24 



Hardwood Record — Veneer & Panel Section 



August 10, 1919 



tions and conditions affecting the manufacture of the 

 plywood were studied sufficiently to bring out their rela- 

 tive influence upon the water resistance of the panels as 

 shown by the boiling test. 



There was no question as to the effect of thickness of 

 the veneers, since, as shown in the table above, the 1/20" 

 veneer gave 85 per cent of the panels perfect and the 

 1/10" gave only 25 per cent perfect. The gradual de- 

 crease in the percentage of perfect panels, as the plies 

 increased in thickness, strongly indicated that the cause 

 of failures lay within the plies and that external conditions 

 probably had little effect. These conditions, however, 

 were not ignored in the investigation. 



Thickness of Veneers 



Sample pieces taken from a large panel usually gave 

 extremely varied results when panels of the thicker 

 veneers were tested. One part of the panel might be 

 practically perfect, while total failures occurred in another 

 part of the same panel. This seemed to point toward 

 a lack of uniformity of pressure. However, later tests 

 proved that the pressure used had little to do w^ith the 

 water resistance of the plywood. Along with this con- 

 clusion, came another — that the variation in thickness of 

 the individual piece of veneer, likewise, had little, if any. 

 effect on the water resistance of the panel. 



Veneers from Steamed zmd Soaked Logs 



It was thought that veneer cut from a steamed log 

 would, perhaps, give different results from veneer from 

 a soaked log, but this was proved to be a fallacy after 

 dozens of pieces of both kinds were followed through a 

 uniform treatment of drying, gluing, pressing, and final 

 boiling. One kind gave just as good results as the other. 



Glues and Various Ingredients 



The various casein glues were quite thoroughly tested 

 out. The methods of mixing and stirring, and the 

 amounts of the different ingredients were varied. Thick 

 glues and thin glues, heavy spreads and light spreads 

 were tried, but none of the changes made had any very 

 marked effect, although the thicker layers gave a higher 

 percentage of failures. 



The purity of the various ingredients was questioned 

 and, where possible, tests were made to ascertain the 

 effect of different brands of commercial products, but no 

 beneficial results w^ere obtained. 



Pressing and Drying Panels 



The length of time during which the panels remained 

 under pressure made no difference in their water resist- 

 ance. Tests were made on those left in for various 

 periods of from 3 to 48 hours, with no difference in the 

 resulting panels. 



There was some indication that when like sides of the 

 veneer were laid together there was an improvement, but 

 not enough to change the routine of handling the veneers. 



Treatment of the panels after removal from pressure 

 had practically no effect. A number of variations in dry- 



ing methods were tried — air-drying for short and for long 

 periods, kiln-drying in low and in high temperatures and 

 in dry and moist air, and very rapid drying in a textile 

 drier. No one method gave more satisfactory results 

 than another. 



Intensive Study of Veneer Moisture Content 



Finally tests were made with veneer containing a higher 

 percentage of moisture than past experience with animal 

 and vegetable glues seemed to warrant. Immediate and 

 striking results were obtained. While changes in any of 

 the other factors involved in the manufacture of this ply- 

 wood produced slight, if any, difference in the water re- 

 sistance, the change to higher moisture content produced 

 enormous and definite differences. The following fig- 

 ures show howr great were these results. Of 600 sample 

 pieces with low moisture content, only I 00 were perfect, 

 and of the 500 failures, fully a third had over 50 per cent 

 of the face plies separated. Of 1,000 pieces with high 

 moisture content not a single piece failed. 



Factors Affecting Amount of Moisture Necessary 



A large number of determinations were made in an 

 effort to find the lowest moisture content of the veneer 

 that would produce 1 00 per cent of water-resistant panels, 

 and also to find the range above that minimum point. 

 This series of tests, along with the factory observations, 

 brought out several important points. The minimum 

 moisture content seems to be different for different thick- 

 nesses of veneer, for different kinds of wood, and for 

 different constructions of panel. At this time only two 

 constructions were considered, 3-ply 1/12" birch, and 

 1/14" birch faces on a 1/10" yellow poplar core. Each 

 seemed to have different minimum points. 



The minimum point for the 1/12" birch was found 

 to be near 1 8 per cent moisture, probably a little below 

 that; while the range above that point, in which the boil- 

 ing test indicated perfect panels, that is no separation 

 of plies, w^as 35 per cent or higher. Three-ply pieces, 

 5" square each containing 50 square inches of glue line, 

 were adopted as standards for testing. Sample panels, 

 one foot square or larger, were made from veneer with 

 different known moisture percentages. The table gives 

 the percentage of failure for each different moisture con- 

 dition, based on square inches of glue line. 



Three-ply 1-12 inch Birch faces on !-I2 inch Birch core. 



liLith Fat'i's 

 and Core 



7 % 



9!/2 



16 



18 



24 to 35 



of <;iiif 

 liine Failiii 



97';-; 



2 

 2 

 

 



No. of Sii. Ill 



of OUu' 

 I.iiii' Failuiv 



560 

 4 



n 





 



Xo. of S(i. III. No. of 



Glup Pieces 



Line Tisliil Boiled 



575 

 200 

 575 

 1200 

 400 



8 

 4 

 8 



24 



Moisture content is based on weight of wood dried at 1 00^ C. 



For the other construction, 1/14-inch birch faces on 

 1 I 0-inch yellow poplar core, the minimum point seemed 

 to be very close to 20 per cent, w^hile the range above 

 that was between 20 per cent and 45 per cent. The table 

 following is based on tests made of about 40 panels, each 

 with known moisture content. About 1 50 pieces, 5 inches 



