marcow

Hi,

I have a couple of cracked seattubes of our frames in the last couple of years. The first two had a shim inside, so we blamed it on the shim. But this week we got another one without a shim.
I only design the frames, buy the tubes and have them made by other companies. And this specific type is on the market since 2002 and about 1000 pieces have been made.
The seattube is 31.8x0.9 4130. And the seatpost is 30.0 mm.
The frames is 3 years old , and the rider is about 75 kg.
The position of the crack is where the seatpost ends.

My questions:
Where does the stress come from that causes the crack? Why doesn't the other side crack?

marcow

Here's one from last year.
Same tube and material. Shim was used with a Thudbuster seatpost.
The crack is at 80 mm from the top. This is the shim length.


I think this rules out that the cause has anything to do with movement of the bottom part of the seattube.
This specific bike has a Pinion gearing system. (but I don't think that has anything to do with it)
I also had another frame from this type. Also with Pinion, shim and a Thudbuster seatpost. But the shim was 100 mm instead of the above 80 mm.
The crack there was also 20 mm down.

jadocs

How long is the seat post being used? does it end where the crack begins?

marcow

It's a 300 mm seatpost. And it ends at the crack.

unterhausen

I have to think about this more, but the first thing that occurred to me is that there could be stress built into the frame. It would have to be in such a way that the rear of the seat tube was in tension. Then the compression from the rider leads to tension/compression cycles, which are the worst possible case for fatigue crack growth.

Do you know the order of operations? It seems like doing the top tube join first would possibly lessen the tension in the frame. If they do the side tack on the seat tube first, then the seat tube will be put in tension when they did the top tube.

marcow

Mystery almost solved. Bad tube.
Wall thickness varies from 0.47 to 1.4 mm

unterhausen

Good job. Problems with distortion leading to heavy reaming.

Andrew R Stewart 

Very good thread! I much like that marcow did more then post images and ask about this. When we made the Helenic frames at Cyclery North we had some concerns about seat tube walls and post fit too. I did see a different shop's attempt at a helenic that cracked at the stay/seat tube joint. We assumed it had been over heated, knowing the person who did the work as we did. Andy

marcow

That's what we tought at frame nr 1.(the green one) It was a normal frame, that was locally converted with a Pinion bridge. The welder admitted that he might have bended the tube a little to make it fit.
Then nr 2 appeared. This was also a Pinion frame, but this one was build around the Pinion Bridge, and was not a conversion. Done by a different framebuilder. This doesn't rule out the tension, but makes it less obvious.
Frame nr 3 (the grey one in the first post). I don't know how they build it.

marcow

Not too much reaming. (not more than 0.05 mm that I can measure)
But it is strange that I measure 1.4 mm on one side, that is supposed to be 0.9 mm.
Somehow the internal diameter is off center.

unterhausen

Okay, you can see that in your pic. I assume it's not high-end tubing? Do you mind saying what brand?

Almost makes me want to get a water table and doing an ultrasonic thickness inspection every tube.

'02 nrs

after finding a hairline crack on an '85 Falcon in an unstressed area it too had me stumped-till I shined a light in to the tube.a very thin wall in the same area as the fracture.guess its not uncommon to have bad tubes of varying thickness.

unterhausen

That's not really unstressed area on that Falcon though, seeing how it's close to the area squished by the binder bolt. How far did the crack go?

Andrew R Stewart 

The Falcon's crack is exactly where the temp control during brazing can be hard to do, twice. I wonder if there's a slight bulge outward of the ST just below the lug's edge. I can't count how many frames I've felt that heat distortion about a finger length down the ST. Both of the frames I have made that suffered cracks did right there too. Both times I knew I cooked the ST. One cracked before the paint, one after many thousands of miles and many years. Andy