Replicating a Hawker spun rivet joint

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,130
Location
Memphis, TN
The big thing is I wouldn’t want to manufacture an obscure faster for a small job. If I had a Hurricane or a Hart or a Demon or Fury; I would be all in with funky fasteners. Those are dream planes for anyone. I would not let no tools stop me on that. But. I would be inclined to just buy some hi-loks. Seems expensive except it’s one and done. Simple tools. You personally don’t have to make lots of little bitty pieces for each hole. The time could be spent better. If you came up with a system to sell where it’s all off the shelf, I could be more onboard with that.
 

Harvardiv

Active Member
Joined
Jan 2, 2007
Messages
41
The big thing is I wouldn’t want to manufacture an obscure faster for a small job. If I had a Hurricane or a Hart or a Demon or Fury; I would be all in with funky fasteners. Those are dream planes for anyone. I would not let no tools stop me on that. But. I would be inclined to just buy some hi-loks. Seems expensive except it’s one and done. Simple tools. You personally don’t have to make lots of little bitty pieces for each hole. The time could be spent better. If you came up with a system to sell where it’s all off the shelf, I could be more onboard with that.
Hiloks that someone mentioned earlier is probably a great solution, as they can be found in a form which mimics the original appearance. With that said, I pretty much agree that on a rare plane its probably better to go original of you can even if it costs 4x more in time and money. After interviewing several top hurricane rebuilders I have concluded that solid rivets were not original to ccf hurricanes, and that a rotating riveting tool appears not to be original either.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,130
Location
Memphis, TN
A rotating tool would be too slow for production. In the field they might have cobbled up stuff like it from time to time. They probably used something that looked like a standard manual rivet squeezer with jaws and a four foot handle. Maybe hydraulic or air if they got fancy or there is some sort of big preload. The question comes were both ends shop formed or was one manufactured and one shop formed. I would be inclined to say one was manufactured to keep consistency and speed of assembling. A machine probably made the head and then chopped to length. It could have been a shop floor tool done manually. Just feed it tube. Make the head, crank out the length, cut off. Some apprentices arm was probably sore every day making 2” sleeves all day long.
 

Mad MAC

Well-Known Member
Joined
Dec 9, 2004
Messages
819
Location
Hamilton New Zealand
Anyone poking around vintage aircraft spec's should have a look at the silver biplanes. Silver Biplanes

If you can get consistent tubes, the easiest way to form them, might be to have a pair of hard forming dies and thread all of them together with a high strength bolt to drive the dies (bolt, die, rivet, die nut) to form the rivet tails.

Actually if you increase the rivet size to allow the use of an aluminum tube, why wouldn't you just use a bucking bar and rivet gun with modifered faces to form the tails.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,130
Location
Memphis, TN
Traditional rivets take some minor skill and if you’ve ever bucked a -5 or bigger rivet by hand, you know it’s not the right way to do this. Also the Hawker setup is a replacement for bolts not rivets. It would be several hundred pounds in savings on something like a Hurricane if you could afford the manufacturing complication.

A regular bolt is pretty easy to use and hard to get away from. This is somewhat a complicated system not in a user way but in a manufacturing way. Tooling that you can’t use on something else. A wrench can be used a million times; tools for this would be one and done unless there is damage. The hardened ferrules would have a CNC lathe working a good bit for one plane. Doing them one at a time at home, forever just with the stopping and starting for your normal life. The clamping tubes, still more labor. And after all that you can put it together; oh yea, you have to make those tools to spec too.

If you intended to machine your own bolts and nuts at home, I would pick the Hawker system to machine. If I’m reaching in a bag of fasteners, bolts are so much easier.
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
311
Anyone poking around vintage aircraft spec's should have a look at the silver biplanes. Silver Biplanes
Wow!
Thought my early fascination with biplane models ended by HS with interest in more streamlined types.
But i "want" one of those!

There does not seem to be much info or my google-fu skills are lacking.
Do you have further links for the home page airplane on that site? Including on methods of assembly?

({edited} PS: found the Hawker Hart & Demon, my error had been looking for a streamlined bulldog)
Still interested in methods of assembly.

The hardened ferrules would have a CNC lathe working a good bit for one plane. Doing them one at a time at home, forever just with the stopping and starting for your normal life
I disagree.
A conventional screw machine will beat a cnc hollow (no pun intended) on stuff like that, and a hand screw machine (turret lathe) with bar feed can probably keep up. I used to run hundreds of somewhat complex fastener assemblies at a go, for furniture that i designed. A typical part would be threaded, hollow, chamfer, formed end, and an integral hex (socket) drive. Another might have 2 different internal threads in the same axis, offcenter to the rotational axis of the part. etc.

If you intended to machine your own bolts and nuts at home,
There's the sticking point for "fasteners" whether rivets or bolts - the materials & engineering.

As you know doubt understand, cut thread bolts would be dangerous in many high confidence apps and would not be optimal where they could not be oversize to compensate, such as in aviation. Of course running a roller threading head in a turret (or cnc) lathe as opposed to say a Geometric die head, is not uncommon. Nonetheless, there would be some non-trivial testing to prove equivalence with ansi bolts.

If suitable size tube can be procured, i think the rivets would be easier, and faster. No threads. a few simple tools, just keep the turret moving around.

smt
 
Last edited:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,130
Location
Memphis, TN
If you can walk out in your garage and make this stuff, that’s cool. Homebuilt airplane style. It’s about a one off use at home compared to a factory. Interesting historically, not practical privately.

The rivets would need to be whatever the ferrule material load rating would be. Getting caught up in the aluminum part is missing the engineering. The ferrule is the key. The aluminum tube is just a clamp. Most rivits would need to be HiLoc strong, which is just a fancy bolt. Im definitely not able to drive a rivet that is meant to replace a bolt at home. If you design the cluster to use AD rivets, it’s going to look a lot different that what this is after.
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
311
The ferrule is the key.
That's the part(s) I'm talking about. For a screw machine, which is most likely how they were run in the first place.

The other 2 parts are just chopping tube to length (originally) though today it might require a boring or reaming op depending on modern available tube options.

However, in re-visiting the diagram of the assembly, I had assumed the "tubular-part-to-be-spun" was stainless tube. Cannot find a spec for "T26" except hints about welding it, which seem to indicate it is actually magnesium based? Anyone know?
 
Last edited:

Harvardiv

Active Member
Joined
Jan 2, 2007
Messages
41
I like all of your thoughts on the matter, but what is the most practical, and also most similar historically to the original tube rivets?
At the same time what solution also answers the original question from 2013?

In addition, what solution has the most quantifiable tensile and shear stress loading?

A stainless steel rivnut in my opinion. It likely has a shear and tensile strength probably equal to or superior to the Original British tube rivets likely 10x the loading of aluminum ad aircraft rivets.

A rivnut, the no screw and with internal threads invisible from the outside I think is a great solution.

I would say not only is this a great solution, but likely nearly a perfect one. As the difference between this and the original would be very hard to see. (That is if they make them long enough)

So why aren't British Hawker restorations using this method? Well, because 99% of these restorations are happening in Canada or England. Enough said.
 
Last edited:

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,110
Location
Thunder Bay
So why aren't British Hawker restorations using this method? Well, because 99% of these restorations are happening in Canada or England. Enough said.
Wait, is that because Canadians and Brits are perfectionists or because they’re idiots?

Incidentally, now isn’t the time but some day I’m going to try and dig up the rumoured full-scale Hurricane replica plans that have a properly engineered welded steel tube fuselage. It would have been designed to be compatible with original Hurri parts which unfortunately have been long sold off and it also may have been a two-seater (just a jump seat in back - not dual control).
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
311
A friend in GB posted this link for me on Practical Machinist:


Apparently the "T26" tubular rivet part is relatvely low tensile ("20ton") mild steel.

A stainless steel rivnut in my opinion
I drove 450+ of those things in the past month or so.
Mostly #10-24 & 1/4" - 20.
I doubt rivnuts have anywhere close to necessary shear unless filled with a tight fitting bolt.
Also doubt they make them long enough in the small sizes, but you could call.

Nonetheless, your note held a valuable hint!
It would be relatively easy/fast to make a simple, adjustable length, die set to pull-squeeze the tubular rivets with a typical large series pneumatic/hydraulic cherry or "pop" rivet puller. I made such a set to pull the riv-nuts, 1.)to save my aging wrists & 2.)to avoid the cost of the commercial spin-pull units. :)

(spin in this case does not refer to forming action, but to the capability of the unit to spin the rivnut arbor in the tool on or off at operator finger control, in a production setting. Rather than manually screwing the rivnut onto the arbor, then manually unscrewing the arbor aftter the rivnut is set)

smt
 
Last edited:

Bill-Higdon

Well-Known Member
Joined
Feb 6, 2011
Messages
1,685
Location
Salem, Oregon, USA
Wait, is that because Canadians and Brits are perfectionists or because they’re idiots?

Incidentally, now isn’t the time but some day I’m going to try and dig up the rumoured full-scale Hurricane replica plans that have a properly engineered welded steel tube fuselage. It would have been designed to be compatible with original Hurri parts which unfortunately have been long sold off and it also may have been a two-seater (just a jump seat in back - not dual control).
As I remember the main spar was made of nested hexagonal tubes, that'd be fun to make at home
 

Chilton

Well-Known Member
Joined
Jun 28, 2014
Messages
185
Location
Jersey, channel islands
However, in re-visiting the diagram of the assembly, I had assumed the "tubular-part-to-be-spun" was stainless tube. Cannot find a spec for "T26" except hints about welding it, which seem to indicate it is actually magnesium based? Anyone know?
T26 would be a tube in a low carbon stel similar to the American 1020 steel, T35 is close to 4130, the old British specs numbered tube and sheet material seperately even if the same alloy was used, so flat sheet would be S prefix, probably S1.
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
311
S80D seems pretty common, too?
In the hardened state it is rather soft. About 27 - 35 Rc. Very similar (in hardness) to a a familiar tool material in the prehardened state, 4140"PH". PH meaning "prehard" in this case, not the usual "precipitation hardening", as for some stainless steels like 17-4) So not gummy, easy to machine with HSS if the speeds are kept low. Listed with specific reference to screw machine parts.

 

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,110
Location
Thunder Bay
As I remember the main spar was made of nested hexagonal tubes, that'd be fun to make at home
Was Hawker doing that too? I know Bristol liked their tapered polygonal spar caps. For the replica Hurricane there was some talk of doing wood wings, basically PT-19 style structure but I never heard of that making it as far as an engineer’s desk.
 

wktaylor

Well-Known Member
Joined
Sep 5, 2003
Messages
389
Location
Midwest USA
This installation makes no sense... from a modern perspective. SIMPLE JOINT VIEWS... without context of the larger assembly... simply does NOT provide enough info [for me] to make sense of WHY this was joint was made this way. Also, although this joint may have been common for British Hawker... but there is no equivalent for WWII US wartime... that I'm aware of. Please give us larger context.

NOTE the jet I work on has some 1950s quirky assembly practices, that required specialized riveting and swaged-collar lock-bolt tooling for 'installs'... but nothing like this joint.
 

Riggerrob

Well-Known Member
Joined
Sep 9, 2014
Messages
2,316
Location
Canada
This installation makes no sense... from a modern perspective. SIMPLE JOINT VIEWS... without context of the larger assembly... simply does NOT provide enough info [for me] to make sense of WHY this was joint was made this way. Also, although this joint may have been common for British Hawker... but there is no equivalent for WWII US wartime... that I'm aware of. Please give us larger context.

NOTE the jet I work on has some 1950s quirky assembly practices, that required specialized riveting and swaged-collar lock-bolt tooling for 'installs'... but nothing like this joint.
Those pinned joints were a hang-over from World War 1 construction methods. The concept was extrapolated from wood and wire frames that were standard in WW 1 airplanes. Inter-war, they gradually shifted from wood struts to metal tubing, but retained internal wire bracing. Wire-bracing was gradually replaced by diagonal tubing braces. Pined joints were the simplest way to join all those tubes with rivets.
The disadvantage was all the small, precisely fitted pieces.
This was a result of British politics and British industrial practices. Fear of communism and industrial unions forced British politicians and factory owners to employ as many semi-skilled laborers as possible. Since high labor costs interfered with investments in tooling ... pinned joints became the norm instead of the welded joints used in American factories.

For a weapons example, compare the British STEN gun with the American M3 "Grease Gun" sub machine guns. The STEN had dozens of small, simple parts riveted together while to M3 was made of only a few pieces of stamped sheet steel welded together. STEN parts were made in dozens of small machine shops while M3 parts were stamped in the same large factories as automobile parts.
 
Last edited:

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
311
. without context of the larger assembly... simply does NOT provide enough info [for me] to make sense of WHY this was joint was made this way. Also, although this joint may have been common for British Hawker... but there is no equivalent for WWII US wartime...
There might have been advantages from the manufacturing standpoint.
If you don't weld the frames, all the individual tube pieces can be made absolutely identical with no fitting/grinding/birdsmouths that need further fitting as the fuselage squirms during weld-up. Unlike with a manually gas welded fuselage (in early days) the completed assemblies are all dead-nuts the same, if built in much simpler fixtures than welding requires. No post-welded straightening and annealing of clusters required. No metallurgical tests to be sure every welder is doing best process. As someone posted on PM, Brit manufacturing used to do just about anything to avoid welding. All your machinists and fitters can work with mechanical joints rapidly and efficiently. All the machinery is familiar, and all of them can make tools and tooling. Welding adds a separate layer of manufacturing complexity and different type of skilled worker.

The US embraced welding as soon as it was available, and did not mind tweaking and post fitting things. More a "git-er-done" attitude, and fix it later.

You can design for perfection, so rework and tweaking is unnecessary with skilled workers.

Or you can design for fastest production, understand that perfection might not be necessary, and assume a lot of unskilled labor can make it fit as the steps accumulate. Or design the underlying structure so it is adaptable, sometimes simply by adding the tabs and hard points in a different fixture at a later step & by allowing bigger holes, slots or similar fitting accommodations in the frame or the add-on parts.

I machine, and weld.
I'll bet that if you started out planning to make quite a few hundred of those airplanes, the riveted joints were faster. A few diesets, and few automatic forming stations for the tube ends, and every part arrives at the assembly jig dead same as the last. You can't look at that joint, even a complex one, and imagine it would be faster to weld it, than to merely insert parts and squeeze them. Semi-skilled labor doing the assembly. The parts that arrive at the assembly station took seconds to make in automated machinery. Screw machines and OBI presses rain (or at least drizzle :) )parts when they are running. etc.

Of course in recent years we are more attuned to any process including robotic welding in serious fixtures driven by the automotive industry coming out "perfect"

smt
 
Last edited:
Top