72 NOTES ON RIVETS AND SPACING OF RIVETS FOR OIL-TIGHT WORK. 
being done on the tank between the various tests. Preliminary tests were carried out 
in the order named, with pressures up to 30 pounds, and the leakages and drops in air 
pressure were measured as accurately as possible under various pressures for equal 
lengths of time. ‘The tank as a whole showed up well under the various tests with the 
exception of the single-strapped seam in the 12-pound plating which began to leak at 
10 pounds pressure through the butt caulking of the seam. The single seam in the 
1s-pound plating also showed leakage at a slightly higher pressure. This made it 
difficult to compare the leakages under the various mediums, but, generally speaking, 
the leakage in section A, which was slightly less than in section B, because of the double 
seam strap in the 15-pound plating, showed up greater for oil and gasoline than for 
water, the gasoline being the greatest. The leakage of air was easily detected by using 
soapy water, and as a medium for testing tightness it is considered that air, as has been 
frequently shown in compartment tests, is fully as searching as water, oil or gasoline. 
The greater leakage of gasoline as compared with oil may be partly attributable to the 
gasoline test having been conducted after the oil test. One result was particularly noted; 
namely, that in the oil test under a 65-foot head the leakage was half that for oil under 
a 45-foot head, being 7 pints as against 14 pints. This is probably due to the greater 
deflection closing up the seam. The electric welding showed no leaks in any of the tests, 
nor did the riveting as a whole, only a few weepy rivets being noted. There was some 
slight leakage through caulking of bounding bars. 
After the preliminary tests it was decided to renew the 12-pound plate, using a 
double-riveted lap instead of a single strap, double riveted, the rivet spacing remaining 
the same. The tank was then again tested with water, air, oil and gasoline, using 30 
pounds of pressure as the maximum. Both compartments were perfectly tight with 
water, air and oil. When gasoline was applied to section A, after 30 pounds of pressure 
had been on for an hour, a slight leak developed in the bounding bar to 15-pound plating. 
After 30 pounds of pressure of gasoline had been on section B for three hours a bad leak 
developed in the seam strap of the 15-pound plating, which showed up in the inner 
strap of section A; also at the corners of the bounding bar, the total amounting to 3% 
gallons in six hours. 
The marked improvement in the behavior of the lap over that of the single-riveted 
strap is worthy of note. It is regretted that a single-riveted lap was not subjected to 
test also. 
Commenting on that portion of Mr. Frear’s paper dealing with naval practice and 
the present tendency to allow a somewhat greater spacing than called for by the navy 
rules for oil-tight work, it may be stated that for some at least of the naval vessels 
building, the following oil-tight practice is observed generally (see table, page 73): 
This practice is used as a general guide, border-line cases being settled on their 
merits or on special considerations of strength. 
On battleship work it is customary to fit double bounding bars on oil-tights between 
torpedo bulkheads, double-riveted 5 diameters through the transverse, and single riveted 
6 diameters through the torpedo bulkheads, both bars being stapled. In service oil 
tanks in double bottoms, double bars, single-riveted 5 diameters are used, both stapled. 
In emergency oil tanks in bottoms (reserve feed and void spaces), double bars, single- 
riveted, 5 diameters are used, one bar only being stapled. 
As regards caulking of oil-tight rivets before testing, the yard endeavors to restrict 
this to a minimum. 
