After reading htis thread today I checked all my forks and stems regarding their insertion past the threaded part of the steerers. I only found one that was a concern, it has a lot of thread below the headset and a slightly shorter shafted ITM Eclypse. I have it as low as it will go and it is slightly overlapping the threaded area. So that one I will be changing the stem out when I can find a suitable one in the same reach. I found another that I just need to keep slammed (same shorter ITM Eclypse type) and the wedge is fully 20mm below the threaded part.
So I am very glad you are ok, and thanks for bringing this kind of thing to my attention!

Maybe the failure comes from the threads being cut, instead of rolled on to the steerer tube. But I don't think threads on steerer tubes can even be rolled on....or could it??

Rolling threads on a tube can be done, of course.
You need a removable plug or roller inside the tube so it won't collapse, and the rolling mechanism has to be hard enough and stiff enough to deform the metal correctly.
But I've never seen a steerer tube with rolled threads.

Typically the threads are cut. The die looks like a pipe cutting die. The depth of the threads start out being small and shallow and as it progresses up the threads become bigger and deeper until they reach the full thread size. The length of this taper varies. The taper is there to spread the cutting load to many facets of the die, but it also has the benefit of creating a transition from the uncut portion of the tube to the full thread. Thread cutting lubricant must be used when cutting threads in steel tube. Machinist and mechanics have their favorites, so we won't worry about which one you prefer to use. The important point is to use a lubricant.

after a quick glance at some old related threads / search - it appears some steel steerer threads are rolled

( but Sheldon Brown ‘never have I seen one that appeared to have rolled threads’ )

hmmm

one thing that was not previously mentioned - gotta be careful to not apply too much force / torque to stem expander bolts

I’ve never been a fan of quill stems with expanders - yeah I know they’ve been around since the Lord was a boy - and I have a number of bikes with quill stems - but I much prefer threadless system

I measured both the Salsa stem and my current Salsa Moto Ace bars and both are 26.0 clamp diameter. Lucky me.


It's looking more and more like I'm going to take this opportunity to upgrade to threadless. I just confirmed that my existing bars and extra Salsa stem play nice together and I've got anther stem in the mail with the correct dimensions. Also ordered up the 1" spacers I'll need. All that's left is a 1" threadless headset. Turns out that's the easiest component to find so all I have to do it pick the one I like.
This does mean I'll need to pull the freshly wrapped bar tape off at least one side and remove the shifter and brake cable housing since the stem is old school and is the one pinch bolt style. One of the threadless stems I have is the newer bolt on face plate style for the bar clamp which will give me much more freedom in swapping stems later should I choose.

Sounds like you are all set.

John

Good news. As I suspected, the local frame builder can repair it no problem and can also convert it to threadless, which I've decided to do. No idea what the time frame will be like on getting it back or how much it will cost but neither is a particularly big deal. I'll get it when I get it and pay however much he wants. In the meantime, I can collect the last couple of parts I need to do the conversion, take the bar tape and shifters off and install the new headset races in the frame.

If you were a cheap bastard (not saying you are, but I am), you could make your existing headst work by removing the threads from the upper screwed race. A burr (rotary file) in a dremel does it well, just grind metal off until the threads are just barely gone, everywhere, checking the fit when you start to get close. The bottoms of the threads are a great visual guide to how much metal to take off.

Me, I do it on a lathe because I have one, but the folks who've done it shade-tree mechanic style say it works perfectly. I can do the thread-removal for you on my lathe but shipping both directions will cost about as much as a new economy headset, plus you end up with a used headset, maybe even well on its way to worn out. The 'threadectomy' operation is only worth it if your current headset is good quality and not very old.

bulgie I'm sometimes a cheap bastard but mainly I'm thrifty. I like to make stuff work that I already have on hand. That's a good idea you've got there on converting a threaded race to threadless. I don't have a lathe but a friend's dad does. It's a tempting idea but I kind of want to save the whole headset for another project or another point down the line for the day when 1" threadless headsets are too impossibly hard to find. It's the original Shimano 1055 and is in great shape with only extremely minor indexing in the races, minor enough that there's no way to feel it while riding. Worth saving and potentially worth machining but for now, I'll go with a good used or NOS unit. Got my eye on a not cheap (but cheaper than Chris King) NOS Dura Ace 7410 headset. Some very functional bling.

I'm a little surprised this happened, tbh. Back in the '70s, we used to get replacement forks that were meant to fit every frame possible and thus had a very long threaded section. Thankfully, I don't recall ever seeing a broken one. Of course, it helps if you have more stem insertion than OP did.

I'm still unclear of the risk being mentioned for 25.4 bar-stem combos.
Thanks to the OP for this thread, time to recheck my steerers as I have only threaded stems, all pulled up to max extension.

Early Salsa was more mountain bike oriented. Handlebar clamps were 25.4mm for flat bars. Road bars have “traditionally” been 26mm, although some are 25.4mm, and Cinelli was 26.4mm.

It really comes down to matching the stem clamp to the bar. In the OP’s case both were 26mm so everything fits.

John

Cinelli made bars in 25.4 also.

Stem and bar compatibility are moot at this point, I've got parts that work together.

After much looking around for a replacement headset and considering things like cost, retro-ish styling (I know function over form is best but let's be real, most in the C&V forum have a thing for keeping that period look) and my generally curious nature, I've decided to modify the original upper cone and make it threadless per bulgie 's suggestion. A good friend's dad has a lathe and enjoys little projects like this and it'll be fun to further learn how to use a lathe.

I have everything I need at this point except a 1" threadless cap, which I'll order today. I'll be getting the replacement stem powder coated but this time I think I'll go with orange instead of purple.

We're you scared when this happened? I'd be scared ****less irrespective of the speed haha...

I had Cinelli 1R break on me a the usual place. Scared the crap out of me, but fortunately I was going about zero miles an hour. I rode it home. I don't want to imagine having a steerer break

My impression is that steerers all have cut threads. BITD, Trek had a steerer threading machine. You can see it in the picture of Dick Nolan and Mike Appel from the vintage Trek brochure.

Just make sure the framebuilder knows not to heat up the rest of the fork very much. Those Klein forks that I have seen are Sakae Litage forks painted by Klein. They are bonded using epoxy, which may not respond well to heat. How much heat is difficult to say, without knowing the exact epoxy and whether or not it has been heat treated already at the factory, which some epoxys are to increase strength and resiliency. I have seen "do not exceed" numbers from 135F up to 300F in my reading of epoxy documentation. It is not the same as not burning the paint off a steel fork. It will require a good deal more care. Heat can and will conduct readily down the steerer and may soften the epoxy bonding the steerer to the crown if precautions aren't taken.

If I were doing the job, I would not take chances. I would submerge the forkblades and crown in a bucket of water, and I would prepare the internal sleeve and extension separately (braze together on the bench) and then silver braze the sleeve and extension to the fork in one quick operation.

That's just my opinion though. I've never had the chance to do it because the only bonded fork steerer failure I've ever had to fix was an early Vitus fork with an aluminum 1" threaded steerer (!!)

Make SURE you have it centred.
Get the dad to show you how to use a dial test indicator - and get it so that needle does not even quiver.
You need to have the DTI run on the bearing races, not of any of the other parts of the cone; they may well not be true to the races, you cannot trust them.

Lack of concentricity is one of the reasons that home-shop refurbishment of bearing races is almost always a non-starter.
The other is lack of circularity - a decent lathe makes this one manageable.
With a lathe in good condition, a DTI, and skill it can be attempted.
If the dad has access to a Quorn or similar you have a better chance of success.

Don't even bother trying any refurbishment by hand; some people do, and the nice shiny surface one can get will fool a neophyte - but it won't fool the bearings; decent ones are accurate to 0.000025", grade 10 to 0.00001.
Cigarette paper is roughly 80 times that in thickness.

scarlson This was all stuff I thought about before even asking to have it repaired.
Jeremy Sycip has been in business for right about 30 years custom building frames. He's been contracted by the big names in components in the past to build one-off stuff to showcase new technologies at trade shows. He's repaired everything under the sun at this point, I have full confidence that he'll do it right and is aware of the inherent risks of heating up epoxy. He knows infinitely more than I do so I won't even attempt to tell him what to do aside from how long I want the steer and threaded or threadless.

oneclick Thanks for the info. While it's certainly good to know going in, this isn't exactly a wheel bearing we're talking about here. The tolerances don't have to be quite as exacting. I'm still aiming for perfect but will settle for good enough. All I'm doing is machining the interface between the steerer and inside of the cone. I've seen headsets, both in bicycles and motorcycles, so worn out from rust that the balls and races were visibly shaped different but still allowed the front fork to turn, albeit feeling like it was lubricated with sand.

As long as you are sure!! I only say this because I have, in the past, not fully communicated my concerns with framebuilders I had thought of as gods, and been burned as a result. Lots of people know lots of things, but no one person knows everything. If the framebuilder understands the concern, he shouldn't be pissed off at you for bringing it up. If he doesn't understand, and is still pissed off, run!! This is his business and this is your safety. There's no space for anyone's ego.

Huh, the ones I did, I just chucked 'em in a 3-jaw scroll chuck and I did not indicate them in. Just whatever concentricity that old well-worn chuck gives you. Those headsets adjusted up just fine as far as I can tell. What''s the downside of doing it that way? Or put another way, what symptoms should I have been looking for?

About those reports from people who did it with a carbide burr in a die grinder, and said the headset worked great: are we chalking that up to they were too "out of it" to notice how bad their headset really were? My working theory is, bike headsets just arten't that sensitive to tiny errors in concentricity.

Yeah, I have a Vitus alu fork that I got for free because the alu steerer predictably (inevitably?) broke off at the bottom of the threaded section. Had it for years, waiting for some stroke of brilliance for how to put it to use on a bike. Or maybe a toilet paper roll holder.

Aluminum threaded steerers, sheesh what were they thinking?

I know!! When I say I "fixed" it, I mean I threaded down a slightly-too-long Litage fork to get the Vitus back on the road. The offending Vitus fork is now on a "mule" bike that only gets ridden on the old RoadMachine trainer. In order to get the length I needed from it, I assembled a partial junk headset with no ball bearings. I joke that it is the next step up from fixed gear: fixed steer!

P.S. I agree with you on headsets. I don't think the precision has to be too great. I have seen quite a few bikes with bent steerers or improperly faced head tubes or some schmutz on the crown that makes the race sit crooked and the headset turns just fine somehow.