And … balloons leak. Even really good balloons can leak … much less the party ones we buy at Target. You might want to try a manometer instead, you know … a marked “U” channel with an appropriate fluid in it. Now take all the other variables into account and you’ll be close. Of course we don’t have an ideal gas to use pv=nrt accurately, but for our purposes it’s close enough. This method substantially removes the question of changing volume due to balloon relaxation et al.

Or flush the inside of the tank with Freon (don’t let the FEDs know), pressurize it a little, seal it, then use a “sniffer” to detect the very smallest micro leak. Flow plenty of air over the wing areas for a while to remove residual Freon pockets. Freon’s molecular size (critical dimension is about 5.3 angstroms) is close to pentane (6 angstroms I think I remember) which means, if you can flow Freon, you can flow pentane. So if you find a leak using a sniffer, you might have a gasoline leak there too. Hey, I might be wrong about the numbers … it’s been a long time since using these dimensions etc, but the sniffer should work even if the numbers are wrong. This method removes the volumetric, atmospheric and temperature variation issues altogether. Just make sure there’s positive pressure in the tank when searching for that micro amount of Freon.

Jim