RandomTroll

I've been cleaning and lubricating my chain regularly and checking its stretch monthly. Recently I've stretched them to .5% in about a thousand miles. Today I measured the chain at .5% after only 350 miles, after its first month. Could this be a result of worn chainrings and/or cassette? The chainrings have many years of wear; the cassette is pretty new.

The chain is an SRAM PC830. I bought 3 on sale for about $6.50 each - is there something wrong with them? I've used them in the past without unusual wear.

Andrew R Stewart 

Cog/ring tooth wear is less an exact process then chain wear (and we know how much band width goes into that subject). But Rohloff makes a cassette wear indicator https://online2.qbp.com/qbponlinestorefront/p/TL4613 as an example of what some shops use for cogs. This tool is more subjective then a chain and ruler but does point out the well worn cogs pretty well. Looking at new tooth profiles and then the worn teeth give you some idea about lost material. But without new at hand it's like watching children grow day to day. Hard to see much change until...

Two more points- Many years ago when SRAM chains were Sachs I asked the tech guy at an industry show (the Pier in NYC IIRC) about measuring chains. He said that new chains had quite a bit of variation in pin to pin dimensions when one was looking at fine amounts. Over the years I sometimes have checked new chain "wear" with various indicators and, of course, a ruler. The ruler has seemed spot on pretty much all the time, start to finish pins centered on 12". But when using an indicator (fill in your favorite- Park's different ones, Rohloff, KMC digital and others) I find differing amounts of "wear" with brand new chains.

There is a thought that chains go through a break in with some initial wear very soon and after this less wear per mile. Of course the writings I've seen about this deal with people who pay attention to chain wear and clean/lube "as needed" and don't reflect the common install and forget that many riders seem to live by. Andy

dsbrantjr

How are you measuring this wear? I only trust a good rule, chain checkers are notoriously unreliable. I have heard reports of chain checkers reporting a new chain as 0.5% worn.

HillRider

I have a nearly new SRAM PC1030 that measured about half way to .5% wear (12" ruler method) after one 30 mile ride the day I installed it so perhaps SRAM's lower line, or even upper line, chains aren't that accurately made.

Lazyass

I use a magnifying glass to check the teeth.

agenkin

And what signs are you looking for?

Jamminatrix

Here's a good article highlighting length variations across brand new chains models: Technical FAQ: Chain wear measurement - VeloNews.com

Even using a typical between-rollers chain wear tool (that should be more prone to overestimating stretch because of roller wear), it tells me a brand new Shimano 9 speed chain is stretched out, while a heavily used Shimano 10 speed chain says less than <.5 stretch. It's down to manufacturing difference.

FBinNY 

The best indicator of a sprocket's condition is whether a new chain will run on it without skipping under load. Trying to predict that is problematic and will likely result in replacing sprockets sooner than necessary.

You can also gauge condition by eye, or by looping a length of new chain halfway around and lifting it away in the middle of the wrap. There's no clear guideline for this, which is probably more accurate (if accurate is appropriate here) with more teeth, ie. chain rings vs. cassettes. I've found through experience that if I can pull the chain up enough to create 1/4" of daylight underneath, I'm nearing (or passed) the ring's useful life. But that's only a rough guide, because the worn shape of the teeth is also a factor, and rings may slip sooner or later than that indicator.

So, IMO gauging sprocket wear may be useful as a rough guide for planning, ie. ordering spares in advance, or before planning a long tour, I wouldn't replace a sprocket based on any wear "measurement".

ThermionicScott 

PC830's wear quickly in my experience. The non-hardened pins (as opposed to better chains like the 850 and up) only seem to take about 1500 miles or less to wear to the 1/2% mark. My PC850s always last more than 3000 miles so I consider them a much better value for a few dollars more each.

Al1943

I've never had a road bike chain ring skip, but last year I had a new experience with chainring wear. After replacing the chain on my Campy Record 10-speed drive train I noticed a grinding noise when accelerating from a dead stop. Checking the derailleurs and making minor adjustments had no effect. But the grinding noise occurred only when running on the inside, 39t, ring. Noise was absent when on the 53t. Replacing the 39 cured the problem, no more noise.

FBinNY 

The relationship between chain wear and sprocket condition is complicated, and depends on what aspect of chain wear you're considering. In my (educated) opinion, worn chains accelerate sprocket wear, but the reverse isn't true, or if so, only in a slight way.

Consider that chain stretch is measured pin to pin, so you're measuring only wear between the pins and the inner link plates that move on them. These are well insulated from the sprocket, and ALL the wear happens when the chain is bending onto and off sprockets on the driving side of the loop (the return loop is always slack or only lightly tensioned so there's little wear, and the rest of the links are stationary).

So if you're using a ruler to measure stretch, the condition of the sprockets will not be a factor.

OTOH, worn sprockets can affect roller wear because the rollers see more movement with a worn sprocket. That can become an issue, if you use a device that measures using the rollers (most or all do). Increased roller wear can give a false high reading for stretch which is what I suspect you're seeing.

Keep in mind that chain wear measurement isn't a precise science, and the rough information gleaned can only be used as a guideline, comparable to changing the oil in your car every 3,000 miles (or so). While you can measure simple stretch fairly precisely, (apart from roller wear) with a ruler, it's only part of the story, because roller wear also affects how chains run on sprockets, and there's no assured ratio of roller vs. pin wear on chains.

My advice, use a ruler to measure stretch rather than a 5" device, so you have at least half the story right, but don't obsess over it.

RandomTroll

Measuring 145 mm to .03mm is more precise than measuring 12 inches to 1/32.

I always measure chains before I install them. They don't start out the same.

By the time a chainring or cog is so worn a new chain skips on them it's pretty late in its life, and a drag to put the old chain back on and buy new chainrings or cogs.

RPK79

Place the item in a tub of water and measure the displacement. Compare this to the displacement at the time the item was new.

Drew Eckhardt 

0.5% is about 1/16" over 11-12". If you're not measuring that elongation on a ruler it's your chain tool and how it interacts with the as-manufactured roller to bushing clearance in your chains.

dwmckee

Have you tried this yourself? What was your experience?

fietsbob

Rohloff cog wear gage https://jet.com/product/detail/93615...YtVxoCrn3w_wcB Or Via QBP @ your LBS
they made chains before they started making the 14 speed Hubs, https://www.youtube.com/watch?v=Lb63caDLcN0

fietsbob

Archimedes (c. 287 BC – c. 212 BC) was so excited when he discovered the logic of displacement,
He ran naked from his bath, where He had figured it out.

Eureka! The Story of Archimedes' Greatest Discovery

FBinNY 

It's not what you measure, it's how you measure it. To start with, new chains are very precise if you measure 24 pitches or, better yet, over the 4'+ overall length. That's because it's easy for chain makers to precisely tolerance the holes in the plates and the pin diameter.

However, all bets are off once you involve the rollers. The rollers have varying amounts of slop or play, even on brand new chains, and that can throw off any short measurement.

As for your 145mm to .03mm, I doubt it's any more accurate. It sounds good, but fractional mm measurements in the range of about 0.001" are highly dependent on touch and how you pick up the end points. So IMO, anyone who thinks he can measure any length on a chain to that kind of tolerance is kidding himself. In any case, as I posted earlier precision in chain measurement is pointless, and comparable to having your car's odometer calibrated to ensure that you change the oil at exactly 3,000 miles.

That said, believe what you wish, and go with any method that suits you.

Reynolds 

I made a copy of that tool (it's very simple) and found cogs have to be really worn with hook - shaped teeth to be considered worn out by the tool.

FBinNY 

Yes, because that's consistent with the intended purpose of the tool, which is to detect sprockets worn to the hook profile bad enough to sang and cause chain suck.

Reynolds 

^ But it detects sprockets so worn that you could as well sort them out by eye...

fietsbob

You could just replace the cassette, as you replace the chain a 3rd time, and not bother measuring the teeth .

1, 2 and Out.

FBinNY 

Absolutely, eyeballing and a bit of judgement are more than adequate for this kind of thing. Or you can not bother at all and just deal with it if/when a problem manifests.

But there's a cohort of people who insist on real numbers and real measurements. so there's a tool made for that.

dwmckee

Do you think Archimedes' bath displacement technique was accurate enough to measure the difference with and without bellybutton lint?

fietsbob

Apparently sufficient to tell the difference between Gold clad Silver and solid Au.