I went to a Trek store to get a new chain and cassette - the chain had a broken link. The tech on duty used a Park CC-3.2 chain checker. He used the .75 side of the checker and since it didn't fall through the chain he said that meant that the cassette was good and didn't need to be replaced - made no sense to me but hey, he works on bikes all day so I deferred to his "expertise". He was sure I could get another chain's worth of use out of it. The chain had a broken link so obviously it needed to be replaced. Subsequent testing showed me the cassette wasn't good - it was slipping on a brand-new chain. I got a cassette elsewhere which I installed myself and the slipping disappeared. Obviously the Trek tech's "test" failed.

Have you ever heard of the above outlined use of a chain checker to evaluate the condition of a cassette? Would you view this as a valid test for a cassette or is this tech just an idi...um, misguided?

Perhaps he misunderstood what you are asking. Particularly if you did a long TLDR type description which in this case will be TLDH . And thought you meant the chain.

The long accepted test by many for a cassette being worn out is when a chain, particularly a new chain, slips on it.

Rohlof makes a tool to test for sprocket wear as do others. However then you have to decide at what point it's worth it to you to get another cassette. Worn cassettes will wear a chain faster. However for what they both cost, it's not going to amount to a big savings whether you wait till the chain slips or change the cassette sooner. Probably actually more expense as is most other preventive maintenance that is done for other than safety reasons.

If any in spec chain at .75 isn't slipping on your cassette, that usually means the cassette isn't worn enough for a new chain to slip either. Both should have been slipping.

It isnt a perfect method, but it is reasonable.

I don't know why he didn't just test ride it after the new chain was on, though.

That's a pretty common methodology. As you found out, it's not perfect and fails relatively often.

This is why mechanics still do it - if it works out they never hear anything so that's confirmation it's a good test. If it doesn't work out, almost as often they never hear anything either so they get a false confirmation. Not assigning any blame to you by the way, just explaining.

For a time I worked at two different big box bike stores practically across the street from one another and would regularly get dissatisfied customers from one store at the other. They rarely left feedback, they just stopped going to the other store.

I don't see how he could have misunderstood - I showed him the chain had a broken link and while I was there I wanted to replace the cassette as well. He thought a cassette shouldn't be worn enough because it was the factory chain, he also said that he had to use the old chain to check the cassette then did the test as described above declaring the cassette good. It wasn't complicated - I wanted to replace the cassette - he insisted I didn't need to.

Going by what you're saying if a worn cassette will wear a chain faster I would think the default would be to just replace both but he was resistant to that.


This time they definitely heard about it - I left feedback online outlining the scenario. How is it supposed to test the cassette - that if the chain isn't worn to the .75% point the assumption is that the cassette shouldn't be that worn?

It's a live and learn experience.

To be honest, the broken link on your chain should have been a red flag. I want to say the your mechanic was probably wanting to save you some money, not a problem. However, some shops ask the mechanics to test ride after a repair, particularly after changing drive train components. One thing: a cassette that was ridden mostly on a single cog can be worn out even if the rest of the cassette is pristine. This is very common for bikes with previous owners. I have known guys who bought multi speed bikes who never changed gears. Why buy a bike with many gears and never use them? But, there it is. a cassette may only have a single cog that is worn out, but you have to test that single cog to determine that the cassette should be replaced

I don't know how you get that out of what I said. Once you use anything it has some wear on it. It doesn't just go for so long and then all of a sudden it's worn.

I was just saying that a worn cassette will wear a new chain faster. But not that much faster. In my case, is it worth putting a $70 dollar Ultegra 11 speed cassette when my Ultegra 11 speed chain only cost $35?

Not to me, that chain will wear faster on the old cassette, but only a little bit faster. So guessing a 1000 or so less miles. Even if it was half the normal chain life I'd still run the new chain with the old cassette till a new chain slipped on one of it's sprockets.

When you replace stuff is a matter for your judgement and pleasure. So if you want to replace them both, then go ahead. But the common thing many of us do is replace the cassette when a new chain slips on one of it's cogs. Same for the rings if the front sprockets get too worn. Which did happen on one of my bikes.

Some replace the cassettes every other time they change their chain. And some replace them every time they change their chain. I'm not certain, but I feel that the majority of those people tend to be involved in serious competition type cycling. Which will seem more reasonable to change both.

Some skin flints will only replace the worn sprocket on their cassette and not the whole thing. But I'm not going to go to the trouble of looking for that one sprocket.

You decide what you want to do. It's your bike. You don't have to even change the chain when it's 100.5% or 100.75% of it's original length. But those are typical times that over the years many have found work well for changing their chains and minimizing the wear on the sprockets front and back.

How did he know it was a factory chain? I have to say I’m surprised a cassette was worn if the only chain it had ever seen was still at <0.75 albeit the comments above about single sprockets are bang on.

If the link was broken, it wasn't under tension when he measured it, so he got a false reading. Also 0.75 is pretty bad if it is 11 speed.

This is truth, and other posters seem to have missed it or glossed over it.

An 11 speed chain is supposed to be replaced at 0.5 wear… You can go quite a ways past that before hitting the .75 mark, and put a lot of additional wear on the cassette.

I was thinking recently after that "trim a cassette" thread what a cog wear indicator would look like. I came up with several ideas that seemed like they would work. Dimensionally it's an easier problem than the chain because there's no tolerance stackup. But they all needed the cog out of the cluster and flat on a bench or within the tool, and it seemed like you would need a single tool for each tooth count. I should revisit

Why do you assume that the section of chain he measured contained the bad link?

I don't, but I am assuming if the link is broken, the whole chain wasn't under full tension, or was off completely. If the chain is not under normal tension, the result will be an under-estimate, whether or not that segment contains the bad link.

I don't know what you mean by "not under normal tension". The chain measuring tool provides its own tension.

Here we go, yet again.

I'm sorry to inform you how the chain tool works.

Stick to electrochemistry.

Oh, yeah. You're the guy who doesn't know what galvanic corrosion is or how it works. I forgot.

Anyway, chain measuring tools get the same result on or off tbe bike.

Yeah, I'm the physical chemist who does not know any chemistry.

This post is about 80% correct, in my opinion. If you have ten gears and keep the bike in just one, it will wear out the gearing 10x as fast as someone that shifts around to all the gears equally (more than 10x if the gear you use is also the one with the least teeth). You can't judge by just chain wear or whatever when some riders wear out gears ten times as fast as others due to nothing more than their shifting habits.

However, I wouldn't immediately condemn a rider with a "heavily favored" gear. There are valid reasons to have gears on a bike that you don't use often enough to wear out - perhaps even one overpass on an otherwise flat route.

In my experience, the only way to be sure that a cassette still works OK is trying it with a new chain. Sometimes cassettes that look "as new" skip with a new chain and others that look slightly worn don't.

He knew it was a factory chain because I told him - I'm the original owner of the bike, bought it at the same store. I've used a different kind of chain checker before but apparently it's been longer than I realized since I last checked it.

Ooer I didn’t know this. I get to 0.5 pretty quickly on mine (less than 2000 miles) even when I’m using immersive waxing. Usually replace at .75 on my gravel / commuter but not had to swap cassette yet, after 4 chains.

If you hit that 0.5% more point in 2000 miles, I have to ask if you run often in gear combos that keep the chain at a angle between the front and back. Or you pedal with high power at a low cadence too much.

I did wear out my first 11 speed chain in about 5000 miles. And it was over the 0.5% mark before I realized it. I dd very often climb the short hills here in the big ring and lowest rear sprocket. And that kept a lot of angle on the chain. Which I can't help but think must worn it faster.

On my current bike with Di2, I have it set up so that the chain won't go to the extreme two low sprockets on the rear when in the big ring. And I currently have 8000 miles on it. And it's still has not reached 0.5%

I use to ridicule the anti-cross chainers. They seemed to be a cult forecasting doom and gloom to any that ran their chains in the big big combos. They made it seem like the bike's drive train would blow up. But now I think they might have been on to something. They just didn't know how to express it other than say don't do it.

Cross-chaining as a source of wear was directionally correct in the past but more of an indictment of chain technology of the time. Modern day mountain bike riders often ride at extreme cross-chaining in adverse conditions and see low rates of chain wear. The surface treatments, steel available, and ability to more accurately harden the pins/links/rollers has improved significantly just in the past 10 years. This seems to have a much larger impact on wear than cross-chaining.

I can clearly see this even in my own road and off-road riding. Same conditions, lubrication, terrain, and riding style but two different drivetrains. My 9s chains with no chromizing treatment and 1990s chain manufacturing technology/materials wear about 30-40% faster than modern 11/12s chains.