darb85

I need to size my new chain on my road bike. Dont have the old one to help but is the large chain-ring to large Cassette cog plus three links still the preferred way of doing it? I will be using the 1091R chain from sram and their power link.

Thanks!

Brad

FBinNY 

Yes, large/large plus 1" or 2 links is minimum, so if you end up with an odd number you go up to 1-1/2".

You can also work from the other end, threading the RD, looping the small/small and pulling up until the RD begins to take tension. This gives you the maximum length.

Anything that passes both tests is OK. If the chainring and cassette ranges add up to less than the RDs capacity, either measurement alone will work, but if pushing the limits I strongly urge doing both tests, and in no case running a chain too short to safely loop the big/big.

redtires

I personally have never used that particular method...and to toot my own horn a bit, I was a very good professional mechanic for many, many years. That being said, that was before the market was saturated with "compact" drive-trains. Although, I think it would work just fine. I always go big ring to small cog...and then tension the chain so the pulley bolts are perpendicular to the ground. If you can't get it right on, just make it a tad shorter. This will generally allow you to have the best chain tension, with the least chain slap and of course, you should not ride small ring/small cog or big ring/big cog.

operator

This method is less exact than large large plus 2 links.

There's really no reason to use this antiquated inaccurate method of chain sizing anymore.

FBinNY 

It isn't that one method is "right" or wrong, or antiquated. Big/big+ determines the minimum length, small/small w/no slack is the maximum length. When there's a range between the two, whatever works best within the range is the best length.

I'm old school and prefer to set chains up much nearer to the max length. This gives me the freedom to increase the cassette size, or cut out a bad link later on. Of coiurse that means that all the bikes I set up have a weight penalty by an inch or two of chain.

BTW- looping big/big is probably the oldest method for finding the minimum length.

operator

The only correct method of sizing is shortest length.

Small/small with no slack is not a valid test for maximum length - because b screw adjustment will affect this.

FBinNY 

I guess you're much more doctrinaire than I am.

I can't fathom why when setting up a chain on a 11/21 cassette, I should cut it so short that a new chain would be required if switching to a 12-25 cassette for a hillier ride. But that's just me, I consider what's necessary, then look beyond to what's possible.

If you can tell me a technical reason why extra chain is problematic (other than weight) I'll be glad to listen, but until then we'll just agree to disagree on this point.

operator

That extra link on a system that is setup to max capcity of the rear derailleur means that it will be slack in the small/small combo. I've seen this before.

I guess you can tailor it to the application, but why bother calculating whether its at capacity or not. Big/Big eliminates any and all confusion.

Also, going back to the above. Setting the chain length with the derailleur pependicular method allows you to set a too short chain length if the derailleur does not have enough capacity. Big/Big + 2 eliminates this problem. It is always reliable, regardless.

redtires

Thanks for making feel old! Although, in one aspect you are correct. But then again, I still use a 53/39 and a 12/25 set-up, and I've never had any issue using it. But hey, I'm young at heart...next time I replace my chain, I'll try the "large/large" method and see if it makes any difference.

Did I mention I used a Mavic SSC group until the the early 90's?

FBinNY 

Nothing is wrong with what you say, but if you read my earlier post, I described minimum, maximum and in between chain lengths. Any length that is longer than the minimum, and shorter than maximum is OK.

operator

I understood that.

I was pointing that minimum is equal to maximum in some cases.