Old Triumph Motorcycle Frame Repairs

I was asked to take a look at breaks in the frame of a 1971 Triumph Bonneville by the guy riding it. He knew I am an engineer as well as doing some machine work and welding.

The guy picked up the seat and the damage was obvious. The rear portions of the frame tubing run horizontally, and have the rear shock mounts welded to them. Just behind the shock mounts, the top of the frame tubing kicks up slightly and then makes a horizontal 180 degree loop, effectively closign the rear ends of the frame.
The tubing had broken through just behind the shock mounts and the rear loop of the frame tubing was held in place by the rear fender. An additional light crossmember or two further forward on the frame (ahead of the fractures and shock mounts) held the fender as well.

I told the guy riding that Triumph he should have parked it and stripped it for repairs to the frame. He claimed it was riding season and he wasn;t pushing the bike hard, running it locally. I wouldn;t have been out on the road with a motorcycle with that kind of frame damage, but that's my opinion. My own belief is there is flexure of the frame in that area caused by the shock mounts putting concentrated loads into the frame tubing.

My ideas for the repair are to turn a couple of steel pins to drive into the open ends of the frame tubing, spanning the fractures. I would bevel the ends of the fracture for a CJP (full penetration) weld, letting the root passes burn into the connector pins. Welding will be done using DC GTAW.

My questions are: has anyone any knowledge of what alloy steel Triumph made their frame out of the 1970's ? Was it a chrome-moly steel ? Any ideas as to filler metal to use ? I've got DC TIG with lift start, argon shielding gas, so should be OK for the actual welding. I will try to get a piece of 4140 or 41340 round stock to turn the connector pins out of.

I believe the fractures occurred at what I call "inflection points" on the frame- the frame tube laid horizontal as it ran over the tops of the shock mounts. Just aft of the shock mounts, the tubing "kicked up", making an inflection point. At the same time, the rear shocks were putting a concentrated load, as well as some "side-to-side" flexure into the frame at that point. It is my thinking that by going to fitted connector pins accross the fractures, I stiffen that whole area and the repair welds are not so critical as the pins carry the loads. I will turn the pins and hot-bend them as required to fit the frame so there will be good bearing contact before I weld things together. I do not plan to turn any kind of "neck" or groove in the pins at the weld root, just leave 1/16" or so as a root-gap & burn the root pass into the pin.
Has anyone done a similar repair ?

Thanks-
Joe Michaels

The pin idea is perfect. That is how repairs to tubular members is usually done. Usually instead of welding through to the insert at the butt, in which you risk creating a fracture zone around the pin, holes are drilled in the tube and plug welds are made into the pin.

Considerable controversy exists regarding TIG welding of 4130. The AWS says don't. 4130 was designed by the aerospace industry for oxy-acetylene welding. Tig welds in 4130 are too heavily quenched and result in a brittle HAZ. However the racing industry has report success in welding 4130 with mild steel fillers and has petitioned the AWS to create a standard calling out that practice.

starbolin

I have fixed and modified many frames over the years that required sections of tube to be joined. Usually use a piece of tube, doesn't need to be a solid bar. Drill holes in the original tube 3/4" to 1" back, if possible, from the joint before inserting the connectors, plug weld the holes to the inner tube, and weld the joint burning into the the inner tube. This type of repair hold up fine. A lot of the motorcycle frames flex quite a bit, if you use a solid pin it may make the repaired area to stiff and it will only crack somewhere else.

Ed

Joe, you are an engineer and I am a carpenter; but with that in mind here are some notes.

First, as long as the straight channel section just forward of the shock mounts that spans the top of frame is in place & the (weld) joints have good integrity, the outboard loop you mention does little more than partially support the rear fender. Not to downplay the hazards of having a fender come loose and get in the wheel; but that area of the frame does not account for much as far as handling, integrity of the running gear, etc. It "merely" partially holds on the fender, and supports the rear of the seat pan to prevent that from cracking.
If the owner is running 2 up at any time, there is some slight hazard, or more likely a discomfort to the passenger from the seat pan beginning to crack, becoming floppy, and sloping backwards.

Is the crack behind the seat hinge pivot? You may have to modify your splice pin on that side to accomodate.

So, the engineering problem becomes, are you satisfied that the repair will support the rear of the seat pan considering the normal sized passenger the owner rides with. Second, will the repair you propose so stiffen the area that the inherent British parallel twin vibration will cause the frame to crack in another area, such as forward of the shock mounts, or at the thin wall large diameter spine frame that does double duty as the oil tank?

I have seem these completely lopped off back of the shock mounts for "Bobbers" and flat trackers with a solo seat and doubt there is any effect to the integrity of the frame as a chassis. I went down cellar to check that my mind was on the same area you were describing, and was surprised to see that one of my 71's bought in a shipping crate as a project had a (very professional looking) weld repair on one side only in exactly the area you mention. The other side had an over sized bead on the hinge, but that was it.

As far as welding, I would torch weld it, and torch normalize the area, similar to standard aircraft welding technique. In an aircraft, an oversized tube would probably be put over it, with a fish mouth weld at each end to spread the stress out. Pins inside would certainly be more than adequate as you describe. But again, there can be unseen penalties for over stiffening local areas of a structure. I do not know what alloy the frame tubes were during that era, Traytop Johnny might.

BTW, if he has frame cracks/failures in an Oil-in-Frame model, do a thorough inspection of where the tubes connect to the spine tube, and at the bottom where the sheet metal connections locate the swing arm pivots. Those areas have been known to crack, and they create far more serious issues.

It will be very interesting to hear how you do proceed with this one, but it should be a relatively minor repair.

smt

Are you sure the Triumph frame is 4130? I can't picture them using that in a production Bonneville. If they are, the thickness is way overdone. The ones I worked on seemed to be plain old 1020, in which case ordinary TIG or even MIG will work fine. Even if it is 4130, standard race car practice includes TIG with ER70-2 rod, no preheat. 4130 joints needing maximum ductility are done with 308SS rod, preheated to 300°F.

Rather than pins, I would sleeve it with tubing so there is not such a major cross section change at the repair, so that it does not break at the ends of the pins in the future. I would rosette weld the sleeves on both sides if easy to do so, with one pair of rosette welds at 90° to the other pair, again if easy to do so. Weld the tube in 4 passes, one in each quadrant, 12 o'clock to 3, then 6 to 9, then 6 to 3, then 12 to 9.


I concur with smt that the end of the frame does not add much strength, having seen lots of bikes with that portion removed. I suspect the breakage is vibration induced, aggravated by the design of the frame. The triangulation stops right in front of the break, at the shock mounts, correct? That would concentrate the vibration 'whip' right there and fatigue the tubing. Bike with bobbed rear fenders seldom broke the frame there IIRC, the bobbed rear fender would reduce the vibrating weight reducing the load on the frame extension.

Aahhhh, the joys of Triumph frames, cracking there we always put down to vibration, being Brit you could say I've done a few! and have found this method to be the most reliable.

I'd sleeve it with tube (preferably of the same wall thickness, 45deg prep the broken ends and bronze weld it with OA, (preferably using nickel bronze rod for steel)

Bronze avoids metallurgy mis matching problems, did quite a bit of welding for a frame builder when I was a kid, he always bronze welded chrome moly.

We also repaired Yamaha TR & TZ racer frames with bronze, the high frequency vibration usually cracked the tube just before the actual TIG or MIG weld - in the HAZ.

2 cents from where they came

Take care. Sami.

Oh yes, bronze. I only did that once, repairing suspension bits on a british formula car. Carroll Smith expounds at length on it in Build To Win (I think), says it's the only sensible way to join chrome moly.

Perhaps TIG with Everdure?

I dunno, bronze is absorbed into the steel if overheated and makes it less ductile and hard to repair again. But in books/magazines the Brits did commonly go on about using "sif-bronze?" for all sorts of joints, even butt joints, especially on alloy tube. It can be used to make beautiful joints. My exposure to it was in body panels - it works fine for a repair, but if the same panel gets creased and needs repair in the future, it almost needs to be cut out anywhere the bronze flowed to get back to ductile metal. That sort of turned me off on it compared to hammer welding which is somewhat easier anyway.

Many aircraft frames are (were) commonly oxy-acetylene welded up out of 4130 thinwall and I believe it is still the preferred repair method. Mild steel rod is used, and since the weld bead is thicker than the tube walls, weld strength tends to be better than the parent metal. The mild steel prevents hardening and does not add carbon to the HAZ which could make the tube brittle. After welding, the clusters are heated with a larger tip or rosebud to dull red and the flame played out slowly on each cluster. For years after TIG was introduced, this (Oxy-acetylene)was still considered -the- superior process if furnace heat treat was not possible for the complete assembly. There are alloys and processes that now make TIG favored for production in most cases. It is faster when welding varying thickness/size parts. But it is not automatically or necessarily _better_ than O-A & mild steel rod.

Phil Irving points out in "Motorcycle Engineering" that in the late 50s/early 60s when he wrote it Reynolds "B" quality tubing was the most common for production frames. "B" wasn't a good/bad quality designator though. The B tubing was a "low-carbon, stable mild steel that can be brazed or welded by any method". Sounds like 1018/1020 pretty much.

I can't imaging that Norton, Triumph or any of the other manufacturers in the 60s/70s were using 4130 - they were too damn cheap to buy fancy tubing like that (which wasn't needed in their applications anyway). Reynolds 531 was used on racing frames. 531 is a manganese molybdenum steel. T45 has also seen use in racing frames and is a carbon-manganese alloy.

Brazing is generally not recommended for 4130 due to the potential of intergranular penetration, but I certainly know of 4130 frames that have been bronze welded/fillet brazed without failure. I've got a reference that hypothesizes that a low aluminum (used for "killing") content 4130 will have more problems with the increase in grain size as it lets it grow continuously with temperature. A relatively high aluminum content (and keep in mind this is still a very small percentage) "pins the grain boundaries and prevents growth up to about 1050C. This is above the temperature reached by skilled bronze welding. The grain structure therefore remains acceptably fine at bronze welding temperatures whereas that of low aluminum 4130 does not."

Even 531 can be subject to the cracking if overheated during bronze welding.

If it were up to me, I'd just do the sleeve/rosette weld using 1018. I'd have no qualms on that application with TIG, O/A fusion or bronze welding, whichever seemed most appropriate, was most available, and the weldor was most confident in using.

cheers,
Michael

Thanks to everyone for the learned and experience-based responses to my inquiry. I am going to go with making my own "tubes" to do the repair. Since the tubing on the rear section of the frame is not all that big, nor do I need much of a repair piece, I will simply turn some mild steel bar stock to size and drill/ream it to create "tubing". I figure reaming is in order to minimize tool marks and the stress risers they create- given the vibration and flexure that rear section of the Trump frame sees.

My gut feeling when I first saw the job was to use the old tired-and-true oxyacetylene welding. It's what I learned many years ago and what I still use for occasional oddball jobs on thin or dirty work. I do have TIG in my shop nowadays, so was tempted to go with it as opposed to oxyacetylene. However, I have this kind of "blacksmith" sense in me that has me often using the oxyacetylene torch flame to anneal or at least post-heat O/A welds on jobs where fatigue failure or high stress areas have been repaired. Thanks for bearing out my own sense of the thing !

I'll admit my ignorance here: what is a "rosette weld" ? Is it a plug weld (made by drilling a hole in the outer tube, bevelling the hole, and then filling with a weld that burns into the inner part) ?

I'll go with the idea that Trump did not use anything exotic for their frames, so will go with the idea it is a plain low carbon steel tube vs. chrome moly. I've got some 70 ksi TIG filler rod in my shop, and that is what I will use.

Meanwhile, I told the owner/rider he was Ok in riding around with the frame in its present condition so long as he does not get to riding two-up. He's wanting to defer the repair job until the fall, when the riding days are almost done, and I cannot blame him there.

Thanks again to everyone. It is really special to have this kind of forum with people contributing the right mix of practical as well as metallurgically-based expertise.

With best regards and thanks-

Joe Michaels

Nuts - just lost my reply to an internet glitch.

Joe - yes, that's a rosette weld. On thin tubing (.065-.095)w/TIG I generally drill a 1/8" to 3/16" hole w/no countersink, and the welding fuses a bit of the outer tube into the weld leaving a 1/4 to 5/16" weld. It sounds like you know exactly what to do, good luck.

Gooday to everyone
I can speak to welding a Triumph frame. The material is not 4130 but made from mild steel. They only used 4130 in production racers. Slipping a goesinto tube into the existing frame is the answer and then weld. These usually break from people carrying to much weight without a proper rack. Good racks are hard to come by. Nortons also craked do to the length and this was corrected in the later frames with a gusset at the shock mount on the inside. Hope this helps.