What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Tank removal for testing

K

Karee Brinlee

Active Member
Hey group,
I have a QB built in 2003. I think before I attach the wings I should make sure the tanks do not leak.
From what I understand, you must remove to test so you can inspect the rear of tank that is hidden when attached.
I also understand there is a procedure for testing.
I understand you should not pressurize it a great amount.

Anyway,
I have two questions.
One is the test procedure we should follow?
The other question is this. The tank is attached by screws and re attaching those does not worry me. My concern would be the spar attach bolts. Could we over tighten them and cause a leak when we re attach the tank to the spar?

I have not removed it but I am hoping that will be no problem since the attach points will be flushing up metal to metal and not be pulling on anything that could warp and leak.

Thank you for the help guys and gals.
I am moving full speed ahead on my project.
Just got all my Dynon stuff for my AFS panel today.
It is like Christmas at my house....

Time to get this 7 flying...
 
Before removing an already installed tank, I would 'balloon test' it first. If it passes the balloon test, then you're done.
 
If you have the RV7/7A, removing the tanks should be without much drama, especially when the wings are not attached yet. My friend who removed the tanks from his flying RV8A, built in the early 2000, without any issue.

You must do a full soap bubble leak test with the balloon providing the positive pressure inside the tank. The balloon only test is not sufficient as said from the various older posts
 
tank test.

Ok,
Understood. I assume the balloon is attached to the vent line but should remove tank to bubble test rear of tank.

KB


PhatRV said:
If you have the RV7/7A, removing the tanks should be without much drama, especially when the wings are not attached yet. My friend who removed the tanks from his flying RV8A, built in the early 2000, without any issue.

You must do a full soap bubble leak test with the balloon providing the positive pressure inside the tank. The balloon only test is not sufficient as said from the various older posts
Click to expand...
 
If you do the ballon test, or a manometer test (better), and it doesn’t indicate a leak, why would you need to bubble test the back of the tank? I would just test the tank as attached to the wing. If the ballon doesn’t go down at all, or the manometer remains the same, you don’t have a leak.
Most likely place for a leak on your first test - the fuel cap. Make sure you tape it up really well.
 
leak test

This makes sense to me.
I assume all we are doing is using air. No fuel needed. Since air is thinner than fuel. If it holds air with no leak and tapped well we are good.
KB



Scott Hersha said:
If you do the ballon test, or a manometer test (better), and it doesn’t indicate a leak, why would you need to bubble test the back of the tank? I would just test the tank as attached to the wing. If the ballon doesn’t go down at all, or the manometer remains the same, you don’t have a leak.
Most likely place for a leak on your first test - the fuel cap. Make sure you tape it up really well.
Click to expand...
 
Scott Hersha said:
If you do the ballon test, or a manometer test (better), and it doesn’t indicate a leak, why would you need to bubble test the back of the tank? I would just test the tank as attached to the wing. If the ballon doesn’t go down at all, or the manometer remains the same, you don’t have a leak.
Most likely place for a leak on your first test - the fuel cap. Make sure you tape it up really well.
Click to expand...

My opinion is there are far too many variables (temperature, atmospheric pressure, etc.) to use the balloon itself as an indicator for whether the tank leaks or not. The balloon test is to slightly pressurize the tank so you can spray all the rivets/seams with soapy water to visually inspect for leaks.

Pulling an unpainted tank is a non-event. 50 screws, a dozen bolts. Probably 30 minutes off, 30 minutes back on, and 30 minutes to test in between.
 
tank removal

At this point with the wing on the table. It will only take a few minutes to remove the tank.
I think I will do that.
I suppose I can do an air bubble soap test first. Then if no leaks are found. Do a 100L gas test with about 10 gallons. rotate the tank and leave fuel in it for days. If no leaks I would assume its good.
These QB wings are 20 years old. Maybe I will get lucky.. I am not sure how well they sealed tanks in 2003. Hoping it was real good!
KB

Kyle Boatright said:
My opinion is there are far too many variables (temperature, atmospheric pressure, etc.) to use the balloon itself as an indicator for whether the tank leaks or not. The balloon test is to slightly pressurize the tank so you can spray all the rivets/seams with soapy water to visually inspect for leaks.

Pulling an unpainted tank is a non-event. 50 screws, a dozen bolts. Probably 30 minutes off, 30 minutes back on, and 30 minutes to test in between.
Click to expand...
 
Ballon test is not conclusive

I have QB wings and, as Vans recommends in the build manual, I conducted a balloon test. They stayed inflated for two days. Three years later I started getting seeping at a few rivets.

While there has been much discussion around seeping rivets on QB tanks, I don't believe there has been any definitive cause determined. But a balloon test on its own does not necessarily mean the tank is leak free.
 
Last edited:
Scott Hersha said:
If you do the ballon test, or a manometer test (better), and it doesn’t indicate a leak, why would you need to bubble test the back of the tank? I would just test the tank as attached to the wing. If the ballon doesn’t go down at all, or the manometer remains the same, you don’t have a leak.
Most likely place for a leak on your first test - the fuel cap. Make sure you tape it up really well.
Click to expand...

My balloon burst after about 3-4 hours sitting the warm garage when the air inside the tank expanded. Yet I still found small bubbles at the fuel cap after the failed attempt at hermetically sealing it. This shows the rate of air expansion is greater than the leak but there was a leak. If there are minute pinhole leaks, and the leak rate is smaller than the expansion of the air, then using the balloon only test will not detect the leaks. Just one data point from an actual test.

Edit: for those who have fuel tanks that are built years ago, as most RV projects take a long time to complete, leak test them again before you do the final assembly at the hangar. I have heard and seen a few proseal reversion to goo that it is a sanity check to ensure the proseal in your tanks are still okay.
 
Last edited:
Pulling a tank

Kyle Boatright said:
My opinion is there are far too many variables (temperature, atmospheric pressure, etc.) to use the balloon itself as an indicator for whether the tank leaks or not. The balloon test is to slightly pressurize the tank so you can spray all the rivets/seams with soapy water to visually inspect for leaks.

Pulling an unpainted tank is a non-event. 50 screws, a dozen bolts. Probably 30 minutes off, 30 minutes back on, and 30 minutes to test in between.
Click to expand...

Isn't quite that simple. There are 28 AN-3-3 bolts, many of which are hard to get at through access holes. Then there are the screws that strip and require an easy-out to remove them. If your not good at using an easy-out and break it, things get even more difficult!
 
Fred.Stucklen said:
Isn't quite that simple. There are 28 AN-3-3 bolts, many of which are hard to get at through access holes. Then there are the screws that strip and require an easy-out to remove them. If your not good at using an easy-out and break it, things get even more difficult!
Click to expand...

I've done it on my -10 wings and found it to be an inconvenience, rather than a chore. Glad I did it, because it turned up leaks around the baffle on one of the tanks.
 
Tank screws and 21 bolts removed

As far as I can tell, the tank should come off now with all bolts and screws removed.
I am going to look at the plans again and make sure nothing else needs to be removed, but the tank is not coming off even with a fairly snug pull.

Any advise is appreciated.
KB
 
Karee Brinlee said:
As far as I can tell, the tank should come off now with all bolts and screws removed.
I am going to look at the plans again and make sure nothing else needs to be removed, but the tank is not coming off even with a fairly snug pull.

Any advise is appreciated.
KB
Click to expand...

It’s a very tight fit. You just have to work at it. Keep at it and it will start to give.
 
Karee Brinlee said:
At this point with the wing on the table. It will only take a few minutes to remove the tank.
I think I will do that.
I suppose I can do an air bubble soap test first. Then if no leaks are found. Do a 100L gas test with about 10 gallons. rotate the tank and leave fuel in it for days. If no leaks I would assume its good.
These QB wings are 20 years old. Maybe I will get lucky.. I am not sure how well they sealed tanks in 2003. Hoping it was real good!
KB
Click to expand...

I would NOT test with avgas because when the avgas evaporates the tank sealant can degenerate and revert to goo. This happened to my tanks and I needed to completely clean them out and re-seal. Testing with air pressure and bubble solution is all the factory recommends.
 
Kyle Boatright said:
My opinion is there are far too many variables (temperature, atmospheric pressure, etc.) to use the balloon itself as an indicator for whether the tank leaks or not. The balloon test is to slightly pressurize the tank so you can spray all the rivets/seams with soapy water to visually inspect for leaks.

Pulling an unpainted tank is a non-event. 50 screws, a dozen bolts. Probably 30 minutes off, 30 minutes back on, and 30 minutes to test in between.
Click to expand...
Well hello Karee,
This ^, but then again, that is pretty much what you and I discussed last week.

For those that think a balloon will tell you if you have a leak or not are wishing in one hand and hoping in the other. Way too many variables that would affect the pressure in that balloon.
 
Karee Brinlee said:
As far as I can tell, the tank should come off now with all bolts and screws removed.
I am going to look at the plans again and make sure nothing else needs to be removed, but the tank is not coming off even with a fairly snug pull.

Any advise is appreciated.
KB
Click to expand...

There is a bolt on the inboard nose attached to brackets on the fuselage also. Those have to be removed. There are 21 AN3 bolts on the spar.

Give me a call if you wish to talk through it all.
 
Definitely remove and do the soapy water test.
As mentioned above, the balloon alone is of no use.

Now, Assuming you have all the fasteners out…
The tank skin screw dimples hang up in the splice plate at the outboard end of the tank.
Get a small piece of timber block and place it against the rear edge of the tank skin at the outboard end and tap away gently.
Tank R&R is one of those many RV jobs that are nowhere near as bad as you think in advance.
 
Karee Brinlee said:
As far as I can tell, the tank should come off now with all bolts and screws removed.
I am going to look at the plans again and make sure nothing else needs to be removed, but the tank is not coming off even with a fairly snug pull.

Any advise is appreciated.
KB
Click to expand...

My friend had similar problem removing his RV8A wing tank. Even after 21 bolts were removed, it was difficult to pull out the tank. The flange dimpled for the #8 screws limits the movement somewhat. Double check to ensure all 21 bolts are removed by actually feeling the holes. Have a 2nd helper do that just to check the work. Clamp the wing tightly against the workbench and firmly wiggle the tank loose. Try to wiggle the wing root side first to see if it jiggle. We found out that if the wing is allowed to move even a little bit, the tank is difficult to remove. Keep it firm on the bench.
 
Manometer

I did the manometer. Tanks sat in the living room and held 26" for weeks except for minor changes due to temp. I can look it up. Seems like about 1/2-3/4" per degree temp change.
If there was any kind of leak, I would expect the manometer value to change a lot.
 
You must log in or register to reply here.
Back
Top