top of page

ISO 6892-1
Tensile Testing of Metallic Materials at Room Temperature

Universal Grip tensile fixture for ISO 6892-1 metallic tensile testing

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

ISO 6892-1 is the international standard for tensile testing of metallic materials at room temperature. It defines the properties nearly every metal specification is written against: upper and lower yield strength ReH and ReL, proof strength Rp0.2, tensile strength Rm, percentage elongation after fracture A, total extension at maximum force Agt and reduction of area Z. It also fixes how fast the test may run, which is where most of its length goes, because the 2009 edition introduced strain-rate control (Method A) and the ±20 % rate bands that European mills, aerospace primes and automotive steel buyers now specify. Steel, aluminium, copper, titanium, nickel alloys, rebar, wire, tube, castings and forgings are all tested to it; in Europe it replaced EN 10002-1 at the end of 2009 and it is the direct counterpart of ASTM E8/E8M.

This guide covers the current edition, ISO 6892-1:2019, and what changed from 2016; every property and symbol defined; the three control methods A1, A2 and B with their numeric rate ranges and a decision tree for choosing one; the proportional gauge length rule k = 5.65 and the specimen tables from Annexes B, C, D and E with machining tolerances; the force and extensometer classes; the step-by-step procedure, the fit-together elongation measurement and the L0/3 fracture rule; a worked example with the arithmetic shown; the errors that most often produce an out-of-tolerance result; a numeric comparison with ASTM E8/E8M; and the grips, extensometers and machines Universal Grip supplies for each type of test piece.

In this guide: What ISO 6892-1 Measures · Editions: 2009, 2016, 2019 and What Changed · Test Rates: Methods A1, A2 and B · Test Pieces: the k = 5.65 Rule and Annexes B, C, D and E · Machine, Grips, Alignment and Extensometer Classes · Step-by-Step Procedure · Calculations and a Worked Example · Common Errors and Their Size · ISO 6892-1 vs ASTM E8/E8M · Grips, Extensometers and Machines for ISO 6892-1 · FAQ

What ISO 6892-1 Measures

ISO 6892-1 covers the tensile testing of metallic materials and defines the mechanical properties that can be determined at room temperature, which the standard sets at 10 to 35 °C, or 23 ± 5 °C where controlled conditions are specified. A test piece of known cross-section S0 is pulled to fracture while force and extension are recorded; stress R is always force divided by the original cross-section. The properties, with the symbols the standard uses, are:

  • ReH, upper yield strength: the maximum stress before the first drop in force in a material that yields discontinuously (mild steel, some aluminium alloys).

  • ReL, lower yield strength: the lowest stress during plastic yielding, ignoring the initial transient; the value most structural steel specifications quote.

  • Rp, proof strength at non-proportional extension: the stress at a stated plastic extension, almost always 0.2 %, written Rp0.2. Used for materials without a yield point: quenched and tempered steels, stainless steels, aluminium, titanium and nickel alloys. Rp0.1 and Rp1 are also used.

  • Rt, proof strength at total extension: the stress at a stated total (elastic plus plastic) extension, for example Rt0.5; the ISO equivalent of ASTM's extension-under-load yield.

  • Rm, tensile strength: maximum force Fm divided by S0.

  • Ag and Agt: plastic and total percentage extension at maximum force, measured with the extensometer; Ag is the uniform elongation used in forming simulations.

  • A, percentage elongation after fracture: (Lu − L0) ÷ L0 × 100, measured by fitting the broken halves together; written with the gauge-length convention, so A (for k = 5.65), A50mm or A80mm for non-proportional gauge lengths, and A11.3 for the longer proportional length.

  • At: total percentage extension at fracture from the extensometer, and Ae, the yield-point extension between ReH and the start of uniform work hardening.

  • Z, percentage reduction of area: (S0 − Su) ÷ S0 × 100, from the minimum cross-section Su after fracture.

For materials that yield discontinuously, Rp and Rt are not normally determined unless the specification asks for them. All strengths are reported to 1 MPa, Ae to 0.1 %, other extensions and elongations to 0.5 % and Z to 1 %. Our metals testing page gives the lab overview; this guide is about the standard itself.

Editions: 2009, 2016, 2019 and What Changed

ISO 6892-1 replaced ISO 6892:1998 and EN 10002-1 in 2009 and was the first metals standard to put strain-rate control into the main text, based on the European TENSTAND programme that showed yield strength shifting by more than 10 % between the slowest and fastest speeds the old standards allowed. The 2016 second edition split Method A into A1 (closed-loop strain control from the extensometer) and A2 (open-loop crosshead control at an estimated strain rate). The current third edition, ISO 6892-1:2019, was published in November 2019, systematically reviewed and confirmed in September 2025, and has no revision in progress. Several equipment pages still call 2016 the current version, or cite a 2017 or 2020 edition that does not exist.

  • 2019 changes are minor: the extensometer used for Method A1 no longer has to be a clamped type, any device that measures the extension of the test piece is acceptable; the figure that shows the rate ranges now labels extensometer control as A1 and crosshead control as A2; a note to the Annex B flat-specimen table allows the nominal width to be used for S0 when the machining tolerances are met; a note to the Annex D tolerance table requires every piece to be measured when they are not; and the regression quality measure is called the coefficient of determination.

  • Still in force from 2009/2016: the four strain-rate ranges and the ±20 % band, the stress-rate Method B with its two ranges by modulus, the k = 5.65 proportional gauge length, the 10 to 35 °C room-temperature band and the annexes on computer-controlled testing (A), compliance correction (F) and measurement uncertainty (K).

  • Companion parts: ISO 6892-2 covers elevated temperature and was reissued in 2026; ISO 6892-3 covers +10 to −196 °C; ISO 6892-4 covers liquid helium.

Test Rates: Methods A1, A2 and B

Rate matters because yield and proof strength of most alloys rise with strain rate. ISO 6892-1 offers two approaches. Method A controls the strain rate in the test piece and is the preferred method; Method B controls the stress rate in the elastic region and is kept for compatibility with older specifications, with a note that the committee intends to withdraw it. Within Method A there are four strain-rate ranges, each with a tolerance of ±20 %:

  • Range 1: 0.00007 s⁻¹. The slowest band, for ReH, Rp and Rt where a specification calls for it, typically rate-sensitive aluminium, titanium or nickel alloys, or where results must match older slow-speed data.

  • Range 2: 0.00025 s⁻¹. The recommended rate for ReH, Rp and Rt, and for ReL and Ae if the lower yield is determined. 0.00025 s⁻¹ is 0.015 min⁻¹, the same nominal value as ASTM E8 Method B.

  • Range 3: 0.002 s⁻¹. Permitted for ReL and Ae, and for the properties after yield.

  • Range 4: 0.0067 s⁻¹ (0.4 min⁻¹). The recommended rate for Rm, Ag, Agt, A, At and Z once yield is past. If only Rm is required, the whole test may run in Range 3 or 4.

Method A1 closes the control loop on the extensometer, so the rate in the gauge length is held directly, and the machine needs a controller fast enough to do it without overshoot and an extensometer that stays on through yield. Method A2 sets the crosshead speed from the parallel length: vc = Lc × ė, so for Lc = 60 mm at Range 2 the crosshead runs at 60 × 0.00025 = 0.015 mm/s, or 0.9 mm/min, and at Range 4 at 24 mm/min. Because the load string also stretches, the true strain rate in the piece under A2 is lower than the nominal while the force is rising; Annex F gives a compliance correction and most labs simply aim near 0.0003 s⁻¹ so that the actual rate stays inside the band. Up to about half the expected yield the standard does not constrain the speed, so the elastic region may be crossed quickly.

Method B sets a stress rate in the elastic region according to the modulus of the material: 2 to 20 MPa/s for E below 150 GPa (magnesium, aluminium, copper alloys, titanium) and 6 to 60 MPa/s for E of 150 GPa and above (steels, nickel alloys, tungsten). Under Method B the strain rate while determining Rp or Rt must not exceed 0.0025 s⁻¹, and after yield it must not exceed 0.008 s⁻¹; the machine must switch from stress control to crosshead control before the material yields, which in practice means before about 80 % of the expected Rp0.2, because a stress-controlled machine will run away when the stress stops rising.

Choosing a method. If the specification names a method and rate, use it. Otherwise: a servo-controlled machine with an extensometer that can stay on through yield should run A1 at Range 2 to yield and Range 4 to fracture; that is the configuration that gives the smallest spread between laboratories. A machine without closed-loop strain control runs A2 at the same nominal rates. Method B is for legacy data or for a specification that still quotes MPa/s. One exception: materials with a yield-point elongation (Lüders plateau) are not well suited to strain control through the plateau, because the strain rate is set by the Lüders front rather than the machine; test these in crosshead control (A2) through the yield region. Whichever method is used, the report must state it, together with the rate, because a Rp0.2 at Range 1 and one at Range 4 can differ by several percent on the same bar.

Test Pieces: the k = 5.65 Rule and Annexes B, C, D and E

ISO 6892-1 does not use a single fixed specimen. It defines proportional test pieces, whose original gauge length is tied to the cross-section by L0 = k√S0 with k = 5.65, and non-proportional test pieces with a fixed gauge length. For a round bar k = 5.65 gives L0 = 5 d0 exactly, because √(π d0²/4) = 0.886 d0 and 5.65 × 0.886 = 5.0. Where the section is too small for 5.65, k = 11.3 is the preferred alternative; L0 must be at least 15 mm and may be rounded to the nearest 5 mm if the rounding is under 10 % of L0. Elongation measured on different gauge lengths is not comparable, which is why ASTM E8 (4 d0) and E8M or ISO (5 d0) results on the same bar differ by a few percentage points of A. The shapes are set out in four annexes:

  • Annex B, flat products 0.1 to 3 mm thick: Type 1 is 12.5 ± 1 mm wide, L0 50 mm, parallel length Lc at least 57 mm (75 mm recommended), free length between the grips 87.5 mm; Type 2 is 20 ± 1 mm wide, L0 80 mm, Lc at least 90 mm (120 recommended), free length 140 mm; Type 3 is 25 ± 1 mm wide, L0 50 mm, Lc at least 60 mm. Transition radius at least 20 mm. Width machining tolerance ± 0.05 mm on the 12.5 mm piece and ± 0.10 mm on the 20 and 25 mm pieces; if those tolerances are held the nominal width may be used for S0. Proportional pieces are also allowed for sheet.

  • Annex C, wire, bar and sections under 4 mm: tested unmachined, full section. L0 is 200 ± 2 mm or 100 ± 1 mm; the grip separation must be at least L0 + 3 b0 and at least L0 + 20 mm, or 50 mm if elongation is not measured. S0 to ± 1 % from the mass and length or from diameter measurements.

  • Annex D, sheet 3 mm and over, bar and sections 4 mm and over: the machined round proportional pieces are d0 = 20, 14, 10 and 5 mm with L0 = 100, 70, 50 and 25 mm and Lc at least 110, 77, 55 and 28 mm; Lc must be at least L0 + d0/2 and the transition radius at least 0.75 d0. For other shapes Lc is at least L0 + 1.5 √S0, the radius at least 12 mm, and rectangular pieces are kept to a width-to-thickness ratio of 8:1 or less. Non-proportional flats are 40 mm wide with L0 200 mm, 25 mm wide with L0 200 mm, and 20 mm wide with L0 80 mm.

  • Annex D machining tolerances: diameter or width tolerance ± 0.02 mm up to 6 mm, ± 0.03 mm over 6 to 10 mm, ± 0.05 mm over 10 to 18 mm, ± 0.10 mm over 18 to 30 mm, with a shape (variation along the length) tolerance of 0.03 to 0.05 mm by band. When these are met and each dimension is within 0.5 % of nominal, S0 may be taken from the nominal dimensions; when they are not, every piece must be measured.

  • Annex E, tubes: a full-section length with plugs in the ends (the plug must not extend more than D0/4 past the gauge mark into the gauge length and not more than D0 beyond the grips), a longitudinal strip that is flattened only in the gripped ends, or a round machined from the wall. Walls under 3 mm follow Annex B and 3 mm and over follow Annex D for the strip dimensions.

The original cross-section S0 is measured to ± 2 % (± 1 % recommended) at enough points to establish the mean; the standard asks for the measurement at several sections along the parallel length and takes the average, where ASTM uses the smallest section. Gauge marks are applied to ± 1 % of L0 with a marking tool or fine scribe that does not notch the surface; a punch mark that initiates fracture invalidates the elongation. Machined pieces must be free of cold work and decarburisation in the gauge length; the surface finish matters most on high-strength and fatigue-sensitive alloys.

Machine, Grips, Alignment and Extensometer Classes

  • Force measurement: ISO 7500-1 class 1 or better, that is ± 1 % of reading over the range used, verified at least annually; see our calibration page. A 600 kN frame reading 15 kN for a wire test is usually outside its class 1 range, so a second, smaller load cell is the normal answer.

  • Extensometer: ISO 9513 class 1 or better for ReH, ReL, Rp and Rt; class 2 is acceptable for the properties measured at extensions above 5 % (A, Ag, Agt, At, Ae). Modulus determination is more demanding and class 0.5 is the usual recommendation. The extensometer gauge length Le should sit centrally on the parallel length and be as large as practical, with 50 mm and 80 mm the common choices; it must be at least half of L0 and should not reach into the fillets.

  • Gripping and alignment: the force must be applied as axially as possible. Any straightening preload must not exceed 5 % of the specified or expected yield force and the extension it produces must be corrected for. The force reading is zeroed after the load string is assembled and before the piece is gripped, and not changed during the test. Bending introduced by misaligned grips raises Rp and lowers A; alignment is checked with a strain-gauged piece per ISO 23788 or ASTM E1012, and spherical-seat couplings or alignment adapters correct it.

Step-by-Step Procedure

  • 1. Identify the product form and pick the annex. Sheet under 3 mm → Annex B flat; wire or small bar → Annex C full section; thick sheet, bar, forging or casting → Annex D round or flat; tube → Annex E.

  • 2. Machine or cut the test piece within the annex tolerances, with the fillet radius and parallel length required, and check that the width-to-thickness ratio and Lc ≥ L0 + d0/2 (or + b0/2) rules are met.

  • 3. Measure the cross-section at three or more positions along Lc to ± 2 % or better and average; or confirm that the piece is within the tolerances that allow nominal dimensions.

  • 4. Calculate and mark L0. For proportional pieces L0 = 5.65 √S0 rounded to the nearest 5 mm; for non-proportional pieces 50, 80 or 200 mm. Mark lightly to ± 1 %.

  • 5. Set up the machine. Load cell in its class 1 range for the expected Fm; grips matched to the piece (see the equipment section); extensometer of the required class fitted centrally; force zeroed after assembling the load string.

  • 6. Grip the piece without bending it; apply at most a 5 % preload if a straightening force is needed.

  • 7. Run the test. Method A1: Range 2 (0.00025 s⁻¹) under extensometer control to beyond Rp or the yield plateau, then Range 4 (0.0067 s⁻¹) to fracture, switching to crosshead control if the extensometer must come off. Method A2: the equivalent crosshead speeds from Lc. Method B: the MPa/s rate to just before yield, then crosshead control at a rate that keeps the strain rate at or below 0.008 s⁻¹.

  • 8. After fracture, fit the two halves together on their axis, measure Lu to 0.25 mm and the minimum diameter or width and thickness at the fracture to obtain Su to ± 2 %.

  • 9. Check the fracture position. A is valid without qualification when the fracture is at least L0/3 from the nearest gauge mark; a fracture outside that middle third is acceptable only if A still meets the specified value, otherwise repeat or apply the Annex I subdivision method where agreed.

  • 10. Calculate and report the properties with the rounding rules, the method and rate used, the extensometer class and Le, the gauge length convention (A, A50mm, A11.3) and the test temperature.

Calculations and a Worked Example

The formulas are simple; the care is in S0, L0 and the offset construction. R = F ÷ S0 for every strength. Rp0.2 is found by drawing a line parallel to the elastic slope, offset by 0.2 % of Le on the extension axis, and reading the stress where it cuts the curve; Rt0.5 is read at 0.5 % total extension. A = (Lu − L0) ÷ L0 × 100 and Z = (S0 − Su) ÷ S0 × 100. Agt is the total extension at Fm read from the extensometer and Ag is Agt minus the elastic part Rm ÷ E.

Example: Annex D round, quenched and tempered alloy steel. d0 = 10.00 mm, so S0 = π × 10² ÷ 4 = 78.54 mm². L0 = 5.65 × √78.54 = 5.65 × 8.862 = 50.07 mm, rounded to 50 mm (= 5 d0); Lc = 60 mm (≥ 50 + 5); Le = 50 mm with a class 1 clip-on. Method A1: Range 2 to yield, a gauge extension rate of 50 mm × 0.00025 s⁻¹ = 0.0125 mm/s (0.75 mm/min); then Range 4 at 0.0067 s⁻¹, which in A2 terms is a crosshead speed of 60 × 0.0067 = 0.40 mm/s, or 24 mm/min. Force at 0.2 % plastic offset 47.1 kN; maximum force 55.0 kN; Lu after fitting together 61.5 mm; minimum diameter at fracture 7.20 mm; fracture 18 mm from the nearer gauge mark.

  • Rp0.2 = 47,100 N ÷ 78.54 mm² = 599.7 → 600 MPa.

  • Rm = 55,000 ÷ 78.54 = 700.3 → 700 MPa.

  • A = (61.5 − 50) ÷ 50 × 100 = 23.0 → 23.0 % (reported as A, the k = 5.65 value). The fracture is 18 mm from the mark and L0/3 = 16.7 mm, so the result is valid.

  • Z: Su = π × 7.20² ÷ 4 = 40.72 mm²; Z = (78.54 − 40.72) ÷ 78.54 × 100 = 48.15 → 48 %.

  • Capacity check: Fm of 55 kN on a 100 kN frame is 55 % of capacity and in the class 1 range of a 100 kN cell; the same bar at 20 mm diameter would need 220 kN, a floor-standing machine.

Common Errors and Their Size

  • Rate outside the band. Running Rp0.2 at Range 4 instead of Range 2 raises the proof strength of a rate-sensitive steel by 2 to 5 % and of some aluminium and titanium alloys by more; TENSTAND found spreads above 10 % between the slowest and fastest speeds of the pre-2009 standards. Report the method and rate every time.

  • Method A2 without allowing for compliance. Setting the crosshead at exactly Lc × 0.00025 on a stiff piece in a soft load string gives a true strain rate 20 to 40 % below nominal while the force is rising, which is outside the ± 20 % band at the low end. Aim higher or use Annex F.

  • Class 2 extensometer or crosshead displacement used for Rp. Grip settling and machine compliance show up as apparent strain; an Rp0.2 from crosshead travel is typically several percent low and the modulus can be wrong by a factor of two.

  • Nominal area on an out-of-tolerance piece. 0.1 mm undersize on a 10 mm round is 2 % of area and 2 % of every strength; the tolerance for using nominal dimensions is ± 0.02 mm in that band.

  • Fracture outside the middle third with a result below specification: the elongation is invalid and the test must be repeated; it is not permitted to average it in.

  • Gauge marks that notch the surface, especially on high-strength steels and titanium, which then fail at the mark; use a non-marking tool or a fine scribe within the allowed depth.

  • Preload above 5 % of yield or an un-zeroed force channel after fitting heavy grips, which shifts the whole curve.

  • Wrong edition. Specifications written to EN 10002-1 or ISO 6892:1998 permitted speeds that are not allowed in Method A; a report should name ISO 6892-1:2019.

ISO 6892-1 vs ASTM E8/E8M

The two standards test the same thing and a laboratory equipped for one can run the other, but the numbers differ enough that results are not interchangeable without stating the method. Our ASTM E8 guide covers the ASTM side; the main differences are:

  • Rate control: ISO A1 0.00025 s⁻¹ ± 20 % against ASTM E8 Method B 0.015 ± 0.006 min⁻¹ (the same nominal 0.00025 s⁻¹ but ± 40 %); ISO A2 ± 20 % against ASTM Method C 0.015 ± 0.003 min⁻¹; ISO Method B 2 to 20 or 6 to 60 MPa/s against ASTM Method A 1.15 to 11.5 MPa/s (10,000 to 100,000 psi/min). After yield ISO recommends 0.0067 s⁻¹ (0.4 min⁻¹) where ASTM allows 0.05 to 0.5 min⁻¹.

  • Gauge length: ISO proportional 5.65 √S0 (5 d0) against E8 4 d0 and E8M 5 d0; ISO non-proportional 50, 80 and 200 mm against ASTM 2 in and 8 in. Elongation values are not comparable across k.

  • Cross-section: ISO averages measurements along Lc to ± 2 %; ASTM uses the smallest section with a resolution of 0.001 in on pieces over 0.2 in.

  • Elongation validity: ISO requires the fracture at least L0/3 from a gauge mark (unless A meets specification); ASTM discards a fracture within 25 % of the elongated gauge length of a mark. ISO measures Lu to 0.25 mm, ASTM to 0.01 in.

  • Extensometer and force classes: ISO 9513 class 1 (0.5 for modulus) and ISO 7500-1 class 1 against ASTM E83 class B-2 and E4 ± 1 %; the accuracies are close but the verification procedures differ.

  • Symbols and rounding: Rp0.2, ReH, ReL, Rm, Ag, Agt, A, Z against YS (offset), UYS, LYS, TS, Elu, El and RA; ISO rounds strength to 1 MPa, ASTM to 100 psi, 500 psi or 1 ksi by band.

  • Temperature: ISO 10 to 35 °C against ASTM 10 to 38 °C. Uncertainty: ISO has Annex K on measurement uncertainty, with a rule that it is not combined with the result for conformance; E8 has no such annex.

Specifications such as ASTM A370 for steel products and EN 10025 or EN 10204 certificates call one or the other; dual-certified mills run ISO 6892-1 Method A1 at Range 2 and report both sets of symbols, which is within both standards' tolerances for rate.

Grips, Extensometers and Machines for ISO 6892-1

The annex you test to decides the grip. The common failure modes are slipping under slow Range 1 and 2 control, jaw-initiated fractures on hard or brittle alloys, and bending from grips that do not centre the piece; each has a grip answer.

  • Annex B flats, 0.1 to 3 mm sheet and strip: mechanical wedge grips with serrated faces for steel and most alloys, smooth or diamond-coated faces for soft aluminium and copper that serrations would cut; pneumatic grips for thin foil and strip below about 1 mm where a constant, repeatable clamping force prevents crushing; hydraulic grips from 100 kN up for high-strength sheet. Jaw faces must be longer than the gripped end so that the 87.5 or 140 mm free length is respected.

  • Annex D rounds, 5 to 20 mm: wedge grips with V-jaws for plain ends; bolt and threaded grips for threaded-end pieces (M10 to M20 ends on the 5 to 14 mm pieces, thread engagement at least one diameter), which cannot slip and give the best alignment; shoulder grips with split collets for button-head pieces of hard or brittle material where threads would fail. Threaded and shouldered ends are the standard choice for Rp0.2 on high-strength steels, titanium and nickel alloys and for anything that will later be tested hot under ISO 6892-2.

  • Annex D thick flats and plate, 3 mm up: wedge grips rated for the force, hydraulic from 250 kN; flat serrated faces as wide as the piece. A 40 × 25 mm plate of 550 MPa steel needs 550 kN, so a 600 kN or larger floor-standing frame.

  • Annex C wire and small bar: wrap grips (capstan or bollard) that take the load in friction around a drum and eliminate jaw breaks on high-strength wire; small wedge or vise grips for soft wire with the long L0 + 3 b0 separation.

  • Annex E tubes: full-section tube in wedge grips with snug internal plugs so the walls do not collapse, observing the D0/4 and D0 plug-position limits; strips in flat-faced wedge grips.

Capacity. Fm = Rm × S0, with 20 to 30 % headroom so the cell works in its class 1 range and the grip faces are not at their limit. A 12.5 mm mild-steel round (S0 = 122.7 mm², Rm about 450 MPa) needs 55 kN, so a 100 kN frame; a 10 mm round of 1,000 MPa steel needs 79 kN; the 20 mm round of the same steel needs 314 kN. Our 50 kN and 100 kN benchtop machines cover sheet, wire and small rounds; the 200 kN, 600 kN and 1000 kN floor-standing machines cover plate, large rounds and rebar, and the hydraulic machines go beyond that for heavy sections. All run closed-loop strain control for Method A1 from the extensometer channel.

Extensometers. A clip-on axial extensometer of ISO 9513 class 1 (class 0.5 for modulus work) with a 50 or 80 mm gauge length is the workhorse for Rp0.2 and Agt and can stay on to fracture on ductile steels; for brittle alloys it is removed after yield and the controller switches to crosshead. A video extensometer measures the whole test without touching the piece, which suits thin foil, wire and anything that fractures violently, and our Vector 3D extensometer adds width strain for r-value and Poisson's ratio on sheet. Gauge marks for elongation after fracture are applied with the marking tool; the die press and cutting dies produce Annex B blanks from sheet before finish machining.

ISO 6892-1 FAQ

What is the current edition of ISO 6892-1?

ISO 6892-1:2019, the third edition, published November 2019, confirmed without change in September 2025. It is a minor revision of 2016; the strain-rate Method A (split into A1 and A2 in 2016) and stress-rate Method B are unchanged.

What strain rates does ISO 6892-1 Method A use?

Four ranges, each ± 20 %: 0.00007 s⁻¹, 0.00025 s⁻¹ (recommended for ReH, Rp, Rt), 0.002 s⁻¹ and 0.0067 s⁻¹ (recommended for Rm, A and Z after yield). A1 controls on the extensometer; A2 sets the crosshead speed to Lc × strain rate.

What is Method B in ISO 6892-1?

Stress-rate control in the elastic region: 2 to 20 MPa/s for materials with E below 150 GPa (aluminium, copper, titanium, magnesium) and 6 to 60 MPa/s for E of 150 GPa and above (steel, nickel alloys). Strain rate must stay at or below 0.0025 s⁻¹ while finding Rp and at or below 0.008 s⁻¹ after yield, and the machine must leave stress control before yield.

What is a proportional test piece and what is k = 5.65?

A test piece whose original gauge length is L0 = 5.65 √S0, which equals 5 d0 for a round. Elongation A measured this way is the ISO default; k = 11.3 is the alternative for small sections, and non-proportional gauge lengths of 50, 80 and 200 mm are used for sheet and wire. Elongation on different gauge lengths is not comparable.

What are the standard ISO 6892-1 round specimen sizes?

Annex D: diameters 20, 14, 10 and 5 mm with gauge lengths 100, 70, 50 and 25 mm, parallel lengths of at least 110, 77, 55 and 28 mm, fillet radius at least 0.75 d0 and diameter tolerance ± 0.02 to ± 0.10 mm depending on size.

What are the ISO 6892-1 flat specimen dimensions for sheet?

Annex B, for 0.1 to 3 mm sheet: 12.5 mm wide with L0 50 mm and parallel length 75 mm recommended (free length 87.5 mm); 20 mm wide with L0 80 mm and parallel length 120 mm (free length 140 mm); 25 mm wide with L0 50 mm. Width tolerance ± 0.05 mm on the 12.5 mm piece and ± 0.10 mm on the others; fillet radius at least 20 mm.

Which extensometer class does ISO 6892-1 require?

ISO 9513 class 1 or better for ReH, ReL, Rp and Rt; class 2 is acceptable for A, Ag, Agt, At and Ae measured above 5 % extension. Class 0.5 is the usual recommendation for modulus. Crosshead displacement is not acceptable for yield or proof strength.

How is Rp0.2 determined under ISO 6892-1?

From the force-extension curve with a class 1 extensometer: a line parallel to the elastic slope is offset by 0.2 % of the extensometer gauge length, and Rp0.2 is the stress where it meets the curve. In the worked example above 47.1 kN on 78.54 mm² gives 600 MPa.

When is an elongation result invalid because of the fracture position?

When the fracture is closer than L0/3 to a gauge mark, unless the measured A still meets the specified minimum, in which case it is accepted. Otherwise the test is repeated or the Annex I gauge-length subdivision method is used by agreement. The halves must be fitted together and Lu measured to 0.25 mm.

What is the difference between ISO 6892-1 and ASTM E8/E8M?

Same test, different numbers: ISO uses 5.65 √S0 (5 d0) gauge lengths and 50/80 mm non-proportional pieces, ASTM E8 uses 4 d0 and 2 in; ISO A1 is 0.00025 s⁻¹ ± 20 % against ASTM Method B ± 40 %; ISO Method B is 2 to 60 MPa/s against ASTM Method A 1.15 to 11.5 MPa/s; ISO averages the cross-section, ASTM takes the smallest; ISO invalidates fractures outside the middle third, ASTM outside 25 % of the gauge length. Symbols and rounding differ.

What machine capacity is needed for ISO 6892-1?

Tensile strength times cross-section plus 20 to 30 % headroom: a 12.5 mm mild-steel round needs about 55 kN (100 kN frame); a 10 mm round of 1,000 MPa steel 79 kN; a 20 mm round of the same steel 314 kN; a 40 × 25 mm plate of 550 MPa steel 550 kN (600 kN or larger floor-standing machine).

Which grips are used for ISO 6892-1 test pieces?

Wedge grips with serrated faces for sheet and plate (pneumatic for foil, hydraulic above 100 kN), V-jaw wedge, threaded or button-head shoulder grips for rounds, capstan or wrap grips for wire, and wedge grips with internal plugs for full-section tube. Threaded and shouldered ends give the best alignment for Rp0.2 on high-strength alloys.

Need wedge, hydraulic or threaded grips, an extensometer or a complete machine for ISO 6892-1? Contact Universal Grip with your specimen type, expected tensile strength and machine connection and we will quote the right setup.

bottom of page