What's new
Van's Air Force

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

Why is my Pro-Seal turning to gooe?

pa38112

Well Known Member
I am doing my third tank leak repair in 8 years. Each time I find that the Pro-Seal (I actually use ChemSeal, but they are the same ???) has turned to a gooey past that I can scrape off with my fingernail. I always use frash material, mix it and apply it per the instructions, and it always seems to set-up properly. 4 years later and it has turned to gooe :( .
My tanks are almost always full, 100% 100LL usage. Can anyone provide some insight or guidenance so I dont have to doo this again?
 
external seems the softest. Ill pay more attention to the internal when I go back over to the hanger.
 
External is common, and so far, no one has been able to tell us why it reverts to paste. I did ask the Flamemaster/ChemSeal tech rep.

It appears to happen only in the presence of a fuel leak, which suggests any external leak repair with proseal must seal thoroughly or it won't last.

There is an old industry report suggesting peroxide involvement. I soaked cured samples in ordinary hydrogen peroxide for a year or more without reversion.

I have seen softened samples from inside a tank, but never goo.

I would love to know the answer, but the chemistry is beyond my pay grade.
 
My research

Here and across the web and my A&P IA

Seems that the dye ( blue ) in the avgas may be the culprit. Likely to see evidence external say due to a leak . The avgas evaporates off quickly leaving the dye. I recently replaced float sender by removing the root inspection panel. I too had softened sealant as described my OP. Seems full tanks / slosh/ wet environments, no issues, no evaporation. YMMV and no I don’t believe all on the internet. My A&P is sharpe and experienced.... he sees the same from time to time.
 
I think* the dye is a red herring. It is just what remains after fuel has leaked and evaporated. Basically, it is telling us "There's a leak here.". But I don't think it is causal.

The only leak I ever saw where I could clearly identify a root cause was a friend who hadn't mixed his proseal thoroughly. The streaks were evident in the material.

On my own tanks (filled with gas for 20 years), I have two rivets which are starting to show problems. Both on the right tank, both on the same rib. One on top, one on the bottom. Makes me wonder if I got the ratio off when mixing that batch (I mixed a small batch for each rib), or if I inadequately mixed that batch.
 
solution to leaks

I solved my leak problems . My 2001 tanks were built to leak. gas smell in cockpit since I bought it 3 years ago. Bought it knowing I was going to build new tanks. Finally got around to it. Needed flip flop for sure.
New ones all tested (48 hours and bubble tested good) and one installed. Working on the other today.
Oh ya the Proseal was soft and gooey inside of the tanks. And was slopped all over the place like somebody thinned it for install. Real mess.
I really like Scott's video for building. Last set was 28 years ago and I used a whole can of goop. This time I only used half a can and know I did a better job.
This time I used a digital scale. Great tool. For anybody not got it figured out yet. Put mixing pad on scale then turn it on. Automatically discounts the pad. Add Proseal then multiply weight factor by 1.1 Add hardener up to the new weight and you'll have the EXACT proportion of hardener. NO GUESSING!! WooHoo!!
I've been burning Costco premium for the last 2 years. 100 hours. Brown stains don't look good on white paint.
Art
 

Attachments

  • IMG_0294.jpg
    IMG_0294.jpg
    350.9 KB · Views: 367
Add Proseal then multiply weight factor by 1.1 Add hardener up to the new weight and you'll have the EXACT proportion of hardener.
Art

A suggestion (good for mixing epoxy, proseal, and paint):

Weigh out your smaller component first, then determine how much of the bigger component to add. Let's say the desired ratio is 1:10 by weight and you want to net 55 grams.

You weigh 5 grams of the "small ratio" component then 50 grams of the other one. If you mistakenly weigh out 51 grams of the second one, your proportions are off by about 2%.

On the other hand, if you first weigh the 50 gram component, then miss your 5 gram component by even a half gram, your proportions are off by about 10%.

Point is, it is easier to fine tune the larger mass than the smaller one, and missing a little on the larger mass isn't as bad as missing a little on the smaller one.
 
Mix it

Scott's video is the best media Vans has ever produced! ... a haul of great technique.

He doesn't describe tho how he mixes his sealer.

My method is to use the scale as described by Kyle on a stiff card (hotel room keycard), then pile it on a stiff stainless steel plate (1/16x12x12" ish) and mix using a 3" putty knife. A plate of glass would likely work as well.

You can scrape the plate clean in a pass, roll it over and squeeze it down again. Very quickly you'll see the color even out and there will be NO unmixed material anywhere on the plate. Zero

Having the putty knife is also a great way to spread the mix onto rib flanges, getting that pyramid shaped line of goo that Scott describes.

All of my mixes looked exactly the same shade when dry.
Cleanup: Scrape it all off and then a quick wipe with solvent.

Very tidy :)
 
Back when we were all chasing this stuff, I mixed three samples at three different ratios, 10-2, the correct 10-1, and 10-0.5. There was no significant difference in cured sealant. They all cured, and they all cheerfully withstood a long soak in straight MEK.

Remember, the black manganese is a catalyst, not a linking component.

The very dark color in the hydrogen peroxide was mostly due to corrosion...i.e. rusty can.

All samples were well mixed. Poor mixing would surely bring different results.

Polysulfide sealant was not intended for avgas or mogas. There are no test standards for avgas use with this sealant.
 

Attachments

  • Soak.jpg
    Soak.jpg
    923.1 KB · Views: 399
I solved my leak problems...(snip)
New ones all tested (48 hours and bubble tested good) and one installed.

Thing is, no one knows for sure how good their tanks really are until they get G'd up when full, bounced around on a rough day, left in the sun, etc. They're structure, and when loaded they flex like everything else.

I'm not saying balloon tests are a waste of time. They find the 1G leaks ;)
 
Scott's video is the best media Vans has ever produced! ... a haul of great technique.

He doesn't describe tho how he mixes his sealer.

My method is to use the scale as described by Kyle on a stiff card (hotel room keycard), then pile it on a stiff stainless steel plate (1/16x12x12" ish) and mix using a 3" putty knife. A plate of glass would likely work as well.

You can scrape the plate clean in a pass, roll it over and squeeze it down again. Very quickly you'll see the color even out and there will be NO unmixed material anywhere on the plate. Zero

Having the putty knife is also a great way to spread the mix onto rib flanges, getting that pyramid shaped line of goo that Scott describes.

All of my mixes looked exactly the same shade when dry.
Cleanup: Scrape it all off and then a quick wipe with solvent.

Very tidy :)



I put a wrap of saran wrap on my digital scales then used Dixie cups to mix mine in.
 
spot on

Yo Dan.
You are absolutely correct and the best we can do, is have VERY good construction techniques. I'm happy with Scott making the video so at least some people can get it right. The first set I did had no leaks.
The 6 I'm working on now is a mess in the fuel tanks. To much beer or ???
So this set with better processes and more attention to detail I feel confident of no leaks. I'll see when I giter up in the air.
But that said: Ain't nobody flying with me for the first few flights after I get them done today or tomorrow.
I agree with the test results you show.
Thanks for your input on the stuff we ain't supposed to be using. Ha Ha
G's are the bees knees
Art
 
Used a digital food scale to measure ratios. Place a smooth white porcelain tile on the scale and tared it, then measured out base material and added the additional 10% in catalyst. Mixed with a combo type putty knife to a nice homogenous color. I did have some batches cure darker than others but they all cured in time. Both tanks were just finished a few days ago some I'm eagerly waiting for them to final cure to find out how I did.

The tile could be wiped clean or allowed to cure and scraped clean with a fresh razor blade.

I was fortunate enough to just at the baffle attachment stage when Vans published the video, and I referenced it several times before continuing to close everything up. It would be great to see them produce more how to video for the more critical processes of the build (wing and tail attachment, canopy fitting maybe?.
 
I've owned 4 Mooney's, 3 of them were wet wings. I've had all three opened up at one point or another to repair seeps.

I never have seen the sealant turned into goo on a Mooney! Mooney's use an overcoat CS3600, a see through maroon material, which is brushed over the proseal sealant. My guess is this overcoat protects the proseal from attack.

The only way to apply, on an RV, would be to make an access panel on the rear bulkhead of every fuel bay in the tank: a lot of effort but it may be worth it.
 
Last edited:
I've owned 4 Mooney's, 3 of them were wet wings. I've had all three opened up at one point or another to repair seeps.

I never have seen the sealant turned into goo on a Mooney! Mooney's use an overcoat CS3600, a see through maroon material, which is brushed over the proseal sealant. My guess is this overcoat protects the proseal from attack.

The only way to apply, on an RV, would be to make an access panel on the rear bulkhead of every fuel bay in the tank: a lot of effort but it may be worth it.

WOW, this may be the answer! From their web site: "CS 3600 is a one-part high bond strength adhesive
which finds application in a wide variety of purposes,
especially where resistance to oils, gasoline, aromatic
fuels and corrosion is desired. Most materials such as
synthetic rubber, metals, glass and plastics can be
bonded very effectively with this adhesive. CS 3600
is also used as a top coat for polysulfide sealants in
integral fuel tanks for aircraft."
 
I've owned 4 Mooney's, 3 of them were wet wings. I've had all three opened up at one point or another to repair seeps.

I never have seen the sealant turned into goo on a Mooney! Mooney's use an overcoat CS3600, a see through maroon material, which is brushed over the proseal sealant. My guess is this overcoat protects the proseal from attack.

So why are Mooney's such known leakers if they "top cote" the proseal?

-Marc
 
So why are Mooney's such known leakers if they "top cote" the proseal?

-Marc

Done a lot of Mooney work in my time and would say that the sealant has a approximate useable service life of 25 years. This is also the approximate service life of a fuel bladder in the planes that use those. The bulk of the Mooney fleet is now older than 25 years (a lot are much older) so aging sealant is requiring attention. I would speculate that a 30 year old RV will have the same issues. I also think the top coat sealer works well and does add some protection to the pro seal.

Don Broussard
RV9 Rebuild in Progress
57 Pacer
 
I am doing my third tank leak repair in 8 years. Each time I find that the Pro-Seal (I actually use ChemSeal, but they are the same ???) has turned to a gooey past that I can scrape off with my fingernail. I always use frash material, mix it and apply it per the instructions, and it always seems to set-up properly. 4 years later and it has turned to gooe :( .
My tanks are almost always full, 100% 100LL usage. Can anyone provide some insight or guidenance so I dont have to doo this again?

I have the same issue on my -6. A previous owner opened the fuel tanks to perform the anti-rotation SB, and when re-sealed, whatever material that was used just didn't harden. Poorly mixed, or broken down due to fuel/dye/etc. I don't know. Eventually it leaked, so I pulled the covers and re-did the seals. The new proseal is a much darker shade than what I removed, if that helps. What I removed was uniformly coloured, no streaks.

When the previous owner did the covers, he also slopped extra along the outside of the perimeter seal of the end rib... Maybe he had extra and figured it couldn't hurt. Unfortunately, that too has gooed up now and that seam is now leaking. Which suggests that proseal that is gooing up will destroy existing proseal that was working just fine... The original sealant from 1996 elsewhere on the rib is still solid.

Now I need to clean off as much of the residue as I can, and hope that a clean application of modern sealant on the outside will seal. What I wonder, is how far into the joint the proseal has gummed up, and whether it's going to come back and eat away the new proseal from the inside.

The only alternative I can conceive is to remove and replace the end rib, something I think would be nearly impossible on a completed tank. I'd need to open the sender cover again so I could get in and put a bead on the inside after it's riveted back in place, or cut a new opening on the back flange for the same purpose.
 
A few comments to comments made by others.

I mix sealant with a digital scale that measures out 3 decimal places. The third place is in .005 oz increments.
I have found that same as Dan mentioned, as long as you are using some method to measure with a reasonable amount of accuracy, the sealant will perform as expected. For a long time.
I still have in my shop a home made balance scale (not sure whether it was made by Van or Art Chard) that has arm lengths sized to a 10:1 ratio.
All of the earlier prototypes, probably prior to the RV-10, had all of the sealant ratio'ed using it.
The RV-6A prototype being one of those and it is still flying at almost 35 years old with no leaks, so I do not prescribe to the idea that the sealant has a finite life and all RV's will begin to leak at some point.
I believe the likelihood of leaks over time is likely more related to the design of the structure. I.E., what level of movement adjoining parts have in relation to each other with the influence of flight induced loads.
Perhaps Mooney tanks get flexed more since they are an integral part of the wing structure?

I have worked on a number of RV's where reversion (softening) of the sealant had taken place.
I have never personally seen it occur inside a tank.
It seems to be related to there being a constant exposure to evaporating fuel, and ambient air. To me, it is a condition that is not developed on the inside of a tank. Perhaps because of the high percentage of fuel vapor compared to other gasses that make up ambient air. Other than that thought, I have no explanation for it.
 
Before this thread, I'd never heard of proseal going bad. In my -6, the proseal is a good 21 years old (24 if you count the time between building the tanks and first flight). Due to a sunk float I had to remove one sender access plate a year ago and I can say with assurance that the proseal was as good as ever. I have never had any leaks.

When I built the tanks I measured carefully with a scale but on one part (access plate I think) I realized later that I'd measured wrong and used too much catalyst. Called an A&P friend who said "heck when I was working on airliners we always increased the catalyst, otherwise you'd have to wait days to put that plane back into service!" Since then when using proseal for non-essential items, I've tended to go strong on the catalyst and never had any go gooey on me. Anecdotal, but still.

It's very strange to hear of these problems. Could it be some interaction with a specific paint, or primer, or cleaner?

EDIT: Or a change in the formulation or QC problem of some sort. Or could some people be trying to thin it? It seems like all these reports are from third parties i.e. people who didn't do it themselves. I really doubt it has to do with precise measurement of the catalyst, I use leftover proseal all the time to hold stuff together in non-essential places and usually just guestimate the ratio. Even those places it's always proven to have better and longer lasting adhesion than any other flexible adhesive, never seen it go bad.
 
Last edited:
I mix sealant with a digital scale that measures out 3 decimal places. The third place is in .005 oz increments.
I have found that same as Dan mentioned, as long as you are using some method to measure with a reasonable amount of accuracy, the sealant will perform as expected. For a long time.
Scott, is is reasonable to assume that the two smaller kits Vans sells (small jar of sealant with little tube of catalyst, or the single pump-tube) can be mixed directly without measuring/weighing? For the small quantities i've needed so far, I purchased two of the smallest kit, and mixed it all in the small jar with a popsicle stick. I was able to get it extremely uniform in colour (about a medium to medium dark grey) and the leftover bits in the jar did harden cleanly.

I have worked on a number of RV's where reversion (softening) of the sealant had taken place.
Have you ever seen this situation cleaned up, new sealant applied, new sealant appears to set properly, and then at a later date the new sealant starts to revert?

It seems to be related to there being a constant exposure to evaporating fuel, and ambient air.
This would be consistent with the reversion happening at the location of a leak... So perhaps not the cause of the leak initially, but a side-effect after the leak has occurred.
 
"Mix it Hot" - LoL

Read and Follow the directions that come with the ProSeal

1) There are different types of Pro-Seal. One type is used for applications involving fuel.
2) People don't mix it right. Inadequate mixing and incorrect component ratios.
3) Inadequate preparation.

Regardless, the instructions are very clear.

Problems I've seen are always user error - always. And they always fall into the aforementioned reasons.

I guess people don't know how to read or they think they're smarter than the PhD Chemical Engineer who invented the stuff, they're cheap and don't want to buy the right stuff, and they're lazy and don't want to prep the material they're trying to seal.
 
My day job and Proseal observations

I'm not a rep for Proseal, but I have been around it almost daily for over 40 years..In the big plane world it is the mainstay of wet wings, and pressurized fuselages. Currently, I currently manage the KC10 Heavy maintenance program, and these badboys carry 355,000 pounds of fuel when topped off. They are 40 years old, and while they do get leaks, the OEM sealant is for the most part like the day it was built. We use no scales, but typically utilize the Semkit premeasured tubes, or pre-measured cans. The failure of most sealant preparation is often someone trying to mix just a partial kit and getting the ratio wrong. For the RV builder, the Semkit tubes are a no-brainer, and ALWAYS make a sample card to put away, especially if you are closing a tank. We do this as part of the post maintenance inspection sign-off. We also utilize some sealants with a looong work life/cure time. It can take weeks for some to fully cure. Additionally, when I do my RV seal jobs, I assemble with "B" (thick) sealant, then topcoat all external seams with "A" (brushable) that has consistency of honey. Pay attention to shelf life when doing seal jobs..although I have used years old tubes for non-tank work and never had a cure issue.
 
I learned about ProSeal 40-years ago at Hughes Aircraft Company. Surfaces must be clean and scuffed for a good seal. Sealant must be mixed well at the correct ratio and we only used SemKits.

I will only use the SemKits spoken about above. I also use a mixer to make sure that it is mixed according to direction or a little bit longer mixing.
 
Last edited:
Read and Follow the directions that come with the ProSeal

1) There are different types of Pro-Seal. One type is used for applications involving fuel.
2) People don't mix it right. Inadequate mixing and incorrect component ratios.
3) Inadequate preparation.

Regardless, the instructions are very clear.

Problems I've seen are always user error - always. And they always fall into the aforementioned reasons.

I guess people don't know how to read or they think they're smarter than the PhD Chemical Engineer who invented the stuff, they're cheap and don't want to buy the right stuff, and they're lazy and don't want to prep the material they're trying to seal.

The problem with my pro-seal turing to gooe is not user error, improper mixing, or using the wrong material. It is a chemical interaction of the proseal and 100LL vapor. As discussed eariler in the thread, this problem occures external to the tank where a small weep occures allowing 100LL vapor to exist at the right concentration in the gap between the tank and the wing spar. Over time this vapor turnes the pro-seal to gooe no matter how much you pay for it, or how well you mix it. Proseal was developed for use with Jet-A. It works fine for 100LL unless the previously mentioned conditions exists.
 
Desoto Proseal..I accept no substitute.

In my observations, I haven't seen the goo breakdown with Desoto Proseal(unless improperly mixed)..I have never messed with Chemseal, Flamemaster or other variants, and I am not "bashing them", however, I have seen my share of non-Proseal goo(I can tell by color) that I have repaired, particularly on RV's. I will only use one brand..the same one used by virtually every OEM that has wet wings.
 
As I have mentioned in past discussions, I have seen reversion issues with all sealant brands that I have worked with over the years (including Proseal).

The occurrence of this is NOT exclusive to instances where the sealant was not mixed at the proper ratio (though I suppose that could have an impact, I have never seen an occurrence where I suspected that to be the cause).
 
Blue Dye

The problem with my pro-seal turing to gooe is not user error, improper mixing, or using the wrong material. It is a chemical interaction of the proseal and 100LL vapor. As discussed eariler in the thread, this problem occures external to the tank where a small weep occures allowing 100LL vapor to exist at the right concentration in the gap between the tank and the wing spar. Over time this vapor turnes the pro-seal to gooe no matter how much you pay for it, or how well you mix it. Proseal was developed for use with Jet-A. It works fine for 100LL unless the previously mentioned conditions exists.

It has been posted elsewhere here on VAF that the softening of pro-seal isn't due to fuel vapor, it's due to high concentrations of the blue dye when exposed to oxygen (air). The softening occurs in areas where leakage results in constant drip and evaporation on a pro-sealed surface. This concentrates the dye and outside of the tank there is air available to cause the oxidation. I have seen this issue on my own tanks when I had a fuel sender gasket leak. The proseal on the lower root rib fillet softened where the fuel was able to pool/evaporate, but places that did not have concentrations of the dye (along the same fillet) did not soften. After If fixed the leak, I cleaned the area up and laid fresh pro-seal over the softened stuff and I've had no issues since.

Skylor
 
Anyone run 100% mogas with or without ethanol and get goo proseal reversion?

Ethanol could have issues that skew this- but any way to prove a 100 LL blue fuel dye cause?
 
My notes (taken from a UK fuel standard) say the blue dye is 1,4 dialkylamino-anthraquinone at a max concentration of 2.7 mg/l.
 
Last edited:
C.I. Solvent Blue 98 searches, the trade name , has TDS and MSDS info that gives no help to this non-chemist, storage coordinator or firefighter.

Anyone know if it concentrates and acts like Poly-gone?
 
C.I. Solvent Blue 98 searches, the trade name , has TDS and MSDS info that gives no help to this non-chemist, storage coordinator or firefighter.

Anyone know if it concentrates and acts like Poly-gone?

I'd bet against a dye harming proseal. However, it is left at the scene of the crime when fuel evaporates near the source of the leak. Kind of like a broken window may signal a burglary, but the broken glass isn't a causal factor.

There's been a lot of speculation on the topic for the 25 years I've followed this forum and others. But I don't think there has ever been real data that helped us understand the problem.
 
leaks

When I bought my RV-6A 3 years ago was leaking/weeping fuel and I planned on replacing the tanks.
Reason for replacement instead of sealing was that there were more problems than could be fixed by sealing. The builder did a terrible job on these tanks. They were not structurally sound.
I have been using Costco premium the last 3 years and they still leaked (ha ha). Upon removing the tanks for replacement I found the Proseal on the outside of the tanks to be soft and gooey. The stuff on the inside was soft but not gooey.
Some people think that if you put it on thick enough it can't leak. Well 1/16" is enough if you cleaned and prepared the area good enough.
Clean clean clean
New tanks installed, flying, and yes with Costco premium, so if they leak I will find out if the ethanol is maybe the cause. I'll see.
Enjoy the trip Art
 
Last edited:
12, 14 years ago, I ran some reformulated (ethanol) auto gas in my plane. Some time after that, I sprang several tank leaks. No chemist, but I believe that nasty blended gasoline garbage was the cause.

Also, regarding mixing. In the past I've seen people mixing various two or three part products. Some people's idea of mixing is far short of actual mixing. Just sayin. You have to mix it very, very thoroughly.
 
Last year I replaced a tank because it was leaking all along the lower spar edge of the tank.

The theory was that there was an initial leak and someone tried to fix it by slathering Proseal all along the seam. But that there was fuel in the tanks so the Proseal turned to goo.

I made sure the Proseal in the new tank fully cured before I put gas in the tank. No leaks
 

Attachments

  • Tank_Leak_bad_Proseal.png
    Tank_Leak_bad_Proseal.png
    176.9 KB · Views: 164
Dye

I have the same issue...sweating along my bnc and rear flange....fixed the bnc...no sweating and proseal is hard.

Proseal at Baffle has gone sticky....covered in dye. No where else has it gone bad. Tank is coming off to fix it now.
 
We've been discussing sealant reversion here for some time. The old VAF (i.e. 10-15 years ago) had far fewer readers than today, so perhaps the community now includes a few good chemists. For sure, I am not.

The one thing I think we've established is that sealant can revert to a very limited extent inside tanks, but does not fully revert back to paste form until exposed to fuel on an external surface.

Prior searching turned up these two documents:

https://www.danhorton.net/VAF/Can Proseal Decure/Accelerated Peroxide Formation in Fuels.pdf

Note the changes based on atmospheric exposure.

https://www.danhorton.net/VAF/Can Proseal Decure/Fuel Peroxides.pdf

See the notes for PR1440B-2 on page 4.

Unfortunately, I have never found ay documents referencing 100LL, and I remind readers than polysulfide sealants have no avgas test standards.

BTW, 100LL can include antioxidant additives, list attached. I assume they may evaporate quickly when exposed to atmosphere, but I have no clue as to their volatility.
 

Attachments

  • Anti Oxidant 100LL Additives.JPG
    Anti Oxidant 100LL Additives.JPG
    68 KB · Views: 125
Last edited:
Here is a post of discussion with a 3M PhD on the subject:
https://vansairforce.net/community/showpost.php?p=1116704&postcount=31

It is still relevant.

If we started using 3M product and documented having this issue I am sure we would get some help from them. If we found solid evidence that there is incompatibility with the fuel, then we may get engagement to validate this theory and then get the ASTM or other governing tests modified to accommodate 100LL.

Meanwhile, would anyone have a proposed bench procedure to test a product?

Options:
1. Dip test - Proseal is coated on a test strip and dipped with a rotating crank to slowly dip and raise the sample into a pool of 100LL.

2. Drip test - a sample is smeared with pro seal, and 100LL dripped onto the sample at a rate that would allow it to nearly dry before the next drop. Ventilated periodically like a wing root in flight.

3. Dunk test - A test strip set into a glass of 100LL and allowed to evaporate until 50% gone, then repeat. No ventilation.

There are many variables, product age, mixing ratios, quality of mixes, manufacturers, and likely temperature.

As long as it took volunteers to gather information for failing 6-7 steps and accidentally get it solved by an evolutionary process, I guess this one will take about 20 yrs.

Our best bet might be to get the FAA to add this material compatibility into the standards for the 100LL replacement projects. i.e. get someone else to solve our problem.
 
Back
Top