-

Co
py
rig
hte
dM
ate
ria
l

GENERAL

CONSIDERATIONS

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

CHAPTER 8.1
AUTOMATIC TEMPERATURE,
PRESSURE, FLOW CONTROL

Co
py
rig
hte
dM
ate
ria
l

SYSTEMS

Edward B. Gut, RE.

"Donald H. Spethman
Honeywell, Inc., Arlington Heights, Illinois

8.1.1

CONTROLBASICS

8.1.1.1 Control Systems

Elements of Control Systems. Control loops consist of several elements and are
used to match equipment capacity to load by changing system variables. Figure
8.1.1 is a block diagram of a control loop and shows the relation of the elements.
The controlled variable is the condition being controlled; for HVAC systems this
is typically temperature, humidity, or pressure. A sensor is the device that measures
a variable and transmits its value to the controller. The controller compares the
value of the variable with the set point or desired value, and outputs a signal based
on the difference between the variable and the set point.
The final control element responds to the controller signal and varies the manipulated variable. Control elements may be valves, dampers, electric relays, or
electronic motor speed controllers, and manipulated variables may be air, water,
steam, or electricity. The process plant is the equipment being controlled and whose
output is the controlled variable. It may be a coil, fan, steam generator, or heat
exchanger.
Types of Control Loops. There are two basic types of control loops, open loop
and closed loop. With open-loop control, the system sensor measures a variable
external to the system yet has some relation to the controlled variable. An example
is sensing outdoor temperature to control heat flow into a building to maintain
indoor temperature. Thus, a fixed relationship between outdoor temperature and
required heat input is assumed and the control system programmed accordingly.
*The parts of this chapter covering boilers, refrigeration, central plants and building management systems
were written by Donald H. Spethman for the first edition and were updated by Edward B. Gut for this
edition.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

DISTURBANCE

CONTROLLED
ENVIRONMENT
MANIPULATED
CONTROLLED
VARIABLE
VARIABLES

Co
py
rig
hte
dM
ate
ria
l

CONTROLLER

SET
POINT

ERROR

CONTROLLER
OUTPUT

FINAL CONTROL
ELEMENT

SENSOR
(FOR CLOSED-LOOP CONTROL ONLY)
FIGURE 8.1.1 Basic elements of a control loop.

Closed-loop control pertains when the system sensor measures the controlled
variable, resulting in variations in the manipulated variable to maintain the desired
value of the controlled variable. Closed-loop control is also called "feedback control," and results of a corrective action are fed back within the controlled system,
therefore providing true control of the controlled variable.
8.1.1.2 Modes of Feedback Control

Feedback-controlled systems are categorized by the type of corrective action a controller is designed to output. For all types, the set point is the desired value of the
controlled variable to which the controller is set. The control point is the actual
value of the controlled variable as maintained by the controller's action.

Two-Position Control. The final control element may be in one or the other position, i.e., maximum or minimum, except for the brief time when it changes positions. There are two values of the controlled variable which establish the position
of the controlled element: set point and differential. Differential is the smallest range
through which the controlled variable must pass to move the control element from
one position to the other. Figure 8.1.2 shows a temperature controller or thermostat
with a 7O0F (21.10C) set point. At 7O0F (21.10C) this electric thermostat would open
its contacts and stop a burner. For the thermostat contacts to close, turning on the

DIFFERENTIAL

FIGURE 8.1.2 Two-position control.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

burner, the temperature must drop below the set point by the amount of the differential, 20F (1.10C) in this example. Differential may be subtracted from or added
to set point depending on controller design.
Two-position control is a low-cost device and provides acceptable control of
slow-reacting systems that have minimum-lag between controller outputs and control-element response. Fast-reacting systems may overshoot excessively and be unstable. Examples of two-position control are domestic hot-water heaters, residential
space-temperature controls, and HVAC system electric preheat elements.

Timed Two Position Control. The final control element may be in one of two
positions, as for a two-position control, but a timer is incorporated in the controller
so that it responds to the average value of the controlled variable rather than the
peak fluctuations. Timed two-position control greatly reduces the variations or
swings in the control variable by anticipating controlled-variable changes due to
control-system action.
A typical example of timed two-position control is residential space-heating
temperature control. The thermostat has an electric heating element that is energized
during the on period, the heat from the element warms the temperature sensor more
quickly than the rising space temperature, shortening the on time and reducing
temperature overshoot. During the off period the sensor heater is also off, allowing
the sensor to respond directly to space temperature. This results in a relatively
constant cycle time with a variable on-off ratio dependent on space load.
Timed two-position control is low-cost and may be applied to slow-reacting
systems that have some lag between controller output and control-element response.
The timer will anticipate the response and minimize variations in the controlled
variable.

Proportional Control. A proportional controller has a linear relationship between

the value of the incoming sensor signal and the controller's output. The relationship
is generally adjustable in the controller but once adjusted remains fixed during
operation. There is therefore only one value of the final control element for each
value of the controlled variable within the operating range of a proportional control
system.
The variation in control variable required to move the final control element
through its operating range is the throttling range of the control system and is
expressed in the measuring units of the controlled variable. The variable in the
sensor signal required to operate a proportional controller through its range is called
the "proportional band" and is expressed as a percentage of sensor span.
The "set point" of a proportional controller is defined as the sensor input which
results in the controller output at the midpoint of its range.
"Offset" is the difference between the set point and the controlled variable at
any instant. Sometimes offset is also referred to as "deviation," "droop," or "drift."
Offset results from the fixed linear relationship between control input, sensor signal,
and output. Therefore, under full-load conditions, control input must be offset by
one-half the proportional band for the controller to output a signal at one extreme
of its range. Similarly, at minimum load the offset will be one-half the proportional
band (see Figure 8.1.3).
Proportional control is used with slow stable systems, allowing narrow throttling
ranges and therefore small offset. Fast-reacting systems require large throttling
ranges to avoid instability and cycling of the controlled variable. This, of course,
increases offset.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

OPEN

Co
py
rig
hte
dM
ate
ria
l

FINAL CONTROL ELEMENTS

MAX.
OFFSET
BELOW
SETPOINT OFFSET
ABOVE
SETPOINT

CLOSED

SET
HIGH
POINT
CONTROL
RANGE
FIGURE 8.1.3 Proportional-controller offset.
LOW

Proportional-Plus-Reset Control. A proportional-plus-reset controller has proportional action plus an automatic means of resetting the set point to eliminate offset.
This controller action is also called "proportional-plus-integral," or PI control.
A PI controller's initial output signal has a fixed relationship to a changed sensor
input signal, the same as a proportional controller, but then continues to change
until the control variable equals set point. The rate at which this additional change
occurs is called the "reset rate" or "repeats per minute" and is the number of times
the original proportional change in controller output is repeated per minute. The
reset rate may also be expressed as reset or integral time, which is the amount of
time for the controller to change its output as much as the first proportional change.
PI control may be applied in fast-acting systems that require large proportional
bands for stability but where the resultant offset between set point and control point
is undesirable due to comfort and/or energy-conservation considerations. Typical
applications are mixed-air, duct-static, chiller-discharge, and coil-discharge control.

Proportional-plus-Rate Control. Also called "proportional-plus-derivative," or

PD, control, this control mode adds to proportional control an automatic means of
varying controller output based on changes in deviation or difference between the
set point and the control variable.
When deviation increases, rate action adds to the controller's output, causing the
final control element to respond an additional amount to stabilize the controlled
variable more quickly than proportional control alone can. Conversely, when deviation decreases, rate action subtracts from the controller's output. When there is
no change in deviation, rate action stops and the deviation is determined only by
the proportional band of the controller.

Proportional-plus-Integral-plus-Derivative (PID) Control. This combination of

control modes is useful for controlling fast-acting systems that tend to be unstable,
such as duct static-pressure control. For these applications, the controller may be
set with a large proportional band for system stability, a slow reset to eliminate
deviation from set point yet retain stability, and derivative action to speed control
response when the system is upset due to changes.
Adaptive Control. A PID controller must be properly tuned to the system it is
controlling to achieve stable and accurate control. The proportional, integral, and
derivative parameters of the controllers are dependent on the characteristics of the
system being controlled and can be time-consuming to establish. Even then they

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

will be optimum for one operating condition and compromised for the remaining
operating range.
Adaptive control is the ability of a controller to adapt to the system it is controlling by determining the ideal PID parameters and adjusting itself according.
Two types of adaptive control have been developed, self-tuning and model reference.
A self-tuning controller begins with initial PID parameters. With input from the
controller's output and the control-variable value, it establishes new parameters.
After a few cycles of control-system operations, the controller determines the optimum parameters. The process continues as the system operates, so every time the
system changes, the controller reestablishes the parameters so that they are optimum
for every condition of system operation.
The model-reference controller compares its output with that of a fixed model
and develops the PID parameters to achieve control-system operation for the model.
While the model may not be exactly the same as the actual system, it is very close
and allows the controller to develop the parameter values quickly.

Floating Control. A floating control outputs a corrective signal when the difference between the set point and sensor signal is greater than a set amount or differential. The output signal will increase or decrease a final control element depending on if the controlled variable is below or above set point. If the difference
is less than the differential, the controller output is zero and the final control element
remains in the position it was last driven to. Floating controls may be applied to
systems that react quickly with little lag and have slow load changes.
Time-Proportioning Control. Time proportioning is a method of controlling loads
like electric heating elements. The final control element is either on or off, but the
ratio of on-to-off time is varied depending on system load, therefore varying the
energy inputs.
The sum of on and off time, or the total time per cycle, is constant. Timeproportioning control is also called ''average-position control" and is a relatively
low-cost way to simulate proportional control.
8.1.1.3 Flow-Control Characteristics

Flow Control. Proper volume or flow control of one form or another is essential
to the successful operation of most HVAC systems. Usually the flow of water,
steam, and/or air is controlled to modulate system outputs or capacity as required
by changing loads. As in other control loops, a sensor measures the control variable
and a controller compares the sensor signal to a set point and outputs a corrective
signal as required to a final control element. For water and steam flow the final
control element is a valve, and for air flow a damper.
The flow-control characteristics of valves and dampers are designated in terms
of the flow versus opening based on a constant pressure drop across the element.
The three common characteristics are quick opening, linear, and equal percentage.
As shown in Fig. 8.1.4, quick opening provides for more percentage of full flow
than when the valve or damper is opened. Linear characterization has the same
percentage of full flow as when the valve is open, while equal percentage increases
flow by an equal percentage over the previous value for each equal increment of
opening. In other words, a 10 percent change in opening from 20 to 30 percent
increases the flow by the same percentage of flow at 20 percent opening as an

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

% FLOW

QUICK
OPENING
LINEAR

Co
py
rig
hte
dM
ate
ria
l

EQUAL PERCENTAGE

% OPEN
FIGURE 8.1.4 Flow-control characteristics.

increase in opening from 70 to 80 percent would increase flow from the 70 percent
opening position. These different characteristics are required to match the control
needs of water, steam, and air flow.
Pressure drop across a valve or damper in a system rarely stays constant. Therefore actual opening-flow characteristics vary from manufacturer's ratings, which are
based on constant pressure drop. The amount of this variation depends on how
much the pressure drop changes and is determined by overall system design. The
pressure drop is minimum when the valve or damper is full open and increases as
the valve or damper closes. When fully closed, the entire pressure drop is across
the valve or damper.
For the valve or damper to provide approximately its design characteristic, the
design or full-open pressure drop should be a fairly large percentage of the total
system drop. As a high pressure drop consumes energy, consideration should be
given to design or control a system to provide a more constant pressure drop,
allowing the valve or damper to be sized for a lower pressure drop at full flow.

% CAPACITY

Control of Water Flow. One of the primary uses of water-flow control is to modulate the capacity of a heating or cooling coil. However, the capacity of a coil is
not linear with water flow; instead, as the flow is reduced, more energy is transferred
from the water, partly offsetting the reduction in flow. Figure 8.1.5 shows the relationship of capacity versus flow for a heating or cooling coil. This nonlinearity
is primarily a consideration with hot-water coils due to the large temperature difference between the water and air flow through a coil. For hot-water coils, this
nonlinear variation may be reduced by designing the coil for a higher watertemperature drop or by reducing water temperature as system load decreases.
Since hot-water coils have a significantly nonlinear relationship between heat
transfer and water flow, equal-percentage valves are used for coil water-flow control,
resulting in a more linear relationship between valve position and coil heat output.

% FLOW THROUGH COIL

FIGURE 8.1.5 Heating- or cooling-coil
capacity versus flow.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

The capacity of a water valve is a function of valve design and the pressure
drop across the valve and is independent of the supply pressure. Valve capacity is
rated by a flow coefficient, or Cy, which is defined as the amount of water in gal/
min (m 3 /h) that will flow through an open valve at 1 Ib/in 2 (1 atm, 101.325 kPa)
pressure drop. For valves in systems, the pressure drop increases as the valve closes,
offsetting part of the desired flow reduction. To minimize this, valves should be
sized so that they constitute about 25 to 50 percent of the system resistance that a
valve controls.
Valve pressure-drop changes can be minimized by providing a system bypass
valve to maintain total system flow even when control valves close. Also system
flow may be modulated by an automatic flow-control valve in series with the pump,
or the pump may be operated at varying speeds based on system pressures near the
far end of the piping circuit.
Control of Steam Flow. Control of steam flow is usually applied to modulate the
heat output of a steam-to-water and steam-to-air heat exchanger.
For one-pipe steam systems, line-size two-position valves are used to ensure
proper flow of steam and simultaneous drainage of condensate. Two-pipe steam
systems may be controlled by two-position or modulating valves which must be
sized properly for good control. Since output of a steam heat exchanger is linear
with steam flow, valves with linear flow-opening characteristics should be used for
modulating control.
The capacity of a steam valve is determined by valve design, the pressure drop
across it and the inlet pressure. Valves for two-position applications are sized to
provide the required full flow with minimum pressure drop and to be able to close
against system pressure.
Modulating steam valves must be sized to only full-load flows, which may be
less than full heat-exchanger flow, to avoid system instability due to excessive
capacity. Since steam valve capacity depends on pressure drop and inlet pressure,
it is important that valve inlet and outlet pressures are kept fairly constant to maintain a linear relationship between valve opening and heat-exchanger output.
Supply pressures can be controlled by automatic pressure-reducing valves in the
supply lines or by a narrow differential controller. The effect of variations in return
pressures can be minimized by sizing the valve so that the outlet pressure is near
its minimum value or at a pressure resulting in critical velocity in the fully open
valve, whichever is higher.
Critical velocity in a valve is the velocity at which an increase in pressure drop
will not result in an increase in velocity or flow through the valve. This occurs
when outlet pressure is about 58 percent of inlet pressure. For some applications
with large-capacity modulating, two steam valves in parallel may be used for better
full-range control. The valves should be sized so that one valve has about one-third
full-load capacity and the other valve about two-thirds full-load capacity. The valves
are operated in sequence so that the smaller valve controls during low loads and
the larger valves operates when the smaller valve is fully open.
Control of Air Flow. Air flow in HVAC systems is controlled in an on-off mode
or modulating mode. The on-off mode is generally used to allow outside air into a
building when desired such as during occupied times and to prevent outside air
from entering at other times. Modulating air flow is used to blend air from more
than one source to achieve a desired temperature or to vary the volume of air
delivered to match load requirements.
Dampers are used to control air flow and are produced in two basic designs,
parallel-blade and opposing-blade configurations (see Fig. 8.1.6). The opening-flow
Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l
PARALLEL BLADES
OPPOSED BLADES
FIGURE 8.1.6 Damper-blade configurations.

% FLOW

characteristics of these configurations for constant pressure drop and for various
ratios of system pressure drop without the damper to damper pressure drop at full
open flow is shown in Figs. 8.1.7 and 8.1.8.
However, as with valves, dampers installed in systems have varying pressure
drops as they modulate, being minimum when full open and maximum when closed.
For two-position applications, dampers should be selected on the basis of full-flow
pressure drop, leakage, and closed-pressure differential ability. Modulating characteristics are not important.

CONSTANT AP

% FLOW

BLADE POSITION
FIGURE 8.1.7 Characteristics of
parallel-blade dampers. Curves other
than the constant-AP curve represent
ratios of system pressure drop to opendamper pressure drop at full flow.

CONSTANT AP

BLADE POSITION
FIGURE 8.1.8 Characteristics of opposed-blade dampers.
Curves other than
the constant-A/3 curve represent ratios of
system pressure drop to open-damper
pressure drop at full flow.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

Typical modulating applications are mixed air, face and bypass, and volume
control. Mixed air, or control of outside, return, and exhaust air, requires the coordination of three dampers for modulating outside and return air to maintain a
constant supply volume and for modulating exhaust-air volume as outside air varies.
Face and bypass control is used to vary the amount of air through and around a
coil to vary the temperature of the total air flow after the coil. The face damper
controls air flow through the coil, and the bypass damper the air flow around the
coil. The dampers are arranged so that when one opens the other closes, and the
sum of the air flow through both dampers is constant. To achieve this relationship
it is important that both dampers are selected for linear control.
Volume control of air flow may be used to maintain static pressure in a duct or
space or to match space- or zone-conditioning needs. Variable air flow is achieved
by changing duct system resistance to air flow or by diverting air flow through an
alternative or bypass route. Dampers should be selected to provide equal changes
in air flow for equal changes in control variables, which may be temperature, pressure, or flow volume in these specified systems, for stable control over the full
operating range.

8.1.2

CONTROLEQUIPMENTTYPES

The elements of a control loop are divided into four categories: sensors, controllers,
final control elements, and auxiliary equipment, and may be pneumatic, electric, or
electronic.

8.1.2.1 Sensors

The controlled variable of a system is measured by a sensor. A sensor output signal,

whether pneumatic or electric, may change electrical resistance depending on the
value of the sensed variable. The usual pneumatic sensor-signal range is 3 to 15
lb/in2 (20.7 to 103.4 kPa), while electric sensors output 2 to 10 V dc or 4 to 20
mA. Resistance sensors have a nominal resistance of 500, 1000 and 2000 H. Temperature-sensing elements are usually bimetal, rod and tube, sealed bellows, and
resistance.
Bimetal is the oldest and most common type of temperature-sensing element.
Its operation is based on the principal that the change in size with the change in
temperature is different for different metals. Combining two metals, one with a
large expansion coefficient and one with a small coefficient, into a strip, the strip
will deflect with temperature changes due to the different amounts of expansion
(see Fig. 8.1.9). The amount of deflection is proportional to temperature and can
therefore be used to measure or sense temperature and generate a proportional
pneumatic or electric signal. Bimetal strips may be used as straight elements or
may be U-shaped or spiral-wound depending on the space available and the temperature-deflection characteristics desired.
Rod-and-tube elements also use the different expansion rates of metals to generate movement with temperature changes. However, they are constructed with a
low-expansion rod, high-expansion tube (see Fig. 8.1.10) and are usually used for
insertion directly into the medium, such as water, steam, or air.
Two versions of the remote bulb element have a long, (10 to 20 ft or 3 to 7 m),
capillary tube in place of the bulb. One is liquid filled and senses the average
Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

HIGH-EXPANSION METAL
LOW-EXPANSION METAL

Co
py
rig
hte
dM
ate
ria
l

HEAT

HIGH-EXPANSION METAL;
LONGER SIDE
LOW-EXPANSION METAL;
SHORTER SIDE

FIGURE 8.1.9 Bimetal strip.

HIGH-EXPANSION METAL TUBE

LOW-EXPANSION
METAL ROD

CHANGES WITH
TEMPERATURE CHANGES
FIGURE 8.1.10 Rod-and-tube element.

temperature over its length. The other is vapor filled and senses the coldest (12 in
or 30 cm) along its length.
Sealed bellows (see Fig. 8.1.11) consist of a capsule and bellow evacuated of
air and filled with a vapor or liquid. As a vapor or liquid changes pressure or
volume with temperature changes, the bellows moves, providing an indication of
sensed temperature. A variation of sealed bellows is the remote-bulb element (see
Fig. 8.1.12). A bulb is attached to the bellows assembly by a capillary tube so that
MOVEMENT WITH
TEMPERATURE
CHANGE

BELLOWS

CAPSULE
VAPOR
LIQUID

FIGURE 8.1.11 Sealed bellows.

MOVEMENT WITH
TEMPERATURE CHANGES
AT BULB

VAPOR

BLUB

FIGURE 8.1.12 Remote-bulb element.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

temperature changes at the bulb result in pressure changes which are transmitted
to the bellows, resulting in movement corresponding to temperature at the bulb.
Resistance elements consist of an element with a known temperature life vs.
resistance characteristic. The element may be wire, wound on the bobbin, or a
thermister, which is a semiconductor, or a stable metal line platinum, plated on a
ceramic base, whose resistance depends on temperature.
Humidity-sensing elements are hydroscopic or electric. The hydroscopic elements are based on the fact that certain materials change size as they absorb or
release moisture. Typical materials are hair, wood, leather or nylon, whose size
changes due to moisture absorption or release based on the moisture content of
surrounding air, will indicate humidity of the air. This size change is used to develop
a pneumatic signal proportional to humidity or to turn an electric switch on and
off.
Electric humidity-sensing elements are constructed to provide either a resistance
change with ambient humidity changes or a capacitor change and are generally
used with electronic controllers. They generally respond quicker than hydroscopic
elements.
Dew-point sensors are constructed by winding two wires around a hollow tube
impregnated with lithium chloride. The conductivity of the lithium chloride varies
as it absorbs or releases moisture to the surrounding air. Electric power supplied to
the two wires around the sleeve will flow through the lithium chloride at a rate
depending on its conductivity, which varies with dew points. As the electricity flows
through the wires, the temperature of the cavity of the tube is elevated and is a
measure of dew point (see Fig. 8.1.13). The cavity temperature may be sensed with
any temperature sensor that will fit inside the tube.
Pressure sensors may be high-range (psi or Pa) or low-range (in or cm of water).
High-range sensor elements usually are Bourdon tubes, bellows, or diaphragms to
provide movement based on pressure. Low-range pressure sensors generally use
large slack diaphragms or flexible metal bellows to transduce low pressures into
usable forces for indicating pressure. If one side of the element is open to the
atmosphere, the element responds to sensed pressure above or below atmospheric.
For differential pressure sensing, both sides of an element are connected to sense
pressure variables. Outputs of pressure sensors may be pneumatic, electric analog,
or electric on-off.
Pneumatic air-velocity sensors are of the differential-pressure or of the deflectedjet type. The differential-pressure types use a restriction in the air stream, such as
an orifice plate, or sense static and total pressure to generate differential pressures
that represent air velocity (see Fig. 8.1.14). The deflected-jet type has a small air
jet flowing across the measured air stream from an emitter tube. The air is captured
in a collector tube and generates a recovery pressure (see Fig. 8.1.15). When the
velocity of the measured air stream is low, most of the air jet depinges on the
collector tube and the recovery pressure is high. As the air-stream velocity in-

CONSTANT
ELECTRIC
SUPPLY
TO
CONTROLLER

TUBE IMPREGNATED
WITH LITHIUM CHLORIDE

THERMAL SENSOR
FIGURE 8.1.13 Dew-point sensor.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

ORIFICE PLATE
AIR FLOW

TOTAL
PRESSURE

AIR FLOW

Co
py
rig
hte
dM
ate
ria
l

STATIC
PRESSURE

DIFFERENTIAL
PRESSURE SENSOR

DIFFERENTIAL
PRESSURE

DIFFERENTIAL
PRESSURE SENSOR

HIGH
LOW
AIR VELOCITY
FIGURE 8.1.14 Differential-pressure air-velocity sensors.

EMITTER TUBE

AIR FLOW

COLLECTOR TUBE

RECOVERY
PRESSURE

-RECOVERY PRESSURE

HIGH

LOW

HIGH
AIR VELOCITY
FIGURE 8.1.15 Deflected-jet air-velocity
sensor.

creases, the air jet is deflected and recovery pressure diminishes. The recovery
pressure is, therefore, a direct indication of air-stream velocity.
Electric air-velocity sensors use a heated wire or thermistor placed in the air
stream. The amount of current required to maintain the wire or thermistor temperature varies with the cooling effect of differing air velocities and, therefore, is a
measure of air velocity. A reference wire or thermistor shielded from the air stream
compensates for varying air temperatures. The sensor may be solid state with all
sensing elements on a chip.
Water-flow sensors may be differential-pressure types, such as orifice plates, pitot
tubes, or flow nozzles, that have limited range or vortex-shedding, turbine, or magnetic types that have greater range but are more expensive.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Other sensing elements including smoke and high-temperature detectors, specific-gravity, current, CO, and CO2 sensors are often used for complete control of
HVAC systems.

Co
py
rig
hte
dM
ate
ria
l

8.1.2.2 Controllers

Controllers provide the set-point, and for some, the proportional-band, integral, and
derivative parameters of a control loop. They compare the sensor signal with the
set point and output a corrective signal as determined by the controlled settings.
This signal may be direct-acting, increasing with sensor-signal increases, or reverseacting, decreasing with sensor-signal increasing. Controllers may incorporate a
sensing element for sensing and controlling in one device. Proportional controllers
may also be designed to use remote sensors and are called sensor-controller systems.
Controllers may be pneumatic or electric powered. Pneumatic controllers receive
a sensor signal and output a proportional signal typically 3 to 13 lb/in2 (20.6 and
270 kPa). The controller may be a nonrelay or relay type. Nonrelay types use a
restricted supply air, bleeding varying amounts to the atmosphere to generate a
corrective output signal (see Fig. 8.1.16). Since the capacity of the output signal is
restricted, amplification should be limited to small volume-control elements or
where long response times are acceptable. Relay-type controllers incorporate a capacity amplifier for the corrective signal for greater output volume.
Electric controllers also may have integral or remote sensors. Outputs are twoposition to cycle equipment, floating to open, hold, or close a final-control element
or proportioning to position a final-control element. Proportioning electric controllers may be analog or digital.
Electric analog controllers are similar to pneumatic controllers. That is, their
response to a sensor signal is fixed by their design and only by their parameters,
such as set-point, direct- or reverse-action, proportional-band, and if included integral- and derivative-timing, are adjustable. Digital controllers are microprocessorbased, and their response to a sensor signal is programmable. This provides great
flexibility for the application of a digital controller and allows control strategy
changes after installation.
Digital controllers measure signals from sensors, perform control routines in
software programs, and take corrective action in the form of output signals to
actuators. Since the programs are in digital form, the controllers perform what is
known as direct digital control (DDC). Microprocessor-based controllers can be
used as stand-alone controllers or they can be incorporated in a building management system utilizing a minicomputer or a personal computer (PC) as a host to
provide additional functions. A stand alone controller can take several forms. The
simplest generally controls only one control loop while larger versions can control

MOVED BY SENSING ELEMENT

RESTRICTION
AIR
SUPPLY
FIGURE 8.1.16 Nonrelay pneumatic controller.

Copyright 1997 by The McGraw-Hill Companies

OUTPUT
SIGNAL

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

from eight or ten to 30 or 40 loops. As the systems get larger, they generally
incorporate more programming features and functions.
Pneumatic and electric controllers may also provide indication and/or recording
of the value of the sensed variable for visual checks or for a history of system
operation. Transducers may be used with controllers to convert sensor signals and
controller outputs from pneumatic to electric, or vice versa, as required by the
controller or final controlled elements.

8.1.2.3 Final-Control Elements

Final-control elements are valves, dampers, electric heaters, relays, and motors for
fans, pumps, burners, refrigeration, and other HVAC equipment. All these elements
may be operated on-off or two position, while valves and dampers and motors may
also be used with floating-control and proportional-control modes. Final-control
elements may be normally open, that is, open with no controller signal, or normally
closed.
Pneumatic valve and damper operators have a flexible diaphragm or bellows
attached to a valve stem or damper linkage (see Fig. 8.1.17). Movement is opposed
by a compression spring, while a pneumatic controller signal is connected to the
operator and generates a force depending on the pressure of the signal in the area
of the diaphragm. When the signal pressure multiplied by the area exceeds the force
of the spring, the operator moves, also moving the valve or damper until the spring
force and controller's signal generated force are in balance. When the controller
signal reduces, the spring causes the operator to retract. By selection of springs,
various operator position-controller signal characteristics can be attained. Since operator position depends on the balance between the diaphragm and spring force,
any external force from a valve or damper will offset the operator. For control
systems requiring accurate synchronization of final-control elements, this may be a
problem. For precise positioning, a positive positioner is used. It senses controller
input signal and operator position and feeds or bleeds air to or from the operator
to position it regardless of external load.
Electric motors are unidirectional, spring-return, or reversible. Unidirectional
motors are for two-position operation: opening a valve or damper in half a revolution and closing it in the second half. Once initiated, the motor continues through
half a revolution. When it receives a second signal, the controller continues through
the next half-revolution cycle.

SIGNAL FROM
CONTROLLER
ROLLING
DIAPHRAGM
COMPRESSION
SPRING

PUSH ROD

FIGURE 8.1.17 Pneumatic valve

or damper operation.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

Co
py
rig
hte
dM
ate
ria
l

Spring-return motors are also used for two-position operation. A control signal
drives the motor to one end of its movement and holds it there. When the controller
is satisfied and ends its output, the motor is driven back by an internal spring which
was wound during its initial movement.
Reversible motors are used with floating- or proportional-control modes. The
motor can be operated in either direction, depending on the controller signal; it
stops when the signal stops. For proportional control, a potentiometer on the motor
shaft is used to signal the motor position to the controller.

8.1.2.4 Auxiliary Equipment

Many control systems require auxiliary equipment for complete system operation.
For pneumatic control systems, these include:

Compressed air systems with compressors, dryers, and filters to provide clean dry
air at the proper pressures to power the system
Pneumatic-electric relays for switching electric loads with pneumatic signals and
electric-pneumatic relays for switching pneumatic lines with electric signals
Two-position relays for converting proportional pneumatic signals to two-position
and proportional relays for reversing signals, selecting the higher or lower of two
or more signals, averaging two signals, adding or subtracting a constant from a
signal, and amplifying signal pressure or air-flow capacity
Switching relays to divert signals automatically or manually
Gradual switches to manually vary air pressure in a circuit
Electric systems utilize transformers to provide required voltage, relays to switch
electric loads larger than a controller's capacity, potentiometers for manual positioning of proportional control devices or for remote set-point adjustments, manual
on-off switches, and auxiliary switches on dampers and valves for control of sequence operation.
Other auxiliary devices are common to pneumatic and electric systems. These
include step controllers for operating a number of electric switches by a proportional operator to control stages of electric heating or refrigeration. Power controllers may be solid-state, saturable-core, or variable autotransformers and are used to
control electric resistance heaters with a proportional pneumatic or electric control
signal. Clocks and timers are used to control apparatus or control-system sequences
based on time of day or elapsed time.

8.1.2.5 Pneumatic, Electric, Electronic Comparisons

A pneumatic control system may be as shown in Fig. 8.1.18. Advantages of pneumatic controls are the inherent modulation sensors and controller signals and the
low cost of modulating operators. The system is explosion proof, and the control
elements require little maintenance and are easy to troubleshoot. Disadvantages are
the need for an air-compressor system which may be too expensive for small systems. Compressed air must be piped to all controls, increasing installation costs,
and transducers are required for interfacing to automation systems.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

"BRANCH" OR CONTROL
PRESSURE
SENSOR

CONTROLLER

Co
py
rig
hte
dM
ate
ria
l

UAMPtR
OPERATOR

DAMPER

FILTER

COMPRESSOR

PRESSURE
REGULATOR

"MAIN" OR SUPPLY
PRESSURE

THERMOSTAT
VALVE
ACTUATOR
VALVE
FIGURE 8.1.18 Pneumatic control system.

Electric control systems, as shown in Fig. 8.1.19, can be installed wherever

electric power is available and are low-cost for small simple systems. However,
modulating operators are expensive, and explosionproof housings are required in
hazardous areas. Electronic modulating controls allow remote sensors and set-point
adjustment, provide high accuracy, and readily interface with automation systems.
They are higher in cost and require more skilled personnel for trouble-shooting.
Direct digital controllers (see Fig. 8.1.20), offer many advantages. They have
very high accuracy, so control-loop accuracy is limited only by the sensor and finalcontrol element. They are capable of complex control algorithms which may easily
be changed by reprogramming. This allows flexible building operation during construction, startup, occupancy, full-occupancy, and expansion phases. An entire
building can be controlled from one location, and building-wide energymanagement strategies can be accomplished. Direct digital controllers are higher in
cost but can control multiple loops and share sensors.

THERMOSTAT

BURNER
CONTROL

ELECTRIC
POWER

FIGURE 8.1.19 Electric two-position control system.

Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com

DAMPER
ELECTRIC
MOTOR
ELECTRIC
OPERATOR
VALVE

Co
py
rig
hte
dM
ate
ria
l

DUCT

SPACE

DIGITAL COMPUTER
WITH INTERFACE
HARDWARE

DISCHARGE
SENSORS

E-P
TRANSDUCER

PNEUMATIC
OPERATOR
VALVE

FINAL CONTROL
ELEMENTS
FIGURE 8.1.20 Direct digital control system.

8.1.3

CONTROL APPLICATIONS

These are defined by general functions in the total HVAC system.

8.1.3.1 Boiler Control

The control of steam or water boilers involves three types of functionality: flame
safeguard, load control, and control of excess air. Steam boilers also include waterlevel control. The means of accomplishing each of these functions is influenced by
the size of the boiler involved. In general, small boilers have a single control package which accomplishes flame safeguard and load control with no need for excess
air control. Large boilers can have different control packages for all three functions.
The application of flame-safeguard control is very dependent upon boiler and burner
design and therefore is normally supplied by a complete package by the boilerburner manufacturer. The type of fuel(s) selected, the size of design load, and the
type of approval required are the primary decisions of the HVAC designer that
establish the type of boiler controls that are appropriate. On larger-size installations,
the method of load control, the use of multiple boilers, and the cost effectiveness
of appropriate types of excess air control are additional considerations for the HVAC
designer. This section explains means of accomplishing the three basic types of
boiler-control functionality. It also explains an auxiliary function of monitoring
smoke control.

Flame-Safeguard Control. The objective of flame-safeguard control is to ensure

that safe conditions exist for initiating and sustaining combustion. On small- to
medium-size [up to 400,000 Btu/h (422,000 kJ/h)] burners, the flame-safeguardcontrol function is provided by a package called a primary control. The primary
control starts the burner in the proper sequence, proves that combustion air is available, purges the combustion chambers and proves the burner flame is established,
and supervises the flame during burner operation. It causes safety shutdown on
failure to ignite the pilot or main burner or on loss of flame. In addition, the primary
control checks itself against unsafe failure. Typically, a check for flame-simulating

Next Page
Copyright 1997 by The McGraw-Hill Companies

Retrieved from: www.knovel.com