Fuel Tank Leaks

Flygirl1

Flygirl1

Patrol#241FLYING
Looking for advice on how to fix a leak in our quick build kit fuel tanks. We put 4 gal's of fuel in each tank and two out of four of them have evidence of leaks at the baffle weld. These are very small leaks, more like seeps. I'm thinking not worth the effort of taking them out and welding. Hoping someone has an idea on what I can put on the exterior. I would probably put "whatever" on every single one on the bottom just in case new ones appear in the future. D-n-D N241BP/N240BP
 
I am sure that the zinc-based aluminum brazing rod would work well too. I've done quite a bit of playing and destructive testing with it lately. Melts ~ 700F. Propane torch is more than enough heat. Bought my braze rod at Harbor Freight. And as I mentioned in an earlier post ... if you don't like the looks of it, just wipe it off with a dry cotton rag (I learned the hard way that polyester makes a mess) :(


I guess I don't have to remind you to clean/vent the tank wellfirst. Maybe put a chunk of dry-ice in it for an hour before bringing the torch around.
 
Donna - if your tanks are leaking it would be the very first time ever that fuel tanks made at our factory have had leaks. Lots of tanks made and never once, until now, a report of one leaking. You think it is leaking from one of the rivets that was welded over? Proseal is good stuff but messy. The aluminum surface has to be very clean for it to stick and work. Mark
 
Seeing as you've reported two tanks leaking, whereas many hundreds of earlier tanks from the same production line haven't leaked - it's good cause to investigate what's going on. It seems incredibly unlikely that it's just bad luck.

It would be great to see a photo of the offending tanks. Are you 100% sure they are leaking? Sometime the fuel gets mixed with some pen markings on the tanks, and it looks just like a slow leak (like dried fuel dye).

Alternatively, is there any chance whatsoever the tanks could have been squashed - perhaps stuff stacked on top of them? Or otherwise dropped or impacted in any way? That kind of trauma could lead to small cracks forming in the tanks, and the welds are the first thing which would crack.
 
Just my opinion, now is the time to remove them and make a permanent repair. As frustrating as it may be I would pull the tanks and have them welded if they are indeed leaking. Anything less than that and it would be something that was always on my mind.
 
I'm absolutely sure these are leaks and I can say I'm also sure the tanks haven't been compromised. Also let me re-peat these leaks are SMALL, very small. One had left enough fuel on the tank to realize I had a leak, which led me to closely inspect every other one. We have a total of two leaks at two welds at the baffle rivet on two tanks. I would say only one of those might be worth welding, (probably not) and if I did nothing to the others, I could accept that as they are so small. I imagine most people wouldn't even know they are there. I would post a photo but my camera isn't good enough to show it. I thought if I could "smear" something on all the welds, that would be good enough and I wouldn't worry about it. If I ever over flow on filling and fuel found it's way into the wing, that would be WAY more fuel than what I am seeing here. I just don't know what would work best. Sounds like pro seal might do the job with the proper prepping. While on the subject what are people using to seal the hole around the filler? Thanks for your thoughts and opinions! D-n-D
 
I haven't done it but I remember seeing something with regards to sealing leaking rivets on Van's RV tanks which might be similar.

Basically involved applying a slight vacuum to the tank and using thinned ProSeal or some particular Loctite product applied to the offending rivet.

Would be worth a web search.
 
Empty tank, clean surface with stainless brush and lacquer thinner, wipe dry, apply a coat of jb weld liquid metal, allow time to set up. Your finished.
 
Rather than fight them for years, pull em and make a real fix with a TIG torch. Why use JB Weld or Proseal or anything like that? That's for guys with wet wings.

As for just living with a tiny leak, don't do it. My friend has a leak he's chased for years in the wing of his Maule and it's quite annoying to always be smelling fuel when you're not moving.
 
Last edited:
No Title

Well I found two more teeny tiny leaks in one tank this morning. (One of the guys says having a "teeny tiny" fuel leak is like being a "little bit" pregnant. LOL ) So as far as welding, which ones do you weld. If I put another 15 gal. in will the weight cause more over time? Who knows, so I still think pro-seal is the way to go. Fortunatly these wings have really good inspection panels to keep an eye on things!! Mark G. thanks for your thoughts and your offer to help. Really appreciate that. If pro-seal doesn't work, I'll give a shout. Donna
 

Attachments

  • photo3014.jpg
    photo3014.jpg
    4.9 MB · Views: 0
  • photo3015.jpg
    photo3015.jpg
    5 MB · Views: 0
Autozone, etc, sell a two part epoxy type thing that is specifically made for fuel tank repair. I've used it on several things and it seems to hold up well. Its a two part putty type that you kneed together until its a solid color and then put it in place. It sticks really well, can even be applied to a wet, actively leaking tank. hardens up similar to jb weld, but its made to be fuel resistant. I would use that stuff and then just for fun, put pro seal over that. If you do go that route, one time when I bought it it came as two strips of putty, one black one white beside each other, and the last time I bough it it came as a single piece, about 6 inches long and round about the diameter of a tootsie roll, with the white putty in the middle and the black surrounding it. You just cut off how much you think you need and go from there. Good luck.
 
I like Pro Seal for this kind of repair. On my RV8 leading edge tanks built up with Pro Seal, it was used over screw heads and things very similar to those rivet heads.

Just one thing I will mention after talking to Dennis, the husband of Donna who made this post. Interestingly, they leak tested the tanks by pressurizing with air and placing a balloon on it. The balloon was still blown up the next morning. It is hard to imagine that any leak at all would not let the air out over night. But it didn't. Must be very, very small leaks it would seem. Mark
 
Flygirl1 said:
Well I found two more teeny tiny leaks in one tank this morning. (One of the guys says having a "teeny tiny" fuel leak is like being a "little bit" pregnant. LOL ) So as far as welding, which ones do you weld. If I put another 15 gal. in will the weight cause more over time? Who knows, so I still think pro-seal is the way to go. Fortunatly these wings have really good inspection panels to keep an eye on things!! Mark G. thanks for your thoughts and your offer to help. Really appreciate that. If pro-seal doesn't work, I'll give a shout. Donna
Click to expand...

Sounds like you have 4 or more spots where you have fuel seeping around rivets? If it was me I would pull them and ask Marc for an exchange - just asking for trouble going forward.

Marc - as you mentioned you have shipped hundreds of tanks with no problems before - so not giving you grief - just a curiosity point for me, how are they testing the tanks at the factory?

-paul


Sent from my iPad using Tapatalk
 
I have had good luck with the fuel tank repair compound sold at auto parts stores. Slice off a small piece of the stick, roll it between your fingers to activate it then mash it over the rivet. Inexpensive and a lot less messy than proseal.
 
They test at two pressure levels. One higher and one lower pressure. Then use soapy water with glycerin. I can't remember all the ingredients for the soap[y water test. Mark
 
I am waiting for someone smarter than me to explain how Dennis did the balloon pressure test with the balloon staying blown up over night - and then subsequently finding these small leaks. That doesn't make sense to me. It would seem air would escape from ANY leak, however small. Someone enlighten me how this could happen. Mark
 
I have to say the balloon test didn't work for me either. My test showed no leaks on my fuel system, but when I added fuel I found almost all my joints were not tight enough. I can't say why the test doesn't work well, but I can corroborate that the balloon test is fallible.
 
I'm certainly no expert on this topic, but I read on one of the other airplane building forums (can't remember which one now) that "pressure testing" of fuel tanks is a good way to cause fuel leaks, as the metal responds to the internal pressure by bowing outward, which can actually create the leaks - especially around rivet heads. Their suggestion was to plug all the drain points, cover the tanks with a thin film of talcum powder, then just fill the tanks with water, and look for leaks overnight. I have no idea if this is "sound" or not - just reporting what I read elsewhere...
 
Comparing gasoline to water...gas has zero surface tension compared to water. Picture in your mind Tyler 2000 mesh screen...doesn't exist but imagine how small the openings are. A screen that small will hold water simply because of surface tension. Gasoline goes right through it even though the size of the gasoline molecule is like comparing Jupiter to the moon.


Sent from my iPad using Tapatalk
 
When I welded up my tanks, the last few leaks I chased down were small enough that the balloon didn't seem to change even after 8 hours(apply balloon in the PM when I closed up shop, checked when I left for work in the AM with no apparent visual change, check when returning from work and balloon was very slightly smaller.) I was ready to attribute it to changes in barometric pressure but very careful work with soapy water revealed what I hope are the last ones. Note I haven't put fuel in yet so for all I know there may still be leaks. Perhaps if one was a bit more scientific(put a couple marker marks on opposite sides of the balloon and use a flexible tape to measure/track the distance as opposed to "Does it look smaller?") might pick up the leak quicker.

My IA friend doesn't believe in testing tanks with water in that in his experience he has seen leaks which are small enough to pass gasoline but not water. There is filter media that proves this out(http://www.twpinc.com/wire-mesh-by-project/separate-fuel-from-water)

A leak from a rivet as described I personally would be fairly comfortable fixing with epoxy as it isn't likely to grow. A leak from a seam I would want welded as I could see a hard landing or turbulence causing the crack to "zip open", traveling along the seam out from under the epoxy with unpleasant results.

My experience with other fuel tanks outside of aviation (cars, tractors, motorcycles, etc.) is that both JB Weld and the "Fuel tank repair epoxy" make for long lasting repairs but the JB Weld approach requires the tank to be empty and the leak free of gasoline or it won't stick.

But my IA friend would insist on Pro Seal for an aviation tank on an experimental.
 
I just heard from Donna that they probably did not do a balloon test on the tanks. Just on the fuel system itself.

I also spoke with Bob Barrows about this. He said that over the years he has had a LOT of fuel tank leaks. He has repaired most of them with JB Weld and that has worked well for him. However, I still feel ProSeal is the best choice since it is made just for sealing tanks. Messy to use but a very good product for what it does. Mark
 
I was preparing to do the balloon test on my tanks, but now I'm thinking it might be worthwhile to simply fill them with gasoline (since that's what they're supposed to hold) and see if they leak. Sure, it's somewhat of a hassle to then have to drain the gas into other containers, but it might be an effective test.
 
I would start with the balloon test. Fix all the leaks you find by welding when it is a much easier(and safer) task without gasoline fumes to worry about.

My experience is that "the balloon" will find some pretty small leaks if you are patient and somewhat scientific in your approach. When I completed my leak chasing on the tanks I built, the balloon remained unchanged after 24 hours whereas the "last leak detected" had a visible impact on the balloon size after sixteen hours or so.

My plan is that if I find some leaks when I have already put fuel in the tanks I built, I will likely address them with JB Weld (or Devcon) or Pro Seal
 
Reading this thread is starting to spook me! Both tanks are mounted in my wings, but not hooked up or plumbed. I could remove them, blug all the bungs and fill them up or pressure test them. Re welding leaky rivets wouldn't be too hard at that point. It would be even easier (at this point), to just scuff the welds, clean them with MEK and apply a nice dollop of JB Weld on each rivet. I can think of several reasons to do this "prophylactic" fix and am hard pressed to think of a reason not to. So, what y'all think?

Bill
 
Bill, that is exactly what we are going to do, put a dollop on every rivet weld wether it is leaking or not. Just the ones on the bottom mind you. My tanks will never be full enough for long enough to leak out the top! ;-) I'm with Mark on this one--Pro-seal is made for the job intended. I don't give a hoot how messy it is if it works. Nobody will ever see it but me. The left tank in both our Patrols are 100% free of leaks at this time. We both have about 14 gal's of fuel in them. Also, now that it has been brought to your attention, when you look for them remember you'll have to look close as they are very small leaks and the last two I reported took 4 days to show up. Also the fuel we used was an alcohol free auto fuel. Don't panic at this, we are just putting it in to flush out the new system. But keep in mind some winter blends of auto fuel will dissipate a lot slower than 100LL so I guess it is possible that a leak this small might dissipate as fast as it shows up?? Something to think about. Donna
 
You must log in or register to reply here.
Back
Top