What's new
Van's Air Force

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

Clinching Dimpled Fuel Tank Rivets

I.am.Human.Factors

I'm New Here
To everyone who has said building the fuel tanks are their least favorite activity, you were right.

I’m having trouble with rivets clinching over their dimple as I wet-set them in the fuel tanks. Just this evening every other or so was dumping over as I squeezed them. For what it’s worth, I utilized both Cleaveland’s fuel tank and substructure dies. Unriveted dimples appear about as symmetrical as can be seen with the naked eye. I also haven’t seem to have had this problem this frequently through the build, but it’s popping up now.

Anyone battled clinched rivets in the fuel tanks? I’d like to not drill them out, but I don’t see a way around it now. Any tips for getting these to squeeze properly? Does the sealant causing things to slide around as they set cause these rivets to frequently set poorly?

A quick sanity check: 1. Yes they are the proper length. I checked. 2. No, I can’t just stop using the substructure dies. The entire tank is already dimpled and faysealed. :cool:

I’m thinking maybe a handheld squeezer might be the answer to really go slow while setting these? As for the rivets to be shot, well, I’m hoping someone will say, “a clinched rivet every so often isn’t so bad, so move on.” Any takers?

Any ideas, help, words of encouragement, etc., are greatly appreciated...
 

Attachments

  • 47653B77-1675-451A-95FD-C3C77C04CDEA.jpg
    47653B77-1675-451A-95FD-C3C77C04CDEA.jpg
    289.9 KB · Views: 403
  • 531582E4-AA03-4D10-996A-131ED752CDF2.jpg
    531582E4-AA03-4D10-996A-131ED752CDF2.jpg
    307.4 KB · Views: 443
I had this problem. The way I was able to stop it was to wipe clean the bucking bar after every rivet and the shop end of the rivet before shooting it. I think the issue is reduced friction between the face of the bucking bar and the rivet head due to the sealant.

I cut up a bunch of paper towels into 2" squares and used them to clean the bar and the rivets, using a clean piece of towel for each one.
 
Bar orientation

My experience is the orientation of the bar is what causes the rivet to clinch. If it's even a little off perpendicular to the shank, the rivet will clinch. I made a habit of taking a look or using the skin as a blind reference. If I can't see the rivet, I set the face near the rivet and muscle memorize the position then move it over. I also used little paper towel squares to keep things dry.

I actually enjoyed tanks but I had a broken leg and it was about a I could do sitting all day.
 
Check your drill.. I had a #41 that had developed a wobble, abs was making #37 size holes. . After dimpling, they were even bigger. I fixed it by pre squeezing some -3 rivets to fatten them up. I also threw out that bit!
 
You say you were squeezing them and I assume from your comment about trying a manual squeezer you are using a pneumatic squeezer.

I seem to remember having to wipe both ends of the rivet to clean them of the tank sealer before squeezing them. Used MEK and a Q-tip. Make really sure your squeezer has a nice teasing trigger, sneak up on the rivet, make sure the dies are square to the rivet ends before squeezing. You might also try "pre-squeezing" them just a little bit so they still fit through the dimpled hole.
 
I sanded a popsicle stick end to a 100 degree V, then used the tool to remove proseal from each dimple before inserting the rivet. Also paper towels to keep the skin clean on the flush side. A pneumatic squeezer is guaranteed to give you good a good shop head if you keep the rivet set flush on the manufactured head- look for an equal amount of light around the sides to indicate the rivet set is flush. You'll need plenty of good light to do this. Also important to keep the rivet in the middle of the set, so made a duct tape "stop" to keep the set from sliding down off the rivet as the trigger is squeezed. The "stop" is enough layers of duct tape built up to 1/16-1/8" thick- just rest the rivet set against the side of the tape so you are self aligned. Also important to push firmly on the flush side as you squeeze the trigger- this will prevent "proud" heads. Pull the trigger slowly, backing off if anything seems out of wack. I'd drill out the clinched rivets- not worth the risk of future leaks. Practice on some sample aluminum before you do any more squeezing- always practice any new operation to gain confidence. Good luck, we all make mistakes building these things, it's just part of the job description! :)
 
Last edited:
Another trick

I often use a 1/8" or 5/32" universal squeeze set for the shop head of the squeezer instead of a flat set, which will help keep it centered, and make a beautiful looking bucktail. This trick only works if you have good clearance from the vertical flange. If the rivet line is too close to the radius, the set will not center itself.
 
My experience is the orientation of the bar is what causes the rivet to clinch. If it's even a little off perpendicular to the shank, the rivet will clinch. I made a habit of taking a look or using the skin as a blind reference. If I can't see the rivet, I set the face near the rivet and muscle memorize the position then move it over. I also used little paper towel squares to keep things dry.

I actually enjoyed tanks but I had a broken leg and it was about a I could do sitting all day.

I agree with this; I think the wet rivet and bucking bar made that harder to do. Usually I position the bucking bar with the edge flesh of a finger between the bar and rib on each side of the bar. This helps me align the bar correctly. When shooting the wet tank rivets, I didn't do this as it was making a mess. I did notice that they weren't even just normal clinches when it happened; they often sheared off entirely. It was weird. All that said, by carfully cleaning the bar and rivet before shooting, I was able to prevent clinching problem mostly.

I used the "fay seal and allow to set up before riveting" method, so it was less wet than the full wet method, but I applied sealant to the dimple before installing the rivet, so that's where the wet part came from.
 
For what it’s worth, I utilized both Cleaveland’s fuel tank and substructure dies.

Look at the size of the holes, the natural side effect of oversize dimples. If you can't swell the rivet shank to fill those holes tightly, the rivet is substandard.

I’d like to not drill them out, but I don’t see a way around it now.

Not much choice.

Does the sealant causing things to slide around as they set cause these rivets to frequently set poorly?

Sure. As others note, if wet riveting, remove sealant from the head, tail, and tools.

Here the photos suggest you've fay sealed while assembling with clecos, and then allowed cure time, which is fine. If so, you're applying sealant to the individual rivets just before setting them? It's a natural consequence of the belief that sealant space is required under the head of the rivet, the rationale for the oversize dimples created by "tank dies". In reality, extra space is not desired; per the sealant manufacturer, the finished thickness of sealant under a rivet head should be a film less than 0.001". Excess space filled with sealant merely floats the rivet head on a sealant sponge, in an oversize hole.

2. No, I can’t just stop using the substructure dies. The entire tank is already dimpled and faysealed. :cool:

Order new parts.

Any ideas, help, words of encouragement, etc., are greatly appreciated...

All the best airplanes (1) have rejected parts under a bench somewhere, and (2) builders who learned a lot in the process of creating them.
 
Order new parts.



All the best airplanes (1) have rejected parts under a bench somewhere, and (2) builders who learned a lot in the process of creating them.

Lol I hope that is sarcasm. No way would I order new parts for a minor mistake like this... Especially if he stopped riveting early on before doing all of the ribs.
 
Lol I hope that is sarcasm.

Not sarcasm.

If Tucker can't swell the shanks to fill the oversize holes, the assembly is structurally substandard, and prone to leakage and paint blisters. If he can, great, but it will require a longer rivet, and he is already having problem with the standard length.
 
Not sarcasm.

If Tucker can't swell the shanks to fill the oversize holes, the assembly is structurally substandard, and prone to leakage and paint blisters. If he can, great, but it will require a longer rivet, and he is already having problem with the standard length.

Been there, done that, got the Tee shirt. I found that best solution was to use the next size longer rivets. Swell them slightly, using a hand squeezer, before installing the rivets in the structure. In other words, swell the rivets while they are simply mounted on the squeezer, but not in the structure. That way, you simply throw away any rivets that don't swell properly. Swell as needed to get a snug fit in the oversized holes. Install the swelled rivets into the holes and set as normal. Without the excessive rivet to hole clearance, your results will be much better.
My issue was not on the tanks, but on the curved portion of the turtle-deck.

Charlie K
 
Last edited:
Update:

Thanks for all the helpful hints, tips, and encouragement everyone! You all had some excellent ideas! While there was a little doom and gloom among this thread, I tried the simple act of wiping the excess proseal off before setting and that did the trick nicely. Some good rivets were resulting so I’m much happier now.

I’ll definitely be drilling a number of of rivets out that I had previously set, but I wasn’t overconfident and set all of the ribs before posting this message. I set a rib and a half and stopped to think about it. I learned the lesson to slow down long ago. :D

Thanks again everyone!
 
I found the answer ! For me.....

Use the regular yoke not the longeron yoke. Folding over be gone ! Didn’t even have to clean them anymore
 
Tucker, forget doom and gloom. There is risk in allowing emotional content into what should always be a dispassionate build decision. Dispassionate is the key. If you're rationally sure of what you've done, great. If you're not, replace it. Either way, don't let the decision be swayed by feelings, yours or others.

Exactly half the finished RV's are below average.

Long term, no one has ever regretted replacing a marginal part.
 
One of the issues is using B proseal to Fay seal ( also letting sealant cure before riveting). We never used B to do this, B is for filleting and A is the proper sealant for Fay sealing. We drove hundreds of thousands of wet rivets and never cleaned tooling or rivets until they became so gummed up it made more mess than progress. This is a non issue when using the proper sealant, I have no idea why Vans calls for B in this setting other than it is more forgiving in application than A.

I agree with the others, drill and/or replace parts, leaks suck. You’ll get there and be flying soon!
 
Glad you got the problem figured out. Those dies from Cleaveland have worked for many a builder, no reason it shouldn't work for you too.
 
Not giving advice here

... just reflecting aloud...

I can't see where those flanges, properly fay-sealed, would allow fuel to reach the dimpled and riveted areas, hence not seeing the leak concern. But I know fuel tanks, leaks and paint blisters are fickle and mysterious things and that meticulous technique is never a bad thing.

2. Once cured, I wonder how much the rivets contribute to the mechanical strength of a ProSeal-bonded joint. I really do. Especially after trying to get a fuel tank inspection cover off that was ProSealed on years before. From what I experienced that week, I am skeptical that a few rivets, whether clinched over or beautifully set, will enhance that bond strength by much. But I am not an engineer, I'm not DanH, and won't pretend to be.

Carry on.
 
... just reflecting aloud...

Reasonable reflections Bill.

First, understand there are no standards for polysulfide sealant in avgas. All standards are written for jet fuel. The test fluid is actually called JRF, Jet Reference Fluid. It's important to remember, because we don't know how the physical properties of the cured sealant are affected by 100LL, a more volatile compound. For example, no one I know has yet offered an explanation for sealant reversion...sealant turning back into goo.

Second, cured sealant is not a solid. It is full of tiny air bubbles, like a sponge, and thus cannot act like a hydraulic fluid under load.

Third, it is not vapor proof. It emits toluene for a few weeks, and will pass fuel vapor after cure.

Ok, the test standard closest to what we might expect in our application is a sample soaked in JRF for 14 days at 140F. The minimum performance is 50 psi and 100% elongation at failure. The test article from one lab report on my hard drive scored 128 psi and 600%.

Consider those numbers. They are terrific in the context of a rubbery sealant between two parts. However, if we are expecting that sealant to fixate parts as an adhesive, we would be greatly mistaken. Although 128 psi feels like a lot when we're trying to remove a cover plate, it is Jello pudding in the context of structural requirements. For comparison, ordinary West 105/206 is 7320 psi tensile and 11,500 compressive. 2024-T3 is 50,000 tensile yield, 70,000 ultimate.

So, it's not very strong. It's also not very stiff, per the elongation number. At 128 psi, the sample stretched to six times its original length. Clearly that's not possible in a fay joint, as it would have long since sheared all the rivets. It's like pairing a steel cable with a rubber band to pick up a weight. The rubber band sees almost no strain, and contributes nothing useful. The cable takes all the load.

In our tanks, the sealant essentially contributes zip. Load transfer is all via the riveted joint.
 
Last edited:
"Exactly half the finished RV's are below average."

After I got off VAF and returned to my day job as a bean counter I realized....haha he's exactly right!
 
"Exactly half the finished RV's are below average."
After I got off VAF and returned to my day job as a bean counter I realized....haha he's exactly right!

Reminds me of something I used to say to the audience when I had a band.

"If you've had 1/2 as much fun as we've had tonight, Then we've had twice as much fun as you!"
 
Back
Top