Fuel Tank, Part 3 - Leak Testing

Upon completion of the fuel tank the task of leak testing was scoped.  After researching Van's prescribed "balloon" method of leak testing, it was determined that a water manometer approach would provide a more accurate means of providing an internal pressure.  27 inches of water column is equal to 1 pound of air pressure per square inch.  Reports from other RV-12 builders indicated that the balloon-based approach to leak testing was only providing approximately 1/4 to 1/2 psi of internal tank pressure.

I built the following water manometer for my tank testing.  Note that I suspended the tank by 3 nylon cords in order to have complete visibility of the tank, once it was under pressure.

The zero-pressure water column rested at about 13 inches on the metal yard stick.  After several strokes of the bike pump, the column approached 40 inches, resulting in 1 psi of tank pressure.  I then proceeded to spray and brush on a water solution of glycerin and dish soap to detect any air leaks identified by bubbles.  Unfortunately, I found 3 small leaks in the aft upper corners of the tank and had to end the testing.  Thankfully, they were easy to spot.  The locations of the leaks were marked with a Sharpie, the tank depressurized and cleaned up and the repair process began with another new batch of ProSeal.  After 3 days in a very hot (95-100 degrees) garage with the ProSeal fully cured, the second pressure testing began.  Following is a quick shot of the tank being raised to 1 psi.

The previous leaks held beautifully, but I found 3 more even smaller leaks in the opposite corners of the aft panel and around 1 rivet on the tank bottom and from a couple of rivets which attach the left lower mounting bracket.

Again, the tank was depressurized, cleaned up and another round of ProSealing ensued.  Same 3 days in an even warmer garage and the tank was ready for the third test.  This time as I pressured up the tank, I noticed the water column slowing dropping again, but could not find any leaks from the tank so I began double-checking all of my hose connections.  Sure enough, the bike pump attach point was leaking as was the tank outlet, connected to the manometer.  After removing the bike pump and then tightening the hose clamp on the outlet, the water column remained completely stable!

Needless to say, I was quite happy.  I left the garage and came back an hour later to find that the water column had actually risen 1.5 inches as the garage temperature began to drop as the sun was setting.  At this point, I declared the tank to be stable and free from leaks.