I know that cranks aren't perfectly round and as the crank turns the tension on the chain changes but it seems like on my bike that the change is considerable. When the drive side arm is parallel to the ground and pointing towards the front of the bike, the chain tension is good. When the same arm points to the rear the chain sags almost an inch. Is this much of a difference normal or is something up?

loosen your stack bolts and recenter your chainring. theres a guide on sheldon

what kind of crank and ring do you have? some are rounder than others.

this phenomenon bothered me a whole lot more riding on the street than on the track.

Unless it's a biopace ring, that sounds a bit too much/messed up - check out your ring center-ness as Jenkinsal2 said.

could you explain this? the biopace part. I just received, then donated a pair, so I'm just curious.

Sheldon, as always, has the answer

Ok I'll check that out thanks.

Miche Advanced crankset.

It could be much, much better. If that helps.

i've never played around with miche advanced but it's a high quality enough ring that this *shouldn't* be a problem. definitely try recentering your chainring via sheldon's method.

For those who don't click the link, Biopace chain rings are slightly oval, to try to even out power distribution around the pedal stroke. If you look down as you ride you will see the front of the chainring seem to move forward and back with each stroke.

That doesn't even happen with biopace.

Well I tried this and it doesn't seem to work. The chainring doesn't move at all. I even loosened the bolts all the way and nothing. Maybe the entire crankset is defective?

i tried that loosening of the chainring bolts, i find it not to be effective at all since both chainring and spider are machined indentically..

Re clock the chain ring in a different position. How many bolts does it have? That's how many possible positions it has. Maybe one position is more centered then the rest?

I thought about that but like na975 posted above, the chainring and spider are machined to fit perfectly. There is no wiggle room. If something is off, no matter what position it's at it will be the same.

I "suppose" it is possible you are experiencing the same phenomenon in the rear that we expect to find up front--- a poorly built hub.

cog cross-threaded? stiff link in your chain?

Stiff link sounds like an excellent theory. If it weren't fixed, the chain would be hopping.

think about this one for another minute.

Correct. Biopace rings maintain tension the same way all the way around, just as a regular chainring does.

Interesting, didn't know, thanks.

I hope my Phil Wood hub wasn't poorly built. I'll check the links to see if there is a stiff one. I also try rotation the chainring to see if that does anything.

While you have your chainring off, use a compass to go around the half of the chainring with the arm forward (the half that the chain is engaged on). Then do the same with the arm facing back. A lesser number will tell you that the ring is the problem.

I had thought about using a diameter measurement, but it doesn't seem to work right in my head. I think an arc measurement would be better. If you can't measure the arc properly (get a length figure), then do the longer one first (arm forward) and attempt to 'overlay' the same arc with the loose side (arm backward) - if they're not on top of each other, or even close, you know it's the ring.

If it's not too big of a problem, you can simply loosen the tension on the chain to compensate. It doesn't take care of the problem but it makes it unnoticeable.