Changing the chain when it less than 0.5% worn?
 

Had my bike at the shop for repair for reasons other than the drivechain. Was told that my chain was worn out, would I want it replaced. I said no, I have a spare new chain at home. From what I can gather he says the shifting was poor because of the chain. The chain probably has 7,000 miles on it. I checked with the Park chain wear tool, and it's not 0.5% worn yet. I don't distrust the bike shop's opinion, but I had always thought it was a matter of just checking the stretch on the chain. Should I be changing chains more frequently than the wear indicator suggests? What are some good rules of thumb here. This is an 11 speed Shimano 105 groupset.

The chain and cogs wear together and if you change the chain at the appropriate interval (I've heard it's at 50% of usable life), you don't need to change the cogs and will still have good shifting performance. This should be more economical since it's possible to have the cogs last through a couple of chains. I've never done this opting to wait until the drivetrain needs to be replaced as a unit.

When actually measuring with a ruler, the .5% on my PARK checker is really .25%. I think your LBS is......
I take the chain off and measure a 3 foot section to reduce the human error to 1/3.
I use the 1 & 37" marks so I don't have to deal with the hook end.

https://cimg1.ibsrv.net/gimg/bikefor...ca26fe359e.jpg

Pay attention to Bill's "tricks". Especially like the idea of using 3 feet, not 1. BTW, Bill, one slightly different take is to use the 10" and 46" marks. Avoid the hook, avoid the end of the tape that's worn.

Yep, that's what I follow. I wait until .5mm barely will go in. Once I didn't check for the longest time and the .75mm went in and was also sloppy, lol! That was time for a new chain and cassette. I couldn't believe how smooth and crisp-shifting the drivetrain was. In waiting so long, I basically just saved the cost of a second chain at the cost of smoothness and noise.

Just measure with a 12" steel rule. 7000 is a lot of miles so it's quite possible it's worn out or doesn't have much life left.

Why not use the 43 & 79?
It doesn't matter what you use as long as you can add 36.

Good heavens! Starting with 43, a prime number? It won't work then...

Seriously, its an old machinist's trick, and you use "10" so that you don't mess up and forget to subtract the "1", if you start at 1. Starting at the 10 means that if you forget to use 46 and instead use 36, its more obvious. That's the theory, and folks have been using it for at least 100 years or so if I recall reading it in an old machinist's account. May not be too important here, agreed .

That's 2 prime numbers. It must be my Asperger's.

As a Marine Boilermaker (completed apprenticeship in 1971 at Puget Sound NS) we always used 1" or 1'. Or the hook if it wasn't that fussy.
They actually calibrated/repaired (some got beheaded under the shears etc.) our tape measures (tool room) so they were actually quite good.

If you are going to use one of those chain checker thingies, you would be well advised to calibrate it against a brand new chain.

I've never been able to get more than half that life from the best chains made. If your new chain doesn't skip under heavy pedaling pressure with any of the sprockets, count yourself lucky and change sooner next time. Elongation is not the only type of wear that's important. I've had chain skip occur after replacing a campy 10 chain with 6000 miles on it showing less tha than .25% elongation.

Ok, that trumps my meager experience running a threading lathe and a drill press. Not only did you make things out of metal, but very large, very strong things. I grew up working in my Dad's manufacturing plant. Dad had a great respect for the guys that actually made stuff. Emphasized that I should, too. I don't even want to think what would have happened if I'd pulled "The Boss's Kid" routine, or disrespected a machinist. Lord help me if I'd been rude to anyone with "Maker" in their title! But it probably didn't even cross my mind. Even though I was a clueless rubberhead in high school, I somehow knew that a Mold-, Tool and Die-, or Boiler-Maker man was in an elite class.

Shears seem to be the bête noire of tape measures (and sometimes machinist's scales!). I suppose that, today, there's been more than one smart phone, too!

Thanks for mentioning your Marine Boilermaker skill. That's very cool.

7000 miles is plenty to get out of the chain. I recently replaced the entire drive train: chain rings, chain and cassette, which had about the same miles as your chain. I modified my chain rings to be more like what I needed, 46/30 instead of 38/26. The 38 ring was like shark's teeth.

thanks for all the advice. I'm going to change my chain.

This is how you measure chain wear accurately. You measure to the inside of the rollers at 2 locations and subtract, using calipers. Subtracting the 2 dimensions from each other will give you the dimension from the same side of rollers, so in essence the center to center dimension without the effect of roller wear. Do it at multiple sections of chain and average.

https://cimg3.ibsrv.net/gimg/bikefor...858729496f.jpg

Pedro's make an accurate chain checker at a reasonable price.

Or it might be better to compare the chain checker to a ruler, since you're more interested in the chain checker warning you that the chain is worn out (or approaching the end) than confirming that the chain is new.
My first chain checker was the Park CC-2. I found that on a new chain I couldn't even get the pins into the chain links(!). I contacted Park Tools and was told that the CC-2 is calibrated at the 1% mark. I've since found that this is basically true. I will notice when the reading is 0.75 and this trips the alarm to stock a new chain. The minute it hits 1.0 I replace the chain. By the ruler method, the chain isn't *quite* worn 1%. This has also been my experience with a Spin Doctor "go/no-go" chain checker and a Shimano TL-CN42 (another "go/no-go" tool).
So basically I consider the chain check tools to be "advance-notice" devices, and they are faster and easier to use than a steel ruler, which means I check the chain wear a lot more often than I would otherwise.

The Shimano TL-CN42 effectively does this with much less fuss. The only problem is that it's a "go/no-go" measuring device.

I have the Shimano tool and it's difficult to know when you're just at the borderline of the pass/fail because just a light push will make it go through. When my chain was at a point where the Shimano tool needed a slightly harder push to get through, my calipers measured around 0.4-0.5% stretch. I had a BBB or X-tools or some other generic tool that measures to the opposite sides of the rollers to include roller wear and it measured 0.75%.

I agree. That's why I like the Park tool, even though it doesn't compensate for roller wear and gives the "1%" reading a tad bit early. That's a good time to change the chain. If I ever *really* want the exact state of my chain, I have the steel ruler.