What's new
Van's Air Force

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

Chipped Rivet Heads

J

jwilbur

Well Known Member
This is an odd (and unfortunate) issue. I just finished my tanks and they're leak-free. Great. While cleaning off all the soapy water, I noticed something. Two rivet heads were chipped.

IMG-20120810-00261-400x300.jpg


If I had noticed this before the tanks were closed up, I would have replaced them without hesitation. But the tanks are closed (and did I mention leak-free?) So I then looked at every rivet in both tanks and found 7 others. The following pictures show the position of each chipped rivet head that I found.

LeftWingRivetCracks.jpg


RightWingRivetCracks.jpg


All but one is on the top and most interesting I think is the location of the left-most chipped rivet on the right tank. This rivet is in the root-rib and was squeezed. I can almost believe my bucking technique could be so bad as to cause these chips. But how can they happen on a squeezed rivet (hand squeezer)?

I really, really want to avoid putting a bunch of holes in the baffles unless absolutely necessary - I also want to avoid foreign matter from drilled out rivets tumbling around in the tank looking for a fuel-line to plug up. I'm awaiting a response from Van's on the issue. But some secondary questions are what would cause a rivet to chip in this way? Can it be a bad batch of rivets? Should I be concerned about other rivets that haven't chipped (yet)?

Has anyone else seen something like this?
 
The heads look a little funny in your pictures.

Normally you can still see the dimple in the center of the rivet head even after riveting.

Could it possibly be a round head rivet was forcibly riveted into a dimpled hole?
 
I don't have an authoritative opinion on the issue. However, I do have a concern on why this is happening on so many rivets? Are you experiencing the same in other locations?

My concern is what will happen when the rivet begins to fail? I suspect the chipped head will crack given enough time. The question is how long is that period and how much are you willing to gamble? The aircraft certainly won't fail out of the sky, but a fuel leak (perhaps a blister) or a smoking rivet will ruin a paint job.
 
I'm not an expert but thought I'd mention that I had some trouble with rivets cracking both with the squeezer and rivet gun. Mine did not look like yours which actually seem to have a piece missing. In my case it was due to using some old rivets I had been given.
 
az_gila said:
Could it possibly be a round head rivet was forcibly riveted into a dimpled hole?
Click to expand...

I just double-checked. The rivets are correct. It's just really hard to see the little dimple in the picture.

Lionclaw said:
My guess - The manufactured heads were not flush with the skin while squeezing.
Click to expand...

Only one of the cracked rivets were squeezed and I use a hand-squeezer so I don't think this explains it - at least not fully. If it weren't flush when I started to squeeze it would be before any real pressure was applied.
 
Last edited:
rleffler said:
I don't have an authoritative opinion on the issue. However, I do have a concern on why this is happening on so many rivets? Are you experiencing the same in other locations?.
Click to expand...

Never seen it before.

rleffler said:
My concern is what will happen when the rivet begins to fail? I suspect the chipped head will crack given enough time. The question is how long is that period and how much are you willing to gamble? The aircraft certainly won't fail out of the sky, but a fuel leak (perhaps a blister) or a smoking rivet will ruin a paint job.
Click to expand...


Completely agreed. I'm waiting for some advice from Van's. I really want some magic way to repair these rivets properly without tearing apart the baffle.
 
jimgreen said:
I'm not an expert but thought I'd mention that I had some trouble with rivets cracking both with the squeezer and rivet gun. Mine did not look like yours which actually seem to have a piece missing. In my case it was due to using some old rivets I had been given.
Click to expand...

This ocurred to me - not that my rivets are old, I got them with the kit from Van's and have used them without any issues on every section up until now. The big difference this time is that I soaked these rivets in MEK before dipping in proseal and installing. In some cases, they may have been submerged in a small container of MEK for weeks before being used. Other rivets were in the container for minutes before use. Can MEK cause rivets to harden prematurely? ... I'm kind of grasping at straws here. My immediate concern is how to address what I have with the minimum pain, to understand if every rivet in my tanks are compromised in some way, and to prevent it from happening again.
 
jwilbur said:
Only one of the cracked rivets were squeezed and I use a hand-squeezer so I don't think this explains it - at least not fully. If it weren't flush when I started to squeeze it would be before any real pressure was applied.
Click to expand...

It's possible the same could have happened while driving/bucking. If you apply too much pressure with the bucking bar, the rivet head can be pushed up out of the dimple while driving.

I had this happen a few times. The manufactured head ends up slightly concave, the little dimple in the head disappears (or almost disappears), and you end up with a silver-colored spot on the rivet head where the coating on the rivet wears away.

I would always curse and then drill them out. Not sure how I'd handle it on a finished tank though...

Edit: Forgot one thing. I would have to disagree with "If it weren't flush when I started to squeeze it would be before any real pressure was applied". If the rivet shank expands in the hole while the rivet is not flush, applying more pressure won't necessarily make it flush.
 
Last edited:
Move on

I think Lionclaw provides the most likely scenario, that of not having the rivet flush before rivetin/bucking starts.
The question is, why are your rivets not flush to begin with and the only reason is too much pro seal under the rivet.
Like Lionclaw stated if the shank of the rivet expands on both sides of the skin,
you'll be left with trying to expand both sides of the rivet even the flush side which is not designed to expand.
As to the fix, considering that you built a tank that does not leak I would not
necessarily spend too much time worrying about it. I am quite sure Van will tell you to "move on".
A half dozen split rivets are not going to compromise the structure of the fuel tank and especially glued in with pro seal are not going to start rotating and "smoking". Another dab of pro seal on top, wipe flush and leave it alone.
Congratulations on finishing a set of fuel tanks.
 
If the tanks don't leak, leave them alone.

There's an excerpt in the build manual about even poorly set rivets being within 5% of the strongest correctly set ones, so my two bob is: ignore it and move on.
 
Lionclaw said:
It's possible the same could have happened while driving/bucking. If you apply too much pressure with the bucking bar, the rivet head can be pushed up out of the dimple while driving.

I had this happen a few times. The manufactured head ends up slightly concave, the little dimple in the head disappears (or almost disappears), and you end up with a silver-colored spot on the rivet head where the coating on the rivet wears away.
Click to expand...

I have no doubt this is what happened on the two I have pictured in the original post. It's happened to me in other places and I've always just replaced them. But still, it's worrying me that the rivets chipped. I've never seen that happen before. Do I have a bad batch of rivets? And if so, do I have more to be concerned about?


Lionclaw said:
Edit: Forgot one thing. I would have to disagree with "If it weren't flush when I started to squeeze it would be before any real pressure was applied". If the rivet shank expands in the hole while the rivet is not flush, applying more pressure won't necessarily make it flush.
Click to expand...

You are right! I just checked the squeezed rivet. It stands just a little proud. Not much but clearly it isn't perfectly flush. But still, why would it chip?

This picture isn't the greatest, but it is of the one squeezed rivet that chipped.
squeezedRivetWithChip.jpg
 
KRviator said:
There's an excerpt in the build manual about even poorly set rivets being within 5% of the strongest correctly set ones.
Click to expand...

Correct me if I'm wrong, but this is referring to rivets that are underset or overset or slanted shop heads, etc. I'm not sure cracks or chips are included in that statement.
 
Are you sure they're actually chipped? It looks more to me like the manufactured head got pummeled, causing its perimeter to expand outward and split from the stretch.
 
Relax

Joe,
Relax, those rivets are cracked not chipped.
I would be very surprised if Van told you something other than to "move on".
I am sure you are a meticulous builder who has difficulty accepting a minor
flaw. It really is a minor flaw and not worthy of conducting an investigation into
old rivets or other perceived material defects.
Let us know what Van's tech advisers have to say about it.
 
There is an exact spec. to see if your rivets are OK.

http://mybearhawk.com/MIL-R-47196A_MI.pdf

Section 4.2.1 J

It's my original HTML version of the Mil Spec, apparently on the Bearhawk web site.....:)

Interestingly the military does not allow any defects in the manufactured head of a flush rivet, but says some cracking/chipping is OK in the manufactured head of a standard universal rivet.

Get Vans opinion.....
 
Last edited:
jwilbur said:
- not that my rivets are old, I got them with the kit from Van's and have used them without any issues on every section up until now....... The big difference this time is that I soaked these rivets in MEK before dipping in proseal and installing. Can MEK cause rivets to harden prematurely? ...
Click to expand...

MEK, no.

What kind of dimple dies did you use?
 
szicree said:
Are you sure they're actually chipped? It looks more to me like the manufactured head got pummeled, causing its perimeter to expand outward and split from the stretch.
Click to expand...

I hadn't considered this. It makes sense this might be the case.

DanH said:
What kind of dimple dies did you use?
Click to expand...

I used the tank dies for everything except the baffle. By the way, I dimpled for the baffles rather than countersink (as stated in the plans) and had absolutely no trouble at all. The alignment was trivial and I have no leaks.

az_gila said:
There is an exact spec. to see if your rivets are OK.

http://mybearhawk.com/MIL-R-47196A_MI.pdf

Interestingly the military does not allow any defects in the manufactured head of a flush rivet
Click to expand...

Thanks Gil. This was the first place I looked. In fact, I think you emailed the document to me after we met at your hanger (maybe 18 months ago). It is this detail that brings me the most angst.
 
For some reason I keep thinking that is an an470 rivet. I don't want to sound like a smarta$$ but it looks like the contact pressure discolored the middle and not the perimeter of the head. The countersink on the skin pushing up on the underside of the (not countersunk) rivet head. Are you sure it is not an an470-3 rivet that got mixed in the box?
 
jwilbur said:
I used the tank dies for everything except the baffle.
Click to expand...

Right. That would be the other thing you did different compared to the same rivet batch on the rest of the airframe.

Take a look at the rivet specs Gil linked. The maximum hole size for a 3/32 rivet is 0.103". Do an experiment....drill and tank dimple some 0.032 2024T3 scrap just like you did those tanks, then measure the resulting hole in the bottom of that dimple.
 
DanH said:
Right. That would be the other thing you did different compared to the same rivet batch on the rest of the airframe.
Click to expand...

This is a good point about the tank dimple dies being another difference. I've been following the recent tank dimple die discussion, but I can make no intelligent contribution to that debate. I think if the cause of these chips (or possibly cracks that split) is that too much proseal resulted in driving rivets before they were flush, then using the standard dies wouldn't have prevented this issue.
 
jwilbur said:
Correct me if I'm wrong, but this is referring to rivets that are underset or overset or slanted shop heads, etc. I'm not sure cracks or chips are included in that statement.
Click to expand...
It does mention cracked heads, but doesn't specify if they're the shop or manufacturered.

Courtesy of Dave Cole, as I'm too lazy to retype it:
Dave Cole said:
It has been determined that even badly cracked heads are satisfactory from the standpoint of static strength, fatigue strength, and resistance to corrosion. (Poorly set and cracked) rivet heads were tested in tension to determine how well formed a head has to be in order to develop full strength. The tensile strength of all the rivets were within five percent of the strongest. The test indicated that minor deviations from the theoretically desired shape of head are not cause for concern or replacement. The second rivet that is driven in any one hole is likely to be more defective than the first because the hole is enlarged...
Click to expand...
 
jwilbur said:
This picture isn't the greatest, but it is of the one squeezed rivet that chipped.
squeezedRivetWithChip.jpg
Click to expand...

It's really hard to tell looking at that photo, but that *really* looks like a round headed rivet pounded into a countersunk hole... That would certainly stretch the perimeter of the head, and could crack just as shown if stretched far enough.
 
They look like round head (universal) rivets drove with a flat set but, I think flush rivets were pushed up with a bucking bar
 
Last edited:
Snowflake said:
It's really hard to tell looking at that photo, but that *really* looks like a round headed rivet pounded into a countersunk hole... That would certainly stretch the perimeter of the head, and could crack just as shown if stretched far enough.
Click to expand...

The lighting and discoloration may make it look that way to you in this picture, but it isn't. It is most assuredly a flush rivet. Every one of these rivets were picked up with a small needle-nose pliers out of a small tub of MEK, gently tapped dry, dipped carefully in proseal, inserted into the hole, and set (squeezed in this case). I feel that I know each and every one personally. We've grown so close from the time we spent together. .... And that's why this chip or crack is so tragic. ... It's true what they say, you know. You always hurt the ones you love.

And besides, I double checked it last night. It's a flush rivet.
 
Last edited:
Every time I have had any issues with the quality of my riveting it has boiled down to the air pressure on the gun being too high or too low. Too high and you pound the **** out of the rivet. Too low and you end up work hardening the rivet. Once it is dialed in correctly, no issues (except my clumsiness). You might want to check your pressure regulator(s).
 
I have pounded a lot of rivets, and never had a head crack on a flush rivet, ever, and I have pounded my share of bad ones. I would suspect embrittlement of the rivet heads thin edge, losing its elasticity.
I do not know if leaving rivets soaking in MEK for long periods would do anything to them or contributed to this, but I would not recommend that practice.
I would consult Van's and continue to get opinions, but I am in the "if they dont leak, fly them, camp".
If you have some of the MEK soaked rivets still in your possession, you could have a material analysis done on them. It would tell you if the chemical make up of the alloy still met the standard, but not much more, so not sure how helpful it might be.
 
Rivet Gun Pressure Rules of Thumb

To bounce off of Bruce's post....

Using a 3X rivet gun to seat a -3 rivet set the pressure to 30 psi for starters. And, for -4 rivets, you guessed it, set the pressure to 40 psi. Set the flow control on the rivet gun wide open. You want to control the pressue and trim with the flow. Also, keep the pressure regulator within a few feet of your gun to avoid pressure drops found in long hose runs. I do realize there's lots of opinion on technique. This rule of thumb really helped my endeavors.

Chuck
 
Chipped rivets

If i read you correctly you used the fuel tank dies on the skins but not the ribs.

Then i think what you have is a dimple problem.

The fuel tank dies make a slightly deeper dimple. So if the tanks skin is deeper then the ribs the ribs dimple will try and push against the skin dimple the you would mush the rivet for sure. If you use the fuel tank dies you need to use them on both parts to be riveted together.

I just dimpled all my lt tank skins and ribs to match. And will start the sealing process hopefully after young eagles this coming weekend.

Please confirm for us is that what the case was. Skins dimples with fuel tank dies and ribs with standard dies.

Thanks
 
jc11378 said:
If i read you correctly you used the fuel tank dies on the skins but not the ribs.
Click to expand...

Nope. I used regular dies on baffle and where skin attaches to baffle. I used tank dies every place else - both ribs and skins.

Now, why did I use regular dies on the baffle and baffle-related skins rather than tank dies everywhere? Because I can't fit the tank female die on the baffle and rather than grind the die, I decided to use my low-profile regular die. And because I used the regular die on the baffle itself, I had to use the regular die on the skin where it nests together (or else I'd have the problem you described).
 
jwilbur said:
Nope. I used regular dies on baffle and where skin attaches to baffle. I used tank dies every place else - both ribs and skins.

Now, why did I use regular dies on the baffle and baffle-related skins rather than tank dies everywhere? Because I can't fit the tank female die on the baffle and rather than grind the die, I decided to use my low-profile regular die. And because I used the regular die on the baffle itself, I had to use the regular die on the skin where it nests together (or else I'd have the problem you described).
Click to expand...

Probably still wouldn't have mattered. I don't think you can purposefully cause a crack like this to occur, even if you tried. Look to the rivet, not the technique. They should not crack.
 
Jon is right, in that your rivets are a little less ductile than usual.....one factor in the failure rate.

Examine them closely and I'll bet you find the gaps to be splits, not chips.
 
jwilbur said:
I feel that I know each and every one personally. We've grown so close from the time we spent together. .... And that's why this chip or crack is so tragic. ... It's true what they say, you know. You always hurt the ones you love.
Click to expand...
Okay, i'm convinced. :)

I forget whether this has been asked, but how old are the rivets? Is it possible you've got some "aged" rivets in your bin? The older they get the harder they get, maybe that contributed?
 
DanH said:
Jon is right, in that your rivets are a little less ductile than usual.....one factor in the failure rate.

Examine them closely and I'll bet you find the gaps to be splits, not chips.
Click to expand...

But why?

If it's a 100 degree rivet in a 100 degree hole there is no real reason for the outer "ring" of the head to expand that much...

The flat set should be hitting evenly on the rivet head.
 
I assume the rivets were at ambient temperature when driven? The rivets I've driven appear to be pretty malleable. Something sure seemed to make yours less so to result in a split. Can a normal rivet even be overdriven to cause a split?
 
az_gila said:
If it's a 100 degree rivet in a 100 degree hole there is no real reason for the outer "ring" of the head to expand that much...
Click to expand...

You're thinking the dimple constrains the head? Remember, this is an oversize dimple.....tank dies.

Someone drill scrap 0.032" 2024T3 with a clean #40, dimple with a tank die, and gauge the resulting hole diameter.
 
Only a guess?

Could the prolonged exposure to possibly decaying Methyl-Ethyl-Ketone have changed the properties of the aluminum rivets?

I think it would release water and some acids over time.

Just guessing, I am not a materials engineer.
 
DanH said:
You're thinking the dimple constrains the head? Remember, this is an oversize dimple.....tank dies.

Someone drill scrap 0.032" 2024T3 with a clean #40, dimple with a tank die, and gauge the resulting hole diameter.
Click to expand...

Then too much pressure and hitting the head too hard?
 
Snowflake said:
Is it possible you've got some "aged" rivets in your bin? The older they get the harder they get, maybe that contributed?
Click to expand...

These rivets came with the wing kit which was delivered one year ago this month. How long before that they were manufactured, I couldn't say.
 
Response from Van's

I've pasted below the response I got from Van's. Imagine my honor upon hearing that Van himself took the time to look at my pictures. If I had known, I would have shined up the skins a little more.


Joe,

Received both of your E-mails. At this point if the tanks are leak free I would not do anything. I actually had Van and a couple of the Tech guys look at your photos. One question was is what shape is your flush rivet set in? Are there any chips or rough spots?

-- After checking I replied that my flush set is as smooth as when I bought it.

We feel the rivet strength is good even though there are some defects. You could fill the chipped ares with some epoxy product (JB-Weld, etc) to improve the looks and to add some feel-good strength to it. Chapter 5 has a section of riveting where Alcoa comments on these types of situations. You may want to re-read that.

Best Regards,
Joe Blank
​

My thanks to all the VAF members who contributed. More theories, comments, or questions are surely welcome. I still don't know for sure what caused this. But I'm glad to not need to open up the tanks and I'll definitely be more attentive when using these rivets in the future.
 
Last edited:
Stalldog said:
I assume the rivets were at ambient temperature when driven? The rivets I've driven appear to be pretty malleable. Something sure seemed to make yours less so to result in a split. Can a normal rivet even be overdriven to cause a split?
Click to expand...

Hard to know exactly. I live in VA and my shop (garage/car hold) is not heated or cooled and I started the final assembling/prosealing of the tanks back in March (3/2012) and it's now August (8/2012). When the first rivet went in the temp may have been as low as the 40's farenheit. When the last rivets went in the temp could easily have been in the upper 90's. There's no way I can easily correlate the temp at the time the failed rivets were installed. Excellent question, however.
 
Joe, you previously asked if MEK can cause aluminum to harden. Good question. How long did you soak the rivets in MEK? In researching this most of the articles I read indicated that MEK has no affect on aluminum; however, I did come across one study that indicated MEK caused aluminum separation from aluminum oxide on a microscopic scale in a paint sample when soaked in MEK for around 60 hours. What this means, if anything, I don't have a clue, but I wonder if excessive soaking in MEK could cause a change in the exterior layer of the rivet. Once a crack starts, all it has to do is propogate. I'm not a metallurgist nor a chemist, so just wondering here. Sure is a puzzling problem, but it sounds like you're good to go regardless.
 
Stalldog said:
Joe, you previously asked if MEK can cause aluminum to harden. Good question. How long did you soak the rivets in MEK? In researching this most of the articles I read indicated that MEK has no affect on aluminum; however, I did come across one study that indicated MEK caused aluminum separation from aluminum oxide on a microscopic scale in a paint sample when soaked in MEK for around 60 hours. What this means, if anything, I don't have a clue, but I wonder if excessive soaking in MEK could cause a change in the exterior layer of the rivet. Once a crack starts, all it has to do is propogate. I'm not a metallurgist nor a chemist, so just wondering here. Sure is a puzzling problem, but it sounds like you're good to go regardless.
Click to expand...

The rivets are anodized, so it could affect the surface layer differently...
 
Stalldog said:
Joe, you previously asked if MEK can cause aluminum to harden. Good question. How long did you soak the rivets in MEK?.
Click to expand...

I had a little plastic kitchen container filled with MEK. I took a handful of rivets from my bin, put them in the bucket. I then pulled the rivets out one at a time for use. At the end of a session, I put the top on the little bucket and set it aside until next time. The next time (usually the next weekend) I'd open the bucket and continue. When the rivets ran out, refill the little bucket. When the MEK evaporated, put in some more. Some rivets might have been in the bucket for weeks, though the MEK would evaporate probably within a few hours (or days??) because of some small holes in the top of the container.

In hindsight, I should have thought this through and not left rivets in a chemical bath indefinitely. Whether or not this caused a hardening, it now seems like bad form.
 
Last edited:
You must log in or register to reply here.
Back
Top