NATIONAL ADVISORY COMMITTEE

FOR AERONAUTICS

TECHNICAL NOTE

No. 1005

4 L4 'T

SWDARD TEST SPECIMENS MW MOLDED PHENOLIC PARTS

By P. S. Turner and R. H. Thomason National Bureau of Standards

1^^^^N A C ^UJF^

Washington May 1946

https://ntrs.nasa.gov/search.jsp?R=19930080796 2020-07-22T01:42:01+00:00Z

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

TECHNICAL NOTE NO. 1005

CORRELATION BETWEEN STRENGTH PROPERTIES IN

STANDARD TEST SPECIMENS AND MOLDED PHENOLIC PARTS

By P. S. Turner and N. H. Thomason

SUMMARY

This report describes an investigation of the tensile, flexural, and impact properties of 10 selected types of phenolic molding materials. The materials were studied to see in what ways and to what extent their properties satisfy some assumptions on which the theory of strength of materials is based: namely, (a) isotropy, (b) linear stress—strain re-lationship for small strains, and (c) homogeneity. The effect of changing the dimensions of tensile and flexural specimens and the span—depth ratio in flexural tests were studied. The strengths of molded boxes and flexural specimens cut from the boxes were compared with results of tests on standard test specimens molded from the respective materials,

The nonuniformity of a material, which is indicated by the coefficient of variation, affects the results of tests made with specimens of different sizes and tests with differ-ent methods of loading. The strength values were found to depend on the relationship between size and shape of the molded Specimen and size and shape of the fillers. The most

sig-nifjant variations observed within a diversified group of materials were.found to depend on the orientation of fibrous fillers. Of secondary importance was the dependence of the variability of test results on the pieces of filler inoorpor'aj.ed into the molding powder as well as on the siz..e of the piece.

Static breaking strength tests on boxes molded from six

representative phenolic materials correlated, well with falling—ball impact tests on specimens cutfrom molded flat sheets. Good correlation was obtained with Izod impact tests on standard test sp ecimens prepared from the molding materials. The static br'eakjn.g strengths of the boxes do not correlate with the resuj.ts o:f tensile or flexural tests on standard Specimens.

NACA Ti No. 1005

2

INTRODUCTION

No thorough investigation of the relationshis between the strengths of molded plastic articles and strength data on their materials has been reported. Most of the available data on the strength pro p erties of molded phenolic olastics have been obtained with standard test secimens and standard methods of test. Specific data of this t ype, obtained in accordance with test methods established by the '.merican Society for Testing Materials, are published, in manufacturers' data books, for examp le, references land 2. These sources acknowledge that a "molding material, which on standard test pieces appears superior, may show uo in actual production as being even inferior to another material which on standard test pieces reveals a lower order of desirable propertiest (ref-erence 1). These discrepancies are further attributed to such factors as peculiarities in mold design, size and sha p e of the molded article, and variations in molding conditions, but not to inherent differences in the materials or to selective characteristics of the standard test specimens.

This investigation of the strengths of molded tarts and standard test specimens, conducted at the National Bureau of Standards, was sponsored by and conducted with the financial assistance of the National Advisory Committee for Aeronautics.

The molding materials for this investigation were suoDlied by the Bakelite Corporation and the Monsanto Chemical Comoany, Flat plates molded in 1/9- and 1 / 4-inch thicknesses were fur- nished by the Bakelite Cornoration and boxes were molded of the same materials by the General Electric Com p any , Plastics Division. The cooperation of these firms has made possible this exploration of the nebulous region between standard tests and structural performance and is gratefully acknowledged.

MATERIALS AND PREPARATION OF TEST SPECIMENS

The materials used in this investigation are listed in table I. Flat sheets were molded from the Bakelite phenolic molding materials by the Bakelite Cor p oration. Rectangular boxes were molded from the same compositions by the General Electric Company. Moldings prepared at the National Bureau of Standards included; dumbbell tensile specimens in accord-ance with type I of Method No. 1011 of Federal Specification L-P-406a; rectangular bars, 1 inch wide and 5 inches long;

NACA TN No. 1005 3

bars 1/2 by 5 inches for impact specimens; cylinders, 2 inches in diameter and approximately 1 inch in length; and disks, 4 inches in diameter and of various thicknesses. The molding conditions are given in table II.

The specimens molded at the Bureau of Standards were prepared with fully positive hand molds heated by conduction from steam—heated platens. The molding was done with a 50-ton—capacity semi—automatic press. Preforms were prepared in the same molds that were used for the particular specimen. Some of the materials did not produce good preforms at room temperature with the pressure available or permissible for the particular mold. Preforms of these materials were made at somewhat elevated temperatures (see table II) with a hand—operated hydraulic press of 18—ton capacity. The platens were electrically heated and thermostatically controlled.

All machined test specimens were prepared at the National Bureau of Standards. The tensile specimens were milled with a machine having a cam—operated milling fixture for duplicating the desired contour.

TEST PROCEDURES AND EQUIPMENT

Tensile Pests

Tensile tests were made in accordance with Method No. loll of Federal Specification L—P-406a, except that the rate Of separation of the grips was maintained at 0.05 inch per minute and the profile of the specimen specified for thick-nses of 2/4 inch or less was used for all tensile tests. Strains were measured with a Southwark—Peters plastics extensometer, Model No, PS-6, and an autographic stress--strain recorder.' The tensile tests were made on the 0— to 2400—pound range of a 06 0,000—pound—capaoity universal hydraulic testing machine.

Flexural Tests

Flexural tests were conducted at rate of loading speci-fied in Method No. 1031 of Federal Specification L—P-406a

'These strain gages and the recording equipment are described in Bulletin No. 162 issued by the Baldwin Loco-motive Works, Baldwin Southwark Division, Philadelphia, Pa.

NPCA TN No.. 1005 -

for obtaining load—deflection data. Calculations of flexural strength, maximum fiber stress, and modulus of elasticity were made as described in Method No, 1031..

In the tests reported in tables III and IV approximate span—depth ratios were obtained with support and pressure pieces having loading edges rounded to 1/8—inch radii. This jig, together with a notched spacing and centering plate, provided adjustment at a limited number of positions. Sub— sequent tests were made with self—centering continuously variable jigs of the type shown in figure 1 with attachments for obtaining load—deflection data. The deflections were measured with Southwark—Peters plastics extensometers, Models PS-6 and PS-7, and an autographic stress—strain recorder. The pressure and support pieces of the jig which was used for spans of 2 inches or less, shown in figure 1, were rounded to radii of 1/32 inch; those of a larger jig of the same type which was used for spans greater than 2 inches had radii of 1/8 inch,

Specimens used in the study of the effect of span—depth ratio were first broken at the largest span. The remaining pieces were used. for tests at shorter spans. Care was taken to insure that points highly stressed in the first test did not coincide with points of maximum stress in subsecuent tests. Comparisons with specimens which had not been used previously indicated that the portions of the specimens which were used again had not been damaged in the first tests. This method of sampling was used to avoid effects of thick-ness and cure which might affect the results.

The flexural tests were made on the 0— to 2400—pound range of a 6 0 1 000—pound. capacity universal hydraulic testing machine and on the 0— to 240— and 0— to 1200—pound ranges of a 2400—pound—capacity machine.

Izod Impact Tests

The standard Iz.od impact test was conducted in accordance with Method No. 1071 of Federal Specification L —P-406a on speci-mens having machined notches. The molded 1/2— by 1/2— by 5—inch bars were cut in half to make two impact specimens. One—half of each bar was notched in the direction of the molding ram motion and the other half of each bar was notched in the direction pe r p endicular to the ram motion. The tests were made with a pen. 1. yp.e a,ct—testing machine of 4 foot—pound capacity, using the 2— and 4—foot—pound. ranges.

NACA Ti No. 10055

A correction for the energy absorbed in tossing the broken pieces of the specimens was obtained as follows: The broken Meces of the specimens were fitted together and subjected to a second im p act. This tossingenergy was cor- rected for friction and windege. It w as assumed that the energies imarted to the severed end of the specimen were r000rtiona1 to the unextended energies after the Iod and tossing tests. A portion of the tossing energy roortionl to the unexp ended energy in the Izod test was subtracted from the Izod imp act value.

Falling-Ball Imp act Tests

Falling-ball progressive_re,eated mact tests, similar to Method No. 1074 of Federal SDecification L-P-406a, were conducted. on 3- by -inch rectangular sections cut from the molded flat sheets. Preliminary tests were made on 14-inch-diarneter disks of various thicknesses supiDorted On a cast-iron- p ipe cap.

The equipment used for testing the rectangular sections is shown in figure 2. The specimens were mounted in a hard-wood frame which rested on a flat steel plate. - The frame provided a 1/8-inch-wide supporting area at the edges. A 1/2-pound steel ball was used for tests on 1/8-inch-thick sheet material and the molded disks; 112- and 2-'oound balls were used for testing 1/ 14-inch_thjck sheet material.

The height of fall was increased in steps of 1 inch starting with the 1-inch height, until comp lete failure occurred. The energy to crack the specimens also 'as noted.

Impact - Flexural Test

Flexural specimens 112 inch wide, i.hich were machined from the 1/8-inch-thick molded flat sheets, were struck at the center of a 3-inch span "ith a ball weighing 0.15 ocund. The euiment was essentially the same as that shown in figure 2 except that a flexural-test jig (see fig. 1) with suoport pieces rounded to 1/9-inch. radii was substituted for the wooden frame. The height of fall was adjusted by successive high-low ap p roximations so as to obtain the height of fell which would crack but not break apart the soecimen of a specific material v ith a single blow. A majority of the specimens received impacts close to the energ y recuired to crack the s p ecimen. All s p ecimens which were not ooletelv broken by the impact were subjected to e. flexur1 test at a

NACA TN No. 1005 6

span-depth ratio of 9:1 With the load, applied at the point of impact. The flexural strengths were plotted against the impact energies applied to the individual specimens.

Tests on Molded Boxes

Breaking-strength tests were conducted on molded boxes using a plunger having a hemis pherical end of 5/9-inch radius, as shown in figure 3 . The rate of travel of the plunger rel-ative to the base of the boxes 1"ac 0.05 inch per minute. The loads were applied in three ways: (1) at the molded hole; (2) at a point on the bottom smmetrical1y located on the diagonal with respect to the center and the molded hole; and (3) at the latter point through a rubber cushion. The load was not app lied at the center of the bottom because of a deeply indented mold number located at that critical point. The rubber cushion was a No. 7 rubber stopp er placed with the larger diameter face bearing on the surface of the box. (See fig. 3.) No attempt was made to analyze the stresses set up in the boxes b y the loads applied.

cdiioi , _ All the specimens were conditioned for at least 8 hours at 25 0 C (770 F) and 50 percent relative humid-ity and were tested in the conditioned atmosphere.

Statistical Analysis

The coefficient of variation which is used as a measure of the variability of the materja1sjg based on the most likely estimate of the standard deviation of the parent p opu-lation (reference 3, P. 145). It is calculated according to the formula

V = .? 1LL±LJJ., x 100 Average

where

V coefficient of variation in Percent

d deviation of the individual result "i" from the average

n number of test results

NACA TN No. 1005

The standard error of the average was calculated accord-ing to the formula:

S.E. (dj'2 / n(n - i)

The standard error for the difference between two aver, ages was calculated according to the formula:

S . E AB = ./sE 2 + S.E.B2

The difference between two averages Is considered to be significant If it equals or exceeds 3 times S.E.AB.

RESULTS OF TESTS AND DISCUSSION

Anisotropy of Molded. Phenolic Plastics

Standard test specimens and standard methods of test do not, as a general rule, take cognizance Of the possibility that molded phenolic plastics may be nonistropic in the three—dimensional sense. It has been reported (reference 4, p. 84) that molded thermosetting plastics are generally isotropic.

Visual examination of molded articles of various sizes and shapes Indicates that In thin sections long fibers of the filler are oriented in planes parallel to the molded surfaces. In thick sections the fibers tend to be oriented In planes perpendicular to the direction of flow In the mold-ing. Peculiar orientations are found around inserts and at abrupt changes in section thickness. Fibers in gradual changes of section are oriented around the contour of the part. Sketches illustrating the orientation of fibers are shown in figure 4.

D ifferences in the directional properties of the various phenolic molding materials in the form of molded cylinders, 2 inches in diameter and approximately 1 inch in length, were investigated. Sections cut from these cylinders were of uni-form appearance when sanded except that the orientation of long fibers was visible. Fibers In the interior of the cylinders were oriented at random in planes perpendicular to the axis of the cylinders, the direction of the ram motion. Fibers near the surface were oriented parallel to the molded surface.

NACA TN No. 1005 S

• F].exural specimens of ap proximately uniform size were

cut from the cylinders parallel and Per p endicul9.r, res-pectively, to the axis. Sp ecimens cut from the circular faces were discarded. Rectangular specimens were machined and sanded to uniform thickness within ±0.001 inch.

The results of the flexural strength tests are pre-sented in table III. All the specimens cut with their long axis parallel to the direction of the ram motion failed with typically brittle breaks. The specimens of the long fiber materials cut with their long axis per p endicular to the ram motion broke with tt gren stick breaks."

The ratio of the flexural strength of specimens cut parallel and perpendicular, res p

ectively, to the axis of the cylinder is used as an index of - isotrop y. The variation of this index with bulk factor of the p owder (see footnote of table I) is shown graphically in figure 5 . The bulk factor is roughly a measure of the size and the shap e of the filler particles.

Tensile and Flexural Pro p erties of Phenlic Plastics

Variation of flexura]. strength with span-depth ratio. The strength of a structure made of a brittle material usually is determined in service by resistance to bending, alone or in combination with axial loading (reference 5, p. 25), It is generally recognized that the flexural strength (modulus of rupture) varies with the material, the form of the section, the method and rate of loading, the sp an-depth ratio, and, in the case of fibrous materials such as wood, upon the size of the piece. The effect of span-depth ratio on the strength of brittle materials - for example, cast iron .ad plaster - is slight except for ratios less than 10 (reference 6, op. 103 and 106).

Few data on the variation of flexural strength with s p an-depth ratio have been reported for plastics although different specifications require different ratios for testing. Federal Specification L-P-406a requires a minimum span-depth ratio of 16:1. The flexural strength data published in the manu-facturers' data books have been determined at a span-depth ratio of 8:1 in accordance with A.S.T.M. methods.

NACA TN No. 1005

9

The variation of flexural strength with span-depth ratio obtained on molded specimens of 10 phenolic molding materials is given in table IV. The depth of the beam in these tests was the molded thickness of the sheet. Curves for BM-45,' BM-120, and. R-6565, which showed statistically significant variations with span-depth ratio, and for BM-250, which showed practically no change, are shown in figure 6. The asbestos-filled material, B}4-250, and the mica-filled material, Resinox 7013, show the least variation with span-depth ratio,

The materials containing large pieces of filler, such as those containing tire cord or macerated fabric, fre- quently broke at points some distance from the center of the beam. These failures occurred at the junctures of large pieces of filler. The results obtained with those materials were too variable to show a significant variation with span-de p th ratio with the number of specimens used.

Although no two materials show the same quantitative variation with span-depth ratio, the flexural strengths of the mo1ded and laminated plastics are usually greater for smaller ra.tos.

The large deflectiona obtained in the flexural tests of some plast-ics at large pan-depth ratios introduce consider-able err-or into the calculation of the bending moment. The reactions at the support pieces are no longer parallel to the applied load. The component of the moment produced by the sidewise-thrust of the support pieces is not considered in the method of calculating prescribed by the F ederal and A.S.T.M, specifications Also when large deflections occur, the specimen may slip and increase the actual span length. The magnitude of these errors would be less in tests at a span-depth ratio of 8:1 than at a ratio of 16:1.

Effect of varying the dimensions of specimens on

The results of flexural tests on specimens cut from- flat sheets and on molded bars are listed in table V. Speo15 were prepared from the flat sheets in three widths for both 1/8- and 1/4-inch thicknesses, The flexural strength appears to be independent of the width of the speci-men. The 1/4-inch-wide specimens of BM-200 and B?4-3510 were hard to m.aeh.jne and had burred edges. The burred edges are believed to -b the cause of the reduced strength of these Specime

NACA TN No. 1005 10

The most noticeable effect is the lower strength ob-tained with the thicker specimens of the long-fiber materials, BM-250, BM-200, and 314-35 10. It should be noted that tht effect would invalidate studies of the effect of span-depth ratio in which different ratios are obtained with specimens of different molded. thicknesses. The effect does not appear in the case of the w oodf1ourfi.].1ed material, BM- L 5, and gradually becomes more pronounced for increasinglv fibrous materials. This selectivity of the thickness effect in-dicates that it is caused by the fibrous fillers rather than by curing effects.

Additional studies of the effect of thickness and cur-ing time on flexural strength were made with BM-120, a mold-ing material which showed onl y a sli ght difference between thicknesses of 1/9 and l/l inch. S p ecimens were taken from 4-inch disks molded in thicknesses bet'een 1/16 inch and 3/ inch. Disks were molded for the minimum length of time re-quired to p roduce sound moldings and for twice that length of time. The results of the flexural strength tests are given in table VI and shown graihical1r in figure 7. These data show that minimum and double cures make only a slight difference for this material. The effect of thickness be-comes more p ronounced for 1/16- and 3/32-inch thicknesses, which are more nearly com-parable to the lengths of the fibrous filler in this material.

Comparative tensile tests with different types of sped-The data for tensile strength which are published, in

manufacturers' data books have been obtained using the "dog-bone" specimen described in Method No, 1012 of Federal SDeci-fication L-P-406a and in A.S.T.M. Method of Test D 61_42T. The use of the "dumbbell" s p ecimen described in ! .T ethod No. 1011 of Federal Specification L-P-406a and A.S.T.M. Method of Test D 638- 1 2T has been considered for re1acing the dog-. bne specimen by A.S.T.M. Committee D-20 on Plastics. The two specimens are shown in figure 4 Reports on comparative test data have been inconsistent.

Comparative tensile test results obtained with molded dumbbell specimens, dumbbell specimens machined from l/-. and 1/4-inch-thick molded sheets, and results obtained in other laboratories with both specimens are listed in table VII. Data from the manufacturers' bulletins are included for com-parison. In general, the dogbone specimens indicate higher strengths with lower coefficients of variation, The test data re p orted by the Bakelite Cororatjon for dumbbell soeci-mens are more erratic than those obtained at the National

NACA TN No. 1005 11

Bureau of Standards. klrnost without excetion larger s p eci-mens of both shapes show lower coefficients of variation. The significance of the results obtained with the doghone specimens has been ouestioned generally because of the shape of the test piece.

The most variable results with dumbbell specimens molded at the National Bureau of Standards were obtained w ith Resinox 6905, a material containing chopp

ed tire cord. The nieces of tire cord have lengths varying between 1/2 and 3 inches. In the molded dumbbell specimen the lengths of cord have a p re-ferred orientation along the length of the s p ecimen. The strength of a specimen would be high if a number of the longer lengths extended through the reduced section and low 'if, as a matter of chance, none bridged the distance. The failure of specimens of this material differed from the failures of the other materials in that the specimens were not com-pletely severed. The variety of stress-strain diagrams obtained with Resinox 6905 is shown in figure 9. The dog-bone specimen would permit most of the cords to be anchored in the ends of the specimens. The dogbone specimen would, therefore, be expected to give higher results.

None of the other materials contain nieces of filler long enough to bridge the reduced section of the dumbbell specimen, but many of the materials contain fibers sufficient-ly long to bridge the reduced section of the dogbone speci-men. These materials show much lower strengths with the dumbbell specimen. The "oodflourfilled material, the fibers of which are too short to bridge the reduced section of either specimen, shows a greater strength with the dumb-bell specimen than is re p

orted for the dogbone specimen. It is obvious from this dicussjon that an erroneous im-pression of the tensile strength of a molded p art may be obtained from tests of standard specimens.

The results obtained w ith the machined dumbbell seci-mens are in good agreement with the results ohta.ined with the molded dumbbell s p ecimens except for the asbestos-filled material BM-250. Since different batches of moiling materi-als were used for pre p

aring the different specimens, perfect agreement cannot be exp ected. The slightly higher results obtained by the Bell Telep hone Laboratory for BM-510 may have been caused by the higher rate of loading.

Stress-strain relationships.- Typical tensile stress-strai diagrams for molded dumbbell s p ecimens are shown in figure 9. These curves were obtained with stronger-than-

NACA TN No. 1005 12

average specimens. The average ultimate strength for each material is indicated on the curves. A stress-strain diagram for cast iron (reference 7, p. 356) is given for comparison. The tensile stress-strain diagrams obtained with eight molded specimens of Resinox 6905 are shown in figure 9.

The stress-strain curves for all the phenoltc molding materials are similar to the curve for cast iron and siso to curves for concrete in comression and tension (reference 9, pp. 119 and 120). These curves indicate that the molded phenolic plastics behaved like ty'mical brittle materials in the tensile tests.

Typical flexural load-deflection diagrams for six p he-nolic materials are shown in figure 10. Other re p orts 1 for examp le, reference 9, p. 122, show si m ilar contrasting curves for woodflour- and fabric-filled materials. The flexural test emphasizes the differences between the materials. In the flexural test the failure of a surface fiber in tension will produce a succession of beams of diminishing depth which will offer diminishing resistance to the motion of the load-ing device. The materials containing short fibers offer. little resistance to the progress of the failure through the beam and consequently show brittle failures. The type of failure depends on the degree of orientation of the fibers across the fracture.

Tensile moduli of elasticity.- The tensile moduli of elasticity of the molding materials are given in table VIII. These moduli were obtained with a nonaveraging strain gage and consequently are affected b y any initial warping of the specimens. The differences between the results obtained for the short-fiber materials with machined and molded speci-mens are masked by the variability of the results. The differences observed for BM- . 250, BM-200, and BM-35 10 are large enom.gh to be significant, although they are not con-sistent

The tensile modulus of elasticity of the asbestos-filled material, BM-250, as determined with molded dumbbell speci-

mens, varied between 1.64 X 106 psi for a specimen 0,2O4

inch thick to 2.22.X 10 6 psi for a smecimen 0.121 inch thick. The moduli of intermediate thicknesses fell in regular sequence between these limits. The moduli of the specimens machined from the flat sheets show only a slight change with thickness and are higher than the moduli of the molded sp ecimens. The other materials did not show this effect.

NACA Ti No. 1005 13

It is thought that this behavior can be exp1ined as follows: The lengths of the asbestos fibers in BT-250 are short in comDarison 1 'ith the length and width of the reduced section of the molded dumbbell specimen, but are corntrable to the thickness. In the central portion of the specimen, away from the edges, the fibers are oriented flatwise in the same way as in molded flat sheets. Along the edges the fibers are, oriented parallel to the surface of the molded edge. As the thickness of the molded specimen is decreased, the orientation of the fibers annroaches the laminar orienta-tion of the flat sheets. The modulus of elasticity corre-spondingly apDroaches the modulus obtained for the flat sheets. The magnitude of the change of modulus with thickness for this material is attributed to the large differences between the properties of the asbestos filler and the resin.

• .

The lengths of the fibrous fillers in PIT-200 and 3M-3510 are short in comarison with the length of the dumb-bell specimen but long in comparison with the width and thickness. Because of these relative dimensions the fibers are oriented along the length of the sDecimen and cause the moduli of the molded specimens to be higher than those of the specimens machined from 'flat sheets. The fibers in specimens cut from flat sheets are oriented in the plane of the sheet, but have a random orientation along the length of the specimens.

Flexural •moduli of elasticity.- Data for modulus of elasticity in flexure are reported in table IX. Thee moduli agree fairly well with the tensile moduli retorted in table VIII. The moduli determined on 112- by 1/2- b.y 5-inch molded bars are affected by the direction of testing with reference to the direction of arplication of the molding pressure. The orientation of the fibers in moldd bars is shown in figure 4 Since the fibers along the edges can be-

oriented both parallel to molded edge and DerDendicular to the direction of the ram motion, these surface fibers are most highly oriented. Higher results are obtained in edgewise tests when the deoth of the beam is taken ocr-pendicular to the direction of ram notion when the most highly oriented fibers are located in the tension and com-pressive faces of the beam. The effect is greater for materials containing long fibers (large bulk factors), as shown in figure 11.

NACA TN No. 1005

Impact Properties of Phenolic Plastics

Izod impact tests. - The p endulum tyDe of imDact test has been found most useful for cornoaring the shock resist-ance of electrical insulating materials of generally similar comoositjon and physical characteristics. The test is re- ported to be unreliable for indicating the relative shock resistance of materials which differ markedly in composition or mechanical properties (reference 10, p. 7).

The Izod impact strength of the six phenolic molding materials are listed in table X in com parison with data taken from the Bakelite Technical Data Book. The discrean-cy between the manufacturers' data and that obtained at the National Bureau of Standards for the macerated fabric-filled materials may depend on the capacities of the machines used. The 4-foot-pound pendulum of the machine used at the National Bureau of Standards was barely sufficient to sever the soeci-mens of BM-200 and BM-3510, although the capacity was three to four times the indicated breaking energy. The specimens which were not completely severed were left attached to the clamped portion by a few threads which acted as a hinge per-mitting the specimen to fold over out of the oath of the pendulum. The other results are in good agreement. The im-pact strengths were consistently higher for s p ecimens notched perpendicular to the rain motion com p ared w ith-those notched parallel to the ram motion.

The energy expended in tossing the broken halves of the specimens amounted to about two-thirds of the Izod im-pact strength in the case of the w oodflour_filled. material, BM-i45. The energy required to toss the broken halves of the sPecimens in all cases was oro p ortional to the oecifjc gravity of the material and amounted to 0.14 foot-mound per inch of notch pr unit of specific gravity. since this amount of energy does not include any breaking energy, Izod impact strengths of 0.20 and 0 .27, values freouently reported for cellulose-filled and mineral-filled materials, respec- tively, indicate little if any imoact resistance. These values aoly only to results obtained t'ith the 2- to 4-foot-pound machine with standard s p ecimens. Correcting the Izod impact strength for the tossing energy b y subtracting a part of the tossing energy proportional to the residual energy of the pendulum after the Izod test, accentu.tes the differ-ences between materials and bet r een directions of test.ng. It throws little light on the actual differences between the impact resistance of the materials.

NACA TN No. 1005 15

The work involved in breaking an unnotched impact bar in flexure is also retorted in table X. The work to maxi-mum load and the total work to break the specimens were determined from the ars of the load-deflection diagrams (fig. 10). The areas were obtained with a planimeter. The work to maximum load does not show any consistent re-lationship to the Izod impact strength. The total work separates the fabric-filled materials from the other because of the large amount of w ork done after failure. Hazen (ref-erence ii) re p orts that static bending tests give mineral-filled phenolic materials toughness ratings more nearly in agreement with ordinary experience, but that the test under-rates the toughiiess of fabric-filled materials such as BM-3510. Hoever, Hazen included only the work done to the maximum load, in his ratings. He re p orted a value of 0.691 foot-pound p er cubic inch for BM-3510 as compared with a value of 0.85 for work to maximum load obtained in this laboratory. The total work w as about 1.5 foot-p ound per cubic inch, indicating that the energy to tear the fabric-filled mate-rials after failure may account for a 1arre Dart of the measured imact strength.

F11ing-ball impact tests.- The results of progressive-repeated filling- ball imp act tests on 14-inch disks of 3M-120 are given in table XI. The magnitude of the last im p act in a series of impacts w hich caused failure is DroDortional to a p ower of the thickness between one and two. Since a number of other factors, such as the diameter-to-thickness ratio of the disks, the number of imoacts, and the velocity of the final impact are variable, the results are considered from a purely empirical viewpoint, assuming oro p ortionality to the square of the thickness. As long as corimarisons are made between sheets of the same nominal thickness, the exact re-lationshi p need not be known. The relationshin between the magnitude of the final impact energy on this basis and the number of impacts is shown in figure 12. This curve shows a trend with thickness similar to that observed for flexural strength shown in figure 7.

On the basis of these results, similar p rogressive-repeated impact tests were made on rectangular sections cut from molded flat sheets of the six phenolic molding materials. The results of these tests are given in table XII. The series of impacts on the 1/4-inch-thick sheets with a 2-pound ball caused failure at about the same impact energy per unit of thickness squared as that of a larger number of higher veloc-ity impacts with the 1/2-round ball. The energy required to crack the tension side of the plates apnears to be

NACA TN No. 1005 16

independent of the filler and conse q uently shows no cor-relation with the Izod impact strength. The degree of cracking required to define failure was arbitrarily chosen as the first visible crack. In the case of the fibrous materials the widening of the crack occurred very gradually. The foregoing conclusions could, therefore, be changed appreciably by a different interpretation as to when failure occurred.

The energy required to disrupt the specimens completely shows a very definite increase for increasingly fibrous materials. A comparison with the results of the Izod impact test is shown in figure 13. since the cracking energy is practically constant for the different materials, it is apparent that high Izod impact strength indicated high tear-ing strength.

Imp act-flexural test.- This test was devised to evalu-ate the damage the simens of the long-fiber materials in the falling-ball test. since the enrgy re q uired. to crack the sDecimens in the falling-ball test was practically ineDendent of the fil1r, the different behavior of the materials must be attributed to their differing ability to sustain partial failure without total loss of strength.

The effect of single impacts on the flexural strengths of simple beams is given in table XIII. Each impact value in the table represents tests on 114 to 21 s p ecimens 1/2 inch wide cut from the 1/8-inch-thick molded flat sheets. Most of these specimens received impacts close to or within the range of impacts which caused cracking. The short-fiber materials BM- 145, B14-120, and BM-6260 did not indicate a range of cracking energies but ere either completely broken or not apparently damaged by the impact. The flexural strengths, including zeros for specimens broken by the impact alone, w ere averaged in appropriate ranges of impact energies. The curves for EM-120 and for BM-3510, a long-fiber material, are shown in figure 114. The impact energies, ex p ressed in inch-pound per thickness squared, required to reduce the average flexural strength to 10,000 and 5,000 pounds per square inch, respectively, were determined grap hically from curves of the residual flexural strength plotted against the impact energy. Comp arison of these imp act energies with the Izod impact strengths of these materials is shown in figure 15.

Good correlation is observed for the short-fiber mate-rials (bulk factors less than 14) and long-fiber materials

NACA TN No. 1005

17

(bulk factors greater than )4), respectively, but different iroportionality factors are involved for the two classes of materials.

Strength Properties of Molded Boxes

The flexural strengths of secimens cut from the molded boxes are compared with data obtained on s p ecimens from the 1/8-inch-thick molded flat sheets in table XIV. The loca-tions of the specimens cut from the boxes are shown in figure 16. The smoother outer surface of the boxes was made the tension side of the beam. The strengths of soeci-mens from the boxes are in good agreement w ith the strengths of specimens from the 1/S-inch-thick sheets, except for boxes made from BM-250.

The reduced strength of the asbestos-filled material 13M-250 may be caused by a number of factors. The boxes of this material were molded from preforms instead of loose powder which was used for boxes of the other materials. B11-250 is the only mineral-filled material represented and has a mold shrinkage less than that contemolted b y the mold designer. The significant difference observed for specimens cut at right angles to one another suggests that the direction of flow from the single preform results in a secial orientation of filler in this molded box.

The breaking strengths of the boxes molded from the six phenolic materials are given in table XV. A comparison with the flexural strengths of soecimens cut from the molded boxes is shown in figure 17. The strength of the boxes does not correlate with any of the tensile or flexural strengths determined in the course of this investigation.

The failure of the boxes made from the short-fjbei materials was sudden and complete. The boxes made of the long-fiber materials showed signs of failure at about the same load which caused complete failure of the short-fiber materials but were able to Withstand considerably higher loads in spite of numerous cracks. The manner of failure was verysimilar to the failure of the flat shets in. the falling-ball test.

A comparison of the strengths of the boxes with the results of the falling-ball test on the 1/14-jnch-thick sheets is shown in figure iS. Correlation of the strengths of the boxes w ith the results of the Izod imoact test on

NACA TN No. 1005 18

standard molded soecimens is shown in figure 19, with the isotropic index as determined, on molded cylinders in figure 20, and with the bulk factors of the molding powders in fig-ure 21. The last two comparisons show a significant differ-ence between the mineral and cellulose-filled materials. Of the data u suall y reoorted in the manufacturers' bulle-tins the Izod impact strength is the best index of the strength of the boxes. The results are in agreement with the current practice of designing on the basis of impact resistance (reference 12). The strengths of articles of other shaoes which would not Dermit the distribution of the load b y partial failure v ould not be exoected to show similar correlation.

Variabilit y of Materials

Coefficients of variation as defined b y the formula on page 6 have been reoorted for most of the test results. The coefficients of variation obtained Yith small samples are themselves ciuite variable, as would be exoected. The materials containing larger nieces of filler are much more variable than the materials containing 'oodflour or short cotton flock, particularly in the flexural tests. Thinner moldings of the materials containing fibrous fillers have higher flexural strengths, as stated by the manufacturers, but at the expense of increased variability (table v). Coefficients of variation of larger tensile specimens of both dumbbell and d.ogbone types are less than coefficients of smaller specimens although the strengths shot' little change with cross section if the length of the specimen is kept constant (table VII).

In the flexural tests high results were usuall y accom-panied by off-center failures. Low results were obtained when a discontinuity of the filler occurred on the tension face of the beam at mids p an. It should be noted, therefore, that both the flexural strengths and the variabilities of the fabric-filled materials determined with the l/-inch-thick specimens would have been lower if the strength had been calculated for the stress at the 0oint of failure instead of for failure at midsan.

The largest sam p les were used for tests of specimens cut from the molded boxes. Coefficients of variation cal-culated for these results are considered to he t yp ical of the materials. Frequency-flexural stren gth diagrams for the six phenolic molding materials are sho w n in figure A.

1'ACA TN No. 1005 19

The difference in the flexural stiength of 3M-200 when tested at span-depth ratios of 16:1 and 8:1, respectively, i s 700 Psi or about 6 percent (table IV). The coefficient of variation considered to be typical of this material in l/-inch thickness is 16 p ercent (table V and fig. 22). In order to establish the significance of tie difference in flexural strength it would be necessary to make 10 tests At each span- enth ratio. The use of the five siecimens usuall y reouired for routine tests of riastic materials is definitel y inadequate for determining small differences for such variable materials. For examle, five secimens of the above-mentioned material w ould be sufficient only to estab-lish the significance of i difference of about 30 Dercent or more.

The nonuniformit y of a material, which is indicated b y the coefficient of vari.tion, affects the results of tests made with specimens of different sises and tests with difforent methods of loding. Tucker (reference l) presents n. treatment of the statistic.1 theory of the effects of dimensions and methods of loading ucon strength properties, wherein the "weakest-link" theory developed indep endentl y by Weibull (reference l4), and the "strength-summation" theory are discussed in relation to the strength of concrete beams. The statistical analyses as verified by tests on concrete beams indicate that the modulus of rupture (flexural strength) is inde p endent of the width of the beam, but is decrea,ed by an increase in thickness or length. The first two of these conclusions are sub-stantiated b y the results reported here of tests on speci-mens of different widths Pnd thickness cut from the molded flat sheets. The most variable materials show the greatest differences with thickness. The conclusion regarding the effect of length was not checked 'because it was not oossible to isolate the indeendent effect of span-depth ratio.

The 'weakest-link-in-series theory prop osed by Weibull indicates that smaller tensile specimens (shorter lengths) should have higher strength. This may partially explain the higher results obtained with the doghone specimens although the effect of orientation of the fillers and the effect of the shape of the s p ecimen may be the principal cause of the difference.

The strength-summation or r 1inks_in_ D aral1e1 tT theory indicates that the tensile strength should be indeendent of the cross section and that the coefficient of variation should decrease with an increase in cross section. This is

NACA TN No. 1005

20

substantially what was observed for the tensile strengths of specimens machined from the 1/8-inch- q nd 114-inch-thick' molded flat sheets.

In the foregoing discussion the statistical theor y has been apr1ied only qualitatively since the coefficients of variation based on small samples are themselves quite variable. The qualitative agreement w ith the statistical theory indicates that further work along this line would be useful.

CON CL US TONS

Conclusions relating particularly to test specimens and methods of test have been presented in the discussion. These findings indicate that the interoretation of the results of tests must take into consideration the charac-teristics of the individual filler in relation to the particular test piece.

General conclusions are as follows:

1. Phenolic molding materials are generally nonisotropic. The degree of anisotro p

y deoends on the size and shape of the fillers and the dimensions and shape of the molded sec-tion.

2. The nonhomogeneity of these materials is r'flected in the coefficient of variation which increases with the size of the iieces of filler and is an imDortant character-istic of each material.

3. The flexural strengths of snecimens cut from molded boxes were found to be ingood agreement with the flexural strengths of specimens from molded flat sheets of aprroxi-matel.y equal thickness. AnAsbestos-filled material 3M-250 appears to be an exception to this statement. The reason for the exceotion has not been established.

4. Tensile and flexural stress-strain curves indicate that phenolic materials are essentially brittle. Fibrous materials, however, are capable of relieving localized stress and distributing the load by partial failure.

5. The breaking strengths of molded boxes correlates well with the results of the falling-ball impact test on flat sheets molded of the same materials. Good correlation

NACA TN No. 1005

21

is also obtained with the Izod irnoact strength as determined on standard test s.ecirnens. Good corrolation with the bulk factor of the oowder also was observed.

6. The trends observed in this investigation for the behavior of standard test specimens agree ouniitatively with conclusions derived from statistical analysis of the effects of dimensions and methods of loading upon the strength proo-erties of concrete beams. It is concluded that further work along these lines is desirable.

National Bureau of Standards, Washington, D. C., July 25,

REFERENCE S

1. Anon.: Bakelite Mding Material Data Book. The Bakelite Corp. (New York, N. Y.)

2. Anon. : Resinox Molding Materials Data Book. Monsanto Chemical Co., Plastics Div. (Springfield, Mass.)

3. Deming, W. B., and Birge, R. T.: On the Statistical Theory of Errors. Reviews of Modern Ph., vol. 6, July 1934, pp. 119-161.

14 • Army-Navy-Civil Corn. on Aircraft Design Criteria: Plastics for Aircraft. ANC Bull. ANC-17, July 23, 1943, P. 5.02200.

5. Roark, R. J., and Eartenberg, R. S.: Predicting the Strength of Structures from Tests of Plaster Models. Univ. of Nis. Eng. Ex p . Sta. Bull. No. 91 , 195.

6. Roark, Raymond J.: Formulas for Stress and Strain. McGraw-Hill Book Co., Inc. (New York, N. Y.), 1Q3, PD- 103, 106.

7. Seely, F. B.: Resistance of Matrials. John 'Tiler and Sons, Inc. (New York, N. Y.), 1925, p . 356.

. Schuman, L., and Tucker, J., Jr.: Tensile and Other Prooerties of Concretes. Res. Paper 1552, Nat. Bur. of Standards, Jour. Res., vol. 31, Aug. 143, oo. 103-124.

NACA TN No. 1005 22

9. Carswell T. S., and Nason, H. K.: Effect of Environ-mental Conditions on Mechanical Properties of Organic Plastics. Modern Plastics, vol. 21, June. 1944, pp. 121-126, 158, 160.

10. Telfair, David, and Nason, Howard K.; Impact Testing of Plastics- I. Energy Considerations. Proc., A.S.T.M.,, vol. e3, 1943, pp.-1211-1219; Modern Plastics, vol. 20, July 1943,.pp. 8588.

11. Hazen, Thomas: Toughness of Molding Materials, Modern Plastics, vol. 21, Oct. 1943, pp. 103-106, 144.

12. Larsen, W. S.: Interpretation of Mechanical Data Necessary in Plastic Part Design. Product Engineering, vol. 15, July 1944, pp. 469-472.

13. Tucker, J., Jr.: Statistical Theory of the Effect of Dimensions and Method of Loading upon the Modulus of Rupture of Beams. Proc.., A.S.T.M., vol. 41, 1941, pp. 1072-1088.

14. Weibull, W.: A Statistical Theory of the Strength of Materials. Proc. Royal Swedish Inst. Engineering Res., No. 151, 19.

NACA TN No. 1005 22

9. Carswe11,T. S., and Nason, H. IC.: Effect ofEnviron-mental Conditions on Mechanical Properties of Organic Plastics. Modern Plastics, vol. 21, June1944, pp. 121-126, 158, 160.

10. Telfair, David, and Nason, Howard K.: Impact Testing of Plastics - I. Energy Considerations. Proc., A.S.T.M., vol. e3, 1943, pp.-1211-1219; Modern Plastics, vol. 20, July 1943,. pp. 85-88.

11. Hazen, Thomas: Toughness of Molding Materials, Modern Plastics, vol. 21, Oct. 1943, pp. 103-106, 144.

12. Larsen, W. S.: Interpretation of Mechanical Data Necessary in Plastic Part Design. Product Engineering, vol. 15, July 1944, pp. 469-472.

13. Tucker, J., Jr.: Statistical Theory of the Effect of Dimensions and Method of Loading upon the Modulus of Rupture of Beams. Proc., A.S.T.M., vol. 41, 1941, pp. 1072-1088.

14. Weibull, W.: A Statistical Theory of the Strength of Materials. Proc. Royal Swedish Inst. Engineering Res., No. 151,1P9.

NACA TN No. 1005 1 23

TABLE I..- DESCRIPTION OF PHENOLIC MOLDING MATERIALS

Manufacturers' designation Filler

Mean-bulk factor1 (Manufacturers' data)

Bakelite BM-45 Woodflour 2445 Bakelite BH-120 Woodflour and ôotton flock 2.58 Bakelite BM-6260 Wood.fiour and cotton flock 3.8 Bakelite BM-250 Long—fiber asbestos 8.0 Bakelite BM-200 Maàerated. fabric 9.5 Bakelite BM-3510 Macerated. fabric 14.5

Resinox 7013 Mica 2.64 Resinox 6565 Long cotton flock 5,5 Resinox 6905 Tire cord. 1 6.5 Resinox 6542 Maoerated, fabric 10.0

1Bulk factor - The ratio of the volume of the molding

composition to the volume of the finished, molding.

24 NACA TN No. 1005

?ABLE II . - MOLDING CONDITIONS USED IN PREPARATION 07 TEST PICONS

MaterielBatch

No.

Mean Bulk

Factors Molder

Typo ofMolding or

SpecimenThickness

(in.)

Moldina Clonditiensb

(07)Pressure (lb/in )

Time (min)

Cooling Time (am.)

BB-45 178 2.15 BBS Dumbbell 0.16-0.25 300 2,000 8 none BBS flexure Bar 0.114_O.20 300 2,000 18 none BBS do. 0.17-0.19 320 3,000 3-1/2 none BBS 2' Cylinder 1.30 300 2,000 18 none MIS do. 0.97 300 6,000 150 none

182A Bak-Corp. Flat Sheet, i/S and 1/4 320 1,700 15 5 G.E. Co. Boxes 1/8' 320 d 7 3 BBS Impact Bar 1/2 310 5:600 8 none

BM-120 638 2.58 BBS Dumbbell O.140.2 300 2,500 2 none BBS Flexure Bar 0. l-0. i 300 2,500 2 oem BBS 2' Cylinder 1.18 300 2,500 24 none

670 BBS ' Disk 1/6 310 3,000 1-1/2' none BBS do. 1/8 310 3,000 3e none BBS do. 1/4 310 3,000 2-1/2' none BBS do. l/ 310 3,0)0 aS none BBS do. 3/8 310 3,000 O none BBS do. 3/8 310 3,000 Ne none

7WC Bak. Corp. Flat Sheets 1/8 and 1/4 320 1,700 l G.E. Co. Boxes 1/8 320 4 BBS Impact Bar 1/2 310 5,600 8 none BBS 49 Disk 1/16,3/32 & l, 310 3,000 2T none BBS do. 1/16,3/32 & 1/8 310 3,000 V none

R-7013 Unknown 2.64 BBS Dumbbell 0.l-0.26 300 209 none BBS Flexure Bar 0.1 0.20 300 ,000 20 none BBS 2' Cylinder 1.12 '5-300 2,000 60h none

BB-6260 Unknown 3.8 BBS Dumbbell 0.l4-0.25 300 2,000 15 none 1221 BBS Flexure Bar 0.15-0.18 300 2,000 15 none Unknown BBS 2' Cylinder 1.13 300 2,000 15 none 1221 BBS do. 1.39 300 (6,000 ( 50 none

(2,000 (15 none 1221 Bek.Corp. Flat Sheet, i/S and 1/4 320 1,700 1

1221 & 1359 0.1. Co. Boxes 1/6 320 4 1221 BBS Impact Bar 1/2 310 5,600 8 none

R-6565 6519 5.5 BBS Dumbbell 0.18-0.3 00 4,000 15 none BBS Flexure Bar 0.15-0.20 00 k,000 15 none !1BS 2' Cylinder 1.63 300 (6,000 5 none

(2,000 15 none

R-6905 476 6.5 BBS Dumbbell 0.12-0.32 300 I,000 1 none BBS Flexure Bar 0.15-0.25 300 ,000 l none BBS 2' Cylinder 1.09 300 (6,000 ( 5 none

(2,000 (15 none

BM-250 1680 8.0 BBS Dumbbell 0.126-0.20 300 3,000 .6 none BBS Flexure Bar 0.12-0.14 300 3,000 46 none BBS 2' Cylinder 1.01 75-300 2,000 60 h none

191B BBS / Impact Bar 1/2 290 ,6o0 16 none 202D Bak.Corp. Flat Sheets 1/8 and l/ 320 ,7OO1 15 5

G.E. Co. Boxes 1/8 320 48, 7 3 207H G.E. Co. Dozes 1/8 320 4 a k 5 none

B11-200 678, 9.5 BBS Dumbbell 0.13-0.28 300 5,000 15. none Black BBS Flexure Bar 0 1 c-0.l6 300 5,000 15 none

BBS 2' Cylinder .07 300 (10,000 5 none 2,000 15 none

27 Brown Bak.Oorp. Flat Sheet, 1/8 and 1/4 320 1,700 1 23 & 27 Brown G.E. Co. Boxes 1/8 320 4

72 Black BBS Impact Bare 1/2 310 5,600 6 none

a-654e 670 10.0 BBS Dumbbell 0.l-0.2 300 ,000 15 none BBS Flexure Bar 0.lte_0.19 300 ,000 15 none BBS 2' Cylinder 1.10 300 (8,000 10 none

(2,000 32 none

BM-3510 16607 14.5 BBS Dumbbell 0.12-0.22 300 4,000 15 none BBS Flexure Bar o.14-o.16 300 2,250 15 none BBS 2' Cylinder 1.10 300 (10,000 (10 none

2,000 (26 none 1985 Bak Corp. Flat Sheets 1/8 and 1/ 320 1,700 l

198 a 2112 G.E. Ca. Boxes 1/8 320 4 1,160 BBS Impact Bars 1/2 310 5,600 8 none

a. Data from Manufacturers' bulletins. b. Preforms prepared at room temperature were used for all moldings at the Nation Bureau of

Standards unless otherwise indicated. Moldings were not breathed except as indicated. o. Preforms prepared at 1500?, 6,000 lb/tn2. 4. Pressure of 1,100 lb/in2 reported by General Electric Co. Since the mold is not fully

positive, pressure on plastic to indefinite. Materials preheated in an oven at 176 0 F for 10 minutes. Molded wthoait preforms. e. Powder preheated in oven for'15 minutes at 1750?, and preformed at 1750?, 3,000 lb/in2 . Bold

ass closed in 15 seconds and breathed 10, 20, 30 and kO seconds after closing. f. No preform. used. Mold was closed in 15 seconds and breathed 10 and 20 seconds after closing. g. Preformed at approximately 200 0?, 15 tons force. h. Powder placed in cold mold and heated to 3000? in 30 minutes I. Mold breathed 3 times at 3 to 5 second intervals, starting 15 seconds after mold was closed. J. Preform 5 by 5-1/ by 5/8 inch prepared at approximately 11.5 tons/tn2. k. Material preheated at 2100?, 'for 10 minutes cooled to room temperature and preformed.

NACA TN No. 1005

25

C, .4 C) P. )4 00 CU LA p4 0' IA 0 IA N- 0 4\ 0'

g o oO 00 '.0 N-N-. LA 4. ('\ 4. 4' rrlk

P dc d d do o d o a o o 0I-4 (0

1-4

04' • 0 LALA 0 '..0 '.0 N- '.0 LA LA LA LA LA 00)

N- 42 a

i- C . LA 0 --f cu N-0 P.0) t0 '.000 '.0 N- '.000 '.0 4. N- N.- '.0 N-

-P

4) 0 00 0 0 0 0 0 0k 0 - 00 CU 0 oO CU

000000000.00 0' P\ ,-4 N-. 4'

0 00 4343 CU O'0' • CU U) 0)P -. - I 0000 14 D .4

mr-I

Os Os r4 N- r4 i-I .-4 ,-4 ,-4 p-I ,-4 ,-4

0 -00 C'.) '.0

0 '.0

0 I1-

00 LAO

0 LA

0 0

0 0's

0 0

0 LA

0 00

•4 .4 0) ' 00 N- 0's 4. N-C's N- N- N- 0 00 0' r-4C1 r4 - S

a,-

bmw (OP

00 0 0 00 0 0 0 0 0 0 P. k.4 00 00's 0 4. 0 '00 00 CU 00 0

C's0 CU

0 4

0 LA

0 00

0).., '.O 0's'eO 0 U" O\0's 00 00 0 ,-I p-I CU

1-4 5 1-4 1-4 1-4 1-4

"4(0 0+' • (1) 1.0 ON 0 LA '.000 00 '.0 '.0 '.0 '.0 U's

0(L) i-I z E-

.0

tO 16 P +'..-I LAO LA LA r-I 0 '.0 LA Li's N- r'\ c"j m C4 43 ............

0 0.0) (0 '.000 N- '.0 t-156 '.0 4. '.0 N.- '.0 N-

-P4, 4'O -

-0000

0 0

0 00 00

0 0

0 0 0 0

4' 0 CU 11-sOs 000 4. 0000 00

0 '.0

0 .-I

0 '.0

0 '.0 CU

U) 0)0 0 00.4 00 N- C's 4. 00 N- LA LA 4. LA '.0 '.0 43

00 0 00 0 0d c! c! d

4)W

LA 11-s p-I N- Ns 1.0 '.000 ,-4 4.

0 00

0 4

p4,-I 0) ,-4 '-'(0

N-0 N- (1-s U'P.0 r1-s CU (1-'. 11-s 11-S

Cd

LIjP k.

00 00

0 0

0 0

00 00

0 0

0 0

0 0 0

0 -. C's r\ N- '.0 CU ON 4. N-0 N-

0 LA

0 0 00

F-F- 00 N-F-'4. 1" td's 4 u

0 a)- OF- 00 CU ,1-s0' 1's Os i-I N- 0 0

r's0' 1r4 P.411-' '.000 0 4 p-I P • ...........

bO.-4.4 . p-b 4 I-I 1-41-4 p-4 p-4 p-I p-4 p4 0,-I-

'4

42- .

W IALAO LAO LA

(0 CU CU CU LA '.0 00 C's 0 . FL I p-I

-4 0 P's '.0 LA LAO 0 CU k 4) LA 4-

CU ,-I

p-I 0

CU '.0

'.0 LA

0 C's

Li's CU

0 CU

4.UN LA ON

+' N.. I '.0 '.0 I I 'so I

C)

C) '-I

0 z

0.

0

0

0

0

0.

E-0 U) 1-4

1-4 1-4

0-I

0 "-4

0) 0 0 4) (J) 0)

4) 00- -p-i

4) P P. '4 0)0)

'4 (0 4) (4 (00 W .p4 (0 I-. 0. 400 4)

'40) '4 0.

..o

(0 0) . 00 0

14 S-,-'44' 0 0(04)40 0) 004 P. 0.-lE (0 .10) .0

4' 4 0 0 '4 '4(0040) 0 $. d

S-s 0)(00) --'-I .0P..0

040 43 (0 0. 43 (00)5.fl

0) • '0 MO E,-4.4 4)

(0

'4-ri 04) .4 05 S-ir40 -.40.00) 0)(0

0'kM '4 (00.(0 0) .4 (0 0) - .4 .0 .001.-4 p (OC) 0k00 +'a,+2p

.4 '-44).P..30 '4 •+' CU

CU odP-rIO 0\ 4-4 (0.-s

0-40 . C)4's $1)

4) 4) 0. C) p--I 40 (0040 01(0.0

(0k0ku 04 0.41 0-.-4-.--w 0)0) 0.0 0)0Tj4) . ti00 8 C) a) os P(0u)-,4.,ia, 0 .404 00.00.

0 'd 4) 0) U) (U '> 0.00.0)0.

O 640-. (0 Old) (4

26

NACA TN No. 1005

0 '.4

0. (0

P4.0.

P4, 01

c u000 N. 4 H'. t(

.4.1 Cu 0

Cu•..

cu g2 _ 1.4

1Igo +1

U'. +1

H'. +1 0' •I

IA 1?1 • Cu

.4

w -

88 s 8 8 88888 k. - r.4 .4 H'. tt It - N.

.4 _ .4 .4 .-4 .4 M .4 - m 4 .4

(0 t 4

'-a CU .4 U'. H'. IA '.0 IA '.0 IA 4

CU Cu Cu N. Cu ) 0

4 4 0 .4 Cu I- '.0

.'..0 14 - F4 .4.4

cu •08 00

00N. 000 4'0 % Cu 4 N. H'.

, •I • 4, •. •* •s •s P4 4' - W

88 888880 F- IL (A 0'. ) '.0 4 0'.

0 0'. - - 000 A; 11 Cu 0 CU 0 _- .4 r4 _4 .4 .4 .4 .4 .4 .4 .4 P4. --

4(0 04

VO

'0CU U'. IL'. H'. IA '0 4 '0 IL'. IL'.

40 U".I'.J 0 H'. .4 4 14 '0 (C'. 4 0 CU ,.'.o .- 6 .4 .4 .4 .4 .4 .4 .4 .4 .4

?N. at

-'.0 0

.4a' 4

.4.4 '-4

Cu

0 0 0 0 0 0 0 0 0

Cu.4 4

CU 4

0 H'.

U'. N-

U'. Cu

4 IAU'. H'.

.4 I.

U'.

41 •I 41 *1 •, •I +4 +1 44 ON

•I 4. U'. -

88 8 8 88 8 8 8 8 8 8 ii 4' N- 0'. F- u: a'

00 0.-4 ,-4.4.-4 0' .-4 a' r4'.O r4 r4 .4 .4 r4 r4 .4 .4 .4

%4 04.

'.0 CU '.0 U'. 0'H'. IC'. '.0 (A '0 IA (C'.

1.000

f 4 0... - N. 4 % 4 IA ma

W939 N. 'OH'. 0 F- 10 10 '0 4,4 1 4 14 .-4 Cu CU .4 .4 .4 .4 .4

(0

00. 0J 0 F-N. F-

4'0

P-130 4Cu IA

It'. N.

4 cu CU

oo66 6 6 66 6 6 6 6 6 o

-(0 Ok

U0 1101

2 u'. 28 '. U'.Cu4'0 Cu U'.

0.-I 14 Cu CU H'. '.0 IL'. (C'. 0' 0

I '.0 '.0 '.0

00

44-I 4.

0.4 0 Iid P445

140 45 OCu

0.-4 45.4 •4 . W

(04, 50(0

0_I • 0

u0 p.

OP 0

NACA TN No 1005 27

- 2P.. .4 1 . .4 40 0 0 4040 444

ji ii ii ij.0 o

cu cu 0. cu 0 cu MIC : : :: : :. : Io e 1044

4* 0010 445 444.4 .4 .4

cum, '.

cm0'0s .5 01040444 444 O1004S S7 1(4 .4 '...4 14.4 .4 0 '..a iA.4 OL* 4(4.4 .4 .4 'USU4..4

Ooo 6 o dd o 4coo o oodd 6 Od6 6 o Odd o o on

54 0 04004 US 0 04 *..US 04 144 04

01 ow '*.U'* e US040 04Us o ..S

U5 04 US 04010041(4 0 .4 "..W. 40 04 IA

04004 U5 0 04 444 04004444 0 04 '... 04 0' US 0 1(54. US oo 0 e 66d 6 6 1666 o 6 666 6 - 4000 0 0

I I *4 I *4 *4

cdd OO .04 p .4 .4 .4 .4 .4.4 .4 .4 .4 - .4 .4 .4 rI .-I .4 .4 .4 .4

4(4 40 '.0 0 0S0 0 '0 0 US 10 '0 '0 '.0 '.0 '.0 '0 0 0.

-

6 o 6 6 66 6 6 6 6 6 6 6 6 6 6 6 6 .

o 0 vw^

-

a. 0 04 US 040 . US 0 44% IA 0 .4 US 0 01 14.4 0 04 • "I.0 04 US 04 6 6

40 d

1(40' 66

01 US 04 6 e 6

444 IA 0' 6 o e

04 US 0 6 6 ..

49 144 04 6 6 o

445 44W4040 I'- 804US '.0 SO44.44114 0' .40.4USU4 49 401*4I'-l'-O0 0 P*-A14044*40 '. d.Ao6,rs • 0 .-C44 U'. r-w,--.o..r- .1 4.'044US44*.O'0 ,-.-I0'.0'.0

- .4.4 .4 04. ..4 -.4

44 .4.4 .4 .4.4444.4.4 4 .444.4 4

P- g g g g go a •'* 0

01 *4.sUS .

44% •

4.'- 04 44.4 444014.445 .4 I0'.00Cu 44.4 *4+4 ........ .4 +441 ........ '

' .'.

HH. .4;0-0- i::iI::i i i

0.'0 IA.0%0'.0SO40 '.0 44% 4040 10 '04040405040 '0 W4.0'0104010 US '.0'040'D'.0',0 US '.04050504050

44 4S

00 44*

I

T.c -

6 666666 6 0 00 0 000000 0 000000 0 666666 6 666666 0 44

8 8 8 8 8 8 4

.. 666666' . 6 66 . 660666 666666- 6666 666666

44 440 000.94 44. .N .N

04040404004040 040I040 05) 040000 0400rn00040443o 5)04045)ø5)0004 5)04a4o04oww

U 0d d a 00

u.ji or .4 04004

.4 '.0 01

I

go 00 00 00

C .0.4

iii .00100 00 004*4* 4)5) 040101 4004

044.4+1 04+1*0 9. I. og. 140

- j0 050 010 .444.4.4 .4_n .0 .04 4 . 44 0 0414 00000 44._4454 00-v 0-I. .44*4*4)

- .400 .44400

044)0.4 4*0.454 0.4 44004. 44 01

444 0 .4400 gt .054 44 0404*

.0 .6

too

44oc0 .t., -.4 0.444 C q 0 .4 0

44 04)4. 0 0 01 04 0 s-..w 0.-. 00

4;

28

NAOA TN No. 1005'

0i-I-4

iC

rx4o

E-Z

E.-.0 (/DC\J

H

PEI

E-E-4-4

e'r

00

ZC)

CD 121 w 04

C)

CID

cx4i-i

1z1

'-4

'.00 r'o.* N- LC'LO H'O.O. . D WO (T N- N- LCU 0

'-I

C'J000000 000000 bO N- N- r'\ N-C ('4 N-N- LCYO 0 Hr4CJF'\r-1N\ OJr-4C'JHr

o p +,+I+I+I+$+I

H

000000 000000 000000 000000

bf r, Nr'\C\J Lfl N- H LL.* O Cu a ft ft a V. S b

a) a) r N Cu CU (ft CU C'4H F4 bO H H H H H H v-I v-I H H H H

ccc! H 000000

O CU.000000 0 LLCU H H 0 H 0 t\ O - - a - a a a a - a

Pf 000r-,4c' HHr4 H

a)- - HQ)- (1) a) rs4ba\J -i 000000 k 000000

Wo i 000000 i 000000 -4 .rI C) r\ O' CU Q\ Cu H C) CU N- CU W r'O

a)P E CJr-4H0C\Jr-4

,.l 0)

- - a N\CUC'JHHH

HHr-4HHv-I HHHHHH E 0

•r4 -

CHW 0+'

• U) 0.0 N-000 H 0'4N-0010

00) C\JH HH'OO CJ r-4 H ZE

C\JC'J bOcd ..— •

• Hr-4

F-4 r4

o

C) . -P 0 z- OOO

PQ

r1W ( 0) .i•4 -

•ri H EOOr4

'.0 CU r-4OOO HOO.O

E4 4-4 -_. ---. 0 H v-I H H r H H H H r

v-I

•rI Il .- 0

p -40

•r4

ON

- 0

4.) +' dP1 cd

(1) •ri cd

Cd o00)00

I—I k Cd k 4.)

+) 4-' 0) Cd Cd d

d 'i Cd a) 0) d

H 0) Cd 0) 00)4.' 0

; c;.:;

0 .4.0

0 0 4) 14 4, 14 14 0 .1. 040 a) 44 0.00

040 14 14.-I 01414

0.0 1401410

•Q0k 00 I-I 4) 1414 40A 14 14.4 0.

• .4.44' •O 0 0 0.14 0000

0.00.04' 4. 0344 4)41 .4000 .4 141414.-4.4 cis 0 1414144) 144' '0.4 10.4 00140140

514141403.0 4,14.4 .4 0 141001400-4 .4 04) 4) 0.14 014441441140 .4 .4

00030 0. 04-0 0140 0 to OliVS 0.4 o .44 0

04,01)4414 U $'4)0.00 14

tO 01 0.4' 14 I-I p.01.03O1.4'

I.

NACA TN No. 1005

29

- 030 0.10 I 00% 1"%O IA.* 03 ,.

.4

0L0 0

co 00 .* ..'. 2. 00 0 0 0

14 . CU .-4,4 0.4.4p..2 I• tt.0

01.4.-I 4' 30 to M

*1+1 •,.l •.., •,., •,., •, •s.,

- 144 C) 13 00 00 o gg g eo

0.4 10 47, lAO'. 0%03 0.1- . Ft 03 14'... ,0 -

F-.1.--

101- '.0.0 .o'.o .* IA IA ir..o .-4 .4

14..4 0•

Q . 14'i %.D cu P".1- ro 000 • %0 *0 4,-v_b 4. IA l\ 10 l4. 0 '-401 .l4'. P1(44 .401 .401 .4 01.4

00 00 00 00 00 00 C,

10 0 0

'414 0.4 • 0 '.0.0 '0.0 '.0.0 '.0.0 %0 1..0 '.0.01-

014 -

03

a'..". • . •

00 00 0 3.4 1-IC'. (441'-I., 4.'

•-. ) p

.44 •l•i 444.4 •i.I

4)C) 0 U C) C) 4

.4 14.1 0 IA It'. 8 88 0 U\4' 4. IA.4 00 IA 0 1.. U'I IA 0 14.0 '.0 00030) .0 030014- P.- 1.0 '.0 '0 1'- '.0 4

14,4

I. 4) C

o 300 .0000-

0.10 034. 034' C o 0 3 0 4) .4 .4 .4 ,-4,-I

• W. 0).0

o 0 'C 0

14 4) 014

4

00 za)

pU

03

- CU CU'.0'.0 030 0'. -0%.". 0% IA 4. 4' Cu .4 4)

U14p4,4

0 '44

I 0 .4 At

LI '0 03

a -

CU0 0

82 280 l'-'.O

0 00 0 .4 0 0 p4 .4 4) ...

'.0 .4 0101.4.4.4

..4 .4 .'4454 .4

.'0

.4 .. p .4 4' 14- p4

14 3,0 *4 •1*1 •l•S *1 •i *1,1*1 *1 *1 *1*1 H •p H 4..1

0 14w - 4)

- 0 O) .-40'..4 lAP-.

0 O 0 IA

0 '.0

O0 0 0 0

O4) 0 14.4

% 0'. IAO.%0'.00 4. Iv

IAC4I.* r-..o p (44 (44*0 000

4". 03

0 03

.4 0'.

0014'-. PIH 0,0 S0 '.0'.0'.0 lAP- N\ '.0.0 P- (A 4. 4' 4'-'. IA IA IA

4)) '-4 -

I 14 014

-.1 IA 14-CU 0100 .4.40101

IA (44 go 00 40 1- 40 0 063 '0 0 • 03 .4 0 4)

4)14 400 .4 -03.* 03* . I". O%0'. 0 0 (44 0 0 • II .44 0

00 .4 .4.-4.4.-4.4 CU .4.4.4 01 .4 .4.4

40 .4

03 .4

10 .4

00.4 14.4'-' 0

• 0

• 0 . 0 0 0 co 0 0 0

•0.0

1 0 14 IA 01 0 at .4 0 '.0 IA 0% (44

1- J

I '.0 '.0 I I '.4,

I-I 0 143 0. 03

00 -4 14

No 1420 1. 4-4 00

0 XI) I-0 4-• .4

10.0

4.

.44. 0

0I-I-'

0.0 14

0 140 no

0

0. I.X C,

'-.4,

M

'-4

30

NACA TN No. 1005

H

3-I

K4 4-2

0 - .r4 z 10

a) •r3

•r4 ' o aS

p1o. ccLzt-

o i 0)

114 0+3

CO F-. H 0 H Q) 'r4

HO E-O Cl) .rI

1110)0 ,a)

114 +3 P4 00)

Q) 1-4 H

a) aS

a) 0 ad

0)11. IxlE

HO

ZP4 14 CD

H H H

5

• o N- r-I 0 H O F\ 0 N- L(C4 N- LC\ 0 CU N-¼O N- CU CU L(\H LC\\O N\Lf\ UCU * i-o O\ O NN-'.0 CU a, IC\H '.0

H H

CU - cu

0 ..-rI

000000r400000000000 0000 H -.0 (1) r-$

0 '.0. +3+1+3 +3 +3+) +3 +3 +3 +3 43 43 +3 $1 +3+3+3 +1 +3 +3 +3 +3 o o Lf H CU -

• CU N- CU H N\\0 H (Y Lf\CJ CU 0 . LC\ 0' LC\'.0 bo N-'.0 N- r'\ U) ç CU CU H CU H H ffN H CU CU N-.- r H 0 0 0 (71 r\ 0' CU '.0

G)..-4 H bLi-, H -1 H H H H H H H CU CU CU H H ,-1 i-4 H 0 .-4 0 H CU

oco 0000 0 H HHHHHHHHHH CU C\Jr-IOH000HOOCU

+3

aS - C.) CU

0)bO.r4 cd N-LC\0O H r-4 \-O CU N-NL(\LN-LC\NO.'.0

0) k a HH0HH0N\HHCUCU000'0'0'H'U)H* r-I .,-3

(1) rI ri() HHr-lr-IHHH.Hr4HCUCUHH,H000HOHCU

H -

Cs.4 CO

• 0) '.0 .* L) '.0 WO tf\\10 '.0 uO '.0 '.0 rjo '.0 '.0 N-O O .1-00) ZE-

t'0 Lc\'.o 0O4- LL0 N\N-CU N- r' CU u

0) -. 0000000000000000000000

N-HHCUCUHN-N-N-N\rk4 co'.o tc- - 0 CU . j- LC\C\J r'.o CU r LC' CJ CU u\-* O CU LC\

0) HHCUH Ht\HHCUr4HCUHHCUHHCUHHHH 0c)00000000000000000000

C)

-

a0 0' N- t\ 0 N- CU N- U\0 W 1(0 CU 0 N-'.0 N L(CU 0 N-

CUN-.- C'U LO.*CU 0'.0OOWLC\.*N\ 0' N- tf'.O O LC\ Lr'.O O N\'.0 '.0 '\'.0 LCI CU -Q . E • HHCUHHHHCUHHC\jHHCUHHCUHCUHCU ddddddddddddddcdcd

CDd .

0) 0) 0) 0) 0) 0) - d d d 'd d. d.d1rid

(3)a) •r4 0)q.4 0).-4 Ø..4 0).r4 Q).rI 0)0)0)0) ' 'i.'4 'd. d d. d. ø'cSd

p .-Io •H C) . r-40 . -4o • .-4C) • r4C) •HHr4H OCOOO(t00(d0O(d0O00(d0OOOO

00 •r4 0 0 0 r4 \O 3.4 LCt CU 0 CU 4. \40 0 H

4. H CU CU (\ U-N tO 0 +3 I I I I

•co

0 .r-4

aS

.3-s

0 k

0)0

d4-'

0)0) b •.-i

CV .1-4 4.) H U) tH

a) Li 0

0

II

aSP

NACA TN No. 1005

31

'I-, 10 0XU 0+' I1

'.4 a' 0 C-) 0 .400)+' 10 - 0 0 0 0 -1-442 0.0 . . .

+' k (d r-4 0 ,.4 r4 - r4 r4 0(6 an

0 (6-

- 'o r-cj CU 4 .-4r-4O 4- 4- ,' ,-1 4. ,-.4 N- I\N- 0 Ho

0' 00000 0' 0' r CU r -4 ,4 CU 0,01\ .itD. 0.-41414.414 00 r4r4 CU CU rll 00

r40.0 (6..-4(6.-4

I'II' 0'

Ii ii tO tO I N-

Ii .3 .O K%8 150 N-0\C) CU CU 0' 0 l . tt 10 CU L('0 1

a,0 0' 00' 60 0 0, 0' 0, 0 CU ,-4 4" 0 CU 600'

1.4(6 rM4 'D F4 0-I00.-40 00 14 ('UC'.J r4r-4 00

- 4.. ,-40 CU

+'d'- 4-04-'-4CUO r'. r" OC(J ,4'.0 0 L 60r4 -.4,414.0 O'.0000C) O\0' rC\J. ,4(4 \ r4CU 10(Y%

>0 0 4 4 P4 F4 r4 00 rl ,-4 C') CU .-4 r-4 0' o 040

'-4

'44 U) 04-'

• CO CU CU '.0 '.0 '.0 '.0 CU CU CU CU CU CU CU CU CU C") 00) ZI-.

+3 1060 '.0 '.0 '-.0 '.0 6060 6010 6010 6060 1060 p.0(6

. bO. 0'. 10 0' 60 F4 ')\ C') '.0 C') r4 l 60 (6 0'. ?-.-cu 0 r'\ 0 0'. N- 010 0 N- 0 ON 060

C I-i V. 4.4- 0' Lf'.0' Lr% 4.4. (t\ LI\4. LL\- LC'-

000000 00 00 00 00 00 4

A L- LC'.0'.0'. 04. Lt\ '.00'. N- F4 10 C") CU 0) 00 CO N- a' P '.0 '.0 r- ON 600 N- 0 00 (7\0

4.4. 4 r4 CU CU 4-4. 4. Ln -t tf'. tfi W 4.

000000 00 00 00 00 00 4

0.4 4)0)0)4)6)4) 0)0) 0)0) 4)0) (D4) (D0) -.44' U)U)CO(flu)CO (A in U) (1) (ACO U) (a (((CO 4' 4) -.4.4 -.4.-, -.4 -.4 .4 .4 -.4.4 "4 14 .4 .4 -.4.4

• 04) 4)6- (4

4,6)4,4,4,43 (6DCO(6(6(6

430) (6 .

+'Ø (660

434) (660

434) (660

43 (6

444 F4F4F4F4r.4 ,-4'd r4d r4 10 ,-4d r-4• n r.. r.. r.. r.. gi4 w rk. w fa. 1.1 r.r.i

4,434343 006)4) 004)0) (0 CO U) (0 434,4)43

r-4 F4

g g 00 0000

ø 4)0.4.4.4 .4 0(D 4)0) 0)0) 00 00)

P.O 'd I I I I V d-(6 V F4 r"4

e p.0) ,-4,-46010.*4. r-lr.4 r-1r4 ,-4r4 ,-4,-4 £- 0 0O'-.-.--. 00 00 00 00 00

U) 4F4.-4F4

'-4 (6 0 C' .4 0 0 0 (4 CU 0 (A Lt'% CU 4) 4. CU Cu re-'. \() 42 I I I I I I

a)

'-4

C) '-4

0 z

0 0

0

I,. 0

p.4 C-)

P-1

0 .4

0 0

4

4..

p4 '-4

I-.

•

0.4 4.3 4'O 0E

(((4 14

0 '4-44.3 0

6 OCU 0 r- 04. .4 +'O r-4 0-c.' k r. (60

4)

p...

4) 0) 0112 4)

4-4'4'4 0) 000.0

4' U) 43+3 r4 F4 P6 P4 (D

cc 0) OM

p. I II 43

Q) 00)00) (0 0) li.0

434)46(6 (660+).3

(6(6

(6 4. 44

ma

U

.00

Ca

00

a a .444 a .0

4. 00 44 ..4a a o .0 .044 0 4' 04

40 0

0 44.0 04' 0 0 a

.0a .0-4 o 44 - Ml. a .04.. 00

V 0 00 • 5.. 0 4.0 .0 a 4) 4.4) I-b. 4. .4 40 o a O 4. a 4.0 -0) 06too O 4.p. aa 4

4)0 .44.4 C 0) 0.4 00 a .0 0 4) 4' Q,-440

0 40 a+'0000 - a 04. 4.

.4 .004.0 0 0 .4Q0(6 .4 0 44 a4 4.0 44 0 .4 4) 1 -. 0a .0 0 44 .

o 04.00 0 0 4.. aaa,a ..E (6

0 .0 .04.4. 40 4.0 0 00 at. 044 a 6- 0 00040 (6 0

0 6-k4..-.a 4. a .0 0 04.0.44-4 ao a 14 .4 Pa4.V 0 .4 0 4) 00Ca 440 0 .0 14 .401.10 0

0 4' 04.0 6-..4 40 44 0*10 .4 0+' 0 a a .4 .0 004.0 00 4) 0

(6 .0 43

44+' 0.-I 0 05.. (6

(4 0 .4 - • 0 O 0 4.004.6- MoW 0) 0 00. 00 4..0440.0 00000 04)0 .00004 I.4'(0.0

44.04.0 .4 .40 .00. (6 04.0.0.1

6-00 4) 4.440 04.00 00*4000.4

4.0.00-14d+'

04.

CM .4.406- 4400 0..00. a 0 0.0 0 5. *46-000 .-I.4_s(6a 00'4) 4. .504. 000044 I-s 4. 5 40 .40001.

000 .(66- .-44)*4 oa EEaI.o (60.04.

0 0.04) 5444000 .404) (6.404.0.04)0014

4.0 (604.00440.0.0 .0O4 000 6.0 6-06- 00 • 00004400 00 4.4.04.6-0044 4.0000 *40Ok O (60400.0000004' .4 0 4' 4' 00 0 0 44 .0 4. 14

4. 00540 1.0001, O 000440000IOa.64a1. 6.04.0 4'.4O(6 440 044

04.. 0004.6-0.0.0 0 (6 4040*4 4. 4.4.4.0.0 .4014004.. 000 004' 04)0.404.000 00444.0 4' 4004. 0.44' 4.4' 6*4.44. 0*4 4.0004.0 440 0 044 0 0.00g.-WOo). 4.000.00.14 0.E44 V U4 14.0 04.00.00000 .4'(6444'

-

44.00 .0 0

32

NACA TN No. 1005

I-.

a .'.. -4 440

00_I o ... IA r IA

U:. UN

00 0 0

.4 06-I.-

4. 000

9 .0LAIt\

10 '.

4'

I1 N N-

IA to

d c d

06-k • 00.0 l.a a a

004 4' (04. .00

0'.0 0.-I

.4 O .-4.4

jtw .4*4

10

4.00I o i. .446-

00 00 00 00

a II 0- 6-0

A o 04)0

4., 4.

44

a0 0'.0 0.4 %00 U*.0 o..o N .4 N-I

0 IA IA .4

WLA .4.4

440J .0 Lr% It4 .* Q).0 .4.4

adc dd cd cd ddc dd d

(6.444) 00W

4 6-. -

(0

..4 N-.-4 80. CU

dd dd. dd a 6-0

00 14 00 dd

cu 44 44 6-

000 .4 a

o 0 6- 6-0

4. .4 .44-4-U\ r\r'.4)'.\

ALAIA %I'.'

t. %4"%N. I\4'\4"%

.4.4O O\' 4

Oo.r-. AWW 0.40 a 4. a 0.00.

4H,4 4.4.4 .4,.4 H•

.4.4.44. .

.4'.4.4 .4.4.-I6-44

U U p. •-. .40 N-0 '.00 (AlA 6-

IM 441.00

0.-I 00

0.-s 00 00

00C4'\ 00 .40.-4

00.-I'.O .4

0 . _so -- .. .. •. a..0,-s ..

0 a 44 I. •,+l •II •I+I •I•I I •i*I *5*4 0.

a44.0 0

N NN

mm 4'9O 4'r

0'Q0 0 .* .4 l0 40

a_s 0 t'. '.*i. .4 '.%I.ø 0 -4W .444 000 000 000 0.-s_s a (j( 490J5'.* a

-4.0 - 4'

.

4• .

44(6

H NJ NJ tLJ NN

:: 4.0

06-4'000 .000 .4 0.1

000 00 0.4

000 V0

0.4000 aua

(6.4000 00.0

000 0 00

(044.4)).4 (044.4))

4' (0.0'3

. (0.(0

a .4 %.w

(64' m%

14 *4

.

,-4 .4-4

ft

.4

ft

*4

ft a a 8 8 8 8

0 01, all Ila 119 M 112 119

CV

NACA TN No. 1005 33

TABLE XI.- PROGRESSIVE-REPEATED FALLING-BALL IMPACT TEST ON 4-INCH-DIAMETER DISKS OF BM-120.

Test No.

Thickness (in.)

Height of Falla (in.)

b Energ'y to Break Specimen

(in.-lb) (in-lb/in,) (in.1b/i&d)

1 0.123 7 3948 2893 230

2 0.123 70 .3.48 28.3 230

3 0.124 70 3.48 28.0 . 226

4 0.126 6d 2.98 23.7 . 188

5 0.170 11 5.47 3292 189

6 0.251 18 8095 35.6. 142

7 0.253 19 9.44 37.4 148

8 09253 19 9.44 37.4 148

9 0.253 16 7.98 31.4 124

10 0.253 19 9.44 37.4 148

11 0.379 35 17.4 46.0 121

12 0.379 33 16.4 . 43.3 114

13 0.380 40 1909 52.4 138

a. Also indicates number of impacts, since height of fall was increased from 0 in steps of 1 inch.

b. Energy.of last impact of series. Tests made with 0.497 lb ball 'and with edges of the specimens supported on a 3.5 inch pipe cap.

c. Cracked by impact of 2.98 in.-lb in 6 in. fall.

d. Cracked by impact of 2.48 in.-lb in 5 in. fall.

34

NACA TN No. 1005

r=i

'-I

-0

C,

0-

A. >4 U)

0 I.. U)

F-. F-. C) A. 4-4 I..)

z '-4

.14

A.

Go fn

--

c as 4.,- 4. 0 0 U) '-0 '.0 U' op ,-4 CU N- \IO\ ,-I CU r'\'.0 0.1 '.0 4. '0 .* ZE wt^ A A A

Ar- LINO" J\ L2\ all ri\

CU ON p-I 01 CU U rI '.0 r\ L(\ .4. .4 -S-0 01 ' 10 U) N- 0 0.0

a) ,0 W I4. 0 '.0 ICJ P'\4. N' L('. r-4 U) 01 U) ,-4 N- U)C r-I CU U I p.4 F'\r.4 CU U) . U p.4 p.4 p.I H.M I go 0 0 C)\O'. 0". 0'-. 01 U) LCW '.0 CU a-. r

a Jr!i FLO!J J\ ) r!4t L2-j-(!) ,L r'\8 '.0 U' N-N-LC to N- N- CU Cu '.0 if'. N- 0.1

CU r-I i-I p.4 ,-I p.4 CU r-I p.4 N-4 .4. ((\ 0 '.0 p.4 p.4 1-4 - p.4

43

MU) It,) Ig tf\C'.J N- U) p.4 N- '.0 CU 0 LfO a' 0 0 U) 0

14 A m .4. N-N- U) U) N- .4. to 0 CU--& 0 0'. 60 zt 4-a i o CU r-I p.4 p.4 r4 p.4 f'\p.4 CU U) .4. LC\ LI'. p4 N- p.4

-4 -4 4 •4 1-4 1-

i.

U) -' 43 60 CU

CU U) N- fl- '.0

00 10 N- U) r\ U) CU '.0 p.4 CU CU U) .( I I0 I- Lt

.4J\ :rL z CU 01 p.4 CU 01

N-rI N-,-4 zt

: i-I a) - E C) CU P I '.o r-a-- 060 N- U) CU 0 01.4- CU N-.--44- 0'sCU U)

00 10) 04 f%-r-- U)U) N- r-I Lf''-O f'.".r-I '.0 ON r4 ,-4 0 0'. 0 04 H H p.4,-I p.4 CU p-I p4 CU Cu rI I". CU CU -CU H CU J!4 i J\C!)C!) C!4CJ

( gx.,iL.i N-4 LI'. LC+.0 LI'. a'F'\CU 0 N-4' t'.0 0'. 606060 H p.4 p.4 p.4 p.4 p.4 p.4 p-I p.4 CU r-4 r.4 i CU r'I p-I p.4 p-4

0 4'

hO) 'U) - U) U).0Iu)

-4 c:: .4. C'.) N- LCO r\ 0.1 N- r1 \ N- Lr'.4- '.0 U'104- 0'. 60 I-. I '.0 '.0 N-'.'.O 04.4. CU (7'. if'. N- 00 0' 60 0'.

F-4 p.4 p.4 p.4 r-4 p.4 CU I p.4 CU p-I p.4 t ('U ('U r-I ,-4

p-I rI 60 N-N-C'. 4. .'. 0 60 L(\U) '.0 CU 0 ON if'. O'.

,.4 CU CU p.4 l\F\ i-I CU CU p.4 CU CU p-I CU CU H CU ('U a)e . . . bO. 000 000 000 000 000 000

C'.C'. J'.0 4d '.'.b - CU '.0 If\ r".oZ, r'\'.o

I.4. CU Lf\ U". 4-4- LI'.

N-- - N

Cu p.I CU CU p.4 $\

k4 -L J'.

4.4 4 p. CU CU - N p- CU CU p.401 p.4 CU C'.) 0-.-4

C) 000 000 001.0 000 000 000 (ho) -

• 4. '.0 '.0 4. p.44 N".N-'.0 N\'-.O '-.0 4. LfV.0 N. If N-v-I N-'.O 0'.0 .-4 6004. ONO 0'.4- 4 r\ 4. N- N-

o 14 4 p-I 01 CU p.4 4V". p.4 CU ('.1 H 01 01 p.4 C'.J ('.4 ,-1 CU 01

-,-I U)_4 I . p- '- 000 000 000 000 000 000

LC\N-N- L(\C'.J p-I .4. 60 t4\¼0 N- 4. U N- t'.O N-

p.4

0'. a-s N- C'. O'.N- 0'. 0". N- 0'. 0'. N- 0'. 0'. N- 0'. ON N- B M 0 4.4.0' 4-4.. 0'. 4-4-0'. 4-4.0' 4-4- a'. 4-4-0'.

.4 i-I00,-I 00 p.4 004 00'I OOr.4 00 p.4 in 0

43p.4 ' N-N-N-N-N-N- N- N- N- N- N- N- N- N- N- N- N- N-

I;;1 0 0 .4 0 '.0 0 0 14 LI'. CU CU U". 0 a). 4. v-I '.0 CU CU 4' I I I I I I Id

r. U) U) U). U) U) U)

I 4-4 '-4 >4

N

420)

a)4rI E cu :w 140)1-. 'i-. Rt Q)

4-,

O) 0) -.-4 00i 00-,-4

.4 U) r. (0

0) a)• A3 14

0)4' 0 a) .tjo 0) (a

p.4-I-,. wrI. .0 40 U) +3,0.443

0)0) r4.0 a)

40a) 4) tI4' 0.0 41)01-40 +3 v(0 1-44 a) 0 U) 0. 014 p. •wo a) E'11 U) +'-.-4

tO 0'ci a) v-I (I) 0) 1-. () U);i

r-4

41)0)0) U) 0) .0 Ii a) 4'+) a) .00) 0 t..

0 )4 00) -.40 0

Ii 0) 01,0 WE

+34-'.1-4 p.41 00 I .0 (1) 4))

4. 0 00. 0 (0

C) •0 1.00) 60+20) CU'--.. 14 '--- t4-\ 0 .11 '-4 OF-. ,0

00) U) 0 0.,0 0 0.0 a) bD 0 o E 014.1-4

04' p-I 0)0)0) P4 a) (0CM

0) 1-. 0

-

40

NACA TN No. 1005

35

cu Ia a)

a o :: •. .;j NC.)

-0 r4 o-t c 4) .) 0 CD 10 N 11) CD

o 0 N N CD 0)r1 r4

-P0 - •dt -I

-iQ) d -

1 4-' -

C) C\i r4 a) es %J u

bO-I1

dJa).00

*D4-i .c f' 1') 0 0 •C\) 0 043 (0 N N.0)0 02

-P D- r-I r4 o OD Id f-I P4 01

H

a) o 0

• AS 0 r4 '4 r4 c-I

j Cl 0 o O. 0 -P r4 rIO 43 Cd

• c 4-443 U) F4 d U) U)

i-I 0 00 • 0 0 •,-I 0 00 0 U) Cd 0

ø .*

Cdr-14)

'I) w

U') 0 0 10 10 • • S • • •

p:Cd 02 02 t) CD 0) "i' r1

r-4 0 cd CD 0

•r4 0 02 0 0 r4 -. Cq O 10010

a '44 r4 I 0202 tO 4) I S I I I Cd

m AD M ø rQ

0 z

I-I C,)

0Cl)

PO co Izlt-4

rz1

J4

E4 00

'-I

0

C,)

El

C,,

I-f '-I '-I

p4

E-4

36 NACA TN No. 1005

0

..4

0.i r..

0 0.4

I-I op 1(4.4 CuP. 001

0

COP. :u)

4..

F. U)

r 4-I

8

p. '-4 1.4

F.

) Cua'0'4IL\ .*l(WIA 4U'4.40 0".4P-t-- w 110i 040 %M'.0'.O f,%UN,4%D .IO'4I'. .4 .4- - .1 .4 .4.4.4.4 .1

0 '4.4 0000'

00 .4.OF-.4L4

0000 P$'\Cu'.0 0000 '0E-Or- 0000

04"O 0' 0000 cu PIN •-. C%h-4.4 .1.4 .4 CU ('U 4(4.4 4 CU 1%.-4 .*4 4('%C'U r'v Cu 11

)0 .4

•I.I.I+,.,,I.I.,., ,,*I.i,, *141•1•1 •I•i•I•I •I*,.I.I 01 14 - 88888888° 8888 00 0000

.000000,\8 00 0000 8888 ,".r,.ou-. cuo'.o'.o '.oioo Cu Kkt CU $'.W\ VOOMO US- 44I'4' 11 14_I.-l.4.4,-I.-4,4 r4 .4.4.4.4 .4 .4 _4•4.-4•-4 .4.4._4r-4

1440.0 c!c, C?)000 d4 Cl .-4 0 0000000 CuCUIe'.CU44O\CUo.s o4(-A$Cu 00 or-oo ncs.'.

0.40OC'UCuOOjO'Pa •c'.aso ou- 00%.l go

. 000040 .4.4.4.4.4.4.4.4.4 0 14 r4 r4 01 -

04

888888888 8888 00 0a' 00 0000 0000 8888 40.4 $4".. a-1'4lo0IrN.4 r4 or-0 .'asooa

01 410 4CUCUCu l'4t". 0.4_I_I '.000r- C'.-4 4. I". CU \ .4

p _I .4.4.4.1.4.4.4.1.4 .4.4.4.4 .4.4.4.4 .4.4.4.4 -

o 0.4 • 01 4*44 W4'44 0.4 P-4040 I"- 40 .4 01I 4$4.bJ.11 Ol...0

$i 001 .4. _I 004 4.4.-lU". '.00 24'. p4..4 102. or-.4'..o ('.4 IC'. ZI.

- c'..* ,_a'. a' '.

01 4.-4.I.I.4.-I.4 .4.4.4.4 4.4.4.4 i*('.a.* .4.4.4.4 4(\*r .4.4.1.4 • 000000000 0000 0000 0000 0000

U)01 JsU.JJI&I&,L,L 140('U.4.4e'UlvCu.4.4

,!,Jj'. J'.I '.J I44 . .-4.4.4_I.4p444p4 4("1U-4.4 .44-4.4.4 4("CIaQ0 r44.4.4 pl".I(\.1.4 CU4('.PI('U

dddcddd dccd dcdd cddd oodd 0

40 .4 .".0 CU 0'00.4 O'Or'.-.4 r. '.IAOO' M LnmN $411

p4-.4"1.4 4(%4(".1%Cu 44-'. .I.4.4.4.4.4-4.4.4 ".i4-'.(%I P'. .4.-41I.-4 r\* .4 P4-i .4.4.4.4 I4-'.4('.CU CU .4.41114 1\P4-4Cu 4(\ .-4.4_I14

-

44.4

L000000000 0000 0000 0000 0000

4' .1 0 14 ,. u-s CU 0.4 .4

C" >4 0 0 '.0 41'. 0 8 0 0 40.4 .-. tV 4 4

Cu 40 '.0

0 '.0 fl *4 •. *1 +1 •

.4.4

00 0 0 0 0 401.1 0 0 0 0 0 0 .4.4 4 4 .4 CU 0'. 010 01

'• ('U . N- 44'. 41'. .4

'.0 .4It'. .1

M40.O 0 O ( 0.-I .4 '.0 44'. 44'. 40 Pa .44 .1 CU 0'. 0'. 40 0". E-..

If .4 I -

00 0

0 0

0 0

0 0 ( 40.4 0'. CU Cu '.0 .4 0 • 001 .4 4 .4 ('U CU Cu

14 .1 .4 U--kS401 4 0.4

.01 N- 0 40 40 CU ('U 0 001 .4 4 14 .4 ..-I .4

-

o N- is-'. 4'.0 44'.

4 01 .1 .-1 .1 .4 (-4 - .4

d d d o

01

In

CU 14 .4

CU 14 Cu 4- CU

0 d d d d 0

.4 ('.4 0. ".0 4 1440 .4 .4 .4 .4 .4 .4 01.4 4.-

.. 0 0

. 0 0 0 0

(0 .4 (0 0 0 .4 0 '.0 0 0 0

.4 IC'. Ii

0)U'. Cu 4 .4 CU

14IC'.. (U CU 44'.

0 04 01

U01 04 0

.01.4 44 .44 04:201 .OP4 0

0044 •4( 40 0 010114.0

• .040 01

CU 044.-bOa 44

0 o

.4.44) .4 0400101 40 401 I.. VOl.40 04 p.401..

.M 0$414liO - 00144

(40.0.11.

0400101(4 44) 00.4.4 11 .4 4$4) .4 010104001 0 (0 (4 s. t4) 04 O.411.4 I. 01004W O 1401.-4.--4'.D4) 11 1114 .4 . 0.01.4.4040 11 40 •''r-4.0 0(4 (40404E

40 •11 040 $4 0101114001040'.-. 0.00 (41414.0 144'.-44t'..40Q 01 o ..t.4.0.O'.-.

ow B .4 44U).4000-.4 14(4 400 VO 4.'.. 11001001

4114 0004W.'11. .414. 4''.'. 01 41 on .4.4 0 .4 44400140.0,-.

'-4 4.1104144040400 01(40.4(4000 44 0.4 (5 4. I. 4. k,_ 40400 '.-.'.-..,-.o 40.4'.., 1.. 0+' 0 (1 15 (5 01 4)

.411111144 (0 0144)00404.4 .4 011.440014001 1.. U.4.4.4.4.-.0

p. 0000. 4' 1(4 II 1.. 04 04 4) 0) 40 .0 40.4 X co

4

NACA TN No. 1005

37

C')

0 U)

0

z 0

U)

Cl)

C.,

Ixl

E-4 Cl)

fri

U).

- 0 CU w 4-a)— t-4 O 4- N-bo. CU ,-4 CU CU

.0'-4

I-

I 1 o CU '.0 CU .%_• O O 0' CU CU

>. .0 cd '-4 r-4 CU 1 m

'-4 004 a)_ C)

a)r4(d-q Sd NP 0 O v-4 0 O r

() r1 D CO a) I 0 O r-4 \ ON o'— CU I CU CU CU f\

cri Id .9 0)4'

440.,i Or 04'

0 • r r r\ ,-i r 00)

00 42

la 0 +21 '.0 N- r-\ 0 CU 4- 0 43 Q a) F— O O 0 CU 4- 0'

bD. CUr-i CU CU

d .44 c!i ca A 16 - N- O N- 4-0 N- LC\ (0 ca rl, 1 H CU

0 00 404 424.)

0) .r4 0) 0 rrlcd • CJJ

Cd 0) r-4 N-4-

O

rl W N- r-4

O 4- N-

0) rl r-4r.4 CU CU

cdc I4 0+'

bbo1

-r4 4 4-1 V) +) 04'

• U) N\ r4N r'\ CU N\ CU F4\ 0 00)

W. E-41

rio CU 0 0 NIN 0 CJ '.O Cd a— 0 r CU '.0r CU 1500

CD bO • i—I i-I i-4 r-i r-4 CU CU CU

CICU Cib Cd — W i-I 0 r\CU 0 r0' o r'4 r-i rlrl (U CU r-4

W t9

004 td 4 .0 O ri N\ i'-a t 0'

cd 0' CU r-4 4-CUrl 4-0 4-31 4) lip i-I i-I r-4 F-4 CU CU CU

•rl 44 U) 0+3 . (n F '\ ON rr N\

o 0>

C)CU

C) r-I0' LC\ CU N- • 4' 0

cdZ O4- 4-

CU cu n. 00

N- r\ cu cu 150r-4

U) i-I N- r4 .CUCU O\r-I .-ICU

0 0 -rI 0 110 0 0 P4k LC CU CU LEN 0 u 0) 4- r4 '.0 CU CU t\ 42

cdI I I 1 I I

i-I 4'

o a) 0 •-. c

r-1 a) Q) 1) 0 .

C) 0 4-' ci, v U) -.1 0) a) e .i k >-C) a) 0) 0

,L bOçi.

1l) E 4-'r4 a) . -

bOi0 U). U) -_4 4-'

, cd

0-0 (1) U)

a) ..-iO 0 4' bL) co ll 0 42

0P-.4-' Cd

,c'I-I

ri- C) '0 0 0.0k 0 0)

a) -I 43 0 0

(Op. CM a) C) r

+2 4.' 0) +2 0

'0 (D 0)

U)Cd0) --.4 ca4.'

Cd 0) -'-4

O '0ka, .

- Cd

Cd42

-p4 0) - -'-4 E )

42 co o C) ir Cd

0 .Sd '0 0 k

'0+' Cd 0 k S.4 0)

.r4PO C)Cd

0) '0 r4 rl ,r • a) rI

Cd0-CiCd

Cl)

a) ''0

C) 0 0 '004)

4

to C.) a 0)0k i-I .r4(4.40 00

r4p4.p +'ooE t Al 0 0

a)OCO k-. '0-rakOk a)IfC) Cda)a)O

•.4) Ii

4) .0.04) 0a)a) I'4'0.0 I4cs4)<p4)4 0 0OC)0P., U) 4)Z4.'U)PU) Cd

NACA TN No. 1005

Fig. 1

0

Figure 1.- Flexure-teat jig.

NACA TN No. 1005

Fig. 2

Figure 2.- Apparatus for falling-ball Impact teat.

NACA TN No. 1005 Fig. 3

Figure 3.- Breaking strength tests on molded boxes.

0

-43 o

C) -4.

Cd

I;a)

I I'-4

ill =

tub I'll

Ii

Ii 'I'll

iti Ii

1

C 43

(d •r4

a, 43 U)

(4.4

Cd 0

".4 0

r. 0

P4

C) a)

Cl)

4-;ict 1 • 11 IIII

?fA':i'

IF

%II(t

MII -11,1111111111

ii 1.' ((III I I

LI

a) E

P4 C) U, a U)

a) rl P4 (I,

a) 4'

'-4 rI

a)

r-4 0+'

a) to

U) 4 P4 a-

Cd (Ii43 OW

0 Al '4+' 43,4 0 It a'

CD

I

0 P4 43

0 E

4.

bD

'4 lmd

0.

0 (4.4,4 04'

Cd

•1-4 U) 43,4 Ok a)0 k

'dO

Cd dk

C) '40)

P4C) •1-4

0'4

04 tdcd

a) r-lk CdCd

XH 4' 43)

MO

I U, a) a CIS

U)

a,

'-4 0 La

'-4 Cd

P4 C) a) a. U,

•1I

U) 1.4 0) rl '-I •1.4 (4.4

U)

0 k

'a •1-4

"-I

"-4

0

0)

0

43

Cd

U) '4 1-. 0

rl Cd C)

P4 a.

NACA TN No. 1005

Fig. 4

Q) E

•1-4 _ -- Th fl

1 11 1 1,1 C)

(t a a)

\\ IIIij a)

7 a)

v ii 1-1 r1 a)

,2 & II'i 1

II \\) I? tII

L iIIItl U)

ir II 0 I:IIId •c

0

- U)

O

/ I I II-

tIduu;:;u" C)

U)

'-4 0

a

MEN2-Som_____

MkONNIMMEMMEMEMMENOMONEE

MMEMETIMMENOWN __

11011-

1OC

8C

$

6C 0 .4 4 2 0 N

1.4

4C

2C

NACA TN No. 1005 Figs. 5,6

010 13

14 Mean bulk (actor ( Aufactiarer's data)

Figure 5.- Anisotropy of 2-inch cylindrical moldings.

MENM IN Ubi_

MMI

lull MIMM

rn_

111 MOMM

M11MM111111___ IrI M11MMEMI,

11 liiMMMIAI M MMMMEMIMIMMMM U _

%J .4 .4 74 A .4 Ted ot 'q2ua.;e Tvzrxau

c .4

'I

4

p.

D

4 4.. .4

-

4-inch disk moldings of B2-120 X Minimum cure, Batch No. 7440 - o 100 percent overcure, Batch

No. 744C Minimum cure, Batch No. 870

0100 percent overcure, Batch --

No. 670

I

13

-4 a

0 ' 12

I..

a

.; 11

a

10

-4 a

a a a

a a .4

NACA TN No. 1005

Figs. 7,8

lflb afaa 1/5 114 3/8 Thickness, in.

Figure 7.-.Variation or flexural strength with thickness and cure of moldings.

Strain, in./in.

figure 8.- Tensile stress-strain diagrams for individual specimens of Resinox 5906

NACA TN No. 3005

rig. 9

1• co

Tod Lot 'g STISVOL

0 co LV co 0

UOZT II I

14 6mo jo; Ge0

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

-------

-S

-

Ix

--

mS.-

.- I

9 43 a --

In OU).4t N. g we-

4r)

- --a-

--s-

-

oeetd .to; O0g I I I

o o k .4 -4

td U C) .4

(0 C) U) 06 a)

a) -4 -4 a) a a, 4.

C) -4 -4 0 a a,

a a.

.4

a -4

4. a)

a) I

C.) 0) -4 0) 4. U)

&Q C..

0. (C) a)

0C•. .-4 00 -4

• 104 Q) -z- a

U) 0) - 4•

Cl) d U

0-.

0)

Cli

led 201 'ssa.t,s etTsue.L

o 0 Oc000.-4

U) N N 2 0 U) .-4 CON t) r)

, S S I IS

SI till

7i

U) 14I -.

_____

d CO

10-4 0 --- -----------------

so 1p

- I - • LI) 4

as

-

C)ld-

14

o I44 0 a) —4-- --------

Is

-.4--.- ---------------- -

IL)'d aa

0,4

- _;;= T 0.0 -.--.---

4O1

v g•4

o.o

-----

- --

-

- ----

- --

-

U) 5;; .4 14 a,

4..

C)

—4

-4 0

V

0-

-4 0 a

I-. 0 5.-.

Ca a 1.

C)

0 -4 4., C) V -I 5.-. a,

0 0 -4

'-4

C)

0 -4

I,

NACA TN No. 1005

Fig. 10

8 0

-4

qt Ipo1

NACA TN No. 1005

Pigs. 11,12

111111111 I (odulus ratio, percent

_

fle al modulus of elasticity , i2rtwise x 100 flexural modulus of elasticity, edgewise

71atwie - molding direction

- * fl - f--

0 +1/2 in. -.---- K

--

-

- ---

-5---

-

0

80

CY---

b 7 9 11 13 15 Mean bulk factor (manufacturer's data)

Figure 11.- Effect of direction of molding pressure on flexural modulus of elasticity of molding powders with various bulk factors.

300 -- -

N

26(

q)

U

22(

S.--4

.4

18( ba '4

C

C) I -1 -4

'U

10 15 20 25 30 35 40 Number of impacts

11

13 101

8

7(

10%

Tests made with 1/2-lb steel ball on 4-inch disks of BM-120 mounted on a cast iron support.

OAverage breaking energy for in-* Two points dicated thickness.

DAverage or energy.

-

--- ---xTbroe points

-4" -- .._

)igure 12.- Relation of energy of final impact to break in ball impact test to number of Impacts.

NACA TN No. 1006

Pigs. 13,14

l6OC 0

-4

0 -4

4

. 120C 0 .4

4.

'-4

. 80C

to

'S

V

4,40(

It '-4 m

----------_- -------

-

- ---

---•--.- -

-

-ir17ei sheets - - 01/2-lb ball, 1/4-in. sheets A 8-lb ball, 1/4-in.

Plates were 3.5 by

sheets

4.5 inches

- -

-

- - --- --

I I - 0 .4 .8 1.2 11.6 2.0 2.4 2.8 Isod impact energy, ft-lb/in, of notch

figure 13.- Comparison of energy of final impact to break plates in falling ball impact test and Izod impact strength. -

20

16

-I 'S PS

r) 0 -4

4.

'S

4-'S

V '-4

0 20 40 60 80 100 120 140 Impact energy, iflrlb/in.2

Figure 14.- Effect of single falling-ball impacts o. flexua,i. strength.

--

i - - -- -- - -

Weight of ball .15 lb Specimens 1/2 I 1/8 Distance between supports

XflI-35lO 0 DM-3510 (averages) a 120

I 5'3' - -

'--. - -

---^I k-

- -

-

0 -

- -------------- _____ -BM-33:0---

P

3M-626

BM-45

BU-250 - -

1- * Reduction in strength to 5,000 psi XReduction in strength to 10,000 psi

16(

Ce

.4

.0 I 12(

.4

4.

8 4. U) 4) 4.

4

NAOA TN No. 1005

Figs. 15,17

0 .4 .8 1.2 1.6 2.0 2.4 2.8 hod impact strength, ft-lb/in, of notch

Figure 15.- Relation between single blow impact energy required to reduce average flexural strength to a selected value and the Isod impact of phenolic molding materials.

500 -

400

•0 -4

U) 4)

Ox 300 .0

0

U)

r200

100

0o VLV J. id 13 14

Flexural strength, 10 psi

Figure 17.- Strengths of boxes compared with flexural strengths .of specimens from boxes.

Strengths of boxes were determined by loading with a plunger having a hemispherical end of 5/8" radius.

Rate of head travel was .05 in./minute,

)( Loading at molded hole on bottom. 0 Loading at point on bottom symmetrically located

with respect to'molded hole. Loaded at point "0" through rubber cushion.

- - .Aa- -

Strength tests made on one lot of boxes; flexural specimens out from a different lot. Two lots

- molded from different batches of molding powder. -

NACA TN No. 1005 Fig. 16

Figure 16.- Locations of specimens cut from molded boxes.

I , I I I I IStrengths of boxes were determined by loading with a - plunger having a hemispherical end of 5/8-inch radius.

--

-- -

Rate of head travel was .05 Inch/minute.

X Loading at molded hole. - 0 Loading at point on bottom symmetrically located with

- - -

respect to molded hole. - _jojg j_pin_' throu_ rubber_cushion. - ;;;;:;.

-----

400

.0 -1300

2 14 0

• 20C

7. 1OC

NACA TN No. 1005

Figs. 18,19

0 2 4 6 8 10 12 Falling ball impact breaking energy, 102 In.-lb/(thickness)2

Figure 18.- Strengths of boxes compared with results of falling ball impact test on 1/4-inch-thick flat sheets.

Soc --40C

.0 -4

a) a)

30c .0 Ii 0

4,

20< 0,

10(

0 .4 .8 1.2 1.6 2.0 2.4 Isod impact strength

Figure 19.- Strengths of boxes compared with Xaod impact strengths.

I I I I I I ... Strengths of boxes were determined by loading with a plunger having a hemispherical end of 5/8 .-inch radius. - -

-

Rate of head travel was .05 inch/minute.

- X Loading at molded hole on bottom. - o Loading at point on bottom symmetrically located with

- -

respect to molded hole. Lc nt 00 through rubbercushion. .- - -

-

-

-

- - -- 0

-7 _x . -

- --

0

NACA TN No. 1005

500

50

10

0 30

01

Figs. 20,21

Strengths of boxes were determined by loading with a plunger having a hemispherical end of 5/8-inch radius.

Rate of head travel was .05 inch/minute.] I XLading at molded hole on

bottom. 0 Loading at point on bottom

symmetrically located with respect to molded hole.

A Loading at point "0" - through rubber cushion. -

a Asbestos-filled material. 0

-

-------- -

.-x a

-.-- ---

Ra

------

bO 70 50 90 100 Isotropic index

Figure 20.- Strengths of boxes correlated with Isotropic index. FT-1 i I' . Ill Strengths of boxes were determined by loading with a plunger having a hemispherical end of 5/8-inch radius.

Rate of head travel was .05 inch/minute.

X Loading at molded hole on bottom. o Loading at point on bottom symmetrically

located with respect to molded hole. A Loading at point "0" through rubber cushion, a Asbestos-filled material.

- 0-

a- -- - - - .

.--

- —

---

---

X-

- a

- oo __ - 0

-' 1i 11 13 15 Mean bulk factor (manufacturer's data)

Figure 21.- Strengths of boxes correlated with bulk factors of powders.

40C

.0 '-4

'a30C

'a .0 4. to200

4-U)

100

40

•0 -4

'a 'a 0 .0 "I 0

0

4, to

20(

4, U)

NACA TN No. 1005

Fig. 22

V Coefficient of variation

10

20

10

20

0Q

010

0 20

0 0

10

Cr 0

20

10

20

10

2046 8 10 12 14 16 18 20 22

Flexural strength, 103 psi

Figure 22.— Frequency—flexural strength diagrams, specimens from boxes.

BM-3510' Bulk factor 14.5_ I = 13.3 percent 56 specimens

--BM-200

Bulk factor 9.5 T 16.7 percent 56 specimens

BM-250

• 1Bulk factor 8.0

16.0 percent -

54 specimens

BM-6260 factor 3.8

r 10.5 percent -Bulk

r.54 specimens

A BM-120 Bulk factor 2.58

= 6.5 percent -

__ --

57 specimens

F BM-45 factor 2.45

= 6.9 percent - -Bulk

57 specimens