Replicating a Hawker spun rivet joint

Help Support Homebuilt Aircraft & Kit Plane Forum:

Autodidact

Well-Known Member
The first one is how hawker did it, the second one is how I'm thinking about doing it; will it work, am I over thinking this, should I just put pull rivets on each side of the tube (instead of one long one all the way through) and forget about the spacer tube, ferrule, and washers?

Jay Kempf

Curmudgeon in Training (CIT)
What does it do in the end? Is this a hard point in a tube somewhere? If so you need the hole through the middle. When I looked at the first picture what I was thinking was a through bolt with two cones on either end. Put the assembly together, tighten the bold and flare the end just like a plumbing flare tool. If you have access to both ends you could do it with a large C frame press and a couple cones. Then there is the, how do you get the inner compression sleeve in and in place part of the whole equation. Again, that is an access thing but I am sure solvable with some sort of coat hanger or string thing.

BBerson

Light Plane Philosopher
Most joints need more than one rivet. Getting the spacers in tube is hard to do and must fit exact.
Pop rivets that long are hard to find.

Last edited:

Autodidact

Well-Known Member
Most joints need more than one rivet. Getting the spacers in tube is hard to do and must fit exact.
Pop rivets that long are hard to find.
Yes, yes, and yes. I did find a source for those long rivets, though, and that's what prompted me to make this thread. But Jay had an idea that i think will let me make the spun rivets without the complicated and rare tool that Hawker (and now Retrotec) used. Here are a picture of a typical fuselage joint from a Hawker Hind, I think, and also a drawing of the rivet spinning tool. It works like an old hand cranked drill press in principle; you turn the chuck and there is also a wheel to turn that increases the pressure on the work piece (I didn't make the drawing of the rivet tool, Retrotec emailed it to me a while back):

When I looked at the first picture what I was thinking was a through bolt with two cones on either end. Put the assembly together, tighten the bold and flare the end just like a plumbing flare tool. If you have access to both ends you could do it with a large C frame press and a couple cones. Then there is the, how do you get the inner compression sleeve in and in place part of the whole equation. Again, that is an access thing but I am sure solvable with some sort of coat hanger or string thing.
That is genius. A fine thread bolt with a cone that slides up to the bolt head and then a special nut with the forming tool as part of it. You could let the bolt turn a little bit slower than the tool/nut and vary the pressure infinitely just like you can with the C-frame press but without building or tracking down a rare and hard to get tool or adapting an old drill press. The fuselage joint above is one of those hard points - the front cabane strut mount just behind the firewall (I think).

By the way, the rivets are formed with a flange on one end with a regular drill press and are then finished off on the truss with the C-frame tool.

Last edited:

BBerson

Light Plane Philosopher
Why not long solid rivets? Much stronger than hollow.
Solid rivets can be peened down with a hammer. The technique is described in FAA 43.13.

Still looks like much fiddly labor to me. What was done for the war is probably too labor intensive today.
I worked in a small airplane factory building a modern version of a ww2 design (Interstate L6). The tiny spacers and rivets and bits for each wing rib took several hours per rib to assemble. It was a nightmare compared with a one piece modern stamped rib.

Autodidact

Well-Known Member
It still looks like much fiddly labor to me. What was done for the war is probably too labor intensive today. I worked in a small airplane factory building a modern version of a ww2 design (Interstate L6). The tiny spacers and rivets and bits for each wing rib took several hours per rib to assemble. It was a nightmare compared with a one piece modern stamped rib.
I can't justify it with logic, but imagine a scaled down, open cockpit version of the pic below powered by a UL Power or similar engine and if the structure was almost a dead ringer for the Fury biplane or the Hurricane? My gut tells me that you are right about the extra work, though:

TFF

Well-Known Member
The tool would not be that hard to make. C Clamp, brake pad rivet tool, and a tubing flare tool all have aspects. Finding the right metal tube stock to use would be harder.

Matt G.

Well-Known Member
I can't see any reason to do this other than historical accuracy, so why not just use a blind rivet on each side of the joint? That way at least you know the joint strength allowables.

Autodidact

Well-Known Member
I can't see any reason to do this other than historical accuracy, so why not just use a blind rivet on each side of the joint? That way at least you know the joint strength allowables.
Yeah; I think I just needed to hear a couple of people say that. It is an awfully complicated way to do something. And time consuming, I would guess.

Harvardiv

Active Member
How would you control the quality and guarantee the strength using a peened hollow rivet? Why not just use the 2 ferrules with a solid rivet? A Hurricane rebuilder in Canada said they did their static one that way. So I would think if this technique is in 43.13 you could do it.

cvairwerks

Well-Known Member
For those joints, the washer, ferrule and the distance tube are required....They weren't added to make things difficult.

Ian has some good construction details from his Typhoon, with uses bolts instead of rivets in most joings.

Harvardiv

Active Member
Not interested in doing things differently.

wsimpso1

Super Moderator
Staff member
Log Member
Hawker was bolting their airframes together, and there were places where they apparently needed to lower the bearing stress at a joint, but it appears they desired less weight than the solid bolts. Even their bolted joints used bushings to get the bearing stresses down while using bolts of smaller cross section and weight than a full section bolt. They appear to have joints backup and/or more preload was required and they used bolts, or less back/strength/preload was needed and they set these tubular rivets. As our Canadian friend described, these were sometimes special parts, sometimes standard Hawker parts. It all smacks of Hawker engineering a series of fasteners for their own use. Just looking at the illustration says "Highly Engineered System" to me.

There are other features of such a system that I just can not like.

The system looks like a sieve for water entrance to the structure. One would have to come up with a way to seal every joint as it is built, then prevent corrosion from the modest air and water vapor that is built in - perhaps wet each fastener with epoxy when installed and then coat the insides with linseed oil in welded steel tube structures.

Sleeve placement for every hole.

All these bolted joints with forged ends, etc, when we can just weld in a bushing and straps to locally reinforce for a needed engine or wing attach point?

I would build a steel tube truss and gas weld the joints, just like is done reliably in so many other homebuilts. Inventing your own fastener system to use - ugh.

Billski

Harvardiv

Active Member
Actually the bushings were not likely used to reduce bearing stress, but more likely for an interference fit. Hawker aircraft didn't use any welding anywhere so it wasn't an option.

cvairwerks

Well-Known Member
It would be worth contacting the guys down in Australia that were restoring Jack McDonald's Hawker Demon, as it uses the same rivet construction. They had the airframe completely dismantled and reriveted it.

Harvardiv

Active Member
Thanks for the info, I really like the method used on the Canadian Hurricanes of a solid rivet through bushings instead of a tube rivet through bushings. I believe CCF modified the technique.

Riggerrob

Well-Known Member
Actually the bushings were not likely used to reduce bearing stress, but more likely for an interference fit. Hawker aircraft didn't use any welding anywhere so it wasn't an option.
Yes.
Britain suffered from a shortage of certified welders and welding torches up until the end of World War 2.
Try water-proofing an armored fighting vehicle to ford even shallow rivers. Hah! Hah!
That is why Hawker airplanes used complex, precisely reamed pin joints long after Americans had converted to welded tube joints.

The modern equivalent is the bolted, aluminum tube airframes widely used on modern ultralights. See Robert Baslee's Aerodrome Airplane kits.

Last edited:

cvairwerks

Well-Known Member
Fairey and Hawker used the rivet/bolt joint design way back, long before WW2 started. Didn't take a highly skilled person to assemble the joint.

Staff member
Log Member
Redacted.

Aviacs

Well-Known Member
What TFF said - That hawker tool is a dead ringer for my stainless brakeline double flare spinning tool. Well, after you cross it with a forged structural C-clamp (actually, have to say "G-cramp" if it is to be properly Bitish, of course) & spin grind the cones for the desired contour.

Point being, for maybe $90 -$150 and some fun time at the lathe, milling machine, with perhaps some spinning on the surface grinder, you can have a dead ringer for the original tool, to form original looking functional connections.

Dunno what your scrounging options are, but sometimes shopping fleamarkets and junk shops are productive for the raw materials (sacrificial donor tools).

If the facility exists to make your own dies, it might even be possible to start with one of the low-buck soft metal tube tools from china.
If spending over \$30, be sure the tool is eccentric/spinning type, and was designed to include stainless brake lines.

smt
PS: the pulled rivet joint might work ok, but per your sketch, i don't see it compressing the ferule and "everything" against the spacer tube with quite the positive effect of the original. To fill all the space in the connector plate holes as well as the tube. Part of the function of the ferule appears to take up that role in the original. I think it is getting spun/displaced as the rivet is, rivet taking most of the tension function?

Do you have a turret lathe? How many of each part do you need?
I guess these days, should consider cnc. But i still set up turrets for parts like that. Sometimes still faster, but certainly monotonous after the first few 100.

Last edited: