J

Jasonm

Well Known Member
Hi all,
Before we start riveting the VS and rudder we started some test riveting on our C-frame and hydro-pneumatic hand squeezer and issues quickly arose with the results. All the rivets squeezed seem to "push" or lay over to one side. We cannot get a rivet to squeeze perfectly for anything. We've tried adjusting the air pressure, travel length of the squeezer, different dimple dies, and using different backing plates. Nothing will produce an acceptable result. Take a look at these pictures and help me figure out what's going on. All the rivets are 426 3-3. After match drilling using a #40 bit the hole was .101-.103 but after dimpling the holes opened up to .106-.110! The pictures seem to indicate the dimple hole is too big.

photo5_zps95922ad0.jpg

This one was set in the C-frame by itself and squeezed.

photo9_zps1a8c40dd.jpg

Here's our C-Frame

10894402075_738ccd62e1_b_zpsb64b51ef.jpg

Take a look at the two on the right. They're "pushed" over.

photo12_zps69769a46.jpg

Another one "pushed over"

photo13_zps696789d1.jpg

Last one

Are we being unreasonable or is there something wrong here?

Thanks all.
 
Last edited:
The real experts will weigh in here, but that stand alone rivet did not look straight. If I put one in my hand or pneumatic squeeze, they compress straight.

I would say something is wrong. Something is wobbling or deflecting or not square.

Oh - your pictures don't show. Check in preview before posting and you should be able to see them. Sorry, I don't know how to fix them either.

Good luck in your investigation.
 
That is one fancy C frame, I would check for any measurable deflection or any longitudinally movement while squeezing the rivet. And check for Parallel die surfaces while under load. I built my own frame and bought the head unit, but had to go back and reenforce my frame with more material to minimize load induce deflection.
 
several reasons

Rivets will do that if:
1. pneumatic squeezer is not perfectly vertical to the rivet itself.
2. rivet is too long for the material
3. rivet will not form correctly by itself, needs material to form to
 
To my way of thinking, this is a dimpling tool, and not a riveting tool.

photo9_zps1a8c40dd.jpg


Where did you get the pneumatic setup??? I like it.
 
Thanks Bret. Deflection on the frame is less than 1/16" under full load. We've checked to make sure the head is square and perpendicular to the bottom plate. What makes me think it's the hole diameter after dimpling is that if we hand squeeze a 426 the result is the same.
 
Where did you get the pneumatic setup??? I like it.
Click to expand...

Thanks Mike. The head came from Numatx and my father built the rest.

It does make one heck of a dimplier though!
 
Last edited:
Mike S said:
To my way of thinking, this is a dimpling tool, and not a riveting tool.

photo9_zps1a8c40dd.jpg


Where did you get the pneumatic setup??? I like it.
Click to expand...

Yep, I have tried to use my DRDT a time or two for riveting and they turn out just like that. For a test, try a hand squeezer and see if they don't turn out as they should.

I never bothered to try and figure out the way to make my DRDT smash a good looking rivet.
 
Don't bother trying to use your C-frame for riveting. Just clamp a bucking bar in a bench vise and hold the part in one hand while operating the gun in the other. That's how I did all the doublers on my rear wing spars.
 
rivet problem

I agree. Try setting a few by hand with a manual squeezer and with a 2x or 3x gun on a back plate. They look like something is not square at impact. I try to use a squeezer first then back rivet set then gun. I will also use the manual squeezer if the pneumatic won't cit.
 
Deflection

I'll bet good money your frame is deflecting. The plans for the DRDT-2 show a tube wall thickness of 3/16". I used 1/4 wall but there is still noticeable deflection. My frame isn't even close to being strong enough for riveting. I think limiting frame deflection enough to set rivets would take a much deeper tube section. Shim out the top of your piston assembly 1/32 or 1/16 and see if that makes a difference.
 
"C" Frame

I can't find my old college book for the equations to back me up(some active engineers will chime in I hope), but I remember that structures like this, even when "perfectly" aligned, produce some sort of deflection. When they deflect, they start to twist, once they twist, the force is no longer aligned and they twist more.

Mine works fairly well, but I used the DRDT plans and adjusted for a 3/8" wall thickness of the tubing, PLUS 1/4" stiffener back in the throat of the "C". This resulted in a 108# "C" frame! After connecting the hydraulic ram and base plate, I shimmed the base plate (one piece of card stock was all it took for me). I am able to dimple/rivet no problem with mine. I measured the displacement on mine and with 90 PSI air(I think it makes 3,000 -3,500 psi ate the ram), I adjusted the ram to be way to far out (so I would get max deflection possible for the pressure) and measured with my dial indicator. I got 0.003" deflection at about a 20-30 degree angle which gives me around 0.0028-0.0025" deflection. I tried a hard line from the intensifier to the head but it was far too small a diameter, hence the slow actuation, since been removed.

Here is the link to Youtube of my measurements.http://www.youtube.com/watch?v=5opG2IGIiSE


Here are pics of the "C" frame
http://www.nicksrv14.com/nicksrv14/My_Albums/Pages/%22C%22_Frame_Dimpler_Riveter.html
 
Don't take this the wrong way, but some first timers approach rivets in the worst way. I have seen some amazing things done to drive rivets where all that was needed was a good quality rivet gun, the correct set, a bucking bar and the target clamped so it will not move.

There are 20 to 40 thousand (?) rivets in any RV and most have to be set with a gun. Before you are done you will be able to rivet with either hand at any angle standing on your head and wondering why it seemed so hard at the start!

The best approach is to practice driving a bunch of rivets. Don't bother with two pieces of material, just drive a bunch. The hardest to drive correctly are the first you see, the 1/8" in the rudder spar!

Just as important is to practice drilling out bad ones. Even after three metal airplanes, I still mess up rivets (mostly when using my hand squeezers). Any Van's kit has far more than enough rivets and they are the least expensive thing you will ever use.
 
Thanks for all the help and info all, it's greatly appreciated. We've determined that we're asking our c-frame to perform outside its capabilities and will either modify it to squeeze rivets or leave it setup as a dimplier. No hard feels at all Bruce. We are striving for perfection in every rivet and this c-frame was built to help us achieve that.
 
Numatx input

At Oshkosh 2010, when the Numatx powered head was introduced that would fit a DRDT-2 frame, it was intended for Dimpling of skins. The primary advantage of the powered head is it allows for holding the skins with two hands, permitting rapid dimpling. The hydraulic powered head produces ~ 3,000 lbf, comparable to single piston pneumatic squeezers. With a reach of 22", the bending moment at the C-frame root is HUGE, at ~ 60,000 in-lbs! This is the reason the C-frame needs to be beefy to resist bending. Some twisting can also occur if the entire powered setup is not well centered on the frame, in a sideways sense. I have measured a frame deflection of ~ 1/4" at the frame tip. with frames built with 3/16" tube. This has driven some to build frames from 1/4" tubing, and also add doublers at the frame root. I can not begin to attest to all that have used the Numatx powered heads for riveting, but I know many have with good results.

Many good points have been raised about the myriad of factors which can lead to malforming a rivet head in general. Two factor which have not been mentioned are:

1. The skin needs to be lying on a table surface, so as to keep it perfectly square to the compression pin action. If one is hand holding the peice to be riveted, it is difficult to ensure it is well square. Jason, after our phone discussion yesterday, I can see your c-frame setup, and I suggest that you somehow recess the frame so that the skin can be supported by the table to be level with the C-frame lower leg upper surface (where the action is occuring). It looks like you may have been using the wooden block to achieve the same result. This is critical IMO.

2. Rivets are forged like popcorn, and their ends are not always perfectly square. There is no getting around this, but it is a factor. The picture of the rivet squeezed by itself doen't really affect the discussion, since rivets unsupported will typically buckle to one side when squeezed (column buckling). the slightest amount of angle on the rivet will introduce a side force that will cause the rivet to shift just prior to achieving full force. The human eye cant see a few thou of movement, but that is what is happening, and Jason's pictures are the end result.

A final suggestion is that it should be possible to make a spring loaded cup for pressing the skin to the C-frame just prior to riveting. This is akin to the back riveting die set used with rivet guns. I'm sure I could devise one down the road, to fall in the Numatx offereing.

Hope this may help.
 
Nice machine, but, stick to dimpling with it. When you rivet or dimple with a "C" frame you're not actually using the frame to form the dimple or squeeze the rivet. The hammer blow straight down on the sets or dies does the work and the force is not exerted to the "C" frame. The frame is mearely used to align the sets and dies. With the hydraulic force mounted to the frame, it can easily contort the frame and skew the rivet. The yokes on pneumatic squeezers are pretty hefty hardened steel to take that kind of force and are a lot smaller lessening the likelihood of bending and contorting.

Roberta:)
 
Hi Mark,
Thanks for the reply to this and all the info on the phone yesterday. The picture of the C-frame I posted does not show the support tables that were made.
Here's a shot with them installed.
36CCB4C0-BBF1-485A-B900-1C2716931E55-1284-000000D5F4797CEF_zpsfb2338ae.jpg


You can also see in the yellow container at the bottom left there is a spring loaded rivet cup that will compress everything just before squeezing. It's the larger black piece.

I really think we are getting too much deflection to produced an acceptable rivet. We'll play around with it this week and see what the readings truly are. We've got a few ideas on stiffening things up a bit.

Mark- we have no doubt your head unit is more than capable of doing what we want it to. That is really a nicely made piece of equipment.
 
hardened steel makes no difference in flex

Robert,

While your general advice to not use the powered C-frame may be reasonably valid for most, it does not mean one cant get there by taking sufficient measures to maximize stiffness and minimize play.

Hardened steel makes no difference in the FLEXIBILITY of steel. This is true, since it does not change the modulus (stiffness) of the steel, which is nearly 30E6 psi. So when one does FEA, as I have done for this C-frame and other frames, jaws, yokes, etc. the level of heat treat has zero to do with it. Only the Modulus. And before I get a retort back asking "why do manufactures bother to heat treat?", the answer is it raises the yield/ultimate point of the material. This affects wether the product (C-frame) can live forever under repeated cyclic loading, or if the life will be limited (s-n curve for a material and heat treat level).

I have further measured the flex of a 3" yoke with 3K of force to be ~.090". This is very visible to the naked eye. This is .090/3" or .03 versus the frame flex I mentioned of .25/22" or .011. So the angle of distortion is less or equal for the C-frame.
 
Last edited:
slightly more compact powered head

Jason,

Thanks. The new head is slightly more compact, and can accept the cylinder from the 3060 hand squeezer.
 
You must log in or register to reply here.