themp

3000 miles ago I had my Shimano 10s CN-HG95 chain wear out at 1500 miles based on using this tool:

https://www.parktool.com/en-us/produ...dicator-cc-3-2

I have always got 3000 miles plus on previous chains. I am on my second chain now and it has now worn out again with the above tool at 1500 miles. I went and got a good ruler and this chain checker tool and it shows no wear:

https://pedros.com/products/chain-checker-plus-ii

Thus, taking the roller wear out of the checking my chain is still good. But I am at a loss on why I am getting roller wear now. I did do the following maintenance at 25K miles and the wear started after that:

1.) Free hub replaced, it totally failed. But, I now have a wobble in the cassette with the new free hub when coasting. The Park Tool manual says this is normal and not to worry about it. Here is a link to that:

https://www.dropbox.com/scl/fi/dnuok...t1p8amkmv&dl=0

2.) Large chain ring, did not have to pull the crank to replace it. Small chain ring was replaced a few years ago.

3.) Derailleur pulleys, Shimano exact replacement.

4.) Cassette and chain. Now on the second chain that shows roller wear at 1500 miles.

Anyone, have an idea on what might cause just roller wear and no pin wear? I clean and oil the chain every 100 miles or so. I also think I have a real Shimano chain, but I did purchase the chains off Amazon. They came in packaging and have the Shimano pin for connecting. Finally, the square tapered bottom bracket is showing some play, so I have ordered a new one for replacement next.

KerryIrons

The fatal failing of many chain checkers is that their measure of the roller spacing is not accurate. Some will show a chain 50% worn right out of the box. Believe the ruler.

Iride01 

Might be grit getting in things. Since you replaced the freehub, which I've never had to do, That even more sounds like you might be riding where conditions are harsh. Do you ride in the dirt or sand quite a bit or is this all paved road and paved trails?

I use the Park Tool chain checker and I've never had roller wear that caused it to fail a chain early. When it said it was bad, measuring with a steel scale also showed the chains bad.

I too just lube my chain every 80 to 100 miles with a wax/water based lube currently. I've never replaced a chain before 4000 miles. However that one that only went 4000 miles probably should have been replaced sooner. But I'd gotten lazy about checking.

Perhaps you put a lot of muscle into the drivetrain. Pedal a easier gear at a faster cadence and you'll be just as fast but put less wear and tear on the drivetrain, IMO.

urbanknight

I first bought that same, cheap, unreliable chain checker and it did indeed show new chains as partially worn. Got a longer one (Park CC-4, similar to your Pedros one) and it was much better, but I later learned how easy it is to check with a good metal ruler since the pins should be exactly a half inch apart all the way down the line.

urbanknight

I suspect the problem is manufacturing tolerances. You got a good one, but the OP and I got a dud, and there are probably variations in between as well.

jadmt

apparently there were a lot of counterfeit shimano chains being sold and that was the big issue was they wore out really fast. sounds like many ebay and amazon shimano chains were bad.

maddog34

so your checker shows wear and your chain is only half the age of other ones... hmmm... sounds like there are other factors to look at.
rate of lubings..
type of lube used..
type of riding...
Wear On Associated Parts..................
weather factors..
chain and/or bicycles "cleanings"and frequency of these events..

and with all these OTHER FACTORS to consider, some choose to immediately scream "COUNTERFEITS!" and "Bad Manufacturing!" without wanting additional INFORMATION. SMH.

When were the chainrings and cassette/freewheel last changed? what quality level were they replaced with?
did you change lube types?
are you stripping all the lube out of the chain during cleanings?
does the chain feel looser when inspected by hand?

I have never bought a "Chain Checker", and never will.
i had one come in in a load of old parts once... i gave it away later that day.
funny thing about chains... they need to be "partially worn out" to MOVE.... it's called "Bearing Clearance."
most Bell chains and many "brand new bike"chains in Walmart are "Badly Worn OUT" from the factory, according to a chain checker.

The Sideplate/Pin Interface is where "chain wear" measurable with a chain checker occurs.

Barry2 

It's for this reason I no longer purchase chains from non-bicycle specific vendors (You know who I mean).
I use a LBS or a trusted sports store like Sports Basement & REI.

Barry

deacon mark

Use and machinist straight steel ruler with markings of at least 1/32. I have one that goes to 1/64 but I also repair guitars so need it. Never use or rely on a chain checker I have been a mechanic for years and they show all sorts of things on new chains and used chains. Simple metal ruler and pull the chain apart 12 inches and see what you get. I get 6-8k on my chains but I don't live in the mountains and tend to spin, I also keep the chain reasonably clean and lubed with plain old 3-in-one oil.

FBinNY 

Chain wear patterns are complex and full of subtleties.

Factors may include differences in materials. The pins, plates and rollers are made of different alloys and manufacturing considerations often (but not always) mean the rollers are the softest.

Also, wear at the pin is 100% internal to the chain, while roller wear - both ID and OD - can depend on the condition of the sprockets.

Perfectly meshing (new) chains & sprockets have minimal internal movement as the chain winds onto and off the sprocket. However, with wear the roller will roll on the backs of the teeth under high load, as the chain winds on and off.

So, the ratio of pin and roller wear will change as the drive train ages.

Hopefully, that explains the OP's observation.

Final thought. Pin to pin measurement is most accurate, especially when taken over 12", but doesn't factor roller wear, which does matter. So one needs to understand the limitations and take stretch measurements only as a rough guide rather than precisely meaningful.

msalvetti

When I brought my new Trek in for its first service after about 2600 miles I got scolded by the tech for letting the chain wear too far. I think he was right, I had started to miss shifts but it never occurred to me that it was due to chain wear.

I just replaced the new chain, as it had worn to 0.75% based on the Park CC-4 tool after just 1700 miles or so. I clean and lube my chain every 200-300 miles, ride only pavement, and never in the rain.

Pretty likely these were both real Shimano chains. The chain I'm running now is a KMC, we'll see how long that one lasts.

Mark

Paul_P

If chain-checking tools are possibly inaccurate, and metal machinist rules are accurate, why not use the rule to check the tool, file it if necessary and then have an accurate tool ?

deacon mark

Assuming you already have a chain-check tool then maybe that is ok. In my case I already have a few machinist rulers so no need to expense a tool I don't need or get to work by checking with my already accurate ruler. However chain-checking tools do not measure the whole length of the chain in 12 inches so I simply think they are inferior to begin with.

cuevélo

Because it isn't that the tools are manufactured inaccurately, it is that they measure something other than chain pitch elongation (other than Park CC-4, Pedro's chain checker plus II, and similar tools). The real important thing to measure is the distance between rollers, on the same face of the roller, when all rollers are pushed in the same direction. Traditional chain checker tools push the rollers apart, and measure between them, rather than truly measuring the pitch.

cyccommute 

The point of the chain checking tools is to have a simple way to a go/no go decision on whether a chain is worn or not. I did testing of a couple of chain tools and a measuring method in a now closed thread () that illustrated that neither chain tool is all that wrong on chain wear nor inaccurate. I have no idea how people mess up using a chain checking tool but they seem to be able to do it. A lot of the error probably comes from inaccurate estimation of chain use and/or other environmental factors that they fail to account for.

Frankly we don’t need highly accurate chain wear measurements anyway since chains are something that inevitably wears out and isn’t all that expensive to begin with. The rule measurement method could, in theory, give you an extra 200 miles of use but in the grand scheme of things that’s only a minor bit of mileage that’s not worth the hassle. The commonly used 1/16” elongation over 12” is only 0.5% chain wear which means that you might be losing much more than 200 miles of useable chain life. 0.75% wear is closer to 3/32” over 12”. And, I’ll point out again, that using a 12” rule and estimating 1/16” is not “accurate”. Any “estimate” is by definition a guess can’t be accurate.

FBinNY 

The entire notion of precision or accuracy in chain wear measurement is oxymorinic.

This is not a binary pass/fail thing where there's a magic point where things suddenly change. Talking about precision here is about as meaningful as obsessively changing motor oil at exact 2,500 mile intervals. Obviously a few hundred miles either way won't matter, and the exact time you do it will depend more on convenience than anything else.

Moreover, specific, untrackable variables make ANY specific interval meaningless. For example, 2,500 highway miles are very different than 2,500 urban miles, meaning a 2 or 3:1 difference in actual engine hours.

Likewise, with a chain. As the OP's experience shows, the relationship between pin wear (stretch), and roller wear varies, so the best one can hope for is a rough guide for a very approximate time to replace.

For my part, I tend to be conservative about replacing the first chain when the sprockets are still close to new, and get progressively more lax as the drive train gets older.

themp

Thanks for the link to the closed thread. It was a doozie... But, in the end here I have the Park chain checker that shows 0.5 wear a 'No Go', the Pedro shows 'Go', and my new metal ruler of 16 inches that shows no wear, a 'Go'. I can be conservative and replace now, or wait for the Pedro or ruler to go to a 'No Go'? Any advice on that question? Thanks....

Iride01 

IMO, you are going to have far less worries if you continue to use it with the roller wear and no pin wear than if you had a chain with no roller wear and a little bit of pin wear. But how much is too much is just a guess. I'd say replace the chain whenever it's convenient between now and then. But I wouldn't be in a big rush to find a chain right now nor would I not ride the bike.

With no pin wear the chain will have the 1/2" pitch from pin to pin it is supposed to have. The rollers are going to position themselves in the valleys of the cog where they need to be, and most of the cog wear will be about the same as if the rollers weren't worn. Or at least the wear won't be as bad as a truly worn chain.

But chains are cheap IMO, so there is nothing wrong with replacing it now if you prefer.

cyccommute 

Are you measuring the chain under tension with the ruler? Unless you are hanging it or at least measuring it using the tension of the rear derailer (bottom of the chainring to the derailer), you can introduce error into the measurement. Using any instrument has error involved but some have more places for error than others. A rule has a lot of places for error that the chain checker doesn’t.

FBinNY 

Like with most things, the devil is in the details. Ultimately, any considered judgement about chain replacement depends on the relative cost of chains vs. cassettes, and the condition of the cassette.

If you're using relatively inexpensive chains, and the cassette is newish and pricey, then being conservative makes good practical sense.

OTOH if this is the second or third chain on a cheap cassette, then you.dont stand to lose much by leaving chains on longer.

If you're still on the fence, there's another option. Consider replacing the chain now, but saving it to rotate back in when the new chain wears.

For years, I rotated 3 chains every 1,000 miles or so, ensuring that chain and sprocket age was always decently matched.

I stopped when modern chain closures evolved to one and done, taking the option away.

themp

The ruler is 18 inches long and squared off at the end at 18. I am measuring with the chain under tension from the bottom of the derailleur to the chain ring. I can put the square end in the pin indent(it catches slightly) and then look at the 6 inch mark at the derailleur and it is right on the 6 marking at the same point on the pin.

https://www.amazon.com/Staedtler-Sta...ps%2C87&sr=8-6

I found this statement on bikeradar:Another type of chain wear is ‘slop’. This isn’t as easy to measure, but it is common with riders who don’t put a lot of torque into their drivetrain, or perhaps spin a higher cadence.

Worn-out rollers increase the side-to-side play in the chain. As a result, the shifting is sloppier because the shifting ramps of the cassette or chainring need to work harder to derail the chain onto the desired cog.

It’s important though to bear in mind that chain slop won’t wear out other components.

And at Sheldon Brown's site this statement is made:

There are also special tools made to measure chain wear; these are a bit more convenient, though by no means necessary, and most -- except for the Shimano TL-CN40 and TL-CN41 -- are inaccurate because they allow roller play to confound the measurement of link-pin wear.

And this is an interesting article on the subject:

https://bike.bikegremlin.com/3370/ro...th-chainrings/

cyccommute 

There’s still a lot more places for error in the rule measurement method than when using a chain wear tool. At least you aren’t “estimating” the measurement and calling it more “accurate”.

These three statements seem to be at odds with one another. Roller wear is likely not as much of an issue as people make it out to be as stated by bikegremlin. Pin wear needs to be measured and chain wear tools do a good enough job for the task. As cyclists, we obsess far too much about chain wear anyway. Chains are cheap (or should be) and they wear. Replacing them at 3000 miles or 1700 miles if they show wear isn’t a huge lose of money. If you spend more than about 10 minutes per month worrying about chain wear or even chain lubrication, you are spending way too much time worrying about chains.