ISO 34-1
Rubber Tear Strength: Trouser, Angle and Crescent

Written by Chris Thomas, Universal Grip Company · Reviewed by Frank Bertini · Updated October 24, 2026

ISO 34-1 is the international standard for the tear strength of vulcanized and thermoplastic rubber. A trouser, angle or crescent test piece is pulled apart in a tensile machine and the force needed to tear it is divided by its thickness to give a tear strength in kilonewtons per metre. It is the ISO companion to ISO 37 tensile testing and the counterpart of ASTM D624: seal, hose, belt, glove, footwear, tyre and automotive rubber specifications set tear limits against one or the other, and many specify both.

This guide covers the current 2022 edition and what changed since 1994; all three methods with their speeds, cut and nick depths and the tolerances the standard actually gives (the nick is 1.0 ± 0.2 mm, not the 0.5 mm of the ASTM dies); the 2.0 ± 0.2 mm sheet, cutting, conditioning and the 24-hour rule; the two-direction protocol; the self-tightening grip requirement; the step-by-step procedure; the calculation with a worked trouser example using the ISO 6133 median of peaks; the precision data that explains why tear results scatter; a numeric comparison with ASTM D624; and the fixtures, grips, dies and machines Universal Grip supplies for the test.

In this guide: What ISO 34-1 Measures · Editions: 1994 to ISO 34-1:2022 · Test Pieces: Trouser, Angle and Crescent · Machine, Speed and Grips · Conditioning and Temperature · Step-by-Step Procedure · Calculation and Worked Example · Precision: Why Tear Results Scatter · Common Errors and Their Effect · ISO 34-1 vs ASTM D624 · Equipment for ISO 34-1: Fixtures, Grips, Dies and Machines · FAQ

What ISO 34-1 Measures

ISO 34-1 defines three methods for determining the tear strength of rubber, each with its own test piece and its own definition of the result. They are not interchangeable: clause 4 of the standard says that results from the different test pieces do not correlate with each other, so a specification must name the method, and a result must be reported with it.

• Method A, trouser test piece: a rectangular strip slit along its centre line so that the two legs can be gripped one in each jaw. The tear propagates along the slit at a steady rate equal to half the jaw speed. The result is the median force of the peaks on the force trace, evaluated per ISO 6133, divided by thickness. The method measures tear propagation and is insensitive to the length of the starting cut, which is why the standard calls it the preferred method.

• Method B, angle test piece: a strip with a 90° internal angle in one edge. Procedure (a) tests it unnicked, so the result combines the force to initiate a tear at the apex and the force to propagate it; the two cannot be separated. Procedure (b) tests it nicked at the apex, which removes initiation and measures propagation only. In both cases the result is the maximum force divided by thickness.

• Method C, crescent test piece: a curved test piece with a nick in the centre of its concave edge. The result is the maximum force needed to extend the nick, divided by thickness. The rate of tear propagation is unrelated to the jaw speed for this shape.

All three results are expressed as tear strength Ts in kN/m, numerically the same as N/mm. The tear strength depends on the shape, the speed, the temperature and, for anisotropic material, the direction of tearing relative to the grain, so the standard requires the test piece type, procedure and direction to be stated with every result. The tear testing application page gives the lab overview of tear methods across materials; this guide covers the rubber standard itself.

Editions: 1994 to ISO 34-1:2022

The current edition is ISO 34-1:2022, the fifth edition, published June 2022; earlier editions were 1994, 2004, 2010 and 2015, and several equipment pages still cite the withdrawn 2010 edition. The 2022 revision moved the method descriptions into clause 4, redrew the trouser-insertion figure and clarified wording; it did not change dimensions, speeds or nick depths. Two earlier changes matter for comparing old data: force accuracy rose from grade B (2 %) in 1994 to ISO 5893 Class 1 (1 %), and the thickness-variation rule between pieces in a group was relaxed from 1.5 % to 7.5 %. ISO 34-2 covers the small Delft test piece for finished products and small samples.

Test Pieces: Trouser, Angle and Crescent

All three test pieces are die-cut in a single stroke from sheet 2.0 ± 0.2 mm thick, moulded or cut from product, on a slightly yielding support so the die edges stay sharp and the cut edges are square. The dies must have sharp, perpendicular cutting edges with minimal concavity, because a ragged edge starts a tear early and lowers the result.

• Trouser (Method A, Figure 1): a strip of the proportions in the standard's Figure 1, nominally 100 mm long and 15 mm wide, with a cut 40 ± 5 mm long made along the centre of the width from one end. The cut is made with a sharp blade; the last millimetre or so is finished with a razor so the tear starts from a clean, sharp front rather than a crushed one. The same die serves DIN 53507 and ISO 8067 Method A for foam.

• Angle (Method B, Figure 2): a strip with parallel-sided tabs at each end and a 90° re-entrant angle cut into one edge at the centre. Procedure (a) uses it as cut. Procedure (b) adds a nick 1.0 ± 0.2 mm deep at the apex of the internal angle. The unnicked angle piece is geometrically the same family as the ASTM D624 Die C and ASTM D1004 die.

• Crescent (Method C, Figure 3): a curved test piece with parallel tabs, nicked 1.0 ± 0.2 mm deep at the centre of the concave inner edge. It is the same family as the ASTM D624 Dies A and B, but those are nicked to 0.50 mm, so a crescent nicker bought for D624 is the wrong depth for ISO 34-1.

Nicking. The nick is made with a razor blade or knife clamped so that it cuts perpendicular to the major axis of the test piece and to its surface, wetted with water or soap solution so it does not drag. The standard requires the nick depth to be measured to an accuracy of 0.05 mm at a magnification of at least 10×; that 0.05 mm is the measurement accuracy, not the tolerance on the nick, which is ± 0.2 mm. A guided nicking jig with a depth stop makes the 1.0 mm repeatable; a hand-held blade does not. If the test pieces are to be aged, nick them after ageing, and in every case test them within 24 hours of cutting or nicking.

Direction. Rubber sheet has a grain from calendering or moulding flow, and tear strength along and across it can differ by 20 % or more. Where the sheet allows, cut test pieces in two directions 90° apart and report each separately. The trouser tear runs parallel to the length of the piece; the angle and crescent tears run perpendicular to it, so define the direction relative to the grain, not to the die.

Thickness. Measure with an ISO 23529 Method A gauge where tearing will occur: along the trouser cut path, at the angle apex, at the crescent nick. No reading on a piece may differ from that piece's median by more than 2 %, and the medians of pieces in a group must lie within 7.5 % of the grand median; otherwise reject. The calculation uses the median for the piece.

Machine, Speed and Grips

• Machine: a tensile machine meeting ISO 5893 Class 1, with force measured to within 1 % of the reading. For the trouser method the standard says a low-inertia machine with autographic (continuous) recording is essential, because the force trace is a saw-tooth of tear peaks and a sluggish or under-sampled system flattens them and biases the median. A load cell of 100 N to 1 kN suits most compounds; typical tear forces on 2 mm sheet are 10 to 250 N, and reinforced or hard compounds reach about 500 N.

• Speed: 100 ± 10 mm/min for the trouser piece, 500 ± 50 mm/min for the angle and crescent pieces. The trouser speed is slower because the tear front advances at half the jaw speed and the force must be recorded as a steady propagation, not a snap.

• Grips: self-tightening grips that apply a uniform pressure across the tab and keep the test piece axial and symmetric as the force rises. The angle and crescent pieces are gripped only in their parallel-sided tabs, never on the curved or angled portion. The trouser piece is inserted with one leg in each jaw, so that the uncut portion hangs centrally between them and the legs are not pre-stretched. The jaw faces should be at least as wide as the tab, 25 mm square faces being the usual choice.

• Sampling rate: the trouser trace needs enough data points to resolve every peak. At 100 mm/min a 60 mm tear takes 72 s; recording at 50 to 100 Hz gives several thousand points and a reliable median. Peak-hold force gauges that record only the maximum cannot run Method A.

The ASTM D624 tear fixture page shows the jaw arrangement for crescent and angle pieces; the same hardware runs ISO 34-1 Methods B and C with only the nick depth changed. Roller grips and pneumatic grips both meet the uniform-pressure requirement; the equipment section below compares them.

Conditioning and Temperature

Conditioning follows ISO 23529, the general rubber conditioning standard. The sheet must be at least 16 hours past vulcanization before testing (for test pieces cut from product, 16 hours after manufacture), conditioned for at least 3 hours at the standard laboratory temperature before cutting, and the finished test pieces tested within 24 hours of cutting or nicking. The standard laboratory temperature is 23 ± 2 °C, or 27 ± 2 °C in tropical countries; humidity control is only required for compounds known to be moisture-sensitive. Tests at other temperatures use the ISO 23529 list and the shortest soak that brings the piece to temperature, to limit ageing during the soak.

All test pieces in a comparison must be tested at the same temperature, because rubber tear strength falls steeply with temperature and the ranking of compounds can change between 23 °C and 70 °C. An environmental chamber with pneumatic grips inside is the usual way to run ISO 34-1 hot or cold.

Step-by-Step Procedure

• 1. Confirm the sheet is 2.0 ± 0.2 mm, at least 16 hours past cure and conditioned 3 hours at 23 ± 2 °C. Mark the grain direction.

• 2. Die-cut at least five test pieces per direction in a single stroke; inspect edges for nicks or tears and discard damaged pieces.

• 3. For the trouser piece, cut the 40 ± 5 mm slit along the centre line and finish the last millimetre with a razor. For angle (b) and crescent pieces, nick 1.0 ± 0.2 mm with a wetted blade in a guided jig and check the depth at 10× or more. Nick after ageing if the pieces are aged.

• 4. Measure thickness at the tear location with an ISO 23529 gauge; take the median per piece, check the 2 % and 7.5 % rules, and record the median.

• 5. Set the machine to 100 mm/min (trouser) or 500 mm/min (angle, crescent); fit a load cell such that tear forces fall well inside its calibrated range; check the sampling rate.

• 6. Mount the piece symmetrically and axially: trouser legs one per jaw with the uncut portion free; angle and crescent pieces gripped only on the parallel tabs. Do not pre-tension.

• 7. Start the test immediately after mounting and record force continuously until the piece has torn through or the tear leaves the intended path.

• 8. Trouser: evaluate the trace per ISO 6133 and take the median of the peaks. Angle and crescent: take the maximum force. Discard tests where the tear runs off the path, the piece slips or breaks in the grips, and replace them.

• 9. Calculate Ts = F/d for each piece; report the median and the range for each direction, to the nearest kN/m, with the method, test-piece type, procedure (a) or (b), temperature and conditioning.

Calculation and Worked Example

The tear strength is Ts = F / d, where F is the force in newtons (median of the peaks for the trouser piece per ISO 6133, maximum force for the angle and crescent pieces) and d is the median thickness of the test piece in millimetres. The result is in kN/m, which is numerically identical to N/mm. For five pieces in a direction, report the median of the five values and the range from lowest to highest. To convert ASTM results in lbf/in to kN/m, multiply by 0.1751.

Worked example, trouser piece. A nitrile seal compound is tested at 100 mm/min. The thickness readings along the tear path are 2.01, 2.03 and 2.02 mm, median 2.02 mm. The force trace shows seven peaks: 17.4, 19.0, 18.1, 18.6, 17.9, 18.3 and 19.5 N. Sorted, they are 17.4, 17.9, 18.1, 18.3, 18.6, 19.0, 19.5, and the median (the fourth of seven) is 18.3 N. Ts = 18.3 / 2.02 = 9.06 kN/m, reported as 9 kN/m. Had the maximum peak been used instead of the median, the result would have been 19.5 / 2.02 = 9.65 kN/m, about 7 % high; that is the error a peak-hold gauge introduces on Method A.

Worked example, crescent piece. The same compound nicked 1.0 mm and tested at 500 mm/min gives a maximum force of 85.0 N on a piece of median thickness 2.10 mm. Ts = 85.0 / 2.10 = 40.5 kN/m, reported as 40 or 41 kN/m. The crescent result is more than four times the trouser result on the same material, which is normal: the crescent measures the force to drive a tear from a nick through a wide, stretched section, while the trouser measures steady propagation at a constant tear front. It is also why the standard says the methods do not correlate.

Precision: Why Tear Results Scatter

Rubber tear is one of the least repeatable mechanical tests. The interlaboratory data in the standard's informative annex give, for the trouser piece, a within-laboratory repeatability r of roughly 45 % of the mean and a between-laboratory reproducibility R of roughly 70 %; the ASTM D624 precision statement shows reproducibility of the same order for the nicked dies. The causes are physical rather than instrumental: the tear front follows the path of least resistance through a heterogeneous filled network, the crack tip radius after nicking varies with blade sharpness and wetting, and the grain direction changes the result by tens of percent.

In practice a specification limit should be set against the median of at least five pieces, a change of nick blade, die or operator can shift results more than a change of compound, and comparing two laboratories needs an agreed method, die, nick depth and direction, not just a common unit. The ASTM D624 guide covers the same point from the ASTM side.

Common Errors and Their Effect

• Using a 0.5 mm nicker for ISO work. D624 crescent dies are nicked 0.50 ± 0.05 mm; ISO 34-1 requires 1.0 ± 0.2 mm. A shallower nick raises the crescent result by roughly 10 to 20 % on a typical compound because more force is needed to start the tear from the shorter crack.

• Reporting the maximum force on the trouser piece. The standard requires the ISO 6133 median of peaks. Using the maximum biases the result upward by 5 to 15 % depending on how knotty the tear is.

• Running the trouser piece at 500 mm/min. The speed is 100 ± 10 mm/min. At 500 mm/min the legs stretch, the tear becomes unstable and the force peaks are smeared by machine inertia.

• Gripping the angle or crescent piece on the curved portion. Clamping outside the parallel tabs introduces bending at the tear site and typically lowers the maximum force.

• Pre-stretched trouser legs. If the legs are tensioned while clamping, the first part of the trace is leg extension, not tear, and the first peaks are invalid.

• Thickness measured away from the tear path. A 0.1 mm error on a 2.0 mm piece is 5 % in the result. Measure where the tear will run.

• Testing more than 24 hours after nicking, or nicking before ageing. The crack tip blunts and the result rises.

• Mixing directions. Reporting a pooled median from pieces cut along and across the grain hides a difference that can exceed 20 % and makes the data useless for comparison.

• Too few test pieces. With a repeatability near 45 %, three pieces give a median that is barely better than a single value; the standard's minimum is five per direction.

ISO 34-1 vs ASTM D624

The two standards use the same physics and the same unit, and the angle and crescent shapes are the same family, but the details differ enough that results are not interchangeable. ASTM D624 is the usual choice for North American specifications, ISO 34-1 for European and Asian ones; automotive and footwear specifications often cite both.

• Crescent nick depth: ISO 34-1 Method C 1.0 ± 0.2 mm; D624 Dies A and B 0.50 ± 0.05 mm. This is the most important difference and the one most often got wrong.

• Angle piece: D624 Die C is tested unnicked only; ISO 34-1 Method B has both an unnicked procedure (a), which is the Die C equivalent, and a nicked procedure (b), which has no D624 twin.

• Trouser speed: ISO 34-1 100 ± 10 mm/min; D624 Die T and Die CP 50 ± 5 mm/min. Angle and crescent speed is 500 ± 50 mm/min in both.

• Thickness: ISO 34-1 2.0 ± 0.2 mm; D624 2.3 ± 1.0 mm (1.3 to 3.3 mm permitted), with the result corrected by the measured thickness in both.

• Trouser result: ISO 34-1 median of peaks per ISO 6133; D624 permits the mean or median of the peaks and, for Die CP, specifies the evaluation separately.

• Thickness measurement: ISO 23529 Method A gauge versus ASTM D3767; conditioning per ISO 23529 versus ASTM D1349. The practical requirements are nearly identical.

• Units: kN/m in both; D624 also reports lbf/in (× 0.1751 = kN/m).

The ISO tensile counterpart is ISO 37 and the ASTM one is ASTM D412; the same grips, dies and machine run all four standards, and a rubber laboratory normally qualifies its equipment for the set. The elastomers testing page lists the rest of the rubber methods we equip.

Equipment for ISO 34-1: Fixtures, Grips, Dies and Machines

ISO 34-1 needs a correctly dimensioned die for each test piece, a nicking jig that cuts 1.0 mm repeatably, grips that hold a 2 mm rubber tab without slipping or cutting it, and a machine with a continuous force trace and the two speeds. Universal Grip supplies all four, and the same hardware runs ASTM D624, ISO 37 and ASTM D412.

• Tear fixture and grips. Our ASTM D624 tear fixture holds crescent and angle pieces on their parallel tabs with uniform pressure and keeps the tear site centred between the jaws; it runs ISO 34-1 Methods B and C unchanged. For the trouser piece, self-tightening eccentric roller grips are the classic choice: the clamping force rises with the tear force, the rubber cannot slip, and the leg is not crushed at the jaw. Pneumatic grips with 25 mm serrated or rubber-faced jaws and a regulated pressure of 2 to 4 bar give a constant, repeatable clamp and are the better choice for high-throughput QC and for tests inside a chamber. Vise grips with flat faces work for angle and crescent pieces at low cost but need a hand-tightened clamp that is operator-dependent.

• Cutting dies and press. Trouser, angle and crescent dies to the ISO 34-1 figures, in hardened steel for long life or ribbon steel for low cost, cut in a single stroke on our die press. The same press cuts ISO 37 dumbbells. See the video below on hardened versus ribbon-steel dies.

• Nicking jig. A guided razor-blade nicker with a 1.0 mm depth stop for crescent and angle pieces; the 0.5 mm stop for D624 is a separate setting, not a substitute.

• Machines. Tear forces are small: a Mark-10 test frame with a 500 N or 1 kN load cell and continuous data capture runs all three methods on a bench, and a 2.5 kN or 5 kN universal testing machine adds the stiffness, speed accuracy and software needed for ISO 5893 Class 1 and ISO 6133 peak analysis, plus capacity for ISO 37 tensile on the same frame. Choose a load cell so that typical tear forces sit between 10 % and 90 % of its range: a 500 N cell for most compounds, 1 kN for reinforced or hard rubber.

• Chambers. For tear strength at service temperature, an environmental chamber with pneumatic grips mounted inside, −70 to +250 °C, and extension rods to keep the load cell at room temperature.

Tell us the test-piece type, the compound and expected tear force, and the machine the fixture must fit, and we will specify the die, nicker, grips and load cell as a set. The tensile grip selection guide explains the general grip choices for rubber.

ISO 34-1 FAQ

What is ISO 34-1?

ISO 34-1 is the international standard for the tear strength of vulcanized and thermoplastic rubber using trouser (Method A), angle (Method B, unnicked or nicked) and crescent (Method C, nicked) test pieces. The tear force is divided by the thickness to give tear strength in kN/m. The current edition is ISO 34-1:2022.

What is the test speed for ISO 34-1?

100 ± 10 mm/min for the trouser test piece (Method A) and 500 ± 50 mm/min for the angle and crescent test pieces (Methods B and C).

How deep is the nick in ISO 34-1?

1.0 ± 0.2 mm for both the nicked angle piece and the crescent piece, measured to 0.05 mm at 10× magnification. ASTM D624 crescent dies are nicked 0.50 ± 0.05 mm, so a D624 nicker is the wrong depth for ISO 34-1.

How thick is the ISO 34-1 test piece?

The sheet is 2.0 ± 0.2 mm thick. Thickness is measured at the tear location with an ISO 23529 gauge; readings on a piece must be within 2 % of its median, and the medians of pieces in a comparison within 7.5 % of the grand median.

How is tear strength calculated in ISO 34-1?

Ts = F / d, with F in newtons (median of the peaks per ISO 6133 for the trouser piece, maximum force for angle and crescent) and d the median thickness in mm, giving kN/m (equal to N/mm). Report the median and range of at least five pieces per direction, to the nearest kN/m.

How many test pieces does ISO 34-1 require?

At least five per sample, and where the sheet allows, five in each of two directions 90° apart, because tear strength along and across the grain can differ by 20 % or more.

Which ISO 34-1 method is preferred?

The trouser method (A). It measures steady tear propagation, is insensitive to the length of the starting cut, and gives the lowest scatter. The angle and crescent methods are faster to run and are common in QC specifications.

What is the difference between ISO 34-1 and ASTM D624?

The crescent nick is 1.0 ± 0.2 mm in ISO 34-1 versus 0.50 ± 0.05 mm in D624; the trouser speed is 100 mm/min versus 50 mm/min; ISO sheet is 2.0 ± 0.2 mm versus D624's 2.3 ± 1.0 mm; ISO has a nicked angle procedure that D624 does not; and ISO requires the median of peaks for the trouser piece. Results are not interchangeable.

What grips are used for ISO 34-1?

Self-tightening grips with uniform pressure: eccentric roller grips or pneumatic grips with 25 mm faces. Angle and crescent pieces are gripped only on their parallel tabs; the trouser piece has one leg in each jaw with the uncut portion free and unstretched.

What is the conditioning for ISO 34-1?

Per ISO 23529: at least 16 hours after vulcanization, at least 3 hours at 23 ± 2 °C (or 27 ± 2 °C) before cutting, test within 24 hours of cutting or nicking, and nick after any ageing.

How repeatable is the ISO 34-1 tear test?

Poorly, by the nature of rubber tear: the interlaboratory data give a repeatability of roughly 45 % and a reproducibility of roughly 70 % of the mean for the trouser piece. Specifications should be set against the median of five pieces with the method, die, nick depth and direction fixed.

What machine capacity is needed for ISO 34-1?

Very little: tear forces on 2 mm sheet are typically 10 to 250 N, up to about 500 N for reinforced compounds. A 500 N or 1 kN load cell on a Mark-10 frame or a 2.5 to 5 kN universal testing machine with continuous data recording and ISO 5893 Class 1 force accuracy is the normal setup.

Need a tear fixture, grips, cutting dies or a complete tear-testing system for ISO 34-1? Contact Universal Grip with your test-piece type, expected tear force and machine connection and we will quote the right setup.