576 



(f) Effect of increasing mass of diaphragm on damage . Since the 

 density of the diaphrasm material occurs as a parameter in the damage 

 equations of the Kirkwood Theory (cf. Section III), a partial test of the 

 theory could be made by "loading" the diaphragm so as to change the ef- 

 fective density without altering the mechanical strength of the diaphragm. 

 The following methods of loading were tried. 



(1) A lead disk was soldered to the back of the diaphragm with two 

 series of grooves cut into the disk almost down to the diaphragm; they 

 were l/k in. apart and at right angles to each other so as to reduce the 

 strength of the lead. (A few solid lead disks were tried and proved un- 

 successful.) 



(2) The gauges were hung face down, so that the diaphragm was 

 parallel to the charge-to-gauge line, with (a) mercury and (b) putty 

 lying free on the back of the diaphragm. 



(3) Putty was packed into a thin-walled steel tube soldered to the 

 back of the diaphragm. The gauge was set in the normal vertical position 

 and a cardboard disk was used to hold the putty in place. 



(U) Mercury was loaded on the back of a vertically held diaphragm 

 by means of a second diaphragm clamped in a fitting inside the gauge. 

 This second diaphragm consisted of (a) cellulose acetate sheeting, and 

 (b) this sheeting plus a l/^ in. sheet of rubber. 



A load of about 530 gm, which is 6.7 times the weight of the steel 

 diaphragm (79 gm. ) was used. The charge range was 200 to 2^+00 gra. at 

 distances of k to k-\/2 ft. Results are given in Table XIX. Since In 

 some of these tests the diaphragm was heated to solder material to it, 

 five diaphragms were given similar heat treatments and then damaged with 

 no loading. The heating was fovind to affect the damage by no more than 

 2 to 35^- 



In general it may be concluded from these tests (a) that a rough 

 agreement with theory has been found; (b) that in the cases of mercury 

 loading and of lead leading where the lead completely left the diaphragm, 

 the lower damage of loaded diaphragms results from the fact that the 

 loading material left the diaphragm in the deceleration stage carrying 

 energy with it; (c) that the horizontal gauges show less damage and some- 

 what less effect of loading than the vertical gauges. 



(g) Effect of bending radius on nature of ruptxire of diaphragms . 

 The daxnage gauges used at UERL have been constructed with a bending radius 

 of 1/16 in. (Sec. II. 1). It was observed that steel diaphragms in these 

 gauges ruptured at the edge, while copper diaphragms burst in the center. 

 It was suggested that ruptvire should occur at the edge because of the 

 finite time required for the plastic wave to propagate to the center. 

 Kirkwood in his theory ignores this propagation time and predicts that 

 rupture should occur in the center, provided that the bending radius is 

 great enough to avoid rupture by bending as opposed to tension. In order 



