In producer- gas, there is a certain relation between the total
number of molecules of carbon contained in combination, and the
molecules held by the different gases, CO ,, CO, CH .. The knowl
edge of the volumetric contents in CO , will allow a fair estimate
of the remaining components being made.
In Fig. 67 is illustrated a section through the analyzer of the
Ados 1 gas-machine. The gas to be analyzed is drawn from the
flue by means of a very ingenious double -acting air -pump, operated
by the draught of the furnace stack . A fixed volume of gas, 100 CC . ,
is passed through the absorber, A , containing a strong solution of
9
KOH ; thence the gas, minus the CO 2, is sent to the measuring

22
vessel , G 2.2 A line proportional in length to the remaining volume

of gas is drawn automatically on the paper wrapped around the
drum , U. A sample may be taken every four minutes, or at greater
time intervals, the apparatus being regulable at will . Glycerine
is used to fill the apparatus. A gas-filter is provided , as it is neces
sary in the case of bituminous producer -gas.
The apparatus has been introduced in a number of large Euro
pean works, especially in connection with boilers. Fig. 68 shows
the 24 -hour record of a boiler. The firemen are paid premiums
according to the carbon contents of the waste products .
The apparatus has also been installed , lately, on the producer
batteries at the Hazard works of the New Jersey Zinc Company.
1 Made by E. Dégremont , Le Cateau , France.
CHAPTER XIII .
CALORIMETRY .
Data concerning fuels cannot be complete, or even regarded as

satisfactory, unless the heating value or calorific power be in
cluded . This expression of its latent energy is of primary im
portance in the valuation of most fuels, as it affords a basis of
comparison both as to relative value and as to the possible heat
utilization .
It would seem , at first, possible to calculate the calorific power
of coal from its chemical analysis. And so we could were the
compounds present accurately known , but though many attempts
have been made, none have met with signal success . The most
reliable formula for the heating effect of a coal, burned to liquid
water, is that of Dulong, which gives the calorific power in calories
per kilogram .
Calorific Power equals 8080 C + 34460 (H – 10) + 2250 S.
Attempts to plot the calorific power and analysis graphically and
thus arrive at a general rule, or graphic method, of arbitrarily de
termining the heating value of an unknown coal, have not been
satisfactory .
Knowing the chemical analysis or composition of an oil or gas,
we can calculate its calorific power quite accurately, but in many ,
perhaps most , cases the complexity of composition is such that a
direct determination is preferable.
The calorific power, expressed in the centigrade -kilogram or
Fahrenheit-pound units, is found by actually burning the fuel
under known conditions and noting the heat evolved ; usually by
absorption by water. It is, therefore, not simply a theoretical
property of the fuel, but represents actually the heat evolved when
at particular fuel is burned .
CALORIMETRY . 193
The form of apparatus employed will vary somewhat with the

state of the fuel, whether solid , liquid or gaseous ; but omitting the
Thompson calorimeter, which was liable to as much as 15% error,
we have at present various modifications of three standard forms:
( 1) The Bomb Calorimeter, originally devised by Berthelot.
(2) The Parr Calorimeter .
(3) The Junker Calorimeter .
The first of these, under the Mahler, Atwater, etc., modifica
tions, is the most accurate calorimeter known, and is suitable for
the combustion of fuel either solid, liquid or gaseous. The Parr
is suitable only for use with solid or liquid fuel, and the Junker for
liquid or gaseous fuel only .
The following descriptions will render clear the mechanism and
actual procedure to be followed in the use of these instruments.
THE BOMB CALORIMETER .
The general principle upon which this apparatus is based is the

combustion , by means of oxygen, of the material in an enclosed
space without addition or subtraction of material during the op
eration . The total heat evolved can thus be absorbed by sur
rounding the apparatus completely with water, the rise in tem
perature being noted. To insure an excess of oxygen gas, by
which a complete combustion is secured , a pressure of about 20
atmospheres is used . The ignition is brought about by electrically
heating to incandescence a small coil of wire in contact with the
substance to be burned .
The vessel or bomb within which the combustion takes place is a
strong, seamless, cylindrical steel shell provided with a close-fitting
cap which screws over the open end . To prevent oxidation of the
iron shell during the combustion and corrosion by the nitric and
sulphuric acids formed , it is necessary to line the shell with plati
num or enamel the interior (Mahler) . More recently Atwater
has used a gold -plated copper lining which has advantages over
the older forms. A description of this latter form of apparatus,
with a few explanatory remarks , will suffice for all the modifications
now in use . To enumerate the different parts of the apparatus

we have :
( 1 ) The calorimeter, including the bomb, a thin brass cylinder

or pail to contain the water surrounding the bomb, a thermometer,
and the stirring mechanism .
(2) The outer envelope, enclosing the calorimeter system ;
designed to render the loss or gain of heat as uniform and slow as
possible.
(3) Supplementary apparatus for filling the bomb and com
pressing the fuel and also effecting its ignition , etc. This in
cludes the pressure gauge and couplings, a clamp and spanner to
hold and close the bomb, a briquetting device for compressing
the fuel, and a motor for stirring the water.
The bomb shown in Fig. 69 is self -explanatory, the substance
being held in the capsule C. Oxygen is admitted under pressure
at 0, and is retained by closing the valve V. The joint between
body and cap is made tight by a lead gasket, which is pressed by
the annular V-shaped edge of the body. The V-shape is a great
improvement over the older flat surface. Ignition is effected
by passing a current of about 5 amperes through the insulated
post , P, to the coil of fine iron wire, W, just above the combus
tible matter. This wire is heated to ignition by the current, which
finds its way to the body of the bomb and thence completes the
circuit by a connection at V.
The lining which covers the interior surface of the shell and cap
has been made of platinum , enamel and gold -plated copper. The
two former labor under the disadvantage that platinum is very
expensive and the enamel will crack and flake off, exposing the
>
iron . The gold - plated copper is satisfactory and durable, but

it or any other removable lining must be carefully fitted to prevent
the compressed gas from finding its way between it and the shell ;
otherwise the lining will be completely destroyed upon the re
lease of the interior pressure. The convenience of removal, how
ever, outweighs this objection, which may be avoided with care.
When in use the bomb is immersed in water contained in a thin
brass pail , of such dimensions that the bomb is completely covered
CALORIMETRY . 195
and sufficient room is left so that there is a certainty that no water

shall splash over during the stirring. This pail must be kept
‫המון‬
‫שיחותי‬
ilkiliilliillllll
VICT
Fig . 69. - CROSS -SECTION THROUGH THE MAHLER CALORIMETER (ATWATER

MODIFICATION ).
bright and clean to render the thermal losses low and regular.
The outer envelopes, in the Atwater form , are of indurated
196 INDUSTRIAL FURNACES AND METHODS OF CONTROL,
fiber having an air space between the adjacent walls. The covers
are of hard rubber, arranged also with an air space and perforated
for the stirrer and the thermometer. In some of the other forms
this outer envelope is a felt-covered, water - filled metal jacket,
but this is not necessary or convenient. Whatever the outer
envelopes be, they must not absorb water and they must be easily
cleaned and dried for use .
Fig. 70. - BERTHELOT-MAHLER APPARATUS READY FOR OPERATION, IN THE

METALLURGICAL LABORATORY OF COLUMBIA UNIVERSITY.
Stirring is effected by the passage of two perforated annular

brass rings which encirele the bomb and fit within the calorimeter
pail. The up-and -down movement is best brought about by a
motor connected by a stout cord or system of levers, as shown in
Fig. 70. For this a low-priced motor may be used , as is shown in
Fig. 70, instead of the rather expensive type wound for high
voltages. Fig. 70 shows the apparatus assembled and the dis
position of the different parts.
CALORIMETRY . 197
Most materials are in powdered form on account of the ne

cessity of securing a representative sample, and in such cases it
is best to press the powder into a compact pellet, so that small
particles may not be blown about within the bomb and escape
combustion . This is done in the pellet press A , shown in Fig. 71 ,
which is capable of exerting a pressure of about ten tons on the
material being bricked. Such a pressure consolidates most sub
Fig. 71. –VARIOUS PARTS OF THE BERTHELOT-MAHLER CALORIMETER.
stances , but in the case of some anthracite coals, etc. , the mixture
of a little benzoic acid in known amount will give the desired
result. The benzoic acid must be known, that we may make
the proper correction by deducting its heat of combustion from
the total calories evolved .
METHOD OF USE OF THE APPARATUS. — The quantity of substance
taken for combustion will vary somewhat with the different forms
of apparatus , the rule being that the observed temperature rise
should be 2.5 to 3.5° C. This rise is affected by the hydrothermal

value of the calorimeter system , with a large hydrothermal value
it being necessary to take a correspondingly large amount of the
material to be burned . The hydrothermal values of various calorim
eters range from 2000 to 3000 grams, and from this it is clear
that such a weight should be taken as will yield 5000 to 7000 cal
ories, or 7500 to 10,500 calories, according to the apparatus. For
most good coals, etc. , the weight will fall between 1 and 1.5 grams,
which should be weighed out approximately and compressed into
a solid pellet, which is brushed free from loose particles and weighed
accurately in the capsule in which it is to be burned. These cap
sules are of nickel, about 0.4 mm . (#4 in . ) thick, and of a proper
size to fit in the receiving dish of the bomb, which is thereby pro
tected from rapid deterioration . The nickel capsule should be
ignited in a Bunsen burner, to superficially oxidize it .
The wire used for ignition should be of soft iron , about 36 gauge
and 24 to 3 in . in length . To form it into a coil it is wound around
a large needle of wire, leaving half an inch of each end straight.
Its weight should be known in order to make the correction for its
combustion . This is most conveniently done by winding a number
of turns about a cylinder of the proper circumference and cutting all
at once. With care the weights of these wires should correspond
very closely, and thus obviate repeated weighings. One of the coils
is bridged between the platinum wires attached to the cap ; then
the capsule containing the substance is put in place, the wire being
brought over it in close contact .
With some substances special devices are required to secure
combustion . This is true of refuse, bituminous slates and shales.
Ash , etc. , the calorific power of which may sometimes be quite
high, falls in this class. These substances may be completely
burned by mixing with a proper amount of benzoic acid before
compressing. This amount should not approach 50 % of the total
calorific power, and so much should not be used that the remaining
cinder is fused by the intensity of the heat generated, for this may
prevent complete combustion .
Heavy oils, etc., may be absorbed in fibrous asbestos which has
been ignited and weighed in the capsule. The gain in weight after
CALORIMETRY . 199
addition of the oil represents the sample taken for combustion.

Volatile oils and liquids should be enclosed in gelatine capsules,
the weight of which is known . Trial by combustion of a sample
of the capsules will give the proper correction to be applied to each .
Having everything in readiness, the bomb cover is placed on
the bomb, which is gripped tightly by the clamp B, Fig. 71. The
>
cover is screwed firmly into place by means of the spanner, using

oil on the exterior rubbing parts. To fill with oxygen the bomb is
taken to the oxygen tank and connected at O by a brass tube, the
valve V being slightly open ; the pressure is admitted from the oxy
gen tank until the pressure- gauge registers 20 atmospheres . The
pressure is now cut off and the valve V closed . After disconnect
ing, the bomb is ready for immersion in the calorimeter-pail. It is
a wise plan to test the tightness of all joints at a low pressure, say
5 atmospheres . This can be done for large leaks by closing the
oxygen - tank valve and noticing whether the pressure-gauge re
mains constant. More delicate tests for leaks are made by moist
ening the joint, when bubbles will indicate the exact location . Any
perceptible leak about the bomb must be remedied, lest it render
the results inaccurate.
The water weighed into the calorimeter- pail should be about 30
below that of the air, in the inner air space. This difference in tem
perature is such that, under average conditions, the radiation cor
rections to be applied will be at a minimum . Such a weight of
water should be taken that the total hydrothermal value of the
system shall be a round number, say 2000, 2500 or 3000 grams .
Therefore, the weight of water to be weighed out will be the hydro
thermal equivalent of the system subtracted from the desired hydro
thermal value. The weight of the water need only be to the nearest
gram .
The calorimeter-pail is now placed in the inner compartment,
the stirrer is placed in position , the bomb immersed, with the elec
trical connections in place , the covers are put on , and finally the
thermometer is introduced .
The thermometer used must be one capable of being read to
0.001 °, and it should be graduated to hundredths. Either a series
of short-range fixed zero theriaometers must be provided , or one
of the “ metastatic " Beekmann type of adjustable zero. The choice

of the latter introduces the necessity for a correction, which cannot
be disregarded, depending upon the amount of mercury in the ex
pansion bulb ; but it is rather preferable to the series arrangement,
and is certainly far less expensive. A reading -lens is necessary for
estimating the smaller divisions, and it should be so arranged as
to avoid parallax.
Electric current is needed for the ignition , and it is also the most
convenient source of power for actuating the stirring mechanism .
Rather the best source of current for this purpose is the ordinary
lighting circuit, for which the motor driving the stirring apparatus
may be wound, or a much cheaper motor intended for battery
1
work may be used in connection with a lamp resistance, and the

speed be reduced by a large band -wheel, as is shown in Fig. 70.
For ignition , about 5 amperes should be supplied by means of five
32 - candlepower incandescent lamps. These also give the signal
when the circuit is closed , and ignition takes place by their bright
ening. In case the lighting current is not available, six Edison
Lalande batteries may be used .
The stirrer may be set in motion at a speed of not less than 40
strokes per minute. A higher speed may be used with advantage,
but care must be exercised that it is not so great as to splash the
water against the cover, or out of the calorimeter -pail. The ther
mometer bulb should be slightly over half -way down , or about
opposite the central mass of the bomb. It passes through a good
sized opening in the stirrer, and is held firmly by an exterior arm
attached to the outer envelope.
When the thermomenter shows, by its uniform rise, that the
system has attained an equilibrium (this requires 5 or 6 minutes) ,
we are ready to take the observations which compose the initial
period. Just before taking the thermomenter reading it is a
good plan to tap the thermometer lighly with a pencil , so that
the surface tension and frictional resistance of the moving mercury
may have no effect upon the accuracy of the readings.
Observations Required.-We shall require the actual temper
ature of the system to the nearest half-degree in order to make
the correction of the metastatic thermometer, if this is used, due
TJERARE
O THE
UNIVERSITY
OF
CA IFORNIA
CALORIMETRY . 201
to the varying quantity of mercury in its bulb. This may be

taken with an ordinary thermometer.
For the initial period we need six temperature readings at one
minute intervals. These readings are taken to 0.001°, and must
show a perfectly uniform rate of rise in temperature. At the
moment of the sixth reading — that is, at the expiration of the
fifth minute — the switch completing the ignition current is closed .
The ignition should take place promptly, with only a momentary
brightening of the lamps. The darkening of the lights indicates
the breaking of the circuit by ignition of the wire, and this should
take not longer than about a second; if it takes a longer time
it indicates that the wire is too coarse or the current not strong
enough . Short- circuiting may prevent the drop of the lights
and the ignition of the charge . Immediately after the ignition
has taken place the switch should be opened.
The main or combustion period is now on , and five readings
covering the five minutes of the main period are taken by con
tinuing the previous time intervals . During the rapid rise of the
temperature it will be impossible to take the readings exactly,
but the last three readings may be taken with great exactness,
the last especially. The maximum should be attained about
the third minute, and it should not be followed by any sudden
drop or fluctuation in the indicated temperature, as this points
directly to insufficient stirring. Should the rate of fall not be
come perfectly uniform before the expiration of the main period
of five minutes, the period must be extended until it does become
so , and the length of time noted for future use in calculation .
Following the main period is a final period of five minutes in
length, at the end of which a reading is taken, giving the average
rate of fall for use in the calculation of the radiation correction .
The bomb can now be removed and secured tightly in the clamp.
The internal pressure is relieved by opening the valve V gently,
and the cap is removed by the use of the spanner. The interior
of the bomb and cover are rinsed into a beaker with distilled water.
The nitric and sulphuric acids in this are estimated as follows:
The solution in the beaker is titrated with a sodium carbonate
solution containing 3.706 grams ( Na , CO2) per liter, using methyl
orange as an indicator . Each cubic centimeter of this solution

equals a calory of heat evolved by the combustion of the corre
sponding nitrogen. Sulphuric acid is included in this titration,
and may be allowed for by precipitation by means of barium
chloride . The barium sulphate resulting is weighed and each
0.008 gram equals a calory to be deducted from the nitric-acid
estimation .
As the heat of the burning nitrogen has no connection with that
of the burning fuel, the heat thus evolved is to be deducted from
the total. It usually amounts to from 10 to 30 calories. The
heat evolved by the burning sulphur is a portion of the calorific
power of the fuel , but its only significance here is that it may
lead to a high estimation of the nitric acid . Its equivalent ( roughly
0.01 gram barium sulphate equals 1 calory ) is to be deducted from
that of the nitric acid. It is usually insignificant, but should
be looked for in case the nitric acid value seems high .
Any iron wire remaining unoxidized should be weighed or meas
ured, and the proper correction made in the accessory combustions.
CALCULATION OF Results.—To find the heat of combustion
of the substance burned, it is necessary to ascertain the total heat
evolved , and to deduct from the total the heat liberated in the
accessory combustions. To find the total heat evolved , we must
know the hydrothermal value of the calorimeter system and its
corrected rise in temperature. The hydrothermal value of the
system is the sum of the weight of water used , and therefore
known, and the hydrothermal equivalent of the system . This
latter is a constant; later the method of ascertaining its value will
be shown. The corrected rise in temperature is found by the ap
plication of the laws of radiation to the observations taken .
It is very important that such a form be made out prior to the
tests , as confusion and the possibility of error are thereby avoided .
The computation , though simple, can become very intricate with
out some such systematic arrangement.
In this sheet the general data are entered at the top and the
records of observation in the left column. No corrected ther
mometer readings are shown in this example, for the reason that a
fixed - zero instrument was used.
CALORIMETRY . 203
The method of calculation is best shown by an example taken

from practice, which is shown on the accompanying form .
CALCULATION 11 .
CALORIMETER TEST
Sample No. 85,544 . Description , Bituminous Coal. Date, Jan. 18 , 1904 .
Temp. Room , 19.5. Time, 3:20 . Temp. Water , 16.7.
CAPSULE No. 5 CORRECTION FOR ACCESSORY COMBUSTIONS
Capsule + substance . 5.1835 Wt . Fe wire, .0325 .. 52. cals.

..
Capsule .... 4.0005 Titre of HNO 3, formed .. 25 .
Substance W 1.1835 Correction for accessories.. 77. cals.
INITIAL
PERIOD
Corrected
Min . Readings INITIAL PERIOD FINAL CALCULATIONS
Temp
1 L
17.235 Fall . .017 05 ..... 20.347
2 - 17.237
3 - 17.241 Rate V ... -.0036 8o ..... 17.252
- 17.245
- 17.248 Mean 1°, 17.243 03-80 ..... 3.095
00 17.252
PERIOD
Rad . Corr .... - .0187

MAIN
os 20.347 Corr. Rise . 3.1137

7 01 - 19.3
8 82 20.31 10
80 17.252 X 3000
983 - 20.350
10 4 20.350
115 20 80.310 8. + 80 . 37.599 - Total Heat . -9341 .
18.799 Accessories 77 .
8s + 0 Corr. Heat ... -9264 .

MINUTES
18.799
PERIOD
)FINAL
2 log. Corr. Heat - 3.96679

FINAL PERIOD 1
Sum 99.109 " W .07362
Fall.....
(5
86.215 + .026
50
+ .0052
Diff. log .. 3.8937
Diff . 12.894 Rate V '
log. diff . 1.11050 Antilog. equals 7.832
V .0036
log. V' - V. 2.94448
CALORIES 7832 .
Colog . O -0 - 1.52073 V: V + .0088
X 1.8
log . 1.57571
Antilog + .0367
Mean tº , 0' 20.329
B. t. u . 14097 .
+ 5 V .0180
Rad. Corr 0187 om 17.243
Ꮎ . 0 ... 3.086
16 20.321
The correction for radiation is calculated by means of the Reg

nault - Pfoundler formula :
V' – V Og + O 0
1
C = NV +
( ºs2
+
Σ 0 - Ne
Where N = number of time units (minutes) in combustion period .

V rate of fall of temperature of calorimeter during
initial period.
V ' = rate of fall of temperature of calorimeter during
final period .
= mean temperature of calorimeter during initial period.
Ꮎ = mean temperature of calorimeter during final period .
02 , 02......0 , = temperature at the end of first, second
1
.... nth minutes of the combustion period.

e = temperature at moment of ignition.
Applying this formula to the example given , we find that the

first reduction of the observations is made in the second column.
This being completed , a summation of the first four observations,
in the main period, is made in the left column (80.310) . This
is combined with the other expressions within the bracket by
transferring their values from the center column. Logarithms
are used to effect the reduction with the fractional form , and the
final reduction is also assisted in the same manner. The radiation
correction is added to the difference between the final (e's) and
initial (80) temperature of the main period . This gives the
corrected rise , which , multiplied by the hydrothermal value of the
system , will give the total heat evolved . From the total heat
we subtract that due to the accessory combustions, and this cor
rected heat, divided by the weight of the sample, is the calorific
power of the substance, expressed in calories per gram. It is
usual to give the value of a fuel also in British thermal units, which
is done by multiplying the calorific power by 1.8.
The correction for accessory combustions is made in a very simple
manner from the known weight and heat of combustion . Thus,
in the example, the weight of the iron wire used was 0.0325 gram ,
and this, times the specific heat of combustion, gives the calories
(52) to be deducted.
CALORIMETRY . 205
The following specific heats of combustion are accurately known :

Pure iron or soft iron wire, 1601 calories per gram .
66
Naphthalene .. 9628
Cane-sugar . 3959
Benzoic acid 6322
The Hydrothermal Equivalent. — This constant is usually furnished

with the instrument, but in case of change of parts or loss of its
recorded value it may be easily found by burning a known weight
of a standard substance, such as benzoic acid , the heat of com
bustion of which is very accurately known .
The following quantities are involved , and the relation is ex
pressed by the formula
H == (W + X) T
-
Where H the total heat liberated ,

W = the weight of water used ,
T = corrected temperature rise,
X = the desired hydrothermal equivalent.
H , the total heat , is found from the weight and heat of com
bustion of the standard substance used plus the heat derived
from the accessory combustions, including nitric acid and the iron
wire, etc.
ACCURACY OF THE METHOD. — The actual accuracy attained
will largely depend upon the care and expertness of the operator.
Conditions outside the operator's control may considerably affect
the accuracy attainable . Violent draughts of air, changing tem
perature of the room , dampness of the interior chamber, or loose
ness of covers by which currents of air may pass through the in
terior chambers, are sources of error. The chief source of error
lies in the temperature observations, and it follows that these must
be very carefully made.
With care and reasonably good conditions the total error of
estimation should not be greater than 0.5% , and greater deviations
should be cause for repetition or investigation of the apparatus.
The time required to make an estimation with a compact and
conveniently arranged apparatus should not exceed 40 minutes,

and with a systematic arrangement the time may be much short
ened .
THE PARR CALORIMETER .
This calorimeter is a modification of the Thompson form , in

that it uses the oxidizing power of chemicals to effect the com
bustion of the fuel. It avoids one of the serious sources of error
of the latter by the use of sodium peroxide, which, in decomposing,
oxidizes the fuel and at the same time absorbs the products of com
bution so that they do not escape through the surrounding water.
Fig. 72. -PARR CALORIMETER.

. CALORIMETRY . 207
The exterior jacket is not unlike the outer envelope of the At

water-Berthelot calorimeter. The bomb is replaced with a small
chamber, D, capable of revolution about a vertical axis. A proper
weight (0.5 to 1 gram of the fuel) mixed with 16 to 32 grams of
sodium peroxide and the accessory chemicals is placed within this
chamber and the cap is fitted on tightly. It is then immersed in
the calorimeter-pail containing 2000 cc. of water. Stirring is
effected by the propeller wings fastened to the rotating chamber;
a speed of about 150 revolutions is recommended . When equili
brium is reached , as shown by the thermometer, the charge is
ignited by means of a red -hot wire introduced through the valve
stem. Ignition takes place with rise of temperature. The obser
vations to be made can be recorded exactly, as those outlined for
the bomb calorimeter, or the directions given with the apparatus
may be followed .
The same precautions apply to this calorimetric measurement
as to others, and in addition the mixture between the fuel and the
peroxide must be thorough , and both should be finely powdered ,
the fuel to 100 -mesh and the peroxide to 50 -mesh. The operator
should tap the cylinder gently on the bottom , so that all the con
tained material shall gather there, for any adhering to the sides
or about the valve will escape combustion . Dampness must be
carefully avoided , as it is likely to ignite the charge and ruin the
determination .
With anthracite and some other substances not readily com
bustible it is necessary to add an accelerator to the charge. This
is composed of tartaric acid 33% and potassium persulphate 66% .
Even with a liberal addition of this mixture the combustion is not
always complete. One and one- half grams of this mixture is used ,
with 12 grams of sodium peroxide, for the combustion of 0.5 gram
of coal .
The corrections needed are : ( 1 ) For radiation losses . These are
made exactly as in the bomb calorimeter, though the printed direc
tions from the makers do not mention this point . Reliance is
placed in exactly balancing the loss and gain by starting at a suffi
ciently low temperature, that recommended being 3° F. below room
temperature. Experience with any calorimeter shows that this
is rarely the case , and it will not do to ignore this point. (2) Cor
rections due to auxiliary chemical reactions, aside from the union
of the oxygen with the fuel . These consist of (a) the decomposition
of Na202, ) forming Na 20 + 0 ; (b) the union of sodium oxide with
carbon dioxide, Na , 0 X CO , = No , C0g; (c) the union of sodium

oxide and water, Na ,0 + 1,0 = 2Na OH ; (d) the decomposition
of sodium peroxide by water, Na , 0 , + H20 2NaOH + 0. If
2
the accelerator is used , the following also enter, namely : (e) De

composition of the potassium persulphate into potassium sulphate,
oxygen and sulphuric anhydride, with the subsequent formation
of sodium sulphate due to the presence of sodium hydroxide; (f)
the combustion of the tartaric acid, with its attendant series of
reactions.
The first four of these are corrected by the use of an arbitrary
factor which has been found by experiment to be within about
1 % of the truth . This factor amounts to the deduction of 27%
of the heat evolved after making all other corrections. Corrections
for (e) and (f) are made by running a charge consisting of 0.5 gram
of tartaric acid, 1 gram of persulphate and 6 grams of peroxide.
The calories evolved are to be corrected for the iron wire and de
ducted from like combustions with the accelerator.
The makers give the tartaric acid equivalent as 1779 B. t. u.
(988 calories) and the persulphate equivalent as 330 B. t. u.
( 183 calories ). The iron wire equivalent (wt . of iron being 0.4
gram , its temperature 700° F. , and sp. ht . 0.114) is 0.4 X 700 X
0.114 = 31.9 B. t. u . ( 18 calories ). This rather crude approxi
mation probably gives results within the limits of accuracy of
the apparatus .
( 3) A deduction is made for the estimated heat carried by the
hot wire used for ignition. The temperature of this wire will vary
from 700 ° to 1200 ° F. ( 3899-667 ° C.) above the room temperature.
The temperature difference, times the specific heat, times the weight
of the wire gives the correction.
The specific heat of soft iron is from 0.111 to 0.114.
copper 0.092 .
From the observations made, a computation parallel to that shown

CALORIMETRY . 209
in full for the bomb is made, or, when accuracy is not required , the
directions accompanying the apparatus may be followed , and for
this purpose they are here given .
DIRECTIONS FOR USING THE PARR STANDARD CALORIMETER .
The calorimeter should be placed on a good firm desk or table.

The power needed is exceedingly slight, the smallest electric or
water motor being ample. Revolve the pulley by means of a
loose cord at the rate of about 150 r. p. m .
The general arrangement of the parts is shown in Fig. 72. How
ever, the can should be removed from the instrument for filling
with water, and care observed that no water remains on the outside
or is allowed to spill over into the air spaces of the insulating vessels.
Exactly two liters of water (preferably distilled ) are used, and it
should have a temperature about 3° F. ( 1.65 ° C.) below that of the
room . That is, approximately three- fourths ( 1 ) of the total
rise in temperature should occur before the temperature of the
room is reached .
The glass jar is for the chemical, and should be kept carefully
closed and clamped to prevent absorption of moisture from the
air . For this reason , also, the contents of only one small can of
chemical is emptied into the jar at one time. There must be also
room left for the measuring-cup and handle entire.
The cartridge for filling should be perfectly dry inside and out.
If water has been admitted to the valve, remove it from the stem
by screwing off the knurled top and drying all parts thoroughly.
To prepare the cartridge for filling, screw on the bottom firmly so
as to avoid any possibility of admitting water to the interior, then
place on a sheet of white paper.
The coal is prepared by grinding in a mortar and passing through
the sieve of 0.25 mm . mesh , or 100 to the inch . Coals containing
over 2.5-3% of water should have the water removed . In such
cases the exact charge of the commercially dry coal is weighed out
and dried for an hour at 1050-110° C. ( 220 °-230 ° F.) , then trans
ferred to the cartridge.
PROCEDURE FOR ALL ORDINARY SOFT COALS.— (Not anthracitic
or true anthracites . ) Exactly 1 gram is taken and dried as above

indicated .
One full measure of chemical is put in the cartridge and the
coal added . Screw on the top with valve or loop of fine wire for
electrical ignition , and shake vigorously to thoroughly mix the
contents. In shaking the cartridge it is well to hold the thumb
against the valve-stem (at E, Fig. 72 ) to prevent its opening and
admitting the material into the stem . When the mixing is com
plete , tap the cartridge lightly to settle the contents and to shake
all the material from the upper part of the cylinder. Tap the
valve, also, to see that it works freely. Put on the spring clips
with vanes. The cartridge is now put in place, the can with water
being already in position. Adjust the cover. Insert the ther
mometer so that the lower end of the bulb will be about midway
towards the bottom of the can , place the pulley on the stem and
connect with the motor. The cartridge should turn to the right,
or as the hands of a watch , thus deflecting the currents downward .
After about three minutes the first reading of the thermometer
may be taken .
Ignite with the short piece of iron wire which is held or rests
on a wire gauze in a Bunsen flame till red hot; drop quickly into
the opening at the upper end of the valve . With the pinchers
the valve is now pressed completely down and released with a
quick motion , so as to prevent the escape of heated air from the
interior.
All the time, of course , the revolving of the cartridge continues.
The combustion should be indicated by a rapid rise of the mercury ,
which reaches its maximum height after from four to five minutes .
Make note of the final temperature, and compute thus:
Calculation . - Subtract the correction factor for the heat of the
wire - namely, 0.022 ° F. (0.0122 ° C .) — and multiply the re
mainder by 3110. The product will be the number of British
thermal units per pound of coal. Where the electric method of
ignition is used this correction for the wire may be disregarded .
To dismantle, remove the thermometer, pulley and cover ; then
ake out the can and contents entire, so that the lifting out of the
artridge will not drip water into the dry parts of the instrument.
CALORIMETRY . 211
Remove the spring slips and unscrew the ends. Place the cartridge
in a dish of hot water. The contents will dissolve out quickly.
The cartridge and ends should be at once thoroughly dried, when
they will be ready for a new test.
Note.-( a ) With the electric method of ignition the fine wire is
wrapped firmly, with good contact, around the ends of the ter
minals , and bent (:-shaped so as to extend below about half an inch.
Do not have too great a length of wire for the current. Make
a preliminary test without screwing the stem into the bomb, so
that the action can be seen . The wire should quickly become in
candescent. ( 6 ) Lignites and fuels of low calorific content may be
too greatly diluted in the above ratio of material for perfect com
bustion . In such cases weigh out one gram and dry. Take
14 measure of chemical. The factor for computing will be 1555 in
stead of 3110.
FOR ALL ANTHRACITES AND ANTHRACITIC SOFT COALS , COKES ,
ETC. - Grind and sift the coal as usual. Weigh out exactly 4 gram
for the test ; drying may be omitted if the water content is below
24 or 3 % . Transfer the sample to the cartridge and add exactly
14 grams of the “ special chemical” for anthracites , petroleums,
etc. Then add one and one- quarter measure of the ordinary chemical .
Shake thoroughly, place in the can , assemble and complete the
combustion in the usual manner .
The correction factor for the special chemical and hot wire
is marked on the label. Subtract this number from the total
rise in temperature and compute thus: Multiply the remainder
by 3110. The product will be the British thermal units per pound
of fuel .
FOR PETROLEUM , ETC.- Proceed exactly as for anthracites. Pre
pare a small, light, 15 c.c. weighing-flask with perforated cork
and dropping -tube, with common rubber-bulb cap . Fill the
flask about 3 full of oil and weigh carefully. Place in the cartridge
measure of chemical and 14 grams exactly weighed out of the
“ special chemical.” Then , by means of the dropping -tube, add
about 0.300 gram (25–30 drops) of oil . Stir the mixture in the
cartridge thoroughly with a piece of wire. Add the remainder
that is , one measure of ordinary chemical. Remove the wire,
cleaning it as well as may be in this upper layer and without jar

ring off any particles. Screw on the cap, shake well, place in the
calorimeter and ignite as usual .. Weigh back the flask carefully
and determine the amount of oil taken by difference. Compute
by formula instead of using the factor, thus: Correcting as before
for the wire and extra chemical employed , letting " r " equal the
“ r " x .73 X 2130 .
corrected rise in temperature; then wt . of oil .
British
thermal units per pound of oil .
NOTES .
(a) The factor 3110 is deducted as follows : The water used,

plus the water equivalent of the metal in the instrument, amounts
to 2130 grams. In the reaction 73 % of the heat is due to combus
tion of the coal and 27% is due to the heat of combination of CO ,
and H ,0 with the chemical. If, now, ļ gram of coal causes
2130 grams of water to rise " r " degrees, and if only 73% of this
is due to combustion , then 73 X 2130 X 3 X “ s ” equals rise in
temperature that would result from combustion of an equal weight
(2130 grams) of coal -73 X 2130 X 2-equals 3110. Where i
gram of coal is taken this factor, of course, becomes 1555.
(6 ) The ignition wire is of soft iron 21 mm . in diameter, or No.
11 gauge, and 1 cm . long. It should weigh approximately 0.330
gram . It looses very little by use . When by oxidation the weight
falls much below 0.300 gram a new wire should be substituted.
The correction for the wire may be calculated thus : Taking 0.114
as the specific heat and 1200º - 1300 ° F. (667° -723 ° C.) as the tem
1250° x .330 x .114
perature at a red heat . Then 0.022 ° (0.0122 °
2130
C. ) . This agrees closely, also, with the factor obtained experi

mentally.. A variation of a hundred degrees in the initial tem
perature of the wire causes a difference in the final reading of only
0.002° F. , or six thermal units, and may be disregarded.
(c) Chemicals vary, moreover the sodium peroxide may have
become contaminated with moisture from the air. It is advisable ,
CALORIMETRY . 213
if question arises, to run a blank thus: Weigh out 11 grams of the

“ special chemical” mixture of tartaric acid and potassium per
sulphate, place in the cartridge, add also one measure of the ordi
nary chemical sodium peroxide ; shake and ignite in the usual
manner. Duplicates should agree closely with each other and with
the standardized factor marked on the label .
( d ) Do not bring the instrument from a cold room to work at
once in a warm room , or vice versa . An hour, at least, should be
given for equalization of temperature.
(e) Do not throw a mixture of chemical and unburned coal into
water. It may ignite violently. Similarly, a drop of water left
inside of the valve may work into the charge during the shaking
and ignite it prematurely.
( 1) It is well to test the action of the valve by dropping the wire
through on to a sheet of paper a number of times, holding the stem
in the hand and dropping the wire as in igniting a charge. It will
be noted that the wire is allowed to lodge at the valve and then is
released by pressure on the top . Too sudden a pressure and release,
or a failure to press the valve completely down, may result in
catching the wire before it has cleared the valve. In such a case ,
of course, the charge would fail of ignition .
(9) The above directions presume the use of a Fahrenheit ther
mometer. In case a centigrade thermometer is used, the use of the
factor 3110 gives the result in kilocalories. To change kilocalories
to British thermal units, multiply by 1.8. Also the correction
factors as herein given should be divided by 1.8. Thus, for the
hot wire the correction would be 0.012° C.
(h) A piece of fine emery cloth around the cartridge will facili
tate holding the same when unscrewing the ends.
The advantages of the Parr calorimeter over the bomb are its
independence of compressed oxygen and its lower price. The
possible error is much greater than that of the bomb, though dupli
cates may check very closely. The inaccuracy due to temperature
readings on the thermometer furnished (divided to 26) may
amount to over 1 % and the error of estimation of the correction
of the hot wire may be 0.1 % .
A liberable estimate of the probable error will be from 1 to 2% .
The time taken for an estimation is about the same as that for
the bomb .
JUNKER's CALORIMETER .
The principle upon which Junker's calorimeter ( for gas and oils)
depends is that of burning the substance with a free access of air,
and passing the products of combustion through a cooling chamber.
The thoroughly cooled gases are discharged into the atmosphere,
and the rise in temperature of the cooling medium affords the
measure of the heat evolved .
It is readily seen that this is very similar to industrial heating ,
and in fact the Junker's calorimeter is simply a small tubular
boiler of relatively large heating area . Fig. 73 shows the calorim
eter proper in section plan and elevation. It will be seen that
the gases of combustion ascend to 29 , where they are directed
downward through the water- cooled tubes, 30, and finally emerge,
fully cooled , through exit , 32, to the atmosphere. The cooling
water enters at 1,. part of it overflowing at 5 , the constant -head
device ; the feed-water descends in tube, 3, to the valve, 9, and
thermometer, 12 , whence it enters the calorimeter chamber at 13 ,
and is distributed by a perforated annular sheet, 14. The water
gradually rises through the mixers, 17 , past the thermometer, 43 ,
and is discharged without suction over the funnel lip , 20.
To maintain a constant and measurable supply of gas to the
calorimeter, a very delicate pressure regulator and a gas meter
divided to 0.001 liter are needed . A U -tube pressure gauge should
also be used to read the pressure under which the meter is working.
The thermometers used for measuring the temperature of the in
flowing and outflowing water are divided to 10 °, and have a
range from 0 to 50 °. The temperature of the air and exit gases
is taken with an ordinary thermometer.
Apparatus Arranged for Use . — For use the calorimeter and ad
juncts should be arranged as follows : To the gas main or supply
attach the pressure regulator, and to this connect the gas meter.
Between the meter and the calorimeter burner is inserted the
U -tube pressure gauge charged with water, colored if desired. The
A.
‫رام‬
43
)(۱
ULLE
20)
7:10:00
6
519 .:
38 21
10
29
28
36
12
27
39
31
32 33
-34
35/
40
15
30
Fig . 7. - JUNKER Gas CALORIMETER .

Bunsen burner to which the tube from the meter is attached must
afford complete combustion of the gas. The water-supply tube is
attached to the faucet, and the two overflow pipes run to a waste
pipe which must be close by.
Starting the Calorimeter .-- Water is turned on of sufficient volume
to overflow continually through tube, 5, the valve, 9, being suffi
ciently open to allow a flow of, say, 200 c.c. per minute through
the calorimeter. Note that no drip of water takes place from tube,
35, as this would indicate a leak in the cooling chamber. Being
satisfied as to this point, a glass graduate is placed here to catch
the condensed water of combustion .
Gas is turned on from the main , the Bunsen -burner valve, 22,
being closed. The needle of the gas meter should become stationary
and remain so, otherwise a leak of gas is indicated beyond the meter,
and this must be remedied. Everything being in proper condition ,
the Bunsen burner is taken out and lighted, such a regulation being
given that the flame evolves about 1000 to 1200 calories per hour ;
this will represent a gas consumption in the case of
Illuminating gas.. 100 to 400 liters per hour.

Hydrogen .. ..200 " 800
Water - gas. .300 " 1000
Mixed gas. 400 1600
(6 66
Producer- gas .500 2000
After replacing the burner, the flame of which must not come
below the water- jacketed level , a few minutes are allowed for the
heating to take effect, and then adjustments in the rate of flow of
the water are made, such that the exit gases are cooled practically
to room temperature, and the rise in temperature of the water is
about 10° C. This rise should be equally distributed on each side
of the room temperature - that is , the feed -water should be as
much below that of the room as the exit-water is above. Error
from radiation is then at a minimum , but this condition can
not always be had . Corrections for radiation cannot well be
made, on account of the gradual change in temperature from
CALORIMETRY . 217
top to bottom, and therefore this loss should be kept as low as

possible.
The damper for the exit gases should not be shut off so much
as to cause hot air to escape through the air inlets, and likewise
draughts should be avoided, as they may cause back currents. The
gas pressure should be recorded, and it should remain perfectly
constant .
Taking Observations. — Testing gas . - A few minutes must be
allowed for the attainment of equilibrium and the steady reading
of the thermometers registering inflowing and exit water, also the
exit gas.
The following observations are to be taken :
W, the amount of exit water flowing during the time of burning.

G, the amount of gas burned .
T ,, temperature of inflowing water to 0.01 C.
1
T 21 exit water
T , room or gas.
Tu exit gas.
P 1 ) pressure of gas to burner in mm . mercury.
P2, barometric pressure of the atmosphere.
When the temperature becomes steady the tube carrying the

exit water is changed, exactly as the meter needle crosses zero,
to a weighed receiving dish . Record of T, and T , is now taken
at equal intervals of time, frequent enough to bar the possibility of
an unnoticed change. T3, ) T.,
T4 P ,, and P ,, can be taken at once
and should not vary , though it is well to check T ,
Exactly as the meter needle crosses the determined volume to
be burned , say 3 liters, the exit tube is replaced in the waste-pipe
and the receiving dish is weighed . The gain in weight is W.
Record is also taken of the water of condensation which has col
lected during the same time as W.
Testing oils.- In testing oils and inflammable liquids the general
procedure is the same as above, the only variation being in the
manner of estimating the amount of material burned .
The liquid is burned in a lamp, which first volatilizes it and

then burns the gas, as is done in the Turner gasolene -blast lamp.
Supply of liquid and pressure is insured by compressing air in a
chamber above the liquid . To light this lamp it must be pre
heated , either by a Bunsen flame or by benzine ignited in a receiv
ing cup for this purpose . When lighted the lamp is placed on a
balance, similar to the standard - candle balance, and brought to
position in the calorimeter. When the condition of constant tem
perature is reached, the balance is brought nearly to equilibrium .
The loss of weight, due to burning, will soon cause the pointer to
move past the zero point; at this moment the test begins. The
tube is then transferred to the weighing vessel and the observations
taken . A weight representing the quantity of substance it is
desired to burn is taken from the balance; the equilibrium thus
destroyed will be regained by loss in burning, and when this occurs
the test ends, just as in the case of testing gas . The total heat
generated , W X R, divided by the weight of liquid burned , will
give the calorific power per unit weight.
W XR
Calculation . The formula H , where H is the calorific
G
power per unit ( volume or weight) and G is this volume or weight

burned, is applicable direct in the case of liquids, but gases , on
account of the effect of pressure and temperature on volume, must
be reduced to standard conditions. For this purpose we employ
PXG X 273
the formula G 0
where
760 XT
G. 0
volume at 0 ° C. and 760 mm .
P = observed pressure of gas supply P , + Pz. 1
G volume of gas indicated by meter at T.

T = absolute temperature of gas = T , + 273 .
3
W XR
For standard conditions, then , M ‫ون‬ The calculation
G
and application of this formula are rendered clear by the accom

panying example.
CALORIMETRY . 219
CALORIFIC POWER OF Gas .
G 3.000 liter.
P 2.4 mm .
P2 748 . mm .
750.4 mm . Pi + P2 - P.-
T, 17.63 deg.
T, 2 1st , 26.80
2d , 26.74
3d , 26.73
66
4th , 26.73
5th , 26.75
6th , 26.77
Mean T 2, 26.753 deg. 26.753 17.63 9.123 = R.

T, = 22 . 22 + 273 = T.
T4 = 22
W 1.823 liter, W XR 16.6312
PXG X 273 750.4 X 3.0000 X 273 .

G. 2.74 liter.
760 XT 760 ( 22. + 273. )
16.6312
H 5.897 calories per liter.
2.82
Calorific power per cubic meter 5.897 X 1000 5897. cal .
The use of a wet meter, and often a wet -pressure regulator , re

sults in the delivery of gas saturated with water vapor. This
usually is not of importance, as the gas in the mains is already
saturated, but should a dry gas be furnished or results be called
for on a dry basis, a correction depending upon temperature and
pressure will have to be made.
Accuracy of the Instrument. - With care as to details the Junker
calorimeter is an accurate instrument. Its main source of error
is in the lack of an exact radiation correction , but this may be partly
avoided by the precautions previously mentioned . The accuracy

with which the thermometers may be read controls the total rise,
R, and hence the amount of water, W. With delicate thermometers
the value of R may be so controlled as to nearly eliminate the
radiation loss, but less sensitive thermometers necessitate a large
value for R, that the error of reading be not too great, and this
involves a corresponding increase in the radiation loss.
The Low Thermal Value. — Most uses to which fuels are put do
not presuppose the cooling of the products of combustion to ordinary
temperatures, or even to the boiling point of water, and hence
the latent heat of condensation of the water vapor present, as well
as the sensible heat of the gases, is not utilized. Ordinary cal
orimetric measurements include this inefficient heat, which should
be deducted, according to the particular use intended, in order
to give relative practical values. The sensible heat lost should
be low, and as it varies with each case no general correction is
possible, but the latent heat of the water vapor may be 10% of
the calorific power, and it should be deducted, the amount de
pending upon the analysis of the fuel. Its value is readily cal
culated , knowing the amount of hydrogen present.
This corrected value is known as the low thermal value .
APPROXIMATE DETERMINATION OF THE CALORIFIC POWER OF A

FUEL BY MEANS OF THE MAHLER AND GOUTAL FORMULAS.
Few industrial laboratories are equipped with calorimeters,

and it is very often useful to be able to find , within 1 % , the ap
proximate calorific power of a fuel . A number of formulas have
been proposed for the calculations of the calorific power from
the ultimate analysis or the approximate analysis of the fuel.
Among them may be mentioned the Vivien, von Jüptner, Goutal
and Mahler formulas.
The only two that have stood the test of practice are the Goutal
and the Mahler. Fortunately they complete each other. The
Goutal makes use of the data obtained from an ordinary proxi
mate analysis, the Mahler of the ultimate analysis.
CALORIMETRY . 221
Mahler Formula . Calorific Power Q.

8140 C + 34500 H - 3000 ( O + N )
Q' . 100
This is further simplified if we write

O + N = 100 – (C + H)H
Substituting in the equation, we have finally
Q.
Q '. 111.40 C + 375 H -
3000
in calories per kilogram of fuel.

The comparison of columns 10 and 11 in table 22 shows how
closely the computed results agree with the experimental deter
minations. The formula does not give accurate results in case
of cannels , lignites, peat and woods.
Goutal Formula . — The data required are the following:
C Fixed Carbon .
V Volatile matter.
a A variable factor.
" a " is a function of the volatile matter of the fuel, abstraction

being made of the moisture and ash .
Calorific Power = 82 C + a V
For all fuels with volatile matter V' below 40% , the values of
" a "" are given in table 12 .
Moisture Determination . — The moisture is determined on à
2-gram sample heated for an hour in an air bath at 115° C. The

constant weight should be verified by two weighings.
Volatile Matter.-In a crucible of 30 c.c. capacity are placed
5 grams of the powdered fuel ; the crucible is placed uncovered
over a Berzelius burner. The flame is to be maintained so as to
surround completely the crucible. After all the illuminants have
been entirely driven off, the heating should be continued for three
more minutes.
Ash . —- Two grams of the fuel are burned in a muffie, the heating
to be done very carefully and not too suddenly.
222 INDUSTRIAL FURSACES AND METHODS OF CONTROL .
Example . The analysis of the anthracitic coal of Le Creusot is

the following:
Fixed carbon ... .86.70% С
Volatile matter . . 10.05 V
Ash ..... 1.45
Moisture . 1.80
The volatile matter V' , calculated as if there was neither ash or
moisture, is
V X 100 10.05 X 100
V ' 10.4
C + V 86.70 + 10.5
TABLE 12 .
GOUTAL FACTORS.
V" a &
% Calories % Calories
156789012345GTBU938
1 to 5 100 23 105
145 24 104
142 25 103
139 26 102
136 27 101
133 28 100
130 29 99
127 30 98
124 31 97
122 32 97
120 33 96
117 34 95
115 35 94
113 36 91
112 37 88
110 38 85
109 39 82
108 40 80
22 107
From table 12 we find that it corresponds to a value of “ a ”

equal to 129 calories.
Substituting in the formula, we have
Calorific Power 82 X 86.7 + 129 X 10.05 8406 calories .
By actual determination of the calorific power by means of the
Mahler bomb it was found to be 8404 calories . The results ob
tained are generally within 1 of the experimental determinations;
an error of 2 % is quite exceptional and is observed only with the
anthracites or lignites.
CHAPTER XIV .
ELEMENTARY FUEL ANALYSIS — THE ULTIMATE ANALYSIS OF
FUELS, COAL, COKE, ETC.
Determination of Carbon , Hydrogen and Ash . — These determina

tions may be made in a single combustion . The apparatus used
is an ordinary combustion furnace, the variations being that either
a porcelain or a wrought-iron tube may be used , and that lead
chromate is used in addition to cupric oxide.
All material used in charging the tube, and the tube itself, must
be freed from moisture previous to the combustion , by thorough
ignition with the passage of a current of dry air. The tube is
charged in the central portion , for about 15 c.m., with coarse ,
strongly ignited cupric oxide ; the forward end contains a sheet
iron boat, 20 c.m. long, charged with powdered lead chromate, to
retain sulphur, mixed with coarse pumice. In case a wrought
iron tube is used in place of the porcelain , it should be strongly
ignited, previous to charging, and steam passed through it. This
oxidizes the inside and destroys the porosity of the untreated iron .
The absorption train consists of a U -tube filled with fragments
of pumice, wet with strong sulphuric acid , a second U -tube of cal
cium chloride, a C -tube charged with active soda -lime, a U -tube
of sulphuric acid and pumice, and a second U -tube of sulphuric
acid and pumice to act as a guard -tube. The entrance train is
composed of a soda - lime tube and a sulphuric acid pumice U-tube .
The same drying agent should be used at each end .
The apparatus should be tested , by ignition and passage of air,
until satisfactory constant weights are obtained for the sulphuric
acid absorption tube and the ( 3) calcium chloride, soda -lime and
sulphuric -acid tubes weighed co'lectively.
To make the combustion , al ow the furnace to cool and intro
duce a weighed porcelain boat, containing 0.5 to 1 gram of the
ՄՄ WW
Calcium
Chloride
Sulphuric
Acid
Sulphuric
Acid
Calcium
Chloride
Soda .
Lime
Sulphuric
Acid
sulphuric
Acid
Fig. 74. - TRAIN OF APPARATUS FOR THE ELEMENTARY ANALYSIS OF FUELS.

ELEMENTARY FUEL ANALYSIS . 225
fuel sample, dried at 104-107 ° C. The boat should rest on a sheet

of platinum foil . The tube is now closed and the burners under
the lead chromate and cupric oxide lighted. The lead chromate
should be kept at a red heat and the cupric oxide at a bright red
heat . When the rest of the tube is properly heated the burners
under the boat may be lighted, and it gradually brought to a bright
red. The air current should be drawn through the tube at such
a rate that about two bubbles per second pass through the sul
phuric acid . The combustion should be complete in 30 minutes
and the boat can be drawn, cooled and weighed. Its gain in weight,
plus f the sulphur content, equals the mineral matter in the coal.
Its direct gain in weight is the ash .
The gain in weight of the sulphuric -acid tube equals the water
produced and absorbed, which is converted into hydrogen by 0.1119.
It is not necessary to take the weight of the calcium chloride tube
unless back -suction has taken place. The gain in weight of the
three tubes is the carbon dioxide absorbed, and this is converted
into carbon by 0.27273 .
These results must be converted to the basis of the air -dried
sample, using the percentage of moisture found.
Determination of Moisture. - Dry one gram of the sample in an
open dish in an air bath , at 104-107° C. , for one hour. Cool in a
desiccator and weigh quickly. The loss in weight is the moisture
present.
Determination of the Sulphur. - Eschka's method modified accord
ing to Miller. One gram of the finely pulverized coal is mixed with
one gram of light magnesium oxide and 0.5 gram of anhydrous
sodium carbonate, in a platinum dish having a capacity of 75 to
100 c.c., or a Meissen porcelain crucible, and heated with an alcohol
lamp. When great accuracy is not required a gas flame may be
used , the dish being protected, as far as possible, from the products
of combustion which contain sulphur. The mixture is stirred
frequently with a stout platinum wire or glass rod , and the heat
applied gradually, especially with soft coals, until strong glowing
has ceased ; then the heat is increased until in fifteen minutes the
bottom of the dish is at a low red heat, which is maintained until
all of the carbon is burned out. The residue, which should be
nearly white and porous, is transferred to a No. 2 beaker with

about 50 c.c. of water ; 15 c.c. of bromine water are added, and the
whole boiled for at least 5 minutes, allowed to settle, and decanted
through a filter; boiled a second and third time with 30 C.C. of
water, and then washed very thoroughly with hot water.
The filtrate is acidified with hydrochloric acid so as to have an
excess of about 1 c.c. and boiled to expel any free bromine. Then
to the hot solution barium chloride ( 10% ) solution is added , drop
by drop, until 10 c.c. have been added , and the solution boiled or
heated on a water bath until the precipitate settles quickly ; filter,
ignite and weigh .
From the weight of barium sulphate the percentage of sulphur
is calculated . Factor, 0.13736.
Determination of Nitrogen . - Kjeldahl's method. Place 1 to 11
grams of the fuel in a long narrow -necked flask , preferably made of
Jena hard glass, of about 150 c.c. capacity, and add 20 c.c. of con
centrated sulphuric acid . Add 0.5 gram of phosphorus pentoxide
and place the flask in an inclined position over a flame. Heat the
contents to boiling and boil for about two hours. A vigorous re
action takes place, resulting in the production of carbon dioxide
and sulphur dioxide while the fuel slowly dissolves. Solution must
be complete and the liquid a brownish yellow before the next step
of oxidation .
Small portions of potassium permanganate are added with care,
the oxidation being accompanied by detonations; the solution
clears up and finally becomes a bluish - green. This is a signal that
the oxidation is complete, and the contents are now digested at a
gentle heat for ten minutes, then allowed to cool and poured into
a 750 -c.c. distilling flask , containing about 150 c.c. of water. The
distilling flask is closed by a double-bore rubber stopper through
which passes a separatory funnel and a glass tube, bent to form an
inverted U. This glass tube is connected with a 50 -c.c. pipette,
which in turn enters a 250-c.c. Erlenmeyer flask , connected to a
U -tube, both of which are immersed in a vessel of water. The
U -tube and flask contain 30 c.c. of } normal sulphuric acid . Now
40 c.c. of 1.3 sp . gr. (30%) sodium or potassium hydroxide is run
in from the separatory funnel , and gentle distillation is begun, con
ELEMENTARY FUEL ANALYSIS . 227
tinuing until one -half of the contents has passed over, or, say,
20 to 30 minutes. The addition of aa few pieces of zinc will prevent
bumping during distillation, but should nitrates be present in the
chemicals used) they will be reduced to ammonia and thus cause
error . Any foaming or tendency of the liquid to spray over with
the steam is to be carefully avoided .
The total distillate is collected in the flask and the excess acid
titrated with alkali, standard solution, using litmus, methyl orange
or phenacetolin ( phenol phtalein ) as an indicator. The ammonia
equivalent, calculated, is reduced to nitrogen by the factor 0.82278.
Note.- The solution of the fuel in the acid is assisted by metallic
oxides and some salts. The most active are cupric oxide and
mercuric oxide. The latter must be prepared wet, as the dry
preparation contains nitrogen compounds that will be reduced, in
the distillation, by the zinc. The sulphates of these metals answer
as well. Use about 0.75 gram. If these are used they must be
precipitated , in the distillation flask , by an excess of 4% potassium
sulphide solution. The proper amount can be found by testing
0.5 gram of the salt in solution. An excess must be used to
decompose the mercuro -ammonium compounds. Even twice the
amount actually required will usually do no harm , but more may
result in a slight evolution of hydrogen sulphide.
Determination of Oxygen . - Oxygen cannot be directly deter
mined with accuracy . If it were certain that the sulphur was
present as pyrite, and that this is completely converted to ferric
oxide in the ash, the oxygen would be, by difference, 100 minus the
sum of carbon , hydrogen, nitrogen , ash and of the sulphur.
This is the best rule for the present, but the ash should be examined
for sulphates and the rule modified where cause is found for its
alteration .
PART III
DESIGN OF FURNACES
POWDERED FUEL
CHIMNEYS, REGENERATORS, RECUPER
ATORS, REVERSING VALVES, PORTS,

HEARTHS, POWDERED FUEL
BY
A. L. J. QUENEAU
PART III .
CHAPTER XV .
CHIMNEYS .
Chimneys are always necessary for evacuating the waste products

into the atmosphere, no matter what the system of draught may
be - natural or forced. In order to insure a draught in the chimney,
the products of combustion must be delivered hot at the base of
the stack. It will be shown that a temperature of 200 ° C. is amply
sufficient. As the chimney operates through the difference of tem
peratures between the atmospheric air and the products of com
bustion , it follows that, the lower the temperature of the latter at
the base of the stack, the better must be the construction of the
stack , in view of keeping the heat radiation at a minimum . This
is an important consideration if it is desired to operate the regen
erative furnaces with the maximum fuel economy. Brick stacks
are to be preferred for such purposes .
Calculation of the required height of chimney.-(The chimney
height is taken from the level of the producer grate to the summit
of the stack . )
Let L the height of the chimney .
ti the average temperature of the waste products
in the chimney .
a the dilatation coefficient of the air.
t the temperature of the atmospheric air.
P 1, the weight of a column of waste products in the
stack .
P the weight of a column of atmospheric air of a
like volume.
Ph the pressure at top of the stack .
The difference in pressure at the grate is :

( Ph + P ) – ( Ph + P. ) = P – P , ( 1)
This difference in pressure will produce the draught in the stack

equal to a column of atmospheric air of height , h .
The gases issue from the stack at a velocity
V = 12 g h (2)
( L + h ) ( 1 + at) = L ( 1 + at ,)
h
La (t , – t )
1 + at (3)
at being a very small quantity, it will be sufficiently accurate to

write
h La ( t , – t) ( 3) 1
Substituting this value of h in (2) ,
V = V2 g La (t , –- t))
Therefore :
V varies as v L (t, - t)
V is proportional to the square root of the height of the chimney
and to the difference in temperature. Let us find what temperature
will give the maximum velocity.
Let 8. be the density of the air at the temperature t , and pres
0
sure Ph + P1
The weight of air, Q , passing through a section S, per second ,
is
Q = SVO , S8 , 12
V2 ga L (t , – t)
Let 8o. be the density of the air at 0° C. and pressure Ph + P ,
8
Ö
Q = So 12 ga LL (t , – t) (4)
1 + at , 1 + at ,
CHIMNEYS . 233
In order to find the maximum value of Q , it is sufficient to find

-
the maximum of theexpression ( vt , -t since So. 2ga L = K

1 +at, &
is a constant for a given height , L.
-
Q = K V ti - t
1 tati
Differentiating and equating to 0, we have
1 tati 2a (t , t)
1
ti 2t + a
1
If t = 0 , a
T , = 273 ° C.
273
t - t
Let us find the values of the expression V g for differ
1 + at
ent temperatures for
t, 100 ° y = 7.4
200° 8.16
273° 8.255
300 ° 8.251
400 ° 8.110
500 ° 7.89
It follows that when the average temperature tı , of the waste

products, is 100 ° C. , the draught will be quite sufficient, and it will
seldom be necessary to go beyond 200 ° C. This will give about
300 ° C. at the base of the stack .
From equation (4) it is seen that the quantity Q is directly pro
portional to the section S , and also to the square root of the height;
it follows that it is more advantageous to increase the diameter
than the height. A height of 30 to 40 meters is generally sufficient
to insure a good draught. 1
The nature of the surface is very important. Peclet has shown
1 The case of deleterious fumes to be discharged into the atmosphere is not here taken
into consideration .
experimentally that the velocity of gas in chimneys made of brick ,

plate iron and cast iron are as the numbers 2.00, 3.16, 4.47.
It is, then, possible to pass as much, per unit of time, through a
cast- iron pipe one foot in diameter as in a brick flue two feet in
diameter, the other factors being equal.
The friction increases with the perimeter of the conduct. As,
perimeter
however, the section increases, the ratio decreases rapidly,
section
the two factors increasing and the friction becomes less important
perimeter
after a certain ratio has been reached. The ratio section
must be
as small as possible ; the circular section is, therefore, the best, and
other sections will become less efficient the farther they are removed
from the circle .
Elbows and contractions increasing greatly the friction , they
should be avoided as far as possible. All bends should have generous
diameters.
Section of Stack . – Redtenbacher's formula gives
K h
S
42L 84, L
S is the section in square meters .
K is the kilograms of coal burned per hour.
L is the height in meters.
He also gives the following formulas:

L?
Height of chimney
100 q ?
L
Cross- section in sq . m . = 103 h
When L Weight of waste products per kg. per hour.

q Cross -section of stack in sq . m .
· h = Height of stack in meters.
CHIMNEYS . 235
For the velocity of the gas leaving the chimney, we have Peclet's
formula ,
V = 0
V2gh (T – t)
273 + t
T and t are temperatures.

h is the height of the stack .
g is the acceleration of gravity, 9.81 .
o is a coefficient varying between .38 and .85.
Giesker gives for it a value of .33.
Redtenbacher also states that the height of the chimney should

not be less than 16 meters and equal to 25 times its smallest di
ameter .
Pinzger gives 2 meters per second as the allowable minimum
velocity in the stack. Gruner states that, for temperatures be
tween 2009 and 300 ° C. , the velocity may reach 5 meters per second;
for higher temperatures it may reach 10 meters per second. The
reduction of temperature in an underground flue has been found by
Jüptner and Toldt to be, in one instance, 1.46° C. per running
meter, and they consider that , for chimney calculations, it is safe
to assume a cooling of about 3° C. per running meter of chimney
height.
It is, in general, preferable to give to each furnace its own stack, as
the regulation of the draught is then made easier and more effective.
The draught is better regulated by means of a damper placed on
the top of the stack than by a register in the flue leading to the
base.
Example.-- A furnace burns in three gas-producers 11,077 kg.

of anthracite coal per 24 hours. In allowing for the carbon content
of the ash , the composition of the anthracite is as follows:
Carbon burned .. 52.9 %
Carbon lost in the ash . 5.5
Ash .. 30.2
H20 .. 4.85
S. .67
H2 ., N2 , 02 ... 5.88
Total . 100.00
The gas has the following analysis:
Carbon contents
% by volume % by weight in grs . per 100
liters of dry gas
CO2 10.21 18.09 4.930
CO 20.31 22.92 9.837
H 18.71 1.51
N 2. 50.77 57.47
H20 . 5.50 3.99
Total .. 14.767
The carbon coming wholly from the coal, it is possible to find the
volume of gas produced per minute.
105.50 liters of wet gas at 0° C. and 760 mm . pressure carry 14.767
grams of carbon .
128 grams. of coal per second 128 x .529 = 67.82 grams of
carbon. Number of cubic meters of wet producer -gas obtained from
67.82 X .10550
67.82 grams carbon =
.4845 cu.m. at 0 ° C.
14.767
and 760 mm . At 300 ° C. we have .4845 X 2.1010 =

1.018 cu . m .
Composition of the waste products:
Carbon contents
% by volume % hy weight in grs. per 100
liters of dry gas
CO2 14.51 20.86 5.684

02 4.85 5.07
N2 . 80.64 74.04
H20 6.10 3.59
Volume of the waste products per second, at 0 ° C. and 760 mm.

pressure ,
67.82 X .1061
1.266 cu.m.
5.684
Weight of the waste products per second = 1.691 kg.

L = Weight of the waste products per hour = 3600
=
X
1.691 = 6087 kg.
=
CHIMNEYS . 237
According to Redtenbacher,
L
q 103
h
Making h = 36 m .
6087 6087
q 1.015 sq. m.
103.36 6000
V
The stack is to be built with aa circular section ; therefore the mean

diameter,
D = 1.136 m .
With a batter of 13 mm . per running meter, the top diameter

will be
1.136 – ( 18 x .013) = .902 m .
and the base diameter,
1.136 + ( 18 x .013) 1.370 m .
In Prechtl's table, the stack, for a fuel consumption of 460 kg.

per hour, should have a diameter of 2.0 m . at the base and 1.33 m .
at the top. (See table 23. )
ERARY
THE
UNTERSITY
OF
CA ! IFORNIA
CHAPTER XVI .
REGENERATORS .
In the general theory, no distinction was made between the

two systems of recuperation, the Siemens and the parallel counter
current. Now that we have reached the design of furnaces, it
is necessary to make distinction between them ; each has its own
field, its advantages and drawbacks.
In the Siemens system the chambers are partly filled with a
checkerwork of firebricks. The bricks are heated by passing the
products of combustion through the chambers during one period.
By reversing the valve the flow of waste products is stopped, and
in their place either gas or air enters to be pre -heated. The heat
stored in the bricks during the passage of the hot products of com
bustion is now restored to the incoming cold gases until the end
of this period , when the valves are again reversed and the waste
products admitted anew.
It follows that at each reversal there will be a sudden rise of
temperature, and also that , by having shorter periods between
the reversals, it will be possible to increase rapidly, within limits, the
temperature of the pre-heated gases and thereby their temperature
of combustion ; this will result in an increased temperature in the
hearth .
In the parallel counter- current system the air alone is pre
heated. The recuperator is composed of a series of tubes through
which the secondary air circulates, the tubes being entirely sur
rounded by the waste products traveling in a reverse direction . In
this case the calories stored by the outer surfaces of the tubes are
conducted through the thickness of the walls and are restored by
the inner surfaces by radiation and convection. The gaseous cur
rents flowing always in the same direction, valves are dispensed
with . Once the equilibrium has been reached in the furnace the
air will be pre -heated to practically a uniform temperature.
REGENERATORS . 239
From the manner in which the calories are accumulated and

restored in the two systems, it follows that the Siemens will be
the more efficient. In practice the difference of temperature be
tween the waste products leaving the hearth and the gases entering
is about 300 ° C. in the first system , while in the second system the
difference is over 600° C.
It is impossible to increase the temperature of the hearth rapidly
in the second system , while this is easily done in the first. The
gas cannot be pre-heated in the parallel counter -current system
on account of the soot deposition , which would choke the tubes ;
this, however, is true only of the gas made from bituminous coal,
as with anthracite gas, practically free of hydrocarbons, there is
little soot deposition .
In case of low temperatures, the tubes can be made of cast iron ;
metals must be discarded if the temperature of the pre-heated gas
reaches above 500 ° C. Beyond that point it becomes necessary
to make use of perforated firebricks. The efficiency drops very
much in passing from one material to the other, on account of the
low coefficient of conductivity of firebricks.
With the use of bricks it is very difficult to keep the passages
air-tight. The hollow potteries can be made only in short lengths,
and numerous joints are necessary . When the furnace is started it
is seen , by the examination of the furnaces illustrated in Figs. 33, 35
that the heating of the recuperators begins endways ; there is, then,
an unequal expansion which causes some of the joints to tear loose,
and they remain so to the end of the campaign. A. Lodin * gives
an example of a recuperative illuminating gas furnace. Samples
of the products of combustion were taken at the points of entrance
and exit of the recuperator. The average of twelve analyses is as
follows:
Entrance Exit
CO 17.04 10.34
CO 1. 20 none
0 1.67 2.05
80.09 80.61
1 A, Lodin , " Metallurgie du Zinc, ") Paris . 1905 .

In a parallel current recuperative zinc furnace that came under

the observation of the writer, the average of five analyses taken at
the points of entrance and exits of the products of combustion
gave the following :
Entrance Exit
CO2 16.72 13.96

02 . 2.72 6.33
N2 80.56 79.71
The large admixture of atmospheric air to the products of com

bustion on their way to the stack tends to lower their temperature ,
and if care is not taken to detect the error, the furnace will appear
to have a very good fuel efficiency when in actuality it has not.
There remains the very great advantage of operating without
valves .
The parallel counter -current system from its nature is well suited
to furnaces fired at medium temperatures, as zinc-reduction fur
naces , brass-melting furnaces , illuminating-gas muffle 'furnaces,
billet reheating furnaces, etc.—all furnaces requiring a uniform
temperature or a progressive rise of temperature distributed over a
number of hours. This uniformity of temperature of the hearth may
result in economies which will easily offset the low-fuel efficiency .
In muffle furnaces , in which the vessels are made of fireclay, the
varying temperatures of the Siemens furnace may cause a higher
loss of vessels than in a parallel counter- current furnace, and the
increased expenditure of costly muffles will offset the saving in fuel
due to better heat recuperation .
Maximum Allowable Velocities of Gas Currents in the Various Parts

of Regenerative Furnaces. 1
M. per Sec.
Velocities of the air and gas through the reversing valves .... 3.5
Velocity of the gas entering the regenerator.... 3.0
Mean velocity of the gas through the regenerator ( not over ) 1.0
Mean velocity of the gas through the brick checkerwork ( not
over ). .. 2.0
1 From F. Toldt, loc . cit . p . 262.
REGENERATORS . 241
Velocity of the gas in the free space over the checkerwork ... 0.5
Velocity of the gas issuing through the ports . 8.0
Velocity of the air in the flue leading to the regenerator .. 3.0
Mean velocity of the air through the regenerator (not over) .. 2.0
Velocity of the air in the free space over the checkerwork .. 0.5
Mean velocity of the air through the regenerator (under) ... 1.0
Velocity of the air issuing from the ports .... 8.0
Velocity of the products of combustion through the hearth .. 2.0
Minimum Times of Sojourn of the Gaseous Currents in the Various
Parts of Regenerative Furnaces.
No. of Sec .
Gas in the regenerator . 4
Gas in the checkerwork . 3
Air in the regenerator 5
Air in the checkerwork 3
Products of combustion in the hearth. 2
The figures given were considered proper for the following tem
peratures:
Temperature of the producer -gas in the flue leading from the
reversing valve to the regenerator.... 400 ° C.
Temperature of the air at the reversing valve 30 ° C.
Temperature of the air in the flue leading from the revers
ing valve to the recuperator. .. 270 ° C.
Temperature of the gas leaving the regenerator . 850° C.
Mean temperature of the gas in the regenerator. 625° C.
Temperature of the air leaving the regenerator. 950 ° C.
Mean temperature of the air in the regenerator . 610 ° C.
Temperature of the products of combustion in the hearth . 1600 ° C.
Temperature of the products of combustion entering the re
generator... . 1600° C.
Temperature of the products of combustion after passing
through the reversing apparatus .. 300° C.
The velocities and the times of sojourn will vary according to

the temperature .
1 From Le Chatelier, " ('omptes Rendus, " Vol. ( XIV , 1892 , p . 470 ,
It is well known that the design of furnaces cannot generally be

made to follow the dimensions given by mathematical calculations.
The available room , as well as economy, will often necessitate cur
tailments of the dimensions as well as change in the grouping of
the various organs. In all cases, however, it is well and necessary
to know what should be had in order to make the best use of what
is at hand . Paul Héroult said that success followed him only as
far as his tests had been based on previous theoretical analysis
and calculations.
REGENERATORS.
As a general rule, regenerative chambers have too small volumes .

Nevertheless, it is well to remember that regenerators may be made
too large. Discounting the valuable space that they would occupy,
and their increased first cost , they present also thermic disadvan
tages. The radiation becomes very high, and we have seen that
this item is far from being negligible . It amounts in most cases to
18% or more of the total available calories. As the volume of
the gas chambers is increased , the loss due to the gas filling it at
the time of reversal is also increased . Finally the temperature of
the waste products may be rendered so low as to impair the chim
ney efficiency. We have seen that the temperature of the waste
products should not be lowered much below 200° C. (see page 233) .
The chambers, as far as possible, should be grouped in one block
approaching the cubic form in order to reduce the loss due to
radiation . Chambers of cylindrical shape offer the minimum perim
eter ; if they are isolated, as in the Batho furnace, the radiation
losses are higher than in grouped cubical chambers. For the same
reasons , the walls should not be made less than 440 mm . in thick
ness, and better 550 mm ; the dividing walls need, also, to be of
the same thickness, in order to reduce the leakage through them .
The chambers may be constructed with the longer axis either ver
tical or horizontal. It may be stated , in a general way, that vertical
chambers have a better thermic efficiency .
It is of the greatest importance that the flue extending under
the checkerwork have a sufficiently ample cross -section , so that
REGENERATORS . 243
the gaseous currents will be forced to distribute themselves through

out the whole mass of the checkerwork .
Moving masses do not change their paths of motion unless
compelled to do so. This elementary law of mechanics is very
often disregarded, not only in regenerators and recuperators, but
also in many engineering applications. It is a common occurrence
to see a flue leading to a number of subsidiary pipes, through which
a gas current is to be divided , of such diameter that practically the
pipes furthest removed will receive all the gas.
In a certain recuperative muffle furnace, the distributing air flue
had a cross- section smaller than the total section of the subsidiary
uc
Fig. 75.-INCORRECT FLUE CONNECTIONS FOR A HORIZONTAL REGENERATIVE

CHAMBER .
The hatched area represents the path of the gas currents .

From F. Toldt, “ Regenerativ Gasöfen."
pipes; the result was that only a few of these received any air. The
heat recuperation became practically nil, and barely 25% of the
total available heat was utilized in the hearth .
In the Schoenwalders furnace, this principle is very carefully
emphasized ; the lower flue is made very large and the chamber is
Fig . 76. - INCORRECT FLUE CONNECTIONS FOR A VERTICAL REGENERATIVE

CHAMBER.
The hatched area represents the path of the gas currents.
From F. Toldt, “ Regenerativ Gasöfen ."
subdivided by a wall reaching to the upper level of the checker

work, each chamber having its own flue connecting with the re
versing valve.
The upper flue , collecting the heated gases (air or producer-gas),
s
s
Fig. 77.-- CORRECT FLUE DESIGN FOR A HORIZONTAL REGENERATIVE CHAMBER.

SS, SLAG AND Dust CATCHER .
From F. Toldt's " Regenerativ Gasöfen."
REGENERATORS . 245
should also be of generous cross-section , in order that , on the waste

product phase, they may distribute themselves uniformly through
out the entire volume of the checkerwork.
The design of flues leading to the chambers requires careful
attention and consideration in order to utilize the whole of the
checkerwork . ( Fig. 75, 76, 77. )
In Fig. 32, the chambers are built according to these considera
tions ; it may be further mentioned that the chambers are very
happily grouped with respect to the radiation losses.
In the case of muffle furnaces, the chambers are conveniently
and properly placed below the hearth . On the contrary, the
hearths of steel furnaces or glass tank furnaces have no connection
with the chambers, as the support given to the hearth would be
very precarious, subjected, as it would be, to the sudden changes
due to the heating and cooling of the chambers.
The chambers should be provided with ample dust- and slag
settlers; they will protect, in large measure, the checkerwork from
the accretions, which rapidly reduce the open spaces and soon
necessitate laying the furnace up for repairs.
In zinc metallurgy peculiar conditions obtain which necessitate
the air- and gas-chambers being made of equal volumes. On
account of the leakage of the zinc vapor through the walls of the
muffles, as well as through the cracks of broken muffles, the waste
products carry an appreciable amount of zinc oxide. This settles
on the checkerwork and in time closes the free passages . If, how
ever, the gas is turned into the air -chambers, there is a reaction
between the zinc oxide and the carbon of the soot brought in by
the gas, a metallic zinc being formed . This is carried away by the
waste products, thus cleaning the chambers.
The ratio of the width of the open channels to that of the divid
ing walls should be in relation to the specific heats of the gases and
of the firebricks. The open spaces should be larger in the gas
than in the air-chambers, in which the ratio may be made equal
to unity . In most instances, however , the ratio of the volume of
the checkerwork to the free space is much below unity , and this
point cannot be directly taken into consideration .
There is a great diversity of opinion among authorities in regard
to the proper dimensions for regenerators.. Breslauer gives the

following figures as ratio between the volume of a set of cham
bers in cubic meters and the steel production in metric tons per
hour.
Dowials . 14.60 Terre Noire .... 30.00

Graz ... 15.50 Krupp (3-ton furnace ). 36.00
Haleside ... 16.75 Swedish furnace .. 36.60
Krupp ( large furnace ). 18.32 Borsigwerk ..... 44.00
Landore .. 20.00
Pateg .. 20.60 Steelton (50- ton furnace) ........ 37.45
Siemens recommends a total surface of 51 sq . m . of checkerwork

per 1000 kg. of coal consumed in 24 hours, or 25.15 sq. m. per
chamber .
In regard to the weight to be given to the chamberwork , Grüner
states that for each kilogram of coal consumed between two re
versals 50, and better 60, kilograms of checkerwork should be al
lowed for one set of regenerative chambers. Toldt goes a step
further, and gives the following rule for the determination of the
volume and weight to be given to the checkerwork :
“ A volume of 6 cu . m . and a weight of 2850 kg. should be pro
vided per cubic meter of gas or air ( at 0° C. and 760 mm . pressure)
for each 100° C. rise in the temperature in the regenerator.”
This rule, in terms of Grüner, would give 71 kg. per kg. of coal
consumed between two reversals.
H. H. Campbell states that the combined volume of checker
work per set of chambers should be , as a minimum , 50 cu.ft. per ton
of steel in the furnace , and double that amount to obtain the best
results. In a 40-ton furnace of the Pennsylvania Steel Company,
at Steelton, Pa. , the volume is 65 to 70 cu . ft. In a 50 -ton furnace,
the illustration of which is given in Fig. 39, the volume is 100 cu. ft .;
in a 30-ton furnace at Donawitz , Austria, the volume is 110 cu . ft.,
while it is 50 cu . ft. in a furnace at Sharon , Pa .; also 50 cu . ft. for a
50-ton furnace at Duquesne , Pa. , burning natural gas and having
only two chambers, the air-chambers, affording nearly the capacity
recommended .
1 Breslauer, " Der Stahlguss in Masseformerei. "
2 Computed from data given , H. H. Campbell , loc. cit.
3 H. H. Campbell , loc. cit. page 190.
REGENERATORS . 247
In cubic meters and metric tons, we have the ratio

142
3.12 or 3.12 cubic meters
45.4
of the combined capacity of one set of chambers per metric ton of

furnace capacity. In terms of Breslauer, the 50 -ton Steelton
furnace has a ratio of about 37.45. It is more direct and simpler
than any of the rules giving the volume of the checkerwork as a
function of the weight of coal burned , or of the volume of gas
burned per unit of time.. In a 40 -ton Bain furnace, given by F.W.
Harboard ?, the ratio is 1.24.
Comparison of H. H. Campbells with Toldt's rules: The figures
given are from the thermic analysis of the work of a 50-ton furnace
( Fig. 39) . Assuming that a 50 -ton charge is worked in twelve
hours
Coal consumed per charge, 50 X 200 = 10,000 kg.

Total carbon in the coal, 75.60% .
10,000
Coal consumed per second = .231 kg.
12 X 3600
If deduction is made for the carbon in the ash, the carbon contents
are 73.78% . The total carbon burned per second is then .231 x
.7378 = .1704 kg.
Carbon contents
Analysis of producer - gas in grams per 100
liters of dry gas
CO2 5.7 % 3.06

CO 22.0 10.90
CH 2.6 1.40
CH 0.6 0.64
H 10.5
Oz . 0.4
N 58.2
Total. 16.00
? F. W. Harboard , “ Steel, " page 145.

" J. W. Richards in H. H. Campbell, loc. cit ., page 128.
The gas arried .0375 cubic meter of steam per cubic meter of
dry gas. Cubic meters of dry gas produced per second ,
.1704
1.065 .
.160
Cubic meters of wet gas produced per second ,

1.065 X 1.0375 1.105 .
If we assume the same rise of temperature as Toldt , the volume of

the gas chambers will be
1.105 X 6.0 X 4.5 29.84 cu . m . or
29.84 X 35.3145 1053 cu . ft .
The 50-ton Steelton furnace has about 3315 cu . ft . of checkerwork

for one gas-chamber. Thus the required capacity, according to
Toldt, is about one-third of that of the standard American practice.
In terms of the weight of coal burned per minute, the Grüner rule
should be amended as follows:
The weight of the checkerwork should be of 230 kg. per kilogram
of coal burned between two reversals.
The air- chambers should have a larger volume than the gas
chambers. A ratio of 1 : 1.37 is proposed by H. C. McMillan. 1
He recommends in the same paper 15 ft . as the minimum depth of

chambers, and preferably 20 ft . Ledebur favors a ratio of
>
the volume of air to gas of 4 : 3 or 3 : 2 .

Calculation of the checkerwork necessary for a furnace consuming
17,000 kg. in 24 hours . The ash , weighing 4000 kg. , has the
following composition :
Water . 28 %
Carbon 40
Ash . 32
The elementary analysis of the coal is as follows :

Carbon .. 82.5 !
Hydrogen .. 5.0
Nitrogen 12
100.00
Oxygen . 6.0
Ash 4.1
Moisture , 1.21
Calorific power 8210 cals .
TH . C. McMillan , Proceedings Institute of Cleveland Engineers , Jan. 1903 .
REGENERATORS . 249
Carbon in the coal, 17,000 X .825 = 14,030 kg.

Carbon in the ash , 4,000 X.40 = 1600 kg.
The carbon lost amounts to 11.41 % of the total carbon of the coal.
Recasting the elementary analysis, we have,
Carbon utilized . 73.1 %
Carbon lost in ash 9.4
Hydrogen 5.0
Nitrogen . 1.2 100.0
Oxygen , 6.0
Ash 4.1
Moisture 1.2
The gas has the following composition :
Grams of C per
Gas By Volume By Weight 100 liters of
dry gas
CO 2 5.21 9.12 2.79

CO 20.31 22.65 10.90
H 13.2 100 % 1.05 } 100 %
сн . 3.1 1.98 1.66
N2 . 58.2 65.20
H20 . 2.5 1.79
Total , 15.35 Gr .
17,000
Weight of coal consumed per second , .197 kg
86,400
This coal contains .197 X .731 -.144 kg. of carbon .
Cubic meters of producer -gas per second,
.144
.937 cu . m . of dry gas, or
.1535
.937 X 1.025 = .960 cu . m . of wet gas at 0° C. and 760 mm . pressure.
The products of combustion liave the following volumetric

analysis:
CO2 14.3
2 84.0 100 %
1.7 )
H20 . 8.7
The air required for combustion, per unit volume of gas , taking
into account the excess of oxygen in the waste products ( see table13) :
250 INDUSTRIAL FURNACES AND METHODS OF CONTROL ..
20.3 vol. CO require 10.15 vol. 0 , + 38.85 vol . N ,

2
13.2 H,2
6.60 0.2 + 25.13 N
3.1 CH , 4
6.20 0 , + 23.61
2
N
28.6
Excess X 1.7 0 , 3.40 O , + 12.94 N,
14.3
Total, 26.35 100.53
Total volume of air, 126.88
or 100 volumes of gas require 126.88 volumes of air, one volume

requires 1.2688.
Volume of air in cubic meters per second :
.960 x 1.2688 1.218 cu . m ., at 0° C. and 760 mm . pressure .
Minimum Section of the Gas- and Air - Flues Leading to the

Regenerator.
Temperature of the gas, 400 ° C.

Temperature of the air, 270° C.
Volume of the gas at 400 ° C .:
.960 (1 + at) 2.468 x .960 = 2.370 cu.m.
Maximum allowable speed , 3.0 meters per second.
2.370
Minimum section .79 sq. m.
3.0
Volume of the air at 270° C.:

1.218 (1 + at) 1.218 x 1.9909 = 2.425 cu . m.
2.425
Minimum section .808 sq. m.
3.0
From the temperature given on page 241 , the gas has a temper
ature of 850° C. , as it comes from the regenerator, and 400 ° as it
enters the checkerwork . The rise of temperature is, therefore,
450° C.
According to Toldt's rule, the cubic contents and weight of the
checkerwork for the gas regenerator is found as follows:
REGENERATORS . 251
Cubic contents .960 X 6 X 4.5 25.92 cu . m .

Weight of checkerwork 2.850 X 4.5 X .960 – 12,312 kg.
Volume occupied by the checkerwork,
12,312
6.87 cu . m .
1800
Volume left vacant, 19.08 cu . m .
Volume of the gas at 625° C. , .960 X 3.2938
1 3.162 cu . m .
19.08
Time of sojourn 6.03 seconds.
3,162
Air Regenerator.
Volume of air 1.218 cu. m. per second, at 0 ° C. and 760 mm .
pressure .
The air enters at 270° C. and leaves at 950° C. , a rise of 680° C.
Volume of the checkerwork 1.218 x 6.80 X 6 49.46 cu . m .
Weight of the checkerwork = 1.218 X 2850 X 6.8 = 23,500 kg.
Taking the weight of firebrick at 1800 kg . per cu . m ., the volume
occupied by the brick is
23,500
13.05 cu . m .
1800
leaving 46.41 cu . m . for the air.
Volume of the air at 680° C .:
1.218 ( 1 + at) = 1.218 X 3.2387 = 3.945 cu . m .
The air would remain
36.41
9.23 seconds.
3.945
Cross Section of the Ports.
Allowable velocity of the air and gas, 8 m . per second.
Temperature of the air, 950º.
Temperature of the gas, 850°.
Volume of the air at 950° , 1.218 X 4.4865 5.465 cu . m .
Volume of the gas at 850° , .960 X 4.1195 = 3.954 cu . m .
5.465
Cross-section of the air -port, 8
.683 sq . m .
3.954
Cross-section of the gas-port, .494 sq . m .
8
CHAPTER XVII.
RECUPERATORS.
Calculation of the Dimensions to be Given to the Recuperative Sur

faces for a Muffle Furnace . - The data given in the design of аa chim
ney will again be made use of in connection with the recuperator.
Volume of gas generated per second at 0° C. and 760 mm . pres
sure .4845 cu. m.
Volume of Air Injected Per Unit Volume of Gas.

20.31 vol . of CO require 10.155 vol.of 0 2, + 38.67 vol . of N ,
( See table 14. )
18.71 vol . of H , require 9.355 vol . of 0 , + 35.62 vol . of N ,
Excess of 0 2, .... 10.53 vol. of 0 2, + 40.10 vol . of N ,
( See page 236. ) Totals 30.040 vol. O , 114.39 vol. N ,
Total volume of air 144.43 vol .
or 100 volumes of gas require 144.43 volumes of air. One volume

of gas requires 1.4443 volumes of air.
Volume of air injected per second :
.4845 X 1.4443 = .700 cu . m .
At 0° C. and 760 mm . pressure .
Weighing .700 X 1.293909 = .906 kg.

Volume of the wet waste products, ' 1.343 cu . m .
The waste products enter the recuperator at a temperature of
1340 ° C. We will assume that the temperature has fallen to 400 ° C.
at the recuperator exit ; their average temperature is then 870° C.
The total heat capacities of the products of combustion at 1340 ° C.
and 400° C. are found as follows:
1 The volume of the products of combustion has been assumed to be 1.343 c.m. instead
of 1.266 used in the stack calculations .
RECUPERATORS . 253
They contain per weight

.361 kg. of CO2
.088 kg. of 0,
1.282 kg. of N,
.066 kg. of H 20
From Fig. 88 the heat capacities at 1340 ° C. are :
Gas
Heat Capacities Calories
Weights in Calories
co 361 411.0 148.5

2. .088 318.0 28.0
N 1.282 366.0 469.0
H20 . .066 897.0 59.1
Total.. 704.6
At 400 ° C. we have :
Gas Weight Heat Capacities Calories
CO2 . 361 X 91 , 32.9

038 X 88 7.7
N. 1 282 X 100 128.2
H20 066 X 205 13.5
Total.. - 182.3
Then the available number of calories per second is

704.6 182.3 522.3
The air enters the flue leading to the recuperator at 30° C. and
leaves the recuperator at 930 ° C. , a rise of 900° .
The volume of air to be pre-heated is .700 cu . m ., weighing .960
kg. It contains .698 kg. of nitrogen and .208 kg. of oxygen .
Heat capacities of the injected air at 930° C .:
02. 208 207 43.4

N 2 699 234 . 163.3
Total.. 206.7
Heat capacities of the air at 30 ° C.:
02 208 6.55 1.36

N .698 7.50 5.25
Total.. 6.61
Calories to be taken from the waste products between 30° C.

and 930 ' C. ,
206.7 - 6.6 200.1 calories.
200.1
Thus 38.3% of the total available calories contained in
522.3
the products of combustion are utilized , leaving a large margin
for losses through radiation , etc.
105
*oe
92
oe
30 76 76
** 21 18 / 30
Fig . 78. - SECTION OF PERFORATED BLOCK .
The perforated blocks are shown in Fig. 80. In order to sim

plify the calculations, all the formulas being in terms of cylin
drical tubes , we will make use of a perforated block of circular
section, equivalent in area to one of the perforations of the fire
brick block ,
Area of the cross -section of one perforation ,
.076 211
x .105) = .01251 sq.m.
+ (0.076 X
4
RECUPERATORS . 255
The diameter of the equivalent circle ,

d = .126 m .
The diameter of the outer cylinder,

D = .126 + (2 X .030) .186 m .
Area of the inner surface of block , one meter long,

ADI = .397 sq. m.
Volume of the enclosed air- column,
IID 21
.01251 .
4
Weight of the air - column,

.01251 X 1.2939 .0162 kg.
The total number of calories to be absorbed by the air per hour
200.1 X 3600 720,360 calories.
30 120 30
180
Fig. 79. - SECTION THROUGH CYLINDRICAL TUBE, EQUIVALENT TO ONE
HALF OF THE PERFORATED BLOCK .
Let us find the number of calories, W, that can be transmitted

per hour through the equivalent cylindrical tube, one meter long,
W CF - ( 1)
In which C -
Coefficient of heat transmission .
F Inside surface of cylinder.
T
t , - tz. 1
t , = Mean outside temperature 870 ° C.

ta = Mean inside temperature 480° C.
t2
T 870° - 480 ° 390° .
The formula giving the value of the coefficient C is as follows:
1 1 d 1 d D
+ + log . nep . (2)
С a 1
D а 2 21 d
In which a , and a z are the coefficients of heat transmission for
the outer and the inner surfaces .
D = outer diameter.
d = inner diameter.
^ = coefficient of heat conductivity of bricks =: .7..
λ
å , for both surfaces = 8 .

a ,2 is given by the formula,
1.00771 1.0077' 2
(
)
3
az 125 S + .55 b (t , - t2) - 2 33
t, 2
Where Si radiation coefficient of firebrick 3.60 .

bconvection coefficient of air. ·
Valerius gives for confined air, b = 4.
For air in a free space , b 5 to 6 .
The air being in motion, the higher coefficient will be taken, or,
substituting in ( 3) ,
b = 6
1.0077870 - 1.0077480
a
2
125X 3.6 + .55 X 6 X 390.233
390
2 454.5 + 13.25 467.75 .

A very interesting paper has recently appeared on the quantitative radiation of fuos
and furnace surfaces. by J. Bied , in Revue de Metallurgie, Sept. , 1905 .
RECUPERATORS . 257
Substituting in (2) ,
1 1 .126 1 .126 .186
Х + + log . nep .
С 8 .186 467.75 1.4 .126
1
.1218 .
С
с : 8.21 .
Substituting in ( 1 ) ,
W = 8.21 X .397 X 390 = 1271 calories per hour.
Length of tubes required to transmit the necessary number of
calories,
720,360
567 meters .
1271
As each block has two perforations, the length of blocks required

is
567
= 283.5 m .
2
The blocks are made conveniently in 292 mm . lengths. The

bricks used as spreaders between the perforated blocks are
66 x 182 mm . , taking 33 mm. from the length of each block ; thus
reducing their effective length to 259 mm.
Total length of blocks,
292 X 283.5
320 m.
259
For the general design of ports, see Fig. 31 and 32, while for the 7
recuperator, see Fig. 33 and 35.

The blocks are placed side by side ( Fig. 80) , and only about
one-half of their inner surfaces are placed at the distance of 30 mm.
from the outer surfaces, as was assumed in the calculations ; the
total length found should be correspondingly increased.
320 X 3
Revised total length , = 480 meters.
2
For practical considerations the over- all width of the recuperator

cannot exceed 5.18 m . To insure the proper air distribution
throughout the entire section of the recuperator it will be divided

into two chambers. The secondary air will be offered a free ver
tical path from the lower distributing flue to the upper collecting
fue . The outer walls are made .55 m . thick, the division walls .44 m .
雅
Fig . 80.- PERFORATED FIREBRICK BLOCKS FOR THE PARALLEL COUNTER
CURRENT SYSTEM OF HEAT RECUPERATION.
The width remaining for each chamber is

5.18 ( 1.10 + .44)
= 1.82 m .
2
480
Each chamber has 240 m . of perforated blocks.
2
The length of the chamber is 6.200 m . Each chamber will have
6 rows, 6.20 m. long, with 14 blocks in each , or for each meter of
height 6 X 14 = 84 running meters of perforated blocks.
Each row will then have a height of
240
= 2.86 m.
=
84
RECUPERATORS . 259
In order to use only full lengths of blocks we will take for the
height the nearest common multiple of their length, 292 mm . or
2.920 meters.
Each chamber offers for the passage of the air 84 columns of per
forated blocks, each with two perforations of .01251 sq . m. area ,
a total of
.01251 X 84 1.05 sq . m.
The flue distributing the secondary air to the perforation should
have a cross -section 50% larger than this, or
1.05 X 3
1.59 sq . m .
2
The width of the chamber is 1.82 m . Height of flue

1.59
.875 m .
1.82
Volume of the air at 480 ° C. for each chamber,
.350 X 2.7616 .967 cu. m.
Mean velocity of the air through the perforations,

.967
.92 m .
1.05
-a good velocity for small rough tubes.

Volume of the air at 930° C. ,
.350 X 4.4131 1.544 cu . m .
For the proper distribution of the air in the hearth 11 ports

per side are needed . For practical use they are made .150
X .150 mm.
The velocity of the heated air issuing through them is
1.544
6.24 m . per second .
11 X .150 X .150
The same number of gas-ports are placed alternating with the air
ports, 11 per side. The gas issuing through the ports has a tem
perature of 400 ° C.
Volume of gas per side, at 400 ° C. ,

.4845
X 2.468 .598 cu . m.
2
For the sake of simplicity in construction , the gas- ports are

made of the same size as the air- ports.
Velocity of the gas issuing through the ports,
.598
2.41 m . per second .
11 x .1502
Owing to the pressure with which the gas comes from the pro
ducers, its velocity is much higher than that indicated by its vol
ume and its temperature of 400 ° C. and the atmospheric pressure .
In practice it is found that the gas velocity is higher than that of
the air.
The air is to be sent through the recuperator by means of a fan ;
this allows for a simple and efficient regulation of the volume re
quired .
Following is a calculation of the velocity of the products of
combustion through the recuperators.
Volume of the products of combustion per side, their mean tem
perature through the recuperator being 870 ° C. ,
1.343
x 4.1929 2.82 cu , m .
2
In order to secure a thorough contact of the blocks with the
products of combustion throughout the recuperator , they are
made to circulate in three consecutive horizontal paths (see Fig. 33
and 35) , on their way to the stack -flue.
Thus only one-third of the available space of the recuperator is
usable at one time.
In each chamber we have passages 6.20 m . long by .182 m. wide
and 259 mm . high. There are 50 such passages; the total vol
ume , then , is
6.20 x .182 X .259 X 50 .
And as only one -third is available, the volume is ,
6.20 x .182 X .259 x 50
4.87 cu . m .
3
RECUPERATORS . 261
Mean velocity of the waste products through the recuperator,

4.87
1.73 m . per second.
2.82
The methods and formulas used in the calculation of the proper
radiating areas to be taken for the absorption by air of a given
number of calories can be made use of for all engineering problems
involving the same factors.
If, instead of a firebrick recuperator, we had chosen to select, for
lower temperatures, cast -iron pipes, the radiation and conduction
coefficient of cast iron would have been introduced into the formula .
It is often desired to find the length of pipe required to cool to
a known temperature a gaseous mass, traveling in cast-iron , wrought
iron or lead pipes. The same formulas will apply by making use of
the proper coefficients.
CHAPTER XVIII .
REVERSING VALVES.
There are a large number of types of valves in use , from the

original Siemens valve to the modern water -seals . A good valve
should be such that it will preserve its true form , no matter to what
range of temperature it is exposed. Cast - iron valves, when first
made, may have perfect bearing surfaces; these are not preserved
long, however, as the cast iron expands unequally, according to the
shape of the casting, and the resulting distortion prevents the valve
ht
Fig . 81.-WAILES WATER-SEALED VALVE .
from fitting tight on its seat . Once the valve has become warped ,
it should be removed at once, as its condition will become worse
on account of the high temperature caused by the gas burning at
the leak . Also small amounts of dust, soot , etc., settling on the
seat, will prevent tight closing. The seats should be selected with
the view of avoiding the settling of solid particles, and should be
self -cleaning.
The only valves that remain tight at all temperatures are the
water-seal valves, providing that the level of the seal is kept hor
REVERSING VALVES . 263
1000 2400
FIGO
PINO
K3001
50H2101200
OOH
會
800
OSTEOC0095
+0019
009
Ai Regenerator
RE100
NA
Passie 160
700 B002 306 )
El F
Ventilator
Admission
Ventilator
Gus
Gus
All
000
Feeli Valve
Air Regenerator
FIG . 82. - REVERSING VALVE .

From F. Toldt, " Regenerativ Casöfen .”
izontal. They have one grave defect, however, that of vaporizing

a large quantity of water, the vaporization being a function of the
temperature of the gas passing through the valves. This increase
in the water vapor has an unfavorable effect on the thermic effi
ciency of the furnace, and in the case of steel furnaces may cause
increased oxidation of the metallic bath .
TLET
Fig . 83. – FORTER WATER - SEALED REVERSING VALVE , HAND -OPERATED, WITH
Casing RAISED , SHOWING HOOD AND WATER-SEAL .
For most purposes, water -seal valves are to be preferred. Im

portant leaks of gas are sometimes caused by dry valves, a portion
of the producer-gas short-circuiting to the stack.
As an instance, we have the following sample of stack gas taken
from a zinc furnace in which it was intended to use a thoroughly
oxidizing flame:
lToldtoc
.c,From
L.it
OVEN
.84.
—COKE
FIG
% %
by Volume by Weight
CO 13.75 20.21
CO 2.50 2.34
02 . 3.75 4.01
H2 . 1.93 0.13
N 78.07 73.31
H20. 4.50
Entailing a dead loss of 16.67% of the fuel used.

It is well to have the gas inlet and the reversing valves distinct,
the inlet valves being removed from the high temperature of the
reversing valves.
HEARTH .
In the original Siemens furnace the roof is depressed, as it was

thought necessary to bring the flames to impinge on the material
to be heated on the hearth . This form has very serious drawbacks,
among which are the rapid deterioration of the roof and the incom
plete combustion of the gases.
In a furnace of so small a volume the velocities of the gaseous
masses are so great that, notwithstanding the high rate of iffusion
between the highly heated gases, they reach the ports of exit before
complete combustion takes place. Furthermore, this disposition
compels the gases to distribute themselves in thin blade -like streams,
which burn only at their surface of contact.
Friedrich Siemens strongly advocates high , dome-shaped roofs,
claiming that heating should be obtained by radiation. This form
is doubtless the best for bituminous producer-gas , and in general
producer-gases rich in illuminants. For producer-gas made from
coke , anthracite or charcoal, a high roof would not be economical
in fuel, as the flame is not very luminous.
It is well known that the contact of relatively cold materials
cools the gases below their ignition points, thus promoting incom
plete combustion.
In industrial operations, however, it is the ultimate economy
which is the only important one. While the heating by radiation
may be conducive to fuel economy, the time of melting, as claimed
1
by H. H. Campbell , ' is reduced by making the flame impinge on
the stock .
1 H. H. Campbell, loc. cit. , page 213.
THUND inntinn mun
LLLLLLLLLLLLLLLLLL
THIS M
WHIRU
SK SKI
1.UZ
2211
Fig. 844. — COKE OVEN .

From F. Toldt, loc. cit.
SK SKI
i 11
FIG. 846.- COKE OVEN.
From F. Toldt, loc. cit.
nim
12
Il
N
16
Fig. 85. – VARIOUS TYPES OF GAS AND AIR PORTS.

From F. Toldt, loc. cit .
Under American conditions it is of more importance to reduce

the time of melt, thereby increasing the furnace capacity per day,
than to save some fuel, while the reverse is probably true on the Con
tinent, where fuel is very expensive and labor comparatively cheap.
The tendency of the present practice in the United States is to
have a slight rise in the roof. However, many furnaces have a
practically straight roof along a median line, passing through the
hearth's longer axis ( see Fig. 39) . Roofs built along these lines
are long-lived, as they are not cut by the flames.
The volume of the hearth should be such as to allow the velocities
of the gases to decrease very materially as they issue from the ports,
allowing the products of combustion to remain as long as practi
cable in the hearth. The increase in the time of sojourn permits
complete combustion , as it affords a better opportunity for the
diffusion to take place.
PORTS.
Many designs and dispositions of ports have been advanced,
most of them quite complicated. With the gas and the air highly
pre-heated, it is not necessary to go into great refinements of con
struction in order to secure complete combustion .
It is to be remembered that the ports have to withstand the most
severe conditions, and that, above all , their construction should be
such as to allow of long life and quick repairs.
The suggestions first brought forward by Siemens are still good.
He recommended that the gas be prevented from coming in contact
with either the roof or the furnace sides, the air-ports , therefore,
to be placed above the gas-ports and overlapping these at both ends
( No. 5, 10, and 13 , Fig. 85) .
In Fig. 85 are given several dispositions of ports. For small
furnaces , the ports should either be inclined toward each other, or
the air-ports given a sharp angle toward the gas-ports ( No. 4 and 8 ,
Fig. 10) . The point of intersection of the gas and air currents
should neither be too near the ports, as it would endanger their
lives , nor too far, as the combustion would hardly have begun before
the flame would meet the hearth material.
For large furnaces (50 to 70 tons) the air -port is given only a
-VN 270
W —FIG6.
ELLMAN
FURNACE
INDEPENDENTLY
,STILTING
PORT HOWING
THE
BRACED
.8BLOCKS
slight angle toward the gas-port. The present practice in America

for these large furnaces is to have two ports only, one for the air
and the other for the gas, the air-port overlapping the gas-port at
each end .
In the Schoenwalders furnace, the gas and air meet in a com
bustion chamber preceding the hearth ; this chamber and the ports
themselves must be exposed to extreme temperatures; the loss by
radiation is very high , and serves no good purpose.
In order to increase the length of the campaign of open-hearth
furnaces, inventors have sought to enclose the ports, with their
sub -dividing walls, in a removable cage with its own system of tie
rods and braces . This construction allows the removal of a block
needing repair and its replacement in about an hour's time. The
design of the block shown in Fig. 39 is that of C. E. Stafford .
Fig. 86 shows the construction of the self-contained , removable
port-blocks of a tilting Wellman furnace .
The introduction of air-cooled cast -iron blocks under the ports
also tends to lengthen the life of the port-blocks. In the case of
tank glass furnaces ( Fig. 38) , or with muffle furnaces, where the
flames have to heat a very large hearth at a uniform temperature,
the ports are alternated, being placed side by side, or better in
two horizontal rows, and are all built with horizontal sides, in order
to have a gradual diffusion . The construction shown in Fig. 32
would tend to create an intense combustion at the zone of inter
section of the two currents, which would endanger the life of a
vessel in close proximity.
The conditions necessary to insure complete combustion, in the
opinion of Ledebur , '1 are as follows:
( 1 ) High temperature in the combustion chamber.
(2) Intimate mixture of the gases and air.
(3) Diffusion of the chemically reacting gases.
(4) Excess of oxygen .
All these conditions are practically met in modern metallurgical
furnaces.
1 Ledebur, “ Die Gasfeuerungen . "
CHAPTER XIX .
POWDERED FUEL.
Whenever practical, the use of powdered fuel offers considerable

economy over direct or gas firing. The heat obtained is also more in
tense than that which can be obtained from either solid coal or pro
ducer- gas if care is taken to pre-heat the air required for combustion.
Dust- coal firing has the further advantage over gas furnaces of
doing away with the gas -producer and its inherent losses. These
are considerable, for it is seldom that the average thermic loss, in
the gasification of the fuel by premature combustion to carbon
dioxide , heat radiation and convection , is under 20% of the avail
able calories; to this should be added the loss due to the carbon
contained in the ash and the coal consumed in the boiler grate for
the production of the steam required for the gas-producer. A
total loss of 30 % is a conservative figure for good practice.
With coal-dust firing there is only one combustion , that to CO 2 ;
further, the combustion is complete, the ash being free from
residual carbon. The air required for combustion is also sus
ceptible of close regulation , it being added in proper volumes and
at the same time, mixed with the fuel.
The preparation of the powdered fuel, however, is a serious mat
ter. For proper combustion the coal must be crushed and com
minuted so that 95% of it will pass through a 100 -mesh screen .
Previous to crushing the coal must be freed from all moisture.
After going through a rotary drying kiln, the coal is sent to a
tube-mill, half filled with French and Danish flint pebbles, 2 to 7 cm .
in diameter. The grinding and screening take place jointly. From
the tube-mill the fine dust is conveyed to the furnace.
When reduced to such fineness and kept dry, coal-dust will flow
very much as a liquid. There is very little danger of the dust clog
ging up pipes or hanging in hoppers, etc.
LT
90
80
Calories
10 HAA
e Good
an xid
e
th
Djo n
60
Me
rbo H₂9
so Kad
r
vapor
Wate
30
e.co
1927 NalH
Gases
20 Perfect
10
TemperatureC.s
200 600 800 1000 1200 2000
1800
2200
2400
1600 2600 2800 3000
Fig.
THERMAL87.
DIAGRAMOF-
GIVING
THE
CAPACITIES
GASES
PER
MOLECULAR
VOLUME
FOR
ANY
TEMPERATURE
BETWEEN -
3200
AND
DissOCIATION0(°).
C.
BEING
NEGLECTED
!
R000
wood
!
oood
!
8000
Caloryes
!
7000
H2
!
6000
gen
d ro
sood Hy 4
CH
e
han
Met 10 or
Vap k
Wato
nood
!
toe
idel
rbon
Diaox
C xi
Mo de
carlrp
e onen
!
ooo Wi tz og
oxygen og
Temperatures
1980 100 1896 2007 200 200 6600 2800 SON
200 400 600 800 1000 Ipod
o
BETWEEN
TIMPERATU'RE
OFFOR
KILOGRAM
PER
GASES
CAPACITIES
THERMAL
THE
GIVING
-D
80ANYIAGRAM
. 8.
FIG
)NEGLE
D
(C.
3200
AND
BEING IATION
ISSOCCTED
POWDERED FUEL . 273
The danger of a subsequent explosion , due to the ignition of

coal-dust, has been largely overestimated . Since a terrible coal
dust explosion took place, in the grinding-mill of aa cement company,
care has been taken to avoid the accumulation of the powdered
coal on the floors, and also to provide sufficient ventilation . The
main point to bear in mind is to avoid a confined place in which the
coal-dust may become well mixed with air .
Powdered coal, while lying in a heap, will burn as slowly as so
much punk; a sudden complete combustion , giving, by the prac
tically instantaneous elevation of temperature, rise to destructive
gas explosions, can only take place if the coal - dust is thoroughly
mixed with air. Such conditions will prevail when the powdered
fuel is carried along by an air current. The coal-dust explosions
of dry coal mines occur under precisely such conditions. If an ex
plosion , small or large, of fire-damp takes place, the thick layer of
coal-dust covering the walls, timbers and floor is at once trans
ported by the powerful air current, and a secondary explosion may
>
take place in the fuel cloud, which may prove more disastrous than
the primary one. The same thing may occur with any combus
tible material , as flour, etc. Practically all the cement produced
in the United States , outside of the natural gas or oil fields, is
burned with coal-dust, and explosions are rarely heard of.
9
Dust-coal firing is admirably adapted to the obtention of high

temperatures in long hearths. The pre-heating of the air required
for combustion gives rise to a particular difficulty with regard to
the ash-dust carried by the waste products.
The regenerator should be designed with generous ash -settling
chambers and large passages in the checkerwork to prevent their
undue contraction by the accumulation of ash -dust; at the same
time the total area should not be so large as to cause the velocity
of the gaseous currents to become too small , this tending to increase
the ash -dust deposition.
The parallel counter- current regenerators are well adapted for
this purpose , for, as we have seen , besides avoiding the use of
valves, it is possible to clean out the passages for the waste products
by a jet of steam or compressed air, or even with small and sharp
water -jets. Just enough water should be injected to have instan
taneous vaporization ; some qualities of firebrick will stand sudden

cooling without spalling. For regenerators a design similar to that
of blast-furnace hot stoves would be advantageous, these being
formed of a series of thin solid walls; the cleaning would thus be
much simplified. Such a chamber would resemble that of the
Ferraris Zinc Gas Furnace, U. S. patent 714,685.
Coal- dust firing, like gas firing, requires for great efficiency that
the combustion take place in a chamber in contact with highly
heated material, firebrick being the best. A comparatively cool
urface will cause a chilling of the fuel and air immediately in con
tact with it below the ignition point, resulting in incomplete com
bustion . It follows that, if dust firing is to be used in connection
with boilers, a firebrick combustion chamber should be provided in
order that only the oxidized waste products come in contact with
the boiler flues.
The powdered coal may be supplied to the furnace in different
ways, according to its dimensions. Practically the same remarks
made in connection with gas firing apply to firing with coal-dust.
For a small hearth , the fuel should be fed in a diffuse state, that is,
already mixed with the air required for combustion . For long
furnaces, as rotary cement kilns, roasters, driers, etc. , the fuel
should be sent axially under high velocity, the air being admitted
by ports, parallel to the coal-dust port. A powerful elongated
flame is created , the coal -dust burning practically the whole length
of the furnace, which may be 150 ft . long . "
Coals rich in volatile matter, as gas coals and lignites, are to be
preferred . The following analysis is a fair average of the West
Virginia gas coals :
Fixed carbon . 54.00 %
Volatile matter .. 32.75
Ash ... 12.00
Moisture . 1.25
With such coals, the combustion takes place practically in two

stages. The coal as it emerges from the stoker is coked , the vola
tile matter being liberated and burned at once, while the solid coke
is first brought to bright redness before it ignites. During these
1 Rotary clinkering furnace of the Edison Portland Cement Company.
..
POWDERED FUEL . 275
successive operations, the coal has traveled a large portion of the

length of the furnace.
COAL - Dust STOKERS.
Few fields of industry have been more studied than the applica
tion of powdered fuel to furnace firing. On pages 283–286 is given
a list of the United States patents covering the field . All are based
on the injection , under pressure as high as 50 lb. and as low as
8 oz . , of the air carrying with it the powdered fuel.
>
The greatest field of coal-dust firing is in the cement industry .

Practically all of the cement produced in the United States is made
in rotary kilns, and this type is fast superseding all other forms
abroad . The rotary cement kiln has not a high thermic efficiency ;
its simplicity of operation and large capacity, however, make it on
the whole very economical . Its thermic efficiency is low, for the
following reasons :
( 1 ) High velocity of the gases rushing through it .
(2) Large excess of air, there being a number of openings some
of which cannot be closed, such as the necessary clearance
space between the terminal hoods and the kiln proper .
(3) Practically no air recuperation.
(4) High radiation of the metallic wall .
(5) High temperature of the issuing clinker.
The radiation is unavoidable, and, furthermore, necessary , the
life of the shell depending on it. The efficiency can be increased
only by reducing the losses due to the excess of air and by recu
peration . A number of recuperators have been proposed ; most
of them , however, are faulty in principle and cannot give good air
recuperation .
Rotary kilns are built with a second rotary kiln lying in the
same or in the reverse direction . The cement mixture enters the
upper kiln, the clinker being discharged into the second . An air
current is made to pass in a reverse direction , first through the
lower kiln , then through the upper, to the combustion of the fuel .
In long kilns, 5 to 8 ft . in diameter, the heated air will rush along
the upper surface, while the hot clinker occupies the lower part .
The heating of the air can be done only through direct contact ,
not by radiation . The air is in contact only with the shell of the
kiln, which is in turn heated by contact with the clinker. The
calories carried away by the white clinker form a large percentage
of the fuel used .
The following example is taken from practice:
A rotary kiln producing 200 bbl. of clinker (weight of a barrel is
173 kg.) per 24 hours consumes 54.5 kg. of fuel per bbl. The clinker
leaves the kiln at a temperature of 1300 ° C. Taking the specific
heat of cement clinker to be, at that high temperature, .30 , the
calories contained in the clinker produced daily are
173 X 200 X.30 X 1300 = 13,494,000 calories.
The analysis of the coal is given , page 274 .
The calorific power of the coal can be found by means of Goutal's
formula ( see page 222) .
V
' = Volatile matter, neglecting ash and moisture.
100 X 32.75
86.75
= 37.75 %
From the table of factors :

a = 85.75 calories.
Substituting in Goutal's formula,

Calorific power = 82 X 54 X 85.75 X 37.75 7665 cal.
The calories contained in the clinker are equivalent to

13,494,000
1760 kg . of coal.
7665
The kiln consumes per day, 54.5 x 200 10,900 kg.
The calories carried away by the hot clinker are equivalent to

1760 X 100
16.14 %
10,900
A rather important percentage of the total available calories.
POWDERED FUEL . 277
It is necessary to cool the clinker for handling. This is usually

done by first quenching it with water, and then finishing the cooling
by air currents. For the latter treatment, the clinker is stacked
in an iron tower, with baffle plates, the air being forced to go through
the clinker piles on its way to the atmosphere, whence it passes as
soon as it has been heated .
Efficiency Test of a Rotary Clinkering Furnace for the Manufacture

of Portland Cement.
A rotary kiln 20 m . in length burns .131 kg. of powdered bitu

minous coal per second, and sinters .656 kg. of raw rock mixture
in the same time. The whole of the ash in the coal is collected in
the dust chamber, none contaminating the clinker.
Analysis of the Products of Combustion .
CO , 26.53 %
() 2..
CO . 1.70
Na 71.77
Total . 100.00
Water vapor . 8.50 %

The elementary analysis of the coal is as follows:
Total carbon ... 76.5 %
Hydrogen .. 4.91
Ash . 8.50
The analysis is given on dry basis ; the coal carried 0.94% of

moisture .
Calorific power (dry basis ) .......... 7804 calories.
Both the raw rock and the powdered coal were fed into the
bins at a temperature of 52° C.
Analysis of Raw Rock.
Carbon dioxide..... 31.81 %

Water ... 2.62
Graphite.... 1.05
Loss by ignition .. 35.48
Temperature of the products of combustion ... 800 °

Temperature of discharged clinker.. 1320°
We will first find the total amount of calories brought in the

furnace :
Kg. of dry coal burned per second, .9906 X .131 = .1297.

Calories brought by coal, .1297 Xx 7804 = 1012.7.
Kg. of graphite in raw rock , .656 x .0105 = .00689.
Calories liberated by graphite, .00689 X 8147 = 56.11 .
Sensible heat brought by coal and rock, taking their specific heat
as .21 for both ,
( .656 + .131 ) X.21 X 52 = 8.6 calories.
Total number of calories brought by the coal and the raw rock,
1012.7 + 56.1 + 8.6 1077.4 .
We will take the specific of the clinker to be .30 at the tem

perature of 1320° .
Clinker produced per second , .656 X .6452 = .424 kg.
Calories held by clinker at 1320 °,
.424 x .30 X 1320 168 .
Total weight of carbon brought by the rock and coal per second :
Carbon in coal, .1297 x .765 = .0993 kg.
Carbon as graphite, .00689
Carbon dioxide in rock , .656 X .3181 = .209.
12 x .209
Carbon in carbon dioxide, .057 kg.
44
Total carbon , .0993 + .0069 +.057 = .1632 kg.
POWDERED FUEL . 279
One cubic meter of waste products at 0° and 760 mm . pressure

contained 265.3 liters of carbon dioxide, weighing
.2653 X 1.96503 .52132 kg .

12 x .52132
holding .1422 kg. of carbon.
44
The 17 liters of carbon monoxide weighed

.017 X 1.2505 .02126 kg .
12 X 21.26
and contained = .0091 kg. of carbon .
28
Total weight of carbon per c.m. of waste products,

.1422 + .0091 .1513 kg.
Since we have accounted for all of the carbon entering the fur
nace , it becomes possible to calculate the volume of the waste
products issuing from the stack , per second :
.1632
Volume of dry gases 1.08 c.m.
.1513
Volume of wet gases , 1.08 x 1.085 1.17 c.m.
Weight of the water vapor, per second ,

.085 X 1.08 x .80105 - .0737 ky .
Total weight of carbon dioxide, per second,

.52132 x 1.08 .56232.
Calories contained in .56232 kg. of carbon dioxide at 800° :

From Table 5 we find the thermal capacities per kilogram ,
.562 X 208 = 117. calories.

Weight of nitrogen in 1.08 c.m. ,

1.2544 x .7177 X 1.08 = .9711 kg.
Weight of the carbon monoxide :
1.2505 x .017 X 1.08 .023 kg.
Thermal capacities of carbon monoxide and nitrogen at 800°,
207 calories.
Calories held by the nitrogen and carbon monoxide ,
( .9711 + .023) X 207 = 205.76 calories.
Latent calories held in .023 kg. of carbon monoxide ,
.023 X 2403 55.10 calories .
Calories held by the water vapor, 461 calories per kg. at 800 °,
.0737 X 461 33.98 calories .
Total number of calories, latent and sensible, lost ,
117. + 205.76 + 55.10 + 33.98 = 411.84 calories.
According to our definitions of lost and utilized heats we have :
Heat lost through the stack ( sensible and latent),
411.84 X 100
38.22 %
1077.4
Heat lost through incomplete combustion (latent heat) ,

55.10 X 100
5.11 %
1077.4
Heat carried away by the clinker,

168 X 100
15.60 % .
1077.4
Heat utilized in the furnace, per difference, 46.18% .

What volume of air was injected in the kiln ?
POWDERED FUEL . 281
The total weight of carbon burned per second is .1022 kg. , of

which .0098 kg. was burned as carbon monoxide and the remainder,
.0924, as carbon dioxide.
One kg. of carbon burning to carbon dioxide requires 9.02 c.m.
of air ( Table 2) .
Volume of air required by the carbon burning to carbon dioxide ,
9.02 x .0924 .8334 c.m.
One kg. of carbon burning to carbon monoxide requires 4.48 c.m.

of air ; the volume required for .0098 is
4.48 x .0098 .044 c.m.
Weight of hydrogen burned per second,

.0491 X .1295 = .00637 kg.
One kilo of hydrogen requires 26.88 c.m. of air.
Volume of air required for the hydrogen,
.00637 X 26.88 .1712 c.m.
Total volume of air,
.8334 + .044 +. .1712 1.0186 c.m.
The inner cross -section of the kiln being 1.824 sq.m., what is
the mean velocity of the gases through the barrel ?
The temperature of the gases near the discharging point of the
clinker is 1340 ° :
1340 + 800
Mean temperature 1070°.
2
Volume of the waste products at 1070 ° ( Table 19) :

V = V ( 1 + at ) = 1.17 X 4.9269 = 5.766 c.. m .
'
Mean velocity, in meters :

5.766
3.16 m .
1.824
By increasing the length of the kiln from 20 to 40 meters the

temperature of the waste products has been lowered to 300° . What
percentage of heat will then be lost through the stack, and what
will be the resulting economy?
At 300 ° the thermal capacities of the carbon dioxide, carbon mon
oxide, nitrogen and water vapor are respectively:
68.1 , 76 and 153 calories.
By using the weights found previously we have a total loss in
calories, through the stack , of 125.14.
125.14 X 100
% of heat lost 11.61 %
1077.40
a difference of
38.22 – 11.61 = 26.61 % .
Through proper recuperation of the calories carried away by

the clinkers (15.60 % of the total available calories), it is possible
to bring back in the furnace, as heated air, about 10% of that
lost heat. Thus, by proper kiln dimensions and recuperation, a total
saving of 36% can be effected .
CHAPTER XX .
LIST OF U. S. PATENTS COVERING THE STOKING OF POWDERED FUEL.
53,208 — March 13, 1866. Improved Method of Burning Waste

Coal. J. D. Whelpley and Jacob J. Storer.
Reissue 3,857- March 1 , 1870. Improvement in the Use and
Application of Fuel in Metallurgic and Other Furnaces. J. D.
Whelpley and Jacob J. Storer.
103,695—May 31 , 1870. Improvement in Feeding Fuel to Metal
lurgic and Other Furnaces. J. D. Whelpley and Jacob J. Storer .
103,804-May 31 , 1870. Improvements in Feeding Powdered
Fuel to Metallurgic and Other Furnaces. J. D. Whelpley and Jacob
J. Storer.
111,614 - Feb. 7 , 1871. Improvement in Apparatus for Feeding
Pulverized Coal to Furnaces. Thomas Russel Crampton .
111,615—Feb . 7, 1871. Improvement in Apparatus for Distribut
ing and Feeding Powdered Fuel to Furnaces. Thomas Russel
Crampton .
111,616—Feb. 7 , 1871. Improvements in Furnaces for Burning
Pulverized Fuel under Steam -Boilers, Evaporators, etc. J. D.
Whelpley and Jacob J. Storer.
116,903 —— July 11 , 1871. Improvement in Treatment of Ores by
means of Fuel , Chemicals and Fluxes . J. D. Whelpley and Jacob
J. Storer.
120,007 — Oct. 17 , 1871. Improvement in Apparatus for Feeding
Fuel into Furnaces. John Y. Smith .
120,008 - Oct. 17 , 1871. Improvement in Apparatus for Feeding
Pulverized Fuel to Metallurgical and Other Furnaces. John Y. Smith .
120,680—Nov. 7, 1871. Improvement in Apparatus for Feeding
>
Pulverized Fuel to Furnaces . John Y. Smith .

184,122—Nov . 7, 1876. Improvement in Pulverized Fuel Feeder
for Smelting -Furnaces. William West .
185,592 —— Dec. 19, 1876. Improvement in Apparatus for Intro

ducing Powdered Fuel into Furnaces. George K. Stevenson .
227,176 — May 4, 1880. Feeder for Pulverized Fuel. John G.
McAuley and William West.
234,395—Nov. 16, 1880. Hydrocarbon -Furnace. William D.
Dickey.
238,891 — March 15, 1881. Device for Feeding Fine Fuel. Au
gustin Greiner.
240,265-Apr. 19, 1881. Apparatus for Feeding Fuel. Charles
H. Palmer .
243,593 — June 28, 1881. Furnace. John G. McAuley.
245,427 — Aug. 9, 1881. Mechanism for Feeding Fine Fuel . John
D. Averell .
247,570 — Sept. 27, 1881. Fuel-Feeding Apparatus. John G.
McAuley.
261,864 - Aug. 1 , 1882. Feeding Light Fuel . Henry Mason.
265,347 – Oct. 3 , 1882. Burning Pulverulent Fuel. Ernest Tour
angin .
292,237 — Jan. 22, 1884. ‘ Apparatus for Feeding and Consuming
Fine Fuel . Julius Leede .
295,145—Mar. 18, 1884. Coal-Dust Feeder. Henry A. Bradley.
317,712 — May 12, 1885. Hydrocarbon Injector. William W.
Brisben .
327,210—Sept . 29, 1885. Feeding Fine Fuel . William West
lake .
332,975 —- Dec. 22, 1885. Feeding Fuel to Boilers. Samuel W.
Valentine .
338,105 — Mar. 16, 1886. Feeding Fine Fuel . William Rogers.
>
406,753—July 9, 1889. Apparatus for Burning Culm . Allan

Mason .
411,555—Sept . 24 , 1889. Apparatus for Burning Coal and Hy
drocarbon Fuel in Combination . Allan Mason .
414,322—Nov . 5, 1889. Coal- Dust Feeder. Edmond Pait.
438,852 — Oct. 21 , 1890. Apparatus for Burning Coal and Hydro
carbon Fuel in Combination . Allan Mason .
457,589-Aug.
— 11 , 1891.. Rotary Furnace for Burning Cement ,
Lime, etc. Jose F. De Navarro.
PATENTS CONCERNING POWDERED FUEL . 285
464,514 - Dec. 8, 1891. Rotary Furnace for Burning Cement.

Jose F. De Navarro.
475,715 — May 24, 1892. Firing Apparatus for Use with Coal
Dust. Carl Wegener.
494,375 — Mar. 28, 1893. Pulverized Fuel Feeder. Hugh S.
Grigsby.
510,788 — Dec. 12, 1893. Pulverized Fuel Burner. Frederick H.
Hawkins.
511,004 — Dec. 19, 1893. Firing Apparatus for Use with Coal
Dust . Carl Wegener.
516,652 — Mar. 20, 1894. Fine- Fuel Furnace. Hermann Kluep
fel.
517,632-Apr. 8, 1894. Fine-Fuel Furnace. Carl Wegener.
519,784 — May 15, 1894. Coal-Dust Firing Apparatus. Carl
Wegener.
524,579 — Aug. 14, 1894. Fuel-Feeding Device. Joseph Davies.
531,160—Dec. 18, 1894. Apparatus for Burning Granular Fuel .
Colin W. Claybourne.
537,108 — Apr. 9, 1895. Coal -Dust Feeder for Furnaces. Alfred
Friedeberg
540,114 - May 28, 1895. Method of and Apparatus for Mixing
Coal -Dust and Air for Combustion . Constanz Schmitz .
551,074—Dec. 10, 1895. Feeding Appliance for Coal -Dust. Fer
dinand De Camp.
551,098—Dec. 10, 1895. Apparatus for Burning Fine Fuel. Wil
liam Melvin Russel .
553,696 - Jan. 28, 1896. Fuel- Burner. Gamaliel C. St. John .
554,327 - Feb . 11 , 1896. Coal-Dust Firing Apparatus. Carl
Wegener.
558,875—Apr. 21 , 1896. Apparatus for Burning Coal - Dust.
Georg Hillinger.
Reissue 11,8164- Apr. 3 , 1900. Apparatus for Burning Coal-Dust.
Georg Hillinger.
558,875 — Apr. 21 , 1896. Apparatus for Burning Coal- Dust .
>
Georg Hillinger.
563,789 — July 14, 1896. Feeding Pulverized Fuel. Edward C.
mith .
568,599 — Sept. 29, 1896. Method of and Apparatus for Calcining

Cement. Clifford Bonneville .
571,139—Nov. 10, 1896. Fuel- Feeder for Furnaces. Wilhelm
Ruhl.
577,995 — Mar. 2, 1897. Furnace for Burning Coal- Dust. Gustav
Unger.
581,244 - Apr. 20, 1897. Apparatus for Feeding Comminuted
Fuel to Furnaces. Charles J. Allen and Frank R. Tibbitts.
582,587 — May 11 , 1897. Apparatus for Utilizing Heat of Fur
nace Products. Alexander Marshall Hay.
626,981 —June 13, 1899. Feeding Device for Steam -Boiler Fur
naces. Joseph Davies.
645,031 - Mar. 6 , 1900. Apparatus for Burning Pulverized Fuel .
Edward H. Hurry and Harry J. Seaman .
658,069 — Sept. 18, 1900. Method of Feeding Pulverized Fuel .
Frederick H. Lewis.
661,700 — Nov . 13 , 1900. Apparatus for Burning Pulverized Fuel .
Frederick H. Lewis .
691,337—Jan . 14 , 1902. Apparatus for Feeding Fine Fuel . Rolla
C. Carpenter.
759,356 — May 10 , 1904. Method of Burning Portland Cement
Clinker, etc. Thomas A. Edison .
759,357 - May 10 , 1904. Apparatus for Burning Portland Cement
Clinker, etc. Thomas A. Edison .
TABLES.
CHAPTER XXI .
TABLE 13
COMPOSITION IN WEIGHT OF THE ATMOSPHERIC AIR .
02 N2 Air 0, N2 Air 02 N2 Air

1
0.1 0.33 0.43 3.3 11.05 14.35 20.0 66.96 86.96
0.2 0.67 0.87 3.4 11.39 14.79 21.0 70.26 91.26
0.3 1.00 1.30 3.5 11.72 15.22 22.0 73.56 95.56
0.4 1.34 1.74 3.6 12.06 15.63 23.0 77.00 100.00
0.5 1.67 2.17 3.7 12.39 16.09 24.0 80.16 104.16
0.6 2.01 2.61 3.8 12.72 16.52 25.0 83.56 108.56
0.7 2.34 3.04 3.9 13.06 16.96 26.0 86.96 112.96
0.8 2.68 3.48 4.0 13.40 17.40 27.0 90.36 117.36
0.9 3.01 3.91 4.1 13.73 17.83 28.0 93.76 121.76
1.0 3.35 4.35 4.2 14.06 18.26 29.0 97.12 126.12
1.1 3.68 4.78 4.3 14.40 18.70 30.0 100.48 130.48
1.2 4.02 5.22 4.4 14.74 19.14 31.0 103.84 134.84
1.3 4.35 5.65 4.5 15.07 19.57 32.0 107. 20 139.20
1.4 4.69 6.09 4.6 15.40 20.00 33.0 110.54 143.54
1.5 5.02 6.52 4.7 15.74 20.44 34.0 113.88 147.88
1.6 5.36 6.96 4.8 16.08 20.88 35.0 117.22 152.22
1.7 5.69 7.39 4.9 16.41 21.31 36.0 120.56 156.56
1.8 6.03 7.83 5.0 16.74 21.74 37.0 123.90 160.90
1.9 6.36 8.26 6.0 20.04 26.04 38.0 127.34 165.24
2.0 6.70 8.70 7.0 23.44 30.44 39.0 130.58 169.58
2.1 7.03 9.13 8.0 26.80 34.80 40.0 133.92 173.92
2.2 7.37 9.57 9.0 30.14 39.14 41.0 137.22 178.22
2.3 7.70 10.00 10.0 33.48 43.48 42.0 140.56 182.56
2.4 8.04 10.44 11.0 36.78 47.78 43.0 143.82 186.82
2.5 8.37 10.87 12.0 40.08 52.08 44.0 147.12 191.12
2.6 8.71 11.31 13.0 43.48 56.48 45.0 150.56 195.56
2.7 9.04 11.74 14.0 46.88 60.88 46.0 154.00 200.00
2.8 9.38 12.18 15.0 50.24 65.24 47.0 157.16 204.16
2.9 9.71 12.61 16.0 53.60 69.60 48.0 160.32 208.32
3.0 10.02 13.02 17.0 56.94 73.94 49.0 163.73 212.72
3.1 10.38 13.48 18.0 60.28 78.28 50.0 167.12 217.12
3.2 10.71 13.91 19.0 63.62 82.62 60.0 200.96 260.96
TABLE 14
COMPOSITION IN VOLUMES OF THE ATMOSPHERIC AIR .
02 N2 Air 02 N2 Air 02 N2 Air

OOO
0OWN
12.94
15.86
16.34 ន 79.97
96.16
100.97
00
13.32 16.82 105.78

0.4 13.70 17.30 110.58
NN
0.5 14.08 17.78 115.39

COWN
0.6 3.8 14.46 18.26 25.0 120.20

0.7 3.9 14.85 18.75 26.0 125.01
0.8 4.0 15.23 19.23 102.82 129.82
NOOOOORWNO
0.9 4.1 15.61 19.71 106.62 134.62

WONNNNNNNNNN
.0 4 2 15.99 20.19 29.0 110.43 139.43

5.29 4.3 16.37 20.67 30.0 114.24 144.24
.2 4.4 16.76 21.15 31.0 118.05 149.05
行路仍以 2 门 GBn8 衍 5309
3 4.5 17.13 21.63 32.0 121.85 153.85

.4 4.6 17.52 22.12 33.0 125.66 158.66
.5 4.7 17.89 22.59 34.0 129.47 163.47
.6 4.8 18.27 23.07 35.0 133.28 168.28
4.9 18.65 23.55 36.0 137.09 173.09
5.0 24.04 37.0 140.90 177.90
6.0 28.85 38.0 144.70 182.70
7.0 33.66 39.0 148.51 187.51
8.0 38.46 40.0 152.32 192.32
9.0 43.27 41.0 156.13 197.13
10.0 42.0 159.93 201.93
393
110 .89 43.0 163.74 206.74

12.0 44.0 167.55 211.55
12.49 13.0 49.50 45.0 171.36 216.36
12.97 14.0 53.31 46.0 175.17 221.17
13.45 15.0 57.12 47.0 178.98
13.93 16.0 60 . 48.0 182.78 230.78
14 42 17.0 64.74 49.0 186 59 235.59
14 90 18.0 68.54 50.0 190.40 240.40
15.38 19.0 72 . 60.0 228.48 288.48
TABLES . 289
TABLE 15
WEIGHTS AND SPECIFIC GRAVITIES OF Gases FOUND IN FURNACES.
SPECIFIC GRAVITY Weight of 1 Lit.

at 0° C. 760 mm .,
Gas Formula 45° Latitude
Hydrogen Air = 1 and at Sea
| 2 , Mol. Weight Level
Acetylene. C2H2 25.947 89820 1.16143 Grms.

Atmospheric Air . 1.00000 1.293052 "
40
Ethane. C2H6 29.947 1.03667 1.34047
Ethylene.. CH4 27.947 0.96744 1.25095
Carbon Monoxide CO 27.937 0.96709 1.25050
Carbon Dioxide. CO2 43.900 1.51968 1.96503
Hydrogen .. H2 2.000 0.069234 0.089523 "
Hydrogen Sulphide . H2S 34.000 1.17697 1.52189 06
Methane.. CHA 15.974 0.55297 0.71502
Nitrogen .. N2 28.024 0.97010 1.25440
Oxygen .. O2 31.927 1.0521 1.43908
Sulphur Dioxide . SO2 63.927 2.21295 2.86146
Water Vapor . H2O 17.963 0.62182 0.80405
Landolt and Boernstein , lve, tri . , page 77 .
TABLE 16
SPECIFIC HEATS PER MOLECULAR VOLUMES.
Gas Under Constant Pressure Under Constant Volumo
PerfectGases ( O2 , N 2, H2, CO ) . 6.83 x 10-3 + 1.2 X10-6t 4.83 X10-3 + 1.2 X10 - 0
Water Vapor, H 20. 8.08 X 10-3 + 5.8 X 10 -60 6.08 X10-3 + 5.8x10-66
Carbon Dioxide, CO 2 8.52 x 10-3 + 7.4 X 10 -6t 6.52 X 10-3 + 7.4 x 10-64
Methane , CH ... 9.78 X 10-3 + 12. X10 - 6t 7.78X10-3 + 12. X10-04
TABLE 17
SPECIFIC HEATS OF GASES PER KILOGRAM .
Gases Under Constant Pressure Under Constant Volumo
Oxygen .. .213 + 38 X 10 - 6t .150 + 38 X 10 - Ot

Nitrogen and Carbon Monoxide. .243 + 42 x 10-60 .171 + 42 X 10 - 6t
Hydrogen . 3.400 + 600 X 10 -- 60 2.400 + 600 X 10 -67
Water Vapor .447 + 324 X 10 - 6t .335 + 324 x 10-60
Carbon Dioxide .193 + 168 X 10 - 60 .150 + 168 X 10 - 6t
Methane . .608 + 748 X 10 - .491 + 748 X 10 - Ot
TABLE 18
MEAN SPECIFIC HEATS, BETWEEN to' AND t , ° OF GASES PER KILOGRAM .
Oxygen .. .213 + 19 X 10 - 60 .150 + 19 X 10 - 6

Nitrogen and Carbon Monoxide. .243 + 21 X 10 -67 .243 + 21 X 10 - 67
Hydrogen . 3.400 + 300 X 10 - 6t 2.400 + 300 X 10 -
Water Vapor , 447 + 162 X 10 -- 6t 335 + 162 X 10 -67
Carbon Dioxide . .193 + 84 X 10 -6t .150 + 84 X 10 - 6t
Methane or Marsh Gas .. .608 + 374 X 10 - 6t .491 + 374 x 10-6
TABLES . 291
TABLE 19
VALUES OF FACTORS at AND ( 1 + at) FOR TEMPERATURES BETWEEN 0 ° AND 1800 ° C.
t at
|(1+at) t at
(1+ at: t at ( 1 + at ) t
at |(K ( 1 + at)
10 0.0367 1.0367 310 1.1377 2.1377 610 2.2387 3.2387 910 3.3397 4.3397
20 0.0734 1.0734 320 1.1744 2.1744 620 2.2754 3.2754 920 3.3764 4.3764
30 0.1101 1.1101 330 1.2111 2.2111 630 2.3121 3.3121 930 3.4131 4.4131
40 0.1468 1.1468 340 1.2478 2.2478 640 2.3488 3.3488 940 3.4498 4.4498
50 0.1835 1.1835 350 1.2845 2.2845 650 2.3855 3.3855 950 3.4865 4.4865
60 0.2202 1.2202 360 1.3212 2.3212 660 2.4222 3.4222 960 3.5232 4.5232
70 0.2569 1.2569 370 1.3579 2.3579 670 2.4589 3.4589 970 3.5599 4.5599
80 0.2936 1.2936 380 1.3946 2.3946 680 2.4956 3.4956 980 3.5966 4.5966
90 0.3303 1.3303 390 1.4313 2.4313 690 2.5323 3.5323 990 3.6333 4.6333
100 0.3670 1.3670 400 1.4680 2.4680 700 2.5690 3.5690 1000 3.6700 4.6700
110 0.40371 1.4037 410 1.5047 2.5047 710 2.6057 3.6057 1020 3.7434 4.7434
120 0.4404 1.4404 420 1.5414 2.5414 720 2.6424 3.6424 1040 3.8168 4.8168
130 0.4771 1.4771 430 1.5781 2.5781 730 2.6791 3.6791 1060 3.8902 4.8902
140 0.5138 1.5138 440 1.6148 2.6148 740 2.7158 3.7158 1080 3.9636 4.9636
150 0.5505 1.5505 450 1.6515 2.6515 750 2.7525 3.7525 1100 4.0370 5.0370
160 0.5872 1.5872 460 1.6882 2.6882 760 2.7892 3.7892 1120 4.1104 5.1104
170 0.6239 1.6239 470 1.7249 2.7249 770 2.8259 3.8259 1140 4.1838 5.1838
180 0.6606 1.6606 480 1.7616 2.7616 780 2.8626 3 8626 1160 4.2572 5 2572
190 0.6973 1.9973 490 1.7983 2.7983 790 2.8993 3.8993 1180 4.3309 5.3306
200 0.7340 1.7340 500 1.8350 2.8350 800 2.9360 3.9360 1200 4.4040 5.4040
210 0.7707 1.7707 510 1.8717 2.8717 810 2.9727 3.9727 1220 4.4774 5.4774
220 0.8074 1.8074 520 1.9084 2.9084 820 3.0094 4.0094 1240 4.5508 5.5508
230 0.8441 1.8441 530 1.9451 2.9451 830 3.0461 4.0461 1260 4.6242 5.6242
240 0.8808 1.8808 540 1.9818 2.9818 840 3.0828 4.0828 1280 4.6976 5.6976
250 0.9175 1.9175 550 2.0185 3.0185 850 3.1195 4.1195 1300 4.7710 5.7710
260 0.9542 1.9542 560 2.0552 3.0552 860 3.1562 4.1562 1400 5.1380 6.1380
270 0.9909 1.9909 570 2.0919 3.0919 870 3.1929 4.1929 1500 5.500 6.500
280 1.0276 2.0276 580 2.1286 3.1286 880 3.2296 ' 4.2296 1600 5.872 6.872
290 1.0643 2.0643 590 2.1653 3.1653 890 3.2663 4.2663 1700 6.239 7.239
300 1.1010 2.1010 600, 2.2020 3.2020 900 3.3030 4.3030 1800 6.606
1
7.606
TABLE 20
INTERNATIONAL ATOMIC WEIGHTS FOR 1905 OF THE MORE USUAL CHEMICAL ELEMENTS.
RESU55834
O = 16 H - 1
Aluminum . 27.1 26.9
Antimony . 120.2 119.3
Arsenic . 75.0 74.4
Barium Вя 137.4 136.4
Bismuth . 208.5 206.9
10.9
SE 11.0
Boron..
Bromine. 79.96 79.36
5E-:
Cadmium 112.4 111.6
Calcium.. 40.1 39.7
Carbon .. 12.0 11.91
Chlorine. 35.45 35.18
Chromium . 52.1 51.7
Cobalt 59.0 58.55
Copper. 63.6 63.1
Fluorine. 18.9
Ezoz247EUFFED>
19.0
Gold... Au 197.2 195.7
Hydrogen . 1.008 1.000
lodine 126.97 126.01
Iridium . 193.0 191.5
Iron . Fe 55.9 55.5
Lead.. 206.9 205.35
Lithium Li 7.03 6.98
Magnesium Mg 24.36 24.18
Manganese. Mn 55.0 54.6
Mercury. Hg 200.0 198.5
Molybdenum . Mo 96.0 95.3
Nickel Ni 58.7 58.3
Nitrogen 14.04 13.93
Oxygen . 16.00 15.88
Phosphorus. 31.0 30.77
Platinum . 194.8 193.3
Potassium . 39.15 38.85
Silicon . 28.4 28.2
Silver . 107.93 107.11
Sodium 23.05 22.88
Strontium 87.6 86.94
Sulphur . 32.06 31.82
Tin .. 119.0 118.1
Titanium . 48.1 47.7
Tungsten . 184.0 182.6
Uranium 238.5 236.7
Vanadium . 51.2 50.8
Zinc . 65.4 64.9
TABLES . 293
TABLE 21
Conversion Tables !.
CONVERSION OF TEMPERATURE SCALES .
Fahrenheit F
Centigrade = C
Absolute A
90
F + 32
5
F
9 (A — 273) + 32
5
5 ( F - 32)
C
9
C = A - 273
5 (F – 32)
A + 273
9
A C + 273
One large calory = 3.968 32 B. t.t u . Log. 0.598 6067.

One B. t . uu . = 0.251 995 8 ; large calory 1.401 3933.
One inch = 2.54 centimeters ; log. 0.404 8346 .
One foot = 30.480 centimeters ; log. 1.484 0158.
One centimeter = 0.393 700 inch ; log. 1.595 1654.
One square inch = 6.541 63 square centimeters; log. 0.809 6692.
One square foot = 929.034 square centimeters ; log. 2.968 0317.
One square centimeter = 0.155 000 square inch; log. 1.1903308 .
One square centimeter = 0.001 076 387 square foot; log. 3.031 9683
One cubic inch = 16.387 16 cubic centimeters ; log. 1.214 5038 .
One cubic foot = 28 317.0 cubic centimeters ; log. 4.452 0475.
One cubic centimeter = 0.061 023 4 cubic inches; log. 2.7854 962.
-
1 Carl Herig , “ Conversion Tables,"

One cubic meter = 61 023.4 cubic inches; log. 4.785 4962,.

One cubic meter 35.314 5 cubic feet; log. 1.547 9525.
One kilogram = 2.204 62 pounds (av.) ; log. 0.343 3342 .
One pound ( av . ) = 0.453 593 4 kilogram ; log. 1.656 6658.
Pressures in Terms of Water, Mercury, Atmosphere :
1 kilogram per sq . mm. 100 kg. per sq. cm.

1422.34 lb. per sq . in. 3.153 0034
1 kilo per sq. cm . 2048.17 lb. per sq . ft . 3.311 3659

32.8083 ft . of water 1.515 9842
= 28.9572 in , of mercury 1.461 7564
735.514 mm . of mercury 2.866 5910
.967782 atmosphere 1.985 7774
1 atmosphere 2116.35 lb. per sq. ft. 3.325 5885

760. mm. of mercury 2.880 8136
33.9006 ft . of water 1.530 2068
14.6969 lb. per sq . in . 1.167 2260
1.03329 kg. per sq . cm . .014 2226
1 pound per square inch 703.067 kg. per sq. m . 2.846 9966
51.7116 mm . of mercury 1.713 5876
= 2.306 65 ft . of water .362 9808
2.035 88 in . of mercury .308 7530
= .070 306 7 kg. per sq . cm . 2.846 9966
.0680415 atmosphere 2.832 7740
1 inch of mercury column 70.7310 lb. per sq. ft. 1.849 6095
25.40005 mm . of mercury 1.404 8346
1.13299 ft . of water 0.054 2278
= .491187 lb. per sq . in . 1.691 2470
= .0345337 kg. per sq. cm . 2.538 2436
= .0334211 atmosphere 2.524 0210
TABLES . 295
1 foot of water column 304.801 kg. per sq . m . 2.484 0158

62.4283 lb. per sq . ft . 1.795 3817
22.4185 mm, of mercury 1.350 6068
.43353 lb. per sq . in . 1.637 0192
.0304801 kg. per sq. cm . 2.484 0158
= .029498 atmosphere 2.469 7932
1 meter of water column 204.817 lb. per sq . ft. 2.311 3659

73.5514 mm . of mercury 1.866 5910
3.28083 ft. of water .515 9842
2.89572 in. of mercury .461 7564
1.42234 lb. per sq . in. .153 0034
.1 kg. per sq . cm . 1.000 0000
.0967782 atmosphere 2.985 7774
1 mm . of mercury column 13.59593 kg . per sq. m . 1.133 4090

2.78468 lb. per sq . ft. .444 7749
.044606 ft . of water 2.649 3932
.03937 in. of mercury 2.595 1654
.019338 lb. per sq . in . 2.286 4124
.01359593 meter of water 2.133 4090
: .00135959 kg. per sq . cm . 3.133 4090
= .00131579 atmosphere 3.119 1864
1 inch of water column 25.4001 kg . per sq. m . 1.404 8346

5.20236 lb. per sq . ft. .716 2005
- 1.8682 mm . of mercury .271 4256
= .073552 in . of mercury 2.866 5910
= .036128 lb. per sq . in . 2.557 8380
= .0254001 m . of water 2.404 8346
= .00254001 kg. per sq . cm . 3.404 8346
= .0024582 atmosphere 3.390 6120
Linear Velocities .
One meter per second 3.280583 feet per second 0.515 9842
One foot per second .304801 meter per second 1.484 0158
22
TABLE
FUELS
.'OF
POWERS
CALORIFIC
AND
ANALYSES
ELEMENTARY
ELEMENTARY
ANALYSI
.A sh
AND S
ANALYSIS
.
ELEMENTARY MOISTUR
F
- REE E
Basis
Remarks
FUELS
100
Mahler Formula Qol =

H34500 + C8140—-3000 ( 0 + N )
Calorific Power Calculated from
Experimental Calorific Power

Exp. Calorific Power on Ash and
Moisture -free Basis = Qo
Nitrogen
Oxyen and
Hydrogen
pic Water
Carbon
Nitrogen
% % %
6%% % % % Calories Calories Calories
19 Hydrogen
•
09 Oxygen and
09 Carbon
.Anthracite
Pennsylvania 2.426
2.201
95.373
5.900
3.450
2.199
1.995 7844
86.456 8256 8450
Muresère
)(Ifrom
La
Anthracite 3.255
1.504
95.241
4.400
4.700
2.969
1367
86.564 7468 8216 8172
D
)(T uong
- onking
Hay
from
Anthracite 2.157
4.988 7533
92.855
4.000
3.260
4.626
2.000
86.114 8121 8152
Coal
.Kebao
Anthracite 3.479
3.065
93.456
5.450
2.800
2.731
2.733
85.746 7828 8532 8531 0.70
%S
.
Coal
Inthracite
Commentry 5.392
115
3.
91.493
5.400
1.775 7850
5.005
2.892
84.928 8456 8260 S
%
0.797
Coal
Anthracite
Blanzy 6.828
3.172
90.000
6.300
1.760
6.278
2.916
746
82 7773 8206 8215
.
Coal
Anthracite
Combe
Grande 4.591
3.930 7850 8540 8670
91.459
7.250
0.831
4.220
€1
3
84.068
.
Coal
Anthracite
Creusot 3.827
3.784
92.389
1.450
1.800
3.703 8404 8687 8712
3.661
89.386
Anthracites and
Anthracitic Coals
Shaft 4.475
4.269
91.256
1.700 8392
1.350
4.338
4.139
88.473 8656 8767 S
%
0.673
9 Marc
, t.
SAnzin
.Shaft
t.
Jean
,Combe
SGrande 4.353
4.462
91.185
3.800
0.610
4.161
2
4 65
87.164 8371 8756 8832
10 collected
%Ash
3.10
aoliere
M.-lLa
Roche 5.512
4.382
90.106
4.000
0.900 8417
5.242
4.167
85.691 8767 8681
11
calorimeter
in
Seam
F 5.494
4.402
90.104
4.000
0.625
5.240
4.198
85.934 8426 8834 8688 S
%
0.810
, erdinand
Aniche 8689
Shaft
Seamncles
,,Ehick
TAnzin 6.135
4.670
89.195
5.000 8051
1.100
5,761
4.385
83.754 8574
Shaft
,Mdagny mp
Seam 6.273
4.838
88.889
9.700
1.200
5.589
4.311
200 7839 8787 8717 S
%
2Roncha 79 s0.556
%
0.858
.8
S haft 5.168
4.800
90.032
1.550
1.000
5.036
4.678 8839 8829
N,Seam
9 0
1+1 o.
Lens 87.736 8614
7.295
4.870
87.835
1.500
7.076
4.724
200
85. 8380 8639 8611
Carmaux
.. 8788
5.632
4.838
89.530
3.000
1.350
5.386
4.628
85.636 8482
8867
5.743
5.026
89.231
4.000 8857 8829 0.591
S
%
Seam
, th
8
Etienne 1.250 8392
5.432
4.772
81.546
Bituminous and
St. 8667 8452
191 ),( ard
G
Ports 8.640
4.8401
8.910.86.520
0.7801
7
. .700 7646
4.370
78.2401
23456789
Semi-Bituminous Coals
20 Bethune
. 5.374
87.031
4.100
1.200
7.193
07.595
5: 89
82.418 8210 8668 8710
21 Seam
.5SNhaft
,D
Lenso.
usouich 7.303
5.436
87.2611
3.000
1.050
7.007 8395
5.216
83.727 8749 8758
22 Firminy 9.134
5.578
85.388
3.600
1.225
8.593
5.309
273
81 8161 8573 8601
23 Seam
,T
Montrambert
hick 9.935
5.543
522
84
3.000
0.844
3
9.553
530
81273 8268 8598 8494
Commentry 8.732
5.604
85.664
3.400
3.000
8.173
5.245
80.182 7870 8408 8644 S
%
0.591
.
Coal
C
L annel
(). ancashire
Wigan 5.715
5.718
88.567
10.900
0.600
5.058
5.060
78.382 7761 8768 9011
.
Coal
Cannel
Niddric 9.633
6.574
4.700.83.793
3.950
8.800 7703
6.005
76.545 8431 8800
Cas Coals
10.416
5.635
83.949
4.800
4.300 7790
9.468 8570 8465 S
%
0.687
27 .
Montvic 5.122
76.310 0.481
s
%
Shaft
,S
.M arie
3dte
Blanzy
Seam 10.462
5.273
84.265
1.900
3.900
9.855 7866
1.967
79.3781 8350 8364
Shaft
,.E
S te
ugenie
,1st
Blanzy
Seam 11.142
5.684
83.174
7.800
1.700
10.083
5.144
75.273 7486 8270 8397
Bourran
,Decazeville 7408 8083 8197
S
%
1.28
12.827
5.638
81.535
4.850
3.500
11.756
5.167
74.727
Coals
T
., ramont
Decazeville 7494 7837 7897
S
%
0.82
31 15.606
5.673
78.721
80
2.800
5
14.922
15.425
75.273
Lignitie
23.045
5.942
71.013
18.500
16.500
14.979
3.862
46,159 4885 7039 7141 S
%
0.273
32 deeu
Terre
-F 5.058
25.707
235
69
4.750
0.710 6284
24.303
4.782
65.455 6646 6610 %S
1.47
33 (S) tyria
Trifail 28.632
5.007
66.361
6.750 5536
3.144
25.799
4.512
59.795 6076 6270 S
1.50
%
34 Vaugirard
..
Lignites
PB35ohemia
| eat 36.824
5
, .692
57.214
0.950
6.125
34.230
5.542
53.183 5489 5903 5609
ir
dried
,APine
Norwegian 42.900
6.019
51.081
0.333
6.940
39.780
5.581
47.366 4477 4828 4497
36 43.685
5.878
50.437
0.750 4689 4823
.Lorraine
Oak 6.920
40.335
5.427
46.568 4329
37 49.290
'6 .170 4200
44.444
0.750
6.920 4200 4264
H 0
0ellulose
1C.(C12
I38 49.390
6.170
44.440
Woods
39 ,-C
Grand
0.231
97.353
7.800
0.500
2.215
0.212
273
89
2.416 7010 7920 7932
ofombe
Metallurgical 8001 7957
40 2.745
0.463
96.792
4.200
0.500 7665
2.629
0.444
92.727
Commentry 1.637 7787
0.654
97.709 8044 8130
41 Anzin
3.200
0.500
1.585
0.633
94.582
....
the
at
(Made
42 Pennsylvania
From 2.286
0.730
984
90.
5.900
0.233
2.146
0.685
91.036
School
Paris 7528 8036 8078
Cokes
Anthracite
Mines
of
23.911
3.437
652
72
3.600
0.233 6155
23.053
3.311
70.036 6384 6382
43 Commentry
. 16.362
4.732
78.906
4.600
3.600
15.020
4.314
72.436 7119 7754 7565
44
15.029 st
Coals
,1Oxidized
Blanzy
Sample
5.20079.459
3.500
5.03213.722
72.546 7169 78521 7918
45 d
;5Sample
,2.512
.Blanzy
Combustibles
."sur
les
,“E
Mahlerrom
1Ftudes
P. 7
TABLE
23
.
FURNACES
REGENERATIVE
FOR
MEASUREMENTS
STACK
OF
TABLE
PRECHTL
INNER
INNER
WIDTH
>
CON
HOURLY K
,-
-.pP OKS
AT AT
OF
SUMPTION 68,-7 1do
- .223
3
V
Top BABE
THEORETICAL
Draugbt
Stack
REAL
V - Vagh
,
-
P
the
of
Weight
,
Without With
K
Coal Wood Gas Air
Column Column
at at
D
F
P -- P
C250
°. C0° . 1.293 F
Section
Section
Section
Section
h
Due to Flues
Resistance
Resistance
Due to Flues
Velocity
Mean Diameter
Mean Section
Top
Stack of Square
Stack of Square
Stack of Circular
Stack of Circular
A
Height of Stack
.
m
.Sec
per
m
Sec
m3 3
0.04
0.2415
B Velocity
Soo
0.10
0.592
11 0.140
0.86 3
6.0
0.43 18 0.197
180
888
17
con
18 13.05 0.247
480
18 0.352
75 20 18 13.60
0.62 18 0.477
100 200 22
0.741
昭限44 好引
150 300 26 0.73 18
GooerWNN
0.83 0.93 18 0.995

200 400 23
438 吃品必 % 3%
250 500 30 0.91 1.03 18
MODAW NOB Volume of Enclosed Gas Column

300 600 31 18
350 700 32
400 800 33
28
B 以 % 83323
450 900 34 1.17

888888888
500 1000 34 ..23

1.29 117.40
550 1100 35
600 1200 35 .34 98.4 128.50 50 17.35
118.0
27 153.00 350
700 36 1.44
800 134.2 173.00 83.00 17.70
900 61 157 204
1000 68 178.0 231.00
2012 135.0 260.00
1100
1200 216.0 144.5 280
以脱胎础昭四孔孔小刀
232.0 300.00
四四张
1300
1400 253.0 327
OOOO
1500 272.5 353.00
SUNNN
四
888
ទំ និ ត
జరు
888888888.nooo
行 299
CSR8
TABLES . 299
TABLE 24
Coefficients of heat conductivity in calories (sq . m . -hour- ° C . "):

.
Masonry 1.3 to 2.1

Firebrick . .7
Air ... .0175 to .0205
Cement. .059
Water . . 44 to .56
Iron 40.00 to 70.00
Copper .. 330.00
66
1 Lanboldt & Boernstein , Phys. Chem . Tabellen." Also Ingenieurs Taschenbuch ."
Vol. II , pp. 408-411.
BIBLIOGRAPHY.
FUELS AND CALORIMETRY .
Metallurgy, Vol. I. Refractory Materials and Fuel . John

Percy, M.D.
Fuel and its Applications. E. J. Mills, D. Sc., and E. J. Rowan ,
C.E.
Fuel - Its Combustion and Economy. C. W. Williams, A. I. C. E. ,
and D. K. Clark .
Fuel . R. Galloway, F. C. S.
Fuel and Water. F. Schwakhöfer and W. R. Browne, M. A.
Introduction to the Study of Metallurgy. W. C. Roberts
Austen , F. R. S.
Fuel and its Efficiency. B. H. Thwaite, J. I. S. I. 1892 , Vol. I.
Geology of Coal. Chemistry of Coal . Coal. Edited by T. E.
Thorpe .
Coal. T. H. Huxley, F. R. S.
Sulphur in Coal . G. H. Bailey , J. S. C. I. 1889 .
Briquette Fuel. W. Colquhoun , P. I. C. E. Vol. CXVIII.
Heating Power of Coal. W. Thomson, F.C.S. J. S. C. I. 1889.
Fuel Purification . P. Bateson , J. S. C. I. 1891 .
Liquid Fuel for Mechanical and Industrial Purposes. E. A.
Brayley Hodgetts.
Liquid Fuel . L. T. Wright , J. S. C. I. 1887 .
Liquid Fuel. G. Stockfleth , J. S. C. I. 1894 .
Liquid Fuel. H. Aydon , P. I. C. E., Vol. LII.
Natural Gas . F. C. Phillips, J. S. C. I. 1894 .
Gaseous Fuel. Sir E. Lowthian Bell, J. I. S. I. , 1889, Vol. II.
Fuel Gas . B. Loomis, J. I. S. I. , Vol . II .
BIBLIOGRAPHY . 301
Use of Gas as a Heating Agent compared with Solid Fuel. G. E.

Davis, J. S. C. I. 1883.
Generation and Application of Gas. A. Wilson , J.S.C. I. 1883.
Gaseous Fuel . L. T. Wright, J. S. C. I. 1883.
Dowson Gas. J. S. C. I. 1888 .
Water Gas. J. F. Bell, J. S. C. I. 1889.
Fuel Value of Gases. H. L. Payne, J. S. C. I. 1893.
Gasification of Coal and Oil. P. Dvorkovitch, J. S. C. I. 1893.
Gaseous Fuel. G. Ritchie, J. W. of S. I. S. I. Vol. I.
Note on the Efficiency of Gaseous Fuels. F. A. Matthewman ,
J. W. of S. I. S. I .; Vol . I.
Sacrifice of Heat in Making Producer -Gas. J. Ormiston, J. W. of
S. I. S. I. Vol . I.
Notes on Water Gas. G. Ritchie, J. W. of S. I. S. I. , Vol . II.
On Gas Producers. H. J. Bowman , P. I. C. E. , Vol . LXXXIV .
Efficiencies of Gas Producers. C. F. Jenkin , P. I. C. E., Vol .
CXXIII .
Mond Producer-Gas. J. H. Darby, J. I. S. I. 1896, Vol. I.
Notes on Fuel. B. H. Thwaite, J. I. S. I. 1892 , Vol. I.
Manuel pour l'Essai des Combustibles et le Controle des Ap
pariels de Chauffage. F. Fischer.
Analyse Chimique et Essai des Combustibles. L. Campredon.
Tableaux des Essais de Combustibles Minéraux . Ad . Carnot.
Conference sur le pouvoir Calorifique des Combustibles et l'Ap
pareil de M. Mahler. E. Goutal .
Les Matières Volatiles et le Pouvoir Calorifique des Houilles.
E. Goutal .
Étude sur la Fusibilité des Cendres de Combustibles. Le Cha
telier et Chantepie .
Contribution à l'Etude des Combustibles. P. Mahler.
Etudes sur les Combustibles, Solides, Liquides et Gazeux . P.
Mahler.
Les Combustibles, Solides, Liquides et Gazeux . D. H. J. Phillips.
Pouvoir Calorifique des Combustibles, Solides, Liquides et Gazeux .
Scheurer-Kestner .
Essai des Combustibles. D. Siderski.
Fuels and Refractory Materials.. A. Humboldt Sexton .
Die Untersuchung und Bewerthung der Brennstoffe. P. Fritzsche.

Das gas als Leucht-, Heiz-, und Kraftstoff. 0. Pfeiffer.
Heizwert der Kohlen Osterreich -Ungarns und Preussisch - Schle
siens. Schwackhöfer.
Chemische Technologie des Brennstoffe. F. Fisher.
Ueber die Bestimmung der Nutzbaren Verbrennungswärme.
Dr. K. Kroeker.
Chemische und Calorimeterische Untersuchung von Brennstoffen .
Dr. H. Longbein .
Die Bestimmung des Heizwerthen von Brennmaterialen . H. von
Jüptner.
Metallurgical Laboratory Notes . H. M. Howe.
Ready Reference Tables. Carl Hering.
New Form of Calorimeter. W. Thomson , J. S. C. I. 1886 .
Traité Pratique de Calorimétrie Chimique. M. Berthelot.
Das Parische Calorimeter. Dr. G. Lunge.
COMBUSTION .
Flame. L. T. Wright, J. S. C. I. 1887.

Flame . F. J. Rowan , J. S. C. I. 1889 .
Flame . A. Smithells, J. C. S. 1892 .
Combustion . F. Siemens, J. I. S. I. 1886 , Vol. II.
Combustion of Coal in House Fires. J. B. Cohen , D. Sc . , and
G. Hefford , J. S. C. I. 1893 .
Luminosity of Flame. Profs. Lewes and Smithells, J. C. S. and
J. S. C. I. 1892 et sequa.
Combustion of Powdered Fuel. J. R. Crampton, J. I. S. I. 1873 .
Die Lehre von den Brennmaterialen . R. Krüger.
Die Gasfeuerungen . A. Ledebur.
Die Feuerungen mit flüssigen Brennmaterialen. J. Lew .
GAS ANALYSIS .
Volumetric Analysis . Francis Sutton , F. I. C. , F. C. S. 1896 .

The Chemical Analysis of Iron . Andrew Alexander Blair. 1902 .
Traité d'Analyse des Substances Minerales. Adolphe Carnot.
Ausgewählte Methoden des Analytischen Chemie. A. Classen .
BIBLIOGRAPHY. 303
Quantitative Analysis . J. C. Olsen .

Analyse des Gas. Pozzi -Escot.
Method of Gas Analysis. W. Hempel . Translated by L. M.
Dennie.
Techno -Chemical Analysis. G. Lunge. Translated by A. T.
Cohn .
Analytical Chemistry. F. P. Treadwell, Ph. D. Translated by
Wm. T. Hall, S. B.
The Gases of the Atmosphere. Wm . Ramsay.
THERMAL CHEMISTRY.
Application of Thermal Chemistry to Metallurgical Problems.

A. Pourcel , J. I. S. I. 1889, Vol . I.
Thermal Chemistry . M. Pattison Muir and M. Wilson .
Technische-thermochemische Berechnungen zur Heizung. A.
Naumann .
Lehr und Handbuch der Thermochemie . A. Naumann .
Thermo -dynamics and Chemistry. Duhem-Burgess.
CHIMNEYS.
Boiler Chimneys. R. Wilson Crosby .

Mechanical Engineers' Pocket Book . W. Kent .
Trans . A. S. M. E. , Vol . XI . De Volson Wood .
Steam Engines. Prof. Rankine.
Treatise on Heat. Thomas Box .
Trans. A. S. M. E. Vol . XI. R. H. Thurston .
Anlage und Betrieb der Dampfkessel. Reiche.
Encyklopädie. Prechtl's.
PYROMETRY .
Heisch & Folkard's Pyrometer. F. Hurter, J. S. C. I. 1886.

Le Chatelier's Pyrometer. W. C. Roberts - Austen , F. R. S. ,
J. I. S. I. 1891 , Vol. I.
Autographically Recording Pyrometer. W. C. Roberts-Austen .
J. I. S. I. 1892, Vol. I.
Recording Pyrometers. W. C. Roberts-Austen , J. I. S. I.

1893, Vol. I.
Platinum Pyrometers. H. L. Callendar, J. I. S. I. 1892, Vol. I.
A New Optical Pyrometer. B. H. Thwaite, J. I. S. I. 1892,
Vol . I.
High Temperature Measurements. Le Chatelier- Boudouard
Burgess.
Iron and Steel Institute . 1904 .
FURNACES.
Coke Ovens . A. B. Cochrane. Pr. Inst . Mech . Eng. 1801 .

New Form of Siemens Furnace. J. Head and M. P. Pouff,
J. I. S. I. 1889, Vol . II .
Regenerating Heat in the New Form of Siemens Furnace. R.
Ackerman , J. I. S. I. 1890, Vol . I.
Improved Design for Open -hearth Steel Furnaces. B. H.
Thwaite, P. S. S. I. , Vol. VI .
Calorific Power of Puddling Furnace. J. Cubillo, J. I. S. I. 1892 ,
Vol. I.
Efficiency of Reverberatory Furnaces. J. Cubillo, P. S. S. I.,
Vol. VI .
Die Oefen für Metallurgische Processe . A. Ledebur.
Chemische-calorische Studien über Generatorem und Martin
Oefen . Toldt .
Allgemeine Anweisung für den Bau und den Betrieb der Regen
erativ -Gasöfen . R. Schneider.
Regenerativ -Gasöfen . F. Toldt.
Travail des Metaux dérivés du Fer. L. Gages.
Metallurgical Calculations. Electro - Chemical and Metallurgical
Industry, Vol. III . J. W. Richards.
Le Chauffage Industriel et les Fours à Gaz.. Emilio Damour.
METALLURGY.
Iron , Steel and Other Alloys. Henry M. Howe.

Manufacture and Properties of Iron and Steel. Harry Huse
Campbell
BIBLIOGRAPHY . 305
Metallurgy of Steel . F. W. Harboard, Assoc. R. S. M. , F. I. C.

The Metallurgy of Steel. H. M. Howe.
The Metallurgy of Lead . H. O. Hofman.
Modern Copper Smelting. Edward Dyer Peters, Jr.
The Metallurgy of Zinc. Walter Renton Ingalls.
Introduction to the Study of Metallurgy . Sir W. C. Roberts
Austen .
Die Anlage und der Betrieb der Eisenhütten . E. F. Dürre.
Das Eisenhüttenwesen Schwedens . J. V. Ehrenwerth .
Abhandlungen über Metallurgie (deutsch von Kupelweiser) .
M. L. Gruner.
Des Ingenieurs Taschenbuch . Hütte ( Verein ).
Handbuch der Eisenhüttenkunde. A. Ledebur..
Ausfürliches Handbuch der Eisenhüttenkunde . H. Wedding
Traité de Métallurgie Générale. G. Schnabel .
Metallurgie de Zinc. A. Lodin .
Traité Théorique et Pratique de Métallurgie Générale. L. Babu.
Métallurgie. Urbain Le Verrier.
Aluminum . Dr. J. W. Richards.
INDEX.
A PAGE
PAGE Berthelot . .7 , 10
“ Ados " automatic gas ana Berthelot -Mahler apparatus
lyzer, section of . 187 ready for operation ... 196
Air, atmospheric, in a furnace, calorimeter .. 197
value of .. 59 Bibliography .300-5
leaks , when of importance . 136 Bicheroux furnace . .37, 73 , 76
necessary to have excess of Bied , J. , cited . 256
in aa direct- fired furnace . 71 Biedermann .. 46 , 95 , 116
pre-heating of . 49 Blacksmith's forge as a type of
primary 40 furnace .. 75
primary effect of pre-heat Block , perforated , section of .. 254
ing .. 122 Boëtius furnace . .72 , 73, 76
primary, secondary, and Boiler, steam . 1
gas, diagram showing steam , practice — field for
theoretical economies improvement in ... 84
from the pre -heating Boilers , for torpedo -boats, data
of ... 126 concerning 83
ratio of primary to see (marine ) multitubular,
ondary , modified by consideration of ... 78
steam . 63 multitubular, heat lost and
regenerator . 251 utilized in ... 84
secondary ... 11 steam , follow furnaces . 76
secondary, furnaces that steam ; heat lost when
recuperate by means of . 86 combustion is effected
total , advantage of recu by large excess of air .. 78
peration by. 110 Boiling points, apparatus for
total, pre-heating of.. 108 the determination of ... 157
volume of injected in the Bomb calorimeter . 193 , 194
chambers 137 calorimeter, calculation of
Asphaltum , action of in gas results from use of .... 202
producer .. 65 Breslauer, cited . .246
Atwater calorimeter modifica Bromine water. 180
tions .. 193 Buire -Lencauchez producers . 64
Bunte burette . 179
B.
C
Bardolle, L ... 46
Barus.. 156 Calories, hypothetical perfect
Batho furnace . 242 exchange of between
Belleville boiler. 2.78, 82 hot and cold gases..... 142
boiler and economizer, ser lost and utilized in furnaces
tion through . 79 , 80 fired with Siemens gas,
Belleville , M .. 79 double recuperation .... 95
308 INDEX
PAGE PAGE
Calories lost and utilized in fur Chloride, cuprous solution .... 180
naces fired with Siemens Coal , bituminous, temperature
90 of combustion of .. 22
gas, single recuperation .. 10 hydrocarbons of, action of
Calorific power defined ...
power under constant pres in gas- producer.. 65
sure . 10 powdered, met hod of using. 274
Calorimeter test, calculation of . 203 ultimate analysis and cal
Calorimeters . . 193 orific power of.. 131
Calorimetric corrections . 208, 209 Coal-dust, explosion of . 273
method of Berthelot . 129 proper care of .. 273
tests , accuracy obtainable stokers . 275
in ... 205 Coke oven . 265, 267
Calorimetry 192 Combustion , books on . 302
Calory , large, used in this book . 7 calculations . 235, 236 , 237
Campbell 50-ton basic fur complete, in boilers , with
nace .. . 106 , 107 , 109 out excess of air, desid
H. H. , cited . .246 , 266 eratum . 85
Cane, pyrometric , for protect general equation of the
ing wires of Le Chatelier temperature of. 14
thermo - electric pyrome graphic method for deter
ter .... 152 mining the temperatures
pyrometric, use of in tests . 135 of (diagram ) . 15
Carbon dioxide, action of in heat of.... 10
gas- producer. 65 phenomena, scientific data
dioxide, dissociation of... 31 necessary for the study
dioxide, effect of in gas of .. 7
producer 116 principles relating to.. 58
dioxide, injection into gas products at 1500 °, calories
producers of pure hot.. 117 contained in . 138
dioxide, proposed injection products, so -called regen
of in gas-producers.... 116 erated .. 51
dioxide, ratio of the vol products, method of as
ume of to that of nitrogen 59 certaining 22
hydrogen and ash , deter products, thermal capaci
mination of. 223 ties for the volumes
monoxide. 186 of ... 19
monoxide, dissociation of 35, 36 products, volumetric an
monoxide, temperature of alysis of. 134
combustion of. 19 reactions of .97 , 116
temperature of combustion in recuperative gas- fired
of . 17 , 21 furnaces, hypotheses re
Case, W. L .. 46 quired for theory of....67 , 68
Cement industry, points con specific heats of.. 205
cerning. 275 temperatures of, defined .. 14
Ceramic furnaces. 73 temperature of, how com
Checkerwork , calculation con puted ... 14
cerning 248 Commission du Grisou . .7 , 14
Chemistry, thermal, books on . 303 Compagnie Parisienne du Gaz.
Chimney (ofgas furnace ), office 87 , 92, 99
and limits of .. 41 Parisienne du Gaz , retort
Chimneys. 231 gas furnaces of .. 86
books on 303 Coquillon platinum spiral 186
required height of, by cal Couple, Le Chatelier thermo
culation . 231-233 electric..... 171
INDEX 309
PAGE PAGE
Couple, standardization of, Firing, direct, type of furnaces

method of accomplishing using 77
155 , 156 Fisher -Gautier . 46, 47
Couples , some types of ...... 149 Fletcher furnace .. 159
Currents, gaseous, minimum Flue connections, incorrect, for
times of sojourn of in a horizontal regenerative
regenerative furnaces ... 241 chamber.. 243
Cylindrical tube, section connections, incorrect, for
through, equivalent to a vertical regenerative
half of the perforated chamber.. 244
block . 255 design , correct , for a hor
izontal regenerative
D chamber .. 244
Dampers, importance of in Focusing device of the Féry
furnaces . 43 heat -radiation pyrome
150 ter, mirror type .. 172
Day ..
Dellwick . 45 Freezing point of metal, appar
Dellwick - Fleisher water -gas atus for determining ... 160
producer. 61 Fuel, calorific power of, ap
Dowson .. 46 proximate determina
Draught, improvement of in tion by Mahler and Gou
gas-producers . 46 tal formulas ..... 220
Drehschmidt tube . 186 , 188 character of to be
64 considered in making
Duff gas-producer.. calculations .. 143
Dulong formula for heating
effect of coal ... 192 combustion, air required
in (table ). 9
Dust -coal firing, adaptation of . 273 .1 , 2
economy
E elementary analysis of. 223
gaseous, use of in boilers
Ebelmen fusion producer... 43 not recommended .. 84
gas -producer 45
.94 , 110 powdered, application of
producer. to furnace firing .. 275
Ecole Nationale des Mines . 14
powdered, list of U. S.
Edison Portland Cement Co., patents covering the
furnace of ... 274
stoking of..... .283-286
Efficiency , recuperative, de powdered, preparation of . 272
fined . 56
Elliott apparatus for gas an and....calorimetry, books
Fuels on ..300-302
alysis .. . 178 , 179 apparatus for elementary
apparatus fitted with 224
analysis of....
Drehschmidt platinum Furnace, best , points concerning 52
tube for determination calculation ofth e effi
of hydrogen and me ciency of a ..69 , 136
thane.. 185
determination of the effi
E. m . f., necessity of vari ciency of a 56
ability of in couples .... 150 dimensions. 3
F
direct- fired , diagram show
ing percentage of heat
Ferraris zinc gas furnace ... 274 utilized and lost ... 75
Firebrick blocks, perforated , direct -fired , elements of
for the parallel counter the value of ..... 73
current system of heat with double recuperation ,
recuperation ... 258 heat balance -sheet of .. 93
310 INDEX .
PAGE PAGE
Furnace , electric -resistance, ad Furnaces, condensed classifica

vantages of.. 159 tion of various types.... 123
gas , description of parts ... 40 direct - fired , characteristics
gas - fired, some advantages of .... 71
of ... 39 direct- fired, heat losses for
gas, industries employed alltemperatures ....73, 74, 75
in ... 38 direct- fired , limit of tem
gas, reasons for its success . 39 peratures of .... 17
1
glass.. direct-fired, principal in
94 dustrial applications of .
glass, Göbbe system . 71
glass-pot, defects of. 39 with double recuperation
glass tank , fired with Sie for secondary air and
mens gas.. 92 gas .. 92
glass, typical example of efficiencies of when fired
supplementary • recuper with gas obtained from
ation . 115 producers injected with
Hoffman .. 1
part of the products of
horseshoe, disposition of combustion .. 119
parts in ... 49 electric ... 161
horseshoe, movement of electric, suited to pyro
flame in ... 50 metric work .. 161
how to determine the value gas, improvements in . 43
of a .. 129 followed by steam boilers. 76
illuminating -gas muffle, of gas, classification of .... 37
Compagnie Parisienne gas, description and func
du Gaz . .88, 89 tions of essential pa ts .. 40
new Siemens ... .48-51, 121 , 187 gas, origin and develop
parts, improvements in ... 39 ment of .. 37
proper ( in gas furnace ), gas -fired, different applica
limits of .. 41 tions of.. 3
puddling .. gas- fired , when economi
puddling, calculation of cally substituted . 128
lost heat in . 76 gas- fired reverberatory, ra
puddling, of the forges diation in ....... 50
d'Eurville 77 gas- fired, second group of . 97
puddling, with water gas- fired , twelve systems
cooled hearth .. 77 considered . 53
Queneau electric resistance glass, heat utilization in . 76
crucible .. 162 heat balance - sheets of .. 129
re -heating . 1 large , with long flames .. 3
rotary clinkering, for mak making use of regeneration
ing Portland cement of combustion products. 116
efficiency test of ..... 277-282 medium recuperative,prog
selection of proper type.. 1 ress to be sought here . 95
study of the heat efficiency without recuperation ... 73
of a . 129 medium temperature, best. 123
triple recuperation, special with one or two pairs of
interest of .. 122 chambers , difference of
Furnaces (see also under Gas ) . efficiency in ..... 123
books on .. 304 proper and improper meth
ceramic .. 73 ods of estimating value
classified 141 of.. 143
comparative table of all proper, recent modifica
systems. . 124 , 125 tions in 49
INDEX 311
PAGE PAGE
Furnaces that recuperate by Gas firing, with recupera

secondary air alone.... 86 tion . 2
small, with short flames .. 3 furnace, description of
for temperatures of about parts. 40
1000 °, classification of .. 128 furnace , industries em
with triple recuperation , ployed in .. 38
high efficiency of ..... 114 furnace, reasons for its
what type is superior theo success . 39
retically .. 144 furnace, recuperative il
luminating .. 239
G furnaces , classification of . 37
furnaces, description andl
Galvanometer of Cambridge functions of essentia
Scientific Instrument Co. 148 parts .... 40
Cambridge Scientific In furnaces, improvements in . 43
strument Co. direct furnaces, origin and devel
reading , for the Féry opment of .. 37
heat- radiation pyrometer 173 furnaces, retort, Lencau
d'Arsonval . 171 chez system . 86
operation of 167 furnaces, retort, Siemens
Pellin registering . 148, 166 system .. 86
where to place for tests .. 155 masses , reacting only here
of Siemens and Halske ... 148 considered .. 66
Siemens-Halske recording .. 167 mixed , advantages of in
Galvanometers, pyrometric ... 166 recuperation ... 46
types of..? 148 mixed , in case of recupera
Gas ( see also under FURNACES ). tion by total air . 110
analysis .. 133 , 175 mixed , defined ... 51
analysis , calculations in 189 mixed, higher efficiency of
-analyzing machine, auto due to better recu per
matic self -registering ... 191 ation .. 108
analysis, books on .. .302-303 mixed , how obtained . 46
and air ports, types of.. 268 mixed , for single recuper
and air flues leading to ation , slight advantage
regenerator, minimum of .. 101
section of. 250 mixed , tendency to use . . 102
calorific power of . 219 ( natural, Iola , Kan .), an
chambers in Siemens re alysis of and calculations
cuperative system ... 19 based on .. 28-33
currents, maximum allow natural, not capable of
able velocities of in efficient pre-heating.... 65
regenerative furnaces .. 240 ( natural, Pittsburg ), com
enriched, how obtained . 14 bustion temperatures of. 25
-fired furnaces, when eco (natural, Pittsburg ), data
nomically substituted . 128 concerning combustion
-fired furnaces, improve of .... ... 26 , 27 , 28
ments in ... 39 (natural, Pittsburg ), vol
-fired furnaces, recuper ume of oxygen required
ative organs in perfect, for combustion . 25
from one point of view . 69 pre -heating of 49
-fired furnaces (recupera -producer, cooling of,
tive ), selection of method omitted in hypothesis .. 141
for calculating efficiency -producer, equation of re
of .... 67 action in . 86
312 INDEX
PAGE
PAGE
Gas - producer, injection of hot Göbbe, E. 116
products of combustion furnace. 99
into .. 118 Goutal formula . .220 , 221 , 276
producer, office and delim Grate , problem of preserva
itation of . 40 tion of .... 127
-producer, office of in a Grates, horizontal and inclined . 43
furnace .. 40 Gruner, calculation by .
-producer, what it includes . 235
40 rule, amendment to . 248
-producer, Taylor. 42
-producers, how aided in H
operation by steam ... 64
-producers, steam and Harboard , F. W., cited .....
other auxiliaries in ... 60 247
Harvey . 46 , 95 , 116
samples, method of taking Hearth conditions here taken
in tests . 136 as the same in all cases ..
Siemens . 57
44 in furnaces . 261
Siemens, diagram show in gas furnace .
ing theoretical temper recent modifications in ..
41
49
ature of formation of .. 87 Heat balance -sheet of furnace,
specific heat of a, at any with double recupera
temperature, how de tion ...
termined .. 11 93
balance- sheet, necessary
varying temperature of . 98 factors of.. 130
Gases, analysis of various in balance -sheet of a Sie
dustrial.. .46, 47 mens furnace , with
auxiliary to fuel gasifica double
tion .. 64 fired with recuperation,
mixed gas,
burned or oxidized , action data required for ...
of ... 131
58 of combu stion and calorific
(or currents ), classification power .
of. in a complete gas fur 7
cowe
nace .. efficiency of a furnace ,

..41 , 43 study of...
influence of pre -heating .. 3 latent, how affected by
129
influence of the velocity of. 3 differences in gas-burn
thermal capacities of be ing.
tween certain points, 62
lost, definition of . 55
how expressed ... 11 lost in furnaces, knowledge
thermal capacities of, per
of essential point .
13
kilogram (table ). 70
thermal capacities of, per lost, recuperation of ..
lost and utilized in gas fur
kilogram in Centigrade naces fired with mixed
degrees (table ). . 12
gas , double recuperation .
thermal capacities of, per
molecular volume ( table) 13 105, 108
lost and utilized in fur
thermal capacities of, per nace fired with mixed
molecular volumes in
absolute temperatures gas, recuperation by sec
(table ) ... ondary air : 101
12 lost and utilized in gas
thermal capacities of, per furnaces fired with mixed
molecular volumes in
Centigrade degrees ( table) 12 gas , triple recuperation . 112
lost and utilized in fur
thermal capacities and spe nace fired with Siemens
cific heats of .... 10 gas, with double recu
used in firing furnaces . 47 peration. 93
INDEX 313
PAGE I
Heat lost and utilized in multi PAGE
tubular boilers ... 81 Injector, Koerting steam ... 46

lost in a puddling furnace, Inversion organs (of gas-pro
calculation of.. 76 ducer ).. 40
lost in steam boilers when
combustion is affected J
by large excess of air .. 78
recuperation, progress Junker calorimeter.... 193
made in .. 47
calorimeter, accuracy of .. 219
recuperation , theory of, in calorimeter, description of. 214
the light of experimental calorimeter, method of us
data .. 140 .216-2
ing..
specific, at any tempera gas calorimeter. 215
ture , how calculated ... 13 Juptner, von , formula . 220
total available, principles
relating to . 58
K
total , component parts
of.... 56
Koerting steam injector ... 46, 120
transmission , coefficients
of ... 256
Klönne furnace . .94 , 108, 111 , 126
utilization, consideration of
various phases . 2 L
utilization considered, an
E6 % E5 %
alyzed and defined . 55 Laboratory ( in gas furnace ).. 41

utilization , curve of, in any Le Chatelier. ..7 , 14 , 35 , 241
given furnace..... 55 thermo-couple, cost of. 151
utilization in furnaces, fun thermo - couple, endurance
damental principles of.. 58
of in tests . 151
utilization in glass furnaces 76
thermo -electric pyrometer.
utilized, definition of. 55 129 , 147 , 148
utilized , equation of. 56 Le Creusot, analysis of coal of. 222
utilized and lost in direct Liebig bulb . . 177 , 178
75 Ledebur . .57 , 96 , 271
fired furnaces..
utilized , how to calculate .. 138 Lencauchez, M. . 76, 90
Heating , industrial, statement brass furnace , fired with
mixed gas . 99
of problem . 1
problem , early difficulties producer 94
and confusion ..... 2 Leybold .. 47
Lurman . 47
Heats, specific, of fires under
constant pressure, how
used in calculations.... 11 M
Hempel apparatus for gas an
alysis .. 178 Mahler, P. 177
Hempel, Dr. 179 calorimeter . 129
Heraus, Dr.. 159 , 162 , 163 calorimeter (Atwater mod
Héroult, Paul. 242 ification ), cross - section
Holborn . 150 through .. 195
Holman , S. W 153 calorimeter modifications. 193
Hydrocarbons, heavy, occur formula .. .220, 221
rence of.. 175 Mallard .. 7
Hydrogen, combustion of .. 186 Mariotte flask . 177
temperature of combustion Martin -Siemens furnaces, class
of in air (diagram )... 16 ified .. 92
Hydrothermal equivalent. 205 McMillan , H. C., cited . 248
314 INDEX
PAGE PAGE
Melting point of metals, ap Parr standard calorimeter, di

paratus for determining. 160 rections for using. . 209-213
Melting point of precious met Peclet, experiments by . 234
als, wire method for · formula by ... 235
determining .. 161 Pennsylvania Steel Co .. 246
Metallurgy , books on . .304-305 Pinzger, calculation by. 235
Metals, apparatus for determin Platinuin , determination of
ing melting and freezing melting point of . 161
points of .. 160 Ponsard furnace . 37
precious, wire method for Ports, cross section of . 251
determining melting in furnaces . 266
points of .. 161 Potash solution . 180
Moisture, determination of .. 225 Producer blown by steam , re
Molecular volume, definition of. 7 actions in ... 61
weight of fuel burned , grateless, advocated by
computations in this Ebelmen . 96
work based on ... 16 -gas, calorific power of at
Morgan gas -producer .. 44 600 ° .. 136
gravity -discharge billet -gas, method of finding
heating furnace.. 100 temperature of combus
suspended -roof billet tion of ... 18
heating furnace . 102 -gas (Siemens ) , tempera
Muffle furnace, calculation of ture of combustion of
the dimensions to be when burned with pre
given to recuperative heated air .. 23
surfaces for .. 252-254
-gas, volumetric analysis
of .. 134
N -gases, heat regeneration
applicable to . 51
New Jersey Zinc Co .. 191 Producers without grates . 43
Niclausse, M. 83 Pyrogallol , alkaline solution of. 180
boiler . .78 , 80, 81 , 83 Pyrometric cane for protecting
multitubular boiler, front wires of Le Chatelier
elevation of.. 82 thermo- electric pyrome
multitubular boiler, sec ter ... 152
tion through .. 81 Pyrometer, Féry heat -radia
Nitrogen , determination of . 226 tion .. .170 , 171
heat -radiation . 171
0 Féry heat -radiation , lens
type .. 171
Oil, heavy, action of in gas Féry heat-radiation , mir
producer. 65 ror type .. 172
Orsat apparatus for gas an Féry heat-radiation , mir
alysis .. 178 ror type, focusing de
Oxygen , absorption of . 184 vice of .. 172
determination of . 227 Féryorn
optical 169
Holb and Kurlbaum .. 169
P Le Chatelier thermo- elec
tric .. 147
Morse . 169
Parallel counter - current sys
tem of recuperation . 238 optical, of Le Chatelier. . 168 , 169
Parr calorimeter. 193 principle of the thermo
calorimeter, properties electric. 150
. 207, 208 uses of. 153
of ...
INDEX 315
PAGE PAGE
Pyrometer, Wanner. 170 Recuperative chambers, useful

Pyrometers , registering, advan surfaces of, should be
tages of... 153 adapted to exchange of
registering, where to place. 153 calories.. 142
Pyrometry, books on .. 303-304 efficiency defined . 56
efficiency measure of rel
Q ative value of a fur
nace .. 57
Queneau, A. L. 46 organs in gas - fired fur
Queneau, A. L. , system of . 113 naces perfect from one
point of view . 69
R system , Siemens.. 48
Recuperators . .47 , 252
Radot furnace, Göbbe system . . 86 Redtenbacher, formulas by ... 234
Radot -Lencauchez furnace... 99 on
Regenerati , carbon , no gain
Record of the CO , contents of from . 119
the stack gases from a Regenerators. . 47 , 242
hand - fired boiler . 190 proper dimensions for .. 246
Recuperation, adaptability of. 2 Regnault.
Regnault -Pfoundler formula
11
double, by primary air and 203
secondary air .. 108 Richards, J. W. , cited ... 247
double, by secondary air Roofs in gas - fired reverbera
an d gas. 102 tory furnaces. 49
as delimited in this work .. 57
double, primary and sec S
ondary air . 94 Saillers producer .. 96
double, how best secured .. 51 Sample-tube, manipulation of. 181
fundamental principles of. 65 Sampling in gas analysis . 176
meaning of term in this Schoenwalders furnace.. 269
book .. 53 Schnabel -Gautier.. 47
methods of . 49 Siemens, Frederick... 37 , 50, 110,
organs ( in gas furnace ) . 40 120 , 264
by secondary air ... 99 furnace . .94 , 123, 124
by secondary air alone . 86 furnace, new 110 , 116 , 120
Siemens'invention of. 2 furnace, new , circulation of
single adaptations of . 115 gas currents in ... 59
theory of . .3 , 7 furnace, new , fallacies in
by total air, advantage of. 111 regard to . 59
total , when impossible ... 66 furnace, new , with horse
triple, adapted to high shoe hearth 48-51
temperatures. 114 furnace, original, briefly
triple, alone may give com described . 37
plete recovery of calories 67 furnace, original, how im
triple, by primary air, sec proved .. 52
ondary air and gas . 110 furnaces classified . 92
triple, by primary air, sec gas . 123
ondary air and gas gas defined . 51
(Queneau system )... 113 gas, diagram showing the
triple, worth serious consid oretical temperature of
eration ... 110 formation of 87
without inversion . 47 gas under given conditions,
Recuperative chambers, neglect diagram showing tem
of the cooling of: hy perature of combustion
potheses . 142 of .. 92
316 INDEX
PAGE PAGE
Siemens gas, properties of..... 11 Temperature, high, and recu

glass tank furnace ... ..99, peration connection be
103-105 tween 2
producer ... 94 measuring for tests . 134 , 135
recuperative system . 48 ruling, defined . 54
recuperator with inversion . 47 ruling, and efficiency of
siphon .. 16 furnace , relation between 76
siphon, drawbacks of . 94 ruling, influence of .... 54
suggestions by. 269 influence of on the heat
system , extension of to utilization in a fur
boiler firing unsuccessful 76 nace .. 53
system of recuperation .... 238 ruling, in variableness of.. 141
William . 37 ruling, lack of in
Siphon , Siemens. 46 boilers ... 81
Société des Métaux . 99 Temperatures of about 1000 °,
Spectrophotometer, König's .. 170 classification of furnaces
Stafford , C. E..... 269 for .... 128
Standardization , method of of combustion . 14
Holborn and Day ... 150 of combustion, general
Steam as an aid in operation of equation of 14
WN
gas -producers.. 64 of furnaces, high . 2
an auxiliary of recupera high , production of. 3
tion .. 63 low ( as in steam boilers ),
effect of use in furnaces .. 63 have little to expect
for single recuperation, from gas recuperation .. 128
slight advantage of .... 102 usual, given by furnaces
Sulphur, determination of . 225 with single recuperation ,
Swindell gas -producer, with fired by Siemens gas .. 90
double face .. 62 Thermal capacity defined . 10
Thermo -chemical reactions , ta
T ble of .. 8
-couple, apparatus for the
Table 12 (Goutal factors ) . 222 protection of in roasting
13 .. 287 kiln .... 154
11 288 -couple, to replace injured
15 , 16 , 289 one by new one . 153
17 , 18 , 290 -couples, care of .. 153
19 291 -couples, diagram showing
20 . 292 e . m . f. curves of dif
21 293-295 ferent . 148
22 296 297 -couples, some types of . 149
23 298 Thermometer for experiments.
24 . 299 200 , 201
Tar, action of in gas-producer.. 65 Thompson calorimeter .. 193
Taylor gas -producer, water Three -way cock , F , various
sealed , with Bild auto positions of . 181
matic feed ... 60 cock , G , various positions
gas- producer, with me of .. 182
chanical eed ... 12 cock , H , various positions
43 of . 183
grateless producer. Toldt, F. , cited .. 48 , 240 , 243 ,
producer.. 94
Temperature of combustion of 244 , 262, 265, 267 , 268
Siemens gas under given Tubes, sample , used in gas an
conditions, diagram of. 92 alysis. 176
INDEX . 317
V PAGE
PAGE
Water vapor, determination ,
Value, low thermal... 220 how effected.. 177
Valve, Forter water-sealed re vapor, dissociation of . 35
versing ... 263 vapor in gases, apparatus
reversing.. 262 for determination of... 177
Wailes water- sealed . 261 Weights, atomic . 7
reversing. 261 molecular. 7
Vielle .... 7 Wellman tilting furnace. 269, 270
Vigreux, C ... 46 Wire method .. 161
Vivien formula .. 220 Wires, highly heated , life of
Volume, molecular, definition during furnace tests.... 150
of ... 7
Z
W Zinc , apparatus for determining
the boiling point of.... 158
Water gas, manufacture of.... 45 furnace , parallel current re
gas, production of . 97 cuperative... 240
gas, theoretical, tempera metallurgy, conditions pe
ture of combustion of.. 20 culiar to .. 245
LIBRARY
OF THE
UNIVERSITY
OF
CALIFORNIA
17
D2
Mining
dept,
in
3
11. ҮС 33830
Соися .
ерлері
154166
2oh

Industrial Furnaces and Methods of Contr

Uploaded by

Copyright:

Available Formats

Industrial Furnaces and Methods of Contr

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Industrial Furnaces and Methods of Contr

Uploaded by

Copyright:

Available Formats

This is a reproduction of a library book that was digitized

by Google as part of an ongoing effort to preserve the

Authorized Translation with Additions

A. L. J. QUENEAU , B.S. ( Paris) , E.M. , A.M. (Columbia)

New York and London

THE ENGINEERING AND MINING JOURNAL

STATIONERS ' HALL COURT, LONDON , ENGLAND

Fig. 1. Graphic Method for the Determination of the Temper

Fig. 24. Puddling Furnace of the Forges d'Eurville.

Fig. 45. Apparatus for the Protection of the Thermo- Couple in

Fig. 61. Apparatus for the Determination of the Water Vapor

Fig. 70. Berthelot-Mahler Apparatus Ready for Operation, in the

1. Table of Thermo -Chemical Reactions .

Heat of combustion and calorific power–Tables 1 , 2

Origin and rapid development of gas furnaces - Descrip

Influence of the temperature on the heat utilization in a

First group of gas - fired furnaces — Siemens gas-recuper

Second group of gas-fired furnaces — Reactions of com

Furnaces making use of the so -called regeneration of the

Classification of the various types of furnaces — Table 11. . 122-128

Experimental study of the heat utilization in furnaces

Discussion of the theory of heat recuperation in the light

Pyrometric practice and instruments — Forms of couples 1

-Standardization - Crucible methodElectric fur

Gas analysis — Sampling — Sampling tubes—Water-vapor

Calorimetry — Types of calorimeters—Method of using

Elementary fuel analysis - Determination of elements .... 223—227

Regenerators - Velocities and sojourn of gas currents in

Recuperators -- Details... 252-261

Reversing Valves - Hearth - Ports ... 262-271

Powdered fuel - Coal-dust stokers - Details ... .2724282

List of United States patents covering the stoking of

Fuels and calorimetry - Combustion - Gas analysis — Ther

The knowledge of the lack of any book in English covering

Analysis, which have only been touched upon in the work of

puzzling English, rouses our applause, Mr. Queneau gives us easier

Columbia University in the

The problem of industrial heating is very complex and often

designs, all ingenious, the advantages of which are often more

2. Production of high temperatures; influence of pre-heating

Authorized Translation and Additions

THEORETICAL STUDY OF RECUPERATION .

SCIENTIFIC Data NECESSARY FOR THE STUDY OF COMBUSTION

work , is what we will call the Molecular Volume.

2.33 4.46 6.79

1.57 1.91 3.48

3.67 8.93 12.60

2.00 3.33 5.33

When the body under combustion is a fuel, coal, coke, etc. , at

sorbed by the gas in this operation is called the Thermal Capacity

stants are the basis for all calculations of exchange, in calories, in

Capacity. It is not possible, inversely, to derive from the specific

it is evidently impossible to go from the derivative to the primitive

Perfect Gases Water Vapor Carbon Dioxide Methane

It follows that the total Thermal Capacities of a Molecular Vol

to = T. — 273° and t T- 273 °

will be expressed by the formulas given in Table 3.

Gas Under Constant Pressure Under Constant Volume

T - TO0 T2 – T.2 T - T T2 - T.2

Water Vapor, H 20 . 6.5