Paint Blisters/Fuel Tank Rivets

[rv72004]

The planes may fly well with blisters, but it still sucks. It doesnt make it better or right. Of the RV's I have looked at, a lot more than 1% have blisters.
My QB proseal had gone sticky on one wing root. That same wing blistered like crazy.
I personally think [and would put money on it] that either the proseal was badly mixed or the batch was bad. A airline engineer who saw it said the proseal was bad. I wiped what I could off with MEK [it came of easily] and redid the outside fay seal. It has sprung a leak once again so I redid a small area at the spar side.
I have photos to prove the mess. You would be blown away at how bad it looked. Maybe one day Ill fiqure how to post pics. Are there instructions on this site how to ?

[mdredmond]

Maybe one day Ill fiqure how to post pics. Are there instructions on this site how to ?

EJ, yes, here: http://www.vansairforce.com/community/showthread.php?t=1696

[DanH]

My QB proseal had gone sticky on one wing root. That same wing blistered like crazy. I personally think [and would put money on it] that either the proseal was badly mixed or the batch was bad. A airline engineer who saw it said the proseal was bad. I wiped what I could off with MEK [it came of easily] and redid the outside fay seal. It has sprung a leak once again so I redid a small area at the spar side.

EJ, we now have a direct line into Flamemaster. Obviously your sealant did not cure; it should not wipe off with MEK. The question is why. If you would like them to look at this sealant problem, well, no problem. They may want a sample.

Interesting case; I don't recall another blister report with clear indications of uncured sealant.

Ok, you have obvious fuel leaks as well as blisters. Can you relate more detail? Leak locations, blister locations, blisters opened for examination...anything at all. Photos are good too.

EDIT: These just in from EJ, by email. Wow....

http://img261.imageshack.us/img261/8599/dscn1013.jpg (http://img261.imageshack.us/i/dscn1013.jpg/)

http://img340.imageshack.us/img340/8076/dscn1015c.jpg (http://img340.imageshack.us/i/dscn1015c.jpg/)

http://img525.imageshack.us/img525/5743/dscn1016j.jpg (http://img525.imageshack.us/i/dscn1016j.jpg/)

[RV505]

Speaking of research....

The mystery "oily substance" applied to QB components is ordinary WD-40. I've reviewed the subject with Ken at Flamemaster. The MSDS says it contains the following hydrocarbons:

64742-47-8 - Deodorized kerosene
64742-88-7 - Straight run white spirit (Stoddard Solvent)
64742-88-9 - Naphtha (petroleum), hydrotreated heavy
64742-65-0 - Lube Oil

...and a proprietary surfactant (wetting agent).

Kens says the hydrocarbons would not cause a deterioration or reversion of the sealant material. The surfactant and any possible activity is unknown.

Common sense says wicking action would place WD-40 under fastener heads and inside open voids. The application of surfactant and hydrocarbons means we must clean throughly prior to paint, and even then we probably don't remove it all from the hidden spaces.

However, the above would be true for the entire airframe, not just tanks. One could argue tank rivets are the least likely to hide WD-40; they're the ones with seal material under the heads.

Probably not an issue, and no point in debating the matter right now anyway. The real purpose of identifying the oily stuff is to catalog any possible chemical traces which may show up in later analysis.


WD-40 was not allowed at the airlines or even allowed in the Flamable locker.. I asked once and they said there was someting about the chemical properties of it and that was good enough for me. We used Aerokroil?

[DanH]

WD-40 was not allowed at the airlines or even allowed in the Flamable locker.. I asked once and they said there was someting about the chemical properties of it and that was good enough for me. We used Aerokroil?


Are you in a position to make a phone call or a visit, and nail down the specific reason why it was banned?

[szicree]

I've got no skin in this, but you guys have really drawn me in to this mystery. I'm excitedly waiting for the chemists to do their thing.
Also, I think it's worth noting that fuel does very little to modern paint, but will cut right through most primers. The akzo stuff holds up to it, but I found it would cut through cured dp50.

[RV505]

It has been so long and and half way across the country.

! jave found other posts on the internet though.

Feb 9th, 2009, 08:16 PM #2
steve299
Warming The Wheels
Ride: 2008 Sportster XL883c




Join Date: Dec 16th, 2008
Location: Greer, SC
Posts: 132
Re: The Magic of WD-40

--------------------------------------------------------------------------------

Be real careful with WD40. We used to use it on chrome aircraft struts to clean them until the chrome started delaminating. The engineers researched and told us not to use wd40 anymore that it was the cause of the drome breaking down.

[RV505]

I found this also
http://whitts.alioth.net/Page91PTS%20Engine%20and%20Airframe%20Paperwork.ht m

I scrolled down and found this!

WD-40 and Cleaning

The use of WD-40 is not recommended on aircraft. It is not recommended because only a light lubricant is left after the solvents evaporate. Use Kroil or ACF-50 available through aircraft parts houses. The use of steam cleaning and pressure washers on aircraft is not good practice since lubrication points are not sealed. Open bushes, pivot points must be dismantled to grease. Pressure washes flush out the oil and will wet the interior making it subject to corrosion.

[rvbuilder2002]

[QUOTE=DanH;366246]EJ, we now have a direct line into Flamemaster. Obviously your sealant did not cure; it should not wipe off with MEK. The question is why. If you would like them to look at this sealant problem, well, no problem. They may want a sample.

Interesting case; I don't recall another blister report with clear indications of uncured sealant.

Ok, you have obvious fuel leaks as well as blisters. Can you relate more detail? Leak locations, blister locations, blisters opened for examination...anything at all. Photos are good too.

http://img525.imageshack.us/img525/5743/dscn1016j.jpg (http://img525.imageshack.us/i/dscn1016j.jpg/)

[QUOTE]

One point to consider in this case that may or may not be relevant...Tank sealant will soften and then become mushy with extended exposure to wet fuel with an oxygen source (being on the exterior of the tank) such as from a constant seep from a leaking tank access cover.
You can confirm this with flame master.

Just mentioning it since if there was an active leak at the root end of the tank related to a breach in the cover seal or fuel level sender, it can produce the exact result shown in the photos with enough exposure.

[Mike S]

Thanks Scott, good info to know.

[Rick6a]

My QB proseal had gone sticky on one wing root. That same wing blistered like crazy....I personally think and would put money on it that either the proseal was badly mixed or the batch was bad....
As far as I've been made aware, yours is the first tank to exhibit what appears to be uncured proseal. Is that an isolated local problem or does the softness extend to other areas of the tank? For instance, can you easily dissolve the cured sealant on the opposite side of the tank with MEK? If so, it is likely your problems are more extensive than a series of mere blisters.

Speculation on my part, but the possibility does exist that in your particular (and seemingly unique) case, the sealant was not thoroughly mixed and/or the operator mixed an improper ratio of A to B. That possibility seems (to me) more likely than sealant being of defective manufacture. Admittedly, I am opining like everybody else is at this point but as more information comes out it becomes increasingly difficult to dismiss fuel or fuel vapors as THE supporting actor in this little play. Even old proseal if properly stored and correctly mixed several YEARS past its expiration date will cure adequately. I have done just that many, many times in non-fuel applications.

My concern is that if that proseal is soft in other areas throughout the inside of the tank, leaks and blisters can only get worse over time. When you say "that wing blistered like crazy".....define crazy. How many blisters? Top and bottom? One end of the tank to the other? If the blistering is indeed extensive and aided and abetted by defective proseal, your tanks may suffer from more than a mere "cosmetic" problem. I have to think that deteriorating proseal collecting inside the tanks *may* or *can* effect flight safety at some point.

I am very interested in how extensive your blistering problem is and if sealant softness exists elsewhere on your fuel tanks. Nothing tells the story better than photographs. If you cannot figure out how to post pictures, I can in addition to Dan H. post any pics if you forward them to me.

[Mike S]

My QB proseal had gone sticky on one wing root. That same wing blistered like crazy.
I personally think [and would put money on it] that either the proseal was badly mixed or the batch was bad. A airline engineer who saw it said the proseal was bad. I wiped what I could off with MEK [it came of easily] and redid the outside fay seal. It has sprung a leak once again so I redid a small area at the spar side.
I have photos to prove the mess. You would be blown away at how bad it looked. Maybe one day Ill fiqure how to post pics. Are there instructions on this site how to ?

Have you checked out the fuel filter lately???

I would suggest watching for any evidence of the softened material going into the fuel system down stream from the tank.

[N5XL]

One point to consider in this case that may or may not be relevant...Tank sealant will soften and then become mushy with extended exposer to wet fuel with an oxygen source (being on the exterior of the tank) such as from a constant seep from a leaking tank access cover.
You can confirm this with flame master.

Just mentioning it since if there was an active leak at the root end of teh tank related to a breach in the cover seal of fuel level sender, it can produce the exact result shown in the photos with enough exposure.


I wouldn't think that exposure to wet fuel would cause tank sealant to do what we are seeing in the picture but maybe so...

The reason I say hesitate to agree is that is one of the tests that I see mentioned in product technical data sheets is to evaluate the sealant to specification compliance, by soaking a "test piece" (test piece = wire mesh that has been bonded with sealant to an aluminum substrate) in a mixture of simulated fuel (JRF II, type I) and 3% aqueous sodium chloride solution.

The test piece stays in this challenge fluid for a period of 7 days at 140F. At the end of 7 days, the sample is taken out and the wire mesh is hooked up to a device to measure the force required to peel it away from the aluminum substrate at a 180 degree angle, measured in pounds of force per linear inch. PPG's Proseal 890 B specification indicates it takes 39 pounds to peel the wire mesh away from the aluminum substrate after a 7 day soak in the above challenge fluid.

Unfortunately, there is no peel strength given for CS3204B, although it states that CS3204 meets AMS-S 8802, which is supposed to be derived from the old MIL-S 8802F. I would imagine that CS3204 should meet the minimum MIL-S 8802F specification which is listed as 20 pounds minimum under the same exposure.

Perhaps Flamemaster can give us the peel strength data for both anodized and alclad aluminum pieces in the challenge fluid as well as comment on the characteristics of the picture. That picture looks really bad, and I'd sure as heck not fly anywhere with that going on.

[DanH]

Tank sealant will soften and then become mushy with extended exposure to wet fuel with an oxygen source (being on the exterior of the tank) such as from a constant seep from a leaking tank access cover.
You can confirm this with flame master.

The photos went to Ken this AM before posting here. We'll see what he says. I say the tank headspace has the same % O2 as outside.

POSTSCRIPT Just talked to Ken at Flamemaster. His opinion, based on EJ's photos, is that we're looking a a pure case of uncured sealant, not reversion or other chemical failure.

As for sealant failure due to "fuel with an oxygen source"....there is some basis in history dating from the bad old days of lead-cured sealant. Lead-cured sealants got a coat of Buna-N for protection. Those days are long gone, but apparently stories still circulate.

Not applicable to AMS-S-8802 sealants. The only common chemical with any effect is methylene chloride stripper.

Dave - check your hotmail account for the latest CS3204 qual test report.

[szicree]

http://img261.imageshack.us/img261/8599/dscn1013.jpg (http://img261.imageshack.us/i/dscn1013.jpg/)

http://img340.imageshack.us/img340/8076/dscn1015c.jpg (http://img340.imageshack.us/i/dscn1015c.jpg/)

http://img525.imageshack.us/img525/5743/dscn1016j.jpg (http://img525.imageshack.us/i/dscn1016j.jpg/)

I'd say that the blob between your fingers would stop an engine if it got into the fuel system. There is quite a lot of sealant inside the tanks and the potential for such a blob to get loose is great. No offense, but you wouldn't get me to fly in that plane until the tanks were checked out completely.

[rvbuilder2002]

As for sealant failure due to "fuel with an oxygen source"....there is some basis in history dating from the bad old days of lead-cured sealant. Lead-cured sealants got a coat of Buna-N for protection. Those days are long gone, but apparently stories still circulate.

Not applicable to AMS-S-8802 sealants. The only common chemical with any effect is methylene chloride stripper.

Dave - check your hotmail account for the latest CS3204 qual test report.

Ok...
I can't provide the name of a person at flame master that provided the info but that is where it came from. I was not base on urban legend. So it seems that either someone there is providing tech. info and they don't know what they are talking about, or the person you talked too doesn't know.

Dan, I only provided info that had been provided by someone at Flame master in the context of the sealant currently supplied and used by Van's. Just trying to provide valid info in the quest for answers. I assumed info provided by the manufacturer was better than a hunch such as "I don't think the difference in oxygen between inside to outside would be a factor" but it looks like it's not.
I still find it interesting that in the one photo it looks like the fuel level sender has been removed for some reason (a leak perhaps?). It has what appears to possibly be a fuel stain below the opening for the fuel sender? Not definitive because it is not a photo with a great level of detail to know for sure (on my monitor at least).

[rv72004]

Here is a full review on affected tank,

The tank close out was leaking and the fuel had run over the proseal at the bottom of the outer rib. The proseal had gone totally soft in this area as the pictures show. At the front top and back it was also softish in that MEK could dissolve the outer layer with a wipe, but not as soft as the pics show. There may be some truth to the oxygen softening the proseal in the presence of avgas. But this bad ? Is it possible ?

The cover was removed and the proseal inside the tank [outer rib] was not of the same consistency ,feel or wipe-ability as the one outside .It seemed OK from a non scientific point of view. This tells me the outside seal was possibly of another batch.

The fuel filter has been checked several times since the pic and there is not contamination. So the severe problem appears to be only the outside.
But here is the snag. I have never built tanks , and don't know how many times proseal must be mixed in the process. Are there any other bad batches? And this does not pertain only to my tanks. What about all the others out there? Surely the fellow/s who mixed the proseal on my tanks
has worked on other tanks too?

There were 2 fuel leaks on this tank. One on the top rear rivet line and one on bottom rear rivet line. After some time say 6 months [if I recall] a blue streak developed from these 2 rivets. I drilled them out , reprosealed and fitted 2 cherrymax rivets. No more leaks.

As far as blisters were concerned, I never took photos or counted how many, but here is what I recall;
There were none on the bottom of the tank. The top had many a blister, say 20 or more. There were not huge as I would prick them the moment they would start. In many cases the paint would come off leaving some rivet exposed. The I touched up with a fine artist brush making sure not to close the rivet seal to the skin. This stopped further blistering.

Things have quietened down on that tank . Blistering has stopped .
The soft proseal was replaced.
But no one can tell me for sure why all these problems with this tank?

The other side had a few blisters, say 5. No leaks and no soft outside proseal. Different builders in the QB facility? Different batches of proseal?

[DanH]

Ok...
I can't provide the name of a person at flame master that provided the info but that is where it came from.

Don't sweat it Scott. Can you share why you addressed the issue with Flamemaster?

[Rick6a]

Please forgive what may be a stupid question, but if fuel vapors under pressure are what are creating the blisters, couldn't I just pop the first one to avoid more?

......we have popped many, and it does not stop them from coming..Ive popped em early, squeegied em down in the hot sun when everything is flexible. Bubble gone..Month later, bubble back.
.....As far as blisters were concerned.....I would prick them the moment they would start...I touched up with a fine artist brush making sure not to close the rivet seal to the skin. This stopped further blistering.
Review the three recent quotes pulled from posts to this thread.

Interestingly, EJ commented he pricked the paint blisters, did a touch-up around the holes and the blisters stopped growing....not Kahuna's experience with the blisters but there is a fundamental difference in how each builder decided to approach the problem. EJ intentionally did not allow the paint to come back together, thus providing a relief hole for escaping vapor to vent. Kahuna on the other hand essentially laid the blisters back down which could allow fuel vapor to soften and essentially re weld the paint blister back together...in effect sealing the blister making it air tight again. Conceivably, this could over time allow fuel vapor constantly originating from within the fuel tank to regenerate the blisters and we know in Kahuna's case those blisters did in fact redevelop a month later.

Here is another interesting comment from EJ:

There were 2 fuel leaks on this tank. One on the top rear rivet line and one on bottom rear rivet line. After some time say 6 months [if I recall] a blue streak developed from these 2 rivets. I drilled them out , reprosealed and fitted 2 cherrymax rivets. No more leaks.

If I read this right, EJ may have had leaking rivets along the rear baffle and remember, those rivets are OUTSIDE the pressure boundary. If a rivet can leave a visible blue streak via a defective fay seal on the rear baffle, what makes anyone think fuel vapor cannot find its way though tiny voids in that same defective fay seal...voids so small it would take a magnifying glass to visually detect? If the eerily similar experience described in the Pazmany newsletter is any indication, a routine pressure test could not possibly detect such gradual losses anyway:

http://www.pazmany.com/newsletters/PL-1_and_2/64.pdf

In one way or another, every scenario I come up with consistently links the obvious; the possibility (probability) of fuel or vapor working its magic in conjunction with the misapplication of proseal.

I am hoping someone out there has submitted samples for scientific testing by now. Anyone?

[rvbuilder2002]

Don't sweat it Scott. Can you share why you addressed the issue with Flamemaster?

A builder had a long term leak located at the access cover. Sealant did just as shown in the photos.
Flame master was asked what they thought...they said it was normal when sealant has extended exposure to fuel on the outside of a tank with oxygen and evaporation.

From EJ's description it sounds like it is the same scenario.

[DanH]

Scott, did you personally call Flamemaster, or did you get this story via another party who said they called?

[rvbuilder2002]

Scott, did you personally call Flamemaster, or did you get this story via another party who said they called?

Another party who personally told me what he was told by the person he spoke with at flamemaster. I'm not sure why that would matter. Particularly since it seems to jive with the scenario that AJ has detailed.

[DanH]

Thank you Scott. I understand.

Without disrespect to anyone, we should expect certain realities. It is the way the world works. Remember "trust but verify"?

This particular technical point should be easy to check. I've already contacted Ken Chenard at Flamemaster, but you won't bother me at all if you confirm for yourself, with Ken or another company contact. I'm only interested in truth.

You could call one of the other sealant manufacturers, as all AMS-S-8802 sealants are likely very similar (and they probably check each other's formulations, another reality).

You could set up a simple experiment. I'd suggest three pickle jars, one with properly mixed sealant, one with 1/2 the specified quantity of Part B, and one with 1/4. Give the samples a week to observe the degree of cure. Then place each in an individual jar, add 1/8" of avgas (leaving lots of headspace oxygen), cap 'em and give them another week. Anyone with sealant in hand care to volunteer?

We can also consider the basic polymer chemistry. I'm not a chemist; perhaps some of our trained members can comment. I'm told the liquid pre-polymer molecules each contain 2 or 3 SH-groups (sulfur-hydrogen). The SH groups in each molecule react in the presence of a peroxide (contains oxygen), split out water and form an S-S bond (sulfur-sulfur), producing the chain extension or crosslinking. The black curing agent (catalyst-accelerator) is manganese dioxide (MnO/2). Note the role of oxygen.

Anyone care to comment on the likelyhood of this bond reverting back to goo? I'm also curious about the generation of water as a byproduct of cure....mystery liquids being of interest here.

[kaye7877]

Dan,

Consider that softening of the proseal or other equivalent material could be caused by the continuous evaporation of fuel from the material surface possibly leaving behind a higher concentration of a substance found in the fuel. Sealing the jar wouldn't show the same result as you would have by continual wetting of the material as it is evaporated.

Brad

[rv72004]

Ok, I counted the blisters on the "bad" tanks. About 60 in total
The other wing only had about 8 blisters.
Just a question about oxygen and prc, If it does soften the sealant howcome the sealant at the top of the rib which was not subject to a leak , was also soft and wipeable? Must admit the area under the cover was a bit more gooey.
Not saying that oxygen would not cause this, but why is it also prone in areas that didnt have a leak ? Could the problem be exagerated by bad prc/mix?
Also why is this the tank with leaking rivets and 60plus blisters?
The other tank does not display the same characteristics.

Somehow i wonder if various factors didnt combine to cause all the problems with this tank.

[rvtach]

I will be at COPPERSTATE Fly-In (before, during, and after). If you're going to be there and have wet blisters but no way to collect a sample drop me a PM. I'll bring stuff to give you to collect a sample.

I've got no blisters myself (also no fuel in the tanks yet or paint) but would like to see an answer to this.

[Ted Johns]

Anyone care to comment on the likelyhood of this bond reverting back to goo?

Dan, high enough peroxide concentration in the fuel will deteriorate polysulfide type sealant. I suspect that ozone would as well. They are both powerful oxidizers. Of course, regular old air is not in the same league, not even close.

See: http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA087267&Location=U2&doc=GetTRDoc.pdf

In the strongest (10 meq.) peroxide concentration, there was serious degradation. This concentration also degraded neoprene. The study also implicated fuel acid, a result of the breakdown of the peroxide. Note that the sudy was done with JP5, not avgas.

From a practical standpoint, I have to think that if gasoline + air equals destruction of pro-seal, we'd see a lot more goo at the top of old gas tanks. As the study said, concerning MIL-S-8802 sealants: "These tank sealants are intended to provide very long service and are difficult to repair."

[DanH]

Brad, I noticed the evaporation-concentration angle in Scott's proposal. Hopefully we'll learn more about that theory in the near future. The jar experiment was intended to explore the effect of fuel on sealant mixed in improper proportion. My fault, poor presentation.

Ted, good find. Note the date, 1980. Made me curious, so I did some reading. Looks like the report had some effect; this is a note regarding current jet fuel:

Antioxidants - Hydroprocessing of aviation fuels removes naturally occurring antioxidants that provide protection from peroxidation. Peroxides
are known to attack elastomers causing embrittlement while also contributing to gum and particulate formation. The use of antioxidants effectively prevents peroxidation from occurring and under JFSCL and Def Stan 91-91, 17 to 24 mg/L of an approved antioxidant must be added to the proportion of the fuel blend that has been hydroprocessed.

http://www.exxonmobil.com/AviationGlobal/Files/WorldJetFuelSpecifications2005.pdf

Avgas also contains a lot of antioxidant, mostly for storage stability.

Everybody realize Ted's point is "strong oxidants can damage sealant", which is surely correct. I offer the above information before somebody goes off on a bad fuel goose chase. So far I'm only aware of one possible fuel culprit, methanol. Two of our blistered brothers have used auto fuel (David and Bill), so fuel alcohol was of interest. Ken says ethanol hasn't been a big problem, but methanol will fail sealant. I think it also eats aluminum, so I'm guessing David or Bill may have tanked some ethanol blend but probably not methanol. Anybody know for sure if methanol blends are sold in Missouri or Georgia? And EJ, what have you been burning?

[N5XL]

Dan, high enough peroxide concentration in the fuel will deteriorate polysulfide type sealant.

See: http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA087267&Location=U2&doc=GetTRDoc.pdf

In the strongest (10 meq.) peroxide concentration, there was serious degradation.

From a practical standpoint, I have to think that if gasoline + air equals destruction of pro-seal, we'd see a lot more goo at the top of old gas tanks. As the study said, concerning MIL-S-8802 sealants: "These tank sealants are intended to provide very long service and are difficult to repair."

Ted,
Great find. What the paper is confirming is a mechanism for the breaking of previously crosslinked polymer linkages, and to Dan's point, proof that such a mechanism exists. Ted I know the paper specifically discusses JP5, but I'll make the case that the same potential exists in avgas. I've copied two relevant passages out of the paper to form the hypothesis.

This degradation is attributed to a reversion effect caused by polymer chain and/or crosslink scission. Polysulfide polymers are susceptible to attack by acid catalyzed hydrolysis at the backbone formal linkage. Another polysulfide degradation route is by direct oxidation at the formal linkage but this usually occurs at temperatures around 300 T. The polysulfides are generally considered to have good oxidation resistance at temperatures below 200'F. They are also cured by inorganic and organic oxidizing agents. From the available data it is uncertain whether catalyzed oxidation or acid catalyzed hydrolysis at the formal linkage is the primary degradation process. The latter mechanism is favored.


MIL-S-8802 elastomeric polysulfide sealants, particularly manganese dioxide cured type, are prone to severe degradation in 10 meq. peroxide fuel (1000-1350 hrs.) where the fuel acid number exceeds normal limits. It is uncertain whether a catalyzed direct oxidation or an acidic hydrolysis mechanism is responsible. The latter mode is believed more likely to occur. Elevated fuel acid numbers are believed due to peroxide decomposition by-products possibly present in the original prepared JP-5/peroxide concentrate and additionally accumulated in the stagnant immersion media during the test phase. Data on the effects of lower peroxide concentrations and acid numbers is needed to further assess damage potential with MIL-S-8802 sealants.

With evidence of the effects of high PV given above, the charge is made that avgas could exhibit the same tendency should it have elevated PV and acid value (AV) numbers or given the right mechanism for PV / AV formation. If such conditions exist, polysulfide sealants, particularly those that are cured with manganese dioxide (which is probably the most commonly used sealant in RV construction), are prone to the effects as seen in the pictures given several posts ago. So the question is, does avgas have a tendency to have high PV and subsequently high AV?

One potential route for avgas to have high PV would be the inadvertent blending of small quantities of jet fuel with avgas. Perhaps in the transferring process at the refinery, or perhaps in the transportation of both fuels, common pipes and or hoses would contain enough residual jet fuel to contaminate avgas, giving it the materials it needs to start forming PV and AV. My thinking is that PV contamination by this route isn't likely, but its a possibility.

The second route would be the formation of PV's in the avgas storage tank itself. This possibility is discussed below.

Storage Stability
Avgas instability involves multi-step reactions, some of which are oxidation reactions. Hydroperoxides and peroxides are the initial reaction products. These products remain dissolved in the fuel but may attack and shorten the life of some fuel system elastomers. Additional reactions result in the formation of soluble gums and insoluble particulates. These products may clog fuel filters and deposit on the walls of aircraft fuel systems, restricting flow in small-diameter passageways. Instability of avgas during storage is generally not a problem because of the way the fuel is manufactured (see page 66), and most fuel is used within a few months of its manufacture. Storage stability can be an issue at locations where fuel is stored for occasional or emergency use. Avgas that has been properly manufactured, stored, and handled should remain stable for at least one
year. Avgas subjected to longer storage or to improper storage or handling should be tested to be sure it meets all applicable specification requirements before use. (pg 50, Cheveron Aviation Fuels Technical Review FTR-3, 2006)

I take away from Chevrons explanation that, it shouldn't be a problem because fuel is used relatively quickly, but , the potential is there should the fuel be abused or mishandled. What Cheveron does verify is that avgas is susceptible to oxidation reactions that can cause PV formation. This leads into the third mechanism of PV and AV formation and its already been touched on several posts back...that is that a long term, slow fuel leak will cause the damage we've seen in the picture.

If Chevrons explanation is true, then fuel leaking from a tank, over a period of time would have high PV...not in the tank, but on the outer surface where the fuel is actively exposed to fresh oxygen. Obviously, oxidation reactions need...oxygen. That exposure occurs on the surface of the fuel wetted tanks. The fuel dribbling down on the outside of a leaking tank is thin, thin enough and spread out to have a very high surface area perfect for oxygen exposure. Over time, this oxidation reaction forms peroxides and given enough time, acids, which we now know have the ability to degrade polysulfide based sealants. Peroxides tend to be heavier than the base component from which the are formed, which means that as avgas is constantly dripping and wetting the surface, and as that avgas is evaporated away, it leaves behind peroxides that will stay in intimate contact with, you guessed it, polysulfide tank sealants. The reason you don't see (or shouldn't see) softening of the sealant inside the tank is that the residual oxygen levels inside a fuel tank are much lower than in free air. This coupled with fact that as you're out flying about, avgas is actively washing away any peroxides that might have formed and the engine consumes them, hence they never have a chance to build up inside the tank like they would on the outer surface of a fuel leaking tank.

Of course this is one possible explainable of the ONE case that we just happen to have photographic proof of, and its only a theory on how this kind of damage could be occurring.

Dan, you've mentioned mogas, and I was going to do a bit more research on that part of the PV equation before I commented, but I do believe that mogas is heavily "doped" with various goodies that could easily oxidize and form PV's and AV's. Its clear that the mechanism (PV / AV degradation) is a concern. Mogas, and the potential for it to have ethanol in it, (ethanol tends to be weakly acidic) significantly increases the chance to have PV and AV in enough concentration to cause problems. In an actively flying RV, this might not be a problem, but what about those planes that don't fly often that have mogas in the tanks? Interesting thought, and your point about bad fuel is correct. Its not bad fuel causing the problem, its the potential for any particular fuel to form peroxides and acids that is concerning.

[rv72004]

Dan,
I have only used Avgas 100LL . But remember this only happened on the RHS tank. So if it was bad Avgas both should be badly affected?

[Rick6a]

.....I have never built tanks , and don't know how many times proseal must be mixed in the process......The other side had a few blisters, say 5. No leaks and no soft outside proseal. Different builders in the QB facility? Different batches of proseal?

.....Also why is this the tank with leaking rivets and 60plus blisters? The other tank does not display the same characteristics......

.... have only used Avgas 100LL . But remember this only happened on the RHS tank. So if it was bad Avgas both should be badly affected?EJ

You pretty much answered your own questions. I don't know how the foreign QB production facility is set up but I do know something about how domestic aircraft production works. You ask why do things happen in one tank but not the other? Well in the real world, it is extremely unlikely any given worker, without help is going to start and finish a pair of fuel tanks by himself and do all that work in one shift! Unfinished work might be passed off to someone on the next shift or if there is no next shift, the worker picks up where he left off the next time he passes through the factory gates. It is also highly likely that fuel tanks are assembled using MULTIPLE batches of proseal. As any standard kit builder knows, among the first fuel tank jobs to be completed are the fuel flange, drain fitting and stiffeners. Only after that work is completed....and that work is substantial....does work proceed to installing all the ribs. Finally, the rear baffle is installed. You'd have to install a rear baffle yourself to understand how tricky that is to assure a good fay seal remains intact when you slide the baffle into place. The reason those rear baffle countersinks are machined is because getting that tight fitting baffle past series of dimples would be "problematic" at best. So basically, there are three major phases of assembly in fuel tank construction and each phase will likely require a fresh batch of proseal. A worker can only work on one fuel tank at a time. Someone might be assigned to work on the left tank and someone else assigned the right tank. Those two workers may team up at times to shoot rivets. Who knows how many workers...and believe me...every worker comes to the table gifted with a certain level of skill and craftsmanship...have a hand in assembling any given set of fuel tanks. Ask anybody who has ever hired people. We used to have a saying where I worked.....for every ten workers the company employed, 7 weren't worth a darn, 2 made money for the company and 1 was outstanding.

Now I am not saying that is the way things are where your tanks were assembled but labor is labor, production practices are production practices and people are people no matter if you build the fuel tanks yourself or you hire someone else to do the work for you.

[RV505]

Sometimes when to many peope are involved in a project you get into Analysis Paralysis. Analysis Paralysis is where you can?t make any forward progress because you bog yourself down in details, tweaking, brainstorming, research and anything but just getting on with it. You need to pull them tanks apart or live with it.

[N5XL]

Dan,
I have only used Avgas 100LL . But remember this only happened on the RHS tank. So if it was bad Avgas both should be badly affected?

The mechanism of degradation that Ted pointed us to earlier, explains one additional failure mode, given certain conditions. Why I personally found it interesting was the premise made earlier, that leaking fuel on the outside of tanks, (in constant exposure to air and with sealant), could cause the sealant to soften and degrade over time. Based on what I know of oxidation chemistry from my regular job and given the evidence presented by Ted from the paper, I find this very specific failure mode entirely believable. This specific scenario could explain the soft sealant on the outside of your tank, however much more evidence would be required to prove that this is the route of failure that would cause blistering on your tank. I think it difficult to expect that bad fuel would be the cause of all of the blister scenarios, and I think Dan even cautioned against this. Ricks comments on application and preparation mode failures, would be a much more likely scenario, given how the tanks are actually put together.

[N5XL]

Sometimes when to many peope are involved in a project you get into Analysis Paralysis. Analysis Paralysis is where you can?t make any forward progress because you bog yourself down in details, tweaking, brainstorming, research and anything but just getting on with it. You need to pull them tanks apart or live with it.

Its not analysis paralysis...its a lack of analysis that is causing paralysis. Up to this point, its all been opinions, speculation and morning coffee talk...not that there is anything wrong with that.

IMHO, it would be foolish to accept that pulling the tanks apart (to redo them) or living with the problem is going to solve anything, other than getting something done. Redoing the tanks, with the potential to make the same unknown mistake (because we've not proven the nature of the problem adequately) would be a bad choice. Just my opinion, and worth exactly what you paid for it. :)

[RV8R999]

I'm glad I read this thread..

one conclusion:

1. I'm going with polished aluminum in and around the tank area of my SB -8.

[DanH]

"Analysis Paralysis".....I like it.

Let's keep this investigation focused. I had a note the other day suggesting EJ's gooey end rib sealant would become a distracting sideshow....and it certainly has the potential. It is interesting and we'll surely learn something, but it is not a blister. It did appear in concert with blisters, so look for the link.

A word on theories. Do your best to think it through. Consider how you might prove or disprove your theory by experiment, or by reference to reliable sources. And please, do the work. I do not wish to damp enthusiasm or contribution, but lobbing in a theory blaming kryptonite and then hiding under the bed isn't going to move this along.

Don't forget the KIS principle. There's a huge difference between possible and likely.

Last, one good measurement is better than a mass of speculation. We still don't have a liquid or gas sample from a blister. I know a few guys are working on it, and more could be. It's important, so let's not let relax the effort.

.....we now return to your regular programming ;)

[WSBuilder]

Two questions come to mind: 1) how many with leaking SB tanks did not solvent clean their rivets before installing them, and 2) how many use WD-40 or similar to lubricate their clecos?

Rivet forging machines are high-speed, lubricated mechanisms that impart trace amounts of slime on each rivet. Without a good solvent wash & dry that lubricant makes an excellent parting agent against the sealant, not to mention the possibility of a chemical (blistering) reaction with fuel, thus a vapor/fuel leak.

If the clecos are getting lubricated with WD-40 or similar they are imparting trace amounts of lubricant into the holes and surrounding fay surface as we build. If true, it makes even more sense that the close out bulkhead gets lots of blisters first because the clecos got cleaned and lubricated after fabricating the forward part of the tank?

These are two subtle process controls that could be easily overlooked, especially at the QB facility (BTW, I'm beginning to envision that place as a dirt floor, palm leaf-roofed enterprise!).

Perhaps WD-40 should be included in a contaminant test?

[Rick6a]

I refer interested parties to review this thread:

http://www.vansairforce.com/community/showthread.php?p=369934#post369934

http://i37.tinypic.com/imnpzb.jpg

[n2prise]

I have been following this thread for a LONG time. I have slow-build tanks. You can see in this photo where I have scuffed the inside of the RIGHT fuel tank before dimpling the skins. I soaked all my rivets in a baby food jar of methyl ethyl ketone (MEK). I would pour the rivets into a strainer after the soaking, then let them dry on clean paper towels. I used a pair of hemostats to dip each rivet into the sealant when the riveting process began. The inside of the fuel tank skins were cleaned with MEK just before applying the sealant and riveting them together.

http://www.n2prise.org/Photos/DSCF0102.jpg

These photos were taken April 15 & 16, 2003. This is the LEFT fuel tank that has some fuel seepage on only ONE rivet, and three others with blue tint around them under the paint. The airplane has epoxy primer on it before a pearl white coat of paint, followed with clear coat. My right wing tank built second has NO leaks at all. I am sure my problem is just not enough ProSeal being applied in the LEFT tank during assembly.
http://www.n2prise.org/Photos/DSCF0123.jpg

I did not put enough sealant BETWEEN the rib and the tank skin on the first tank I built. It is visible in this photo at the rib FLUTE where there is not enough sealant to fill the gap between the rib and skin at the flute location. The rivet with the worst seepage is adjacent to that flute in the rib.
http://www.n2prise.org/Photos/DSCF0124.jpg

You can see the outboard rib has a liberal covering of sealant on all the rivets and between the skin and the rib. If I had done that on the other ribs, there would not be a blue stain to be found on this fuel tank. From the first flight on June 9, 2005, it took about a year and a half before I noticed the fuel stains. The RIGHT fuel tank is perfect, but then I built it second and my skills had improved by then. I did not take a single photo showing the inside of the RIGHT fuel tank during sealing and riveting.
http://www.n2prise.org/Photos/DSCF0125.JPG

All my leaks are on rib rivets, not stiffener rivets. There are no leaks on the top side of the tanks or the rear baffle plates, just on the bottom.

I did not lube my clecoes, but they did need an MEK soaking after the tanks were completed. IF my one rivet ever starts to drip fuel, I will pump the tank dry and try the green Loctite on it with the tank under a slight vacuum pressure. For now, I just check it from time to time when I am cleaning bugs off the wings.

[N941WR]

This is a most interesting thread. I have seen N2PRISE'S plane and it is well built and after seeing his post, I know he did it "right".

My plane is not yet painted and I wonder if I will have this problem after I get it painted in the spring.

Prior to the taxi incident, there were no leaks, at least none that I could tell.

Any suggestions on how I should seal the rivets prior to painting?

[Mike S]

Any suggestions on how I should seal the rivets prior to painting?

Very well, and from the inside of the tank.

Bet you didnt want to hear that:rolleyes:

[Rick6a]

http://i34.tinypic.com/2jdrjps.jpg

[Rick6a]

This photo is a view inside a quickbuild fuel tank. This tank has yet
to be installed or filled with fuel. The owner plans on polishing his
aircraft. I think that is a good thing.

http://i33.tinypic.com/2hn2y6b.jpg

[WSBuilder]

Rick,
Did you apply sealant to just one fay surface or both? I was conservative with my Flamemaster and did both. I was especially generous on the back plate.

[N941WR]

Very well, and from the inside of the tank.

Bet you didnt want to hear that:rolleyes:
Mike,

You are a smart A$$ but I had a good laugh after reading that comment.

Mine are pretty much gooped up good! I went through a LOT of proseal when making the tanks, more than I should have but that looks like it might pay off.

I went back and looked at the pictures from when I built my tanks. I think I might have reduced my capacity by one gallon per side based on how much proseal is in there. Sill, I'll be curious if I have a problem.

[DanH]

I've been digging deep into the engineering and process control of tank sealing. Ya'll take a close look at Rick's "Attention To Detail" illustration. The single most important physical detail is the filet, for a specific engineering reason. More later, but for now if you're building tanks be sure to tool smooth filets within the specified "application time" for the particular sealant in use.

[Rick6a]

Rick,
Did you apply sealant to just one fay surface or both?.... Bill,

Applying sealant to both faying surfaces cannot hurt a thing. In my case,
I applied a very thick coat of sealer to just one faying surface (such as
a rib) so there was no doubt in my mind when the parts were joined and
clecoed together that voids in the fay seal could not exist between parts.
I think some builders make a mistake when they slide the ribs into place.
Careless rib insertion can displace an otherwise well applied coat of sealer
thus possibly risk causing needless voids in the fay. For that reason, I
installed the ribs in such a way there was no chance of that happening.

http://i36.tinypic.com/15oaxyp.jpg

[Denase]

Okay new guy here, I expect to be skewered for this. If I build my fuel tanks using Rick's method of sealing, but instead of using the wet rivet approach. Drive the rivets dry and use a low temperature 350 degree solder around the head for sealing the rivet, will it work? Well let the sparks fly, what do you think?

[Mike S]

Okay new guy here

Robert, welcome to the force:D

Solder and aluminum??? Far as I know, they dont play well together. For clarification, I am talking good old tin/lead solder.

As far as some special stuff formulated for use on aluminum, I would check into what is used for flux, and any side effects it may have.

350 degrees might be a bit higher than pro seal can tolerate, I have no idea. Probably someone out there will know.

I think if I was going to solder a tank, I would consider just forgetting all the pro seal and go with solder for all sealing. But then I would worry about vibration effects on the solder.

Gonna be interesting to see where this goes:confused:

[DanH]

If I build my fuel tanks using Rick's method of sealing, but instead of using the wet rivet approach. Drive the rivets dry and use a low temperature 350 degree solder around the head for sealing the rivet, will it work?

FWIW, the maximum test temperature applied to qualify for the MIL spec is 250F.

Robert, unless you're ready to build tanks right now I'd suggest kicking back and watching for future guidance.

[Denase]

Kicking back is what I will do. The product I would use is Muggy formulation #1. (http://www.muggyweld.com/1aluminum.html) This uses a flux that is non corrosive and allows the product to flow. There are other products that may work as well. But there are many questions before this can really be considered. What are the consequences of heating 2024 locally to 350 degrees? Will this procedure truly seal the rivet? How will it work over time with temperature changes, vibration, etc? Is the filler and flux truly non corrosive over time? A test tank built to test this idea would be easy enough. And if it did work it may be a very good way to do tanks in the future.

[Bob Kuykendall]

...What are the consequences of heating 2024 locally to 350 degrees? ...

According to AC43.13-1B, Paragraph 4-54(b):

Reheating at Temperatures Above Boiling Water. Reheating of 2017 and 2024 alloys above 212 ?F tend to impair the original heat treatment. Therefore, reheating above 212 ?F, including the baking of primers, is not acceptable without subsequent complete and correct heat treatment.

Of course, we as experimental aircraft builders are not obligated to adhere completely to AC43.13. But it is usually a good idea to do so.

Thanks, Bob K.
http://www.hpaircraft.com/hp-24

[scsmith]

I would be stunned to learn that there is a solder that works on 2024. There are some products out there that will solder (braze) 3000 and 5000 series aluminum, I think. I have tried a couple on 6061 with no luck.

[longranger]

A test tank built to test this idea would be easy enough. And if it did work it may be a very good way to do tanks in the future.

You could at least test the vibration resistance by building and running a test tank on your mower.;)

[RVnoob]

if you want to have a welded tank, make a welded tank...
I don't think the skin (both the structural design and the material) are meant to be welded or soldered.

[jurik]

I have read through this thread (and others about the same problem) but I see that the last response was in 2009 and I do not see a solution. My RV has been flying for one year now and I also noticed the paint blisters over the rivets on my QB wings and also only in the fuel tank area. No fuel in these blisters or any coloring.

Has anyone found a cause / solution to this yet?

[Mike S]

There is work being done on this by one of the members here, he is trying to keep a low profile till he has solid proof of what is going on, then will make a report.

At this time, further speculation is not going to be much more than a waste of time.

[decathlon737]

Has anyone found a cause / solution to this yet?

My QB wings developed major blisters and started about a month after fuel went in them. It progressively got worse for about a year until I had to do something about it. It was a major PITA! I had to take a perfectly good aircraft out of service to repair it. I was involved with the experiment mentioned along with several others. Some have not repaired yet, others have. I pulled both tanks off, made a mess with the paint, I had to strip and repaint them (4 different colors). The fix is to open the rear baffle between the tank ribs to gain access. I will not elaborate on what I found but resealing them properly has fixed my problem. Repainted them and over a year later, no sign of a blister. You will hear they were painted improperly, not cleaned properly and many other things. If you open them up and seal them correctly they will not leak anymore. You won't find fuel in the blisters and the tanks will easily pass the balloon and soap test, yet they will blister if not sealed properly. I have sold my 8 and about to build another, guess who will build my next tanks. I will never order another QB tank. That's my story. YMMV

[Janekom]

Aslo a QB.
The tanks have passed the ballon test.
The wings have been painted a few months before the first fuel went in.
Saw ONE BLISTER a few days :mad: ago on the RH tank top and who knows if any other will develop.

[detlef lili]

Pat, of course you can get general coating problems, including trapped solvents, if the preparation and paint job on a plane are not done properly.

HOWEVER, what we are talking about here is specifically blistering on rivets AT THE FUEL TANK ONLY and where they do not appear anywhere else on the entire plane. This has proven to be a very common phenomenon.

The ONLY logical conclusion one can draw from this particular phenomenon is that the blistering is caused by fuel or fuel vapour weeping through rivet holes.

To date, no-one has reported a case of this occurring on fuel tanks when their shop heads (inside the tank) have been fully encapsulated.

I again request anyone who has experienced blistering on tanks with fully encapsulated shop-head rivets to step forward. If no-one comes forward I think we can conclude what the problem is.


I have a slow build and it happens only on the upper side. My shop heads are all encapsulated. My suggestion is, that I used to much sealer. As I removed the paint, I could see, that some of this rivets out of center. Maybe, as I squeezed the rivet, the redundant sealer could not went out fast enough and now is enough space for vapor to escape and lift the paint.
I also run a paint shop for more then 20 years and think, that I have a lot of experience. But why happens this??

[Rick6a]

.....My suggestion is, that I used to much sealer.......I really doubt the amount of sealer you used has anything to do with your problem. It is the proper application of that sealer that is far more important. Besides, rivet shop head encapsulation is not the only thing that will prevent blistering, a good fay and filet seal is also very important. Some, if not most builders did not encapsulate the rivet shop heads, yet most builders have not experienced paint blistering. How is this possible? See posts #187 and #207 to understand how a fuel tank can still develop paint blisters regardless of how good the rivet shop head encapsulation is.

[Captain Avgas]

I will quote directly from the fuel tank construction section of Van's RV7 Builder's Manual:

"Dab a bit of sealant over every rivet head".

And yet Vans has repeatedly failed to ensure that their QB tank subcontractor follows this specific guideline.

Attached is a photo I took recently inside an RV7 QB tank. Note that there is NO proseal on the rivet heads. It doesn't say much for Van's quality control procedures.

It is now apparent from feedback on VansAirforce that some QB tanks have had the rivet heads sealed but many have not. In other words the QB tank subcontractor deviates in the construction sequence at will.

I have several friends with blistering paint over QB tank rivets. It's a widespread and common problem and Vans needs to address the situation immediately (if they have not already done so). To date they seem to have simply left the QB subcontractor to his own devices.

The real tragedy of the weeping rivets is that normal pressure testing methods will not reveal the problem. You only find out after the tanks have been painted and are in service. That's a very sad situation for those who have beautiful paint jobs ruined.

I'm not saying that lack of rivet head sealant is the only cause of fuel weeping....but it's one of the causes. Rick Galati has clearly outlined on this thread (and other threads) the numerous reasons why unsatisfactory sealant application can lead to leaks in many areas.

Initially the rivet weeping problem caused widespread confusion and many theories were put forward as to the cause. Time and feedback have however confirmed the initial theory of Rick Galati in that poor sealant application is solely to blame.


http://img824.imageshack.us/img824/4778/p1000897m.jpg (http://img824.imageshack.us/i/p1000897m.jpg/)

[Rick6a]

I rotated Bob's picture 90? so the casual viewer has a normal perspective of the quickbuild fuel tank stiffener illustrated in his post. In addition, I inserted a page pulled from the RV builder's manual because that information is available to everybody who builds an RV and describes the cause of the condition of the circled rivet shop head as "bucking bar slipped."

In the real world, that rivet *could* endure just fine EXCEPT that in this case the stakes have been raised dramatically because that defective rivet is set in a dimpled hole through a sealed fuel tank and is obviously not encapsulated. In the real world, all it would take is the slightest microscopic pin hole though the existing sealer barrier to open up a (potential) pathway to the paint coating on the surface. You could put a balloon test to the tank for a solid week or more and not perceive the (potential) problem. If I were going to apply a labor intensive, multi thousand dollar paint job, I most certainly would not view nor accept that circled rivet as "good enough" and then like so many others, succumb to the siren call of "BUILD ON!" In fairness and to be completely objective about it, that rivet may never weep...apparently the quickbuild production facility is betting it won't...but are you willing to take that chance with your custom paint job and then risk having the blistering problem reported by so many others raise its ugly head in your direction weeks or even months later?

http://i56.tinypic.com/15eva4h.jpg

[David-aviator]

Rick,

You have stated more than a few times that correct application of the sealant is most important and that the fay technique used by your employer is best. Obviously it is, big corporations do not waste money on production techniques that do not work.

WHY is this method better than Van's recommended method of immediate riveting as the parts are assembled with the sealant?

I have thought the matter through and come up with this scenario.

Two parts clecoed together with sealant between them and allowed to cure partially will leave more sealant between the 2 parts than if they were riveted immediately. There is more compression on a joint with bucked rivets and the sealant will be forced out from between the 2 parts and could leave some areas of the joined surfaces rather devoid of sealant.

Is this reasoning in the ball park?

[Rick6a]

....This I also know. When you shoot or squeeze rivets into very wet sealer, in reality only a thin film of sealer remains between the parts because the majority of the sealer is squeezed from between the parts in the riveting process. If the builder fay sealed in a manner that allowed ANY voids to exist in that remaining thin film of sealer...it is all the more reason to suspect that fuel or fuel fumes could eventually find its way into a void or voids hidden within that fay sealed surface..... Dave,

To help answer your question, I quoted myself drawn from post #97.

"Fay" sealing is not just a big corporation issue with its eye focused on the bottom line, it is a sheet metal best practices issue. Fay sealing, filet sealing, butt sealing, encapsulation are all accepted practices outlined in a variety of places that production facilities use because they know the accepted standards work. Saying that, the specific technique I personally use (clecoing wet parts together and then allowing the sealer to partially cure) is my own self developed technique borne out of decades working with sealer. I've always had an interest in keeping the mess to an absolute minimum. I can assure you with an army of Quality Control inspectors, Navy auditors, Air Force auditors, and engineers from a variety of disciplines CONSTANTLY looking over your shoulder, not one time was my technique ever challenged as violating industry standards that McDonnell-Douglas codified into a 6 foot long row of books formally referred to as the "Process Specification."

...WHY is this method better than Van's recommended method of immediate riveting as the parts are assembled with the sealant?


Van's method has to be driven in large part by pure economics. People work by the hour and a cost conscious employer will not buy into the idle workers' excuse "I'm waiting for the sealer to dry." The quickbuild option has to be an affordable alternative or builders would not buy into them. It is very hard to go into this further without getting political. But the reality is that to build those affordable fuel tanks for the mass market, the labor is currently done elsewhere. By and large that labor does a fair job and safety is never compromised. Since blistering has been characterized as a "cosmetic" issue, there is little incentive to increase costs to Van's and the consumer by asking that foreign labor to spend any more additional man hours than they really have to in order to get the product out the door. That may seem a simplistic notion to some, but that's how I see it.

[David-aviator]

As reported in another thread, I am building new tanks and have been reviewing the many suggestions and tips here and else where to get these tanks built right.

In the process of that review, one suggested cause of the blisters is rivets not being wet installed or with inadequate sealant leaving part of the rivet dry. Having set a few rivets getting started on the project, I can understand how this could happen.

I just returned from the hangar after inspecting some of the blisters with a bright light and magnifying glass. I can not see any proseal in the edge between the skin and the exposed rivets, none. Proseal may be around the rivet shank but it is not visible on the surface where the rivet head meets the skin.

One new blister this year (7 years after build) has a blister over part of the rivet. The part of rivet exposed due to the bliser is very shiny between the rivet edge and the skin, there is no proseal in evidence on the surface.

Could it be the problem is as simple as inadequate proseal between the rivet head and skin? Perhaps that's just part of the problem. Inadequate proseal application in the interior with exposed shop heads and voids could be the other half of the equation.

[Lote]

For another data point, we have slow build tanks we did ourselves. Fully encapsulated shop heads inside. Quick wipe with MEK on the outside. One tank has a few blisters, the other doesn't (since 2006). I believe the blisters developed after about two years. Paint is Imron over etched, alodined, variprimed surfaces. About six months ago I cut out the blisters and painted Imron (no other prep) on the spots with a brush to touch up. They have not re-blistered (yet).

LeRoy Johnston
RV-6A N176LD "Esperanza" 440hrs

[rv9av8tr]

Yrs have passed and 317 postings on this thread and still no fix. I purchased QB wings/tanks just 3 yrs ago on the theory that my tanks would be perfect since would be done by someone who knew what they doing. Well... NOT!!

After 6 mo of fuel in them, both tanks developed scores of blisters top and bottom. OK I'm P.Od!! The multiple color paint job is as good as ruined. This IS a problem Vans is responsible for because they didn't, and one can only assume still have not, ensured adequate QC.

My painter had me pin prick the blisters as they occurred to at least limit the extent of the delamination. Even on the really large blisters, there is no evidence of liquid. This is clearly a vapor barrier breach from within the tank.

Has anyone found a successful sealing repair from the outside?
I'm wondering about flush filling the rivets heads with some kind of epoxy
filler.
What about "superglue" around the rivet heads, it seems to have a pretty good wicking action.
Frustrated as h---!

[scsmith]

I have what appears to be a success story. Can't really draw any conclusions, but it is information.

I have RV-8 Quick-Build wings with tanks built in April-May 2007.
Wings were painted before ever having fuel in them, except for a quick leak test.

Before painting, on the slight hope that it would make a difference, I put a drop or two of green Locktite sealant on each tank rivet head. This was allowed to penetrate for two days. I had no mechanism to pull a gentle vacuum on the tank at the time, or I would have. After two days, the excess green Locktite was washed off with a quick wipe of acetone.

Wings were painted with JetGlo system. First washed with a phosphoric acid wash, then soap and water, then dried several days. Then the self-etching primer that is used with JetGlo, then White JetGlo.

Wings were then stored for 6 months before assembling airplane and filling tanks with fuel. First fueled in August, 2009. So, it has been TWO YEARS, 150 hrs hobbs. Airplane has always been hangared except for a few long trips where it sat out for a few days in the summer sun.

NO BLISTERS! NONE!

I can say that I do NOT have particularly good paint adhesion on my wings. When trimming the inboard edge of the flaps when assembled to the fuselage, the aluminum curled up when cut with the shears, and the primer and paint peeled off the aluminum.

I can not say much about the particular level of sealant inside my tanks. I never looked that close. I think I recall significant fillets around shop tails, but not fully encapsulated. I think I recall generous fillets and evidence of fay sealant.

I don't know if perhaps the fact that they where made in April or May in the Philippines helped - maybe there was less need to contaminate the sealant by thinning with MEK if the weather was milder? (Is it even milder there in our spring?)

I don't know if my Locktite sealant application helped? Maybe?

[170 driver]

Has anyone built their tanks with Proseal 890 only instead of the flamemaster sealer from vans? Has there been any blistering on the proseal tanks? I used the flamemaster and have some on one tank. I also had a leak on the rear baffle and the sealer along the lower seam softened from the leaking fuel. I have used proseal at an OEM facility for years and have never seen any blistering or softening. Just wondering if the flamemaster is the culprit here. I know it is supposed to meet the mil specs for 8802, but I am not happy with the results. I wish I had used the proseal 890.

[detlef lili]

Has anyone found a successful sealing repair from the outside?
I'm wondering about flush filling the rivets heads with some kind of epoxy
filler.


Hi!
I sanded down the paint to the rivet and covered with one layer of very thin glass with epoxy (West System).
I have again one blister after 3 month.

Detlef

[Paul K]

I am getting ready to paint my QB tanks and will look inside next week to check the quality of the seal. I might also try the green Loctite as an added effort that may help prevent blisters, can't hurt!

Other than that, I am open to suggestions.

[Cfrisella]

Hi!
I sanded down the paint to the rivet and covered with one layer of very thin glass with epoxy (West System).
I have again one blister after 3 month.

Detlef

Can you tell if the blister raised the glass and epoxy, or has the vapor, creating the blister, seeped through the fiberglass.

[jurik]

I noticed the blisters only after 6 months but it may have appeared earlier. I removed both tanks, stripped the paint and primer and painted it again. This time however I ensured that I paint over the primer within 12 hours as Van?s think it may be the bonding between the paint and the primer. In a way I tend to agree with this as the primer did not blister, only the paint. It has now been 6 months after repainting and no blisters.

[rv9av8tr]

Van's can point their finger at painting all they want. But the reality is it takes a force to lift the paint off the surface and that clearly comes from within the tank. So the root cause is not a paint issue, but a leaking rivet issue. There is no way to fix a poorly assembled QB tank except from the outside. To me, all QB tanks should be approached with great suspecion. There needs to be a reliable way to remedy this externally. Using something like Green Locktite sounds like it might be reasonable place to start. I sure as heck don't intend to build new tanks...

[bsacks05]

I am currently adressing the blister situation with my left tank. The right one is blister free. Both were built by me and not at the same time. I am convinced that I was not careful enough with sealant regarding shop head encapsulation and wet installing rivets.
So....since I am fixing up other issues with the plane, I decided to sand down the paint over the tank rivets and smear some proseal over each head ensuring that it gets between the rivet and the side of the dimples. Some of my rivet heads had gaps between the edge and dimple. :( I know that sealant repairs should be done from the inside, but I am taking the simplest route first.
I will prime over them soon and report the results.

[rv9av8tr]

I would think proseal on the outside would leave a visual telltale of "goop" even after painted.

I stumbled on a local RV builder who used Loctite 290 Wicking formulation (it's red rather than green) to seal a leaking rivet. Rather than drawing a vacuum on the tank, he used a shop-vac to pressurize the rivet head. Didn't leak afterwards.

Also found this interesting article: http://www.seqair.com/skunkworks/Maintenance/TankLeak/TankLeak.html

I detect a big winter project coming up...

[bret]

2006 TO 2012 and no answer yet? I did the 290 Loctite method on my QB wings last night just for extra insurance. I am getting ready to paint the wings this week end. So what is the success rate of the Loctite procedure as of now?

[Jeff Vaughan]

All my fixes done with Loctite have held up to 220 hrs. so far

Have developed a few more leaks on other rivets on my QB wings since paint.

Total leaks to date 8.

[bret]

All my fixes done with Loctite have held up to 220 hrs. so far

Have developed a few more leaks on other rivets on my QB wings since paint.

Total leaks to date 8.

Are you saying you have leaks on non fuel tank rivets?

[ron sterba]

Sitting here thinking about this problem and one thing comes into mind when I worked in the payphone division of the local phone company. The steel screws used in aluminum phone booths on the Oregon coast were affected by electrolysis between the unlike metals. It was the aluminum that turned to white powder next to the steel screws. Another thing to consider is electricity is also used in the booths which may accelerate the electrolysis process. Now consider the plating process on the rivets and was the coating removed prior to painting and just maybe the alloys in the metal of the rivet might have migrated to the aluminum causing a release of gases (ever so small quantity ) to form the bubble. I might be inclined to use a straight pin to pierce the bubble at first sign. Maybe static wicks?

Just my two cents worth but a penny or two for your thoughts. Vi'sa Versa or whatever!@#$%^&*
Ron in Oregon.
QB9A

[Jeff Vaughan]

Brett -

All my leaky rivets are tank rivets

[scsmith]

2006 TO 2012 and no answer yet? I did the 290 Loctite method on my QB wings last night just for extra insurance. I am getting ready to paint the wings this week end. So what is the success rate of the Loctite procedure as of now?

I put a dab of loctite on each tank rivet of my QB wings prior to paint. That was 3 yrs ago, and 2.5 yrs since first fuel in tanks. No blisters.

A few posts back, someone suggested using a shop vac to locally pressurize the outside of each rivet to force the loctite into the rivet. Thats a great idea, I wish I had thought of that -- although it would be pretty time consuming.

Wicking without a pressure differential appears to have been sufficient.

[CPSONE]

Has anyone asked the guys at Willmar Aircraft in Willmar Minnesota about this problem? They are the Mooney leaky fuel tank experts, great guys. They worked out a special process to remove tank sealant from the inside of Mooney fuel tanks and re-sealing them. I bet they would know all about tank rivet blisters etc.

[bsacks05]

I am still convinced, in my case anyway, that I did not use enough sealant along with improper sealing technique when I built my left tank. Specifically: inadequate shop head encapsulation, inadequate fillet seals, and not filling gaps between the manufactured heads and the dimple. I was apparently more careful on the right tank because it has had no blisters. The left tank blistered in numerous location at the heads.
This past summer I stripped the paint at the offending rivets and could actually see a gap on some between the edge of the head and the dimple where there was no sealant. I pressed some fresh sealant over the heads, let it cure, then sanded and brushed on some epoxy. After some light sanding and repaint, the repair is barely noticeable. Unconventional yes, but it's working very well with no blisters yet.

[detlef lili]

I put a dab of loctite on each tank rivet of my QB wings prior to paint. That was 3 yrs ago, and 2.5 yrs since first fuel in tanks. No blisters.

A few posts back, someone suggested using a shop vac to locally pressurize the outside of each rivet to force the loctite into the rivet. Thats a great idea, I wish I had thought of that -- although it would be pretty time consuming.

Wicking without a pressure differential appears to have been sufficient.

A lot of good hints!
Much easier as I did with covering with Epoxy. After six months, I had another leaking rivet. Will try Loctite 290 now.

Detlef

[bret]

When I did the 290 with the shop vac the other night, I did not rig up the water column and what not, I held the hose over the tank fill about 3/4 covered and watched the tank skin, It would take a lot of the vacuum force before even starting to bring the skin in, I would imagine I was way over the 15 or so inches of water vac that others have suggested, my thinking was that it needed a lot more force to wick in the loctite. Also one tank handles about 750 Lbs of force if you use the 6 G load formula...right? so I was not worried about any damage to the tank.
Anyway, I prepped and sprayed epoxy primer today so we shall see......

[Bob Kuykendall]

...Also one tank handles about 750 Lbs of force if you use the 6 G load formula...right?...

Yes, that is so.

However, when you look at the projected area of the tank, it's not much pressure. Eyeballing the tank at 50" span x 15" chord, that's 750 in^2 of area upon which the fuel rests. So at 6g, the fuel applies on average 1 lb/in^2 to the skin. Still, that's 144 lbs/ft^2, not inconsequential.

Thanks, Bob K.

[bret]

Just had a thought, some one else posted a possible cause of the primer or paint not curring and trapping in the reducer-solvents.
The epoxy primer I am using says cured and able to sand in 8 Hrs, that is for a flat surface, the rivets hold a deeper layer of primer, I can see where if you top coated too soon you can trap this uncured primer. After , 24 Hrs I wet sanded the primer and could smell fumes, So my thinking is that if you have a marginally sealed rivet, the uncured primer-paint layer could yield fuel vapors and bubble the finish?

[Dave Stephens]

5 RVs at out field 3 of them have bubbles on tank rivets. 2 of that 3 are QB

Mine is QB and i have 4 or 5 very small bubbles, but all on the top of the tanks I used Upoxy 2 pack etch primer and Automotive base coat with clear laquer


Dave

[bret]

Just trolling around the threads I cam across some one on the 8 sight that has noticed an oil type film around the skins at joints and rivets, apparently they coated the QBs with something for the ocean voyage. I never notice this before, but last week I epoxy primer-ed, wet blocked and left to cure. yesterday I go out to the garage and at the tank seam there was a 1/4" wet line along the seam, look like some type of oil, none of the rivets had this and maybe because I did the Loctite 290 thing. But any way I wiped with grease and wax remover and it did not come back. Painting tomorrow and hope it comes out well. Paint Blisters,Fuel vapors, uncured primer paint thinners, OIL?

[Steve Melton]

perhaps the wing skin opens enough to allow a small amount of fuel around a rivet or flange when it flexes in flight and then traps it after unloading (skin flexes back to ground condition position). most probable area for skin flexing would be the top skin towards the spar. when baked in the sun the fuel vapor finds the easiest way out... around a rivet. the only way to fix this type of problem is from the inside of the tank. I have QB wings in a stand and plan to take a look at the inside with a borescope. thinking about paint, would an acceptable solution be to paint everywhere except the tank or maybe paint everywhere except the tank rivets? that would look cheesy, swiss cheesy.:D

[scsmith]

I have almost 3 years of flying now, 200 hrs, and my QB wings have no blisters. maybe the green Locktite 290 treatment really helped?


Or maybe my wings were made on during a cool spell, so they mixed the proseal right, no MEK, and did a good job getting the rivets coated.

Anyway, I recommend doing the Locktite treatment, can't hurt, costs almost nothing, no reason not to do it.

[chazking]

My QB RV-10 tanks are weeping, profusely, around many rivets. I see weeping only when the tanks are full & in the hangar for several days. All weeping rivets (~3 on each tank) have been circled with blue marker. After last fill-up (two weeks ago) noticed another rivet weeping.

In my case the plane is not painted, has had no paint preparation, and just over 1 year flying. While the weather is much warmer, now, the weeping started in February soon after relocating the plane to it's current hangar.

I spoke with Paul @ Weep No More & he's had no experience with sealing tanks for an RV-10, but thought the process would be no different than other seals (except that the tanks are relatively new). I sent him pix today & expect to chat with him tomorrow. Shipping of the tanks could be an issue due to size; more later.

I subscribe to the idea of pressure differential, but not to the notion that painting has anything to do with the weeping.

Clearly, this condition requires resolution before painting (planned for early Winter 12).

[johngoodman]

I got the rivet blisters about a month after I first filled the tanks with fuel. It was already painted over six months earlier. I have never seen so much as a drop of fuel in the blisters, they are all dry. They are only on the RV-10 fuel tanks - nowhere else. My tanks were Quick Builds.
After one year of flying, I now have one rivet that actually leaks fuel, but it never had a blister and it's the innermost row of rivets on the starboard side. It's on top, and could be explained by passenger feet on it. I plan on fixing it with the green loctite method.

John

[Space Cadet]

Just to add to the database- my 10 just started showing about a dozen paint blisters, top and bottom of fuel tank, all on rivets. Mostly on one side, but a couple tiny ones starting on the other.

Facts: QB wings, bought from another builder but think they are originally from around 2003-5 or so vintage (I'll go double check). Painted myself, SW JetGlo Express with SW epoxy chromate primer, thoroughly cleaned with soap/water, MEK, and a pre-paint cleaner (ammonia-based I believe). Only 60 hrs total flying, hangared, fuel in tanks since early spring. Vents confirmed to function (with empty tanks, think I will double check it again). Only significant outside exposure was my week at Osh this summer, which was also the last time it flew before I noticed them. Only one or two had anything inside them (that light brown oily fluid previously discussed, and those were on the top), rest were dry. No fuel seepage that I can tell, yet. I did not do any real inspection of the fuel tank interior rivets prior to flying, so don't know how well they were covered with sealant.

Dwight

[Erik.37m]

I've read this thread for about 3 hours and man.... what effort has gone into this problem is incredible! I've notice people here now that were reporting on this back at the start! I'm very much impressed with everybody! Some quite funny! I'm way past bedtime and still at work reading this thread.

I'm planning on a RV-14 QB purchase and sure don't want any issues. I'm in manufacturing and I know how hard it is to get good workers so you have to rely on the leadman to keep Q/C intact. I hope they have rules they must follow and they don't skirt them when given a chance.

My take is it is a sealing issue. I'm a welder/ machinist/ fabricator and I know all too well about little tiny cracks, voids, etc. Try sealing for 150 psi pressure in a fatigued-thermally cycled chunk of steel and you can chase leaks all day long.

All joints have to be sealed, mating surfaces, rivets, anywhere there is a joint.

I'm not up to speed with aircraft but "Slosh sealing" is done all the time. Maybe the weight is the issue. I have seen bladder tanks and they are heavy.

Vans should emphasise this to the vendor in the Philippines in the up most. This phase of building should definitely rank a stamp of conformance by their inspector.

I feel very sorry for the poor guy way back when when his pride and joy show plane was pulled from "Best of Show" contention at SNF and Oshkosh.

Erik

[Erik.37m]

I gave this more thought last night. I thought of a inspection technique that could be used after assembly. I worked at a company called hi shear in Torrance , Ca.. They make aerospace fasteners for the likes of Boeing. For heat treating Titanium parts they have vacuum furnaces. When they weren't able to pull down a vacuum to purge the retort of air they would call in an outside vendor to do a "Helium leak test". Those tiny molecules can pretty much find any opening.

I don't know really how big of a problem this is in the real scheme of thing but it is a fast inspection technique and cheap. If the Q/B team had this for an inspection process it would ease the buyers mind I'm sure.
Erik

[Mike S]

I'm not up to speed with aircraft but "Slosh sealing" is done all the time. Erik

Slosh used to be pretty standard for fuel tank sealing, but over time it has been a problem------eventually comes off and plugs up the fuel lines.

I do not know if the issue is the ever changing formula of our avgas, or if the slosh compound has a limited service life or somethinf, but many folks have had to remove the stuff from their tanks.

[rv9av8tr]

I definitely recommend that any builder using QB fuel tanks, that they also seal all tank rivets externally with Lok-Tite wicking formula or filling over the rivets with micro and a strip of fiberglass tape. You really don't know for sure what you're getting with QB tanks.

My tanks are a mess, top and bottom. Plan to work them over this winter/spring.

[scsmith]

... they would call in an outside vendor to do a "Helium leak test". Those tiny molecules can pretty much find any opening.
Erik

This is a really good idea, actually. Hellium sniffers are really expensive, but not out of the question for a shop to acquire to do QC on a lot of tanks. The good ones are ball-park $20K. But small hand-held ones for leak checking are about $4K.

So, if Van's really wants to do some quality assurance on QB tanks when they come in, before shipping to customers, they could buy one of the $4K sniffers and leak-check every QB tank. Also, slow-build builders could send their completed tanks in to get tested for a fee.

Now, helium is pretty hard to keep in a tank. It may be that all our tanks would ooze helium. But presumably, it would be a matter of degree, and testing a few would establish a threshold where it is OK vs. one that would leak fuel vapor. It may also be that with some experimentation, a different gas (you can buy sniffers for O2, N2, CO, CO2, variety of others) could be found that would be a more realistic leak rate that would provide a good go/no-go test.

One might argue that it would be more straightforward to train the QB builders to make them better in the first place, but in 3 years, we haven't seen any evidence of that.

Perhaps some vendor testing of some tanks that have been in service and got blisters, and not, would go a ways toward convincing Van's that there is more of a problem that needs addressing? We would need to find a vendor, and arrange for some good samples to test as an experiment.

[g zero]

Working on my slow build 8 fuel tanks , I was thinking as I was applying sealant to the rivets on the amount of coverage on each rivet . Dan H had some photos showing the " sponge " holes in dried sealant . If rivets are covered with a thin layer would the fuel permeate thru as a gas ( not liquide ) . With that thinking I sealed each rivet with a 5/8 " diam - sealant thick enough not to see any signs of a rivet .
Will report back after a year of full fuel with paint .

[nineninefour]

I'm curious about behavior of Proseal in thicker layers. Most 1-, and 2- part sealants will shrink and tear in thick layers, particularly where the thick material becomes thin, even when it stays adhered. Usually we see this with small mechanical stresses or thermal stresses, both of which could be present in these tanks. Any observations?

[pa38112]

My plane was Painted in 2004, QB tanks. 2 blisteres appeared last week after 8 years in service. Tempature has been in the 40's. They were next to each other on the botom of the tank. The tank was empty and open to atmosphere for a week before these appeared because I was fixing a small leak on another rivet. I cut off the paint, and there is no sign of fuel.
After fixing my leak, a new leak appeared on the rear flange. My guess is that the QB tanks don't use pro-seal brand sealant, or that they don't abide by the shelf-life dates. This causes the material to not fully cure and degrade when exposed to oxygen. There is then off-gassing that causes the blister. Heat will accelerate the degradation...

[BobB]

I don't know if my RV-7a is a QB or not, but I have bubbles.
SN is 71963. Has anyone explored whether the tank caps have not been releasing positive pressure? Don't see a way to attach photosto this message.

[BobB]

Here are photos of some examples.

http://i50.tinypic.com/21etd36.jpg

http://i46.tinypic.com/2lcxnxj.jpg

http://i48.tinypic.com/2jcalwk.jpg

[Mike S]

Bob, welcome to VAF:D

The fuel caps are sealed, tank venting is done with a dedicated line, should be at the wing root, or on fuse close to the wing root. Look for a bit of 1/4" tube sticking out a short distance, usually has a slash cut on the end.

Good to have you aboard, Sorry about the blisters.

[BobB]

Thanks for the reply. Obviously, I didn't build this a/c. I will clean up the bubbles and repaint the areas. Hope the bubbles don't come back.

[MPB]

For what it is worth I had the same problem on a motorcycle fuel tank. I stripped it, cleaned it real well and re-painted it. The problem seemed to be fixed for about a year and then the bubbles came back in the same spot. In hindsight I should have been more diligent figuring out why they started in the first place. just my 2cents...

[DanH]

For what it is worth I had the same problem on a motorcycle fuel tank.

Mike, was the tank welded steel, blown or roto-mold plastic, fiberglass, what? Was the blister on a seam or elsewhere?

[edbooth]

Thanks for the reply. Obviously, I didn't build this a/c. I will clean up the bubbles and repaint the areas. Hope the bubbles don't come back.

They will come back in time, you need to fix the source of the leak before going to all the trouble to repaint. I have experienced leaks in factory built tanks on RV-7 and RV-10 tanks. Even a factory replacement tank for the RV-10 leaked. I think there is extremely poor QC on the building and testing of these tanks. The only one I had that did not leak was the ones on an RV-6 that I built.

[JordanGrant]

Just another data point for posterity:
My RV-6 wings now have blisters.
Data:
1. Slow-build (did it myself), in 1999/2000 time period
2. Std pro-seal methods, no slosh
3. Painted w/Sherwin Williams products (JetGlo/Acryglo)
4. Took 6 years to develop. First flight 2006, first blisters 2012.
5. Correlated with heavier use of 91 octane autogas, laced with ethanol, I started in 2012, although the right wing (which is filled w/100LL 90% of the time) also has the blisters
6. The blisters are still intact, no liquid has ever been noted out of rivets

Plan is to monitor - trying to fix/re-paint would probably make it look worse in the long run, and there is no safety of flight problem at this point.

Cheers,

[Stalldog]

Jordan, where on the tank did the blisters appear, i.e. along the rib line or along the rear baffle, interior rib or exterior rib, etc.? So after six years blisters appeared at the same time on both tanks???

[Skyote]

My quick-build RV-8 started developing blisters on the tank skins a few months after first flight. It seemed to accelerate when it was out in the sun for a week at Oshkosh. I ended up waiting for about a year for the situation to stabilize before deciding what to do about it.

I figured I would have to repaint the tanks, so I decided to try an experiment. The biggest blister was on the bottom of the left tank, about 1.5 in diameter. I used an Xacto knife to make a small slit, about 1/32 in long in the middle of the bubble. I then used a heat gun (low setting) hoping the blister might shrink down flat. And it did.

This worked so well that I did this on all the blisters. After making the slit (parallel with the slipstream), I would apply some heat for a few seconds and then massage the blister down with a cloth. Then repeated this several times. The result was impressive. Unless someone knows specifically where to look, the repair is not noticeable. It is least noticeable on the white paint, a bit more an the red trim. I did this a year ago and none of the blisters have returned. The result is good enough (so far) that I will not go through the hassle of repainting the tanks

This is, of course, just one data point, and success may be a strong function of the brand of paint used.

[Mike S]

Ken, welcome aboard the good ship VAF:D

Good info on the blister repair, another data point for the puzzle, thanks for sharing with us.

Sounds like lancing the blister at first notice may be as good as it gets without having to open the tank??

How about a photo or three of the plane??

[Skyote]

Mike, tried to post a picture, but unsuccessful.

[Mike S]

Mike, tried to post a picture, but unsuccessful.

Ken, check your Private Messages, upper right corner of the page.

[g zero]

BMW Motorcycles made a painted plastic tank for their GS line around 1994 - 1995 , similar blisters would appear , I had one that was blister free for 2 years until I painted it ( DuPont Chroma Base / Clear ) . First day in the sun it looked like a teenager with bad acne ! BMW warranty removed most from the market and replaced with a metal one .
Were there some microscope photos of Proseal posted ? I wonder if there is a minimum thickness that should be used to cover the shop head of rivets on the inside of the tank ?
Tom

[DanH]

Were there some microscope photos of Proseal posted ? I wonder if there is a minimum thickness that should be used to cover the shop head of rivets on the inside of the tank ?

Most of us are using Flamemaster CS3204, while "Pro-Seal" is a trademark of PPG Aerospace. Both are AMS-S-8802 (formerly Mil-S-8802F, Type II) sealants.

Cured sealant is actually much like a sponge, although in theory the voids are closed cell, i.e. not connected. The voids have two sources, (1) air entrained during mixing and (2) bubbles of liquid or vapor toluene, the solvent component used for viscosity adjustment. The solvent evaporates out of the sealant while the chemical cure is proceeding, and apparently for some time after. It would seem the sealant is never totally vapor proof, which is why there is a required thickness.

BTW, there is no test standard for sealant performance with avgas or mogas. All testing is done with what you can think of as a standardized jet fuel equivalent.

https://www.danhorton.net/VAF/Paint%20Blisters/Voids.jpg

Thickness over a shop head? 60 mils with a 1/4" fillet.

This is from the original Essex Chemical document "Polysulfide Sealants For Aerospace". They invented the stuff.

https://www.danhorton.net/VAF/Paint%20Blisters/Sealant%20Filets%20-%20Essex.jpg

https://www.danhorton.net/VAF/Paint%20Blisters/Sealant%20Thickness.jpg

[WildThing]

My -8A QB has a herd of fuel tank rivet blisters. Not going to bother with them at this time, but wondering if sloshing is the only way (yet) determined to correct the issue?

[DanH]

My -8A QB has a herd of fuel tank rivet blisters. Not going to bother with them at this time, but wondering if sloshing is the only way (yet) determined to correct the issue?

http://www.vansaircraft.com/pdf/sb11-9-13.pdf

[g zero]

Dan , on the quick build tanks and others that you have seen with blisters are they lacking total encapsulation of the shop heads ?
Tom

[DanH]

Dan , on the quick build tanks and others that you have seen with blisters are they lacking total encapsulation of the shop heads ?
Tom

Not encapsulating shop heads would violate all the industry standards (ex. SAE AIR4069). It is no doubt a contributor to overall failure rate, but not the problem. Consider blisters along the rear baffle line; shop heads are outside the tank.

[Rick6a]

.......on the quick build tanks and others that you have seen with blisters are they lacking total encapsulation of the shop heads ?
TomYes, blisters have been observed on quick build tanks that lack shop head encapsulation. It was the dozens of blisters on his quick built fuel tanks that prompted Dave Domeier to start this thread in the first place. One need only review the images in post #100 to confirm that shop head encapsulation was indeed omitted in the construction of those (not Dave's) quick built fuel tanks.

That said, a more in-depth review of this thread will reveal why blistering can occur along the rear baffle and virtually everywhere else for that matter if the fay seal is defective. A proper filet seal is highly recommended and industry accepted practice too.

[Martin Sutter]

My quickbuild RV7A has developed 6 separate leaks on rivets of the right tank. This is a 2005 vintage quickbuild. The leaks have all developed within the last year. All have been on the bottom ribs. Looking in the filler neck I can see that the shop heads of the rivets are only encapsulated on one side. It looks like they used an applicator and made one pass down the rib flange, leaving ample sealant on one side of the rivet (the front side in the direction of application) and a large void on the aft side. You can clearly see the shop head and the dimple is clear of any sealant on that side. I presume that was the mode of application used on all the ribs. Close inspection also shows tiny blisters on all bottom rivets of the tank, none so far on the top side. The aircraft has been in service since 2006 and has only been fueled with avgas.

Martin Sutter
Building and flying RV's since 1988
EAA Technical Counselor

[WildThing]

http://www.vansaircraft.com/pdf/sb11-9-13.pdf

Thanks Dan. I guess I'll wait for Plan B - whenever that get's here. Or build a new set of tanks down the road if they get unsafe.

[JordanGrant]

Jordan, where on the tank did the blisters appear, i.e. along the rib line or along the rear baffle, interior rib or exterior rib, etc.? So after six years blisters appeared at the same time on both tanks???

My blisters are all along the rib lines, in the front half of the tank. None along the rear baffle. Primarily on the bottom, although I have one on the top, near the leading edge. I might try the razor blade/heat gun repair technique on the bottom ones and see how it goes.

[Janekom]

About to paint a new RV10 and was wondering if someone has the document that Vans released some time ago on this issue. Thanks.

[jcbarker]

QB tanks 2003 vintage.
Professional paint 18 months ago.
One rivet leaked along rear flange. Fixed before paint. No problem
Blisters just showed. 5 on lower left. 1 on lower right.

Is Van's still silent on this issue?

[simatos]

Ouch this is not a discussion I wanted to join. I got these $#%^%$# blisters on my wings too, What the Heck??????!!!!!!!!!
Just noticed them. Quick Build 2007
painted one year ago
leak test no problems
Only on the tops, dark blue paint.

Am I looking and pulling of and resealing??????

G

[rv9av8tr]

My QB wings are of the same vintage, 2006 and both sides, top and bottom blistered like crazy after a year. I was so pissed!! None of the blisters showed any signs of fuel staining, so was strictly internal vapor pressure delaminating the paint. I ended up sanding the paint down to bare metal all along the rib / rivet lines. Then filled the rivet tops with wet micro balloons, laid a ply of fiberglass strip, then dry micro balloons over that. Sanded and filled, no more blisters.

[simatos]

Holy a word I cannot use
**** paint job was 10 grand
very disappointing all around. Im this close to flying the darn thing. Guess Im gonna have to add blister repair to the never ending list. Thanks for the info. I hadn't thought about your fix. You did it with the wings on, is that right??
Thx Gary

[bret]

Just trolling around the threads I cam across some one on the 8 sight that has noticed an oil type film around the skins at joints and rivets, apparently they coated the QBs with something for the ocean voyage. I never notice this before, but last week I epoxy primer-ed, wet blocked and left to cure. yesterday I go out to the garage and at the tank seam there was a 1/4" wet line along the seam, look like some type of oil, none of the rivets had this and maybe because I did the Loctite 290 thing. But any way I wiped with grease and wax remover and it did not come back. Painting tomorrow and hope it comes out well. Paint Blisters,Fuel vapors, uncured primer paint thinners, OIL?



Just wanted to do an update and add data on the green Loctite trick, over one year with fuel in the tanks and no problems yet, the paint was mechanical etch with maroon scotch bright pads, epoxy primer, 360 wet sand and then base clear coat, fingers crossed. (above post was from 2012)

[airguy]

I've got about a half-dozen blisters forming on my left wing inboard tank, at 40-ish hours and 4 months after paint. Very small but definitely vapor blisters, all in one cluster near the rear baffle at one internal rib. I'll wait a bit to see if any others pop up before doing anything about it, probably this summer.

[DanH]

The last blister survey was circa 2009. It would be interesting to gather fresh data. If you have blisters, and would like to answer a few survey questions, please send a private email.

[PilotjohnS]

I've got about a half-dozen blisters forming on my left wing inboard tank, at 40-ish hours and 4 months after paint. Very small but definitely vapor blisters, all in one cluster near the rear baffle at one internal rib. I'll wait a bit to see if any others pop up before doing anything about it, probably this summer.

Greg, Curious if you had used ?Tank dies? on fuel tanks?

[airguy]

Greg, Curious if you had used ?Tank dies? on fuel tanks?

I did, yes, as that was considered the current vogue thing to do at the time I was building my tanks. I used the wet fay-sealing method on all my rivets as well.

[rv9av8tr]

Holy a word I cannot use
**** paint job was 10 grand
very disappointing all around. Im this close to flying the darn thing. Guess Im gonna have to add blister repair to the never ending list. Thanks for the info. I hadn't thought about your fix. You did it with the wings on, is that right??
Thx Gary

I did it with the wings on. It was a PITA but I?m happy with the outcome.

[simatos]

Thank you guys, food for thought. Would sure like to avoid pulling the tanks:mad:

[dtw_rv6]

Here is an interesting wrinkle on the blister phenomena - My tanks were built 10 years ago, and I only had two minor weeps that I finally made go away with loctite. The tanks have been unpainted their entire life.

Last spring I wrapped them with vinyl along with the wings. The wings are holding up fine except for the tanks - I'm getting blisters over many rivets. I've cut a few open to find fuel pooled inside. Seems like the vapors are escaping where liquid cannot. I've never used anything but proseal (and a few drops of loctite) on the tanks.

Don

[BillL]

Here is an interesting wrinkle on the blister phenomena - My tanks were built 10 years ago, and I only had two minor weeps that I finally made go away with loctite. The tanks have been unpainted their entire life.

Last spring I wrapped them with vinyl along with the wings. The wings are holding up fine except for the tanks - I'm getting blisters over many rivets. I've cut a few open to find fuel pooled inside. Seems like the vapors are escaping where liquid cannot. I've never used anything but proseal (and a few drops of loctite) on the tanks.

Don

Sorry I can not help with blister solution at the moment, but a question: Would you recommend just applying vinyl wrap ( or a good primer and top coat) to the tanks as a test until the plane is ready for real ($$$) vinyl or paint? Blisters only seem to appear after the big investment.

[Yen]

My tanks built in 2013, scotchbrite and MEK wash, can't remember primer, Poly U paint.
Left tank used solely with Avgas,
Right tank used Avgas for 2 years then put in some Mogas. Shortly after it was all mogas I started getting blisters on the top of the wing, above fuel level.
Tried the locktite method with no good result.
Abraded paint and stripped back to bare metal, applied locktite again, then primed with epoxy primer and repainted.
So far there is no more problem, but I have given up using mogas.
What I think happened was that the vapour from the mogas was either so thin that it could get past the proseal, or it somehow chemicaly affected it

[bret]

My tanks built in 2013, scotchbrite and MEK wash, can't remember primer, Poly U paint.
Left tank used solely with Avgas,
Right tank used Avgas for 2 years then put in some Mogas. Shortly after it was all mogas I started getting blisters on the top of the wing, above fuel level.
Tried the locktite method with no good result.
Abraded paint and stripped back to bare metal, applied locktite again, then primed with epoxy primer and repainted.
So far there is no more problem, but I have given up using mogas.
What I think happened was that the vapour from the mogas was either so thin that it could get past the proseal, or it somehow chemicaly affected it


Do they mix alcohol with the gasoline down under? up here 10 percent is normal.

[dtw_rv6]

I had used Mogas exclusively for the first five years of flying. The last five have been all AV gas.

[mspenc45]

I believe that the tank venting design is part of the issue. Why? Simple, when I leave my tank caps open while on the ground in a hot hangar or in the sun, no blisters. Close the caps, blisters! The standard vent design allows fuel to flow into the vent tubes, this creates head when subsequent fuel vapor tries to escape. It's not much, but when I pop my caps after sitting there for a bit on a hot day, woosh, you can hear the pressure relieve itself past the cap as it is opened. This slight pressure surely encourages vapor transmission through the sealant, where it is subsequently trapped by the paint, which in some cases is apparently less permeable than the sealant, and it give way, separating from the skin and forming a blister. As others have noted, once the paint is vented, i use a hypodermic needle, that rivet will no longer blister the paint that has been vented. Soooo, is there another idea for a vent design that does not allow fuel into the vent tube, or at least lessens the potential for back pressure?

[PilotjohnS]

So I am wondering if the vent lines are not clear but rather have some raw fuel causing a liquid plug? This will allow some pressure to build up in the tanks if the internal pressure isnt enough to push the fuel all the way out, or to suck it all into the tank

[rvbuilder2002]

I believe that the tank venting design is part of the issue. Why? Simple, when I leave my tank caps open while on the ground in a hot hangar or in the sun, no blisters. Close the caps, blisters! The standard vent design allows fuel to flow into the vent tubes, this creates head when subsequent fuel vapor tries to escape. It's not much, but when I pop my caps after sitting there for a bit on a hot day, woosh, you can hear the pressure relieve itself past the cap as it is opened. This slight pressure surely encourages vapor transmission through the sealant, where it is subsequently trapped by the paint, which in some cases is apparently less permeable than the sealant, and it give way, separating from the skin and forming a blister. As others have noted, once the paint is vented, i use a hypodermic needle, that rivet will no longer blister the paint that has been vented. Soooo, is there another idea for a vent design that does not allow fuel into the vent tube, or at least lessens the potential for back pressure?

Interesting theory, but when testing the theory we need to consider that the vent design originates all the way back to the RV-4. There are 1000's of RV's with this design, many of them completed in the 80's and 90's with (as far as I am aware) no history of paint blisters.
It has only been within the past 10 years or so that they have been occurring.
My theory is that it is at least in part related to the EPA driven formulation changes that have occurred in paint systems over that time, but that is just a gut feeling.... I don't have any more evidence to back up that theory than any of the others that have been suggested up to this point.

[airguy]

I have outboard wing tanks with flow-through venting of the inboard from the outboards, with the outboards vented at the wing tip. The result of that is that I cannot have more than about 2" of fuel column pressure at any time on the tanks - I simply don't have enough vertical pipe runs to get more than that - so my tanks can't pressurize in the manner you describe. I still have gotten some paint blisters - on both inboard and outboard tanks.

[Huskyhipdoc]

https://www.dropbox.com/s/3e3khte17hq68ld/IMG_0991.jpeg?dl=0
https://www.dropbox.com/s/15ytz90ramxmz9u/IMG_0992.jpeg?dl=0
Abject despair today... then I found this thread..

2018 QB wings, cleaned with detergent degreaser, wiped down with MEK, then two-pack acid etch primer (P30B) left for 24 hrs then two pack epoxy primer (PR143).
Pressure testing of tanks was fine.
I had 20 litres avgas with fluorescent marker in the tanks for a week with no leaks, then the tanks drained fully and left open for 24 hours, and then the wings stored vertically (leading edge down) with the tanks fully sealed to prevent critters getting in.
Today, I moved the wings and one is fine, the other has 10 blisters, 9 on the lower aft flange and one on a rivet one place closer to the leading edge.
No top coat, and minimal fuel exposure yet Ives still got blisters!
I sanded them all down and obviously no fuel present under them.
Irritating beyond belief, but at least ive found them before spending $10k on a paint job.
So how do I fix the problem? epoxy over the rivets, fibreglass tape and filler?
I feel like setting fire to the bloody thing at the moment! I dont mind sorting out my own ****-ups, but sorting out the result of a "professional' build is maddening!

[Kyle Boatright]

So how do I fix the problem? epoxy over the rivets, fibreglass tape and filler?


I feel like setting fire to the bloody thing at the moment! I dont mind sorting out my own ****-ups, but sorting out the result of a "professional' build is maddening!

How to fix it? Remove the epoxy, clean again, be sure to let the cleaner dry. then re-epoxy. Fiberglass tape isn't going to help.

As far as setting it on fire, most of us have been there. Whether the problem was bubbling paint, the IO-400 engine recall, smoked avionics, or some other unexpected gut punch. Unfortunately, there are likely to be others.

[mike newall]

Go to Hobby Lobby - look in the bracelet/trinket/bangle section and look for tiny, tiny discs.

They are for either bracelets or rings. Made out of copper but plated.

I bought a bag of tiny ones for a dollar or so. Trim the ring tag off, flat it down with 320 paper and put a weeny blob of epoxy on and stick over the offending rivet.

When set, wet and dry over to remove edges and then prime and touch up.

Move on, fly on :D:D

[scsmith]

Interesting theory, but when testing the theory we need to consider that the vent design originates all the way back to the RV-4. There are 1000's of RV's with this design, many of them completed in the 80's and 90's with (as far as I am aware) no history of paint blisters.
It has only been within the past 10 years or so that they have been occurring.
My theory is that it is at least in part related to the EPA driven formulation changes that have occurred in paint systems over that time, but that is just a gut feeling.... I don't have any more evidence to back up that theory than any of the others that have been suggested up to this point.

This theory doesn't explain blisters under vinyl wrap.

Might be changes in formulation of the ProSeal though.

[rocketbob]

This theory doesn't explain blisters under vinyl wrap.

Might be changes in formulation of the ProSeal though.

I've been saying for years its the in-fuselage loop in the vent lines that causes paint blisters. Slug of fuel in the loop puts enough pressure in the tanks to cause outgassing via imperfections in the sealant around rivets.

Many certified aircraft that are non-strutted have a vent line that extends to the tips via a Dukes flapper check valve with a small hole drilled in the body of the valve. This does two things: 1. allows for venting even when parked on a sloped surface and 2. Drains the vent line back into the tank to prevent the slug of fuel from increasing pressure in the tank.

I've never seen blisters on wet-wing airplanes, other than RVs.

[Mel]

I've been saying for years its the in-fuselage loop in the vent lines that causes paint blisters. Slug of fuel in the loop puts enough pressure in the tanks to cause outgassing via imperfections in the sealant around rivets.


I can't quite buy that theory since, as Scott says, there are literally thousands of RVs built between the late '70s and the mid 2010s per the plans without a single blister.

[rocketbob]

I can't quite buy that theory since, as Scott says, there are literally thousands of RVs built between the late '70s and the mid 2010s per the plans without a single blister.

Variations in proseal thickness can explain it, as well as the fact that blisters don't occur on the bottom side.

[FORANE]

blisters don't occur on the bottom side.

They did on the RV-9A I owned; top and bottom had blisters.

[rocketbob]

They did on the RV-9A I owned; top and bottom had blisters.

Mine didn't as it was rare that I parked it for an extended period without filling the tanks first.

[BillL]

https://www.dropbox.com/s/3e3khte17hq68ld/IMG_0991.jpeg?dl=0
https://www.dropbox.com/s/15ytz90ramxmz9u/IMG_0992.jpeg?dl=0
Abject despair today... then I found this thread..

2018 QB wings, cleaned with detergent degreaser, wiped down with MEK, then two-pack acid etch primer (P30B) left for 24 hrs then two pack epoxy primer (PR143).
Pressure testing of tanks was fine.
I had 20 litres avgas with fluorescent marker in the tanks for a week with no leaks, then the tanks drained fully and left open for 24 hours, and then the wings stored vertically (leading edge down) with the tanks fully sealed to prevent critters getting in.
Today, I moved the wings and one is fine, the other has 10 blisters, 9 on the lower aft flange and one on a rivet one place closer to the leading edge.
No top coat, and minimal fuel exposure yet Ives still got blisters!
I sanded them all down and obviously no fuel present under them.
Irritating beyond belief, but at least ive found them before spending $10k on a paint job.
So how do I fix the problem? epoxy over the rivets, fibreglass tape and filler?
I feel like setting fire to the bloody thing at the moment! I dont mind sorting out my own ****-ups, but sorting out the result of a "professional' build is maddening!
Darren, if it were me, after calming down a bit . . . I test the primer adhesion. Take a piece of strong adhesive tape, apply to the painted surface and slowly peel it off. Test the tank in several places for adhesion of the primer. The prep did not mention abrading the surface before the process. ;) Do the adhesion test over the rivets on the tank and some other non-tank rivets too. This is just to eliminate some possibilities contributing to the blisters, some that one might be able to control. If everywhere but the blister areas pull no paint, then your process is sound and the issue is focused on the pro sealed rivets.

[DanH]

Today, I moved the wings and one is fine, the other has 10 blisters, 9 on the lower aft flange and one on a rivet one place closer to the leading edge.

Was the surface with blisters exposed to direct sunshine, or a radiant heater, or a similar heat source?

[ColoCardinal]

I can't quite buy that theory since, as Scott says, there are literally thousands of RVs built between the late '70s and the mid 2010s per the plans without a single blister.
Is it possible that only planes painted with catylized paints or "wrapped" are at risk for this? Both are less permiable than the older enamel paints of the past.
As a 25+ year owner of a Cessna Cardinal, I will second Bob's observation that, I too have never seen one with rivet blisters. Could it be that they use a different sealant in the tank?
Very puzzling, indeed!

[Mconner7]

My RV-10 was commercially painted and has a few bubbles on the top tank rivets within a few months. It was built in 2011 and first painted in 2017. It is a slow build plane.

[MiserBird]

Is it possible that only planes painted with catylized paints or "wrapped" are at risk for this? Both are less permiable than the older enamel paints of the past.
As a 25+ year owner of a Cessna Cardinal, I will second Bob's observation that, I too have never seen one with rivet blisters. Could it be that they use a different sealant in the tank?
Very puzzling, indeed!

Back in '84 I purchased a '68 Cardinal which had recently been over painted with DuPont Imron. It was fine until I started running MoGas (pre-ethanol mandate) blisters galore soon appeared on the top surface of the wing tanks.

Moved on to a '75 Cardinal RG in '93 with good original paint, only running AvGas, not one blister. I had it stripped, primed, and repainted in '10 with JetGlo. No blisters to date.

I purchased my '80 RV-3 about 18 months ago and have only run AvGas since I've owned it; no blisters. One interesting note; my RV was owned by Sun 'n Fun for about 15 years and I'm sure the tanks set dry during this period. I think I'll stick with AvGas based just on my personal experience.

[DanH]

I think I'll stick with AvGas based just on my personal experience.

The key difference, as you probably know, is vapor pressure.

[RVDan]

I built a RV-6 and sealed the tanks with PRC 1422, using the B class for the fay seal, and fillet seal covering the tails with this material. In addition, I overcoated with the A class as a brush application and used this grade for wet setting the rivets. The paint is Imron. No bubbles, no leaks or mushy sealant after 26 years.

The airplane I own now was done with the Chemseal product CS3204, which is qualified to Mil-S-8802 Type II. I did not do this airplane myself, but am sure of the sealant used. After 10 years it has bubbles, and the sealant around the fuel sender plates is mushy and seeping. There is a significant difference in the 2 sealants which was introduced in or around the early 90's. At that time the Mil spec was revised to delete dichromated sealants. PRC 1422 used to be an Mil-S-8802 Type II qualified sealant but the change to the standard caused it to be dropped as a qualified material. PRC had PRC 1440 to take the place of PRC 1422 with respect to the Mil qualification. I had documented this in an article in Sport Aviation back in the late 90's. I am not sure how the Proseal line migrated, and was unfamiliar with Chemseal back then. There were tons of Cherokee tanks sealed in the aftermarket with PRC 1422 and I don't recall blistering as a problem. I am wondering if the change to the sealants has resulted in the current crop of problems. If anyone else has used PRC 1422 instead of either PRC 1440, PS 890 or Chemseal 3204 and has bubbles speak up.

Courtaulds Aerospace (PRC) PRC 1422 - dichromated no longer Mil qualified. PRC 1440and PS890 are Mil-S-8802 Type II qualified- dichromates replaced with maganese.

CHEMSEAL CS-3204 is Mil-S-8802 Type II qualified maganese based.

The products have clear differences from the application standpoint. If you mix up both they have some differing properties. I am sure there are smarter chemist type here that might recognize what effect the replacement of the dichromates with maganese might do to the sealant with respect to resistance to Avgas.

[BillL]

I built a RV-6 and sealed the tanks with PRC 1422, using the B class for the fay seal, and fillet seal covering the tails with this material. In addition, I overcoated with the A class as a brush application and used this grade for wet setting the rivets. The paint is Imron. No bubbles, no leaks or mushy sealant after 26 years.

The airplane I own now was done with the Chemseal product CS3204, which is qualified to Mil-S-8802 Type II. I did not do this airplane myself, but am sure of the sealant used. After 10 years it has bubbles, and the sealant around the fuel sender plates is mushy and seeping. There is a significant difference in the 2 sealants which was introduced in or around the early 90's. At that time the Mil spec was revised to delete dichromated sealants. PRC 1422 used to be an Mil-S-8802 Type II qualified sealant but the change to the standard caused it to be dropped as a qualified material. PRC had PRC 1440 to take the place of PRC 1422 with respect to the Mil qualification. I had documented this in an article in Sport Aviation back in the late 90's. I am not sure how the Proseal line migrated, and was unfamiliar with Chemseal back then. There were tons of Cherokee tanks sealed in the aftermarket with PRC 1422 and I don't recall blistering as a problem. I am wondering if the change to the sealants has resulted in the current crop of problems. If anyone else has used PRC 1422 instead of either PRC 1440, PS 890 or Chemseal 3204 and has bubbles speak up.

Courtaulds Aerospace (PRC) PRC 1422 - dichromated no longer Mil qualified. PRC 1440and PS890 are Mil-S-8802 Type II qualified- dichromates replaced with maganese.

CHEMSEAL CS-3204 is Mil-S-8802 Type II qualified maganese based.

The products have clear differences from the application standpoint. If you mix up both they have some differing properties. I am sure there are smarter chemist type here that might recognize what effect the replacement of the dichromates with maganese might do to the sealant with respect to resistance to Avgas.

Interesting - I was reading some book preview on polysulfide sealants (Properties and Applications of Elastomeric Polysulfides
By T. C. P. Lee) and it mentioned that the aerospace versions used dichromate but I looked up the SDS for chemseal and it showed manganese in the hardener. The book also talked about reversion in some formulations , but not the dichromate version. (another failure mode we have seen) I am most certainly not a chemist, but did notice the ingredients. Thanks for sharing. I see the PRC-1422 is available at SkyGeek.

[DanH]

Do note that there isn't any evidence of correlation between reversion and blisters, and blistered rivets have been sectioned and inspected closely. I'm inclined to think they are separate issues.

[BillL]

Do note that there isn't any evidence of correlation between reversion and blisters, and blistered rivets have been sectioned and inspected closely. I'm inclined to think they are separate issues.

Right, Dan, those are separate failure modes and did not mean to imply they were linked.

Hey - just saw a post about Vans -14 demonstrator having rivet blisters. Someone take a picture at SnF so the evidence won't get lost!!

I wonder if helium in the tank will show the leakers before paint??

[skiandtom]

My brother and I have a quick build -7A. It is thirteen years old. We have had two blisters appear on one tank and one on the other.