Fuel tank leak

Over the last few months I have had a rash of fuel leaks. Originally it seemed to be where I had made the tank strengthening modifications so I cleaned the area and coated it with Pro-Seal. I did extensive leak checking and all was well. Put the tank back in the plane and added about 18 gals. of fuel. After about three days there was a leak that flooded the bottom of the plane and removed most of the paint from the outside. I lost about 7 gals. of fuel.
Took the tank back out and found what appeared to be a small leak at the front right corner, cleaned it and applied more Pro-Seal. Did the leak check and let the tank sit for three days with fuel in it and leak checked again. All was OK so put the tank back in the plane. After four days it leaked again, this time about 2 gals. of fuel. I used a piece of paper slid under the tank and the leak seemed to be in the right rear corner. Removed the tank again and in the right rear corner the Pro-Seal was soft. Patched it once more with Pro-Seal and added some fuel and leak tested again but then decided to remove the tank top to seal it from the inside. After draining the fuel and removing the tank top most if not all of the Pro-Seal on the inside was soft and sticky, I was able to pull out the finger strainer with very little force.
The tank had been assembled over several weeks so the Pro-Seal was made up in different batches but all from the same can and had cured for at least a year before adding fuel. All the fuel I have used was auto gas that probably contained Ethanol except for a one time filling of 100 LL. After the tank top was removed and the fuel had evaporated the Pro-Seal started to harden again which, I think, was why it took a few days to soften and leak again. What I would like to do now is remove most of the Pro-Seal from the inside of the tank and coat the seams with a Flame Master A product or something that is brushable to "protect" the Pro-Seal as I will not fly with this fuel tank until I am sure that it will not leak again. I don't think that I mixed the Pro-Seal incorrectly. Is it possible that I got a bad batch of Pro-Seal? Has anyone else seen something like this? I could just make a new tank but will still have the problem of a suitable sealant. Van's has switched to a Flame Master product, is this part of the reason? Any thoughts?
Henry

Did you mix the tank sealant by weight? How old was the sealant? Mogas never caused my tank to leak. I rarely use 100LL.

Are you applying from the out side? If so, that will never give a proper seal.
If you are applying to inside of tank, you are way faster at repairs than me.

The original Pro-Seal used was mixed by weight on a digital scale. The later smaller batches were also mixed by weight on a one-to-one scale using ten quarters for the resin? and one quarter for the hardener, accurate to the weight of a postage stamp. The Pro-Seal came with the kit and was less than 6 months old when I built the tank. I am still at a loss as to why the finger strainer was still tacky and easy to pull out. I am still thinking that the seams of the tank where the surface area of the Pro-Seal that is exposed to the fuel is very small and are OK but the corners where the surface area exposed is large, as is the finger strainer, are degraded and soft. It only takes a couple of hours after the fuel is removed for the Pro-Seal to harden up some for the leaks to stop. I think that is why that when it leaks in the plane and one removes the fuel then the tank it is enough time for the leaks to be gone. I did the leak test at about 3 psi way, above what is recommended, and the tank does bulge but does not leak. The Mil spec. that I looked up did not mention ethanol exposure; Joe is the Mogas that you use have ethanol in it? Most leak sealing was done on the inside through the fuel level sender plate.
Henry

krazydoc33 said:

The Mil spec. that I looked up did not mention ethanol exposure; Joe is the Mogas that you use have ethanol in it? Most leak sealing was done on the inside through the fuel level sender plate.

Click to expand...

Hundreds of RV-12's are operated on autofuel with ethanol (including Vans Aircrafts own demonstrator for most of its 1000+ hour operational life) with no problems reported so I doubt that is the cause unless there is something out of the ordinary with your particular situation.

There is THIS Service Bulletin that was originally issued as a caution to RV-12 owners but I am not aware of a single instance of an RV-12 fuel tank being damaged by phase separation.

Joe is the Mogas that you use have ethanol in it?

Click to expand...

I have purchased mogas from many different gas stations and do not know what the ethanol content was.
Since you mixed the sealant in the proper ratio, I conclude that your sealant was defective from the manufacturer. It should not soften up.
One psi of air pressure is enough to test for leaks. More than that risks damaging the tank. I used a homemade manometer (water in a plastic tube) for monitoring the pressure. The water level varied over time with changes of atmospheric pressure.

I use 93 octane 10% ethanol fuel exclusively with no issues. There's a lot of bad info out there regarding ethanol enhanced fuel blends.

I agree with Joe. The sealant should not be sticky after a day or two. Perhaps the white activator is/was defective. I feel for you. I hope you find a fix short of building a new tank.

I wrote up some notes on what I had found and some tests that I did. Some of it is repetitive from previous posts but I wanted it to be self contained for future reference.

"Monday Oct. 3rd. Morning. Remove fuel tank and leak check with soap solution and air pressure of approx. 2 psi no leaks found. Added about 7 gals. of auto fuel and checked for leaks. No leaks found.
Tuesday Oct. 4th. Checked fuel tank for leaks. No leaks found.
Wednesday Oct 5th. Morning. Checked fuel tank for leaks. No leaks found. Checked fuel tank in the afternoon and found slight leakage at right rear corner.
Thursday Oct 6th. Pro-Seal at right rear corner was very soft, could poke the end of a plastic "q-tip" through the corner. Sealed the corner with Pro-Seal from the outside, not as a permanent fix but to be able to continue leak checking.
Friday Oct. 7th. Spoke with Van's builders support about problems with Pro-Seal and Auto gas with Ethanol, not a common problem. Checked fuel tank for leaks. No additional leaks found.
Saturday Oct. 8th. Drained the fuel tank and removed the front fuel level sensor plate and the top skin. A lot of the exposed Pro-Seal in the bottom half of the tank was tacky, there was no Pro-Seal left in the tooling holes and the fuel finger strainer was very easy to remove. There was a big difference between the Pro-Seal at the top skin seal and then lower rivets that protruded through the tank. Left tank open with top and access plate removed.
Monday Oct. 10th. Inspected the inside of the tank. The old Pro-Seal had hardened up significantly, it was still a little pliable but not nearly as tacky and I could no longer remove the finger strainer after a moderate pull was applied. This is all qualitative but I do not have the equipment to do a more quantitative assessment. I am conducting two separate tests. I have immersed two popsicle sticks that were coated with Pro-Seal in jars. One with auto gas containing Ethanol and the other containing 100 LL. I will check these periodically. The other test is on the tank. The tank is tipped so that the fuel, 100LL, that I added covers the outlet nipple and the finger strainer. I will check that one every few days.
Wednesday Oct 12th. The two popsicle stick samples are about the same, a little less tacky than on Monday but the Pro-Seal has about the same hardness on each of them.
Thursday Oct. 13th. The two popsicle sticks have not changed, no difference between auto gas and 100 LL.
The fuel tank test with 100 LL is a little different. It is possible to remove the finger strainer with some effort. I think that the difference is due more to the length of the time of immersion rather than the type of fuel. The tank was built in early 2012 and the sealing waqs done on 3/17/2012 but it was not until July of 2013 that fuel was added to the tank. The rest of the tank seals seem to be OK, it is only in a few places that there is a problem. I am confidant that the problem is in a few places that were probably contaminated prior to appying the Pro-Seal as other areas that were sealed at the same time with the same batch are OK. I have no idea what that contiminate could have been or how it got there or why it took so long ( from July of 2013 to June of 2016) to be a problem.
My conclusion is that there is no problem with Pro-Seal and the ethanol in auto gas. As I have no idea if there are any other ?weak spots? in the Pro-Seal I will replace the tank with a new one."
Henry

Thanks for the very complete feedback. So often we never hear back the final resolution of an incident. Just a thought. Is it possible the inconsistent curing of the Proseal could be the result of insufficient mixing of the product and catylst?

What's the cure time required for proseal?

krazydoc33 said:

I wrote up some notes on what I had found and some tests that I did. Some of it is repetitive from previous posts but I wanted it to be self contained for future reference.

The tank was built in early 2012 and the sealing was done with Proseal on 3/17/2012 but it was not until July of 2013 that fuel was added to the tank.
Henry

Click to expand...

Interestingly, my fuel tank has the same history, built early 2012 and put to use in September 2013 when the plane first flew. Although I have not experience any leak in the past three years, I just found 2 leaks: one on one screw of the front plate, the other involving several screws from the sight window. The screws involved were not as tight as the rest and tightening them did not stop the leaks. I will open the tank by removing the top skin and try to repair these leaks using Proseal.
I have used Ethanol free gas until six months ago when I had to turn to mogas with Ethanol by lack of Ethanol free gas with high enough octane.
It would be good to hear from builders with a comparable time frame to find out if Proseal degrades in time.

As I have mentioned in previous posts on the subject, the RV-12 prototype (N412RV) has been operated almost exclusively on auto fuel with ethanol (except for when it is traveling cross country). It probably has at least 900 hrs of operation with this fuel.
No sealant degradation has ever been noted..

rvbuilder2002 said:

As I have mentioned in previous posts on the subject, the RV-12 prototype (N412RV) has been operated almost exclusively on auto fuel with ethanol (except for when it is traveling cross country). It probably has at least 900 hrs of operation with this fuel.
No sealant degradation has ever been noted..

Click to expand...

This is good to have this information about N412RV but it is just one RV-12 among a couple hundred that have been flying 3 years or more (I mean all RV-12 that were built and flew before Henry and I flew ours end of 2013). The batches of Proseal involved were produced several years apart and may have different stability in time. The fact that Vans replaced Proseal with Flamemaster is not a confidence builder in favor of Proseal (although if quality was the reason for the switch I believe that Vans would have let it known to its customers).

I am pretty sure that Flame Master is the only sealant that Van's has sold, since the introduction of the RV-12 kit.

rvbuilder2002 said:

I am pretty sure that Flame Master is the only sealant that Van's has sold, since the introduction of the RV-12 kit.

Click to expand...

Scott: from Vans parts catalog

"MC-236-B2 (Quart Kit)
The recommended sealer for RV fuel tanks, replaces older ProSeal. Limited shelf life (approximately 4 months from purchase). Order when ready to seal tanks. ..."

The funniest part is that you cannot find it with a search on Flamemaster, it will popup only with a search on Proseal

I don't think that it was a mixing or a ratio problem. Those would have shown up in other areas too. One thing I forgot to mention was the Pro-Seal in the "tooling holes", it has completely gone! It is still around the holes but the holes are completely open and free of any Pro-Seal. I will post pictures later. This MAY be a good place to check as it is easy to see with a miniature camera through the fuel filler neck or through the fuel sender plate cover.
A new tank, at $400.00 to $800.00 is relatively cheap insurance but I am not happy having to "invest" that much when I have no idea why there was a problem in the first place.
Henry

rvbuilder2002 said:

I am pretty sure that Flame Master is the only sealant that Van's has sold, since the introduction of the RV-12 kit.

Click to expand...

WingedFrog said:

Scott: from Vans parts catalog

"MC-236-B2 (Quart Kit)
The recommended sealer for RV fuel tanks, replaces older ProSeal. Limited shelf life (approximately 4 months from purchase). Order when ready to seal tanks. ..."

The funniest part is that you cannot find it with a search on Flamemaster, it will popup only with a search on Proseal

Click to expand...

I guess I don't understand the point you are trying to make......

It seemed like you were questioning whether a change in sealant at some point could make a difference in the performance outcome that people have with their RV-12 tanks?

As you can see in my post, I said that "Flame Master is the only sealant that Van's has sold, since the introduction of the RV-12 kit.
I never said Van's has never sold Proseal. Van's has been in business for 40+ years and has sold Proseal in the past.... but not to RV-12 builders.

My final note on this topic. I checked the two samples in the two fuels on Oct. 19th. That's about two weeks from immersion. The 100 LL had hardened up a little more and was no longer tacky. The Pro-Seal in the Auto gas sample had come off of the popsicle stick and was at the bottom of the glass container. The small blob of Pro-Seal was just as hard as the 100LL sample and also was no longer tacky. There was still a coating of Pro-Seal on the stick. I THINK I now have an idea of what may have happened. When measuring the two components I use a stick and a container for each part A and B to equal the balance on the scale. I then put the Pro-Seal into the activator container to mix them together. When I applied the Pro-Seal to the fuel tank I used the SAME STICKS to put it on and there may have been traces of either the Pro-Seal or the activator left on the sticks and contaminating some areas. The Pro-Seal and the activator do not migrate together with time as do epoxy resins. It would be permanently weak. I have suggested to Van's that they add to their note on the use of Pro-Seal to discard the mixing sticks and use fresh ones to apply the material to the parts to be sealed.
Comments welcome.
Henry

Use one stick for scooping out the resin and another stick for scooping out the hardener and a 3rd stick for mixing and perhaps a 4th stick for filling a 35 cc syringe (available at farm supply stores). The syringe makes a miniature calking gun. When applying sealant, more is not better. The excess will ooze out of the joint and could clog the fuel system.

krazydoc33 said:

My final note on this topic. I checked the two samples in the two fuels on Oct. 19th. That's about two weeks from immersion. The 100 LL had hardened up a little more and was no longer tacky. The Pro-Seal in the Auto gas sample had come off of the popsicle stick and was at the bottom of the glass container. The small blob of Pro-Seal was just as hard as the 100LL sample and also was no longer tacky. There was still a coating of Pro-Seal on the stick. I THINK I now have an idea of what may have happened. When measuring the two components I use a stick and a container for each part A and B to equal the balance on the scale. I then put the Pro-Seal into the activator container to mix them together. When I applied the Pro-Seal to the fuel tank I used the SAME STICKS to put it on and there may have been traces of either the Pro-Seal or the activator left on the sticks and contaminating some areas. The Pro-Seal and the activator do not migrate together with time as do epoxy resins. It would be permanently weak. I have suggested to Van's that they add to their note on the use of Pro-Seal to discard the mixing sticks and use fresh ones to apply the material to the parts to be sealed.
Comments welcome.
Henry

Click to expand...

Hmmmm - kinda ties in with my theory put forth in post#9 above....

Gas odor in cockpit

I have 167hrs on my RV12 and just started smelling gas when I open the cockpit in the AM. I use ethanol free Autogas. Checked all lines and around the tank but see no fuel puddles or staining. Considering using a dye (ACDelco 10-5045 Multi-Purpose Fluorescent Leak Detection Dye - 1 oz) to show me where my problem is. Has anyone tried this before?

found leak

Well with the dye I found my leak. With the tank out, air testing both with soap and submerging did not reveal the leak. Only when in the plane, filled over 1/2 full did the odor appear. With the dye the leak reveal itself along the right forward support bracket. Apparently the load induced some stress on a rivet which then leaked a small but noticeable amount. I ordered a new tank. If someone wants to try and fix this one let me know.

Fuel Tank Leak

I have also had a fuel smell in my RV-12 ever since making the fuel tank attach modification SB 13-12-19 in early 2014. This past condition inspection in Jan 2017 I finally found a very small leak coming from the right front attach hole. I've recently tried to seal it with fuel resist Permatex like gasket maker but have little faith. Will probably end up purchasing a new tank. This topic has caused concern, thus the reason for my first post. The RV-12 is a great airplane, but would like to figure out the fuel tank issues to make it even better. Does anyone know of a plastic or fiberglass tank that would work? I'm not looking forward to dealing with the sealant again.

Thanks,
Marty
S/N 120439

ben barron said:

Well with the dye I found my leak. With the tank out, air testing both with soap and submerging did not reveal the leak. Only when in the plane, filled over 1/2 full did the odor appear. With the dye the leak reveal itself along the right forward support bracket. Apparently the load induced some stress on a rivet which then leaked a small but noticeable amount. I ordered a new tank. If someone wants to try and fix this one let me know.

Click to expand...

Ben,

This has been an issue I've been concerned about. If I could get your bad tank I would take it to a guy I know who builds racing fuel cells (bent and welded to shape. Would weigh a little more than factory configuration, but would sure give me piece of mind. Once he got the deminsions and angles figured w/ attach brackets figured out I'm sure he would make them for us. Let me know if interested..

Another issue that I'd like to look at at the same time is having a sump in the bottom that would protrude under floor where the existing fuel lines attach...thus eliminating the "no nose up Flight below 4 gallons. I.e. What to do if missed approach has to be made @ landing destination.