1 CIRCULAR 214 OCTOBER 1974 I II Agricultural Experiment Station/Auburn University Auburn, Alabama R. Dennis Rouse, Director TABLE OF CONTENTS INTRODUCTION------------------------------------------------------------------------------ PROCEDURE TFCTTN(fl_ -------------- ---------------------------------------------------------- OBSERVED AND PREDICTED PROPERTIES ----------------------- 7 13 14 Flexure--------------------------------------7 Edgewise Shear----------------------------------------------------12 RESULTS----------------------------------------------------------------------CONCLUSIONS----------------------------------------------------------------REFERENCES----------------------------------------------------------------- 15 ACKNOWLEDGMENT The authors acknowledge with gratitude the assistance of Mr. Hoseid, wood Company in Cedar Springs, Georgia and Mr. Charles Hamilton, plywood operations manager of Scotch Lumber Company in Fulton, Alabama for facilitating fabrication of all experimental sandwich and plywood panels in their mills. vice-president and general manager of Great Northern Ply- Ward FIRST PRINTING 3M, NOVEMBER 1974 Auburn University is an Equal Opportunity Employer Flexural and Shear Properties Of Structural Southern Pine 3-Ply Sandwich Wood Panels EVANGELOS J. BIBLIS and YEN-MING CHIU* INTRODUCTION used in furniture primarily as non-structural components faced with decorative hardwood veneers or plastic laminates. However, the authors are proposing a structural sandwich wood panel with a particleboard core and softwood veneer faces for use in construction of houses and for other structural applications. This structural wood panel combines structural efficiency with favorable manufacturing cost. By utilizing lower quality wood such as residues from other wood industries and species in less commercial demand, it also helps conserve our forest resources. Ordinarily, sandwich panels are constructed with faces that are stronger, stiffer, and denser than their cores. This provides structural efficiency in flexure and increases strength/weight and stiffness/weight ratios when these characteristics are important factors in a design. The proposed structural sandwich wood panels are nonconventional only in that core densities are higher than those of the faces. This adds some extra weight to the panel and proportionally increases transportation costs and, perhaps, handling costs on the construction site. These disadvantages, however, are offset by favorable manufacturing costs in comparison to plywood, by higher shear rigidities and by the important factor of contribution to conservation of forest resources. * Respectively, Professor of wood utilization and former Research Associate, Department of Forestry. S ANDWICH WOOD PANELS with particleboard cores have been Structural wood panels for floors must support primarily uniform and concentrated loads applied perpendicular to the plane of the panel with face grain orientation perpendicular to the direction of supporting joists. Structural wood panels for wall and roof sheathing must resist both flexure and lateral shear loads in the plane of the panel. Thus, flexural properties of sandwich strips were determined representing three types of sandwich. A flexural analysis, applied previously to plywood strips, also was used here to predict flexural stiffness of sandwich strips at various spans. In addition, shear plate tests were conducted for each sandwich type to determine the edgewise shear modulus. An elementary method also was applied to predict the edgewise shear modulus of the sandwich plates. For comparison, flexural and shear properties of two plywood thicknesses also were determined. PROCEDURE The following sandwich panels (4 feet X 8 feet) were fabricated in a southern pine plywood mill: Sandwich panel total thickness S. Pine face thickness inches 7/8 5/8 0.71 Particleboard core, thickness inches 1/8 1/8 1/6 inches 5/8 3/8 3/8 The sandwich panels were fabricated in the following manner: Clear straight grain southern pine veneer, dried to less than 6 percent M.C. (moisture content), was used for all panels. Particleboard (underlayment quality) was used for cores after conditioning to approximately 7 percent M.C. Commercial grade, extended phenolic resin (similar to that used in manufacturing southern pine plywood) was used for gluing veneer faces to particleboard cores. Resin was applied only to the veneers. A spread of 85 pounds per MDGL was applied with a curtain coater. Sandwich panels were first cold pressed with 165 psi. for approximately 3 minutes, then pressed with 200 psi. at 295°F. for 5 minutes. Cured panels were cooled under pressure, then stored in a conditioned room at 65 percent R.H. and 72°F. until testing. In addition, 5/8 inch and 7/8 inch southern pine plywood panels were made simultaneously with the sandwich panels under the same manufacturing conditions. The quality of all veneer used for these plywood panels was select, free of visible defects. [4] TESTING From each sandwich and plywood panel, nine strip specimens were cut, each 3 inches wide and 2) inches long, and tested in flexure to failure with central loading and direction of face grain parallel to a 16 inch span. In addition, from each sandwich and plywood panel, seven flexure specimens were cut, each 3 inches wide and a length of 48 times their thickness plus 4 inches. These specimens were tested non-destnrctivelv in flexure at six span-todepth ratios for determination of flatwise (interply) shear modulus (G..) according to a method used previously by Biblis (2), (3), and Biblis and Chin (4). For determination of edgewise shear modulus (G.,), four small square shear panels were cut from each construction. Side length was 30 times the panel's thickness. This test was conducted according to ASTM-D805-63 (1), as indicated in Figure 1. In addition, edgewise shear strength of each sandwich and plywood construction were determined by a rail shear test according to the method developed by the U.S. Forest Products Laboratory (7) primarily for hardboard. Edgewise shear and interply shear stresses on sandwich specimens are l lllncfr~lif in iinrlrri FIG. 1. Plate shear test used for determination sandwich panels. of edgewise shear modulus of [51 Z Y X (A) Edgewise shear modulus- Gxz Z , .X (B) Flatwise shear modulus FIG. 2. Gxy Illustrations of edgewise and flatwise shear in a sandwich panel. [6] OBSERVED AND PREDICTED PROPERTIES Flexure From specimens tested to failure, the moduli of rupture (MOR) for the sandwich and plywood constructions are listed, along with the observed maximum midspan loads, in Table 1. From specimens tested non-destructively in flexure at six span/ depth ratios, values of pure modulus of elasticity (free of shear deformation) and values of flatwise modulus of rigidity (G1 ,) for each construction were calculated according to the method used previously by Biblis (2) and listed in Table 1. For a more meaningful comparison of stiffness among the various sandwich and plywood constructions, Table 1 also lists the load required to cause 0.1 inch midspan deflection of each type specimen at a 16 inch' span. Experimentally determined values of MOE for each specimen type were plotted against span/depth ratios as shown in Figure 3. Observed values of midspan total deflection at proportional limit (P.L.) load for each type of specimen were predicted by a method which transforms the cross section of a sandwich specimen into a hypothetical cross section of a homogeneous I-beam of unidirectionally laminated veneer with grain direction parallel to span. The transformation is made by reducing the width of the particleboard core by the ratio of modulus of rigidity (flatwise) of the board core (G..e) to that of face veneer (G1 ,.r) thus by the ratio (Gx,../G..) .1 Bending deflection of such an I-beam, when centrally loaded and freely supported at the ends, is calculated by the following equation derived by Newlin and Trayer (8). (1) Y = (PL3/48EI) + (KPL/Gxs.t) where: Y = total elastic deflection at mid-span, inches (in our calculations deflection at proportional limit load was predicted) P = applied load at center to cause Y deflection, lbs. L = span of specimen, inches EI= (Et It +- Ec Ie) Et = pure modulus of elasticity of face veneer (grain parallel to span) p.s.i. 1 The first two subscripts designate the plane of shear stress, the third subscript designates core c or face f. [7] 0. 8 16 24 Span /depth 32 ratio 40 48 FIG. 3. Effective moduli of elasticity of various flexure specimens (strips) at six span-depth ratios. Face grain orientation of unidirectionally laminated veneer (UNI), sandwich (SDW) and plywood (PLW) strips was parallel to span. Particleboard specimens (PCB) were included for comparisons. C8] I -= moment of inertia of face veneer, with respect to neutral axis of sandwich, inches4 EC = pure modulus of elasticity of particleboard core, p.s.i. I -= moment of inertia of core, with respect to neutral axis, inches4 GX,.r = modulus of rigidity of face veneers (GLR) de- termined from unidirectionally laminated veneer, p.s.i. K = coefficient which according to Newlin and Trayer (8) can be calculated by the following simplified equation: 3/2 (h 2 2 - h 2 1 ) h1 b2 h22 h.2 b 10It where: h2 = distance of neutral axis to extreme fiber h1 = distance of neutral axis to flange b1 = reduced width of core of sandwich strip, (web of I-beam) = (G,.e/Gx .) b 2 b2 = actual width of sandwich strip (width of flange of hypothetical I-beam) It -- moment of inertia of the transformed cross section (I-section) with respect to neutral axis. The first term of equation (1) PL3/48EI represents deflection caused by pure bending. The second term (KPL/Gx..t) represents shear deflection. Values of total observed deflection and predicted deflections by equation (1) (pure bending, shear and total) are listed in Table 2. For each sandwich construction MOR values were predicted E (2) by the following equation: MOR - MOR E where MOR and E correspond to properties of sandwich with face grain direction parallel to span, while MORE and EE correspond to properties of veneer obtained from unidirectionally laminated face veneer tested with grain direction parallel to span. Equation (2) was first suggested (FPL-059, (9)) for estimating MOR of plywood at long spans. Estimated MOR values for the three sandwich constructions are listed in Table 3. Values of pure flexure E for each sandwich predicted by the following equation: E = (Edlt + El) /I and also listed in Table 3. [9.] TABLE 1. FLEXURAL PROPERTIES OF SANDWICH AND PLYWOOD STRIPS (3-INCH) WITH FACE GRAIN PARALLEL TO SPAN' Tyef sypeen2 of Pure modulus elasiciy, E PsiPsi 1,656,550 2,047,120 1,667,890 1,613,000 1,715,660 Modulus of igdityofModulus (flatwise). (flatwise)load, 25,360 33,510 25,1.68 20,320 rupture, MOR of SDW-5/8"________--SDW-.71"____-------SDW-7/8"_________ PLW-5/8" 272 8,720 19,960 PLW-7/8"____-_. 'Specimens were conditioned prior to testing at 65% R.H. and 72°F. 2 SDW and PLW designate Sandwich and Plywood respectively. 8 Load applied to specimens 3" in width over 16" span. 4sz designates the sample standard error. O u-a TABLE 2. -- ___. 11,730 12,580 8,810 9,930 s 683 638 378 225 M Maximum midspa P Lb. 572 788 Load causing midspan deflection Lb. 0.1" Effective modulus' of elasticity, E' Psi 1,479,580 1,790,160 109,290 44,430 106 186 843 485 835 265 101 257 1,347,430 1,408,800 1,311,190 53,900 58,320 40,380 ACTUAL AND PREDICTED MIDSPAN DEFLECTIONS AT PROPORTIONAL LIMIT LOADS OF 3" WIDE SANDWICH STRIPS WIT GRAIN OF FACE VENEERS PARALLEL TO SPAN Specimen construction Span to depth ratio Prop. limlit load Predicted pure bending dfltio Predicted shear deflection Total deflection Actual Predicted Error Percentage of shear deflection Lb. SDW-5/8"----------SDW-0.71"------SDW-7/8"----------- In. .74709 .82863 eeconto In. In. .92446 In. .77995 .86618 Pct. -- 15.62 -10.43 - total Pct. 14.96 34.08 15.27 34.25 13.46 31.36 3.74 48 48 24 14 24 14 48 321.4 551.0 380.4 646.7 205.6 160.7 .18677 .06356 .20721 .07170 1.12517 .03286 .03286 .03286 .25393 .10648 .88474 .21963 .09642 .24456 .10905 1.16891 -18.53 4.21 190.2 .03735 .03735 .04374 .03735 .24163 .10144 1.17526 + 1.21 + 7.50 - 2.14 4.31 24 14 411.2 704.9 .28129 .09571 .04374 .04374 .32322 .13585 .32503 .13945 + 0.56 + 2.65 0.54 Values of pure E and G, for veneer faces and particleboard core that were used in equation (1) were determined from additional tests and are listed in Table 4. TABLE 3. PREDICTED VALUES OF PURE MOE AND MOR OF THREE TYPES OF SANDWICH STRIPS WIT FACE GRAIN PARALLEL TO SPAN Type of specimen Predicted pure modulus of 1 Epred.' Eact. elasticity,rpte, Psi Predicted modulus of upreMOR Psi 11,660 MORpred.' M at act.C SDW-5/8"--------------1,982,540 1.196 0.994 SDW-.71________w____ 2,117,870 SDW-7/8"------------_______1,684,020 1.034 1.010 12,450 9,900 0.990 1.124 1'Ratio of predicted value to actual value obtained from tests. TABLE 4. FLEXURAL AND SHEAR PROPERTIES OF PARTICLEBOARD CORE AND VENEER FACES Type of specimen Pure modulus of elasticity, E Psi Flatwise modulus of rigidity, G1 . Psi 20,655 Modulus of rupture, MOR Psi 1,980 Particleboard -----------406,500 2,416,550 Uni-laminated veneer.--- 42,275 14,210 TABLE 5. EDGEWISE SHEAR PROPERTIES OF SANDWICH, CORE, AND FACE VENEER PARTICLEBOARD Type of specimen Actual edgewise modulus of rigidity sat (G1Y)act. Psi Edgewise shear strength s:2 Predicted edgewise modulus of rigidity (G,)pred. (Gxy)pred. (G1Y)act. Psi 3,395 4,225 2,875 3,250 850 Psi 37 42 23 20 19 SDW-5/ 8"______. 131,050 SDW-.71 "_______ 123,680 SDW-7/8"_____________131,350 1,125 1,010 1,090 930 930 128,140 124,570 145,210 _____ 0.978 1.007 1.105 PCB-3/8"---------148,410 PCB-5/8"____------- Face 164,200 veneer____________97,730 3,825 980 22 'SDW and PCB designate sandwich and particleboard respectively. 2 sz designates the sample standard error. [11] Edgewise Shear Experimental values of edgewise shear modulus (Gxy) and edgewise shear strength for each sandwich and plywood construction are listed in Table 5. Edgewise shear modulus was obtained by plate shear test and calculated by the following equation:2 (3) 2h3w where: Gx, = edgewise shear modulus, psi P = load applied to each corner, lbs. u - distance from the center of the panel to the point where the deflection is measured, inches h = thickness of plate, inches w = deflection relative to center, inches Edgewise shear modulus of each of the sandwiches was predicted by the following equation, which is a special form of a general equation proposed for plywood by the U.S. Forest Products Laboratory (10). 1 = Gxy 3u 2 p Gxy- = (Gxy., h + Gxy. he) (4) where: Gay = edgewise shear modulus, psi h = total thickness of sandwich plate, inches Gxy.f = edgewise modulus of rigidity of faces, psi. (determined from plywood plates of matched veneer) hf = thickness of veneer faces, inches Gxy.C he 2 = edgewise modulus of rigidity of core, psi. (de- termined from particleboard plates) = thickness of particleboard core, inches Strictly speaking equation (3) is based on theory of isotropic plates. It has been used by March, Kuenzi, and Kommers (6) for orthotropic thin plates (plywood and specially sectioned wood) after making several specific assumptions with respect to geometry, fiber orientation, homogeneity, construction, and specific anisotropicity of these plates (March, (5)). The use of equation (3) here for sandwich plates is justified only for two reasons: First, it allows a direct comparison between plywood and sandwich plates through their respective G17 values that were obtained by the same test method and equation. Second, use of equation (3) with the sandwich plate (a nonhomogeneous special orthotropic plate) does not violate the original assumptions any more than the use with plywood plates. [12] RESULTS Actual MOR values obtained from tests of 5/8 inch and 7/8 inch sandwich specimens with face grain parallel to the span are larger than MOR values of equal thickness plywood. The larger MOR values of the sandwich constructions might be explained by the fact that the MOR value of the particleboard core is larger than that of crossband veneers in plywood. The most efficient sandwich panel of the three is the one with a 3/8 inch particleboard core faced with 1/6 inch veneers. The MOR value of this sandwich (a measure of flexural strength in relation to its thickness) is 27 percent and 44 percent larger than 5/8 inch and 7/8 inch plywood. The 7/8 inch thick sandwich (5/8 inch core + 1/8 inch faces) can carry larger maximum loads than either of the other two sandwiches or the 5/8 inch or 7/8 inch thick plywoods. MOR values of these sandwich specimens have been predicted with reasonable accuracy by an approximate formula. Values of E for sandwich specimens tested with face grain direction parallel to the span are slightly larger than values of equal thickness plywood. The total midspan deflection of the sandwich specimen was predicted accurately at three span/depth ratios with an approximate method which was developed and applied for plywood by the authors. Total deflections predicted by this method for two of the sandwich constructions are in excellent agreement with actual deflection values from tests. Total midspan deflections at three spans of the third sandwich type were predicted with errors varying from 10.5 percent to 18.5 percent. Edgewise shear moduli GXy of the sandwich panels are approximately 30 percent larger than those of plywood. This is attributable to the larger edgewise shear modulus of the particleboard core that contributes substantially to shear stiffness of the sandwich panel. Edgewise shear modulus of each sandwich construction was predicted very accurately from shear properties and thickness of components. Edgewise shear strength of sandwich panels was found to be higher than for particleboard cores or face veneers. This can be attributed only to the variability of veneer properties and, perhaps, of the particleboard. [ 13] CONCLUSIONS Flexural stiffness and strength of structural sandwich wood panels with grain orientation of face veneer parallel to span can be predicted with accuracy at any span by elementary methods from properties and thickness of particleboard core and face veneer. Edgewise moduli of rigidity of these structural sandwich panels also can be predicted very accurately with an elementary method from the properties and thickness of the components. Modulus of rupture and modulus of elasticity values of 5/8 inch and 7/8 inch thick sandwich panels with grain orientation of face veneer parallel to the span are larger than corresponding values of plywood of equal thickness and veneer quality. Edgewise shear moduli of all sandwich panels are approximately 30 percent larger than plywood panels of the same veneer quality. The most efficient structural sandwich wood panel of those investigated is the one with a 3/8 inch particleboard core faced with 1/6 inch veneers. The MOR value of this sandwich is 27 percent and 44 percent larger than that of 5/8 inch and 7/8 inch plywood, respectively. (14 ] REFERENCES (1) AMERICAN SOCIETY FOR TESTING AND MATERIALS. 1968. ASTM Standards, D805-63, Philadelphia, Pa. (2) BIBLIS, E. J. 1965. An Analysis of Wood-Fiberglass Composite Beams Within and Beyond the Elastic Region. For. Prod. Jour. 15(2): 81-88. (3) BIBLIS, E. J. 1969. Flexural Rigidity of Southern Pine Plywood. For Prod. Jour. 19(6):47-54. (4) BIBLIS, E. J. AND YEN-MING CHIU. 1969. An Analysis of Flexural Elastic Behavior of 3-ply Southern Pine Plywood. Wood and Fiber. 1(2) :154-161. (5) MARCH, H. W. 1942. Flat Plates of Plywood Under Uniform or Concentrated Loads. U.S. For. Prod. Lab. Rpt. No. 1312. Madison, Wis. (6) MARCH, H. W., E. W. KUENZI, AND W. J. KOMMERS. 1942. Method of Measuring the Shearing Moduli in Wood. U.S. For. Prod. Lab. Rept. No. 1301. Madison, Wis. (7) McNATT, J. D. 1969. Rail Shear Test for Evaluating Edgewise Shear Properties of Wood Base Panel Products. Research Paper, FPL-117, U.S. For. Prod. Lab. Madison, Wis. (8) NEWLIN, J. A. AND G. W. TRAYER. 1924. Deflection of Beams with Special Reference to Shear Deformations. National Advisory Committee for Aeronautics Rpt. No. 180. Washington, D.C. (9) U.S. FOREST PRODUCTS LABORATORY. 1984. Bending Strength and Stiffness of Plywood. Research Note FPL-059. (10) U.S. FOREST PRODUCTS LABORATORY. 1951. Design of Wood Aircraft Structures. ANC Bul. 18. Dept. of Defense, Washington, D.C. [15 ] AUbUIh UNIVLRSi Yf With an agricultural research unit in every major soil area, Auburn University serves the needs of field crop, livestock, forestry, and horticultural producers in each region in Alabama. Every citizen of the State has a stake in this research from new program, and more s t since any advantage economical ways of producing and handling farm products directly benefits the consuming public. 1 Research 'in 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Unit Identification Ac E Station, Auburn. Tennessee Valley Substation, Belle Mina. Sand Mountain Substation, Crcssville North Alabama Horticulture Substation, Cullman. Upper Coastal Plain Substation, Winfield. Forestry Unit, Fayette County Thorsby Foundation Seed Stocks Farm, Thorsby. Chilton Area Horticulture Substation, Clanton. Forestry Unit, Coosa County. Piedmont Substation, Camp Hill. Plant Breeding Unit, Tallassee. Forestry Unit, Autauga County. Prattville Experiment Field, Prattville. Black Belt Substation, Marion Junction. Tuskegee Experiment Field, Tuskegee. Lower Coastal Plain Substation, Camden. Forestry Unit, Barbour County Monroeville Experiment Field, Monroeville Wiregrass Substation, Headland. Brewton Experiment Field, Brewton. Ornamental Horticulture Field Station, Spring Hill. Gulf Coast Substation, Fairhope