SonexBuilders.net

I am finishing up the tail section of a xenos and will be flush riveting. I have used the simple dimple die but it works poorly and I have difficulty getting a deep enough round dimple. Wanted to buy cleaveland tools 120 degree dimple die for N and Q flush rivets but have waited since Oct. I have a c-frame and a drdt2 dimple tool.
Almost decided to just use AN solid rivets. Kerry says I can but they are not as strong? I thought they were stronger! (Sonex discourages flush riveting anyway)
Kerry says it may come out looking like a golf ball.

Bob Avery at Avery tool is looking at some rivets I sent him to see if his dies will work or he will modify it for 120 Sonex flush rivets. The standard 120 I bought from Cleaveland was not deep enough for the Sonex supplied rivets.

Want a smooth, flat skin and don't mind spending the time to do it perfect. Thats why I was going to go to AN rivets but if pulled is stronger and I can get the good finish it is easier.
Suggestions, tips, ideas?

Thanks
Lee

Lee,
This email I exchanged with Bob Avery may help:

About 2 years ago AVEX changed some dimensions on their rivets, making the
head bigger. We changed the 120 degree dimple dies from the old dimensions
(smaller / thinner rivet head) to the newer larger / thicker rivet head at
that time. You will need the newer version dimple dies for the new style
rivets. If for some reason that does not work, I can get some of the
rivets from you and machine a set of dies to fit your rivets.

Bob Avery

-----Original Message-----
From: Bryan Cotton [mailto:bryancotton@yahoo.com]
Sent: Thursday, October 03, 2013 12:01 PM
To: boba@averytools.com
Subject: 120 dimple die question

Bob,
Years ago I bought an Avery 120 dimple die for avex rivets. My Sonex uses
some CCC-42 and other size flush pull rivets. The heads seem bigger than the
avex rivets. When I pull them they are not quite flush. Do you have a
version for these rivets?

Bryan

Also, I bought/tried Bob's new dies and I thought they were good.
If I was going to flush rivet my Waiex with solid rivets, I would use AN426-3's and double up. Hard to set -4 in skins nicely.

Wish I would have read this thread yesterday or before. A few months ago, a had the same conversation with a guy from Cleveland tools. He had said he would try to make a custom dimple die for the new flush rivets sonex is supplying. Unfortunately nothing has materialized.

I own the Cleveland tools N/Q die, and as another has said, it isn't deep enough.

For this reason, in addition to me botching a forward wind skin, my building had been stalled for about 3 months. Yes, I've been discouraged over the winter.

Anyway, I found out that one of my coworkers does precision machining, so I talked to him about the situation. I showed him the die I had, and he said that if I supplied him with some 1/2" diameter drill rod, he'd make me one this weekend. I told him it only needed to be 1/128" deeper than the cleveland tools one...but maybe I 1/64" would have been better, as I could have used it as a sub-structure dimple die. 3 feet of drill rod was only $10, so I won't be out much if my coworker doesn't get it right.

I'll try the Avery die next if I have to.

I sure don't understand why Sonex LLC isn't providing a solution for this. I've shown samples of the old flush rivets and the new flush rivets riveted in samples of aluminum, and it's obvious to everyone that the new rivets aren't flush.

Thanks to everyone for your input, it seems I am not the only one with this problem. Bob at Avery will have some of my rivets in a day or two and we'll see how that works out.

Does anyone know for certain if the ccc42 rivets are stronger or weaker than an426. The specs on AN rivets show 16000 lbs tensile 26000 lbs shear strength. The chart in the spruce catalog shows 600 tensile 700 shear on the ccc4 pulled Q rivets N rivets even less but is that what sonex is suppling? Which is really stronger?

Its to late for me to do my tail section with -3 solid rivets as they are drilled out. I will be back riveting anyplace I can such as fuselage sides to get the best result. (no golf ball like dimple around the rivet) Making sure the skin is flat on the back riveting plate.

By the way isn't great that we have such great suppliers who will go the extra mile to help us out. I've had great experience with Mike at Cleaveland but they are having CNC machine issues. Bob Avery is going the extra mile. If anyone needs a DRDT2 dimple tool Paul is very good and helpful there.

We are very lucky to be part of the flying/building community. So many good and kind people.
Thanks Lee,

The strength for the AN rivet you list is per square inch (psi) of the area of a cross section of the rivet shank.

But you have a mix up. The 16,000 psi tensile strength is for a soft aluminum rivet (AN426A-# = 1100 alloy). The 26,000 psi is shear strength for a hard aluminum rivet (AN426AD-# = 2117-T4 alloy). The same rivet tensile strength is 38,000 psi.

For an AN426AD-4 rivet (1/8" diameter hard aluminum) that would be about 320 lbs shear and 450 lbs tensile.

The Sonex rivets are Cherry N equivalents at 450 lbs shear and 600 lbs tensile for a CCC-42

All the bucked rivets on my RV-14 are AN426AD-# or AN470AD-#. I assume the soft aluminum bucked rivets are for non-structural use only or lightly loaded structure. Be careful what you order. (And double check my math! Don't take my word for it) ;-)

Thanks Bob,
My intent was to use AD rivets so thats not an issue. I did the math by deviding tensile and shear strength by 64 which is 1/8 times 8 for a square inch and came up with 593.75 tensile and 406.25 shear. Very close to the cherry N rivet so I wouldn't feel like I was making a weaker airplane. (if my math is right? used to be good at that stuff but that was 45 years ago) 26000 psi devided by 64 = 406 and 38000 devided by 64 = 593. Mine is a little different than yours, anyone have a different answer?

Now the big question, which will give the best result? Don't want to see little valleys around each rivet head. I can always fill but don't want the weight. Guys like me must be a pain in the butt for Sonex, they try so hard to make it easier and we keep making it harder.
I come from years of being chief inspector for maint. on jets and the little details make a difference. If I was just in a hurry to fly I would not use flush rivets and it would work fine. I'm just a little OCD by inspector necessity. I want to savor the building time and be proud not appologetic of the result. Not a perfect show plane just cleanly done and safe.

Thanks for the input
Lee

Unless you are using square rivets your formula won't work but it is a very good first approximation. You need to find the area of a circle 1/8" in diameter.

Area of circle = pi times radius squared. It takes the cross section of about 81.5 rivets 1/8" in diameter to equal a square inch of aluminum. Use 81.5 instead of 64 and it will give you a closer answer for 1/8" diameter rivets.

leeforshee wrote:Does anyone know for certain if the ccc42 rivets are stronger or weaker than an426

(if my math is right? used to be good at that stuff but that was 45 years ago) 26000 psi devided by 64 = 406 and 38000 devided by 64 = 593. Mine is a little different than yours, anyone have a different answer?

AD rivets are made with 2117 aluminum alloy (as described at http://en.wikipedia.org/wiki/Rivet), which has a yield strength of 24-26,000 and a tensile of 38-43,000 psi as per material property references. The 1/8" AD rivet has a cross section area of π * ((1/8)/2)^² ==> 0.0123 sq in ==> 319 yield and 466 tensile. This calculation matches online references stating 300-350# shear and 450-470# tensile. This compares to 450 yield and 600 tensile for CCC-42 stainless rivets.

So, although there are umpteen other considerations when choosing solid vs. pop rivets, Kerry is correct, the SS rivets are generally stronger than AD rivets for the same diameter. That advantage comes at a price, however: my last batch of CCC rivets weighed an average of 0.41 grams per (pulled,installed) rivet vs. 0.19 gr for an AN426 of the same grip length. Also, as of May 2014 the current price for a CCC-42 rivet is 11 cents each vs. less than 1 cent per AN426.

EDIT: I see Bob posted while I was still writing this reply. What he said.

I have used both on my Onex. The reason is that when dimpling a #30 hole, the hole diameter ends up borderlining the rivet specs. I didn’t know that when drilling my horizontal stab. What happens is that when pulling the rivet, the stem is pulled all the way and often protrudes, and consequently the rivet is not compressed and widened according to spec. I tested the strength, and it certainly is strong, even with dimpled #30 holes, so I think it is more of a cosmetic problem than a structural. An AN rivet does not have this problem, it will fill up whatever space is there (within spec’ed limits of course). A #30 dimpled hole is certainly way inside those specs, but it is larger than 1/8, so the (bucked) rivet strength is also larger than those calculated for 1/8 diameter. RVs use 3/32 rivets on the skins (the -4 at least), and with the same rivet spacing as Sonexes, so I certainly don’t think strength of the rivet joints is a problem with the Sonex aircraft.

What I found was that using a #31 drill, ease on the deburring, and the dimpled hole becomes perfect with no protruding mandrel. This required a minor fix of the dimple dye, since it is #30 size originally. I use standard dimples for squeezers and C-frames. I only use pull dimpler when I cannot access the holes with the squeezer or the C frame. I have seen no problems with the dimple dies, maybe I have "new" ones? Avery tools I think.

The AN flush rivets looks much nicer, no comparison really. Along the hinge line I could use a squeezer, which is also much faster than pulled rivets.

Also use 100 degree dimpler on AN and 120 on pulled SS. It may not seem like a big deal those 20 degrees, but it really is, I have tried. The whole idea of flush rivets is to get it tight and flush.

Now I use #31 drill, the modified dimple die and SS pulled rivets. The result is rather good I think, perfectly flush.