Longterm chain/cassette/chainring MTB wear-out test

robobiking

Hello,
My name is Ronald form the Netherlands and I want to share a long term test I started.

I already do a longterm test on my commuter bike in the same way.
But after I got many questions about doing the same test for MTB I contacted my sponsors/supliers,
If they would like to cooperate and they did.

This test will start with these parts:
· Miche cassette 12-29
· 3 chains (Miche) 10speed
· Wolftooth Componants (Elliptical) 38T BCD110 Chainring
· Lube (Pedro’s)

I ride 50km MTB a week, and I will Blog about this test every 2 weeks and post you all updates.
My goal is to see how long I can do with a cassette, chainring and chain.
The chains will be cycled after every 100/200km depending on the wearout.
And I will use my knowledge from my other wearout test for this test to.

These are some of the things that I will do after every ride:
Clean the chain with a cloth, or when really dirty with degreaser.
Lube is after every ride if needed
The chains will be numbered 1, 2 & 3. And will be changed every 2 weeks (after +/- 100/200km).
I will measure the chain every week with a digital calliper gage.

I will be posting my weekly updates here so you guys/girls can follow this if you will.

Bye
Ronald vd Boon
http://robobiking.wordpress.com

The Rookie

As MTB wear is so very dependant on the type of terrain and more specifically the conditions you ride in, your test I'm afraid will be irrelevant to anyone but someone riding where you ride.

cooldad

I still have the original cassette and chainrings on a 1987 Specialized Rockhopper. It's probably done about a million miles.

Doubt many people outside really flat countries would manage with a 38/29 neither.

robobiking

The Rookie wrote:

As MTB wear is so very dependant on the type of terrain and more specifically the conditions you ride in, your test I'm afraid will be irrelevant to anyone but someone riding where you ride.

I see your point, of course terrain type is differant in the Netherlands. But dirt is dirt, rain is rain and sand is sand. All these conditions we have here too. The hills in England will be more defiant with rocks (no we don't have these in the Netherlands). But this test could give a great insight in chain/cassette/chainring wear-out. I will monitor all te conditions I rode in for a full review.

Hope you find something in it.

robobiking

cooldad wrote:

I still have the original cassette and chainrings on a 1987 Specialized Rockhopper. It's probably done about a million miles.

Doubt many people outside really flat countries would manage with a 38/29 neither.

That's cool, I think parts from that era lastet longer because they where build to last.

I drive a 38T and a 29T in the back because I can get up every hill here in the Netherlands with it. Even the steep ones.

cooldad

Isn't the highest hill there like about 3 inches?

The Rookie

Dirt isn't dirt, sand causes much more damage than loamy soil due to the silicate granules.

robobiking

cooldad wrote:

Isn't the highest hill there like about 3 inches?

Well the Netherlands isn't totally flat, I live near the beach where we have lots of dunes. The highest is 52meters.
Ofcourse that's nothing compaired to Great Britan or other country's.

The Rookie wrote:

Dirt isn't dirt, sand causes much more damage than loamy soil due to the silicate granules.

Your right that's true, what I ment to say was that this test could give some insight in wear-out. Even if the terrain where I ride is differant than other people. I ride in the dunes most of the times, so lots of sand there.

davidradar

Great work Ronald. It will be good to see your results!

robobiking

DavidRadar wrote:

Great work Ronald. It will be good to see your results!

Thanks

robobiking

Here's Part 1:

https://robobiking.wordpress.com/2016/0 ... ing-150km/
bye
Ronald

cooldad

"Oops! That page can’t be found."

Excellent.

robobiking

cooldad wrote:

"Oops! That page can’t be found."

Excellent.

Thanks for telling me, it's fixed now.

cooldad

And wow.

For all intents and purposes, isn't ±132.25mm effectively the same as 132.2mm?

What are you measuring 5 100ths of a mm with?

robobiking

You can look at it that way, yes. But for the course of the test I desided to write it this way.

I measure the chain with a digital caliper. And because the wear of the chain is not even on the chain I take average of the measurements I do. And it's just an indication, nothing sceintific.

cooldad

Nope, no idea what you are even measuring. That's 13.2 cm. Or about 5 and a bit inches. Very tiny bike?

robobiking

Aha, I see what you mean. I thought I already mentioned that.
This is how I measure it, between 5 chain-links:
DSC_0449 by Ronald vdBoon, on Flickr

Excuse me for the haziness

cooldad

Including any sag in the chain - It matters if you are playing at 100ths of a mm.

Where's the number on the chain?

Very shoddy science.

robobiking

It's this one:
20160722_120848 by Ronald vdBoon, on Flickr

What kind of sience you would like to see in this test? Maybe we should discus this via PM.

cooldad

I don't really want anything, just bored at work, but you said the chains would be numbered - I see no number. So how do you know which number it is.

And if you want to measure in immeasurably small amounts, you have to do better than a caliper and a vague look.

If you want to sound scientific, at least make the effort.

Herdwick

measuring a chain with a vernier caliper is simply stupid, like that you measure the slackness/tollerance of the chain's rollers, and it's added to the totall 'growth' of the 5 links you measured. how do you define tolerance vs wear?
and 5 hundredts of a mm is out of the accuracy range of the tool you are using anyway

robobiking

cooldad wrote:

I don't really want anything, just bored at work, but you said the chains would be numbered - I see no number. So how do you know which number it is.

And if you want to measure in immeasurably small amounts, you have to do better than a caliper and a vague look.

If you want to sound scientific, at least make the effort.

Herdwick wrote:

measuring a chain with a vernier caliper is simply stupid, like that you measure the slackness/tollerance of the chain's rollers, and it's added to the totall 'growth' of the 5 links you measured. how do you define tolerance vs wear?
and 5 hundredts of a mm is out of the accuracy range of the tool you are using anyway

Thanks for the comments, I can see your concern/doubt in how I handle this test.
Let me explain some things.

1. I use the digital caliper just as an indicator, I could have use a chainchecker but this would only give me to now when the chain is worn. I don't want this test to be scientific in any way. My goal is to check in a simple way how the chain is doing in wearing out. I could have just make a board with 3 nails and hang them side by side and always use the shortest. But have chosen this methode.
2. I don't number the chains. I have 3 zip bags which I have numberd. I put them in there corresponding bag after I switch chains.

But since we are debating here, how would you guys do this?

Best regards
Ronald

cooldad

I wouldn't. Life's too short.

robobiking

8)

keef66

The German Tour magazine did a proper chain wear test a few years ago; proper engineering stuff with a bench rig, repeatable,controlled conditions, contamination with sand to accelerate wear, very accurate measurements. Seem to remember the Shimano 10 speed chains were pretty good, which gave me a nice warm glow since that's what I use.

Not sure your daily riding in random weather, anecdotal observations and measurements over 5 links is going to tell you very much, but if it keeps you amused then carry on!

robobiking

Hello.
Here's an update on the test.
I made a "measuring strip" where I can put the 3 chains on at once. They are at tension by springs.
This way I can see the wearout of the chains side by side.

https://robobiking.wordpress.com/2016/0 ... ing-300km/

Bye

rockmonkeysc

You cannot measure accurately to less than a mm unless you're going to take the chain off the bike and ensure that its dead straight. If you're going to measure to a hundredth of a millimeter then you'd need to do it in a temperature controlled environment and leave the chain in there for several hours before measurement. Your caliper may read to hundredths of a millimeter but accuracy of measurement depends on conditions.
Most riders on here do more than 50km a week and don't rotate chains so your test condition is not realistic.
As mentioned above, drive component wear depends massively on environmental conditions and the type of dirt.

keef66

To be fair his new test rig with the chains off the bike and held under tension by springs is a lot better. You can clearly see the relative wear compared with a new chain, even if the absolute measurements are a bit suspect. Looks like his favourite trails consist almost exclusively of sand, so it's quite a stiff test of chains / lubes

jdayrutherford

I have a problem. Running a Campy super record from 2009. No problems until I changed the cassette and chain, both
new. After that, I get a lot of noise when I pedal. Not really sure if its coming from the cassette or chainrings. Seems like Cassette. Really not rideable. BUT, this only happens in the small ring (50/34). If I put the old, stretched chain back on it
does not happen. The chain is stretched out according to the park tool. I ride in the mountains only. Almost never use
the big ring except on downhills. So most of the wear is on the 34 ring I would guess. The only thing I can think of is that
the small chainring is causing the problem. But I hate to spend the money to replace it (and in a 2009 campy setup this is not cheap as it apparently requires changing both rings and bolt set) without knowing if this is really the problem. Any ideas?

steve_sordy

I've just found this thread and I believe that robobiking is being given an unnecessarily hard time. Measuring to 0.05mm is easily within the capability of a vernier caliper. After all, it is almost 0.002" which is quite large in engineering terms as regards accuracy. When I was an engineering apprentice we were expected to measure to at least 0.0005" which is 4x more precise. If you know what you are doing it can be done reliably. I do not take exception at all to robobiking measuring to 0.05mm, not one bit.

When I measure a chain, I measure it brand new to start with to establish a baseline. I measure over a slightly longer length than he does. I measure inside 12 half links so the dimension is 6" less the roller diameter, which is about 5.7" (the precise dimension varies). I lightly stretch the chain to ensure all the relevant components are touching each other, then use an internal vernier caliper. I measure the chain in eight places around the length of the chain. To my surprise, the dimensions vary around the chain, but in a repeatable manner. I had thought that they would all be the same or very close. At first, I assumed my measuring technique was rusty, but when I checked the same length of chain in the same place several times I got the same measurement each time. I take the average of the eight measurements.

It is completely realistic to measure accurately to this level, engineers all over the world are doing it every day. I accept that temperature can affect the readings, but the vernier is getting bigger as well as the chain when it's a hot day rather than a cold one. The difference of differential temperature growth between the two because the metals are different (steel vs stainless steel) is really minimal. The key thing to do is not to hold onto the gauge too long, or to wear gloves. But this sort of thing is only really necessary if you are measuring to much higher levels of accuracy than is robobiking.

Whatever robobiking comes up with, I agree it is only relevant for him how and where he rides. But the relative information will be interesting. Does he get better life out of his whole system if he rotates three chains until they are all worn out, rather than wearing out each chain in succession? I'm pretty sure it is, but by how much? Is the financial difference enough to be worth the effort?

steve_sordy

@ jdrutherford:

Your cassette is new, your chain is new, the 50t ring is hardly ever used, so it must be that the 34t ring is the one that is worn out. If the cassette was worn out why would the 34t ring up front that has done even more work not be also worn out?
The new chain will be jumping about or even slipping on the ring. If you put the old chain back on it will run better on the 34t ring, but will run crap on the cassette. If you are sure that the shifting is set up properly, then it must be the 34t. Get it changed. You know it makes sense!