Hart Hind Bulldog Wapiti Atlas Siskin Hurricane & others steel testing

Your sample is part of the engine bearer assembly – starboard side, rear, engine support point. Typical of the longer nosed Merlin XX fit. The inspection stamp on the stainless plate is not clear (might be clearer on the other plate?) but the bolts look Canadian, so maybe Canadian MkXII. Please see PM.

Foray, thank you for the quick ID and PM answered, cheers, Ed

Really enjoying this thread.
The genius of these people never ceases to amaze.
Thanks for posting your findings P&P.

Andy

Thanks Andy, It’s good to know that somebody is getting value out of it. What happens is I do a bit of analysis in odd hours and then the data and inspiration build up in the head and I have to do a data dump on the forum otherwise my head will explode!

Sometimes when you examine a part and the light comes on you start to feel umbilically connected to long dead designers- certainly when my time comes to shuffle off I will enjoy walking into that ghost room where they all sit on club lounges reading newspapers and chatting away about engineering design battles lost and won.

Really enjoying this thread.
The genius of these people never ceases to amaze.
Thanks for posting your findings P&P.

Andy

This sample is unverified, as no knowledge of the Hurricane model or aircraft identification is known. It was described as a piece of wreckage from a Hurricane crash in Scotland. Certainly the piece came from the UK and is consistent with Hawker construction. The DTD 166 stainless fishplate has the part number B104964, and I hope that folk more familiar with the Hurricane design can verify this part number and the location of the remnant within the fuselage design. I understand that the Hurricane used 3% Nickel T50 tube in the fuselage, but this remnant reported tube composition results consistent with SAE 8630, about 1% Chromium and 0.4% Nickel. So the astonishing conclusion is that this piece of Hurricane crashed in Scotland is part of a CCF Hurricane built after early to mid 1942, when the US SAE 4130 standard was changed to SAE 8630 across the board. So here we may have a means for the forensic differentiation of otherwise similar UK and CCF Hurricane components that may have tube adhering to them.

Your sample is part of the engine bearer assembly – starboard side, rear, engine support point. Typical of the longer nosed Merlin XX fit. The inspection stamp on the stainless plate is not clear (might be clearer on the other plate?) but the bolts look Canadian, so maybe Canadian MkXII. Please see PM.

Strip steel Mosquito

At first looking at Mosquito components seems to have nothing to do with British strip steel construction technique. At first the sample below looked like a straight forging to me : I assumed that the engine bearer of the Mosquito would have to be a forging, simply from the application. What else would you trust to hold 1,000 horsepower to a fuselage?

Below is the engine bearer bolted to the timber spar of the Mosquito wing. One of the eyelets in this channel section must hold, by the time you divide the load, at least 100 thrashing horses. Ten times as much material is provided for in the towing hitch of a car to harness the same number of horses, so the metal had to be something special. But I underestimated the genius of the designers of the Mosquito.

[ATTACH=CONFIG]238705[/ATTACH]

[ATTACH=CONFIG]238706[/ATTACH]

This piece came out of the ground as a mass of corrosion; it was only until it was blasted did its basic structure become apparent. It is in fact a thin sheet metal bent into a channel section. Where the eyelets are that take the engine bearer there are provided two pieces of flat metal bolted to the sheet metal as reinforcing. What originally looked like a forging that required the use of forge tooling and a separate forging plant could in fact be made in a backyard shed. The piece started to glow with deHavilland genius !

Here was one of the fundamental structural members of the Mosquito design – that did not require a forging plant, when I understand that at the height of the Battle of Britain there was only one forge left capable of forming Merlin crankshafts! It was made out of strip steel. Now this particular fitting came from Australia, so I am not sure of UK or Canadian fittings. But the sheet formed into channel reported as SAE 4130 and the two hardened eyelets bolted to the sheet reported as Managnese alloy steel: consistent with DTD 138 65 ton steel. Again, the whole thing could be made in a back shed form sheet material.

It seems that every time you delve into the guts of a Mosquito component you find the product of inspired thought that whittled out ambitious performance from crude materials and processes. Surely the Mosquito was the Katyusha rocket of British design thought : careful not to upset the pretensions of Spitfires to strategic material supply while getting on with the job of cheaply and cheerfully dropping Cookies on Speer’s Berlin.

So this initial, splendid shock of realization about Mosquito load bearing steel strip structures made me look at the fittings that hold the wings to the fuseleage : all the vim of Leonard Chesire VC dropping pathfinder flares 100 feet over a bridge with 2,000 horsepower connecting to wings held on by only two steel fittings to the rest of the show. What had deHavillands done here?

Again strip steel formed in a backyard press into a sort of ladder section made by laminations of strip. As the section developed more strip was added to make one end thicker than the other where the obvious stress point were. Now each of these laminations was welded on the outside to the other, to make a solid mass on the edges. It was a wonder of simplicity, weldable 4130 sheet made into something that looked like a forging, probably in a sty shared with cattle and pigs. Then the sky opened, the clouds parted and a beam of sunlight came from the navel of Geoffrey deHavilland and hit my forehead : sweet Jesus son of Mary, I wonder if the bunnies crossed the grain flow in their sheets?

Now if you have ever torn an article from a newspaper you will realize that it tears easily one way but not another. As the paper is formed in the paper machine the fibres align with the flow of the felt in the machine, creating this affect. So too when a sheet of steel is formed under the rollers of a mill : the material is stronger across the grain than with the grain. So if deHavillands, who were the experts in cross grained plywood laminations, transferred this inspiration and made a sandwich of cross grained steel laminations, then they might have a cheap, easy to make load bearing structure that would be superior to a forging ! There is only one way to tell, which is to cut open the sandwich fitting and look at the grain flow under a microscope : if it is cross grain then I am going to work for deHavillands!

Keep those samples coming in.

XRF makes a living out of positive metal identification : PMI. It keeps the sons of the east, quick buck merchants and metallurgical pond scum honest. It does this instantly and cheaply, without requiring the destruction of the test piece. It is probably a good tool to stick on the standard shadow board in a restoration facility.

Material identification and quarantine has always been obviously important in aerospace. The following sample of verified Hawker Australian Demon aileron T50 spar below shows that even with all the care assumed in the sepia past, mistakes are made.

[ATTACH=CONFIG]238704[/ATTACH]

The original specifications for the tube in the aileron are 3% Nickel T50 tube. Both port and starboard aileron remnants were recovered from the same location in Australia, being 1935 manufactured items from the OEM dataplate attached. Surprisingly the values for the port remnant showed up as plain carbon steel, the same as an old clothes line or bicycle. This material could never perform like a T50 3% Nickel alloy steel or even T45 Manganese alloy steel : it had to be a mistake. Either the Reynolds tube company supplied Hawkers with the wrong tube in 1935, Hawkers made a manufacturing mistake or the aileron was repaired later by an indifferent RAAF fitter.

In any case the use of this remnant in a modern structure would be attended with the assumption that it was 3% Nickel T50, and perhaps an 80 year old mistake would finally manifest decades after it should have been found. The lesson for me is to be equally suspicious about both the old and the new. We didn’t invent stuff ups.

Below is a verified sample of Bristol Bulldog longeron.

[ATTACH=CONFIG]238700[/ATTACH]

It does not take a lot of sample to tell a lot of story, and this unpromising piece of rust has a lot to reveal. It reports as 5% Nickel – 2% Chromium – 0.5% Molybdenum-0.3% Vanadium alloy steel. All pretty good stuff for Grandpa to be fooling around with 85 years ago. Historic literature establishes that Bulldog longeron was made from Nickel Chromium DTD 99 material.
To understand DTD 99 you must go back to before 1935, after this it is extinct. It did change its name to BS S86 after 1935 but I will tell you a secret : remember the Nickel Chrome BS S88 Hurricane spars I talked about previously ? Well they have an identical chemical composition to DTD99-BS S86. Both materials came from the same mill : JJ Habershon & Sons, who supplied spar material to Hawkers and Bristols. In fact the total British strip steel school of design thought was based on the utilization of one form of Nickel Chromium alloy steel, cloaked in four obtuse and vacant faced DTD – BS steel alloy specifications. I will not confuse you with them, just remember Nickel Chromium alloy steel.

What is interesting in this analysis is that the sample reports 5% Nickel values, when original literature from the 1930s says it should have 3- 5% Nickel. In other words JJ Habershon & Sons have added the upper range of Nickel rather than the lower. The gentleman who I hire the XRF from has, in his principal custom, the verification of stainless alloys in food and pharmaceutical plants. Today’s urgency to cut costs means that stainless invariably comes from oriental climes and there is some justified concern in the use of food grade alloys in modern manufacturing plants. So the XRF is used for positive metal identification in process lines and after many years the XRF man can almost predict the mill the material came from by the composition. The key point is that the expensive European material is always above standard. There is always more alloy in the material than the specification requires, to remove doubt about performance. The oriental material is always just on standard, not a skerrick more of expensive Nickel or Chromium than is required. So its seems this venerable habit of over specifying compositions was alive at JJ Habershon & Sons in the 1930’s. I think this is a healthy thing to carry over into modern calculations concerning substitutions or reconstitutions of old formulations. A stitch in time saves nine.

Below are two samples of Hawker Australian Demon spar material made from DTD54a – BS 88c, the same material as used in the UK Hurricane spar.

[ATTACH=CONFIG]238701[/ATTACH]

[ATTACH=CONFIG]238702[/ATTACH]

Again, this is the identical Nickel Chromium steel alloy. This is not “stainless steel”, even though stainless steel has Nickel and Chromium in it, albeit at much higher levels. This spar material does rust, as the samples eminently prove. But, most interestingly, these samples sourced from Australia also report Nickel values at 5%, rather than the 3% that the original standard allows. Again, this material, as other Hawker Hind spar remnants with the JJ Habershon mill mark on them attest, came from the same mill. This mill shut in the 1960’s, when everything industrial in the UK started to shut down. I have a theory that the only difference between all the different DTD- BS designations around the same chemical composition of Nickel Chrome steel devolve from the mechanical or heat treatment of the steel strip within this mill. I dream of finding an old exercise book filled with the jottings of an old Habershons steel mill foreman. I have tried contacting historical societies in Rotherham, where the mill was located, to try and find this elusive retired mill foreman, but no luck so far. If anyone can connect me with such an apparition, I would be most grateful.

Below is, after much journeying, the most profound and simple connection to the spars of Hurricanes, Hawker Hinds and Bristol Bulldogs : a Rolls Royce Kestrel crankshaft.

[ATTACH=CONFIG]238703[/ATTACH]

It reports a Nickel Chromium chemical composition similar to our Hurricane and other described spars. So here it is : the high performance alloys developed in WW1 to cope with engine crankshaft failure were adapted in the 1920s in strip form to develop Hawker Hart and Bristol Bulldog steel strip structures. I could melt down a Kestrel or Merlin crankshaft and make it into a Hawker Hind or Hurricane wing spar or Bristol Bulldog fuselage. This was a high strength alloy steel that exhibited a certain elasticity that allowed it to survive millions of repetitions of cyclical loadings, whether as the crankshaft of a high power engine or the fluttering wings of a high performance aircraft in a dive. Know this and you can start to put steel strip construction back into the air. There are no SAE 4130 crankshafts in high performance aero engines, as 4130 cannot fully answer strip steel construction needs. Nickel Chromium alloys to the original composition, to the original culture of over specification, can. Unfortunately they are not available off the shelf. In inspired madness, a coil of the original material must be commissioned, which will meet the needs of all. This depends upon a different alloy within the human constitution : cooperation.

And so finally I could spend some time alone with the handheld XRF and do some instant chemical assays of various samples collected. As usual with a session with XRF it creates more questions than it answers. Not in the way of climbing the mountain and slipping backwards, but climbing and seeing new things from higher up. New and splendid things.

I am grateful for all forumites who kindly submitted samples for testing. This art of ‘forensic metallurgy’ will become increasingly important as spare parts dry up and new things need to be made to keep historic aeroplanes in the sky. The topic is utterly devoid of charm for most folk but most important. It is polluted with the worst form of technical jargon : the metallurgical codes of long dead technicians; hieroglyphics and incantations held in the bony fingers of ghost metallurgists spinning in the night mist. It is hard enough prying this knowledge from deep, forgotten dungeons of British engineering thought, and I dare not contemplate digging in the cemetery of French, German and US engineering thought for simple lack of lifetime. But I have a hunch the metallurgy was common under whatever flag it came and as loose, independent little explorations into material compositions are made the jigsaw pieces may come together into a rational whole. So I would encourage you to keep contributing chunks of metal or discovery which I will gladly test and share here or anywhere.

And so some facts, understanding that these are handheld XRF results, not traditional spectroscopy, and the thoughts that develop from them may be relied upon for conversation and dispute, but not for engineering.

CCF Hurricane.

Canadian Car & Foundry operated two steel works and a rolling mill. Being contracted to manufacture the Hurricane design it is self evident that production plans and material specifications were provided by Hawker Aircraft. The logic depended in part on the sourcing of Canadian raw materials rather than supply of metals and tubes from the UK, impossible in wartime conditions. So the obvious starting point is to assume that the Canadians used the same materials as the British, constituting these within their own steel mills. Now it seems that they did not.

Below is a verified section of CCF Hurricane X centre section spar. It is composed of four layers : two roll formed ‘octagonal’ booms with two telescoping cold drawn steel tube liners. I am not overly familiar with the UK Hurricane design, but my understanding is that in the UK the roll formed boom would be made from Nickel Chromium alloy (BS S88c) and the telescoping tube from 3% Nickel alloy (T50) tube.

[ATTACH=CONFIG]238695[/ATTACH]

The CCF section was unfortunately cut with oxy acetylene, allowing an uncontrolled heat treatment process that would make the grain structure and hardness values of the sample unreliable for testing, but the chemical assay would not change. I wondered why oxy was used to cut it. To prepare the sample I waited until the engineering workshop supervisor had left work and crept down to his beloved Brobo saw. The keen, liquid cooled blade made no impression on the Hurricane spar, so I proceeded to hang off the Brobo handle to make it bite. I like hamburgers, so the Brobo started to work. But it wouldn’t cut. Maybe the thin, rusty, ancient Hurricane steel was harder than the blade? I crept out of the workshop. The next morning the workshop supervisor reported that he had to buy a new $230 cutting blade, because the old one had become blunt, and I had to shake my head in sympathy and make clucking noises about “poor quality blades these days” while the expense form was approved. So here’s my tip : don’t use a Brobo saw on CCF Hurricane spars. I eventually used a thin stainless steel rated cutting disc with an angle grinder to cut out my chunks, all the time wondering about the spar alloy composition.

In the CCF section the roll formed octagonal booms are Manganese alloy steel, the same type of alloy used to line wear teeth on excavator buckets. I hypothesise that this strip material is DTD 138, 65 Ton carbon steel strip, based on the literature of the day describing it’s use in 1930’s aeroplane spars. I have not come across an application of DTD 138 before, but it seems that CCF chose to use this Manganese material as a substitution for Nickel Chromium BS S88c. I wonder if the gauge of CCF Manganese centre section spars is different to the gauge of UK Nickel Chromium centre section spars ? I wonder if the gauge or number of telescoping tubes are different? The fascinating thought remains that here is an original substitution that was eminently fit for purpose, that opens the possibility for an alternative approach to Hurricane spar replacement. It needs more work to directly compare the same UK and CCF section, and I wonder if any of the CCF engineering work remains to refer back to ?

In respect of DTD 138 it is an obsolete standard, substituted postwar by BS S517, also now obsolete. This composition was common to the contemporary Manganese alloy T45 tube composition used in the welded tube fuselage of the Avro Anson and deHavilland Tiger Moth. These compositions don’t disappear, they just come in different forms. But there is no shop stocking DTD 138 sheet today. But we can use this knowledge to carefully build cases for substitution using the materials available to us, as it seems the Canadians did in 1939.

I understand that CCF did not operate a tube mill. From where would it get its tube from in 1939 -40 ? My hunch is that it would be sourced from the United States, simply based on the proximity of scale, peacetime US tube supply and the foundries and mills of CCF being fully occupied. A further hunch relates to the use of US SAE 4130 tube and later National Emergency 8630 alloy for tube (NE 8630, later called SAE 8630). This was the familiar SAE 4130 composition with a little more Manganese and some Nickel. This created a weldable composition that could also be applied to non weldable pin jointed structures traditionally using 3% Nickel tubes.

The CCF telescoping tubes within the spar sample report values consistent with SAE 4130, lending weight to the theory of US tube supply. A further sample of Hurricane fuselage is shown below.

[ATTACH=CONFIG]238696[/ATTACH]

[ATTACH=CONFIG]238697[/ATTACH]

[ATTACH=CONFIG]238698[/ATTACH]

This sample is unverified, as no knowledge of the Hurricane model or aircraft identification is known. It was described as a piece of wreckage from a Hurricane crash in Scotland. Certainly the piece came from the UK and is consistent with Hawker construction. The DTD 166 stainless fishplate has the part number B104964, and I hope that folk more familiar with the Hurricane design can verify this part number and the location of the remnant within the fuselage design. I understand that the Hurricane used 3% Nickel T50 tube in the fuselage, but this remnant reported tube composition results consistent with SAE 8630, about 1% Chromium and 0.4% Nickel. So the astonishing conclusion is that this piece of Hurricane crashed in Scotland is part of a CCF Hurricane built after early to mid 1942, when the US SAE 4130 standard was changed to SAE 8630 across the board. So here we may have a means for the forensic differentiation of otherwise similar UK and CCF Hurricane components that may have tube adhering to them.

Why did the US change the ubiquitous 4130 composition to 8630 in 1942, then ultimately revert to 4130 in the current day ? My theory is that higher performance factors from a superior metal composition and a country roused to war by the December 1941 attack on Pearl Harbour demanded it. Postwar, as aluminium moncoque designs carried the performance development burden, only low stress designs like the Taylorcraft – Auster depended on welded tube, where plain 4130 would suffice.

I have also had conversations with veteran engineers in Australia concerning early problems with the 4130 welded tube fuselages of US designs such as the Wirraway and Boomerang, based on the North American Harvard design. In this case weld cracking was resolved by the use of new compositions supplied from the US to the BHP operated British Tube Mills in Adelaide, South Australia. I need to do more work to follow through this topic. Was this SAE 8630 and was this composition more stable for welding?

Again, loose and separate threads can connect in serendipity. The UK deHavilland Mosquito had Managnese alloy T45 tube specified in its landing gear. I have verified Australian built Mosquito landing gear tube that I can test for 8630 and I wonder if Canadian built Mosquito landing gear tube is made from 8630 ? The Canadians also built Harvards. I can compare this with Australain built Wirraway fuselage. Anyone have a verified piece of Canadian Mosquito undercarriage or Harvard tube piece that I can test?

Again, 8630 opens the possibility of an alternative pathway to 3% Nickel T50 tube substitution, except that SAE 8630 is utterly extinct! The real value is in the demonstration of the substitution of local materials within the engineering thought of the day. If modern regulatory and engineering thought can be as fluid as the original then it may be easier to keep sub economic numbers of antique structures flying into the future based on economic supply of modern substitutions. Knowledge of historic metallurgy can allow an informed parley between regulators and engineers, keeping at all times a respectful eye on the wallet of whoever is funding a flying resolution. Stripped of metallurgical jargon, these problems are otherwise relatively simple.

The concluding tests in these series were on further unverified Hurricane fuselage tubes sourced from the UK. In this case compositions reported values consistent with the expected 3% Nickel T50 tube.

[ATTACH=CONFIG]238699[/ATTACH]

It seems the Hurricane fuselage was made from 3% Nickel steel or, in Canada, from SAE 8630 Chromium- Manganese- Nickel – Moly steel. I do not think you can readily make a modern Hurricane with original Hurricane performance out of SAE 4130. Understanding why requires a passage back to an earlier, prewar time, to the Hawker Demon and Bristol Bulldog biplanes.

Treasures

Here are two things which raise my pulse these days :

[ATTACH=CONFIG]237350[/ATTACH]

A piece of rusty steel, not just any common corrosion, but a remnant of Bulldog fuselage vertical strut.:love-struck:
This piece is a mine of information for forensic metallurgy. A small slice can be embedded in clear, solid resin and then polished to reveal the grain structure of the steel alloy, a vital piece of information to assist in understanding its performance. Another small piece can be tested for its chemistry, to provide a confirmation of the alloys used in a known aerostructure, to provide a further dataset to support modern material selection for a modern replica aerostructure. Under the microscope, lost specifications for the type of rivet used to join the section can be confirmed. It is a remarkably useful piece. I would be grateful for anything like this from long lost aeroplanes like the Atlas or Wapiti to help build an understanding of the DNA of these designs. It can look even less spectacular than this piece, as long as it is identifiable to a particular type.

The other thing which excites me is this vintage Avery strip steel fatigue testing machine recently found.

[ATTACH=CONFIG]237351[/ATTACH]

Now we can take modern materials and test them in comparison to each other, to further validate some of the design theory behind strip steel construction. A piece of remnant 3% Nickel T50 tube can be extracted as a flat strip and compared to an identical piece of modern 4130 tube extract. The machine basically grips the test pieces at both ends and applies an cyclical load until the piece breaks.

[ATTACH=CONFIG]237353[/ATTACH]

The longer it lasts, the longer its fatigue life. Particularly good for flying wires and exposing things like the affects of thread cutting on flying wires. All this stuff was bread and butter learning for engineering students a few decades ago. These students then went on to write clever software to simulate this on computers and the pesky old analogue machines were scrapped. Today’s engineering student would probably just laugh at the fatigue tester and ask what it is. But there is nothing quite like watching a piece of metal break in front of your face as a counter clicks the cycles over and you sit rolling in a rocking chair watching it with a cigar and cognac.

Another very useful attribute of this machine is to apply a preload to the test piece, to see how it performs when the actual fitting of the material into a structure creates a pre stress that in some cases may equal service stresses.

[ATTACH=CONFIG]237352[/ATTACH]

An example of a pre stress is a fence wire pulled tight. The fence just sits there but if you apply a ‘service load’ in the form of a prize bull rubbing against it the wire may snap in comparison to a ‘looser wire’.
In strip steel aircraft construction, the steel alloys used are very resistant to forming. So if you have a poorly formed section that is then riveted together the structure can be loaded with a stress that significantly affects actual performance in service. Though on paper the mathematics say the part should survive a service stress, poor forming means it can fail in service. Another indication of this is twisting and buckling in the frame that has to be forced back into shape with jigs. The same problem manifests in welding as weld heat zones expand and distort around cooler zones. Gosh when I flick back through the card files in my mind I remember forcing metal into places it didn’t want to go, when I should have stopped and had a cup of tea and thought it through !

I need this kind of learning because when I was a 17 year old doing up a 1963 Nissan Cedric and I changed the crankshaft bearings over I never thought about checking bearing thickness, just popped them in. Of course somebody had put mixed the bearing shells back in the factory and the crankshaft refused to turn when it was all back together. :confused: So we towed the thing to the top of the hill in neutral then towed it to speed down the hill before I released the clutch and simultaneously my forehead hit the dashboard and the crankshaft loosened up. :p
So now I need to do things with a little more sophistication…

Watch the rotor baldes

A very interesting topic. We have a Cierva C30A frame at the Helicopter Museum which some are keen to restore but which has quite a lot of corrosion etc. I wonder what type of iron that is made of.I assume it didn’t need the strength and format to withstand the same stresses as a Demon etc but the C30s were built by Avro so …..

What a great project ! I do not know the composition of the fuselage but as you mention it is Avro built and contemporary of the Anson so may be welded T45 which won’t be such a problem to source. Metallurgical testing on a loose piece no bigger than your fingernail should confirm the chemistry of the tube and you can get a copy of the T45 standard for the 1930’s from British Standards.

Don’t damage the rotor blades because they will be very, very difficult to replace. From ‘Aircraft Engineering June 1931, pg 143 Seamless steel tubes for Aircraft by Austyn Reynolds’ (Reynolds Tube Co) :

“One recent example now being supplied comprises taper gauge tubes of 15 ft length for Autogiro blade spars. The thickness of these tubes is tapered for practically the whole length. These tubes are made from air hardened nickel chrome steel [ NB same as DTD54a/BS S88c aircraft spars] and are hardened and tempered to give 85 tons per sq inch ultimate stress.”

A very interesting topic. We have a Cierva C30A frame at the Helicopter Museum which some are keen to restore but which has quite a lot of corrosion etc. I wonder what type of iron that is made of.I assume it didn’t need the strength and format to withstand the same stresses as a Demon etc but the C30s were built by Avro so …..

Make it Happen

A divergence into T50 tubing from S88c sheet material for strip steel construction comes from a nagging feeling that this issue cannot be treated in isolation to strip steel construction; that 3% Nickel alloy tube was as much a part of the school of design thought as the Nickel Chrome roll formed steel strip in the spars. Today’s SAE 4130 tube comes from an entirely different school of design thought

There may be a middle ground in SAE 8630.

So here we have the tiniest, hopeful thread that may make pin jointed tube structures easier to restore in the future. If the original engineering allowed 8630, and it is weldable, you MAY be able to convince a mill to run commercial quantities of 8630 tube for absorption in the larger automotive racing market, allowing sub economic quantities to be diverted into pin jointed aircraft structures, while dealing with certification on the basis of a history of safe use.

Now that the T50 elephant in the room has trodden on my toe I have not been able to put it back in its box. Apart from the unique Bulldog, there is no other 1925 – 35 aircraft made entirely from roll formed strip steel. There was Boulton Paul roll formed tube, but its most significant commercial application was in Airship R101, and that is never going to be rebuilt. The principal application of strip steel construction was in aircraft spars. However the principal mass of these aircraft was T50 3% Nickel tube used in the fuselage, so the two materials, BS S88c and T50 Nickel tube, are joined in the Siamese fashion. The art cannot progress until the issue of T50 tube is progressed.

T50 is no longer available because it was obsolete once the last Hurricane or Typhoon rolled off the assembly lines. It characterised pin jointed tube construction from 1925 to 1945. It was not a weldable material because its alloy composition did not allow it to be successfully welded. It was a high strength material that was also highly elastic, which is the principal difference with SAE 4130. You may heat treat 4130 to achieve the strength values of old style T50 but at a cost to elasticity, as 4130 becomes excessively hard. It is said that T45 manganese alloy tube could be tested to achieve T50 strength values, and I believe it, but not the same elasticity.

So what does this mean ? Think of the bow and arrow. The bow needs to be made of a strong material, so it does not break when it is drawn back. The further it is drawn back, the more potential energy it stores for sudden release. If you have a strong, elastic airframe, it can absorb more energy as it is flung about the sky, without breaking. It’s almost like the fine, pliable, strong bones of a bird’s wing…

You can rebuild a WW1 flyer because you can get spruce and birch ply. You can rebuild a WW2 aluminium monocoque flyer because you can get alclad sheet. But you can’t really rebuild a T50 based fuselage unless you rely on carefully found 80 year old material or patches of 4130 that limit performance factors. Nobody is ever going to commission a run of T50 because the key attribute of pin joint construction was the ability to insert tubes of different diameters and wall thicknesses between each pin joint, according to the local stress requirements. Thus a Hurricane or Hawker Demon fuselage is composed of many, many different diameters and gauges of tube, not one simple tube dimension. A tube maker must tool up and economically run a tube dimension in thousands of metres of production, and you may only want 3 feet of a particular type.

I thought SAE 8630 might be a way to resolve this, but after consulting the tube schedules for Demons and Hurricanes there are so many different permutations of tube dimension that even this now seems too commercially difficult to address. Cue wailing and gnashing of teeth.

So let’s go back to basics and start from what we know and perhaps find a way to what we want.

1. Hawker Aircraft Ltd were not the only users of the pin jointed technique but found a formula through the Hart biplane family of being the fundamental suppliers of service aircraft in the 1930s. The pin joint fuselage technique relied on T50 alloy steel tube with 3.75% Nickel matched with wing spars using DTD 54a/Bs88c Nickel Chrome steel alloy roll formed strip steel. These material choices are common to Bristol Bulldog, Westland Wapiti and other types of the era. By virtue of the technique transferring to the Hawker Hurricane design and ultimately Typhoon and Tempest, T50 3.75% Nickel tube persisted as a characteristic material of these designs.

2. T50 tube was generally incorporated in a formula of large diameter and thin wall. Tube for welding, characterised by steel manganese alloy T45, was characterised by smaller diameter and thicker wall to allow weld purchase. Welded steel fuselages were evolved by Fokker and this school of design found favour in Avro aircraft, British licensee of Fokker designs and in the United States. In the US, SAE 4130 was an alloy that found application in weldable tube and US aircraft designs evolved using this technique rather than British pin jointed techniques. The scale of US industry and the uptake of welded tube designs in the aircraft and later automotive/racing industry saw the development of scale 4130 tube supply, while the demand for T50 declined as Hawker Aircraft, the last scale users of the material, exited from Typhoon/Tempest production. SAE 4130 can attain the same strength values as T50 3.75% Nickel alloy tubes, but not the same elasticity.

3. Using SAE 4130 in T50 designs is possible, with curbs that err on the conservative on original performance factors. Either the material is not hardened/heat treated to high strength values, to retain some elasticity, but more material must be added for strength, ie, tubes with thicker walls. In this case the weight of the airframe will increase. This may be deemed irrelevant as munitions/weapons systems are no longer carried in an airframe, but extensive and costly re-engineering is required. Alternatively thinner walled material is heat treated for higher strength values with the compromise in increased hardness, placing limitations on the stresses an airframe may be subject to, and again necessitating extensive and costly engineering. Unfortunately SAE 4130 tube is not supplied in the ultra thin wall thicknesses that T50 was supplied in. You may find the right OD, but not the right ID. A 4130 fuselage must always be heavier. You can’t bore it out, because you are taking strength away. The only way to avoid this is to use T50 material, but available stocks, whether within an airframe for restoration or salvaged from other airframes, are,at a minimum, seventy years old, subject to corrosion and service stresses that require careful testing and examination prior to use. Fully intact airframes are a rare commodity. Wouldn’t it just be easier to use new T50, like new birch ply or alclad ?!

4. Hawker pin jointed construction anticipated new alloys as an intrinsic advantage of the design approach. Thus in 1930 a fuselage could be composed of some T45 steel, lots of T50 steel and some T4 aluminium. It is not a design sin to therefore use titanium tube in a pin jointed structure, as much as sticklers for historical integrity will wail and pull their hair. So the hypothesis that CCF used SAE 8630 in CCF Hurricanes bears on this flexibility in the design philosophy, and the modern day allows for examination of other contemporary aerospace materials to see if they can provide the same performance factors. So far 4130, from an utterly different design philosophy, has established that is can be used with reservation, but what else is out there ? One modern application of thin walled, cold drawn tube is for the hydraulic ‘shiny’ used in high pressure hydraulic rams from excavators to forktrucks to aircraft landing gear. This would require a detailed sifting though myriad suppliers of commercially available tube and an equally detailed process of certification. You may end up matching 40% of all requirements through this method, or via other applications of high strength, modern cold drawn tube.

5. A major philanthropic effort to commission a conventional mill run of 20 permutations of T50 tube may be a circuit breaker, but this would cost a lot of money and leave much underutilised material after twenty or so fuselages have been serviced. Where would such money be better spent on historic aviation, if the ambit of such generosity is limited to this field only? Building a number of structures to house other historical aircraft rotting in the rain ? Keeping unique, intact aircraft like Vulcans and Comet Racers and Lancasters flying ? Ultimately such an event would require the engineering profession to admit a more elegant solution is beyond its wit. The original structures were the product of engineers grappling with a scarcity of funds, management and shareholder tolerance and the comforting blanket of applying proven technologies. It would be sweeter to take on some risk and try and turn pencil shavings into T50 tube.

6. Really we just need to make new T50 tube for Hawker, and other less known, extinct, pin joint, designs. Not one type, but a range of diameters. All tube types follow the same principle of diameter to wall thickness but the later monoplane types have larger diameters in comparison to the earlier biplane designs. All T50 tubes range from 3/4 inch to 2 inch diameter, a fairly small range. There is another key realization : all tube lengths are generally short, perhaps 2-3 feet in length. The only qualification is in the single length used in the tail section, from the rear of the cockpit to the tail. ALL pinjointed designs have a continuous length of tube in this section, generally no longer than 10 feet. So the manufacturing task is to make T50 tube in a tight diameter range in lengths no longer than 11 feet. Simple.

Here is the Hawker Typhoon fuselage and the Hawker Hurricane fuselage, with the key 10 foot length of continuous tube highlighted between the jaws of the calipers. Of course the hybrid pin joint monocoque structure of the Typhoon and Tempest requires only shorter lengths, but these are generally of the largest diameter.

[ATTACH=CONFIG]236133[/ATTACH]

Here is the Hawker Demon biplane fuselage, identical in principal to the later Hurricane.

[ATTACH=CONFIG]236134[/ATTACH]

Here is the forgotten Hawker Henley fuselage, illustrating the wonderful flexibility of the pin jointed technique to arrive at utterly different aircraft types from the same Meccano mix of components. To change the shape, change the length of the tubes. To increase the strength of the structure, use thicker walled or larger diameter tube.

[ATTACH=CONFIG]236135[/ATTACH]

Here is the Gloster Gladiator, after 1935 Glosters were owned by Hawker Aircraft, and utilised Hawker techniques and materials, even though they were old hands at pin jointed tubular fuselages. The key dimension is the single length of continuous tube in the tail section, 126 inches or 11 feet. This characteristic of pin jointed structures is common across designs from the Bristol Bulldog to Hawker Hurricanes, defining the maximum length of tube required. It is not necessary to run thousands of feet of T50 material, just 11 foot lengths.

[ATTACH=CONFIG]236136[/ATTACH]

Now we need to put our minds to a process that can run 11 foot lengths of T50 tube.

Test pieces

Attached are images of very useful material for metallurgical testing that would not otherwise be practical for any other use. These pieces of Hurricane fuselage were purchased on ebay, and while this is not a particularly scientific source, the information around the description is credible and the pieces themselves, after measurement, agree with the tube schedule dimensions for Hurricane I. Additionally, a remnant piece of stainless steel pin joint fishplate is consistent with the Hawker construction technique, featuring tube spacers, S80 hollow rivet ferrules and hollow rivets. A small included remnant of an aluminium instrument case includes a 1939 production date and the aircraft was reported to be shot down in 1940.
I do not have details of the excavation of this material, which would provide a lineal record.

[ATTACH=CONFIG]236040[/ATTACH]

[ATTACH=CONFIG]236041[/ATTACH]

[ATTACH=CONFIG]236042[/ATTACH]

[ATTACH=CONFIG]236043[/ATTACH]

One thing I would also love to get for metallurgical testing is a postage stamp piece of stainless steel (DTD 42H) from the longitudinal tubes of Airship L101, which crashed in France. From time to time a framed display comes up on ebay, but someone pays top dollar for bits of spoons and buttons and no doubt historic interest, but I just want a tiny chunk of the stainless !!

The Hurricane pieces themselves are 1 3/8 OD X 17 G T50 tube that should be 3.75% Nickel alloy tube, different from weldable T45 Manganese tube used in Avro airframes and ubiquitously for engine mounts. In matching these dimensions to the Hurricane I tube schedule, these could only come from the engine mount or cockpit area. It was reported the aircraft dove into the ground, so these may be from the cockpit area. Of interest is the remnant coating of stove pipe black enamel, overlaid by silver dope. I understand the standard finish was black stove pipe enamel for the tubes, but it seems as if the whole structure was overlayed with silver.

A divergence into T50 tubing from S88c sheet material for strip steel construction comes from a nagging feeling that this issue cannot be treated in isolation to strip steel construction; that 3% Nickel alloy tube was as much a part of the school of design thought as the Nickel Chrome roll formed steel strip in the spars. Today’s SAE 4130 tube comes from an entirely different school of design thought, the welded tube Fokker school. As Sir Winston Churchill described Ghandhi as a “half naked Fakir” so strip steel/pinjointed construction described Anton as a “mad Fokker.” These materials and their application in contending schools of design thought never met then and probably should not now.

There may be a middle ground in SAE 8630. It would be good to get samples of Canadian Hurricane tube to see if it was made from 8630, which is SAE 4130 with a bit of Nickel and a bit of Chromium. 8630 was weldable but also took on some of the attributes of fine grain structure that 3% Nickel T50 pin jointed tube boasted. The US aero industry switched from 4130 to 8630 in 1940 as a War Emergency Standard. Why would you make 4130 more expensive by adding alloys if it was already ‘good enough’. The hunch is that it was halfway between British T50 and Fokker 4130, the “best for the boys” and Canada would have relied on North American material rather than import scarce material from the UK to support CCF Hurricane production.

So here we have the tiniest, hopeful thread that may make pin jointed tube structures easier to restore in the future. If the original engineering allowed 8630, and it is weldable, you MAY be able to convince a mill to run commercial quantities of 8630 tube for absorption in the larger automotive racing market, allowing sub economic quantities to be diverted into pin jointed aircraft structures, while dealing with certification on the basis of a history of safe use.

So bring in those bits of CCF Hurricane for testing. Even better, who has the metallurgical information on CCF Hurricane structures and materials ?

powerandpassion

Bless you kind sir.
Mike

4. If, like me, you are in your underpants in the back yard holding a piece of glazing channel that your partner has given you three hours permission to form into a Lancaster, then I can only suggest the following :

a) get someone with a tig welder to tack weld a flat to the channel to form a closed section, later removable. You might try tack welding both channels to kiss each other for same affect, do both sections at the same time if the profile allows it.
b) fill the closed section with sand
c) make up a bending mandrel or former, being the inside dimension of the bend, out of strong timber or a pieces of steel.
d) experiment with a blow torch on the channel while bending
e) Clamp one end and work the bend around, allowing 3 or 4 feet of extra length to allow you to lever the shape around.

In further inspiration :

i) Buy cheap, low tensile, thin walled, welded steel tube of sufficient diameter to enclose the aluminium extrusion
ii) Place extrusion in tube, stand on end, fill with sand, close both ends of tube by crimping or welding.
iii) Go to any tube bender with standard mandrel for the tube size, ask them to bend your sand filled tube as you desire.
iv) Split the steel tube with an angle grinder, extract aluminium extrusion.
v) If you are extremely lucky, you will have a well formed bend in the extrusion. Experiment with short lengths until you get it right.

Indeed, “Flight of the Phoenix” ! Up there with ” Those Magnificent Men in their Flying Machines” and ” Chitty Chitty Bang Bang” and “Mosquito Squadron”.
Started all this off !

The Skysport Bulldog group made a rolling machine and rolled out pieces for the rear fuselage which for some strange reason dissapeared. Two different tubes were made. A 180^ and a 270/90^ tube. The early Bulldog tubes (as well as spar) were riveted together (on the longitudinal flange) and later had the edge folded over with the occasional rivet at a junction. In the case of the fuselage ‘tubes’, they were not round shaped, but ever so slightly egg shaped. Very smart. Very strong for that little bit of extra forming. The metal used in the rebuild was not the original call out. In the immortal words of Tim, ‘not even God has enough money to make this thing fly’. Thus the original premise was to make a flying machine, but money got in the way. I do understand that Guy Black’s group has made original steel (from Switzerland) for his Hawker series. We are talking mill run. Normal man translation: if you ask, you can not afford.

Ed, thank you for illuminating some of the achievements and tribulations of the Skysport rebuild. I wish you had a PA who could also help share some of the extraordinary stuff you are doing with your Bulldog. Go find some young aero tragic who can do some facebooking for you ! Part of the pleasure of the whole thing is showing what CAN be done without being ultra rich. I am fairly confident that you may have started rich, but your Bulldog has fixed that up for you now !

Money certainly helps, but not if it becomes an agent of discouragement to others interested in a subject. Hiram Maxim had money and built an aeroplane. The Wright Bros had no money and built an aeroplane. One of them worked ! I would certainly encourage anybody to look into steel strip construction, because it is under represented, and while the construction techniques were remarkable in 1930, they are unremarkable today and certainly within the capacity of any industrial community capable of making house guttering or a folding clothesline.

[/QUOTE] The U.S. Navy did a lot of testing with the material used in the Bulldog, and was not terribly impressed. This could very easily have been “not made here” syndrome. [/QUOTE]

I have found one US reference to the use of “British type high tensile strip steel” in a Curtiss flying boat, with a discouraging commentary on the difficulty material spring back posed for the builders. The key attribute of this material was its strength and elasticity, so spring back would have been discouraging for unsupported users. Certainly US steel was absent from the promotion of this technique in contrast to the integration of the UK steel industry with UK aircraft firms. It seems the US was more enamoured of Fokker type welded tube construction, which may have offered more volume sales.

[/QUOTE]Interesting comment about not being able to use 4130 today.[/QUOTE]

I wish I could ! Assuming the identical structure, going 1:1 on weight, I might end up with a hardened, brittle structure that would be prone to crack when manipulated through some of the strip rolled geometry. Or I could go for a softer, more flexible structure at double the weight. I may be entirely wrong and am happy to debate the point over a bottle of wine, or two.

[/QUOTE] This leads to the question of what size (and weight penalty) would you use today, and if you had period steel, would anyone pay the extra million quid for that satisfaction under the fabric and paint. Which is more important seeing the machine fly or know that it is 100% original? Knowing the acceptable substitute for todays rebuilds IS worth knowing, so yes Ed2, you are doing us a great service. Keep up the good work. Ed1[/QUOTE]

In trying to understand the period steel it has exposed me to schools of design thought and metallurgical alleys of excursion that I would have no reason to wander down otherwise. The further I wander in the less fearful it all becomes. I do think it is more and more possible. This does not subtract from the usual challenge of building any structure. There are plenty of kit builds and IKEA furniture that come in their totality but never get built. Finding the material ain’t the problem. Money ain’t the problem. If you can do what you have already done, nothing is impossible !

The Skysport Bulldog group made a rolling machine and rolled out pieces for the rear fuselage which for some strange reason dissapeared. Two different tubes were made. A 180^ and a 270/90^ tube. The early Bulldog tubes (as well as spar) were riveted together (on the longitudinal flange) and later had the edge folded over with the occasional rivet at a junction. In the case of the fuselage ‘tubes’, they were not round shaped, but ever so slightly egg shaped. Very smart. Very strong for that little bit of extra forming. The metal used in the rebuild was not the original call out. In the immortal words of Tim, ‘not even God has enough money to make this thing fly’. Thus the original premise was to make a flying machine, but money got in the way. I do understand that Guy Black’s group has made original steel (from Switzerland) for his Hawker series. We are talking mill run. Normal man translation: if you ask, you can not afford.
The U.S. Navy did a lot of testing with the material used in the Bulldog, and was not terribly impressed. This could very easily have been “not made here” syndrome. Interesting comment about not being able to use 4130 today. This leads to the question of what size (and weight penalty) would you use today, and if you had period steel, would anyone pay the extra million quid for that satisfaction under the fabric and paint. Which is more important seeing the machine fly or know that it is 100% original? Knowing the acceptable substitute for todays rebuilds IS worth knowing, so yes Ed2, you are doing us a great service. Keep up the good work.

Ed1

‘Flight of the Phoenix’. Many thanks, you’ve got my query right. I like your image in underpants with 3 hours to produce a Lanc. But really all that sweat etc is not for me. I’m the DO, the seat polishers.
Mike

Tough material, tough question

Asking the impossible – but why not!

Former E of the Lancaster sits between the nose section and the front centre section. While there are some other parts, the former is in two sections, both closed hoops joined top & bottom with short angle & channel parts. The front hoop is an angle section and is part of the nose. The rear hoop is an extruded channel section and is part of the front centre section. The two hoops are bolted together to join the nose & front centre sections together.

A good mate of mine wants to build this former but is in fearful trouble over the extruded channel. As he works the ally channel into shape, it becomes brittle and breaks.
Could this be avoided by a special ally alloy?

Unfortunately the original Avro drawings for the extruded channel itself are no longer on Earth. But I do have an Avro component drawing which suggests the material is SS.3075.
If this is the material code and a modern equivalent exists without costing the earth, then perhaps we are ok.

Is there the slightest chance you or anyone can help us with this.

Which top shelf brand do you like the best?
All the best
Mike

Mike, these are tough and easy questions. Easy : for brand I like Coopers, made in South Australia, served on a hot day by a smiling woman in a bikini, surely not too much to ask ?

Most of the metallurgical effort here is focused on historical steel, as aluminium is still a ‘current’ material, and there would be many more practicing engineers on this forum that could give you a practical response, if you shut the door on their coat tails fast enough !

If your friend is trying to make something fly then I cannot provide any advice.
If it is a static then I can only guess, in the absence of drawings and contextual information, some responses which I don’t feel too much faith in their usefulness :

1. If this is construction/glazing extruded channel bought in a hardware store it is supplied in the hardened condition. Any manipulation would cause the failure you describe. Talk to a heat treating shop about softening options, but you would need to understand the alloy you are working with.

2. Extrusion dies are fairly simple to make and for an economical sum you may find that an extrusion facility can make you any profile you like in any alloy for static purposes. If they are sufficiently engaged they may be able to help select an alloy and assist in forming. I assume you are trying to form the channel into the cross section of the cockpit shape, a big bending ask.

3. I do not know the Avro alloy you have specified, perhaps it is a proprietory specification, typically aluminium alloys of the day are British Standard L1, L3 etc or a DTD spec. The cockpit to fuselage joining channels would have to be very, very strong members, formed in the softened state and age hardened. This would be a high performance alloy and any modern equivalent would no doubt cost a bit.

4. If, like me, you are in your underpants in the back yard holding a piece of glazing channel that your partner has given you three hours permission to form into a Lancaster, then I can only suggest the following :

a) get someone with a tig welder to tack weld a flat to the channel to form a closed section, later removable. You might try tack welding both channels to kiss each other for same affect, do both sections at the same time if the profile allows it.
b) fill the closed section with sand
c) make up a bending mandrel or former, being the inside dimension of the bend, out of strong timber or a pieces of steel.
d) experiment with a blow torch on the channel while bending
e) Clamp one end and work the bend around, allowing 3 or 4 feet of extra length to allow you to lever the shape around.

If all else fails at least you can feel confident that you could participate in that schlocking film where they rebuild a crashed aeroplane to fly out of the desert, I can’t even remember its name…

Skysport

When the RAF museum Bulldog was rebuilt did they re-manufacture this tubing? Or did they create a simpler ‘look-alike’ structure where the original structure was too damaged for repair?

Tony, I do not know, but a forumite familiar with the excellent restoration undertaken by Skysport may be able to answer. From the display of the aircraft there is a diagram which shows original aircraft material and replacement material from other Bulldogs. From this I understand much of the structure is original, with sections from other remnants ‘spliced in.’

I would love to see if anyone has film footage in their attic from this Bulldog flying in the 1960’s