NATIONAL.

ADVISORY COMMITTEE FOR AERONAUTICS

luwm””
ORIGINALLYISSUED
June lgbq
Admnce Redwicted YeportL5108

COLU41f
AND PLATE COMPRESSIVESTR3!lfGTHS

OF AIRCRAFTsTRucmRAL MATRuMs

1~-T AImmuM--Y SHEET

By GeorgeJ. Helmerland J. AlbertRoy

QEm@ey MemorialAeroxiautlcaJ
Laboratory
IangleyField,?a.

-sfkYWl(t,tf’
NACA WARTIME REPORTS are remints of ~aDers ori~inallv issued to provide rap!d distribution of
advance research results to an authorized gr;ui requir;ng ti”em for the war effort: They were pre-
viously held under a security status but are now unclassified. Some of these reports were fiot tech-
.
nically edited. All have been reproduced without change in order to expedite general distribution.
 31176013642651 r
3
M!iC!A
ARR NO. L5F08
NATIONAL ADVISORY COMMITTkE FOR AERONAUTICS

ADVANCE RESTRICTED REPORT

.-
, -. ........ ... ..

COLUMN AND PLATE COMPRESSIVE STRENGTHS

OF AIRCR AFT STRUCTURAL MATE!RIALS

17S-T ALuMINuM-ALLOY SP=T

By George J. Helmerl and J. Albert Roy

sm ARY

Colwnn and ~late conm?essive stren@hs of 17S-T

aluminum-alloy skeet were-determined bo;h within’ and
beyond the elastic range from tests of thin-strip columns
and from local-instability tests of fornmd Z- and channel-
section columns. These tests are part of an extensive
reqeerch investigation to provide data on the structural
strengths of vcrious aircraft materials. Results are
presented in the fo~m of curves and charts that may be
used in the design and amlysis of aircraft structures.

IiTIWCDIJCTION

Column and plate members in an aircraft structure

are the basic elements that I’ailby instability. For
the design of low-weight, structurally efficient air-
craft, the strength of these elements must be known for
the various aircraft materials. An extensive research
. investigation has therefore been undertaken at the
Lan@ey Memorial Aeronautical Laboratory on the column
and plate compressive strengths of a number of the alloys
available for use in eircraft structures. That part of
the investigation for Z?@-T aluminum-alloy sheet has
already been completed and is given in reference 1.

The results of tests to determine the column and

plate-compressive strengths of 17S-T aluminum-alloy
sheet are given herein.

~
 ..

NAOA ARR No ● L5F08

SYMBOLS

len@h of column

radius of gyration

fixlty coefficient used in Euler column formula

effective slenderness ratio of thin-strip column

width of flange of Z- or channel section (see fig. 1)

width of web of Z- or channel section (see fig. 1)

inside radius of bend of’Z- or channel section

(see fig. 1)

thickness of plate

nondimensional coefficient used with bN and t in

plate-buckllng formula (see fig. 2 and reference 2)

modulus of elasticity in ccmmression, taken as

10,600 ksl for 175-T aluminum alloy

nondimensional coefficient (The value of T is

so detemnl.ned that when the effective modulus ‘rEc
is substituted for FCC in the equation for
elastic buckling of columns, the computed
critical stress a,greeswith the experimentally
observed value. The coefficient T is equal to
unity within the elastic range snd decreases
with increasing stress beyond the elastic range.)

nondimensional coefficient for plates corresponding

to T for columns

PoissonIs ratio, taken as 0.3 for 17S-T aluminum

alloy

critical compressive stress

average compressive stress at maximum load

compressive yield stress

 —.—-

NAOA ARR Ho. L5F08 3

METHODS OF TESTING “AND ANALYSIS

. .. ., . . .
...-!.- .-

All tests were made in hydreuli.c testing machines

accurate to within three-fourths of 1 percent. The ends
of the stress-strain specimens and the columns were ground
flat and square.

The methods of testing and enalysis developed

for this research program are described in reference 1
and may be briefly summarized as follows:

The compressive stress-strain curves for the flat

sheet, which Identify the material ~or correlation with
Its column and plate compressive strengths, were obtained
from tests of’ single-tkickness specimens in a compression
fixture of the Mongtomery-Templin type that provides
lateral support through closely spaced rollers. For the
bent material irlthe corners 01’the formed Z- and channel
sections, compression spec!mans were cut from the corner
portion and tested in a specially designed fixture that
provided continuous support along the length.

The column strength and the assooigted effective

column modulus were obtained by the use of the mathod
presented in reference 3, which consists in tests of
thin-strip columns of the meterlal with the ends clamped
in fixtures that provids a high degree of end restraint.
The fixtures used have been improved and the method of
analysis has been modified since public~tlon of refere-
nce 3. The method now used results in Q column curve
representative of nearly perfect column specimens. In
addition, the method now takes into account the fact that
columns of the dimensions tested are actually plates with
two free edges.

The plate compressive strength of the material. was

obtained from compression tests of formed Z- and channel-
section columns so proportioned as to develop local
instability, that is, Instability of’the plate elements
of which the columns are comprised. (See fig. 3.) The
lengths of’the columns were chosen In accordance with
the principles set forth In reference 4. The columns
were tested with the flat ends bearing directly against
the testing-machine heads. In these local-instability .
tests, measurements wera taken of the cross-sectional
distortion, and the critical stress was determined as
the stress at the point near the top of the knee of the
stress-distortion curvs at which a marked increase in
distortion first occurred with small increase In stress.
 .. .
I

4 NACA A.RRNo. L5F08

RESULTS AND DISCUSSION

Compressive Stress-Strain Curves

Compressive stress-strain curves for the 17S-T

aluminum-alloy flat sheet used are given in figure 4 for
both directions of grain. The compressive yield stress
for the cross-grain direction averages about 11 percent
higher than that for tk.awith-grcin direction.

When the flat sheet material is bent to an inside

radius of 3t to form a Z- or cliannel section, the cold
work done on the material evidently raises the compres-
sive yield stress for the corner to a higher value than
that for the flat web or l’lange. (See fig. 5.) The
increase in the compressive yield s~ress at the corner
is indicated to be about 23 percent for the with-grain
and about 15 percent for the cross-grain direction.
Because about 40 percent of the area of the curved corner
specimens represented flat material along the edges of
the specimen, f’or which the compressive yield stress is
less than that for the curved portion, the actual increase
in compressive yield stress for the corner portion may be
somewhat greater than the increase indicated in figure 5.

The Z-, channel, and thin-strip columns to which a

particular stress-strain curve applies are indicated in
table 1, together with the value of the compressive yield
stress acy for that stress-strain curve. These values
of Ucy average about ~1 ksi for the with-grain direc-
tion. ‘The modulus of elasticity in compression was taken
as 1o,6o(Iksi, the present accepted value for 17S-T
aluminum alloy.

Values of’the compressive yield stress of the

material used for these tests are somewhat higher than
the typical values given by the Aluminum Company of
America for 17S-T aluminum-alloy sheet. Because the
compressive yield stress of this material may vary
appreciably, the data and charts of’this report should
not be used for design purposes for material having
values of compressive yield stress appreciably different
from those reported herein unless a suitable method is
devised for adjusting test results for variation in
material properties.
 NAOA m NO. L5F08

-..._ ..7,, Column and Plate Compressive Strength

........--.” --.-,- -, .,. ----
The results of the column and local-instability
tests for the 17S-T aluminum-alloy sheet are summarized
herein. A discussion of the basic relationships Is given
In reference 1.

strength.- The column curves of figure 6 show

Column —.
—.—
the results of the thl.n-strip-column tests for both direc-
tions of grain. The reduction of the effective modulus
of elasticity ‘rEc with stress for columns Is indicated
by the variation of T with stress shown in figure 7.

Plate compressive strength.- The results of the

local-instability tests of the formed Z- and channel-
section columns,- used to determine the plate compressive
strength, are given in tables 2 and 3, respectively.

The plate-buckling curve, which Is analo ous to the

column curve of figure 6, is shown in figure g The
reduction of the effective modulus of elastici~y qEc
with stress for plates is indicated by the variation
of rj with stress, which is shown together with the
variation of 7 irlfl~re 7. The crossing of the T-
and q-curves shown in f’igure 7 occurs because the formed
columns appwently had an appreciable degree of imper-
fection, which resulted in the deviation of the q-curve
from unity at a lower stress than that at which the
T-curve, representative of nearly per.tect columns,
deviated from unity.

The variation of the actual critical stress ~pr

with the theoretical critical stress Ocr/~ computed
for elastic buckling by means of the for&ula and curve
of’figure 2 is shown in figure 9.

In order to illustrate the difference between the

critical stress an- and the average stress at maximum
load ‘~a, the v&iation of acr with 0cr/5m= is
shown in figure 10. Because values of ~m= may be
required in strength calculations, the variation of
cm= with acr/q is Indicated in figure 11. The

.*. , .,, . ,,, ,.

6 NACS Am No. L5r08

effect of grain direction on the plate compressive

strength is shown h figure 12.

Langley Memorial Aeronautical Laboratory

National Mvlsory Committea for Aeronautics
Uangley Field, Va.

REFERENCES

1. Lundquist, Eugsne E., Schuette, Ev~ H., Heimerl,

George J., and Roy, ,J.Alb~rt: Column snd Plate
Compresslvo Strengths of A1.rcr;ft Structural Ziiateriaho
d@-T Aluminlur-Ailoy SlmeL. ~~ACA ARR L5F01, 1945.
2. Kroll, ‘W.D., Fisher, Gordon P., and Helmerl, George J.:
Charts for Calculation of the Critical Stress for
Local Instability of’Columns with I-, Z-, Channel,
and Rectangular-Tube Section. NACA AWl No. 5K04,
1943 ●

3. Luniii~sg: aEugene E., Rossman, Carl A., and Houbolt,

A Method fog Determining the Column Curve
from Te;ts of Columns with ~qual Restraints against
Rotation on the Ends. NACA TN No. 903, 191+3. “
4. Heimerl, George J., and Key, J. Albert: Determination
of Desireble Lengths of Z- and Channel-Section
Columns for Local-Instability Tests. NACA RB
No. L4H1O, 1944.
 TABLE 1 # M

Cmlpresslw yiead
Columns to wh:ch stress-strain curves apply .Stres9-9train stress, ~cy ‘
curve (ksi)
Direction of Designation
Type (fig. 4)
loading (See tables 2 and 3 ) with Cross
grain grain
~in strip With grain and All A l+o.g 45.3
cross grati

With grain All except 5a to 5C B

z
5a to 5C c !$:? .k;:;’
{
i z Cross grain All D 40.2 45.6
I
Channel With grain ~1.1 B 41.5 46.5
I Channe1 Cross grain All D 40.2 “ 45.6

I
I

-J
1
NACA ARR No. L5F08 8

TABIJE
2

DIMENSIONS AND TRST RESULTS FUR FORMED Z-SECTION COLUNNS

“’THATDWi3LOPLOCAL INSTABILITY
,.

‘cr
% Oop =-
bw
(1% (1:.) $ ~ ~ 6* acr
’01- (1:. ) (In. ) kw ) (kal) (ksl) ~u
(a) t

.
With.grain

).ld+ 2.50
.9?2

T
).97
::3 ::g
.12 :n
.1d? 2:50 .9?
.uz 2.51 .921

1.24 13.26
1.2 13.23 ;~7;
1
13.23
1.2l?+
1.71 15.1
1.71 15.19 :%;
m %$ .%
2.21 17.2 .812
2.22 17.21 .784
/
37.0 32.0 .950
3 ●o 32.2 .93
i
%! %:$ :Bj
.8
.819
w %: .590
16.7 16:6 .613
16.5 15.3 .565

Cross grain

41.3 61.9 1.966

I
61.2
$)$ 61.6 %!
J9.6
&:; pj
;;32
3
ii! &
55.5 34. d
73.1 19.3 .707
72.8 19. .711
73.02 19.f? .715

•~rmucs see figure 2.

!3’
of

b Cl=p l#Eot
—. ., Uhre ?2.= 10,600ksl and P = o.3.

NATIOliALAD~80~
COMMIITER FOR AERONA~S

-.
 I I

9 “.
NACA ARR No. L5F08

. TABLE3

DIMENSIONS AND TSST RESULTS P3R FORMtD CHANNEL-SECTION COLUMNS

lTiATDEVELOP LOCAL 1NSTABILIT%

1 I I [ 1 I 1 1 I 1 1 1 1

m
——
D.122 2.51 0.99 10.61 4.220.60 51.0 1.933
.122 2.51 .99 10.60 1..220.60 ~ij:: .9 1
.123 2.51 1.37 12.04 4.820.42 . p:{ .921
.123 2.51 1.37 12.024.8 20. 5
1.37 12.104.8 20.t5
:%? ::% 1.75 13.~5 5.6 19.97
& 4
.9 6
:9 019.?
.123 2.50 1.75 13.75 5.5 20.30 33:9 Z6:;” .931

.123 3.13 I.24 13.264.2 25.4.2 41.9 44.0 .9 2

.12 1.24 13.254.2 25.53 40.6 .96
z
.12 43.1 g:~ .973
.124 ‘:! ::2$ %:8 k:j ;?::;
i 3.1 32.4 .931
.124 3.13 1. 0 15.204.9 25.23 34.4 35:5
1.19 15.19 4.925.53 35.1 ;~~
::3 w 2.1 17.20 5.5 25.3 :2:; 31.2
.123 3.15 2.1i 17.225.5 25.5i 25.1 31.3 .8o2

.123 3.88 36.5 32.0

1.5
%2 !!:;31.56 .3973.76 53.8 .950

LJl-lJl!!
.125 3.88 1.5k 31,1 .4013.7 53.2 32.8 U .956
.123 3.89 31.53 .3 8 3.7 53.8 3Z:Z
.123
.123
3.91
3.91
1.54
2.14 ;:”3$ !:331.7 f?
19:81 4.831.7k! :?4Z::?5
65.8
65.7
24.4 z:;
24.4
34.7
29.3
29.2
:%
.122 3. 0 “i. 18.80
:.: 4.831.9 .5482.53 66.5 23.9 %:? 29.8 jg
.122 3.88 2.72 21.31 5.5 31.9z 80 16.0 17.2 ;;
.12k 3.89 2.72 21.21 5.5 31. 1 :$: ::;1 “z 16 15.7 :k
.124 3.91 2.72 21.27 5.5 31.i!
3 .697 1.71 ~$:8 1
16.
“z 16.5 26.9 k;

Cross grain

0.123 2.22 4.51 .98 0.6282.03

~().()1) ~~.f
;.;?
!s:8

1T
.123 2.23 9.974.51i .11 .6252.05
.123 2.22 1:39 9.7 4.517.96 41.5 g:;
.123 2.54 1.53 12.82 5.0 20.60 :% :::; l&;
.123 2.55 1.53 12.81 5.0 20,71 j++:$
.123 2.55 12.81 5.0 20.71 m H 46:2
.123 .10 ;:3? 15.41 5.0 25.11 .5932.23 55.6 32:6
.123 J?.02 2.43 20.00 5.0 32.63 .6052.16 73.4
.123 h.03 2.43 20.00 5.0 32.92 .6022.1 73.5 ;i“
.;
1 1 I 1 ! I 1 1

aFt)r
values Of km, aae figwu 2.

RATYOSAL U)VMORY
CO-- M ABRCWAVTICS
 NACA ARR No. L!5F08 Fig. 1

—... . . .

NATIONALADVISORY
COMMITTEEFOR AERONAUTICS

Figure 1.- Cross sections of z-and channel-

section columns.

?’

II 11111111111111111111
II I Ill 11111Ill 1 Mm
Fig. 2 NACA ARR No. L5F08

5 I I I I
.
Buckling of web I
W < restrained by flange
/‘ \
/ \
/ \
4

Buckling of flange
restrained by web
3 \

kW

NAllONi,L ADVISORY
{;oMMI1lEE :ORAERONAJIM
o
‘o .2 .4 .8 1.0
b/bw-6

Figure 2.-Values of kW for Z- and channel- section

columns of uniform thickness (from reference ~.
CJ~r ~ kWw2Ect2
n 12(1
-JJ2)b/
 NACA ARR No. L5F08 Fig. 3

. .,,..._, .,.

\ I
\ I

NATIONALADVISORY
COMMlllEEFOR AERONAUTICS

Figure a.- Typical cross - sectional distortion

of columns that develop local instability.
 50
lb

I /
CroSs L
40 J*

-_ 30 A
2

i ) } i
A 0 c D

10 , , ,
ATIONAL ADVISO
B:Y
PWJTICS
ITTEE FCR AERO

0
1+-.002-4
Strain

z
“o
.
Figure 4.- Compressive Stress-strqin curves for 17 S-T r
aluminum - alloy flat sheet. (curves A, 0, C, D, are ~
o
al
identified in table L)
 NACA ARR No. L5F08 Fig. 5

----- .. . . .,.

60

so
Formed material from
curved corner,’ inside
radius, 3t
40

.-
3

a’ 30
m
@
L.
z

20 }

10

t- “002
-q
Strain

Figure 5. - Effect of forminq on compressive

stre3s - drain curves for 175-T aluminum - alloy
Z-section. t = 0.125 inches.
 ‘%1
l--
m
.

40
I
I TT m

cross grai
q . 45 I(si

With grain

0
o 20 40 60 L 80 100 120 140
z
FE 0
.

‘%!
Figure 6,- Column curves for 17s-T qlwninum-allo~ sheet. 0
CD
 .

NACA ARR NO. L5F08 Fig. 7

,.. .,,.. . ...., ..,. ,, ..

Lo

.8 \
\

~ (With groin
&cy= 4. I *i )’
.6

.4- T ~Cmss grain,

CCY =45 ksi)

.2 \
NATIONALADVISORY
COMMITTEEFO~AERONAUTICS

0
o 10 20 30 ‘ 40 50

Stress, ksi “

Figure 7.- Variation of q and z with stress

* for 17S-T aluminum - atloy sheet (q obtaineci
from tests of formed columns).
Fig. 8 NACA ARR No. L5F08

60

\
.

\
50 _--Elastic buckling
\

\
40
\ .
*\

~ z-section
30
0 Channel section
q-r ,ksi

20

10

0-
0 20 40 60 80 100 120

Figure 8.- Plate-buckling curve for 17S -T aluminum-alloy

sheet- loaded’ in the with –grain direction. Obtained .
from tests of formed columns. aCY = 41 ksi.
1111-1~
,–
1,

$Y
..-
1

NACA ARR No. L5F08

50

/ y

40 r

30
/

UCr,ksi

20
A Z-sec+ion

o Channel section-

10

Rgure 9.-Variation of ~Cr with qr/q for plates

of 17’5 -T aluminum-alloy sheet- loaded in the
with –grain direction. Obtained from tests of
formed columns. aCY .41 k.si.
50

40

20,25

I I I I I /
2
iwTt_tl

11111 l/+KIlll
I I
I I I
x-l I
I b.++
I I I I I I I I m
I
X 25
A32_
N \TIONAL),DVISORV
COMMI“TEE FORAERONAll
Tics

00 .6 .8 1.0 o .2 .4 .6 .6 1.0
.2 .4
C3/dmax
!2
Z – section Channel section 0
,.
Is

Figure 10. - Variation of CXCrwith CJCr/6maX for formed 17S-T aluminum-alloy co Iumns loaded %
o _
—
in the wit-h-grain direction. C$Y=41 ksi . a)

s.
#-
/
*- i
 50

J~ z
o
.
4C /

30-

,ksi 4( p

2C

b
3.
t
0 20 o 20
10 X 25
A 32 A 32 .

00 20 40 60 80 100 0 20 40 60 06 100
‘NATIONAL
crCr/q, ksi acr/~ , ksi
ADVISORY
coMMiTT~~ ~oflA~RoNAunm

Z - section Channel section

‘=3

Figure I I.-Variation of ~~ax with OC,/~ for. formed 17S-T aluminum-alloy columns
loaded in the with-groin direction. OCY=41 ksi.
 5C !30 w

40 40 =20 ,25
40 20.25
(-)/
r
4 / 5 32 32
Y 4
30 30
30 With grain With grain
5+
1!--With grain $ ~

Gmax,

‘dmzmI1
u~~ x ,
ksi ksi
20 20

&b bw
t T —
A 18 A 18
Io ❑ 21 “— Io ❑ 21—
0 25 0 25
033— 0 33—

00 20 40 60 00 20 40 60 80
“o 20 40 60
a-r/q, ksi csCr/q, ksi NATIONALAOVISORY
COMMITTEEFOR AERONAUTICS

Z-and channel section Z- section Channel section

ia
?3

z
o
Figure 12: Comparison of the compressive strength of formed Z- and channel-section columns of 17S-T aluminum alloy ●

F
loaded in the with-gy-om and cross-grain directions. With-groin curves are taken from from figures 9 and I I, and test cn
points are for the 6-oss-gro[n direction. ‘3
$
 lllllllllllMltllmT
_.3_
l_176013542551
!