Go Back  Bike Forums > Bike Forums > Bicycle Mechanics
Reload this Page >

Do Chain Checkers really measure the stated %0.5, %0.75 etc. ?

Notices
Bicycle Mechanics Broken bottom bracket? Tacoed wheel? If you're having problems with your bicycle, or just need help fixing a flat, drop in here for the latest on bicycle mechanics & bicycle maintenance.

Do Chain Checkers really measure the stated %0.5, %0.75 etc. ?

Old 12-09-20, 07:18 AM
  #76  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by Bike Gremlin
Makes sense - with one "complaint": 99.9% of the chain gauges don't measure the correct dimension - they are inherently imprecise, by design.
Why no one else makes a tool like Shimano TL-CN42 (i.e. why they make any other chain gauge tool design) is beyond me. Am I missing something?
Pedro's makes Chain Checker Plus, which is a copy of Shimano's predecessor TL-CN41. In any case, such tools are largely not manufactured because the extra benefit they bring in is largely illusory. The elongation of interest is in the distance between the same sides of rollers for subsequent links. Because of this distance being small and representing measurement challenges, distances over a number of links are measured. However, when increasing the link number, all instruments, including TL-CN42, add extra pin to pin distances. For a number of links, the 'right' or 'wrong' side of the roller wear distance elongation contribution shrinks by division over link number. In the limit of a large number of links, you just measure the pin to pin elongation for every instrument/method.

Originally Posted by Sy Reene
Something like the below design from Progold Prolink, though not as compact for fitting in a toolbox, if it were made with the 2 link engagements like the Park or Pedro's might be interesting, as the idea is that it shows wear on a graduated basis.
It is a horribly bad instrument as was discussed before. The idea is OK but maybe difficult to execute.
2_i is online now  
Old 12-09-20, 08:50 AM
  #77  
Sy Reene
Advocatus Diaboli
 
Sy Reene's Avatar
 
Join Date: Feb 2015
Location: Wherever I am
Posts: 8,611

Bikes: Merlin Cyrene, Nashbar steel CX

Mentioned: 14 Post(s)
Tagged: 1 Thread(s)
Quoted: 4717 Post(s)
Liked 1,526 Times in 997 Posts
Originally Posted by 2_i
Well, I am late to the party and most of the important points were already made, starting with the fact that the very question of chain wear is not very precise. In any case, in the past I got a chain checker, ProGold Chain Guage, giving continuous wear values and nominally accurate down to about 0.01%. It was the worst checker ever in my hands, off by 0.7%. I returned it and got my money back
Originally Posted by Sy Reene

Originally Posted by 2_i
It is a horribly bad instrument as was discussed before. The idea is OK but maybe difficult to execute.
I found your reference, but a couple things I'm not clear on. I look at that tool and interpreted the "50%" to mean 'half worn' or maybe equivalent to .5% stretch with another checker? How did you come up with it having a supposed nominal accuracy of .01%?

And, I guess while perhaps that specific brand/model of tool may well in fact be inaccurate, does that necessarily mean that someone else that knows more about machining or manufacturing, couldn't make one of this type of design that was accurate?

Interestingly, the Progold website itself shows a different scale on this tool (1-10 which is "Max"), but no idea how the instructions read:

Sy Reene is offline  
Old 12-09-20, 09:19 AM
  #78  
John_E
Senior Member
Thread Starter
 
Join Date: Dec 2020
Posts: 56
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked 2 Times in 2 Posts
Originally Posted by 2_i

Pedro's makes Chain Checker Plus, which is a copy of Shimano's predecessor TL-CN41. In any case, such tools are largely not manufactured because the extra benefit they bring in is largely illusory. The elongation of interest is in the distance between the same sides of rollers for subsequent links. Because of this distance being small and representing measurement challenges, distances over a number of links are measured. However, when increasing the link number, all instruments, including TL-CN42, add extra pin to pin distances. For a number of links, the 'right' or 'wrong' side of the roller wear distance elongation contribution shrinks by division over link number. In the limit of a large number of links, you just measure the pin to pin elongation for every instrument/method.
I don't think I get what you are saying here. These tools are designed to measure elongation without the addition of roller wear. They assume that roller wear is uniform among all rollers hence when the front and mid rollers are pushed the same direction the roller wear is canceled out.
Another advantage is this design can measure wear for chains with different roller sizes.


Originally Posted by 2_i


It is a horribly bad instrument as was discussed before. The idea is OK but maybe difficult to execute.
Yes the idea is nice but it should be cut precisely.

On the other hand It is not clear what exactly they mean by %100. Does it mean %1 ? or Maybe it corresponds to %0.5 mark. How was it when you checked it against a ruler or other checkers?

Last edited by John_E; 12-09-20 at 09:28 AM.
John_E is offline  
Old 12-09-20, 09:35 AM
  #79  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by Sy Reene
I found your reference, but a couple things I'm not clear on. I look at that tool and interpreted the "50%" to mean 'half worn' or maybe equivalent to .5% stretch with another checker? How did you come up with it having a supposed nominal accuracy of .01%?

And, I guess while perhaps that specific brand/model of tool may well in fact be inaccurate, does that necessarily mean that someone else that knows more about machining or manufacturing, couldn't make one of this type of design that was accurate?

Interestingly, the Progold website itself shows a different scale on this tool (1-10 which is "Max"), but no idea how the instructions read:
I sent that checker back and had to pay the shipping, so financially it did not even make sense. However, I want the seller to feel the pain to help pull a bad product from the market.

In any case, in my memory the full scale was 1% and the marked increments were 0.1%. Given that by eye you can tell 1/10 of the increment, the nominal accuracy for a customer is 0.01%. The problem was that that gauge was showing 0.7% stretch on new chains and I could confirm the expected quality, 0.1% or better, of those new chains with a ruler.

The ProLink gauge looked like made with a laser cutter. I have no idea where the manufacturing problem emerged - maybe in the heating and cooling of the relatively narrow stretch of the metal. You can look up the reviews on Amazon and the issue of readings being completely off is repeated. For me it brought back a vague recollection that falsified weight references were penalized in the remote past with chopped off hands and pondering whether the National Bureau of Standards should get involved in the bike area.
2_i is online now  
Old 12-09-20, 10:43 AM
  #80  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by John_E
I don't think I get what you are saying here. These tools are designed to measure elongation without the addition of roller wear. They assume that roller wear is uniform among all rollers hence when the front and mid rollers are pushed the same direction the roller wear is canceled out.
Another advantage is this design can measure wear for chains with different roller sizes.
You are right that the wear cancels when it is uniform and the measurement is from the same side of the roller. However, the roller wear and diameter differences contribution get suppressed when a measurement is made over a number of links, no matter how these contributions, when different from pin-to-pin elongation, enter. Let us take the roller diameter differences mentioned before, of 0.002". I took a popular Park Tool checker and its hooks are 4.5" apart. We get variation in the apparent elongation from roller diameter difference of 0.002/4.5=0.04%. It is not completely negligible, but when deciding on a 0.7% elongation it nearly is. I do not have TL-42CN, but have TL-41CN. Though the latter may be nominally provide more faithful results than simple checkers, it is a pain to use. When I need to go over several bikes and check the chains at different sample locations along the circumference, I go with the simple Park Tool and it is good enough for me.
2_i is online now  
Old 12-09-20, 01:45 PM
  #81  
Iride01 
I'm good to go!
 
Iride01's Avatar
 
Join Date: Jul 2017
Location: Mississippi
Posts: 14,800

Bikes: Tarmac Disc Comp Di2 - 2020

Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 6099 Post(s)
Liked 4,730 Times in 3,260 Posts
Seems this is all a lot of overthinking on a issue that doesn't change the go no-go decision on a chain by 500 miles to maybe a 1000 miles at the most.

But I guess if y'all had fun thinking and explaining it then that is all that really matters! <grin>

Last edited by Iride01; 12-09-20 at 03:53 PM. Reason: a not and, maybe it should be an... but a works!
Iride01 is online now  
Likes For Iride01:
Old 12-09-20, 03:56 PM
  #82  
DaveSSS 
Senior Member
 
Join Date: Sep 2008
Location: Loveland, CO
Posts: 7,212

Bikes: Cinelli superstar disc, two Yoeleo R12

Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 1090 Post(s)
Liked 554 Times in 443 Posts
If I took chains out of service at .5% elongation, I'd not toss them, just put them aside. If 5 or more chains have been used and new-chain skip has never occurred, it would indicate that you're probably not using each chain long enough and wasting money on chains, just to add a little more life to the cassette. Each of those used, but not worn out chains could be used longer.

All of this depends on how costly the chains and cassettes are. Weight weenies will use $70 hollow pin chains and $350 cassettes on their sram axs 12 bikes. I ride more lowly force axs chains that cost $35 and $185 force cassettes. I have far less invested, but I still don't want to toss either chains or cassettes sooner than necessary. I'll check full length elongation, roller wear and side clearance wear with feeler gages before tossing a chain. None of these wear measurements have a value that's set in stone. I wouldn't toss a chain at .5% elongation if side clearance and roller wear were both good.
DaveSSS is offline  
Old 12-09-20, 04:50 PM
  #83  
davidad
Senior Member
 
Join Date: Feb 2009
Posts: 6,660
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 582 Post(s)
Liked 171 Times in 138 Posts
Originally Posted by DaveSSS
If I took chains out of service at .5% elongation, I'd not toss them, just put them aside. If 5 or more chains have been used and new-chain skip has never occurred, it would indicate that you're probably not using each chain long enough and wasting money on chains, just to add a little more life to the cassette. Each of those used, but not worn out chains could be used longer.

All of this depends on how costly the chains and cassettes are. Weight weenies will use $70 hollow pin chains and $350 cassettes on their sram axs 12 bikes. I ride more lowly force axs chains that cost $35 and $185 force cassettes. I have far less invested, but I still don't want to toss either chains or cassettes sooner than necessary. I'll check full length elongation, roller wear and side clearance wear with feeler gages before tossing a chain. None of these wear measurements have a value that's set in stone. I wouldn't toss a chain at .5% elongation if side clearance and roller wear were both good.
True .5% wear is 1/16" and time to replace the chain.
davidad is offline  
Old 12-09-20, 04:56 PM
  #84  
davidad
Senior Member
 
Join Date: Feb 2009
Posts: 6,660
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 582 Post(s)
Liked 171 Times in 138 Posts
25 or so years ago I bought the Rohloff chain checker. Later I found out that I could go at least 1000 more miles on the chain when I used a tape measure to check my chains instead of the Rohloff. I now have the Pedros copy of the shimano tool and it is fairly accurate. Park has apparently copied this tool.
This thread has gone on to be silly over such a simple topic.
davidad is offline  
Likes For davidad:
Old 12-10-20, 09:29 AM
  #85  
Ev0lutionz
Junior Member
 
Join Date: Aug 2020
Posts: 106
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 76 Post(s)
Liked 11 Times in 7 Posts
Does a generic one work as well? I got one for US 1 dollar.
Ev0lutionz is offline  
Old 12-10-20, 10:20 AM
  #86  
Iride01 
I'm good to go!
 
Iride01's Avatar
 
Join Date: Jul 2017
Location: Mississippi
Posts: 14,800

Bikes: Tarmac Disc Comp Di2 - 2020

Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 6099 Post(s)
Liked 4,730 Times in 3,260 Posts
Originally Posted by Ev0lutionz
Does a generic one work as well? I got one for US 1 dollar.
I'd think a generic tape measure will work as well as a brand name tape measure. May not last as long, but for just measuring, it will work.
Iride01 is online now  
Old 12-10-20, 11:56 AM
  #87  
DaveSSS 
Senior Member
 
Join Date: Sep 2008
Location: Loveland, CO
Posts: 7,212

Bikes: Cinelli superstar disc, two Yoeleo R12

Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 1090 Post(s)
Liked 554 Times in 443 Posts
Originally Posted by davidad
True .5% wear is 1/16" and time to replace the chain.
Don't exclude the opposite situation, where a chain may show .25% elongation, even after 6000 miles of use, but the roller and side clearance wear is overly large. The roller wear at that point can still cause new-chain skip. A Campy chain can do this, which is probably why they suggest using calipers between the outer plates to a maximum length of 132.6mm, rather than measuring elongation alone. Over that short length most of the wear is roller wear.

I started monitoring the elongation on sram axs chains that I had in use that also seemed to elongate very little, but now I've got two bikes that need 55 inch chains, so my previous 54 inch chains can't be used and I'm starting over. I will not be using only one chain on each bike. I much prefer using several chains in a rotation.

I've read reports of axs users with 6000 miles on a chain that still checks OK, but those users don't monitor side clearance or roller wear. Most likely, their second axs chain will skip and their $350 cassette will have a very short life. I've also read of outer plates cracking after 3000 miles.

https://weightweenies.starbike.com/f...2596&mode=view
DaveSSS is offline  
Old 12-10-20, 01:07 PM
  #88  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by DaveSSS
Don't exclude the opposite situation, where a chain may show .25% elongation, even after 6000 miles of use, but the roller and side clearance wear is overly large. The roller wear at that point can still cause new-chain skip. A Campy chain can do this, which is probably why they suggest using calipers between the outer plates to a maximum length of 132.6mm, rather than measuring elongation alone. Over that short length most of the wear is roller wear.

I started monitoring the elongation on sram axs chains that I had in use that also seemed to elongate very little, but now I've got two bikes that need 55 inch chains, so my previous 54 inch chains can't be used and I'm starting over. I will not be using only one chain on each bike. I much prefer using several chains in a rotation.

I've read reports of axs users with 6000 miles on a chain that still checks OK, but those users don't monitor side clearance or roller wear. Most likely, their second axs chain will skip and their $350 cassette will have a very short life. I've also read of outer plates cracking after 3000 miles.

https://weightweenies.starbike.com/f...2596&mode=view
I did an experiment to see (confirm) if roller wear affects chain engagement. It doesn't. Even when taking the rollers completely out, the chain engages cassette properly. Only once there is pin wear (i.e. chain elongation) does a chain start to ride higher on cassette teeth, not engaging properly.

As for the wear "location" - elongation occurs between two pairs of outer plates, while pins on the same outer plate will stay at the same distance regardless of the chain wear, of course.
Bike Gremlin is offline  
Old 12-10-20, 01:24 PM
  #89  
Racing Dan
Senior Member
 
Racing Dan's Avatar
 
Join Date: Aug 2015
Posts: 2,224
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1332 Post(s)
Liked 316 Times in 214 Posts
Originally Posted by John_E
Since you were a machinist I will appreciate your expertise here.

How would you use a digital caliper that has a max length of 6" to measure the elognation (no roller wear)? Where would you take the measurement?

In theory there are many places. But once I tried I realized that aligning the caliper to take consistent measurements at that distance is hard.

My final try was to measure the distance from the end of one outer plate to the beginning of the 6th one. Then I subtract the distance between the edges of two consequent ones.
The distance between two consecutive plates seem to be around 0.172 (I will assume 0.17 to not underestimate the distance). Now new one under tension measured 5.172-5.173, the one that has 600+ miles on it measured 5.173-5.174. The one that shows barely %0.5 wear measured ~5.182. All of these chains are same model KMC chains. Well now it seems the chain which is supposed to have ~%0.24-0.25 wear is shown to reach %0.5.
The only consistent reading I have compared to the chain checker is for the 10 speed chain that is over %75. My readings were 5.208-5.212 and that translates into %0.76-%0.84 wear.
The ruler also does not show anywhere close to %0.5 but of course it relies on my eyesight. I am not starting from the center of a pin since it is impossible to really know where the center is , instead I start from the edge of a pin.

Because I am a new member I can post pictures which would have helped here.
Don't measure, use the calliper as a gauge. Pre set it at whatever length you believe is the max allowable length and tighten the small finger screw. Then its easy to judge by eye if the chain is more or less elongated than that.
Racing Dan is offline  
Old 12-10-20, 01:48 PM
  #90  
Racing Dan
Senior Member
 
Racing Dan's Avatar
 
Join Date: Aug 2015
Posts: 2,224
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1332 Post(s)
Liked 316 Times in 214 Posts
Originally Posted by 2_i
You are right that the wear cancels when it is uniform and the measurement is from the same side of the roller. However, the roller wear and diameter differences contribution get suppressed when a measurement is made over a number of links, no matter how these contributions, when different from pin-to-pin elongation, enter. Let us take the roller diameter differences mentioned before, of 0.002". I took a popular Park Tool checker and its hooks are 4.5" apart. We get variation in the apparent elongation from roller diameter difference of 0.002/4.5=0.04%. It is not completely negligible, but when deciding on a 0.7% elongation it nearly is. I do not have TL-42CN, but have TL-41CN. Though the latter may be nominally provide more faithful results than simple checkers, it is a pain to use. When I need to go over several bikes and check the chains at different sample locations along the circumference, I go with the simple Park Tool and it is good enough for me.
You forget to account for internal wear in the rollers. The hole in the rollers get much bigger as the chain wears, greatly contributing to "apparent elongation" if measured between the rollers. - Just take apart a worn out chain and have an epiphany. I like measuring between the rollers too, but I do keep in mind that the rollers wear too and thus allow for much more "apparent elongation" than 0.5%, as outlined in a previous post.
Racing Dan is offline  
Old 12-10-20, 02:07 PM
  #91  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by Racing Dan
You forget to account for internal wear in the rollers. The hole in the rollers get much bigger as the chain wears, greatly contributing to "apparent elongation" if measured between the rollers. - Just take apart a worn out chain and have an epiphany. I like measuring between the rollers too, but I do keep in mind that the rollers wear too and thus allow for much more "apparent elongation" than 0.5%, as outlined in a previous post.
If the rollers wear by the same percentage as pin to pin distance stretches, a simple chain checker will give the consistent result. It is only the difference in percentage wear that potentially throws off the reading, but that will be suppressed by division by the link number in the measured interval. At that level I do not care. I only check the chain at 2 month intervals and this starting around the time I expect the chain to expire. This in itself may let me miss the threshold by 0.1%. There is no point on insisting on precision in one part of the system while another is flaky.
2_i is online now  
Old 12-10-20, 02:13 PM
  #92  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by Bike Gremlin
I did an experiment to see (confirm) if roller wear affects chain engagement. It doesn't. Even when taking the rollers completely out, the chain engages cassette properly. Only once there is pin wear (i.e. chain elongation) does a chain start to ride higher on cassette teeth, not engaging properly.

As for the wear "location" - elongation occurs between two pairs of outer plates, while pins on the same outer plate will stay at the same distance regardless of the chain wear, of course.
Thanks for the experiment. However, in my memory people claimed that roller wear affected shifting. I am sure that there is more as the rollers play a role in a dynamic engagement of the chain into and out of the teeth and friction. Have you got any comments on these? It is interesting as a principal problem - in practice I will be replacing chain at least as conservatively as now.
2_i is online now  
Old 12-10-20, 03:52 PM
  #93  
Gresp15C
Senior Member
 
Gresp15C's Avatar
 
Join Date: Sep 2014
Posts: 3,893
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 1062 Post(s)
Liked 665 Times in 421 Posts
Originally Posted by Sy Reene
Interestingly, the Progold website itself shows a different scale on this tool (1-10 which is "Max"), but no idea how the instructions read:
I like that design. You can recalibrate it to any scale you want with a calipers.
Gresp15C is offline  
Likes For Gresp15C:
Old 12-10-20, 05:53 PM
  #94  
Racing Dan
Senior Member
 
Racing Dan's Avatar
 
Join Date: Aug 2015
Posts: 2,224
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1332 Post(s)
Liked 316 Times in 214 Posts
Originally Posted by 2_i
If the rollers wear by the same percentage as pin to pin distance stretches, a simple chain checker will give the consistent result. It is only the difference in percentage wear that potentially throws off the reading, but that will be suppressed by division by the link number in the measured interval. At that level I do not care. I only check the chain at 2 month intervals and this starting around the time I expect the chain to expire. This in itself may let me miss the threshold by 0.1%. There is no point on insisting on precision in one part of the system while another is flaky.

From my experience rollers wear faster much than true elongation. For that reason Im not hesitating letting the "between rollers" measurement grow by as much as 1mm between 12 rollers, from ~132,25 -> 133,3mm, as explained in an earlier post. That is more like 0.8%, but at that point true elongation is still less than 0.5%. Beyond that point the chain may technically not have reached 0.5% elongation, but imo shifting and "smoothness" is impacted and id rather just replace it regardless.

Again, no one knows what the old Park, and other gauges that measures between rollers, considers "+0.5%" or "+0.75%" but if they are anything like the campy recommendation, 132.6mm, you be replacing chains way too soon. - Some one needs to measure the gauges before we can judge if they give a reliable indication if the chain is ok or worn out, or they just make you bin your chain way early to play it safe or not enough thought went into the design. Heck, no one even knows if "+0.5%" is the same thing on different gauges.

Also who made "+0.5%" the gold standard for 11s drive trains and for what reason. +1.0% used to be the engineering standard for any chain.

For the above reasons I went with a digital calliper and made up my own rules based on experience. Subject to be revised along the way.

Last edited by Racing Dan; 12-10-20 at 06:15 PM.
Racing Dan is offline  
Old 12-10-20, 06:59 PM
  #95  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by Racing Dan
From my experience rollers wear faster much than true elongation. For that reason Im not hesitating letting the "between rollers" measurement grow by as much as 1mm between 12 rollers, from ~132,25 -> 133,3mm, as explained in an earlier post. That is more like 0.8%, but at that point true elongation is still less than 0.5%. Beyond that point the chain may technically not have reached 0.5% elongation, but imo shifting and "smoothness" is impacted and id rather just replace it regardless.

Again, no one knows what the old Park, and other gauges that measures between rollers, considers "+0.5%" or "+0.75%" but if they are anything like the campy recommendation, 132.6mm, you be replacing chains way too soon. - Some one needs to measure the gauges before we can judge if they give a reliable indication if the chain is ok or worn out, or they just make you bin your chain way early to play it safe or not enough thought went into the design. Heck, no one even knows if "+0.5%" is the same thing on different gauges.

Also who made "+0.5%" the gold standard for 11s drive trains and for what reason. +1.0% used to be the engineering standard for any chain.

For the above reasons I went with a digital calliper and made up my own rules based on experience. Subject to be revised along the way.
With the wear self-accelerating I think that the primary purpose of checking the wear and replacing chain is to avoid getting into a runaway situation. I.e., the extra time/distance that you get to ride from 0.5% to 0.7% is well shorter than from 0.1% to 0.3%. The common expectation is that once you get to 1% is that you likely need to change the cassette even if it is on its first chain. To test the above one would need to map chain elongation against covered distance from the time when the new chain was put on.

Indeed the various 0.5% or other markings could mean anything, including accounting for the estimated roller wear. I went to measure the increase in threshold distance for the simple Park CC-3. Between the 0.75% and 1% markings the distance increases by 0.21% rather than the expected 0.25% or higher if they were accounting for the stronger roller wear.

I never paid much attention to roller wear as isolated from pin-to-pin stretch and may look into that from now on. I even have a relatively new chain on the main bike and I usually have another identical chain in storage that I could use in comparison. I also have various measuring instruments and could measure virtually whatever I would want. However, the issue is of how much time to dedicate to that. The chain is central to the bicycle, which makes it an interesting problem, but there are so many other problems in the world. I skip chain cleaning because it is time consuming bother - it also elevates chain replacing for me somewhat.
2_i is online now  
Old 12-10-20, 08:43 PM
  #96  
DiabloScott
It's MY mountain
 
DiabloScott's Avatar
 
Join Date: Sep 2006
Location: Mt.Diablo
Posts: 9,991

Bikes: Klein, Merckx, Trek

Mentioned: 70 Post(s)
Tagged: 0 Thread(s)
Quoted: 4310 Post(s)
Liked 2,953 Times in 1,601 Posts
Originally Posted by 2_i
If the rollers wear by the same percentage as pin to pin distance stretches, a simple chain checker will give the consistent result.
Chain checkers only see the "slop" in the two rollers they are inserted between, not every roller in the span between them.
So if your tool measures 6 links of unstretched chain, it'll see 3 inches plus two roller slops.
And if your tool measures 12 links of unstretched chain, it'll see 6 inches plus two roller slops.
DiabloScott is offline  
Old 12-11-20, 03:24 AM
  #97  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by 2_i
Thanks for the experiment. However, in my memory people claimed that roller wear affected shifting. I am sure that there is more as the rollers play a role in a dynamic engagement of the chain into and out of the teeth and friction. Have you got any comments on these? It is interesting as a principal problem - in practice I will be replacing chain at least as conservatively as now.
Based on my theoretical knowledge and practical experience (in other words: "to the best of my knowledge", or "as far as I know"):
as a chain gets worn, at the inner plate - to pin interface, it gets more easily bent sideways (shown in the last picture of chapter 4 in this article, and the video linked above talks about it).
This makes it more likely to just "bend", when a derailleur pulley pulls/pushes it sideways, instead of staying more straight, following that movement, and making a gear change.

I don't think roller wear has anything to do with that.
Bike Gremlin is offline  
Old 12-11-20, 08:54 AM
  #98  
2_i 
Senior Member
 
Join Date: Sep 2005
Location: Michigan
Posts: 3,692

Bikes: Trek 730 (quad), 720 & 830, Bike Friday NWT, Brompton M36R & M6R, Dahon HAT060 & HT060, ...

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked 332 Times in 247 Posts
Originally Posted by Bike Gremlin
Based on my theoretical knowledge and practical experience (in other words: "to the best of my knowledge", or "as far as I know"):
as a chain gets worn, at the inner plate - to pin interface, it gets more easily bent sideways (shown in the last picture of chapter 4 in this article, and the video linked above talks about it).
This makes it more likely to just "bend", when a derailleur pulley pulls/pushes it sideways, instead of staying more straight, following that movement, and making a gear change.

I don't think roller wear has anything to do with that.
Thanks, this is what I thought it was. However, I think that the situation with sideways flexibility is different, in that there is no single reference there, but rather some tolerance interval. When you have a stuck link and work to make it operational, you are working to increase the sideways play. I once got a Wippermann chain and it was horrible on my drivetrain yielding missed rear shifts, etc. The problem that I traced the issue to was that it was too stiff laterally. Sram chains that I used there before were quite flexible and worked great in the same configuration. I was pushing it a bit as I was riding custom cassettes at the time and had to go back to a commercial one just to get through that miserable Wippermann chain, but I never went to the brand again.
2_i is online now  
Old 12-11-20, 09:07 AM
  #99  
DaveSSS 
Senior Member
 
Join Date: Sep 2008
Location: Loveland, CO
Posts: 7,212

Bikes: Cinelli superstar disc, two Yoeleo R12

Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 1090 Post(s)
Liked 554 Times in 443 Posts
Originally Posted by Bike Gremlin
I did an experiment to see (confirm) if roller wear affects chain engagement. It doesn't. Even when taking the rollers completely out, the chain engages cassette properly. Only once there is pin wear (i.e. chain elongation) does a chain start to ride higher on cassette teeth, not engaging properly.

As for the wear "location" - elongation occurs between two pairs of outer plates, while pins on the same outer plate will stay at the same distance regardless of the chain wear, of course.
My experiment was using Campy 10 chain on a new cassette for 6000 miles. It showed less than 0.25% elongation, properly measured over the full chain length. The roller wear was huge and so was the side clearance wear. It caused new-chain skip when a new chain was installed. The wear pockets created by the worn rollers caused this, not chain elongation. You may also find that the sprockets that skip with a new chain will not skip using a chain with only a few hundred miles of break-in wear. That can allow a cassette to be used another 3000 miles instead of being tossed.
DaveSSS is offline  
Old 12-11-20, 09:22 AM
  #100  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by DaveSSS
My experiment was using Campy 10 chain on a new cassette for 6000 miles. It showed less than 0.25% elongation, properly measured over the full chain length. The roller wear was huge and so was the side clearance wear. It caused new-chain skip when a new chain was installed. The wear pockets created by the worn rollers caused this, not chain elongation. You may also find that the sprockets that skip with a new chain will not skip using a chain with only a few hundred miles of break-in wear. That can allow a cassette to be used another 3000 miles instead of being tossed.
I still think it has nothing to do with the diameter of the rollers (i.e. roller wear). It has to do with the chain pitch.
If you measure a brand new chain, you might notice it having under 0.5" pitch, on average, because the factory grease won't let the inner plates get all the way to the pins - which gets sorted out after a very short ride.
Bike Gremlin is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service -

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.