Chain Wear and Elongation Calculator

Measure a length across a known number of pitches and this returns the elongation, the remaining life, the GO/NO-GO length to scribe on a rule, and a verdict against the tighter of the limit you chose and the sprocket-dependent 200/N rule. All four published limits are printed with the words their publishers used, the two contradictory statements of 200/N’s range are settled by arithmetic, and the page works out whether your rule can resolve the answer at all.

Last updated 24 Sep 2026

Chain wear and elongation

A measured length → a verdict

Chain sizeOnly the pitch matters here: the nominal length you are measuring against is the pitch times the number of pitches.

Chain pitch (mm)

Pitches measured acrossPin centre to pin centre, counting the pitches between them. More is better and the reason is arithmetic: the resolution of your answer is your rule’s resolution divided by the length you measured. The results below work out how many you need.

Measured length (mm)Measured on the TIGHT span with the drive stopped, away from the sprockets, at the most worn section you can find. Off the drive, apply the measuring load ISO 606 clause 3.6 specifies so the slack is out.

Teeth on the LARGE sprocketNot the small one. The 200/N wear rule takes the large sprocket, because a worn chain rides progressively further up the teeth of the bigger wheel and runs out of tooth there first.

Which published limit are you working to?Four publishers give three different fixed figures and a formula. All four are printed in the results and in the table below; this choice only decides which one the verdict uses — and if 200/N is tighter for your sprocket, the page uses that instead and says so.

Resolution of your measuring instrument (mm)0.5 mm for a steel rule read carefully, 0.1 mm for a long vernier, 1 mm for a tape. This is what decides whether your measurement can tell 1.5% from 2% at all.

Not a circuit, and drawn at two scales because it has to be. The top row is twelve pitches of new chain, rollers and all, at one scale. The bar beneath it is the same chain worn, and the elongation on the right of it is drawn twelve times larger than the length it belongs to — otherwise 2.6 mm in 152 mm would be a third of a pixel and the whole point would be invisible; past about 3.8% it stops growing, which is well past every published limit. That is the first thing this picture is for: the wear that condemns a chain is a fraction of a per cent of anything you can see. The second is underneath. A tooth space with a roller seated in it, and the same roller where a worn chain puts it — lifted by PD·e/2, which is where a longer pitch has to go on a sprocket cut for the old one, and drawn here at a magnified scale so the climb is visible at all. Watch it walk up the flank as the elongation rises. That is not a stretched chain being slack; it is a chain about to come off the teeth, and it is the reason the limits on this page are set by the sprocket rather than by the chain's strength.

64.83%Example

155.0 mm measured across 12 pitches of ANSI 40 chain, on a drive whose large sprocket has 76 teeth, working to the 3% limit with a 0.5 mm rule

Measure a length, compare it with the pitch times the count

elongation % = (L_measured / (N_pitches × p) − 1) × 100 · limit = min(published limit, 200/N_large) · GO/NO-GO length = N_pitches × p × (1 + limit/100) · resolution in % points = 100 × instrument resolution / (N_pitches × p)

elongation: not stretch. The steel has not stretched; the pins and bushings have lost material, so every joint has more clearance and the pitch has grown. That is why it is irreversible and why re-tensioning does not fix it
N_pitches × p: the nominal length. Count the pitches between the two pins you measure to, not the pins — and get the pitch right, because a wrong pitch shows up as a wear figure
200/N_large: the sprocket-dependent limit, N being the teeth on the LARGE sprocket. It drops below 3% between 66 and 67 teeth, so on a small driven sprocket the 3% cap governs and on a big one this does
GO/NO-GO length: the number to scribe on a rule or set a gauge to. It is the only output of this page that does not need arithmetic at the machine
resolution: your instrument’s resolution divided by the length you measured. Nobody publishes this and it decides whether the measurement can answer the question at all

Worked example

155.0 mm measured across 12 pitches of ANSI 40 chain, on a drive whose large sprocket has 76 teeth, working to the 3% limit with a 0.5 mm rule
Nominal length = 12 × 12.70 = 152.40 mm
Elongation = (155.0 / 152.40 − 1) × 100 = 1.706% — that is 2.60 mm over twelve pitches, or 0.217 mm per pitch
The large sprocket has 76 teeth, so 200/N = 200/76 = 2.632%. That is TIGHTER than the 3% selected, so it is the limit the verdict uses
Allowance used = 1.706 / 2.632 = 64.83%, so the remaining life against that limit is 35.2%
The GO / NO-GO length is 152.40 × (1 + 2.6316/100) = 156.41 mm. Scribe that on the rule and you never need to do this arithmetic at the machine again
And the measurement itself: a 0.5 mm rule over 152.40 mm resolves 0.328 percentage points, which is 12% of the whole allowance. To get to a tenth of a point you would need 40 pitches — 508 mm of chain. Twelve pitches is a starting point, not a specification

Four published replacement limits, and what each one is in millimetres over twelve pitches of ANSI 40 chain

Twelve pitches of 12.70 mm chain is 152.40 mm nominal, so the whole difference between the tightest published limit and the loosest is 2.29 mm. That is the number to hold in mind when somebody says to measure with a rule. Note also that the four figures are not four opinions about one quantity: 3% is a sprocket-geometry limit, 1.5% is quoted both as a fixed-centre limit AND, by Tsubaki, as a fatigue-risk threshold, 2% is what a gauge maker marks on the tool, and 200/N is a function of the drive rather than a constant at all. They answer slightly different questions, which is why this page prints all four instead of averaging them. This page sizes a part; it does not certify one. Where the answer carries a consequence — a load path, a lifting duty, a pressure boundary, a fastener holding something that can fall — confirm it against the design code that governs the application, and against the manufacturer’s own rating, before relying on it.
Limit	GO / NO-GO length (mm)	Allowance (mm)	Who publishes it, and in what words
1.5%	154.69	2.29	U.S. Tsubaki, The Complete Guide to Chain: at or below 1.5% wear elongation for transmission chain ‘there is almost no risk of fatigue failure’. Diamond-Drives and Noria’s Reliable Plant both give the same 1.5% as the limit for a fixed-centre drive or one that must run smoothly.
2%	155.45	3.05	The chain wear gauge instructions distributed by RS Components: at 2% elongation ‘the chain must be replaced’, while ‘the sprockets can still be used only if the teeth are checked and acceptable’. Tsubaki give 2% as the conveyor-chain figure.
3%	156.97	4.57	Diamond-Drives by Timken, Noria’s Reliable Plant and P-Flow’s chain maintenance guide all give approximately 3% as the maximum allowable wear elongation for most industrial drives, set by the sprocket’s tooth form. The same RS gauge says that at 3% ‘the chain and the sprockets must be replaced at once’.
200/N, N being the teeth on the large sprocket	depends on the sprocket	depends on the sprocket	Noria’s Reliable Plant, Diamond-Drives and P-Flow all publish it. Two publishers of the 200/N rule state its range of application in opposite terms. Noria’s Reliable Plant: ‘The allowable chain wear, in percent, for large sprockets with 68 teeth or greater can be calculated using the relationship 200/N.’ Diamond-Drives’ FAQ: the formula ‘applies when the large sprocket has 67 teeth or fewer.’ Only the first can be the intent, and arithmetic settles it without appeal to either: 200/N falls below 3% only when N is above 66.67. At 66 teeth it returns 3.03%, which is MORE than the 3% cap, so the cap binds; at 67 teeth it returns 2.99%, and from there up the formula is the tighter of the two. So the crossover sits between 66 and 67 teeth — and neither published statement of the rule’s range puts it there. Reliable Plant’s ‘sprockets with 68 teeth or greater’ is one tooth late; Diamond-Drives’ ’67 teeth or fewer’ is the wrong way round entirely. What the arithmetic does say, and what is worth taking away, is that the familiar 3% limit and the 200/N rule are the same rule: 3% IS 200/N at about 67 teeth, which is a large sprocket on a modest reduction. Below that the 3% cap is the sprocket-independent floor.

The 200/N rule against the large sprocket’s tooth count

Read the 66 and 67 rows together, because they settle a published disagreement. Noria’s Reliable Plant says the rule applies to “large sprockets with 68 teeth or greater”; Diamond-Drives’ FAQ says it “applies when the large sprocket has 67 teeth or fewer”. Those cannot both be right and in fact neither is: 200/N returns 3.03% at 66 teeth and 2.99% at 67, so the crossover sits between 66 and 67. Reliable Plant is one tooth late and Diamond-Drives has it the wrong way round — at 67 teeth or fewer the formula gives MORE than 3% and the 3% cap is what binds. The useful thing the arithmetic shows is that the two rules are one rule: the familiar 3% limit IS 200/N evaluated at about 67 teeth. Below that, 3% is a floor that does not depend on the sprocket; above it, the bigger the driven wheel, the less chain elongation it will tolerate, because the rollers climb its teeth further before running out of tooth.
Teeth on the large sprocket	200/N (%)	Binding limit against a 3% cap (%)	Which one binds	GO / NO-GO over 12 pitches of ANSI 40 (mm)
20	10.000	3.000	3% caps it	156.97
40	5.000	3.000	3% caps it	156.97
60	3.333	3.000	3% caps it	156.97
66	3.030	3.000	3% caps it	156.97
67	2.985	2.985	200/N binds	156.95
68	2.941	2.941	200/N binds	156.88
76	2.632	2.632	200/N binds	156.41
84	2.381	2.381	200/N binds	156.03
95	2.105	2.105	200/N binds	155.61
114	1.754	1.754	200/N binds	155.07
120	1.667	1.667	200/N binds	154.94
150	1.333	1.333	200/N binds	154.43
200	1.000	1.000	200/N binds	153.92

GO / NO-GO lengths over TWELVE pitches, by chain size — and what a 0.5 mm rule can see

Twelve pitches is the span this page opens on and it is a reasonable default, but it is worth being clear that no source consulted in this batch prescribes twelve. Noria say only that more pitches are more accurate; the chain-gauge instructions distributed by RS set the span by chain size and give six pitches for a 38.1 mm chain; Tsubaki’s guide speaks of at least five links. So the honest way to choose is to compute it, which the last two columns do. On ANSI 25 chain the entire 1.5% allowance over twelve pitches is 1.14 mm and a 0.5 mm rule resolves 0.66 percentage points — it cannot tell 1.5% from 2%. Span more chain, or use a vernier, or do both. These dimensions come from a published standard’s table, not from a formula. The standard itself is cited below and the printed values are attributed to the catalogue they were taken from; a different publisher may round differently in the last digit.
Chain	Pitch (mm)	Nominal over 12 pitches (mm)	At 1.5% (mm)	At 2% (mm)	At 3% (mm)	Whole 1.5% allowance (mm)	What a 0.5 mm rule resolves (% points)
ANSI 25	6.35	76.20	77.34	77.72	78.49	1.14	0.656
ANSI 35	9.525	114.30	116.01	116.59	117.73	1.71	0.437
ANSI 40	12.7	152.40	154.69	155.45	156.97	2.29	0.328
ANSI 41	12.7	152.40	154.69	155.45	156.97	2.29	0.328
ANSI 50	15.875	190.50	193.36	194.31	196.22	2.86	0.262
ANSI 60	19.05	228.60	232.03	233.17	235.46	3.43	0.219
ANSI 80	25.4	304.80	309.37	310.90	313.94	4.57	0.164
ANSI 100	31.75	381.00	386.71	388.62	392.43	5.71	0.131
ISO 06B-1	9.525	114.30	116.01	116.59	117.73	1.71	0.437
ISO 08B-1	12.7	152.40	154.69	155.45	156.97	2.29	0.328
ISO 10B-1	15.875	190.50	193.36	194.31	196.22	2.86	0.262
ISO 12B-1	19.05	228.60	232.03	233.17	235.46	3.43	0.219
ISO 16B-1	25.4	304.80	309.37	310.90	313.94	4.57	0.164

What a repair link costs, as a fraction of the chain’s allowable load

Tsubakimoto’s chain guide, clause 2.2.3.3. This table belongs on a wear page because a worn chain is exactly when somebody reaches for an offset link: the chain has grown and taking a pitch out is the quick fix. It is the worst possible moment to do it. A one-pitch offset link is published at a flat 35% reduction, and Tsubaki’s advice on it is plain: “if you use chain with loads that are almost the same as the maximum allowable load, you should avoid using offset links.” A two-pitch offset link is 0 to 25% and an ordinary connecting link 0 to 20%, while a tap-fit connecting link is published as no reduction at all. The page this batch ported quoted about 20% for an offset link, which is the CONNECTING-link figure — every chain has one of those, and very few need the other. This page sizes a part; it does not certify one. Where the answer carries a consequence — a load path, a lifting duty, a pressure boundary, a fastener holding something that can fall — confirm it against the design code that governs the application, and against the manufacturer’s own rating, before relying on it.
Link type	Published reduction	Fraction of the chain’s allowable load left, worst case
Tap-fit connecting link	0%	1.00
Standard connecting link	0 to 20%	0.80
Two-pitch offset link	0 to 25%	0.75
One-pitch offset link	35%	0.65

Elongation, not stretch — and four limits, not one

A worn chain has not stretched. It has lost material. The pin turns inside the bushing every time a link articulates on and off a sprocket, and both surfaces wear away. Every joint gains clearance, every pitch grows a little, and the chain gets longer — which is why the symptom is called elongation and not stretch. It matters because it is irreversible and because it cannot be taken up: re-tensioning a worn chain moves the slack, not the wear. What actually fails is the engagement. As the pitch grows the rollers no longer seat in the tooth roots; they ride progressively further up the flanks, bearing on a smaller and smaller contact area, until the chain climbs out. On a conveyor that is a stoppage. On a hoist, or a drive turning a guard, or anything where a jumped tooth drops a load, it is not.

There are four published limits and they are not four opinions about the same number. Three per cent is what Diamond-Drives, Noria’s Reliable Plant and P-Flow’s maintenance guide all give for “most industrial applications, based upon sprocket design” — it is a statement about how much tooth there is to climb. One and a half per cent appears twice over, and means two different things: Diamond-Drives and Reliable Plant give it as the limit for a fixed centre distance or a drive that must run smoothly, while Tsubaki’s Complete Guide gives it as the level below which “there is almost no risk of fatigue failure” — a strength statement, not a geometry one. Two per cent is what the chain-gauge instructions distributed by RS mark on the tool, and Tsubaki use the same figure for conveyor chain. And 200/N is not a constant at all but a function of the large sprocket’s tooth count. This page prints all four, uses the tighter of your choice and 200/N for the verdict, and says which it used.

The 200/N rule is published with two contradictory ranges of application, and arithmetic settles it. Reliable Plant: the rule applies to “large sprockets with 68 teeth or greater”. Diamond-Drives: it “applies when the large sprocket has 67 teeth or fewer”. Work it out and neither is right. 200/N gives 3.03% at 66 teeth and 2.99% at 67, so the crossover sits between 66 and 67 — Reliable Plant is one tooth late, and Diamond-Drives has the inequality the wrong way round, because below the crossover the formula returns MORE than 3% and the 3% cap is what binds. The more useful thing the arithmetic reveals is that these are not two rules: the familiar 3% limit IS 200/N evaluated at about 67 teeth. Below 67, 3% is a floor that does not care about your sprocket; above it, every extra tooth on the driven wheel tightens the allowance.

How to take the measurement, and the part nobody publishes. Measure on the TIGHT span with the drive stopped, pin centre to pin centre, over as many pitches as the span allows, away from the ends, and take the worst section you can find rather than an average — Noria are explicit that it is the tight span and that more pitches mean more accuracy. Off the drive, apply the measuring force ISO 606 specifies in its clause 3.6, because a slack chain reads short by the clearance in every joint and will tell you it is new. Then the part that is missing from every source consulted here: what your measurement can actually resolve. It is simply your instrument’s resolution divided by the length you measured. A 0.5 mm steel rule over twelve pitches of ANSI 40 chain — 152.40 mm — resolves 0.33 percentage points, which is a third of the gap between the 1.5% and 2% limits and a ninth of the whole 3% allowance. Getting to a tenth of a point needs 40 pitches. The commonly repeated advice to measure over twelve pitches is a reasonable starting point and it is not a specification; the chart and the last columns of the tables turn it into a number for your chain and your rule.

Two things to plan for at the same time as the chain. The sprockets, first: a chain that has run worn has cut its own grown pitch into the teeth, and a new chain on hooked teeth is worn out again in a fraction of its life. The RS gauge instructions draw the line cleanly — at 2% the sprockets may survive if the teeth are acceptable, at 3% the chain and the sprockets go together. Second, and this is the trap: do not take a pitch out of a long chain with an offset link. A one-pitch offset link is published by Tsubaki as a flat 35% reduction in the chain’s maximum allowable load, and a worn chain is exactly the moment somebody reaches for one. Replace the chain. If the centres genuinely cannot move, that is an argument for working to the 1.5% limit from the start, which is what the fixed-centre figure is for. Once the new chain is on, the chain length and centre distance calculator gives you the link count and the centres it needs, the sprocket geometry calculator has the diameters for checking the teeth, and the roller chain selection calculator will tell you whether the chain that keeps wearing out was ever big enough.

Frequently asked questions

At what elongation should I replace a roller chain?

It depends on the drive, and four publishers give four answers that all belong. 3% for most industrial drives with adjustable centres (Diamond-Drives, Reliable Plant, P-Flow). 1.5% for a fixed centre distance, for hardened sprockets, or where the chain must not jump a tooth — and Tsubaki give the same 1.5% for a different reason, as the level below which there is almost no risk of fatigue failure. 2% is what chain-gauge makers mark on the tool. And 200/N, N being the teeth on the large sprocket, which is tighter than 3% from 67 teeth up. This page prints all four and uses the tighter of the one you choose and 200/N.

Why does the number of teeth on the large sprocket matter?

Because that is where a worn chain runs out of tooth. As the pitch grows, the rollers sit further from the tooth roots, and the effect accumulates around the wrap — so the bigger the sprocket, the further up the teeth the last roller in the wrap is riding before it comes off. Hence 200/N: 2.63% at 76 teeth, 2.00% at 100, 1.33% at 150. Below about 67 teeth the formula gives more than 3% and the 3% cap takes over, which means the two rules are really one rule with a floor.

How many pitches should I measure across?

As many as the tight span allows, and the page works out what you get. No source consulted here prescribes a number: Noria say only that more pitches are more accurate, the RS gauge instructions set the span by chain size and give six pitches for a 38.1 mm chain, and Tsubaki’s guide speaks of at least five links. So compute it. The resolution of your answer is your instrument’s resolution divided by the length measured, so a 0.5 mm rule over twelve pitches of ANSI 40 gives 0.33 percentage points — not enough to tell 1.5% from 2%. Getting to a tenth of a point needs 40 pitches of that chain.

Where on the chain should I measure?

On the tight span, with the drive stopped, away from both sprockets, and at the worst section you can find rather than an average — wear is not uniform, and the section that has taken the most starts is the one that will jump first. Noria specify the tight span in terms. If the chain is off the drive, apply the measuring force ISO 606 gives in clause 3.6 before measuring: a slack chain reads short by the clearance in every joint, which is the one error that makes a worn chain look serviceable.

Do I have to replace the sprockets as well?

Usually, and the gauge instructions distributed by RS draw the line for you: at 2% elongation the chain must be replaced and the sprockets can still be used only if the teeth are checked and acceptable; at 3% the chain and the sprockets must be replaced together. The mechanism is that a worn chain cuts its own enlarged pitch into the tooth flanks, so the teeth end up hooked. Put a new chain on hooked teeth and it wears out in a fraction of its life, which is the commonest reason a drive “keeps eating chains”. Compare the tooth form against the sprocket geometry calculator‘s root and caliper diameters if you want a number rather than a look.

Can I just take a link out instead of replacing the chain?

No, and a worn chain is the worst possible occasion to try. Tsubakimoto publish a one-pitch offset link as a 35% reduction in the chain’s maximum allowable load, a two-pitch offset link as 0 to 25%, and their advice is that if the chain is working anywhere near its allowable load you should avoid offset links altogether. Shortening a worn chain also does nothing about the wear: the remaining joints still have all their clearance and the rollers still sit high on the teeth. It buys tension and leaves the failure mode exactly where it was.

Is elongation the same as the chain stretching?

No, and the distinction is practical rather than pedantic. Chain steel does not stretch measurably in service — if a chain has been loaded enough to yield its link plates, the drive has already had an accident. What grows is the clearance in every pin-and-bushing joint as the two surfaces wear away. That means it is irreversible, that it is driven almost entirely by lubrication and contamination rather than by load, and that a chain running in clean oil can outlast one in the open by an order of magnitude at the same tension. It also means the wear rate accelerates: a longer chain rides higher on the teeth, bears on less area, and wears faster.

Related calculators

References

U.S. Tsubaki. The Complete Guide to Chain. “When wear elongation is less than or equal to 1.5 percent for transmission chain, or less than or equal to 2 percent for conveyor chain, there is almost no risk of fatigue failure.”
Diamond-Drives by Timken. Frequently Asked Questions. Maximum allowable wear elongation “approximately 3% for most industrial applications, based upon sprocket design” and “approximately 1.5%” where centres are fixed or the drive must run smoothly; the 200/N relationship, N being the teeth in the large sprocket; and minimum tooth counts against speed — 12 slow, 17 medium, 25 high. Its stated range of application for 200/N contradicts Reliable Plant’s; see the note on the wear page.
Noria Corporation / Reliable Plant. Extend the wear life of roller chains. The 3% and 1.5% limits; “the allowable chain wear, in percent, for large sprockets with 68 teeth or greater can be calculated using the relationship 200/N”; “measure the tight span of the chain”; “the more pitches (pins) contained within the measurement increase the accuracy”; and, for a chain off its sprockets, “the ANSI-specified measuring load should be applied to the chain so that the slack has been removed”.
RS Components. Chain wear gauge instructions, document A700000015008087. At 2% elongation “the chain must be replaced” while “the sprockets can still be used only if the teeth are checked and acceptable”; at 3% “the chain and the sprockets must be replaced at once”. Also the source for the observation that the number of pitches a gauge spans is set by the chain size, not fixed at twelve — it gives 6 pitches for a 38.1 mm pitch chain.
PFlow Industries. Roller Chain Maintenance and Lubrication. “In most roller chain drives, the chain is considered worn out when it has reached 3% wear elongation”; “allowable wear is limited to 200/N (N = number of teeth on largest sprocket)”; and the point this batch used to drop a max-speed column: “the recommended type of lubrication [is] shown in the horsepower tables in the respective standards”, types I/II/III being manual or drip, bath or slinger disc, and oil stream or pressure spray.
Wikipedia, Roller chain. Quoted here only for the ANSI numbering rule, which it states in the same terms as the catalogues: “the first digits indicate the pitch of the chain in eighths of an inch, with the last digit being 0 for standard chain, 1 for lightweight chain, and 5 for bushed chain with no rollers”. Also “one rule of thumb is to replace a roller chain which has elongated 3% on an adjustable drive or 1.5% on a fixed-center drive”, and, on odd link counts, that such a chain “tends to be not so strong”.
Tsubakimoto Chain. Strength Differences Between Chain and the Connecting Links and Offset Links, chain-guide.com 2.2.3.3. Reduction against the chain’s maximum allowable load: tap-fit connecting link none; standard connecting link 0 to 20%; two-pitch offset link 0 to 25%; one-pitch offset link 35%. “If you use chain with loads that are almost the same as the maximum allowable load, you should avoid using offset links.”
ISO 606:1994, Short-pitch transmission precision roller and bush chains, attachments and associated chain sprockets (second edition, 1994-02-15). Cited by number; only the standard’s preview was reachable, which confirms clause 3.6 (a measuring force applied when a chain’s length is measured) and clause 5 (chain wheels, described in its own foreword as ‘the unification of all the relevant national Standards’). The dimensional figures here are taken from five named catalogues instead — see the breaking-load note.
iwis / JWIS. Catalogue: precision chains, British Standard roller chain table. The lowest of the five minimum-breaking-load columns compared in this batch, and the one that matches the figures usually quoted as ISO 606’s own minima (16B-1 at 60.0 kN, 20B-1 at 95.0 kN). Used as the default on the selection page.
chinatransmissions. ANSI Roller Chain Size Chart | Dimension Guide. Second, independent transcription of the ANSI roller diameters and roller widths, agreeing with Martin’s where the two overlap. It is also the source that settles chain 41’s roller diameter at 0.306 in — the page being ported had 6.35 mm, which is 0.250 in, chain 41’s roller WIDTH.