625
STRENGTH OF GLASS.
Strength of Glass.
Two series of tests were made (1) with lights set in frames, bedded in neat Portland cement, (2) tranverse tests of lights supported across each end and loaded across the middle of the span, no cement beino■ used with these specimens.
In the first series the lights were bedded and the cement allowed to set an interval of 21 hours before testing. Loads were applied at the middle of the lights by means of a cast-iron block measuring 2" x 2" in cross-section dimensions. Wet cardboard cushions were placed between the face of the cast-iron block and the light of glass. The central load required to break the glass is recorded as the strength of the material, the load being־ gradually increased until rupture occurred. The actual thickness of the glass varied somewhat from its nominal thickness. The figures over the ultimate strength give the actual thickness of each light.
The results taken as a whole show over what range the strength of the glass may be found. There are variations in the strength of lights of the same nominal thickness, which make it appear that the strength of this material varies in itself; in some cases a large light sustained a greater load than other lights of smaller dimensions.
It is observed that the strongest lights broke into the largest number of fragments. Generally the number of pieces ranged from 8 to 16; An exceptionally strong light was found in the rough plate 1" thick, 16" x 16" size, which displayed upon fracture 46 radial cracks,
I he tabulations of the results with square lights set in frames are followed by the details of the transverse tests, and then tables of comparisons based upon the preceding details.
The results of the transverse tests, in order to admit of direct comparison, are reduced by computation to pounds per square inch, and are given in the column headed “Modulus of rupture.” The formula for this reduction in strength is as follows:
3PZ
^'~2id*' *n which K=m°dulus of rupture, pounds per square inch. P=load.
.	I ־-length of specimen or span.
5= breadth of specimen. d—depth of specimen.
In the first table of comparisons on rough plate from two factories it appears that in four thicknesses each factory shows the stronger glass in two cases. So far as concerns the strength of the glass with respect to thickness, the greater relative strength generally is found in the thinner lights.
In ribbed plate no practical difference is shown in the strength of the glass from the two factories represented.
In the wired plate in each case the sandwich rolled appears the stronger.
In the matter of annealing, one thickness is stronger in the oven-annealed plate, another thickness shows results in favor of Lehr annealing, while in a third thickness no practical difference is found. The method of annealing the thinner lights of this series is not known.
H. Doc. 508-----40