Can anyone explain what caused this chain to break?

For discussions about bikes and equipment.
geocycle
Posts: 1663
Joined: 11 Jan 2007, 9:46am

Can anyone explain what caused this chain to break?

Postby geocycle » 15 Dec 2008, 5:46pm

I thought those of you interested in bike forensics might like to see this picture of a broken SRAM chain. How do you lose every other inner plate from a chain? On returning from a 20 mile ride yesterday I noticed I had lost 4 plates from the inside (left) of the chain! It had been a rough-stuff type ride along the sea front and I suppose I could have hit a rock but I don't recall anything obvious. There is no sign of damage to chain ring or sprocket. I must have ridden a good few miles home and not noticed it was only held together by half the links. For the record it was a PC68 with about 5000 miles on the clock and was running on a rohloff hub gear (hence the relatively high mileage and not being aware of the breaks).

My best guess is I grounded the chainwheel but I really haven't a clue how you damage the inners and leave the outer plates in situ!

Image

**also posted on thorn forum**

User avatar
Mick F
Spambuster
Posts: 45539
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Postby Mick F » 15 Dec 2008, 9:57pm

Very odd, and very good that you managed to get home!

I ain't got no idea what went wrong, except the chain is filthy!
(Sorry!)
Mick F. Cornwall

pete75
Posts: 11186
Joined: 24 Jul 2007, 2:37pm

Postby pete75 » 15 Dec 2008, 10:07pm

Could it be an extreme version of this phenomena along with general wear?

10. Roller Chain Failure - Broken Inner (Roller) Plates

Roller chain failed in service and upon inspection of the failure it was determined to be a "crack" in an inner (roller) plate.

ANSWER: Cracked roller linkplates are often a sign of a fatigue failure due to the cyclic loading. The fatigue life of a chain can be improved by utilizing "Heavy" series chains, by increasing the overall chain size (i.e. #80 up to #100), or by reducing the dynamic load on the chain.

It's from http://www.diamondchain.com/support/techfaqs.php#t10

geocycle
Posts: 1663
Joined: 11 Jan 2007, 9:46am

Postby geocycle » 16 Dec 2008, 9:33am

Mick F wrote:Very odd, and very good that you managed to get home!

I ain't got no idea what went wrong, except the chain is filthy!
(Sorry!)


Sorry if I offended your sensibilities!!! I agree it was filthy, covered in half a beach which followed a fortnight's commuting on gritted roads. It was the fact that it looked so awful that prompted me to clean it and discover that it was half the chain it used to be.

geocycle
Posts: 1663
Joined: 11 Jan 2007, 9:46am

Postby geocycle » 16 Dec 2008, 9:36am

pete75 wrote:Could it be an extreme version of this phenomena along with general wear?

10. Roller Chain Failure - Broken Inner (Roller) Plates

Roller chain failed in service and upon inspection of the failure it was determined to be a "crack" in an inner (roller) plate.

ANSWER: Cracked roller linkplates are often a sign of a fatigue failure due to the cyclic loading. The fatigue life of a chain can be improved by utilizing "Heavy" series chains, by increasing the overall chain size (i.e. #80 up to #100), or by reducing the dynamic load on the chain.

It's from http://www.diamondchain.com/support/techfaqs.php#t10


Pete, thanks for that link. Although the mileage wasn't excessive on a hub geared bike, it had done a lot of offroad miles. The puzzle to me is how the inner links disappeared... only one has a noticeable broken chunk hanging from the rivet.

User avatar
MLJ
Posts: 536
Joined: 15 Jan 2007, 11:48am
Location: Rugby

Postby MLJ » 16 Dec 2008, 9:56am

I have had a fairly new chain running straight on a hub-geared bike break: the side plate just cracked. Poor quality control, I think.

james01
Posts: 2026
Joined: 6 Aug 2007, 4:48am

Postby james01 » 16 Dec 2008, 10:06am

Very strange. In a way it's a tribute to the strength of the chain in that it got you home with only one side functioning. It seems unlikely to be dirt & wear related. It couldn't really be a classic sequential failure ( one part fails & puts more load on remaining parts) because if anything the stress on the failed side was actually reduced when the first link failed, the other side took all the load. The only thing I can think of is that a freak piece of very nasty debris got pulled around the chain, although I'm suprised there's no evidence of damage in the cogs etc.
If you get to the bottom of this one, please let us know.

User avatar
Mick F
Spambuster
Posts: 45539
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Postby Mick F » 16 Dec 2008, 4:56pm

I have a Campag Record 9sp chain that's done 7500 miles and hasn't "stretched" any more than a sixteenth of an inch in 39 inches! I've still got it, even though I'm 10sp now.

Any road up, although it isn't "stretched", it is very floppy. The side plates are slowly working their way off the pins and the chain bends too far sideways to be reliable. I needs some careful tending with a chain tool!

I wonder if that's what happened to your chain. The plates just worked their way off.
Mick F. Cornwall

User avatar
7_lives_left
Posts: 798
Joined: 9 May 2008, 8:29pm
Location: South Bucks

Postby 7_lives_left » 16 Dec 2008, 5:28pm

I have a theory (prepare to shoot it down).

The chain was loose. It rode up on the teeth of (probably) the rear sprocket. Only the inner plates caught on the teeth, the outer plates didn't catch on the teeth. The chain tighted up and the sprocket teeth neatly chomped through a succession of the inner plates as you turned the cranks. Then the chain fell back into place where it should be.

I think it is no coincidence that all the broken inner plates are adjacent to one another. If they happed independently of each other, they would have been scattered all along the length of the chain.

Does that make sense?

PS I think it probably happened at the end rather than the start of the journey, because otherwise the chain would have fallen to bit's at the next 'lumpy bit' you came to. Does that fit?

User avatar
andrew_s
Posts: 4781
Joined: 7 Jan 2007, 9:29pm
Location: Gloucestershire

Postby andrew_s » 16 Dec 2008, 6:25pm

I'd guess that a rock or something tried to push the chain to one side. On a derailleur bike this would just derail the chain (change gear) but on a hub gear or fixed gear bike there isn't enough slack to let the chain lift over the sprocket teeth. If the side push is forceful enough to get the inner link plate riding on top of the teeth, something is going to give as the chain doesn't ride up on the engaging sprocket teeth. The outer plates were unaffected as they are a bit further out and not on top on the teeth.
I'd further guess that the broken inner link plates were on the outside of the chain (away from the bike).

The chain didn't break because the outer plates are still in place holding everything together. The chain will still be able to take as much tension as is required to pull the remaining inner link plate in half.

Clean the sprocket and see if you can see any sign of a nick in the top of any of the teeth.

PS
My unusual bit of missing chain was one that had managed to lose 4 or 5 of the rollers.
Last edited by andrew_s on 16 Dec 2008, 6:31pm, edited 1 time in total.

geocycle
Posts: 1663
Joined: 11 Jan 2007, 9:46am

Postby geocycle » 16 Dec 2008, 6:30pm

7_lives_left wrote:I have a theory (prepare to shoot it down).

The chain was loose. It rode up on the teeth of (probably) the rear sprocket. Only the inner plates caught on the teeth, the outer plates didn't catch on the teeth. The chain tighted up and the sprocket teeth neatly chomped through a succession of the inner plates as you turned the cranks. Then the chain fell back into place where it should be.

I think it is no coincidence that all the broken inner plates are adjacent to one another. If they happed independently of each other, they would have been scattered all along the length of the chain.

Does that make sense?

PS I think it probably happened at the end rather than the start of the journey, because otherwise the chain would have fallen to bit's at the next 'lumpy bit' you came to. Does that fit?


That sounds quite a plausible explanation. It is a while since the eccentric was adjusted to tighten the chain so some movement off road is certainly possible. I rode back about 5 miles along a well surfaced canal towpath after the rough coastal section. As James mentions it is indeed impressive that the chain remained intact. I cannot see any evidence of wearing on the remaining inner plates so some kind of catastrophic crunch is most likely.... I also cannot imagine how a nearly perfect chain line could result in differential wear to this extent.

geocycle
Posts: 1663
Joined: 11 Jan 2007, 9:46am

Postby geocycle » 16 Dec 2008, 6:38pm

andrew_s wrote:I'd guess that a rock or something tried to push the chain to one side. On a derailleur bike this would just derail the chain (change gear) but on a hub gear or fixed gear bike there isn't enough slack to let the chain lift over the sprocket teeth. If the side push is forceful enough to get the inner link plate riding on top of the teeth, something is going to give as the chain doesn't ride up on the engaging sprocket teeth. The outer plates were unaffected as they are a bit further out and not on top on the teeth.
I'd further guess that the broken inner link plates were on the outside of the chain (away from the bike).

The chain didn't break because the outer plates are still in place holding everything together. The chain will still be able to take as much tension as is required to pull the remaining inner link plate in half.

Clean the sprocket and see if you can see any sign of a nick in the top of any of the teeth.

PS
My unusual bit of missing chain was one that had managed to lose 4 or 5 of the rollers.


Andrew, like 7-lives above, I think this may be an explanation. The only point that doesn't fit is that the missing plates are on the bike side which suggests a blow from underneath somehow. The sprocket and chain rings looks fine (fortunately).

PS did you ever work out how you'd lost your rollers.....

User avatar
andrew_s
Posts: 4781
Joined: 7 Jan 2007, 9:29pm
Location: Gloucestershire

Postby andrew_s » 17 Dec 2008, 7:08pm

If the chain was a bit slack I suppose going over a bump with the bike tilted to the gearside could get the chain to start to ride up on the sprocket teeth. Anything physically pushing the chain over would have to get between the chain and the wheel.

I assumed that the rollers had just worn thin enough to crack - it was a pretty elderly chain (~7000 miles).

tad
Posts: 74
Joined: 9 Jan 2007, 11:53am

Postby tad » 17 Dec 2008, 9:49pm

However you did it I think it's very clever, congratulations!

User avatar
squeaker
Posts: 3520
Joined: 12 Jan 2007, 11:43pm
Location: Sussex

Any more part-failed side plates?

Postby squeaker » 18 Dec 2008, 9:14am

It might be interesting to know if any more of the side plates are on their way to failing (radial cracks out from the pin holes).
And no, I'm not offering to degrease / inspect it for you :roll:
"42"