Small leak detecting

[mototopo]

After much work I was able to eliminate a leak in my RH tank. I left the proseal cure for about a week and yesterday I tested it again for leaks.
It passed the "soapy water test" but the hi-tech "balloon method" showed what can be a very small leak. After almost 20 hours sitting in the shop, the balloon has lost about half an inch in diameter. The first tank I made rested for about a week with the balloon fully inflated without any evident pressure drop.

How may I detect a so small leak?

[Beancounter]

Claudio
I suggest trying it again with two balloons. One as a control that is just tied off. The 1/2" deflation in 20 hours could be due to other factors, porous rubber, or a change in barometric pressure. I would want to eliminate any possibility of testing error before more drastic measures.

Matt

[Lycosaurus]

I never understood why the balloon method (24 hours test) is used. It seems that when a leak is suspected by having a balloon lose it's shape over a set period of time, then the actual leak is detected by using soap and water.

Problem, as mentioned in the earlier post, is that temperature, barometric pressure changes, as well as defective/leaky balloons can skew the results.

During our build, I just skipped the balloon method (24 hour test), and just used the balloon as a buffer/pressure regulator for the tank. Filled the tank with some air to inflate the balloon, and went straight to checking every seam and rivet with soapy water. Takes maybe an hour, if that, to do both tanks.

[kentb]

Just use the balloon to keep from over pressuring the tank. Use the soapy water to find any leaks.


Kent

[mototopo]

Thank you guys.
Just to be clear, I used the baloon to have a reference while pressurizing the tank and to show any major leak. Then I proceeded with the first soapy water test searching for small leaks. Nothing found so I was happy for my second leak-free tank. I left the tank pressurized with the balloon just for peace of mind and today I saw that small deflating.

Now, I just re-tested the tank with soapy water and no leak founds. I too imagine the balloon is maybe leaking due porous rubber (the balloon was in my drawer for more then a year...). I just want to be sure there is no leak before proceeding.. repairing a leak now is obviuosly easier then once the tank will be mounted on a completed aircraft.

I'll try with a second balloon just to be sure, but, at this point, I think that there are no leaks... sure at 99,99% ....

[txnbuilder]

Claudio,

Do you have capacitive senders in your tanks?

I used a water column instead of a balloon (much easier to measure the height of water than the size of a balloon!) and also had a very small leak -- only detectable after several days of testing.

It turned out to be in the center pin of the BNC connector for the capacitive senders...

Luis

[mototopo]

Hi Luis,
no, I have the standard floating sender.

I have to investigate how the "water column system" works...

[AltonD]

Its just a manometer. Set up for a 24 inch displacement.

[txnbuilder]

The nice thing about using a water column is that once you pump it up to the right height (about 24' as mentioned before), you just make a mark on the tube. When you come back, it is very easy to see how much the water has moved.

You can either compensate for temperature and pressure differences when you check for leaks, or simply wait until the temperature and pressure are again the same as when you make the mark on the tube -- about the same time the next day during Texas summer

Here is my setup:

http://rv7a.fabs-designs.com/2009/09...ing-the-tanks/

Luis

[AltonD]

Careful though. the 24 inches is the delta between the two levels. I read a post where the builder raised the right column by 24 inches, which meant the left column went down 24 inches for a total displacement of 48 inches. I think he said "I thought those tanks were bulging out a little much. They didn't leak though."