Yuku free message boards

The WW II typhoon story is somewhat more complex with respect to stability.

Two of the Farragut class DDs were right on the ragged edge of acceptable stability. Recall Halsey was trying to refuel. Standard practice for DDs was to ballast fuel tanks with water as they were emptied, so although fuel capacity was going down, the virtual center of gravity was not rising thereby.

However, to refuel meant (in those ships) pumping out the ballast water to make room for the fuel about to be received. Thus the virtual CG did go up, the metacentric height (a measure of stability) shortened, the angle of maximum righting arm (another measure of stability) got smaller as did the angle at which the righting arm would go negative. When it did go negative, the ship was going over like it or not, and regardless of sea state. Clearly with beam winds, this problem got worse. The angle of heel at which this happened was not absolute, but varied with the other variables, as it were.

All the ships went through this process throughout the war, but doing it an typhoon was asking for trouble, and the trouble arrived. It was typical in those days to get down to about 20% fuel - the OTC HAD to fuel his escorts somehow, but chose to get into a typhoon while attempting to do so.

The immediate consequence of the storm was as noted, an increase in minimum acceptable fuel level. Post-war the DE-1040 class inaugurated the concept of compensating fuel tanks whereby oil suction was taken from the TOP of the tank, constantly replaced by firemain ballast water coming in at the bottom. Taking on fuel merely reversed the process, with the ballast being pushed out overflows and over the side. Another post-war design requirement was the ability to resist 100 kt beam winds (if properly ballasted).

"Fuel the DDs every three days" became another post-war mantra in the USN.

It is counterintuitive, is it not?

Keeping full tanks for DDs was as must for stability reasons, as I've disucssed, and putting sea water (ballast) into empty tanks was the only way to do it. Pre-'compensating' fuel tanks, the tank had to be empty to try to avoid the problem you've mentioned. Like most fuel tanks I expect, one could not pump down to actual empty. Water was always going to be in the fuel to some extent, so the tanks were periodically stripped via a connection at the bottom of the tank, pump discharge going into a settling tank where the water and oil at the oil-water interfaced (there always a bit of emulsion there) could separate, and the oily water residue finally pumped overboard. Yes, I know it was not environmentally correct.

Above the stripping connection was the fill connection, and above that the 'service connection.' For a two plant four boiler DD, for example, "fuel oil service' pumps took suction from 'service tanks' (two per plant) discharging via a filter into a manifold that supplied the boiler burners. The service tanks were filled by a fuel oil 'transfer' pump taking suction via that 'service' connection from the storage tanks. Only the latter were ballasted. One service tank was on suction at a time.

So, de-ballasting went on via the stripping connection and over the side.