What's new
Van's Air Force

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

Confirming no leaks in project tanks?

claycookiemonster

Well Known Member
The baffle (RV8) is not on yet, so once I do that, I know how to check out the tightness of the tanks, but how to test the forward part of the tanks for leaks first? Everything forward of the baffles is done, and considering how well everything else was done, I assume the tanks will be fine (Thanks Kent Scherm!) but how do I check this?

Just fill the tanks with water and look for drips? Does water and avgas leak the same?
 
Do a search on fuel tank leaks. There is a lot of data here on it. Search is your friend. Having said that. A manometer is not going to test for leaks. Do not fill your tank up with water. Not a good way to test either.

Get some balloons from the dollar store (or buy Van’s tank test kit). I think they still include balloons in the kit. Seal the tank and put a balloon on one of the fittings. No the balloon is not the test tool. Using Van’s kit you will install a Schrader valve on the sump drain. Pressurize the tank with air. DO NOT over pressurize it. If you do, THAT is what the balloon is doing. The balloon is acting as a relief valve. You just want to add enough air to inflate the balloon. Once it inflates stop the pressure!

First test - does the balloon stay inflated?

If it does then:

Second test - get a spray bottle of soapy water. Spray every seam. Watch for air bubbles. No bubbles - No leaks! Hurray! Bubbles - Leaks! Boohoo!

One last statement. If you find leaks, you will not fix the leak by putting more ProSeal on the outside of the tank. You will have to seal the leak(s) from the inside. Yes Van’s sells repair kits for tank leaks.

Good luck! Hope you don’t have any leaks!
 
I think he wants to test the tank forward of the aft baffle before installing the baffle, so no ballon, or other pressure test yet. There’s nothing wrong with filling it with water with the open baffle end up, and seeing if you have a leak forward of the baffle area. Make sure you let it dry completely after the test before sealing it up, and then do the ballon test and soapy water spray after the proseal cures for about a week.
Is gasoline more likely to leak than water is? I think not. A molecule of water is much smaller than a molecule of petrol, even though the density of the gasoline is lower. If you look at the molecular weight of petroleum, it is much higher than water.
 
Last edited:
I think he wants to test the tank forward of the aft baffle before installing the baffle, so no ballon, or other pressure test yet. There’s nothing wrong with filling it with water with the open baffle end up, and seeing if you have a leak forward of the baffle area. Make sure you let it dry completely after the test before sealing it up, and then do the ballon test and soapy water spray after the proseal cures for about a week.
Is gasoline more likely to leak than water is? I think not. A molecule of water is much smaller than a molecule of petrol, even though the density of the gasoline is lower. If you look at the molecular weight of petroleum, it is much higher than water.
Got it. So, yes what I described is after the tank is sealed.
 
Clay -

I did the same thing on my 10. Before putting the back baffles on, I filled it up with water. I left it in the fixture that I made to build them, put a piece of brown craft paper under it, then filled with water and let it sit overnight.

Once they were complete, I used the kit that Van’s provides to slightly pressurize the tank. I sprayed soapy water all over the exterior looking for leaks.

Both tests worked well.
 
Is gasoline more likely to leak than water is? I think not. A molecule of water is much smaller than a molecule of petrol, even though the density of the gasoline is lower. If you look at the molecular weight of petroleum, it is much higher than water.

I’m gonna question that on anecdotal evidence and explain my reasoning. I don’t believe it has anything to do with molecular weight or complexity. My reasoning = the airframe preservative oils have SGs greater and less than water. Also, no one would argue their molecular chains are less complex than water yet, they will wick through gaps that water will not ex = around bucked rivets.

I would suspect the liquid’s surface tension would be a prime driver here and water’s is relatively high. An example of such would be the behavior of a drop of water on a concrete surface versus a drop of lite oil, preservative oil, gasoline, etc. In summary, gas could find a leak where water might not.

There’s plenty of ChemEs or Chemist on here. If surface tension isn’t a primary driver of such, I’m ready to learn.
 
Last edited:
One last statement. If you find leaks, you will not fix the leak by putting more ProSeal on the outside of the tank. You will have to seal the leak(s) from the inside. Yes Van’s sells repair kits for tank leaks.

When I complained to Vans about leaking rivets on the top of my QB tanks, I was instructed to seal the leaks with proseal applied on the exterior. I was told that they repaired similar leaks in the factory demonstrator that way. YMMV

-Marc
 
Tank leak testing

When I complained to Vans about leaking rivets on the top of my QB tanks, I was instructed to seal the leaks with proseal applied on the exterior. I was told that they repaired similar leaks in the factory demonstrator that way. YMMV

-Marc
Well, I hope your repairs work however you go about doing them. I am speaking from experience but I am just some stranger on a website giving advice. All you have to go on as to whether I know what I am talking about is my word of experience. So accept my advice as wise or don’t accept it.

Whoever you talked to at Vans also is a stranger but, also, he may or may not be speaking from experience. What he does have over me is Brand Recognition. So, you can also accept his advice or not. Doing so does not necessarily mean his advice holds more merit than mine. It then becomes your job to determine for yourself which path to follow. Trust advice from a stranger with Brand Recognition. Trust advice from a stranger with self professed anecdotal experience. Or, do your own experiential work on the problem and find out for yourself the best course of action.

If you choose the latter, be sure to come back here and tell us about your newly acquired knowledge on the subject. I will be one of the first to acknowledge your experience as a valuable addition to the group knowledge on this website.
 
OK OK OK I will add my three cents into these thoughts.
No I'm not a chemist or a brain trust, I'm just old and full of personal experiences.
So, from personal experience water is a lot more dense than fuel or oil.
Think about it. Water beads up on oil or wax. If there is the tiniest bit of oil in/on that area you are testing the water is not going to get it to leak through.
Pour some gas or oil on an area then put some water on it to see for your self.
Now air will go through, But, without a back baffle no air testing can happen.
I built a pair of tanks for my RV-6A this last spring. Balloon tested them to be good. Installed and WaaLaa 20 hours later found a rivet in the bottom of the tank weeping. Hadn't painted them yet. Removed tanks drilled out rivet and installed a blind rivet. All good now 50+ hours later.
All that said I applied some vinyl wrap to the wing tank area and have three rivets on top where the vinyl has bubbled up. I just poked a pin in the bubbles and pressed the vinyl back down.
Question is the rivet leaking? I suppose. But it doesn't leak fuel.
If I was going to "test" before closing up I would use 100LL because it has blue dye in it. BUT, I would never do it!
Watch the Scott's Van's tank build video and have at it.
My three cents worth Art
 
My experience

Just me experience.
I would never fill a partially built tank with water or heaven forbid, avgas without the rear baffle on; there is too much risk of damage if a wood fixture breaks. The rear baffle is structural.

I followed the plans and. Built my tanks. Leak test using a manometer to avoid over pressurizing. ( i just blew in the tank with a swig of water in my mouth, once i got to the point i couldnt blow anymore i put the water in the manometer.) let it sit over night. One tank showed a change in manometer reading. Soapy water showed the leak on rear baffle. I was able to pop the ricets, spread the baffle, add more proseal and re rivet. Next test showed no manometer change ( except for weather changes) and soapy bubbles showed no leak.

When leak checking, one needs to use something that leaks easier than avgas. Air does that, and water with soap ( to reduce water’s surface tension to less than avgas) shows up leaks. In fact, soapy water shows up leaks down to less than any leak gauge seems to be capable of.

My lesson learnt: the rear baffle causes most of the problems. It takes lots of pro seal, is a large area to butter in a short time, and tends to wipe away the proseal during assembly.

Lastly if you find a leak, one can use the RV3 hub caps to patch the rear baffle holes, these will also provide a little more tank volume when all done. ( i am not sure if RV 3 hub caps are still available)
 
I have nothing to say about testing a partially completed tank.

For testing a completed tank, I have a process:
- Do the ballon test as discussed in the plans. Use a lot of soapy water. This test will find the gross leaks.
- For the weeping rivet problem, AVGAS is your medium. With the tanks off the wing, I put 5-7 gallons in each one, then let them sit on the hangar bench for a few weeks. I have white paper towels underneath the tanks to act a witness paper. After I’m happy with one side of the tank, I turn the tank to the other side and repeat. Then I let is sit on the skin sides with leading edge up. Depending on the tank orientation, the appropriate fuel/vent connection is capped and the other left loose so the tank can breath during temperature changes to prevent over/under pressurization.

Much better to find the weeping rivet before the tank is on the plane, and before paint.

A side benefit is this also provides the initial tank flush. The fuel drained out of the tank goes through a cheap inline filter so it can be used again. Once the tanks are on the plane they get another flush during the fuel sender calibration process, and then when verifying fuel flow to the engine.

Carl
 
First off, I agree with you that a repair needs to be done from the inside of the tank. For the record, this was not "next" to the wing walk area; it was the wing walk area that is supposed to take the weight of people walking on it. Also, I wasn't asking for warranty repair, I was willing to pay for it and I informed them of that in a subsequent email.

This is a copy and past of the email I received from Sterling Langrell of Vans who spoke with his "boss":

"I did receive the pictures and have been looking over them and discussed this with my boss. The location of the leaking rivets is a place we have seen from time to time as it is a location that can potentially get some foot traffic being right next to the wing walk area, sitting or walking on sealed rivets can cause the seal to fail. Given that this is a second hand kit paired with the age of it I am not able to offer a "warrantee" repair on this tank.

We have had similar leaks on our factory 10 caused by people stepping and sitting on this area. We have had luck with sanding the paint away and applying a thin dime size dollop of tank sealant on the exterior. There will be a small amount of buildup on the outside but the tank sealant can be painted and is not very noticeable in the end

Sterling"

Whoever you talked to at Vans also is a stranger
 
Last edited:
SNIP
This is a copy and past of the email I received from Sterling Langrell of Vans who spoke with his "boss":

"We have had similar leaks on our factory 10 caused by people stepping and sitting on this area. We have had luck with sanding the paint away and applying a thin dime size dollop of tank sealant on the exterior. There will be a small amount of buildup on the outside but the tank sealant can be painted and is not very noticeable in the end

Sterling"

Yep - I recall other RV-10 owners developing tank leaks after some big bubba sat on the tank.

Have black non-skid on the wing walk, and tell people “if it is not black, do not step or sit on it”.

Carl
 
There’s plenty of ChemEs or Chemist on here. If surface tension isn’t a primary driver of such, I’m ready to learn.

Methinks so too. Indeed, water has the highest surface tension of any room temperature liquid, except mercury.
 
Back
Top