measuring chain wear (again)

For discussions about bikes and equipment.
mercalia
Posts: 14630
Joined: 22 Sep 2013, 10:03pm
Location: london South

measuring chain wear (again)

Post by mercalia »

I normally use a Park tool as I just cant get how you can do the measuring; maybe if there are 2 of you , you can. Now I have a steel tape type measure which has a hook on the end ( they all do?) So can I do it this way, measure the distance between the outside plates edges of 10 links- should be 25.4cm. The hook can easily be secured against the edge of one plate. I put the chain under tension by put a weight on the pedals.
Brucey
Posts: 44693
Joined: 4 Jan 2012, 6:25pm

Re: measuring chain wear (again)

Post by Brucey »

if you use a steel tape and you are using the hook end and measuring in cm then;

a) you definitely will not get an accurate measurement; the hook piece is meant to move on the end of the tape (by its own thickness) but this usually isn't accurate when the tape is new. In addition this arrangement can wear and also the hook can be bent. You can easily get an error of 0.5 to 1mm this way.

b) you are making life very difficult for yourself. Chains are made in inches, so that is the easiest unit to measure them in. 12 full links should be exactly 12".

A 12" steel ruler is the best thing to use; just hold one end of the ruler and the chain together with your finger and thumb, then look at the other end. In each case you need to view square on so that you don't get parallax errors. Measure from a rivet centre; centrality in a circle is a better alignment than a line on a tape with an edge anyway, and conveniently half a rivet diameter is about the tolerable amount of chain stretch (if you want to put a new chain on the same cassette).

If you use a steel tape measure, measure between inch marks on the tape (eg 2" to 14") rather than from the very end; the hook is usually rubbish and if you don't use it, it is in the way and you can't read the end marks clearly anyway. A steel tape isn't quite as accurate as a steel rule but they are usually not too bad, once you get away from the crummy end piece.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
User avatar
Mick F
Spambuster
Posts: 56367
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Re: measuring chain wear (again)

Post by Mick F »

Brucey wrote:A 12" steel ruler is the best thing to use ......

Best thing to use is a 36" steel ruler.
Mine goes up to 39" :wink:
Mick F. Cornwall
NetworkMan
Posts: 727
Joined: 25 Aug 2014, 11:13am
Location: South Devon

Re: measuring chain wear (again)

Post by NetworkMan »

I do much as what Brucey says. I measure a nominal 10 inch length from rivet centre to rivet centre and judge it OK if it's less than 10.05 inches (0.5% elongation).
It's even easier if you measure it when you take it off for cleaning/lubrication. Hang it up on a nail or whatever and apply a little downward force with one hand while holding the rule with the other. I find that easier than grovelling on the ground.
User avatar
CREPELLO
Posts: 5559
Joined: 29 Nov 2008, 12:55am

Re: measuring chain wear (again)

Post by CREPELLO »

I've used a 12" steel rule, marked with an extra 1/6" and 1/8" beyond the 12" mark. Use a clothes line peg to hang the rule at one end and carefully position till the end is in line with the rivet centre. Whilst supporting the other end, apply pressure on the cranks (do this test on the ground, not a work stand) to take out any slop and read the measurement.
mercalia
Posts: 14630
Joined: 22 Sep 2013, 10:03pm
Location: london South

Re: measuring chain wear (again)

Post by mercalia »

those people who measure rivit to rivit must be supermen as I cannt do it. hold the end of ruler in one hand and the other in the other hand; how can you be sure when you are looking at the right hand side your left hand didnt move the ruler a tiny amount ( 1mm is enough to count). Using the "hook" of a metal tape measure ruler to catch the plate end seems to be more secure.

My question wasnt only about the means but also the where. Is measuring the distance between plate ends edges(on a stressed chain) the same as the distance between rivits? I also have have issue with measureing the rivit distance (centres?!? you must have microscopic eyes) - they are so small but are about 1mm or so thick, enough to make a wrong perception matter?

I can do this measurement easy and quick while on the bike.
Last edited by mercalia on 30 Nov 2014, 1:04pm, edited 2 times in total.
User avatar
Mick F
Spambuster
Posts: 56367
Joined: 7 Jan 2007, 11:24am
Location: Tamar Valley, Cornwall

Re: measuring chain wear (again)

Post by Mick F »

Get a long ruler. Not expensive - get one from a cheapo tool store.
Find some sort of clamp - G clamp or use what I use, the clamp for our pasta maker.
Clamp the end of the chain at the beginning of the ruler, and stretch out the chain along the length.
You can see the progression troughout the length.

Here's some very quick shots of my method. I just threw it on the table and took quickie photos, so it's not accurately positioned, but it gives you the idea. Mrs Mick F wanted the table, so I was in a rush! :oops:

It helps to have a clean chain before you drape it on the kitchen table. :lol:
Long View.jpg
Start.jpg
End.jpg
Mick F. Cornwall
bobc
Posts: 495
Joined: 5 Apr 2012, 11:59am

Re: measuring chain wear (again)

Post by bobc »

OK - how I measure chain.....
I use a digital vernier caliper - these things can be found for £9 or £10, they are fairly regularly on offer at Aldi, and useful for loads of jobs.
It's a fair bet if you're reading this you already have one.
Use the "inside measurement" pointy bit at the back.
set to measure in inches.
Put the inside pointy things between two adjacent rollers & pull to measure the distance. SET the ZERO (while applying tension)
Now measure between rollers 5 or 10 links further apart.
You can directly read how many thousandths of an inch the chain has stretched.
While it's on the bike.......
Psamathe
Posts: 17724
Joined: 10 Jan 2014, 8:56pm

Re: measuring chain wear (again)

Post by Psamathe »

When using a rules (12 links=12" or 10 links=10") is it centre of rivet to centre of rivet ?

Ian
User avatar
DaveP
Posts: 3333
Joined: 9 Mar 2007, 4:20pm
Location: W Mids

Re: measuring chain wear (again)

Post by DaveP »

You can choose your own reference points, as long as you do the same for each end. I hope that's not overstating the obvious - but its an important point - and I have seen someone get it wrong!
Personally I measure from the right hand edge of a rivet head, which avoids errors in estimating the centre. I use a 12" engineers rule with no "dead space" at the ends. I stretch the chain along the table on its side (rivet heads up) and align one end of the ruler with the edge of a rivet. Then I just need to look at the other end and estimate how far the corresponding rivet head is past the end of the rule. You are looking for cumulative wear - you don't need to use great precision.
When the cumulative wear over a 12" length (carefully sidestepping debates about what constitutes a link :D ) reaches 1/16" its about time to replace the chain. By the time cumulative wear reaches 1/8" its probably damaging your sprockets and chain wheels.
I replace at about 1/16". If that's a tiny bit early that ok by me - I never got beyond 8 speed... :mrgreen:

Found my source for this: http://sheldonbrown.com/chains.html#wear and scroll to bottom
Trying to retain enough fitness to grow old disgracefully... That hasn't changed!
Brucey
Posts: 44693
Joined: 4 Jan 2012, 6:25pm

Re: measuring chain wear (again)

Post by Brucey »

bobc wrote:OK - how I measure chain.....
I use a digital vernier caliper - these things can be found for £9 or £10, they are fairly regularly on offer at Aldi, and useful for loads of jobs.
It's a fair bet if you're reading this you already have one.
Use the "inside measurement" pointy bit at the back.
set to measure in inches.
Put the inside pointy things between two adjacent rollers & pull to measure the distance. SET the ZERO (while applying tension)
Now measure between rollers 5 or 10 links further apart.
You can directly read how many thousandths of an inch the chain has stretched.
While it's on the bike.......


Like using a ruler, this is a good way of doing it IMHO, because it takes a measurement that (unlike most commercial chain checkers) doesn't include any element of roller wear; that is (provided the roller wear is uniform) taken out when the verniers are zeroed. It is a shame that most vernier calipers are only 6" long, but the measurement is accurate enough that it isn't so important to measure over a long length as when using a less accurate method.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
mercalia
Posts: 14630
Joined: 22 Sep 2013, 10:03pm
Location: london South

Re: measuring chain wear (again)

Post by mercalia »

bobc wrote:OK - how I measure chain.....
I use a digital vernier caliper - these things can be found for £9 or £10, they are fairly regularly on offer at Aldi, and useful for loads of jobs.
It's a fair bet if you're reading this you already have one.
Use the "inside measurement" pointy bit at the back.
set to measure in inches.
Put the inside pointy things between two adjacent rollers & pull to measure the distance. SET the ZERO (while applying tension)
Now measure between rollers 5 or 10 links further apart.
You can directly read how many thousandths of an inch the chain has stretched.
While it's on the bike.......



or you could use the external caliper and measure the distance between the plates edges?

Getting a caliper the best answer I think. I am convinced. I dont under stand though the bit about set the zero and there after.
Brucey
Posts: 44693
Joined: 4 Jan 2012, 6:25pm

Re: measuring chain wear (again)

Post by Brucey »

Bob's method means you measure the nominal pitch length + the wear so for example you might get a reading of 5.010" for five full links. This is 5.000" for the nominal chain length and 0.010" for the wear, which is therefore only 0.2%.

If you measure at side plates you will get a meaningless measurement unless you also zero the calipers in such a way as to allow for the fact that you are measuring something that includes the length of a chain component as well as the pitch of the chain.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
mercalia
Posts: 14630
Joined: 22 Sep 2013, 10:03pm
Location: london South

Re: measuring chain wear (again)

Post by mercalia »

Brucey wrote:Bob's method means you measure the nominal pitch length + the wear so for example you might get a reading of 5.010" for five full links. This is 5.000" for the nominal chain length and 0.010" for the wear, which is therefore only 0.2%.

If you measure at side plates you will get a meaningless measurement unless you also zero the calipers in such a way as to allow for the fact that you are measuring something that includes the length of a chain component as well as the pitch of the chain.

cheers


I assume you dont need to do this with a standard mechanical Vernier Caliper?
Brucey
Posts: 44693
Joined: 4 Jan 2012, 6:25pm

Re: measuring chain wear (again)

Post by Brucey »

mercalia wrote: I assume you dont need to do this with a standard mechanical Vernier Caliper?


well, sort of; you normally can't do that with a mechanical one, which makes the whole business of measuring chains far more difficult. And if you think lining a ruler up with a rivet (which is more or less how most gunsights work BTW...) is difficult, just forget mechanical verniers...

Get digital calipers, exactly as Bob suggests. They are far better, and much more versatile. They are not even very expensive these days. I can live with the occasional battery replacement because they are so much better to use.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Post Reply