Odd number chainring
 

On my way to work today I stopped by a couple bike shops looking for a replacement 52t chainring for my single speed. None of the shops had a 52t ring but my last stop (at Performance) turned up a 53t ring.

I read somewhere a long time ago that having an odd numbered tooth count is bad for your chain. Is this correct? I have been searching google and coming up empty handed.

I'm not an engineer, but I can't see why that would be a problem.

Why are you running that big of a gear anyway?

Absolutely... The reason being that only half the links on your chain 'stretch'(the inner links); but to get the benefit of this you must put the chain back onto the chainring/cog in the same place. As always, sheldon brown was here first.

http://www.sheldonbrown.com/chain-life.html

I guess I spoke too soon. It seems so obvious now.

Wouldnt the odd number be good for skid patches

Sure does... That is why this 'trick' is really only useful for freewheel bikes..

Umm.....what?

You don't really need it. For example, 42/16 gives you 8 patches, 16 if you skid with both feet. 50/18 would give you 9 or 18, etc. That should be plenty.

So that's why you were late?

Skid patches double for ambidextrous skidding only if the number of chainring teeth is odd, mathletics. Forum member fraction proved it with the skid patch theorem and later on the author of Rabbit confirmed this result with a brute force test.

<-- NERRRRRD

Interesting! You've been covering me all day, mander, and thanks again. I hope that "skid patch theorem" is actually referenced somewhere, because I'd like to search for and subsequently read said theorem.

Regardless, 9 skid patches is till better than the usual results of even combinations.

haha sorry about that. Look it up, it's in the diy thread and will come up on a search. I have an arts degree and couldnt follow it , so i had to rely on the brute force proof, but hey, your name is mathletics.

Yes, I found and read it. It seemed sound enough for this crowd, considering the number of anti-math responses it received. The proof of (2) is technically incomplete (it doesn't actually show an "if and only if" relationship), but it followed clearly.

In any case, I now understand. Also, my thanks for your responses is sincere. I'd much rather be corrected and learn something than keep giving misinformation and sounding like a jackass.

Hmmm. Should've known I read that on Sheldon's site. Right now I have a Biopace chainring on the bike and the non-circular shape of the ring causes a slack-tight-slack-tight I can feel more than I care for when I get the rpms high enough.

Why are you running that big of a gear anyway?
When I got the bike the inner chainring (42t) was bent. The outer ring (52t) was in really good shape so I set up the single speed usinging that ring. So I run 52 in the fron and 18 in the rear. It would be plenty easy to take it down to a 42 tooth (do 42/16) but none of the bike shops I went to had 52 or 42. One shop said they have a 42t middle ring (I assume for a triple front) but stated that middle rings are not good for use on a single speed.