What's new
Van's Air Force

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

Third (?) Method for Sealing Fuel Tank Leaks

avatty

Well Known Member
In pressure testing my fuel tanks I found leakage at the top of the tank attach bracket on each tank, in the upper corners of the small “D” formed by the bracket and the tank leading edge (KIA pp. 18:04-05). The leaks were enough that I did not think Loctite would do the job, and the only other technique the factory recommends involves cutting access holes, which I was reluctant to do.

Research on this website uncovered a few posts that mentioned the possibility of sealing leaks by applying tank sealant diluted with MEK to the outside of the tank, and drawing it inward by reducing air pressure within the tank. An equal number of posts said that diluting tank sealant is prohibited, so the idea was a non-starter. This led to further research, including a discussion with a qualified chemist. I also mixed up a few test batches, which seemed to set in the usual way.

The method has been quite effective at addressing the leaks, at least so far. I used a hokey $7 HF “hand transfer pump,” which works both ways, and the manometer (water level) described in the instructions that come with the factory leak test kit. The maximum vacuum applied was 3/4 to 1 pound, which was enough to pull in the diluted sealant and stop the leaks. At one leak source I could actually see the sealant being drawn into the tank, and the water level in the manometer dropping several inches before suddenly stopping. The tanks are now proving airtight after being left pressurized to 3/4 pound for several days.

As mentioned, I was not able to find much discussion of this technique, and what I found was divided. However, in my case it seems to be working, and preferable to cutting access holes unless nothing else works. Maybe I just missed it in my research, but felt the technique was worth mentioning here.
 
I had a Beech Musketeer I work on that had several leaks in the right wing. Using a HF vacuum pump and a stethoscope I was able to find the leaks in a very inaccessible spot and stopped the leaks by applying proseal witih a Semco gun with 4ft of aluminum tube attached to the nozzle end. The vacuum pump was able to draw in the unthinned proseal. The only other way to fix the leaks would have required wing removal.
 
Please post updates on this same thread if they ever do start leaking there. It would be a good data point.
Cutting the holes in the baffle is not very difficult, so do not be afraid of doing it in the future. If you do, sand a chamfer on the bottom side of the repair plate. This will help future removal much much easier by allowing you to get a putty knife under the edge and cut through the proseal.
 
Type A tank sealant is considerably thinner than the more common Type B. Might make a better starting point.

Dave
 
I wonder if this would work for the notorious B-nut leaks at the vent line bulkhead fitting in the wing root (RV-10). Sounds easier than custom wrenches through the sender hole...
 
lose, loose, loosen, >>> look them up

I found my B nut finger tight upon pulling the tank and sender. It explained why that side always vented fuel onto the ramp on a warm day. I tightened it firmly. Probably should have used a bit of Permatex on the threads. Next time.
I do have one small leak left. (my pro seal hardened in one hour outside at 73 F and I lost time to methodically finish.)
Will try to draw a small vacuum and fill the root gap with my latest 2 oz. proseal purchase. A small fluid transfer pump and hose from HF seems to pull about the right amount. The hose even fits the air vent fitting just right... out of the box.
Crossed fingers...
 
third try on leak

I removed the right tank from the RV9A and did a bead of Pro Seal through the tank root opening. Let it cure and it still leaked a bit. Read this thread and ordered more Pro Seal and drew a slight vacuum on tank. Still a small bubble blowing hole (almost as small as diameter of a straight pin) when light pressure applied to the tank. (exterior Pro Seal)
Drew a vacuum and applied third attempt with a bead of exterior Pro Seal.
No bubbles, so assembled all and flew it. Nice flight and I was again a happy owner.
Now, two days later, with the weight of 12 gallons sitting in the tank, a blue stain appeared at the root.
Not a happy owner. Again
Am I learning that exterior applied Pro Seal is just a "bandaid" that has a tendency to come unstuck?
 
Mine needed three applications on one tank, four on the other. It was a painstaking process but much less work than creating an access hole. I was thinking about using loctite at the end, but the tanks tested ok before I got there.

Are you diluting the sealant with MEK? David Paule's statement about different types of tank sealant is interesting, too, but you would have to find some.
 
I removed the right tank from the RV9A and did a bead of Pro Seal through the tank root opening. Let it cure and it still leaked a bit. Read this thread and ordered more Pro Seal and drew a slight vacuum on tank. Still a small bubble blowing hole (almost as small as diameter of a straight pin) when light pressure applied to the tank. (exterior Pro Seal)
Drew a vacuum and applied third attempt with a bead of exterior Pro Seal.
No bubbles, so assembled all and flew it. Nice flight and I was again a happy owner.
Now, two days later, with the weight of 12 gallons sitting in the tank, a blue stain appeared at the root.
Not a happy owner. Again
Am I learning that exterior applied Pro Seal is just a "bandaid" that has a tendency to come unstuck?
short answer! YES!!

Long answer, trying to seal a leaking tank from the outside is a losing proposition. Any leak needs to be addressed from the inside of the tank. Some chemist may decide to chime in here but what I have observed is that when the tank sealant, fuel, and air meet (on the outside of the tank), the tank sealant begins to soften and before long allows the fuel to make it to the “free world”.

Oh, I almost forgot. Ask me how I know!
 
The best band aid of external application of sealant (any kind) was posted here many moons ago. That was where a guy took a syringe and cut the tip off, then used it to seal around the rivet externally and created quite high pressure to extrude sealant into any opening.

Vacuum on the inside, within the structural limitations of the tank are highly suspect - and personally I would have to be shown some physical evidence to believe it pulled in any sealant in the right place. We are talking maybe 0.5 psi where as the syringe could safely provide 50 psi or more. 100X.

Rocket Bobs method of artful application of internal pro seal (if I understood correctly) with a vacuum on the external side of the rivet (suction cup and vacuum pump) is pushing sealant the right direction. But has a limited reach for an RV tank.

Full disclosure - I have fingers crossed for no leaks! None in two years.
 
try again

Back at the beginning of this saga, I thought the leak WAS at a rivet. Turns out a nearby rivet had never been set. ( a paint bubble pointed to the error) I corrected that but it was not the real source of the fuel weep.
I tried the syringe and loctite, but since the hole was elsewhere I was chasing a fast rabbit in the wrong direction.
Now I might go back to the pressure technique for my fourth try. Before I remove all 21 bolts, 52 screws, two B nuts and a fuel sender wire. Not to mention scraping all the new Proseal off the screws where I gooped all the extra on for "good measure". Why is it good measure rarely turns out good?
After drawing a vacuum and seeing the viscosity of Spruce's normal Proseal at about 72 degrees, I am pretty sure none of it was actually "drawn" into any hole anyway.
From all my reading though, I am still mystified, like some others, why Proseal can be exposed to air inside a tank and stay firm and flexible... and then turn to a sticky mess when outside the tank and only minimally exposed to 100LL.
Oh, the mysteries of flight....
 
Last edited:
Back
Top