What's new
Van's Air Force

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

Cleaveland Aircraft Tool, Rivet Squeezer Hydro-Pneumatic System Kit

In my experience with building C-frames and the prototype DRDT/HPRS unit the bottom die receiver should be bolted through oversize holes. This will allow for any lateral adjustment and angular shimming (to account for welding) in order to get the dies to be both perfectly concentric and perfectly parallel.

-
 
Bolting works too

Bolting works too. It does take a couple of fasteners, a few holes drilled,... but it could arguably be the better plan.

Once the powered head is in place, and the compression shaft is being used to align the dies, and thereby the lower die receiver plate, a quick tack weld is pretty simple (after all the heavy welding that has been done in the first place!).

One way or another there should be zero movement of the die axes down the road.
 
grayforge said:
Good info Mike. Can you tell us more about your dies and how they differ from those sold by other vendors?

Thanks!
Russ
Click to expand...

Russ,

Sorry my reply has taken so long. I had made up the page that will be linked below, but didn't want to switch this to a dimple die thread. Since there is a dimple die thread going on, I decided that I would post there. I do still intend on getting a tutorial on how to 'see' the difference in results of dies and techniques.

<snip from: http://www.vansairforce.com/community/showthread.php?t=81949&page=2>

die4264_lg.jpg

At Cleaveland we perfected the 1942 springback dimple die design. The term springback refers to the angle of the 'faces' of the dies rather than the angle of the 'funnel'. The male die has a concave face of a couple degrees where the female has a convex face. These angles as well as some of the other diameters and angles are critical as to how they fit the rivet, and how flat the material ends up after dimpling. Most of our competitors now have attempted to copy our 1992 die geometry, some with more success than others, but they still don't measure up to what a Cleaveland Dimple Die can do. To read more about exactly how our dies were perfected and how we keep them that way you can read the following page that I have put together to answer this question.

http://www.cleavelandtoolstore.com/custompages/springback-dimple-dies.html

As some builders don't know what to look for I will create a short tutorial on how to evaluate the dimpled skin and post it here shortly.

The statement about the tank dies for the substructure is certainly correct. The tank dies are about .007" deeper than standard to account for proseal under the rivet head. They work better for nesting than two dimples from the same die. Most builders don't want to buy another set for the substructure, but for those discerning builders we do carry a "substructure" die (DIE4263SS) that is .011" deeper so that the dimples will nest perfectly.

-
 
Karsten,

What size tubing did you use? 5" x 3"?
3/16" or 1/4" wall thickness?

Curious how much the ends push apart when riveting.
 
A theory question for you Mike. I read your post on Springback Dimple Die history... very interesting.

One thing I wondered was how close to cylindrical/vertical the holes end up being. I saw a reference from a manufacturer somewhere that says the cylinder of the hole after dimpling should be vertical, with no edges pointing in.

Here's an example of what I imagine they don't want to see:
dimple.gif

from: Vans Riveting Specs

In your Springback Dimple Die history, there's a quote from a book stating:
"The cylindrical hole was obtained by drilling the initial hole in the sheet to a diameter slightly smaller than that of the mandrel on the male die (which served to center the die in the hole, as in fig. 3c); the suitably tapered mandrel then forged the hole to the desired shape in the dimpling operation. The hole could not be made too small, however, lest radial cracks appear around its edge."

Kitbuilders typically match drill their holes first, then dimple them. The holes are a bit larger than the dimple mandrel of the die. So wouldn't this cause the non vertical (conical) hole sides as shown in the figure above? Or are the die mandrels actually slightly smaller than the hole when dimpling so that they are formed vertically?

Hope this all makes sense. :) My guess is that our planes are overbuilt, so these non vertical sides are not a factor.

Thanks!
Russ
 
Finished up my DRDT-2 Plus. Powered by the HPRS power head. It was a joy to build. Even fabricated the die receiver from 1/2 flat steel since that only comes with the standard DRDT-2 Front End kit (manual lever) which I never purchased. Went straight to the Hydro-Pneumatic head. :)

Can't wait to try it out (still need the air switch from Numatx).

DRDT-Plus_1.JPG



DRDT-Plus_2.JPG
 
Squeezing technique

Hi Mark,

A question about technique when using the HPRS: what's the typical process/sequence when using the system to squeeze a particular length rivet?

In the build class I took, using the pure pneumatic squeezer, one would start a bit loose, squeeze the rivet, and progressively tighten the adjustable set and squeeze again, repeating until the rivet was properly shaped. It might require some fine tuning on the next rivet since the first may have been work hardened a bit.

Since the HPRS applies constant instead of geometric pressure, I get the impression you don't use adjustable rivet set preload as an adjuster. It sounds like you use pressure into the Intensifier. So adjust the rivet set to be close enough to the rivet to allow full compression, then, say for AD3 rivets, start at a lower air pressure, like 70psi, squeeze a rivet, measure it, then increase pressure in stages until the rivet is sized properly.

Does this sound like the proper sequence?

Thanks!
Russ
 
Adjusting the HPRS to properly form rivet heads

Russ,

Nice looking C-frame!

SHORT ANSWER:

Yes, for the Hydropneumatic rivet squeezer (HPRS) you are correct, you can adjust the AIR PRESSURE to the intensifier slightly to adjust the HYDRAULIC PRESSURE the intensifier creates. Some short trial and error will tell you what combination works for 3/32" and 1/8" rivets. Do a couple of test squeezes on scraps, and record what air pressure settings work. The nice thing about the HPRS, is that it does create a constant force, and thus you don't have to fiddle around as much finding the .030 -.060" sweet spot of travel range for a traditional pneumatic rivet squeezer (TPRS).

You can also play it conservative on each task, by first adjusting the compression pin so that it runs out of stroke just before it touches the rivet. Then you dial out the compression pin in 1/2 turn increments (1/32" lengths) and perform test squeezes until you are getting a properly formed rivet. If the HPRS compression pin runs out of stroke, it doesn't matter how much force it is making, it simply can not over form the head.


LONG ANSWER:

When I was first using a TPRS, I thought I could exactly calculate where the rivet set should end, relative to the stackup of metal I was trying to rivet. I quickly found out this was not true.

For example, lets say you want to rivet two 1/8" thick 2024 alclad sheets together with a 1/8" rivet. A rule of thumb is that the the rivet shank should initally stick 1-1/2D through the prepared hole, and then be compressed to a height of about .5D beyond the prepared hole, where D is the rivet diameter. You could then calculate the inital grip length as:

1/8 + 1/8 + 1.5x1/8 = 2/16 + 2/16 + 3/16 = 7/16

This means you would use a -7 grip length rivet, since rivet grip lengths are in 1/16 increments (diameters are in 1/32" increments).

You would then figure that you could set the compression pin of the TPRS to be the difference of the initial rivet length and the finial rivet length, which = 1.5D -.5D, or just 1D of squeeze (1/8"). So you would then set the gap between the rivet dies to be 5/16" when the squeezer compression pin was fully exteneded.

You would find you were under squeezing the rivet, and not getting the .5D final formed head height.

WHY?

First of all, both the TPRS and my HPRS cause some deflection of the yoke tip when they generate on the order of 3,000 Lbf to squeeze a 1/8" AD4 aircraft rivet. I have measured about .090" of deflection for a typical 3" yoke. This has to be factored into the "equation" when you squeeze a rivet. By "equation" I really mean some trial and error to account for yoke flex. Also, a TPRS force varies, as I show on a graph on my website. You try to adjust the compression pin of the so that it develops it maximum force at the very end of travel, or about the last 1/18 of stroke. But some fiddling with the TPRS is necessary, since you usually have to account for the yoke flex I mentioned.

The HPRS intensifier is sized to product around 3,500 psi of PRESSURE at 90 psi, which then works on a 1-1/8" dia piston in the forming head to generate about 3,500 Lbf of FORCE. This is the force necessary to squeeze 1/8" AD rivets (2117-T4 alloy). It just so happens that the 1-1/8" piston in the C-squeezer or DRDT head has about 1.0 sq. in. of area, so you get nearly a 1:1 pressure to force relationship. The area of an AD4 rivet is .012 sq. in., and an AD3 is .007 sq. in., or about 1/2. Thus, it should take about 1,700 - 2,000 Lbf of force to squeeze an AD3 rivet. WIth the 1:1 relationship I mentioned, it will take about 50 psi air pressure to do AD3 rivets.

For the DRDT head on the end of the frame, you will also get some flex that has to be accounted for. Again, some trial and error on SCRAP pieces will quickly give you an idea what it will take to do the job.
 
Last edited:
update on 18 months use

Just a quick update on using the hydro pneumatic system on building my 7 SB wings and empennage and waiting for SB fuz to arrive.

Ive used the squeezer body, built a dimpling frame (its over engineered and must weigh 100 pounds) and used the rivet puller. The system is still working great. I've only had to bleed things a few times and it takes no more then 5 mins. I use some disposable gloves to keep my hands clean from the hydraulic fluid.

Dimpling skins with the frame is a joy and ultra quick, with both hands guiding the skin, there is little chance for misplace dimples.

The squeezer works great and being so light is fatigue free.

the rivet puller is a must for getting into to some tight spaces, especially when pulling rivets for the nose ribs to main spar.

Due to operator error i broke a part, Mark was immediate in dispatching a replacement and he did not rip me off with after market support.

I'm now thinking of a hand triggered unit for building the fuz. Another excuse for another tool.

It gets 5 star and is my favourite tool.
 
Thanks for the thorough explanation Mark! I'm still doing some prep work in the garage, but am really looking forward to trying the HPRS out.

I may be taking our local builders assist shop's (Axsys Air) "Building the Empennage" class in a couple weeks (work permitting). We're supposed to basically get the tail done in a hard core week of building. I may bring the HPRS system and DRDT-2 in to use instead of their wrist stressing pneumatic squeezer. They'll probably be interested in the HPRS. :)

Thanks again,
Russ
 
David,

Thank you, once again, for you compliments. I'm glad it is a favored tool of yours. BTW, you can convert your intensifier into a more compact solenoid operated unit, by buying a 5-port, 4-way, single acting valve. 12V is what I use for the -2 version.

Russ,

Looks like the class you might take at Axsys air is a good one. Very tidy looking shop and I love the looks of their hangar.

Thus far, builders like David who have purchased my system have been very pleased with it, and I am confident you will be too. From day one, I have known that the best advertising is by word of mouth, and the converse is also true, that a displeased owner can really hurt a product's reputation. Be careful if you bring your C-frame and system to the class, as you may have trouble getting it back!
 
Another happy owner of Mark's Hydro-Pneumatic system here.
A first class product and first class service...

Cheers
Ken
 
WOW

I set up the HPRS last night after receiving the final part for it. I was going to stop on my way home to grab some tranny fluid, but thought I'd check to see if I had any laying around the garage first. When I opened the box containing the pneumatic switch, I was happy to find a container of Transmission fluid included! The kit really includes everything you need.

After filling & bleeding the system, the initial throw and pressure was too low. I just needed to bleed more vigorously. Plus, adding a drain hose to the head unit's bleed fitting let me see all the bubbles still there. This was mentioned in the directions, but I thought I could tell just by watching the fluid come from the bleed fitting. After this, the unit worked like a champ!

Being able to use 2 hands to hold your airplane parts while operating the squeezer with a foot pedal is such an enlightening experience. Full control of your work! Adjusting the pressure was easily done by adjusting the incoming air pressure. -4 rivet? 90psi. -3 rivet? 50psi.I love it!

One thing that surprised me: The arms of my crazy burly custom DRDT-2 (see photos in earlier posts in this thread) deflected apart. D*mn! 3800lbs of pressure is a lot. I hadn't expected any noticeable deflection.

I swapped the high pressure line over to the hand squeezer and bled it too. Works like a charm. So much lighter than the old pneumatic only style.

This almost sounds like a paid infomercial, but when I appreciate the quality of something, I really like to share. Mark, you've done a terrific job designing and manufacturing this system. Thanks!

Russ
 
grayforge said:
One thing that surprised me: The arms of my crazy burly custom DRDT-2 (see photos in earlier posts in this thread) deflected apart. D*mn! 3800lbs of pressure is a lot. I hadn't expected any noticeable deflection.
Click to expand...

This is surprising. What size tubing did you use? 1/4" X what? From the photo I'm guessing 3" X 6". So I'm wondering how one would further decrease deflection. Either by thicker tubing walls or greater vertical dimension (8"?). Maybe Mark or Mike can speculate on how to further decrease deflection.
 
Thanks

Russ,

Thanks you for your positive comments. I hope your builders group like the system too.
 
diamond said:
This is surprising. What size tubing did you use? 1/4" X what? From the photo I'm guessing 3" X 6". So I'm wondering how one would further decrease deflection. Either by thicker tubing walls or greater vertical dimension (8"?). Maybe Mark or Mike can speculate on how to further decrease deflection.
Click to expand...

Mark mentions on his website that some deflection will occur with the DRDT-2 frame. I followed his advice and built his C frame, it too works perfectly.

Cheers
Ken
 
C-frame Deflections - defined ad nauseum

Diamond,

The deflection amount (on the order of 1/4") is not surprising to me, but then again my specialty in engineering is aircraft structures (design and stress analysis). While the the C-frame is made from heavy wall steel tubing and may seem overkill, there it is one Helluva bending moment applied to the frame. The HPRS heads make around 3,500 Lbf of force, at an intensifier air pressure of 90 psi. This translates into a bending moment at the root of the 22" C-frame of:

3,500 Lbf x 22" = 77,000 Lb-in !!!
For this 77,000 Lb-in bending moment, one can imagine >>

1) a load of 1 Lbf at a distance of 77,000 inches (imagine how stiff / light this lever arm would need to be to even transmit the moment)

or

2) a load of 77,000 Lbf at a distance of one inch.

or

3) any combination of force and moment arm between these ranges that equates to 77,000 Lb-inches.

77,000 Lb-in equates to 6,416 Lb-FEET of moment. One way or another, again it is a heck of a big moment.


If you look on my website, I did 3D FEA (Finite Element Analysis) to predict the stress and deflections of a 6"x3" x 3/16" wall C-frame. The predicted deflections correlated well with what was observed in a real welded frame. I also later designed a much stiffer C frame that can be waterjet cut from 1-1/2" mild plate steel. Some will say the steel plate design is overkill, and they are right, IF, they are using a lever operated unit, where they have ample compression pin travel. The compression pin travel of my HPRS heads is 9/16", which matches standard pneumatic squeezers. So there is not a surplus of pin travel, as there is with a hand operated unit.

Thus, if you want to build a lighter-duty C-frame use the a hand operated lever.

If you like the concept of two-handed control of the skin, and foot control of the compression pin, my HPRS shines here, as David Perl mentioned. I might look at what it would take, doubler wise, to stiffen up the basic 3x6" design. Or maybe come up with my own tube design if I get a wild hair. So far, others have successfully used 3/16" wall tubing, just keep in mind that you will need to adjust the dies so there is only about 1/8" of clearance from the female die (mounted at the end of the moving pin), in order to allow for C-frame flex.
 
diamond said:
This is surprising. What size tubing did you use? 1/4" X what? From the photo I'm guessing 3" X 6". So I'm wondering how one would further decrease deflection. Either by thicker tubing walls or greater vertical dimension (8"?). Maybe Mark or Mike can speculate on how to further decrease deflection.
Click to expand...

I used 3"x6" rectangular steel with 1/4" walls. I probably should have run the doublers further forward, IE in a 'C' shape. But my setup will work until I get the inclination to make it yet stronger. :)
 
I am surprised, as David Perl reported only about 1/16" deflection with 1/4" wall at full pressure. Can you guys compare steel types?

While Mark is right about the travel limitations with the deflection, we have also found that torsional rigidity to be the biggest problem in setting rivets with the DRDT frame. And shipping weight to be a problem with any solution. With so much talent in the mix, perhaps a 'groupthink' solution can be attained using locally available materials and methods with some ship-able connectors or specialty machined items?
 
If I recall correctly, I used hot rolled steel. I don't remember the specific type or grade.

On an unrelated note, I made a modification to my HPRS. I separated the foot pedal from the Intensifier and connected them with about 6' of twin 1/4" nylon tubing. This is so I can keep the intensifier on the workbench, providing more slack for the hydraulic line when attached to the DRDT-2. (vs fabricating a hard line for the DRDT-2 running from the head to the rear)

I also slipped a length of braided sleeving around the hydraulic line to prevent it from harming the finish of anything nearby
 
Frame deflections

Mike,

The 1/4" DEFLECTION was what we observed at Oshkosh 2010, with a 3/16" wall frame, and my FEA closely matched this deflection amount. It sounds like the 1/4" WALL that David used knocked this deflection down quite a bit. I think David used 1/4" wall in response to some discussions he and I had at the time, and also the work you (Mike) and I were doing to fine tune the DRDT head.

Putting some gussets on even the 1/4" walled frame, might still help lower the stresses in the welded internal corners of the "C"
 
Relocating the foot valve

Russ,

Relocating the foot valve is perfectly acceptable for your needs. It probably cleans up the hydraulic hose routing for using the DRDT powered head. The other option of course is still the -2 intensifier, with the electronic solenoid.

Also, you might consider an in-line oiler right near the intensifier, to make sure the LP piston it is getting oil. That, or be diligent to drop 3-4 drops of air tool oil at reasonable intervals (every few hours of operation).

Here is one option:

http://www.mcmaster.com/#air-tool-oilers/=gylunv

Type A shown in link above.
 
Since the entire throw distance of the frame arms is probably not always utilized, how about using an adjustable clamp that can moved forward and backward on the arms so as to reduce deflection? Just clamp the top of the upper arm to the bottom of the lower arm just beyond the distance your material requires. Here's a clamp from HF for example. Would this work?

http://www.harborfreight.com/2-piece-3-4-quarter-inch-heavy-duty-cast-iron-pipe-clamp-31255.html

Or, maybe weld up a thick metal sleeve that slips over front of frame and can slide forward or backward to desired point where reinforcement is needed?
 
Last edited:
Diamond, I thought of the same thing and used some of those same clamps that we already had. When applying the force it just stripped them (one on each side) off. A welded collar may work, but it does not solve the torsion problem.

Mark, the 1/4" is what surprised me as both David and Russ used the same thickness with differing results.
 
deflection differences

Mike,

I don't know for sure what to attribute the deflection differences to. Steel has a modulus (E) of 30 x 10^6 (30M) psi. And even if one frame was heat treated, the modulus (basically the spring rate, K = F/X) doesn't change. Heat treat only affects the yield and ultimate strength, not the stiffness. You know all this, I'm reciting it for anyone else who may care and not know.

There could be some differences in the quality of the steel. Another thing that comes to mind, is that David is in the UK, and so maybe his thickness was slightly different than truly 1/4" (1/4" = 6.35mm).

Anyway, some U shaped doublers could help. A little more weight, cost, time...
 
My version

I wanted to go the same route, but like others I wanted to increase the wall thickness to 1/4". Found a local machine shop to do the work for me since I can't weld and dont have time to learn. After quoting the price, they informed me that they couldn't get less than 20' of the 1/4" so they used 3/8" for same price! It has 1/4" plate welded to reinforce the root. At 106# this thing is a behemoth!

Ohh yeah, and I went with Ferrari red wrinkle powder coat...maybe less chance of "missing it" and banging into it...probably not:rolleyes:


34tbtdl.jpg
 
Not sure

I am not 100% sure, but I think it was all A36 hot rolled. The tolerances look to be great, but I won't know for sure until I try to attach the head, etc. I do think it is just a touch (just under 1/32...maybe 1.5/64) short of the 2 " overhang from top and bottom arms. Probably not a big deal at all, once I get the head on, pressurize it and extend it, if the bottom does not line up with the holes, I will either enlarge the holes or just tac the bottom receiver in place.
 
ROTARY7 said:
Diamond,

Yes, you could probably add two lines with a tee at top of the intensifier and a manual shutoff valve to either line.

I may have not made this clear enough, but adding about 10cc of fluid to the head, brought on by some frequency of changing the heads, literally takes about a minute. It is not a complete recharge of the oil in the system, which is only about 50cc anyway. You are only pushing a small quantity of fluid to the head to force any air bubble out of the head.

When launching into a long project like building an RV, everyone decides how best to utilize their time, how much to spend on tooling, etc... If bleeding this system sounds like a lot of work, make sure to get a quick build kit, becasue you are otherwise in for a lot more work than you might realize. I know this personally. Also, I'd get the instrument panel work done by someone else and a host of other things. Bleeding this system is going to be by far the least of your troubles.

In the course of building my RV-7 I've probably spent more time adjusting the compression pin distance of a standard pneumatic squeezer than I would ever spend bleeding my HPRS system from occasionally swapping heads. Also, the weight difference between a standard pneumatic unit and my HPRS heads is significant. Holding an additional 3 lbs over the course of an hour of edge dimpling skins is very noticeable. The weight savings of my unit more than pays for any fuss with bleeding, in my opinion.

Again, seek out some other users experience. Ask David Perl in England or Karsten Woltering in Germany how many times they have needed to bleed their units. They both have the squeezer head, the puller head, and a DRDT head. Each head is powered by a single intensifier (-1 version) and each head has a QD. Their posts are number #3 and #5 respectively of this thread. Although they have never complained to me about bleeding the unit, you can get their feedback directly.
Click to expand...


Ive only had to bleed as couple of times and am using the squeezer, a squeezer on my dimple frame and a rivet puller. I've built SB wings and almost done with SB fuz. Once you get the hang of it, bleeding takes only a few minutes but wear some disposable gloves as the fluids is messy.
 
ROTARY7 said:
Mike,

The 1/4" DEFLECTION was what we observed at Oshkosh 2010, with a 3/16" wall frame, and my FEA closely matched this deflection amount. It sounds like the 1/4" WALL that David used knocked this deflection down quite a bit. I think David used 1/4" wall in response to some discussions he and I had at the time, and also the work you (Mike) and I were doing to fine tune the DRDT head.

Putting some gussets on even the 1/4" walled frame, might still help lower the stresses in the welded internal corners of the "C"
Click to expand...

Just re reading this thread and decided to re-measure the frame thickness, it is made of 3/8 inch thick steel, no-wonder i get a hernia when i try and lift it!
 
Still my favorite, and probably dimples the best to boot and you won't get a hernia lifting it!

12.jpg

Picture courtesy of Avery tools
 
Yeah, but try dimpling a large sheet of aluminum solo with a hammer powered C Frame. Doable, but time consuming.

I love using 2 hands to position the sheet and a foot to operate the squeezer. Dimples appear as if they were on an assembly line. :)
 
grayforge/Lampsguy,

Could you post the dimensions of the homemade DRDT? Length of tubes, distance between the top and bottom tube, etc. I would like to build one and would like some advice. I would really appreciate it.

Thanks
Ken
 
Thanks. That was exactly what I was looking for. Do you think it is worth buying the pneumatic head for the build process?

Ken
 
Same

I used the drawings (sent email to designer asking permission first) provided by experimental aero. I also had a local guy do the powder coating, as I think it might as well look great!
 
I see where others have put reinforcement, but do we know exactly where the flex is happening? I would think it would be in the upper and lower arms, but the reinforcement I see others making is at the rear transition between upper and lower.
 
diamond said:
I see where others have put reinforcement, but do we know exactly where the flex is happening? I would think it would be in the upper and lower arms, but the reinforcement I see others making is at the rear transition between upper and lower.
Click to expand...

Your are right, the flex happens in the arms, but the yield point is at the back where the arms are welded to the vertical portion. Most that build their own frame do it with thicker material to limit the flex and allow the hydraulic front end to do the most work. However the welds at the innermost point are right at the theoretical yield strength, so done correctly, this can be strengthened when constructing.
 
I definitely feel the Hydro-Pneumatic DRDT2 head is worth is, along with the hand squeezer. I just dimpled my forward upper skin and found it a breeze.

Be sure to use cold rolled steel. Stronger than the hot variety.
 
You must log in or register to reply here.
Back
Top