Cracks in Titanium Frame
 

Well, after 26 years I guess it's not a surprise. This morning I discovered two complementary cracks between the seat tube and the top tube on my Airborne "Carpe Diem".
This is not an emergency, though I'm not riding the bike for now to avoid making the crack propagate. It's a "fail-safe" thing; the worst that's likely to happen is the saddle and the top part of the seat tube could rotate. I had the same thing happen to an aluminum folding bike.
It seems I have 3 options:
1) Do nothing. The bike might last the rest of my riding life. I'm 75.
2) Drill "crack stop" holes. I did this successfully on an aluminum bike. Not pretty, but functional.
3) Pursue a repair. There is a shop that welds titanium close enough that I could ride there for an estimate. Sadly, I would lose some of the decals, which as far as I know are no longer available.
I'm inclined to look into option 3. Is there anything peculiar to welding a seat tube that I should ask the welder about? The only thing I can think of is that the inner diameter has to accommodate the seatpost.
Thanks for any advice and/or opinions.

https://cimg7.ibsrv.net/gimg/bikefor...249a0f496a.jpg
This is the non-drive side. There's a similar crack on the other side.


https://cimg2.ibsrv.net/gimg/bikefor...fbe00c57ff.jpg
The seatpost extends several centimeters below the crack; the "minimum insertion" mark is well inside the frame.

I'd bet the seattube will have to be reamed after welding to get good seatpost fit even if the weld does not go all the way through. (Distortion due to the heat.) If this welder isn't a framebuilder he probably does not have the appropriate reamer. A regular steel and softer metals bike mechanic might be unwilling to sacrifice his tools on the much harder titanium. I'd call a titanium framebuilder and ask his advice. Or just get it welded and try the seatpost. That crack is close enough to the the top of the seattube that the inside is accessible and visible so careful hand filing should work. Not recommended but there are probably more than a few steel frames out there with the same treatment.

Buy another old Ti frame for $300.

Getting it clean enough inside will be difficult. Even the amount of oil in a fingerprint can make a Ti weld brittle.
I hope your Ti welder dude knows how to purge the inside with argon. It's easy to end up with some pockets of air inside, even after letting the purge run for minutes before welding, so some though need to be put into where to input the purge and where to let it out.

When I was making Ti frames we did repairs, so I got to see the insides of some pretty expensive bikes with evident contamination inside from inadequate purge. One of the more common repairs was cracked DT on Merlins, right at the DT shifter bosses. I think they did the braze-ons later, and either didn't bother purging at all ("it's just a braze-on, what could go wrong?") or didn't let the purge run long enough to get all the air out.

On this bike, since you will be washing the inside with solvent and/or soap and water, that will go everywhere including inside the TT, through the purge holes. So drying the frame completely will not be trivial.

I don't know many Ti frame repairers so there are almost definitely good ones I don't know about. But the best I do know about is Dave Levy at Ti Cycles near Portland OR. He's been doing it for over 35 years, and he's smart and conscientious.

I found Ti Cycles HERE. Looks promising, though I will have to find out the logistics of shipping the frame anywhere. I don't yet know if my local guy knows about bike frames, but it's good to have a lead on someone who does. Thanks!

Back in the ol' days, this repair - like a Litespeed warranty repair - would involve the replacement of the top and seat tubes. You are probably just looking to get that crack welded back together and the inside of the seat tube reamed. My guess is that a Ti Cycles or Bilenky will charge you at least $600.

Airborne are reasonable quality Chinese made frames, which is why they were about half the price of US made frames and went for maybe $900 depending on model. After you get your frame fixed it will have a nasty 'scar', and it may develop another contamination crack in the future due to the same level of cleanliness that caused the first failure. Do you really want to spend double what you could buy an old US made Litespeed, Macula, Douglas or Mongoose for and still only have a Chinese Ti frame with an obvious repair?

Have you asked Airborne whether this is covered, or if you can use their 50% crash replacement deal?

https://www.airborne.bike/pages/warranty-policy

Dave Levy is awesome, he does excellent work and make some pretty great products. I had considered one of his Ren Yarak frames back in the day but sadly I didn't have the money to purchase and then was going to purchase a few months later and lost my job and took a new one that paid way less than I needed and then opened my own shop and that killed any budget.

The 1997 fatigue test report hosted on the Sheldon Brown site (12 High-End Frames in the EFBe Fatigue Test) includes a section where some of the frame failures are shown, with accompanying descriptions. Here's the photo and text for the Merlin frame that was tested.

The myth of indestructibility collapses: Relatively early, considering the price and the high expectations, a crack spiraled around the down tube of the Merlin Team Road titanium frame. Point of origin: the small weld at the shift-lever boss.

https://www.sheldonbrown.com/rinard/EFBe/tour14ab.jpg

If it comes from a ti weld, it ain't fatigue.

No, because Airborne turned into a couple other companies (Flyte and later Van Nicholas). But I'll look into it anyway. However, as this is pretty much "normal wear and tear", I wouldn't expect it to be covered:
https://cimg3.ibsrv.net/gimg/bikefor...3c0e15e52e.png

I may, after all, just drill some holes to stop the cracks from propagating. I'm not worried about the appearance, as I've certainly gotten my money's worth out of this bike.

It isn't wear and tear. It is the result of a unclean weld - a defect. And it occurred at a relatively low stress part of the frame.

However, I would hope they would extend the courtesy of a crash replacement. Their current rim brake frame would be around $1000 at that discount - which ain't bad compared to repair prices.

This crack is not the normal failure mode I see in this area, which is normally a crack on the front of the seat tube just above the top weld joining it to the top tube; It's a heat-affected zone, plus stress concentration if the weld is not smoothly ground and blended to a perfect fillet (like my old Cannondale), and the tube is loaded in tension (see below), which favors fatigue cracks. But no.

This crack is near the neutral axis in bending of the top tube, but not due to bending of the top tube per se. With rider weight on the seatpost, due to the seat tube angle, you have aft moment, the taller the seatpost (and greater the rider weight), the greater the moment. This is a particular problem with folding bikes and ULTRA long seat posts. And this pulls aft on the top half of the top tube, and pushes forward on the bottom half of the top tube. And in the middle where they meet, you have pure shear. That's what this looks like to my FEA eye. It's not a common failure mode, but does happen. When a tube is under excessive torsional load, it develops a spiral crack, because the shear loads are at 45 degrees to the long axis of the tube. Combined torsion and bending will cause a crack angle between 45 degrees and 90 degrees to the long axis.

A weld repair may work but would eventually have the same failure mode over time, perhaps a long time. If I had to repair it, I'd give long thought to a gusset or patch over the welded section, in a manner that would reduce shear stress for the load, not overly stiffen the area with regard to the rest of the area (so tapering a patch like good lugs instead of rectangular), and of course be reasonably attractive.

That said, the above sounds a lot more expensive and less durable than buying a replacement titanium frame, as others have suggested. YMMV.

I agree to buy a replacement frame would be a more sound idea.

Technically true. They should have named it the "Likelihood of failure within a lesser duration of use than would reasonably have been expected" test.

I didn't know that. The only Van Nicholas I've seen is this one, where the whole head tube fell off:
https://cimg7.ibsrv.net/gimg/bikefor...5bbf827fd7.jpg
https://cimg9.ibsrv.net/gimg/bikefor...8983408b66.jpg

I don't see any crash or car roof-rack damage, so this may have been JRA. Not my bike, not my pics.

WTF are you talking about? Ti frame welds are never ground down, and Ti frames almost never fail in the HAZ.

Hmm, I've seen plenty of Ti frames with smooth/polished welds (Arctos for example, or the ones I made), and I've seen plenty that cracked in the HAZ. In fact I'd say most cracks start in the HAZ though they may propagate elsewhere.

Never say never! Oops.

Whoa, where'd you go to engineering school? Ti welds can't fatigue?

Don't rely on this guy for technical experise!

Don't be silly. You know very well we are talking weld contamination failures on Ti. A bad weld doesn't have anything to say about the general fatigue life of a class of bike frames in general.

And if you are too cool to ever reply to my posts to you, you can put a sock in it with this sort of commentary. Especially if you are the one who misunderstood.

The vast majority of Ti frame cracks start in the weld from contamination. I have looked at hundreds and hundreds of Ti failures and they almost always transect the weld, rather than run parallel to it or fail to cross the weld. Like the OP's. The Van Nicolas you posted might be a HAZ failure - that's the first one like that I've seen. I have seen one Ti tubing failure - well away from the HAZ or any welds.

But the point I was making to Duragrouch is that it is not a normal practice to grind down titanium welds - certainly not to decrease stress risers as he implied. (Nor are the vast majority of aluminum bikes, either. Including current Cannondales and the last of the Kleins.)


Why don't you read the thread, follow along and stop trying to post gotchas? The OP's bike failed, like most cracked Ti, from a polluted weld. The crack started in the weld and radiated out from there. It didn't happen because the welds weren't ground down, it didn't happen because of heat damage to the tubing and it didn't happen because the frame was ridden to some point of metal fatigue.

When I referenced in my post about post-dressing the welds, you'll see I was referring to, not this particular failure, but the normal failure seen in this area of the bike, which is on the high seat tube front, above the weld to the top tube, and referencing aluminum, where it is common to have a fatigue failure at the toe of the weld if it is particularly convex and not a smooth transition to the tube, so a natural stress concentration. Again, this is referencing the more common type of failure in this area of the bike, which I concluded the paragraph by saying, "But no."

But yes, all the titanium welds I have seen are lovely concave with smooth blend to the tube, without post-dressing.

The failure in this case, on the side, very near the neutral axis in vertical/longitudinal bending of the top tube, I propose is due to transverse shear in the top tube due to the aft moment imposed on the seat tube by the seat post. It could also be due to a weld contamination issue, but would seem peculiar initiating in that exact location, unless the entire weld is suffering from contamination, and it just failed in the highest stress location, due to transverse shear on the tube. And noted in the OP, it failed at that location on both sides of the frame. The odds? And as I said, in a way, this is not much different from a spiral torsion failure in the tube, as that would be along the transverse shear direction in torsion, but usually initiating at a point for whatever reason.

Ti weld contamination failures are not enormously connected to stress. You can drill a water bottle hole in a butted Ti downtube and it will never fail. But a tiny contaminated weld in the same spot and the tube will crack. I don't pretend to know the mechanism, but it behaves almost like the way oxide destroys metal or lamination from the inside. Clearly the work cycles on the metal are a factor, but the results are always much more spectacular than cause would suggest.

Which is totally different than how even the thinnest alloy steel behaves under similar or worse circumstances.

My guess, and only a guess, is that what you describe may have an effect at the metal grain boundaries, initiating a sharp crack, whereas the drilled hole you describe has stress relief. Sharp cracks, they move.

The seat tube crack on my (previous) 4130 steel frame, I didn't see nearly early enough to mitigate by drilling a hole at the end of the crack. The crack did not happen due to edge of weld, but started at the inverted T-slot at the front of the seat tube (most bikes put that at the back of the tube where loaded in compression instead of tension like the front, tension stress feeds fatigue), and once started, progressed all the way to the back of the seat tube and into the seatstay weld, and started down vertically, both sides of the seat tube, before I noticed, because I fold my bike so seldom. Early plastic seatpost bushing didn't help, now they are aluminum. Frame was toast. With the replacement frame, I watched for that crack, and when it had barely just begun, filed out the crack and work-hardened the surface with a tapered knife honing steel to introduce residual compressive stresses that counteract tension (like shot peening), and that stopped it.

I'll have to read up on titanium alloys and failure modes.

Thanks to everyone for the suggestions, advice and caveats. I think I'll contact a few repair places for estimates. Meanwhile I'll disassemble the bike and clean the seat tube and inspect the other welds.

Repair seems most desirable to me, if it's not a bank-breaker. Even if it costs the same as a replacement frame, I wouldn't have to source any new components: brakes, fork, headset, drivetrain, etc. I'm very happy with the bike as it is, and I'm not concerned about welding "scars".

A replacement frame would, ideally, have V-brake mounts, the same O.L.D., seat tube ID, and size. That would be like finding a needle in a haystack, I think... though I am not familiar with the pre-owned and NOS frame market.

A new bike is probably the least-likely solution for me. I built the wheels on my bike, and the set-up is exactly what I want.

Meanwhile, I have a nice old mountain bike and a vintage 10-speed I can ride.
Once the cards begin to settle, I will report the outcome.

https://www.bikeforums.net/classic-v...e-headset.html


How tall are you?