POISSON'S RATIOS FOP

GLASS-FABRICABASE

PLASTIC LAMINATES

No. 1860

January 1957

INFORMATION REVIEWEDAND REAFFIRMED

1965

This Report is One of a SeriesIssued in Cooperation with theANC-17 PANEL ON PLASTICS FOR AIRCRAFTof the Departments of theAIR FORCE, NAVY, AND COMMERCE

owl I

11 I Illilaniumr

FOREST PROliUCTS.LABORATORT

MADISON S. WISCONSIN

UNITED STATES DEPARTMENT OF AGRICULTURE

FOREST SERVICE

In Cooperation with the University of Wisconsin

POISSON'S RATIOS FOR GLASS-FABRIC-BASE PLASTIC LAMINATES1

By

ROBERT L. YOUNGS, Technologist

2Forest Products Laboratory, — Forest Service

U. S. Department of Agriculture

Summary

Poisson's ratios were determined by test for several glass-fabric-baseplastic laminates after normal conditioning. Loads were applied paral-lel and perpendicular to the warp direction to obtain Poisson's ratiosalong the natural axes of the materials. A few tests were also made todetermine the effects of wet conditioning and type of resin on Poisson'sratios.

For most of the laminates investigated, the Poisson's ratios were about1/8.

Introduction

Values of Poisson's ratio are frequently required in the calculation ofthe elastic and strength properties of glass-fabric-base plastic laminates.

1—This progress report is one of a series (ANC-17, Item 56-4) prepared

and distributed by the U. S. Forest Products Laboratory under U. S.Navy Bureau of Aeronautics No. NAer 01683 and U. S. Air ForceContract No. DO 33(616)-56-9. Results here reported are prelimi-nary and may be revised as additional data become available.

2—Maintained at Madison, Wis. , in cooperation with the University of

Wisconsin.

Rept. No. 1860 -1- Agriculture-Madison

Few data are available on Poisson's ratios for such laminates, and datafrom different sources are not consistent. In some cases the use of anapproximate value is satisfactory, but in other cases a more reliablevalue is required.

Poisson's ratios have previously been obtained for parallel laminatesreinforced with 162, 181, and 143 glass fabrics, 3— and values based onthese data are included in this report. Limited data on Poisson's ratiofor a few laminates at low stress levels are also presented in the ANC-17Bulletin. 4

The work here reported was done to determine Poisson's ratios for sever-al different laminates, with the principal stresses applied both paralleland perpendicular to the warp direction. Most of the tests were made onparallel laminates after normal conditioning. A few tests, however, wereconducted to investigate the effects of wet conditioning and the type ofresin used.

This research was conducted at the Forest Products Laboratory fromJuly to November 1956, at the request of, and in cooperation with, theANC-17 Panel on Plastics for Aircraft.

Description of Material

One glass-fabric-base plastic laminate 1/8 inch thick, 18 inches wide,and either 18 or 24 inches long was made with Selectron 5003 resin andeach of the following reinforcements:

(a) Plain weave 112 Volan A

128-114

(b) 4-harness satin weave 120-114

3—Erickson, E. C. 0. , and Norris, C. B. Tensile Properties of Glass-

Fabric-Laminates with Laminations Oriented in Any Way. ForestProducts Laboratory Report No. 1853, November 1955.

4U. S. Forest Products Laboratory, ANC-17 Panel on Plastics for Air-craft. ANC-17 Bulletin, Plastics for Aircraft; Part I, ReinforcedPlastics. Issued by Departments of the Air Force, Navy, and Com-merce. 1955.

Rept. No. 1860 -2-

(c) 8-harness satin weave

181 Volan A182-114184 Volan A

(d) Unidirectional weave 143 Volan A

In addition, one laminate 1/8 inch thick, 18 inches wide, and 18 incheslong was made with 181 Volan A fabric and Epon 828 resin with CuringAgent CL.

All panels were parallel laminated except the one reinforced with 143Volan A fabric, which was cross laminated.

Selectron 5003 resin is a laminating resin of the polyester (alkyd-styrene)type conforming with the requirements of Military Specification MIL-R-7575A. Epon 828 is an epoxy laminating resin of the type required byMilitary Specification MIL-R-9300A.

Laminates made with fabric having 114 finish, an older type of fabricfinish, are generally affected more by wet conditioning than laminatesmade with Volan A or one of the other improved finishes. For deter-mination of Poisson's ratios of normally conditioned specimens, however,fabric finish is not considered important. For this reason, fabric having114 finish was used in some of the laminates from which specimens wereto be made for test after normal conditioning, in order to use glass fabricon hand. Laminates that were to be cut into specimens for tests aftereither normal or wet conditioning were reinforced with fabric havingVolan A finish.

Polyester laminates were prepared by means of the usual wet lay-upprocedures. The resin was catalyzed with 0.8 percent of benzoyl per-oxide, by weight, and the resin and fabric were laid up between cello-phane-covered aluminum cauls. The laminates were cured for 90 minutesin a press at a pressure of 14 pounds per square inch and a platen tem-perature that was gradually increased from 220° to 250° F.

The epoxy laminate was prepared by wet lay-up procedures with Epon 828epoxy resin containing 14 percent by weight of curing agent CL. The resin-fabric assembly was cured in a press at 215° F. with contact pressure for12 minutes followed by a pressure of 25 pounds per square inch for 48minutes. This laminate was not post-cured. Fabrication details for this

Rept. No. 1860 -3-

type of epoxy laminate are siven in Forest Products Laboratory ReportsNos. 1820-0 5-- and 1824-C.a

Information on the construction and physical properties of the laminatesis listed in table 1.

Test Specimens

From each panel except those reinforced with 143 or 181 fabric, fivespecimens were cut with their longitudinal axis parallel (0°) to the warpdirection and 5 with their longitudinal axis perpendicular (90°) to it. Fivespecimens were cut from the 143 cross laminate in the 0° direction,which was parallel to the warp of the face laminations. Ten specimenswere prepared from each of the 181 parallel laminates in the 0° direction,and those from each panel were numbered consecutively. Odd-numberedspecimens were tested after normal conditioning, and even-numberedspecimens were tested after wet conditioning.

Each specimen was 16 inches long and 1-1/2 inches wide and was neckeddown at the center to a minimum section 0.8 inch wide and 2-1/2 incheslong. The minimum section was connected to the wide ends by circulararcs of 20-inch radius that were tangent to the minimum section. Speci-mens were cut to approximate size on a bandsaw and finished to finalsize and shape with an emery wheel mounted on a shaper head.

Two metalectric strain gages were mounted near the center of the mini-mum section on each side of each specimen; one to measure elongationand one to measure contraction. On specimens that were to be testedafter normal conditioning, 1-inch SR-4, Type A-1 strain gages weremounted longitudinally to measure elongation, and 1/2-inch, Type A-5strain gages were mounted transversely to measure contraction. Onspecimens that were to be tested after wet conditioning, 1-inch, TypeAB-3 strain gages were mounted longitudinally to measure elongation,

-Youngs, R. L. Supplement to Mechanical Properties of Plastic Lami-nates. Forest Products Laboratory Report No. 1820-C, August 1956.

6—Youngs, R. L. Supplement to Bolt-Blearing Properties of Glass-Fabric-

Base Plastic Laminates. Forest Products Laboratory Report No.1824-C, February 1957.

Rept. No. 1860 -4-

and 1/2-inch, Type AB-5 strain gages were mounted transversely tomeasure contraction.

Specimens that were to be tested after normal conditioning were condi-tioned for at least 2 weeks at 75° F. and 50 percent relative humidity.Specimens to be tested after wet conditioning were conditioned for 60days at 100° F. and 100 percent relative humidity.

Testing Procedure

Specimens were removed from the conditioning area one at a time andmeasured for thickness and width at the minimum section by means ofa micrometer caliper. They were then placed in Templin grips in ahydraulic testing machine and loaded in tension at a constant head speedof 0.035 inch per minute.

One specimen of each load-direction or conditioning group of each lami-nate was mounted in the testing machine and loaded until it failed.These tests were made to show the stress at which the initial and sec-ondary proportional limits occurred in order to serve as a guide forPoisson's ratio tests. Specimens for determination of Poisson's ratiowere mounted in the testing machine, given a small initial load, thenloaded and unloaded four times in one continuous operation with no in-tentional rest or recovery periods. In the first two runs, loads reachedabout three-fourths of the secondary proportional-limit load of the speci-men, as previously, determined from the control test for the group. Inthe third and fourth runs, loads reached about the secondary proportionallimit load. Each loading started at the-initial load, and each unloadingexcept the last ended at the same load.

Elongation and contraction strain data were taken during the first loadingrun, the first unloading run, the second loading run, the fourth loadingrun, and the fourth unloading run with the metalectric strain gages mountedon the specimens.

Tests of normally conditioned material were conducted on three speci-mens at each angle of loading from each laminate. Tests of wet-condi-tioned material were conducted on four specimens of each 181 laminate.

Rept. No. 1860 -5-

Calculation of Poisson's Ratios

Elongation and contraction strain data for each specimen were plottedto the same vertical load scale and suitable horizontal strain scales, asin figure 1. Straight lines were drawn through the plotted points toindicate the slopes of the elongation and contraction curves for each run.Initial and secondary slopes were clearly evident on the elongation curvefor the first loading of each specimen, but were so indistinct on subse-quent loadings that no attempt was made to distinguish between them,since this distinction would have had no significant effect on the valuesof Poisson's ratio calculated from them. All unloading runs were char-acterized by a single slope.

Poisson's ratios were then calculated for the initial and secondary por-tions of the first loading run and for the first unloading run, the secondloading run, the fourth loading run, and the fourth unloading run by divid-ing the strain in contraction perpendicular to the direction of loading bythe strain in elongation parallel to the direction of loading. The ratio ofcontraction in the perpendicular-to-warp direction, ft, to elongation pro-duced by a tensile load in the parallel-to-warp direction, a, was desig-nated as Flap. The ratio of contraction in the parallel-to-warp direction,a, to elongation produced by a tensile load in the perpendicular-to-warpdirection, f3, was designated as t.ipa. Average values of flap and liga were

calculated for the corresponding phase of loading or unloading of all speci-mens in a group.

The indications of the metalectric strain gages used to measure elonga-tion and contraction strains were subject to error resulting from trans-verse sensitivity. This was particularly significant for the gages measur-ing contraction strain, which was quite small relative to the elongationstrain. Each average value was therefore adjusted to eliminate this errorby means of the relationships presented by Baumberger and Hines.—

Presentation and Discussion of Results

Values of Poisson's ratio calculated for each of the laminates tested inthis work are presented in table 2. Also included in table 2 are values

?Baumberger, R. , and Hines, F. , 1944. Practical Reduction Formulasfor Use on Bonded Wire Strain Gages in Two-Dimensional StressFields. Experimental Stress Analysis 2:1:113-127.

Rept. No. 1860 -6-

of Poisson's ratio based on previous similar tests of laminates reinforcedwith 162, 181, and 143 glass fabrics. These values were adjusted fortransverse sensitivity in order to make them more closely comparableto the values obtained in this work, although it is probable that the adjust-ment is of slight practical significance. Modulus of elasticity values cal-culated from the Poisson's ratio elongation data of this report are alsopresented in table 2.

A set of typical load-strain curves in elongation and contraction are shownin figure 1. These curves are based on data obtained in tests of a 181epoxy laminate, and are similar in form to those based on polyester lami-nates tested in this work and to those based on a 181 polyester laminatepreviously tested. 3

The data of table 2 show that the Poisson's ratio was highest for the initialportion of the first loading and lowest for the secondary portion of the firstloading for each material except the 143 parallel laminate that did not havetwo distinct slopes in the first loading curve. In subsequent runs, thePoisson's ratios for both loading and unloading remained about equal tothe average of the Poisson's ratios for the initial and secondary slopesof the first loading.

Since a structural component is usually subjected to repetitions of load,the Poisson's ratio value based on loadings after the first application ofload will probably be the proper value to use in design. The data of table2 indicate that a Poisson's ratio of 1/8 would probably be suitable for mostof the laminates investigated. This is true for both Flag and ppa. The out-standing exception to this was the 143 parallel laminate, for which p. cri_i wasabout 1/4 and ppa about 1/16. An exception might also be made for the182 laminate, for which 1.1.0 was about 1/7 and p.pa about 1/10.

The data on 181 polyester and epoxy laminates tested in both the dry andwet conditions indicate that wetting produced a slight increase in Poisson'sratio in all loading and unloading runs for both types of laminate, althoughthis increase was so slight that it would have little effect on most designcalculations. These same data indicate that Poisson's ratio was higherfor the epoxy laminate than for the polyester laminate in all loading andunloading runs in both the dry and wet conditions. Again, this differenceprobably has little practical significance.

Rept. No. 1860 -7-

It has previously been observed that the tensile modulus of elasticity ofglass-fabric-base polyester laminates that have been preloaded in tensionto a high stress level is generally only slightly higher than that based onthe secondary portion of the first loading curve. Similar relationshipswere observed in this research. Few data are available, however, onthe effect of tensile preloading on the properties of epoxy laminates.

Table 2 lists ratios of modulus of elasticity obtained for the fourth load-ing to the secondary modulus obtained for the first loading. The ratiovaries somewhat among laminates, the final modulus of elasticity ofpolyester laminates being about 3 to 10 percent higher than the secondarymodulus for the first loading. There was, however, a distinct differencein this respect between the 181 polyester laminate and the comparable 181epoxy laminate. The average ratio was about 1.06 for the polyester lami-nate and 1.17 for the epoxy laminate. Similar relationships might be ex-pected for polyester and epoxy laminates reinforced with other fabrics,although the differences may vary. Thus, while the secondary tensilemodulus of elasticity might be used as a design value for polyester lami-nates, it is possible that it may be too conservative for epoxy laminates.

Conclusions

Results of this work indicate that a value of 1/8 would be satisfactory forPoisson's ratio of most glass-fabric-base plastic laminates when loadedat either 0° or 90° to the warp of the fabric. For the parallel-laminated143 laminate, Poisson's ratios of 1/4 for 0° loading and 1/16 for 90° load-ing were indicated. For the 182 laminate, Poisson's ratios of 1/7 for 0'loading and 1/10 for 90° loading were indicated.

Poisson's ratios were slightly higher for a 181 epoxy laminate than for acomparable polyester laminate. Poisson's ratios of the 181 laminateswere increased somewhat by wet conditioning. In neither case, however,was the difference great enough to have any significant effect on mostdesign calculations.

8—Werren, Fred. Effect of Prestressing in Tension and Compression on

the Mechanical Properties of Two Glass-Fabric-Base Plastic Lami-nates. U. S. Forest Products Laboratory Report No. 1811, September1950, and Supplement A, June 1951.

Rept. No. 1860 -8-

The ratios of tensile modulus of elasticity of the fourth loading run to themodulus of the secondary portion of the first run were considerably higherfor the 181 epoxy laminate than for the comparable 181 polyester laminateor other polyester laminates. The secondary tensile modulus of elasticityof this epoxy laminate may be too conservative to use as a design value.

Rept. No. 1860 -9- 1. -13

co r-• n.0 n.0

•• •0 .0 •• 00 so 0 .0 se 0.0 44 04 • •0 •• r• •• •• .0

Ltrn c03 g...J tos, 8 (,z4 CO

• • • • • • • •

H H H H H H H H

•• •• •• •• •• •• •• •• •• •• •• •• • 4. 114. •• •• ••

-Pa)0laa)

1:14

01• 0• 0;) cc) t---• • • • •ON LI-N 4.4N 4-4-N ON ON 0KN Pr\ KN Pc\ CU 0.1

$-1'0cd

I. •• •• • •• •• •• •• •• •• •• •• •• •• •• •• ••S.• •• •• ••

crN c:) CU 0

• • •

a)rEl0

H•• •• •• • •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• ••

PI a)\D CO 0 ON rfN

44-N 4-4

•• •• •• • •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• • •

sio co 1-14-

H H H H H H

o H 1 1 0 0 1 1 1. 1 m0)

a) P4 cd cd cd cd c d I:0

•f-1 (1-1 fa 2k NA 0 P4 PA P4 w Pi 0 Pi

•• •• 1I• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• 4.

44N K1 reN res, PC N KN tr‘0 0 0 0 0 00 0 0 0 0 0 0n rn TN Lr\ in lf \ trl

• 8 0 9 0 8 0 0O N N N /-4 IA N N0 4-) 4-) +3 4-3 4-3 4-3 4-)[4 0000000om0000a)

H H 4-1 H H 4-1 Ha) a) 451) a) a) a) a)C13 cr) to c)) (13 El] CD

•• •• •• • Oil •• •• 41, •• •• •• Do •• 010 t• •• •• •• •• ••

4 4 4 4

A ...* .4. ....„ A A A M152 1 N

i> r.. I> t>.o o o o

I I •co 0 Cu 8 lir) 8 crol

H 4-1 H H H

Jazi

M el-I0

•• • •• •• •• •• •• •• •• •• •• •• •• •• •• ••

zS\ \\ 8 CkS (J.)

if\ U u-N trN

Rept. No. 1860

O

Rept. No. 1060

*-1

00 c0000

ri:f.siO.

on

AF.P

1,10o00

H4,

Aa)ul

" " "

Rtr,0o00

8C-.42H0ua

" ."

1

?i

q,4

P2

HODHO

71,1

og

,

P

14",0o00

8Fa

-P20

co" 4. 41

1

'C'j

0ri

'AN

,DHHI

pg

0. 0 "

ONW00

og

kN-N0on

8i+

1;7

ca0 . "

1

4t;0oo

gwT_,

. .•

MW00

-AP

1

on

RPr \0on

81.-.

'Pt?)H0(000 " "

-*I

Cel

4lo:00

gT

F4A

n-s0ocrlgFa-I.0A0CO

1 0 .0

C'8

AU

t-00,1

00

AA,

-'

a.)

..

1M -1

14P

14co0.1c0

01

P.fr.1" "

R

aD

A

MH0,1

no

A+,

”

A0 $4P

.0-n0o(0

8c.4.00crl

00 0 0 0 .

HN0

q

!=01.9

Ho

no

rHJ

hi

A

n.

fn-,0on0o{-I4)0H000

" 00 "

M

A

,-- ....H........HH

HMHO

no

H L0';',9

- •

Nfn

n0o(0

VI4,

Ha)01" " ..

-

_I-

0

..

R2

no

,r,

,1

F-,0.—

•

T

Zi

24 2

0muN

A ANin

0-Noo00

g CIR.4,

0 0.11a)co

0 . " ..

r

0.-a-80 0m,-Po

0 )4 ,

0a

00

0

8044

20

0,O

, . .

1 A gSo 2

4.4. .0 1 4 , P•0 0 0 ,,N.o m01g t e, 17 2'.0 . o o0 . .0 4.,u-\ V I 0W v Fa 0v 0 0 0g H 0 p

4A '-' 11 Tii• 0 O-. HPa 0 m H

7c-0 figt kN' t .1 '8tiS. C: II: qgPI. . , 'rg

. 1 A E 3, 1 H 0a. 4. I W •;17 .g 1 x 4.1 '-40 4-1 110 0 ,-10 0 . 5,O F0004.k 4.4.• 0 4. 0 0F01-) r., H 0 . la3 0 •0 0 01N, 0 0 0)co 1.--. o., 0 4,m .4 0 V 0g°' +' 9 ,10)1 4. 4 ' 0 rq0 .. 0 0 .0WW00014.m Im m . . 0 to

.2 2 2 4 8M M g .,....,'

N 1 0,1 -71 ..“1

0 00..

oo1.,-nN- -40r N

000,0c0_4-NN

00nMnNNN

00no00c-nN

00HH

00

00Cl1030 Ckrt-s 01

00c0-1-OnnN

00nOMHNN

00-1-OFNnnN

nOHH

H H

0 0LC\ ‘.0

01 01• • • • • • • • .

0000NN

•••••••••

.0nnNN

0 00\ 01NN

••

000000cnt-cow

0oWw0tNN

00M-1-Nz).0NN•• •• ••

0MOHMnN

C‘

-------------

•• • ,00Nn

nn

0 0

nn

00WOC-ChN N•• •• ••

00NWO N100

H

00nn

NN

00Wnmt-NN

00Om00.3cki&•• •• ••

00MnNMnN

nMHo

0

0"'"OV,. 1-V302t'

4 .9 0 •0 0Fa{0 0 ri

'8 4orq- 0.2 8

‘)S

ri

01▪

OO

P,

00r-14c-11

03c)

-• •• •• •• •, .• •• •• •• •• 1. • •• •• •• •• •• .• -• •• •• -• •• •- • •• •• •,

92 W2 ,1ri N1 T` ri ri m 1i 0 r 1 1 6T8 T0

HO nO CHN HO ON nn no NH nv, 0HIH HH HH H HH HH HH No H

o o0 -1

0

Orl

P.0

Hm

O0

o0 0.0 ,0

rglto

Hco

▪ ta .•

WC%

N

474• ••

1,0rJ

••

00nw

NN

0OCU

CU

00c01NN

000N-1-oDNNo7

00(OHMC--NN

NHHH

...