STRENGTH TESTS OF STRUCTURAL TIMBERS. 



MOISTURE DETERMINATIONS. 



Moisture determinations on the untreated wood were made by 

 taking either borings or disks from the tested pieces, weighing them, 

 and then drying them to constant weight. The difference between 

 the original weight and the dry weight divided by the dry weight 

 times 100 is taken as the per cent of moisture at the time of test. 

 Disks taken from the untreated stringers were cut into a number of 

 pieces and the moisture separately determined for each in order to 

 find the distribution of moisture throughout the cross section. The 

 method of dividing the disks is shown in figure 2. The moisture 

 determinations made on treated specimens were handled by dis- 

 tilling the treated shavings cut from the test pieces with water- 

 saturated xylol. For such determinations 

 a definite quantity of treated borings was 

 taken. In all cases a corresponding volume 

 of untreated shavings was obtained, and 

 the dry weight of this sample determined 

 as a basis for computing the moisture con- 

 tent of the treated sample. All test pieces 

 were weighed and measured, the number 

 of rings counted on a radial line, and the < 

 per cent of summerwood and sap deter- 

 mined. 1 Sketches were made and photo- 

 graphs taken, showing the size and loca- 

 tion of knots, checks, and shakes. 



TESTS ON SMALL STICKS. 



i£ a 



V^ 



fk a >| 



' c 



Fig. 2. — Moisture distribution disk 

 for 8-inch by 16-inch stringer. 



After failure occurred in the stringers, 

 small pieces 2 inches by 2 inches in section 

 and 3 feet long were cut from the unbroken 

 portions. These small pieces were selected 

 so as to be free from defects and with straight grain. Their location 

 in a cross section of the stringer was noted, so that data could be 

 secured on the relative strength of the inner and outer portions. The 

 tests of small pieces included bending tests on specimens 2 by 2 by 30 

 inches, compression tests in which specimens 2 by 2 by 8 inches were 

 crushed endwise parallel with the gram, compression tests at right 

 angles to the grain, and shearing tests in which a projecting portion 

 of a small block was sheared off parallel to the grain while the main 

 portion of the block was held firm. 



1 Determinations of summerwood and sap were omitted for some of the Douglas fir. 



