Convert a universal head rivet-based kit to flush

[jfdoyon]

Hello,

New EAA member and first time builder here. I have just purchased the first kit for my future aircraft, and it is on it's way!

The kit is designed around universal headed blind rivets. I would like to convert this to flush rivets for the skin, but am having some trouble engineering the right solution ...

The big issue is that the kit comes pre-punched with holes to final size. This means that dimpling stretches the hole, too large for the rivets normally sized for that hole at its original pre-dimpling size.

So a 1/8th hole gets stretched to .148 to .150 for example.

I've thought up a few theoretical solutions:

- Upsize the holes to an undersized hole for the next size up rivets, and dimple to obtain a final sized hole of the right size. For example, drill the 1/8th hole to something just under 5/32nds, and dimple to 5/32nds final size. Problem is this would require special dimple dies that do not exist (5/32nd dimple die with undersized pilot), whether in the 120 degree or 100 degree head world. Also, not knowing what every hole in the plane looks like, I could run into issues with the dimple being to close to an edge, edge distances, and so on. The A/C appears to also use 5/32nd rivets, which means going up one size from that, which gets quite big!

- Dimple as one normally would, and use oversized cherry max rivets. Turns out that in some circumstances at least, the post-dimpling hole ends up too big for even such a rivet. With 0.016 skins, the hole ends up about .004 too large. On .024 skins however, the hole is just right and could be reamed to #27 (the right size for the oversize CherryMAX -4). That being said, the tensile and shear requirements are far below what the CM provides, so the slightly larger hole may yet still provide the required strengths, even if reamed to #25 or 24. I am not exactly equipped for such testing however! I've sent an e-mail to Cherry Aerospace, but I'm not holding my breath ... This would by far be my preferred solution however.

Has anyone ever undertaken such a project, or know of someone who has? Any sheet metal experts with some ideas on how one might approach this?

Any help or insight would be much appreciated!

Thanks,
J.F.

[lathropdad]

I think your biggest issue will be the type of rivets you use. You can buy rivets that are nominally -4 or -5 but are over sized.

I use a lot of Avex rivets. These rivets expand very nicely in the grip range to fill the holes. With a dimpled hole, the Avex rivet will expand in a way that compressed the dimple as the rivet is pulled, thus closing the hole.

The Avex rivet is no way a substitute for a Cherry Max. It is a substitute for something like a Cherry Q type rivet. Here is where I get my Avex rivets: http://aircraftproducts.wicksaircraf...s-2/avex-rivet

[jfdoyon]

I looked at those (In fact the entire avdel catalog!), the big problem was finding the 120 degree dies ... The only ones I found are at Cleaveland Tools, and only for 1/8th, and apparently made for Van's rivets (CS4-4?). Since I would also need support for 5/32nd rivets, I'm out of luck I think, because I found none of those anywhere ...

The Avex 1604 -4 requires a .128 to .133 hole ... I'm at .140- to .150 depending on sheet thickness ... also, the minimum grip is .093, where I'm going to need to go lower in at least a few spots ... .073 I think for two .016 sheets dimpled for 1/8th.

That being said they claim great take-up on oversized holes, though they don't say by how much, might be worth testing! But without a 5/32nd die, it's still a non-starter.

I'm tooled up for the 100 degree stuff already, hence the appeal of CherryMAX. I would suspect the cherrymax would also do some pulling of the hole closed, but since that would happen after reaming the wall of the rivet and the hole would no longer be parallel. Might be OK for my needs though.

My big question is how much strength does one loose when using an oversized hole and a CherryMAX oversize rivets. I may be able to put a contraption together! I also wouldn't want the rivet to come loose over time from vibrations/forces/etc. passing the strength test is one thing, surviving years of use is another.

Thanks for the tip though, much appreciated.

[DaleB]

Which kit is it? And the big question is, why do you want to use flush pulled rivets? If it's a pre-punched kit you're kind of swimming upstream. Are you looking for a cosmetic improvement, or drag reduction?

[jfdoyon]

Which kit is it? And the big question is, why do you want to use flush pulled rivets? If it's a pre-punched kit you're kind of swimming upstream. Are you looking for a cosmetic improvement, or drag reduction?

TAF Sling 4, and Both!

[DaleB]

From what I have read, at 120 knots cruise there probably wouldn't be enough drag reduction to make a measurable difference in speed or fuel economy. There's an interesting thread on it over at VAF, and I think there's been actual research done that you might be able to dig up if you want to. So it may not have any real effect other than looking better, and to be honest -- even with flush pulled rivets I don't know how much better looking it would be, they still look pretty bad.

Like I often say, there's "Experimental" and then there's experimental.

[jfdoyon]

Good read, thanks! The factory built an ultra-experimental flush-rivet version of the Sling 2 ... but not of the 4 ... and I think they had some issues, as they really did it on a whim ... I'm trying to see whether I can do it better

They did report some speed increases ...

And of course, if I'm building one it's because i think it looks pretty good ...

[Bill Berson]

Avex is very thin head. Just dimple lightly and try it. The hole shouldn't get very big with lightly dimple. The rivet should fill a slightly oversize hole. The dome head Avex might work better. The dome head will crush flat and sink into the dimple and look like a countersink rivet. Hard to tell the difference. I think dome heads are thinner. Try both.

I would not go oversize. That creates edge distance issues. Best to follow the plans.

[jfdoyon]

To be honest, the default rivets on the TAF 4 are not as "domed" as others I've seen ... more of a button head I think some call them. I saw that after having gotten into my head that I wanted flush So it's not as big a deal as if the rivets were full domed heads (Like on my Piper Warrior), which are much bigger and get in the airflow way more than those on the Sling.

I agree going oversize (as in enlarging the holes one size up that is) wouldn't be a sound approach. I've given up on that one.

If I keep going down this road (unsure, getting more complicated than what I originally thought, and also much more expensive), I think I would have to do a shear test on a CherryMAX CR3242-4-2 in a 1/8th dimpled hole reamed to #25 (It should normally be #27) to see if:

- The rivet is still fully secured in place and the shank does expand enough in the hole to hold everything in place and not risk coming loose. So it'll be squeezed in there, but possibly with less force than normal.
- The shear strength is still what I need it to be (even if it's less than the full strength of such a rivet in ideal conditions). The CherryMAX has more than 4 times the shear strength the original rivet has ... so there's some room there to work with.

So I can give up on some rivet strength, so long as it is still well secured in place and own't come loose.

Not quite equipped with proper shear testing machinery, but I should be able to develop a home test that will work. Cherry publishes the testing procedures and such, which will be handy. But of course getting equipped for the testing will also cost money ... but at this point I'm kind of enjoying working the problem, so I may do it just for fun!

[Bill Berson]

No point in using oversize rivets in .016". The larger the rivet the deeper the dimple. And you need to dimple the rib also. Dome head are more durable.
Actually, 3/32" Avex Dome head rivets would be my choice for .016" skin, but not available.
Cessna wing skin ribs are mostly 3/32" solid rivets.