Today i realized that the bottom bracket t47 OD 50.5mm of my last built is heavy ovalized.
one direction 50.3mm the other 50.8mm
Are there some experience? will it become a problem with the bottom bracket cups? will they gradually loosen?
It’s fillet brazed and i think the wall thickness of the bb is not super thick…
Probably not ideal, but the ID and threads are the “important” dimension. It might be a little rough on your taps when you chase the threads. But once you do that then I’m thinking you should be OK.
ID is the same ovalized… threads are okay after cutting with the cyclus tool…
However, when the milling tool is in the thread, there is more play to the right and left than up and down. That’s why I was wondering whether this might affect the fit of the bottom bracket after a while.
That’s probably a specific problem with the larger T47 BB shells as opposed to the smaller BSA BB’s. The wall thickness in both shells is probably about the same, but the smaller BSA BB is stiffer radially as a result of the more aggressive curvature around the outside perimeter.
Let me take a guess, the 50.8mm is the fore-aft measurement and the 50.3 is the top-bottom (or alongside the ST roughly)?
I don’t think this will become a problem though, as long as the BB cup runs ok in the thread. If it can be tightened to the correct torque and the faces are nice and parallel, the cups should be positioned well enough for safe operation.
2 Likes
You have oversized threads. For thread depth of just a bit more than 1mm, that’s too much. One chart I found shows a (edit)difference between Max and Min for thread depth of 0.1mm and you’re talking 0.5mm which is well outside of spec.
If it’s for a customer, I would start over. It’s the hardest thing you’ll ever do but, we have to get used to tossing mistakes. If a chef burns pancakes, they don’t give them to the customer anyway. Don’t degrade yourself. Keep the learning; what caused the ovalization?
If it’s for you, send it with some good locktite and see what happens - there is more learning to be had! If/when it loosens, you can fill the threads with brass and re-thread it; repairs are a good way to make money too so get that learning as well. I wouldn’t do that to a new customer bike though.
1 Like
That’s definitely not what I wanted to hear, but maybe it’s the only right way…
1 Like
in this thread they talked about replacing the bb…? may this is a option?
This spec is taking into account the nominal loading capacity of said thread, which in this case would be something around 1000kN or about 100 tons, and a tightening torque of something like 4000Nm.
A BB thread is not a threaded connection in the classic mechanical engineering sense, so I would say mechanically it should be fine to just have the threads properly supported on half the circumference of the BB. The tightening torque for the BB cup can still easily be achieved and as long as the cup is sat properly against the face, there should be no problem.
I do agree though, that it is a bit of a letdown from a “quality of craftsmanship” perspective…
4 Likes
I suppose you could cut it out if you have a mill. You could also section the shell and then heat it from inside to release the shell fragments from the tube and then heat the brass and brush it off the tube with a wire brush. that would prob minimize the heat going into the tubes.
whatever caused the distortion - loose miters or heat - may haunt you.
1 Like
I’d hate for a customer to call me after a couple months and tell me the BB loosens up after a few rides even after repeated tightening.
Even worse would be if they don’t call me. Instead they take it to their shop and the mechanic does some investigation and they find it’s out of spec. So the mechanic now has a poor opinion of me and my customer posts on a half dozen forums saying how crap my bikes are and that my standards are so poor, I still sent it out.
I think I’ll just make another. It’s what I like to do anyway.
1 Like
That happened with the LBS. Well respected builder, but when they were with another brand the LBS got one of their bikes with a misaligned fork. Still badmouth the builder 20 years later.
1 Like
Can you elaborate on this comment? Something to do with not stretching the fastener? Surface embedment affecting the load of the joint?
Yes, that was exactly what I meant… Usually, you’d design a fastener in a way that it can provide a) sufficient preload for the application and b) is a few orders of magnitude more flexible than the components clamped together, so the pretension can be upheld in changing conditions, vibrations, temperature changes etc.
Sometimes this can only be achieved by artificially increasing the clamping length of the fasteners by using long bushings. You can find a few of these in this picture of the front end of a train:
In the BB case, the size of the thread is chosen for the ability to accomodate the axle rather than for mechanical reasons, and the clamping length between the threads and the face of the BB shell is way too short (and the cross sectional area way too large) for any significant elastic deformation to take place.
Anyway, BTT 
4 Likes
IMHO- You’re gonna be fine as long as you can run the chase and face properly. That amount of distortion was/is pretty common with the heat distortion and pulling power of fillet brazing (YMMV). In my experience with stuff that’s a bit egg shaped, the facing is really key, don’t skimp on ensuring they are as parallel as you can make em. This really allows the bb to snug up on the shell and stay tight. Some light thread lock is good insurance too.
If this is about customer bikes- that’s your call, but once you see an ovalized shell, you can never unsee it.
If it’s your bike, or non critical, run it till it has issues. If it’s truly F’ed up, there’s no loss. Worse case scenario you kill a bb and maybe a spindle.
Janosh, can you elaborate on the “slop” of the cyclus cutter you mention above. There are a few ways to interpret what you wrote, and I have had issues with Cyclus tools not being tight enough to face a bb square.
As described, the cutter for the thread has a little room to move. But on my second to last frame with t47, the cutter cut wonderfully and the bottom bracket went in perfectly.
Oh man, ok, yeah that sounds like a do over on the bb, if after you make the thread cut the tap is wobbly loose! How square are the faces of the bb shell when it’s all done? Im of course talking parallelism, not concentricity.
I’ve had slop in the bore of some cyclus taps that created an out of square bb face. Was thinking that might have been where you were having some issues.
Also, sorry for messing up the spelling of your name.
1 Like
no sorry for that! Faces are square and parallel
In my experience if those faces are parallel say to <.002" (what I do is measure with a caliper across the bb length at different clock positions on the face of the bb, the ST and CS will always pull in a bit) or better and the threads are ugly but functional and take the torque of the installation, I would build it up in the most risk free manner (cheap bb and crank spindle, no paint, etc) and see how it goes.
For your awareness as to where my question came from, and in reference to your original question about experiences with ovalization… I had a drive side t47 cup walk out on me on the first day of a 400 mile bike packing race… it was a facing problem due to slop in the cutters I had- and replaced after this happened. Rode the entire event. The bb cup fretted in the frame, and both cup and frame threads were gone by the end of the route, you could just pull out the bb by hand, no turning to unthread. That bike was tig’ed. In contrast, in 15 years of fillet brazing and the heat distortion and puling power that brazing brings, I never had a bb walk out, even if egg shaped with f’ed up threads. All of those fillets were on 1.5 dia shells to bring up an earlier point on this thread.
One small basic thing I do when I chase threads is start both sides and run them in at the same time to enable the tap guide to control the cut and create better concentricity. Learned that one with ovalization problems like yours. Even going back over the threads again this way can help.
Hope this is helpful.