View Single Post
Old 10-19-15 | 09:42 PM
  #9  
dddd's Avatar
dddd
Ride, Wrench, Swap, Race
15 Anniversary
 
Joined: Jan 2010
Posts: 9,834
Likes: 1,811
From: Northern California

Bikes: Cheltenham-Pedersen racer, Boulder F/S Paris-Roubaix, Varsity racer, '52 Christophe, '62 Continental, '92 Merckx, '75 Limongi, '76 Presto, '72 Gitane SC, '71 Schwinn SS, etc.

Originally Posted by ThermionicScott
I buy SRAM PC-850s in bulk, and use them on anything between 5 and 7 speeds. Double the longevity of the 830 thanks to the hardened pins, for only a couple dollars more.
Good point. I think every chain maker has a sort of threshold in their lineup, below which you get a noticeably inferior chain.

I can most readily identify the cheapies when I use my "5.2 inch minus 1.2 inch" (compensated) stretch measurement method using the inside-measure pins of a digital caliper.

The lesser chains produce inconsistent readings from one measurement to the next, whether new or used.

I've identified Z-series KMC chain, HG30 Shimano chain, TaYa chain and Sunrace chain as having these noticeably-loose tolerances, which makes me suspect that they might increase sprocket wear rate on my most precious freewheels.

My "5.2 - 1.2" measurement method is for eliminating the effects of roller diameter tolerance and roller freeplay on the "stretch" measurement metric, since the caliper pushes rollers in opposite directions.

In use, the chain rollers engaging any single sprocket are all stressed in the same direction, and so cannot have an actual effect on chain link pitch wear.

The subtraction reduces the freeplay component to zero by canceling the freeplay component of each of the two measurements.

A net sum of 4.00" equals new chain, while a sum of 4.020" is at 100.5% of new and is considered worn out, to be replaced immediately if there is to be any chance of using the old sprockets with a new chain.

Measurements need to be repeated until proven consistent, with the sharp edges of the inside-measure pegs centered on the rollers.

Shimano makes a "3-peg" chain measurement tool which accomplishes the same compensation by applying force in the same direction to the two rollers between which the "stretch" length is being measured. But it is a "GO NO-GO" measurement and as such does not tell the user how much that the chain is worn, only whether the chain is still good for a subsequent service interval of some normal duration. It tells the shop whether it is time to replace it. This simple-looking tool is rather expensive, over $50 iir.

I'll add one little tidbit, that the Shimano 9s chains have the biggest difference in terms of how much wider that the sideplates bulge out versus the width of the pins, more than any other chain. They thus give a most aggressive "early" sort of shift behavior toward larger cogs, which works especially well in many friction-shifted setups, and especially with Suntour narrow-spaced freewheels.
...Not to say that other modern chains don't also work very well.
dddd is offline  
Reply