John_E

This is how I took my latest measurement. If someone who has experience with taking consistent measurements can comment on whether this is safe and accurate way to take measurements that would be great.

Gresp15C

Looks good to me. The 0.172 figure could vary from chain to chain, but you can always re-measure it each time you check a chain.

The sources of uncertainty in a chain measurement fall into two categories: Some are constant, no matter how many links you include, for example the "slop" of the rollers and typically the precision of the measurement. Some are proportional to the number of links, such as the overall elongation. For this reason, most of the things you don't want to measure become insignificant, the more links you include.

This is why all of the measurements and gages are longer than just one or a pair of links. Six inches makes sense for go-no-go gages and inexpensive calipers that typically come in a 6" length and are a convenient size to throw in a toolbox. Twelve inches makes sense for a ruler. Anything longer gets unwieldy and might require removing the chain.

One thing I always do is double check in the following way. When I think I'm ready to replace a chain, I hang the old and new chains side by side. Over the full length of a chain, the elongation becomes pretty obvious. I keep the published limits (0.5%, 0.75%, whatever) in mind, but I might also swap a chain at the moment when it happens to be convenient, such as getting the winter bikes ready for a snowstorm that's supposed to arrive soon. A benefit of being an amateur is that we can often make decisions based on what's convenient for our schedule, rather than having to get a bike on and off the stand ASAP so another bike can be brought in.

DaveSSS 

I never use a chain straight out of the box with the grease on it. I clean them of grease before using. At the time of this test, I was using an extremely thin wet lube that's like Prolink lube - mostly solvent and no more than 25% oil. It would not affect the chain pitch. I have measured new chains that are actually a bit short on overall length. I have a AXS chain that's still a little short after 1000 miles of use.

If you have chain skip with an old cassette and new chain at the start of a ride, it should quit by the end of a ride, if it's just the grease that's the problem. If not, keep using the new chain until it quits skipping, if you can stand it.

Racing Dan

The rollers wear much more on the inside than the outside, ie the hole in the roller increases much more than the diameter shrinks. You need to take apart, and measure the parts of a worn out chain and compare to a new one, to understand whats going on.

davidad

But roller wear doesn't matter. They aren't a precision to begin with.

Sy Reene

Actually, it's supposedly the wear inside the bushings, not the rollers.
Here's a nice article:
https://www.bikeradar.com/features/h...bicycle-chain/

Racing Dan

I didnt say they are. All im saying is rollers wear (mostly) from the inside out and contribute greatly to "apparent" chain elongation, if measured between the rollers.

Weather roller wear matters or not, i dunno, - matters how? I do know however, a chain with worn rollers doesn't ride as nice as a new one. For that reason I tend to replace the chain when the rollers have noticeable wear, no matter if the chain is truly elongated or not. Measuring between the rollers will disclose if the chain is either truly elongated, if the rollers are worn or any combination there of. That said I'm fully aware I might bin the chain somewhat prematurely if the objective is max chain and cassette life, but i just don't care that much. - I'm still on tiagra 10s, its not that expensive.

davidad

It ain't rocket science. I said that roller wear is not a factor in chain life and it isn't. When the upset bushings formed on the inner links and the pins wear the chain gets longer. When the chain measures 12and 1/16" in 12 inches it is time to replace it or the rings and cogs will begin to wear and need replacing.

John_E

You are right about outer plate distance and that's why I measured it. I should have stated that 0.172 is for the 11 and 10 speed KMC chains I had, the 8 speed chain has ~0.165 distance between outer plates.

I will try to find a spot to hang almost new and measured %0.5 chains next to each other when I have the time. But there is another problem with looking at the whole chain. The chain unfortunately does not wear uniformly so even if the total wear is less than %0.5 it may be possible to have a section above that point and that begs the question does it matter to have a 5-6" section over %0.5 if the average is below that ?
.

DaveSSS 

Campy recommends changing a chain if any of several measured areas exceeds their 132.6mm length suggestion.

It's easy to use a precision 12 inch rule in several places, if the chain is already on a bench, for a full length measurement, but there's no need for that, unless the chain is getting close to being worn out.

0.5% is not a magic number that will cause a disaster if exceeded. Supposedly, sprocket wear begins to accelerate at that point. If you're changing chains prematurely in hopes of avoiding new-chain skip, about the worst that can happen is one less chain gets used before new-chain skip occurs. If that happens, a new cassette can be used for a few hundred miles, then the old one swapped back for the remaining life of the new chain. Some other chain with some use on it can also be used on that worn cassette and it won't skip. That’s why I prefer using a chain rotation. There is no need to keep track of mileage on the chains and managing them is simple. Hang them on a peg and always use the first in line. A freshly cleaned and lubed chain goes to the back of the peg. I drape mine over the peg about in the middle rather than from the end, so it's easy to place a chain at the back of the line.

Racing Dan

Campys 132.6 mm recommendation cant* be measured with a precision rule, you need a calliper. Look at post 52 to see for yourself.

Id argue the 132.6 mm rule only apply to campy chains or is simply wrong altogether.

*EDIT

DaveSSS 

As an experienced machinist, I'm well aware that calipers are required for Campy's measurement and I never mentioned making that measurement with a rule. Not only are calipers required, most of the tips won't fit between the outer plates. I ground mine down so they will work. If you read earlier posts, the length increase is just over .6% percent. It the same measurement that most early model chain checkers make - adding roller wear to actual change in pitch. The reasoning is sound because a Campy chain will never reach .5% elongation. My Campy 10 chain with 6,000 miles on it had no more than .25%, but the roller wear caused new chain skip on what was initially a new cassette. The chain should probably been changed at 3000 miles.

I also noted the new SRAM AXS are showing this type of wear. I'll be checking roller spacing with calipers, side clearance wear with feeler gages and full length elongation on the eight chains I have. When I use a 12 inch rule, it's done as I described earlier - from the edge of a pin to the very end of the rule. The pin at the far end will be completely covered with a new chain. If half of that pin is exposed, its a little over .5% elongation.

Racing Dan

132,6 implies there is zero initial slop in the rollers and/or the rollers are bigger that KMC and Shimano. I cant believe any of this is true and in the absence of actual posted dimensions, i remain convinced something is up with that particular recommendation. In any case it does NOT apply to either KMC or Shimano, - You be binning your chains waay prematurely. Might as well simply bin it every 1500 mi, and be done with it.

DaveSSS 

To use calipers, you have to also be smart enough to first measure a clean new chain, then add at least .5% to that length. That measurement will still be incorrect, because it adds roller wear to actual elongation. To measure only elongation, you should also measure the roller spacing on a new chain and subtract any increase from a worn chain measurement. Then you're measuring only elongation. A 12 inch precision rule is a lot easier to use.

I've explained why Campy suggests their method, but it's really only relevant to their chains. I never claimed it should be used for other brands. A well maintained Campy chain will elongate very little, even after 6,000 miles, but if only one chain is used for that long, it will cause new-chain skip.

sweeks

Here's that process (new chain on the right). There's no doubt about the elongation. FWIW, the old chain was flagged by three different chain checker tools.

Chains hanging from nails on a level line



Worn chain on the left



Three chain-check tools. The Shimano tool (right) is supposedly not influenced by roller wear.

Racing Dan

Dude, lol!

I used to be a machinist too. I know how to use a calliper, and most other measuring tools related to that trade. Measuring a new chain I exactly what I did and thats the reason I know the 132.6 mm figure makes very little sense. Obviously, if a campy chain is completely different from a Shimano or KMC chain it might make sense in that context, but since no one posted the "new chain" dimensions of a campy chain I remain unconvinced its even a reasonable figure for a campy chain. Why dont you measure one and post the numbers?

Skulking

When the 132.6 measurement was first posted, I did the calculations and it doesn't seem off to me. I didn't do any actual measuring, so this is completely theoretical based on what the dimensions should be. They tell you to measure over 6 outer links which means a chain length of 5.5" (139.7mm), and because they are telling you to measure the interior distance between the rollers, you need to subtract half of the roller diameter at each end for a total of 5/16" (7.9375mm). That brings the new chain length to 131.7625mm. As I said, that is all calculated assuming all dimensions are perfect and there is no initial play between parts, but with that calculation you end up with 132.6mm being 0.64% stretch. If you see anything flawed with my calculations I'd love to know what it is.

Racing Dan

I appreciate you effort but you need to measure a real chain to account for actual roller diameter and slop in the rollers. Its there even if the chain is brand new and the rollers cant be assumed to adhere to any posted standard. If memory serves me right a new KMC 10s roller more like 7.65 mm. That is a significant difference form any nominal diameter, if the difference between a OK or Worn out chain is only a few 1/10 mms.

Skulking

If I had a Campy chain to measure roller diameter I would, but without one all I can go by is nominal measurements based on the standard. Obviously any deviations from the nominal dimensions would become very significant when dealing with such precise measurements. And there must initially be some slop in the rollers or they would be completely bound up. But it seems to me that that calculation is likely where they came up with their 132.6mm number.

Racing Dan

I agree and that is the very reason I believe its wrong :-)

DaveSSS 

Since bike chains are based on English units of measure, I prefer to use inch measurements. Campy's 132.6mm dimension converts to 5.220 inches. With the maximum allowed roller diameter of .306, .305 is a common size, but you may find some as small as .303. Take the .500 nominal pitch and subtract one roller diameter to get the nominal space between the rollers. That comes to .195 inch, if there was no clearance between the hole in the roller and the pin. My calipers find this dimension to be .205 inch, so that means that the holes are .010 inch larger than the pin. A new chain should measure about 5.205 inches between the rollers, which makes the 5.220 dimension only .3% longer. .5% would be 5.231 and .6% would be 5.236. My no-go gage made from a 6mm hex wrench measures an increase in the roller spacing of about .5%. By this time, the side clearance on the chain will be a lot greater than a new one. A new chain will usually have .004-.008 inch of clearance between the inner and outer plates. Twice the original clearance indicates a lot of wear.

Personally, I've never paid any attention to the Campy dimension after using Campy chains for 25 years. I did learn that these chains do not elongate very much, whether measured with a precision 12 inch rule or a full length measurement, so checking elongation is usually worthless. I have had others report significant elongation, but you never know how well maintained the chain is or what kind of conditions it was used in. I learned to measure the change in the roller spacing and the side clearance wear to decide when to change a chain, since the elongation measure was always under .25% on my chains.

I'm into a whole new ball game with SRAM AXS chains, but they aren't much different. They follow the #40 chain standard, with 5/16 (.3125) maximum sized rollers on the same .5 inch pitch. I've used them on cassettes and chain rings made for the #41 chains and haven't had a problem yet.

Racing Dan

That is almost exactly what i predicted 70(!) posts ago in #52. :-) Its useless.

DaveSSS 

Yes, but saying something and proving it are two different things. If followed, this dimension would sell lots of chains. If only elongation was measured, the chain would easily last 6,000 miles, but the second chain used on the cassette would skip. Somewhere in between there's a happy medium.

That's what I'll be doing with the axs chain. It only measures a few thousandths of an inch less than a #41 chain, between the rollers. I'll check roller spacing with calipers and elongation full length, since it shows an increase sooner.

pdlamb

I would be SO dead if I tried that! But it looks like a fairly easy way to check the chain -- aside from taking it off the bike to measure.

sweeks

Hahaha! My wife was none to happy about that. At least I didn't leave grease all over the door!