Don’t miss out on the latest CyclingTips updates.
Your bicycle’s chain is put through hell every time you ride. For every minute of pedaling, approximately 44,000 chain pieces are in motion, creating 320,000 separate instances of sliding surface friction. And all of this is on a component that sits near to the ground and is exposed to the elements.
Just like your tyres and brake pads, chains wear with use. And as a chain wears, friction in the drivetrain increases, your shifting gets sloppier, and worst of all, you’ll quickly start wearing out other drivetrain components. Changing your chain at the right time can save you money and make your riding more enjoyable.
If you’re seeking a quick answer for the easy way to check for chain wear, you should only need to read the first few sections of this article. If you want to go deep down the rabbit hole, well, we can help you with that too.
Basics of chain wear explained
Over time, the chain’s pins and inner links will wear, and as a result, the pitch (length) of each link will grow. Because the chain’s overall length grows with wear, chain wear is commonly called ‘chain stretch’ – even though the metal does not (measurably) stretch.
The standard pitch of a new chain link should sit at half an inch (12.7mm), pin-to-pin. An inner plus an outer (wide and narrow) link of a chain makes an even inch. Chainrings and cassette cogs are designed with this pitch in mind, such that the chain rides at the base of the cog/ring when new. As the chain pitch grows, it rolls higher on the tooth, accelerating cog wear until eventually it just skips over the top. And it’s this dreaded chain skip that you never want to feel when you’ve got all your weight loaded on a pedal.
Once the chain wears, the cassette and chainrings start to wear along with it, becoming ‘hooked’ from the high-riding chain. Replacing your chain before it wears too badly will dramatically increase the life of the rest of your drivetrain (cassette and chainrings). A $40 chain every few months could save you hundreds by preserving your drivetrain.
When to replace a chain
There is no exact science to knowing when to replace a chain. And there’s also plenty of debate about what classifies as a worn chain. However, the information in this article should get you as close to the answer as is known.
It’s common to hear distance recommendations for how often you should replace your chain. But as we covered in our Holy Grail of chain lube article, there are simply too many variables for this to be a reliable technique – and in reality, a simple difference of chain lube and maintenance can result in a longevity difference of anywhere between 500km to 15,000km. This is why you need to manually check it.
How long a chain will last will depend on your riding strength, your choice of chain lube, the chain, riding conditions, shifting habits and the terrain you ride.
Even when cost isn’t a factor, the likes of the top WorldTour teams regularly replace chains approximately every 1,000km (a week in a Grand Tour). Some mechanics claim they get up to three full seasons out of cassettes and chainrings this way, but more importantly, there’s far less risk of a broken chain in the heat of the moment and it’s more efficient, too. Additionally, a worn chain will exhibit greater slack that leads to slower and sloppier shifting.
This argument for regular chain replacement is rather clear-cut if you’re riding on Shimano Dura-Ace, Campagnolo Super Record and SRAM Red components where the cogs can cost as much as an entry-level bike. However, the value proposition becomes a tougher debate if you ride on 105 or below where replacement chainrings and cassettes are far more affordable.
Additionally, you’ll need to consider the installation cost of a new chain if you’re not confident in doing it yourself. Or the cost of a chain breaker if it’s a task you’re looking to take on. Nevertheless, if you value crisp shifting, an efficient drivetrain, or if you often swap between wheelsets, then regularly replacing chains before they develop significant wear is a smart choice regardless of what your components cost.
For that, you should do a manual check (see next section) for wear on a regular basis. If you’re a casual road rider, I’d suggest checking every couple of months, and if you ride most days of the week, then you should check it at more regular intervals again. The slacker you are with basic chain maintenance, the more you should keep a check on wear. And be sure to increase your checking intervals during the muck of winter, too. With time you’ll get an understanding of how long things last, and when you should be watching for wear.
If you ride mountain bikes, gravel or cyclocross, then beware that your chain replacement intervals will likely be far more regular again. It’s not uncommon to hear of riders wearing through a chain in a single poor-conditions endurance event.
Measuring chain wear the free and easy way
So how about measuring that chain wear? Well, the easiest (and free) way is as follows:
1. Shift gears so that your chain is in the big ring and smallest gear on the cassette (e.g. 53-11T)
2. Pull the chain at the front of the chainring as shown. If the chain starts to lift off the top and/or the bottom of where it sits on the chainring teeth, this means that the chain is starting to wear or is worn.
This ‘lift’ is possible because the chain’s pitch has lengthened and so no longer sits properly in the teeth. The photo below shows a brand new chain. Chances are if your chain lifts off more than our worn example, you’ll be needing more than a new chain. However, do beware that worn chainrings can give a false reading with this method, and a new chain on a worn-out chainring will present similar lifting.
However, while this method is free, it doesn’t provide much indication for how worn that chain is, or whether you’ve worn the chain so far that your existing cogs won’t accept a fresh chain. That’s where tools come in.
The easy tool method
So how worn is your chain? Is it so far cooked you’ll need a new cassette? Or is it simply an old chainring telling you lies with the previous method? You’ll need a tool to know for sure.
There is the ruler method which I cover later, but my suggestion is to use an affordable, purpose-built chain checker tool. The simple fixed-length, drop-in style tool is all you need and will quickly give you a go or no-go gauge on your chain wear.
It’s important to know that chains rarely wear evenly across their entire length. And so however you choose to measure your chain wear, you should do it across three to five separate sections and use the highest measurement. And avoid including the quick link or similar joining link in your measurement as these often have slower wear rates.
Using the popular Park Tool CC-3.2 chain wear checker as an example, the .5% marking is there as a recommended replacement point for 11 and 12-speed drivetrains, or as a warning for those on 10-speed or lower setups. The .75% reading is the suggested replacement point for 10-speed and lower. Some tools offer a 1% wear point, too, something best kept for eight-speed chains or lower.
This .5% suggested replacement point is fairly new and comes from an increased understanding of how the narrower cogs of modern drivetrains offer less surface area, and are therefore more prone to material wear. Because of this, many older chain wear checkers on the markets are outdated, and will simply show chain wear on newer drivetrains at a point that’s too late.
If a chain is worn, the tool will drop into the link and sit flush along the chain. Or if it’s not worn, the tool will sit above the link, as shown in the lead photo. For our own bikes running the good stuff, we’ll replace a chain at the .5 marking. This applies to our 10, 11 and 12-speed setups.
However, there are important exceptions to this. If you just wanted the simple answers, you can stop here. I do, however, recommend reading some related chain articles.
If you want the deep dive, then stick with me — this will get a little geeky.
A deeper dive into the mechanics of chain wear
During each articulation around a chainring, cassette or pulley wheel, the eight pieces that make a full chain link are experiencing an enormous amount of friction. It’s easy to see how and why chain lubrication can play such an important role in efficiency and durability.
With each articulation, the riveted pin remains static, with the same surface repeatedly seeing friction. Similarly, the roller is held static when in contact with a cog – it doesn’t roll. As a result, it’s the inner link plates that are articulating around the static pin.
As the inner plates articulate around the pin, the pin is worn thinner, and the inside bores of the inner links expand. This wear leads to play between the pieces, and when the chain is pulled under tension, its length grows. This is elongation wear (aka, stretch).
Traditionally the general goal of chain-wear-checking is to measure this elongation, excluding the rollers surrounding the pin and inner links. However, the cog and chainring are dumb to this measurement, and really, it’s the distance between the compressed rollers that the cogs witness. And just as the pin and inner links wear together, so too can the inside of the roller and the outside diameter of the inner links. I’ll come back to this.
Wearing of the inner and outer plates, known as lateral wear, is also a key factor to consider. This will see the side-to-side play in your chain increase, and with it, you’ll experience slower shifting as the derailleurs and shifting ramps work harder to pull the floppier chain onto the desired cog. It won’t lead to the wear of other components, but it will stop your shifting from working at its best.
A deeper dive in to chain tools
Now back to chain wear tools.
A tool like the Park Tool CC-3.2 is attempting to measure the distance from one pin to another, however, as it sits against the front of one roller and the backside of another, its reading can be thrown off by the rollers. This shouldn’t be an issue, but not all rollers are created equal, and it’s common to find some chains that have rollers that are looser-fitting, faster-wearing or simply different diameters than others.
Because of this, some chain wear checkers seek to isolate the roller wear from the measurement and do this by measuring from equal, and not opposing, sides of the rollers. Examples of this include chain tools from Shimano, Pedro’s, and more recently, Park Tool (CC-4). They should provide a more consistent reading across a greater variety of chains, even if they’re still impacted by roller wear and roller variances (just to a far lesser degree).
“The three-point measurement system used on the Pedro’s Chain Checker Plus II accomplishes this by pushing both rollers in the same direction during measurement, instead of in opposite directions, as is the case with common two-point measurement tools,” explains Jay Seither, Pedro’s head of product management and engineering. “As a result, the Chain Checker Plus achieves a true measure of the pin-to-pin distance.”
For the best chain article, Adam Kerin of Zero Friction Cycling measured, on average, 19% earlier wear rates by using a two-point digital chain checker than when measuring the external pin-to-pin elongation of the chain (I’ll soon explain how to measure it).
The reasons for this discrepancy are down to roller tolerances and roller wear. The fact that some chains start life with looser rollers than others is not news. However, in what’s arguably new information to everyone in the drivetrain space, Kerin found that some chains had rollers wear at a far higher rate in proportion to the pins and inner links. And as mentioned before, the cog’s teeth don’t care about these discrepancies, rather it’s simply the distance between the rollers under load that really matters.
Park Tool’s Project Manager, John Krawczyk, concurs with this. “Whether a chain measures only 0.001% wear or 0.75% “worn” when it is new, this doesn’t change the fact that the cassette and chainrings don’t care how new or how old the chain is,” he said. “All they know is that once the chain goes beyond 0.75% wear (or whatever the replacement metric from the chain manufacturer might be) those rollers no longer fall cleanly in the valley between the teeth and either the chain needs to be replaced or the teeth begin to be re-shaped.”
Given this, and despite such glaring measurement discrepancies, both Kerin and I are of the opinion that the drop-in style chain checker tools are still the best and easiest way to keep a check on wear, and that it’s best to change a chain that becoming worn than one that’s overdue.
When shopping for a chain wear checker, look for one that can’t be damaged through misuse (which will then cause premature wear readings), and one that offers staggered wear readings to give you a rough indication as to the point of wear. Avoid tools that only offer a single point of wear measurement – they’ll only tell you when your chain is toasted, a point that’s too late in my opinion.
I’ve long been partial to the Park Tool CC-3.2 as a budget option, and in most cases, I continue to find it a reliable option. There are plenty of similar choices sold by others, but experience shows that the cheapest ones can be a little hit-and-miss.
For more finite wear-checking, my preference is the KMC Digital Chain Wear Checker. It’s priced high and is certainly overkill for most, but it lets me monitor chain wear in tiny increments. I’ll replace chains once they tip over .4mm on this tool, which is a hair sooner than the .5% measure on the CC-3.2.
SRAM chains are one clear exception to using these suggested tools. Most chains on the market start with a roller that’s 7.63-7.65mm in outside diameter. SRAM’s chains are larger — for example, rollers from a Red 22 chain are 7.69-7.70mm, while an Eagle 12-speed chain uses rollers that measure 7.72mm. And SRAM’s new Flat-Top chain as part of the Road AXS groups is larger again (7.90mm).
Of course, that throws off any tool that measures from opposing sides of the rollers. This is where Pedros’ Chain Checker Plus II, or Park Tool’s CC-4 come in. These backside-to-backside chain checkers will work across all chains, including SRAM.
Because of this, both Pedro’s Chain Wear Checker II and Park Tool’s CC-4 are fast becoming my preferred options, and for not a big increase in cost. The Pedro’s manages to combine other tools into it, while the Park feels a little more rigid in use.
Alternatively, you can use a vernier caliper to measure (and record) the distance between 10 links when the chain is new and then monitor it for .5% wear from there. Use the calipers inside the rollers, just as if it were a chain wear tool. The use of a vernier caliper is what Campagnolo recommends, however, it is a more involved method that involves a potentially more expensive tool (but is useful for more than just chain wear checking).
Regardless of what tool you use, it’s a good idea to learn how it measures on a new chain. If a tool reveals worrisome wear on a new and decent quality chain, then it’s perhaps not a tool to trust.
Using the tools
The amount of tension applied to the chain, and how dirty it is, will impact the reading with any tool used. A dirty chain will likely always read as being less worn than what it really is, likewise for a chain doused in a thick lubricant. While the more tension you put on the chain, the more worn it’ll read. And this is where things can get tricky.
Chains are never ridden without tension. Given that, Jason Smith, CeramicSpeed’s Chief Technology Officer and the person responsible for much of the newer understanding of chain wear, recommends measuring your chain while it’s under load. After all, that’s how it interfaces with a cog in use. Pedro’s Chain Checker and Park Tool’s CC-4 allow you to apply a load directly in using the tool (although it is not across the full measuring span), while with other options you will need to create the load separately.
Here’s my technique. With the rear wheel held in place (easiest if it’s on the ground), I pull on the crank until any easily discernible slack in the chain is taken up. With the chain wear tool in the top-span of the chain (above the chainstay), I then check whether the tool drops into its wear mark. It’s far from scientific, but it’s quick and relatively repeatable.
Most tools will simply drop into place if the chain is worn, and so only ever apply a light load to the tool, and never force it. If you’re having to push down on the tool with any level of effort, the chain is not worn.
This advice is something that Krawczyk of Park Tool reiterates. “Chain and tool tension can greatly impact the effectiveness of any chain wear indicator,” he says. “As wise mechanics say, ‘No matter how good the tool is it is only as good as the mechanic using it.’ With that said we do find that by measuring roller back to roller back (like on the CC-4) this does help minimize the dependence of a set tension. [However] if you push hard enough the tool can and will flex, effectively forcing the tool into an otherwise new or not-yet-worn chain.”
It doesn’t make a whole lot of sense to clean a chain prior to replacing it, but just know that the grittier, greasier and grosser your chain looks, the likelier it’s more worn than what the tool tells.
And remember my earlier point regarding uneven chain wear? Be sure to measure multiple spots along the chain’s length, and use the highest measure.
Chain wear is not linear
When keeping a check on chain wear, it’s critical to know that the wear does not happen in a linear fashion. If you get 3,000km of riding to .25% wear, you’ll unlikely get another 3,000km by the time the chain reaches .5% wear. This is because most quality chains have a number of surface hardening treatments and low friction coatings which will wear away with use – accelerating material wear. Additionally, the contamination inside the chain’s links will increase with use.
Don’t get caught out. Be sure to check at regular time-based intervals, and don’t rely on distance as a measure.
Measuring the chain elongation
It’s often recommended that the most accurate and best way to measure your chain is with a ruler. The theory is that by measuring pin-to-pin you can accurately gauge how much wear has occurred in the components of the chain and it removes any question over roller tolerances and wear, instead focusing solely on the actual pitch of the chain.
However, in my opinion, this process is fraught with the likelihood of user-error. Lining up a ruler from the centre of one pin to another 10 or 12 links away, all while remaining within less than half a millimetre of accuracy, isn’t something that many can do consistently. Additionally, you won’t be able to easily add tension to that chain, and so dirt and lube will have a huge impact on the measurement.
If you disagree (you’re wrong!), then the process is as follows:
With your chain still on the bike, place a ruler’s zero inch mark directly above the center of one of your chain pins. Now count 12 complete links. A complete link equals one inner and one outer. A rivet on a new chain should line up exactly to 12 inches (304.8mm) on the ruler.
As a general rule for 9-speed or lower drivetrains, if the rivet is less than 1/16″ (1.59mm) past the mark, your chain is ok. If it’s between 1/16″ and 1/8″ (3.18mm) past the mark you’ll likely need a new chain, but your sprockets should be ok. If it’s more than 1/8″ past the mark, you’ll probably have to replace both the chain and cassette. For 10, 11 and 12-speed, you’ll want to replace the chain as soon as it reaches 1/16”, or in other words, .5%.
As mentioned, I don’t rate the ruler method. The following, more-involved method produces more accurate results but involves tracking chain wear over the entire length of the chain, starting from when it’s brand new. This is because while all chains should measure an inch per a whole link, it’s rarely the case. Jason Smith previously found a difference in elongation between unused chains of the same brand. There are certainly variances!
For this, once the new chain has been cut to the right length, hang it off a nail or similar hanging space that’ll remain consistent. Measure the total length of the new chain, from the centre of one end to the centre of the other end. If it’s an option, you can mark on the wall (use tape) where the chain reaches.
From here on, you’ll need to measure your chain wear with the chain off the bike, and replace it when the total length has grown by .5% from the original. This is only suggested for chains using a quick link, and you’ll need to do the measurement after the chain has been cleaned and with a weight pulling down on it (Smith suggests a 50lb / 22kg weight at most).
And just when you’re thinking this all sounds like too much work, also remember that chains don’t wear evenly, so it’s quite possible that there are sections of the chain that are far more worn than what the total length measurement suggests.
Yep, chain wear checker tools aren’t perfect, but they’re better than the alternative!
Worn chainrings and cassette
Remember those .5% and .75% suggested replacement points? Well, unfortunately, chain elongation is not the only cause of cog wear — pure metal-on-metal abrasion is a major cause, too.
Myself and countless others have experienced it where a chain may only be slightly worn, but due to a poor choice of lube and a lack of basic maintenance the chain has abraded through a cassette. There is plenty of truth to the old saying that a clean bike is a happy one.
In Kerin’s chain lube testing, he discovered huge differences in cog wear as the direct result of chain lube choice. Some of the poorer-performing chain lubes, such as White Lightning Epic Ride, would see the cogs abraded beyond re-use by the time a chain measured .5% wear. Meanwhile, good wax-based lubes would cause almost no measurable wear to the cogs with the same chain elongation.
Generally speaking, for 10-, 11- and 12-speed drivetrain users, replace your chain when it measures .5%, and you’ll be fine with re-using the existing cassette and chainrings. And you should get three chains to that one cassette, and perhaps as many as six chains to the chainrings. Wait till the chain measures .75% and you’ll likely need at least a new cassette, too.
The discrepancy between chainrings and cassettes is because the former are typically larger and with more teeth, therefore spreading the load across a greater number of teeth at any one point. On this point, yes, smaller chainrings do typically wear out faster than larger rings. Additionally, a chain will wrap itself around more than half a chainring, while the rear derailleur dictates that there will be less wrap on the cassette.
Severely worn cassettes and chainrings are easy to spot as they’ll start looking thin and like shark teeth. The teeth will likely be burred, too. Plus, the chain coming off them will be wrecked.
There are no perfect tools or gauges for determining cog and chainring wear, though fitting a new chain is the surest way to reveal significant wear as it will skip (when under load) and rumble on the worn teeth. That new chain will reveal gaping issues with the worn cassette, and you should be able to spot where the chain sits higher than the base of the cog tooth.
For the chainring, one test is to grab the chain where it sits on the chainring at a 3-o’clock position and pull hard. A worn chainring will likely let go of the chain and let it slip forward – proving just how dangerous such a combination can be. Just be careful to test all of this in a controlled environment before heading into a bunch sprint.
Shimano components are the easiest to measure for wear, with new components measuring 9.5mm from tip to tip on the square-edge teeth. According to Kerin, low to medium power riders should be able to use worn cassettes and chainrings that show up to a 10mm distance from tip to tip, but powerful riders may experience skipping. And by 10.2mm it’s too late for everyone.
According to Kerin, if you keep up on your chain maintenance, use a good lubricant, and replace your chain before it shows significant wear, then you could get as much as 50,000km from your chainrings.
Well done for making it this far.
This is now the second time I’ve taken a deep dive on the subject of chain wear, trying to answer the questions that aren’t asked enough. Unfortunately, like last time, I’m left with some questions that just can’t be answered. That’s the problem when there are far too many variables for something to be an exact science.
Adam Kerin of Zero Friction Cycling summarises it well: “As a general rule I prefer a quality drop-in chain wear checker. For those playing at home, checking wear via measuring elongation with calipers, a ruler, or hanging the entire chain is fraught with some challenges, and chain wear checking needs to be something simple and quick so that users stay on top of, and replace chains before they become too worn and start eating into expensive cassettes and chainrings.”
So, get yourself a trusted chain checker tool, use it often, and don’t be afraid to replace your chain when you think it’s time. Your drivetrain will thank you.