cyccommute 

Demonstrate your claim. How much of a difference is it? Even if I accept your premise, the difference isn’t going to be significant enough to have much impact.

The chain pitch is elongated because the pins wear and allow the pitch distance to increase. You’ve already said that you can separate the two so why try? The distance measurement…independent of how you measure it…tells you that the pitch distance has changed and the chain is worn. Measuring pins or measuring rollers gives the same result, thus there is no need to separate them. In other words, neither measurement give significantly different results so they are equivalent.

Again, the results of the measurement are the same. If the results are the same, the measurements are equivalent. Neither one is superior but one is easier than the other.

I’m not flip-flopping. I completely understand the two measurements and what point you are trying to make. Your point is not worth considering because it is a trivial point of no consequence.

Kontact

Thread OP:

cyccommute 

You don’t seem to understand how this works. You made the claim that roller wear is important and has to be taken into account. So it is up to you to demonstrate your own point. I have shown you that you are wrong by pointing out that rule measurement and chain checker tool measurement give the same results. If you have evidence to the contrary, by all means present it.

At no point have I insulted you. I have patiently explained to you why you are incorrect but I have never been insulting.

Kontact

In my experience on the bicycle mechanic subforum, there are few who get as little respect as actual professional bike mechanics.

I can post almost any fact about bicycle mechanics and at least two know-nothings will come along to say "I don't know about that." Not opinions or techniques - facts.

So I respectfully disagree with your experience.

cyccommute 

As have I. The difference is that I seldom made measurements by direct means.

No. Measurements should only be made to the level of accuracy or precision that is needed. If you only need to make a measurement to one decimal place, you don’t need to make the measurement to 5 decimal places. Measuring chain wear to 0.06” is adequate. Measuring to 0.062547” isn’t needed nor will it tell you anything.

Not in my nor most people’s experience. Changing a chain at 0.75% wear with a chain checker won’t result in skipping gears on the cassette in most all cases. Going much beyond that might or might not…it’s not an exact science. And, the fact that it isn’t an exact science is why you don’t need very high precision and accuracy in the measurement.

Good enough is good enough.

cyccommute 

So? One measurement is wrong and one is right. Which one? Tell us.

Kontact

You said that roller do not wear at all, then you explained how a ruler will show roller wear, then you said it won't show roller wear, then said that roller wear does contribute an extra amount to pitch elongation, then you circled back.

Do I need to quote you? Nope - anyone can go back and read the thread.

Captain obvious.

Kontact

What do you want to measure? The pin pitch, or the actual pitch as the chain engages the sprockets? Both can be correctly measured, only one is useful for chain replacement.

Koyote

You guys are laboring under the misapprehension that most bf posters are here to learn things; that's why you get in these arguments while trying to 'educate' each other.

Also, bragging about credentials and such exemplifies the "appeal to authority" fallacy -- which only makes a person look weak and insecure. An argument has to stand or fall on its merits, not on the person offering it.


PS: if any method of measuring my chain indicates that it's toast, I replace it.

Lombard

You sure have lots of time to argue on these forums. Bike saddle business slow this year?

Kontact

I have two jobs, still make a few minutes a day to do things I like.

I assume you are in the insult and personal attack business, which is why you make time to post off topic character assassination.

Don't you have a short, slow, flat ride you could be on right now?

Lombard

I don't think I'm the one here making insults, personal attacks and character assassinations. What do you all think, folks? BillyD ?

shelbyfv

Wasn't really thinking "professional" bike mechanics, that's a pretty low bar. For example, this pins/rollers/chain wear discussion would benefit from input by an actual engineer of the relevant type. The rest of us just have opinions and don't know what we don't know.

Kontact

I rather doubt the staff exists to validate your put downs in preference to mine.

You were being rude and off topic. The thread will be closed and then you can pretend your hands are clean. But I know and you know it was a personal attack and you are unkind.

Kontact

If having decades of experience being a bike mechanic is not considered expertise on a bike mechanics forum, then nothing is.

No wonder Qanon got so big. Too bad he can't help you fix your $300 shifter.

Lombard

Nice spin. If this thread is closed, it will be all because of you and your inability to admit you are wrong or control your anger.

shelbyfv

You miss the point. I'm confident you could replace my chain. I'm not confident anyone posting here has the expertise/ education/equipment to resolve this pin/roller/ruler/tool discussion. You'd think Shimano or KMC have engineers who could explain this but AFAIK nobody has posted any links. Otherwise it's just noodling and anecdotal.

GhostRider62

I agree

Your approach is neither good nor enough. My approach is sufficiently accurate, precise and repeatable. It is not high precision. And, I do not need your silly lectures on significant figures or red herring examples WRT skipping gears. Seriously, you think you know everything. The resolution of all these devices is insufficient. Prove me wrong, find an R & R study. Unless you can do that, don't bother responding.

Kontact

I get what you're saying, but we haven't even gotten close to engineer territory because we're largely bickering about how simple geometric shapes interact. An academic can't referee that lunacy.

cyccommute 

Let me start by saying that it is technically true that I have never been paid a single penny for being a bicycle mechanic, I’m not inexperienced when it comes to bicycles. I have volunteered my time at my local co-op for most of 13 years, 52 times per year doing a 5 hour shift each Saturday. That’s roughly 3300 hours of mechanic time at my co-op with close to 40 years doing my own work on my own bikes. But, it goes further than that. My work at the co-op has been running the Saturday fix-your-bike hours where I’m in charge (and usually the only volunteer) of 6 stands. Each stand is limited to 90 minutes per customer and is usually full throughout the 5 hour shift. That is 20 to 30 bikes I look at every Saturday. That 3300 hours balloons to the equivalent of 13,000 hours or 6.5 years of full time mechanics work. I would also add that a professional mechanic would never work on 6 bikes at the same time or, perhaps more properly, try to train 6 inexperienced mechanics all at once.

Further, the problem with your “facts” is that I disagree that they are, indeed, “facts”. They are at best opinions and speculation. Show some measurements. Show some data. Yes, I can be guilty of speculation…we all are and there is nothing wrong with it… as well but, more often than not, my speculation is backed up by sound scientific principles. The job at which I am actually a professional required me to understand those principles and be ready to back up my ideas, opinion, and speculations with measurements. See below.

I said that the ruler method would show “wear”. I’ve also said that roller wear doesn’t matter if it exists. The roller contribute to the pitch elongation because they move with relation to the pin but the pin wears causing the elongation. The roller is just along for the ride.

Alright. Point taken. It’s not much of an insult but if I offended, I apologize. However, you did call me a “know nothing”.

So you have no answer?

Now, just for giggles, I decided to demonstrate my point. I found 4 chains…a new one, one with <0.5% wear, one with 0.5% wear, and one >0.75% wear. All but the new one were gauged with a Park CC-3.2 chain checker. I then took apart a link and measured the pin thickness and 2 roller thicknesses on each chain.

Here’s the chain tool measurement for each one, followed by a tape measurement. The tape measure was made from 1” to 13” in order to ensure accuracy.

New chain. I didn’t measure it.






The very first thing I noticed was that the tape measurement showed no elongation at <0.5%. I did not hang or stretch the chain but set it on a bench in the same manner each time.



The 0.5% wear chain shows a little elongation but it is less than the 1/16” that would say that it is 0.5% worn. I’d probably still use this chain but it’s starting to get long in tooth.




The third chain was obviously past the 0.75%. There is a gap that I could see through at the end of the tool. This chain I would call severely worn. The tape measurement is a full 1/8” of elongation.

The picture below is just to illustrate how I measured with the tape measure.



An interim conclusion that I would draw is that using a tape measure (or ruler) may not give proper results. But all the above is only prologue.

I took one link from each chain and pulled it apart. I then used a pair of calipers to measure the pin thickness and roller thickness.

New
Roller wall thickness: 1.5mm
Pin diameter: 4mm

0.5% wear (almost)
Roller thickness: 1.5mm
Pin diameter: 4mm at wear point



>0.5 wear
Roller thickness: 1.5mm
Pin diameter 3.7 mm



>0.75% wear
Roller thickness 1.5mm
Pin diameter: 3.6mm

Each measurement of pin diameter was made at the most worn part of the pin. The picture above is of the <0.5% wear pin. Below is the most worn pin on which you can see obvious grooves.



The most obvious conclusion to draw from the above is that there is no roller wear. I measure each roller at several points around the roller and they all…from the new roller to the roller on the most worn chain…gave a wall thickness measurement of 1.5mm. And I’ll admit that I only used a caliper…and a vernier one at that…while a micrometer might have been better but that still would not have shown that much more wear of the roller. The pin on the worn out chain, on the other hand, was clearly worn and no longer round.

Now, I’ve presented my data. If you disagree, go ahead and present your own.

cyccommute 

Jeez, you guys just don’t understand how this all works. It is not up to me to prove your point. You have to prove your own point. Shimano, SRAM, KMC, and a number of others make chain checking tools that why consider accurate enough, precise enough, and repeatable enough for their chains. Here’s what Shimano says about their tool


Here’s what Park says about their tool

​​​​​​​

Nope, I don’t know everything. But that doesn’t mean I know nothing.

Kontact

When we refer to "roller wear" or "pin wear", that doesn't mean that the roller or pin are the specific things that are worn, but that the contact points at the roller or pin have worn. For the pins that's the pin itself and the inside holes of the inner cage plates. For the rollers that includes the outside diameter of the inner cage plates, the inside diameter of the rollers and the outside diameter of the rollers were they bear against the sprockets.

Before you object, this is no different than saying you have bearing play. The play could be from the balls, cone or race. They combine to become play. You don't fail to adjust out the play because you know the balls aren't worn.


And the other factor in all this is that the relative wear of different parts of the chain is variable between models. A KMC might have a harder roller than a SRAM. Narrower chains, for instance, tend to have harder pins than wider chains (one of the reasons some newer chains last longer than 8 speed chains). The advantage of measuring at the rollers is that you don't have to guess if you are measuring the worn part or not - you're just measuring the final product that contacts the sprockets.

cyccommute 

Several issues here. The roller thickness is the same independent of the wear on the chain. If the rollers were wearing on the inside or outside of the roller, the wall thickness would change. The roller most certainly isn’t wearing on the outside diameter of the roller. There’s a possibility that the inner part of the plate is wearing but that is not roller wear. But it is up to you now to make measurements to back up your opinion. I made measurements and saw no differences. The inner edge of the rollers showed no visible difference between the rollers from new to severely worn. Here are pictures of the new chain, the <0.5% wear, and the >0.75% wear. The pin shows obvious wear but the rollers do not in any dimension.







I showed your mine. Now it your turn to show me yours. You can speculate all you like but that’s all it is until you show some data. The wall thickness is the same all across the rollers.

And, again, any measurement is going to show you that. The two different measurements get you to the same place. I agree that the chain checker is easier. I don’t agree that it gives a more accurate measurement just as I don’t agree that the rule method gives a more accurate measurement. If you have something to show that one method or tool is better than the other, by all means, present it. So far all you’ve done is speculate.

​​​​​​​Measure something!

Racing Dan

Depending on what length of chain you are measuring, a few tenth of mm roller wear can be significant.

If the gauge is only 6" = 152.4 mm and your "wear allowance" is 0.5% (153.16mm) then the max total wear you are trying to measure is 0.76mm (~1/32")

In that context adding, or not adding, a few tenth of a mm roller wear IS significant in determining if the chain is done or not, however hardly visible to the naked eye or a tape measure. You need a proper calliper. At least from recollection measuring new vs worn rollers, a worn roller will take up a significant portion of a 1/32" "wear allowance. I shall see if I can find a used roller and post the measurements.

EDIT:

Found a new and a used KMC 10s roller:

OD: New 7.66mm, Used: 7.58mm

Wall thickness: 1.30mm, Used ~1.17mm (but conical and with pronounced hump in the middle that fits in between the worn inner plates.

At first glance this feels like a lot of hoopla for nothing, however the wear in the rollers will add about 0.3 mm to the measurement if measuring between the rollers. - Almost half of your total wear allowance if you consider 0.5% of 6".

Make of it what you will, but Imo, roller wear should be taken into account or excluded all together. Just ignoring it will likely cause you to bin a perfectly fine chain.

(of course measuring a completely shut chain is going to show the same result no matter what method is used, but that's not the point either)

cyccommute 

At the risk of waking the sleeping dog, I measured the chains again with my Pedro chain checker that isolates the rollers. I measured them against the Park chain checker. The first set of pictures is for a new chain. No difference in the measurement.




Then a chain that measures less than 0.5% with the Park Tool. Again, no difference. Pedros says it is good



For a chain that is over 0.5%, the Pedros would probably fail it. Not a big deal actually. You might loose 200 miles.




The chain that has greater than 0.75% fails with both.




Isolating the rollers makes no real difference.

Just for giggles, I also measured the shoulder on the inside of the inner plate that the roller rolls on. It measured out to 5.3mm on the chain that indicated 0.75% wear. It measured the same 5.3mm on the chain with less than 0.5%. While I was measuring, I measured just the outer edge of the roller on the same two chains. I tried to get as far out on the roller as I could and it was the same thickness as I previously measured. There is no roller wear that I could measure and certainly not enough of a difference in any part of the chain that could explain chain wear except for the changes in the pin diameter.