top of page

ASTM C1161
Bend Fixture - Ceramics

ASTM C1161 Bend Fixture
Universal Grip ASTM C1161 bend fixture for universal testing machines

Downloads: Standard Test Method for Flexural Strength of Advanced Ceramics at Ambient Temperature - PDF · ASTM C1161 Bend Fixture – Universal Grip Data Sheet (PDF)

Written by Chris Thomas, Universal Grip Company · Reviewed by Frank Bertini · Updated October 17, 2026

ASTM C1161 is the standard test method for flexural strength of advanced ceramics at ambient temperature. A small rectangular bar is loaded in four-point (quarter-point) or three-point bending on hardened rollers until it fractures, and the breaking load is converted to an outer-fibre stress. It is the strength test for monolithic and particulate- or whisker-reinforced ceramics and glasses: alumina, zirconia, silicon nitride, silicon carbide, aluminium nitride and the technical glass-ceramics used in electronics, cutting tools, bearings, armour, dental and medical components. Because ceramics fail from the largest flaw they contain, the method is as much about the specimen (its machining, its edges, its alignment) as about the machine, and the standard is unusually specific about all three.

The pages that rank for this standard today give almost none of the numbers. This guide gives all of them: the A, B and C specimen configurations with spans, bearing diameters and crosshead speeds, the strain-rate target, the difference between semi-articulating and fully articulating fixtures and when each is required, the machining and chamfer recipe, the 10- and 30-specimen rules, the four-point and three-point formulas with a worked example, the fracture-location rule that tells you whether a data set is valid, load-cell sizing, and how C1161 maps to ISO 14704 and EN 843-1. It ends with the fixtures and machines Universal Grip supplies for the test. The current edition is C1161-18, reapproved in 2023.

In this guide: What ASTM C1161 Measures · Specimen Configurations A, B and C · Machining, Edges and Chamfers · The Fixture: Semi-Articulating vs Fully Articulating · Speed, Strain Rate and the Machine · Number of Specimens: 10 for a Mean, 30 for Weibull · Step-by-Step Procedure · Calculation and Worked Example · Common Errors · ASTM C1161 vs ISO 14704 vs EN 843-1 · Equipment for ASTM C1161 · FAQ

What ASTM C1161 Measures

C1161 determines the flexural strength of an advanced ceramic: the maximum tensile stress on the outer fibre of a beam at the moment it breaks, calculated from elementary beam theory. The standard covers four-point loading at the quarter points (the load span is half the support span, so each loading roller sits a quarter of the span in from a support) and three-point loading at mid-span. Four-point is the reference configuration, because the region between the inner rollers is under uniform maximum stress, so a large volume of material and surface is sampled and the result reflects the flaw population rather than the single point under a nose. Three-point results are, in the standard's own words, likely to be much higher, because only the line under the centre roller sees the peak stress and the chance of the worst flaw being there is small.

The method is intended for materials with strengths of about 50 MPa and above and for microstructures whose grain size is no more than about one-fiftieth of the beam depth, so the beam behaves as a continuum. Continuous-fibre ceramic composites are excluded; they go to ASTM C1341. Glass is covered, and for glass and glass-ceramics the companion method ASTM C158 is often cited instead. The elevated-temperature version of this test is ASTM C1211, slow-crack-growth testing is C1368, Weibull statistics are C1239, and fractography of the broken halves is C1322, which the standard strongly recommends. For the general choice between three-point and four-point loading, see our 3-point vs 4-point bend article.

Specimen Configurations A, B and C

C1161 defines three proportional specimen sizes. Configuration B (3 × 4 × 45 mm on a 40 mm span) is the standard specimen and the one nearly all published data use; A is the half-size for limited material or small components, and C is the double-size for coarse-grained or large-flaw materials. The values below are from the standard's tables; the standard also sets the spans to within ± 0.10 mm and requires the three-point centre bearing to be centred to the same tolerance.

  • Configuration A: width b = 2.0 mm, depth d = 1.5 mm, length at least 25 mm; support span 20 mm, load span 10 mm; bearing (roller) diameter 2.0 to 2.5 mm; crosshead speed 0.2 mm/min.

  • Configuration B: b = 4.0 mm, d = 3.0 mm, length at least 45 mm; support span 40 mm, load span 20 mm; bearing diameter 4.5 mm; crosshead speed 0.5 mm/min.

  • Configuration C: b = 8.0 mm, d = 6.0 mm, length at least 90 mm; support span 80 mm, load span 40 mm; bearing diameter 9.0 mm; crosshead speed 1.0 mm/min.

Two proportions run through the table: the bearing diameter is about 1.5 times the specimen depth, and the bearing length must be at least three times the specimen width so the roller never loads an edge. The same crosshead speeds are used for three-point loading of each configuration. Note the orientation: the beam is loaded on its 4 mm face (B), so the 3 mm dimension is the depth, d, in the stress formula. Loading a B specimen on edge changes the stress by a factor of 4/3 and the result is not a C1161 value.

Dimensional tolerances. Cross-section ± 0.13 mm for B and C and ± 0.05 mm for A; the two longitudinal faces that touch the rollers must be parallel within 0.015 mm (A and B) or 0.03 mm (C). The end faces are not critical. Width and depth are measured near the fracture after the test with a micrometer resolving 0.002 mm, and reported to the nearest few thousandths of a millimetre. The depth enters the stress as d², so a 0.03 mm error on a 3 mm beam is a 2 % error in strength before anything else has gone wrong.

Which configuration to use. B unless there is a reason not to. Use C when the grain size or the expected flaw size is too large for a 3 mm beam, or when the specification calls for it; use A only when material is scarce, and never for design data: the standard's own error analysis puts the total uncertainty of configuration A above 5 %, against about 3 % for B and C, mostly because the chamfers are a larger fraction of a 1.5 mm beam.

Machining, Edges and Chamfers

Most ceramic flexure specimens break from a machining flaw, so the standard specifies the grinding procedure rather than leaving it to the shop. Unless another finishing procedure is agreed and reported, the standard procedure is:

  • Grind all four long faces longitudinally (parallel to the beam axis) with a diamond wheel under flood coolant. No Blanchard or rotary grinding, because transverse grinding marks become transverse cracks that open under the tensile stress.

  • Coarse grinding: 150-grit or finer diamond, no more than 0.03 mm per pass, down to within 0.060 mm of final dimension.

  • Intermediate grinding: 240- to 320-grit, no more than 0.006 mm per pass, down to within 0.020 mm of final dimension.

  • Finish grinding: 400- to 600-grit, no more than 0.006 mm per pass, removing at least 0.020 mm from each face, so that the damage from the coarser steps is removed.

  • Wheel surface speed at least 25 m/s and table speed no more than 0.25 m/s, with no deep striations or grooves visible on the finished face.

  • Edges: either a 45° chamfer 0.12 ± 0.03 mm wide or a radius of 0.15 ± 0.05 mm on the four long edges, ground longitudinally in the same way; larger chamfers require the correction in the standard's annex.

  • If one face is scratched or otherwise damaged, place it on the compression side (up, under the loading rollers), never on the tension side.

The chamfers matter because an unchamfered edge is the sharpest flaw on the specimen and because the chamfer removes load-bearing area from the extreme fibre. The standard's error analysis estimates the chamfer contribution at about 1 % for configuration B when the chamfer is held to the specified size, and about 4 % for A. Universal Grip can supply specimens ground to this recipe through our testing services, or advise on a grinding vendor.

The Fixture: Semi-Articulating vs Fully Articulating

The fixture is where C1161 differs most from a plastics bend test. The rollers must be free to roll (on needle bearings or in the floating mounts drawn in the standard) so that friction does not add a longitudinal force to the beam, except the centre bearing of a three-point fixture, which is fixed. The fixture must be stiffer than the specimen, and the bearings must be harder and stronger than the material being tested: hardened steel of at least 40 HRC or a yield strength of 1,240 MPa or more, or a ceramic bearing with a modulus of 200 to 400 GPa and a strength of at least 275 MPa, which covers specimen strengths up to about 1,400 MPa.

Semi-articulating fixtures

In a semi-articulating four-point fixture the two inner rollers are parallel to each other within 0.015 mm and pivot together as a pair, and the two outer rollers do the same. This accommodates a specimen that is slightly wedge-shaped along its length but assumes the loaded faces are flat and parallel across the width. It is the fixture for machined specimens that meet the 0.015 mm parallelism requirement.

Fully articulating fixtures

In a fully articulating fixture each roller (or each pair, plus the loading head) can tilt independently about the beam axis as well, so that all four rollers seat on a specimen that is twisted or whose faces are not parallel. The standard requires a fully articulating fixture whenever the specimen's parallelism cannot be met, which in practice means as-fired, as-sintered, heat-treated, oxidised, coated or otherwise unmachined surfaces, and permits it for machined specimens as well. Most labs that test anything other than fully ground bars simply standardise on a fully articulating fixture, because a fixture that cannot seat on the specimen loads one edge first, and the result is an edge failure that reads low. Three-point fixtures come in the same two forms.

What misalignment looks like in the data

The standard gives a usable diagnostic. Breaks outside the inner (load) span are acceptable if they are infrequent, but if roughly 10 % or more of a set breaks outside the inner span, or if the primary fractures occur within 0.5 mm of an inner roller, the fixture is misaligned or not articulating properly and the set should be rejected until it is fixed. Check that the spans are within ± 0.10 mm, that the rollers roll freely, that the loading head is centred, and that nothing in the fixture is touching the specimen except the four rollers.

Universal Grip's articulating bend fixtures are built to C1161 in 2.5 kN and 10 kN capacities with interchangeable roller sets for the A, B and C configurations, adjustable support and load spans with scales, quarter-point and third-point load spacing, and fully articulating heads for as-fired specimens; high-temperature versions serve C1211. They mount on any frame through our adapters.

Speed, Strain Rate and the Machine

The crosshead speeds in the configuration table are chosen to give an outer-fibre strain rate of about 1 × 10⁻⁴ per second, which produces failure in roughly 3 to 30 seconds for most ceramics. The strain rate for a given setup is ε̇ = 6 d s ÷ L², with d the depth, s the crosshead speed and L the support span. For configuration B at 0.5 mm/min (0.00833 mm/s): ε̇ = 6 × 3 × 0.00833 ÷ 40² = 9.4 × 10⁻⁵ s⁻¹, which is the target. If the material is suspected of slow crack growth (glass, some oxides in humid air), faster rates reduce the effect and C1368 characterises it directly; whatever rate is used, record it and the time to failure.

  • Load accuracy: the machine is verified per ASTM E4 and must be accurate in the range used; loads in this test are small, so the load cell is the thing to get right.

  • Load-cell sizing: P = 4 S b d² ÷ (3 L) for four-point. A configuration B bar at 500 MPa breaks at about 600 N and at 1,000 MPa at 1.2 kN; a C bar at 1,000 MPa at 4.8 kN; an A bar at 1,000 MPa at 0.3 kN. A 1 to 5 kN load cell on a 2.5 to 10 kN frame covers the whole method; a 100 kN frame with a 100 kN cell is the wrong tool, because a 600 N break is below 1 % of its range.

  • Preload: no more than 25 % of the expected mean strength while seating the specimen and the fixture.

  • Data rate: ceramics fail without warning; sample at 100 Hz or faster so the peak is captured, and use the machine's break detection to stop the crosshead.

  • Environment: test at ambient temperature and record temperature and relative humidity (humidity affects slow crack growth in many oxides); for elevated-temperature flexure use C1211 in an environmental chamber or furnace.

A 5 kN benchtop universal testing machine with a 1 kN or 2 kN cell and 0.2 to 1.0 mm/min speeds is the usual C1161 system; the benchtop range also runs D790 and D6272 flexure on plastics, D7264 on composites and C1499 equibiaxial flexure on ceramic discs with a change of fixture.

Number of Specimens: 10 for a Mean, 30 for Weibull

Ceramic strength is statistical. The standard requires at least 10 specimens to determine a mean strength and at least 30 when the Weibull modulus is to be estimated per C1239, and more when the fractography shows more than one flaw population. The reason is the scatter: a good structural ceramic has a Weibull modulus of 10 to 20, which means strengths within a single batch spread by 10 to 20 % around the mean, and the mean itself is only known to a few percent with ten specimens. Report the mean and standard deviation, the number of specimens, the configuration, and the notation the standard defines, for example S(4,40) for four-point loading on a 40 mm span, so that nobody compares a three-point number with a four-point one. Fractography per C1322 on at least a sample of the broken halves tells you whether the failures started at surface machining flaws, at pores or inclusions, or at the edges, and that is the information that improves the material.

Step-by-Step Procedure

  • 1. Inspect and measure. Check the chamfers and surfaces under magnification; mark the face that will be in tension; measure width and depth to 0.002 mm at the centre (and again at the fracture afterwards).

  • 2. Set the fixture. Select the roller set for the configuration (4.5 mm for B), set the support span to 40.00 ± 0.10 mm and the load span to 20.00 ± 0.10 mm, confirm the rollers turn freely and the loading head is centred and articulating.

  • 3. Seat the specimen centred on the supports with the tension face down, the long axis perpendicular to the rollers and equal overhang at both ends; for a semi-articulating fixture confirm the loaded faces are the parallel pair.

  • 4. Preload to no more than 25 % of the expected break load to seat everything, then zero.

  • 5. Run at the configuration speed (0.5 mm/min for B) until fracture, recording load at a high data rate. Note the time to failure; it should be in the 3 to 30 s window.

  • 6. Record the break load and the fracture location relative to the inner rollers; keep the pieces for fractography.

  • 7. Calculate the strength for each specimen, then the mean and standard deviation of at least 10 (30 for Weibull). Apply the fracture-location rule to the whole set.

Calculation and Worked Example

  • Four-point (quarter-point): S = 3 P L ÷ (4 b d²), where P is the break load, L the support (outer) span, b the width and d the depth.

  • Three-point: S = 3 P L ÷ (2 b d²).

  • In newtons and millimetres the result is in MPa; report to three significant figures.

Worked example (configuration B, four-point). A sintered alumina bar measures b = 4.00 mm and d = 3.00 mm at the fracture, tested on L = 40.0 mm with a 20.0 mm load span at 0.5 mm/min. It breaks at P = 600 N, 11 s into the test, with the fracture 6 mm inside the inner span. S = 3 × 600 × 40.0 ÷ (4 × 4.00 × 3.00²) = 72,000 ÷ 144 = 500 MPa. The strain rate was 9.4 × 10⁻⁵ s⁻¹ (on target) and the fracture location is valid. Ten bars give 512, 488, 500, 531, 476, 505, 519, 494, 483 and 522 MPa: mean 503 MPa, standard deviation 18 MPa (CV 3.6 %), reported as S(4,40) = 503 ± 18 MPa, n = 10. The same bar in three-point loading at the same break load would report 1,000 MPa, which is why the loading configuration is part of the result's name.

The Mecmesin and several other pages print 3PL/(2bd²) as the four-point formula; it is the three-point one. Using it on four-point data doubles every strength value.

Common Errors

  • Using the three-point formula for four-point data (or the reverse).

  • Loading the B specimen on its 3 mm face instead of its 4 mm face, so that d is wrong.

  • Transverse (Blanchard) grinding, coarse finishing or unchamfered edges, giving low strengths that are machining results, not material results.

  • A semi-articulating fixture on as-fired or non-parallel specimens; edge failures and breaks under an inner roller follow.

  • Rollers that do not roll, worn flats on rollers, or steel rollers that are too soft for a high-strength ceramic.

  • Spans not set to ± 0.10 mm, or the roller set from the wrong configuration (5 mm radius plastics noses on a 3 mm ceramic bar).

  • Load cell too large (a 600 N break on a 50 or 100 kN cell).

  • Fewer than 10 specimens for a mean, or a Weibull modulus quoted from 10.

  • Reporting a mean without the configuration and span, or comparing three-point and four-point values.

  • No fractography, so a change in strength cannot be traced to its flaw type.

ASTM C1161 vs ISO 14704 vs EN 843-1

  • ISO 14704: the international method for fine (advanced technical) ceramics. Four-point loading on a 40 mm span at the quarter points is the recommended configuration, but a 30 mm span with third-point loading is also allowed, and three-point loading on 30 or 40 mm; the standard specimen is the same 3 × 4 mm cross-section with a cross-section tolerance of ± 0.2 mm (looser than C1161's ± 0.13 mm) and the same 0.015 mm parallelism; the same chamfer (0.12 ± 0.03 mm or R 0.15 ± 0.05 mm); rollers 4.5 to 5 mm; steel rollers of at least 40 HRC for specimen strengths up to about 1,400 MPa and 46 HRC up to about 2,000 MPa; 0.5 mm/min adjusted to give 3 to 30 s to fracture; at least 10 specimens, 30 for Weibull; a class 1 machine per ISO 7500-1. It has no half-size or double-size configurations.

  • EN 843-1: the European method, closely aligned with ISO 14704, defines two geometries by span, A with a 20 mm span and B with a 40 mm span, for both three- and four-point loading.

  • Comparability: a B-configuration C1161 test and an ISO 14704 four-point test on a 40 mm quarter-point span are the same test in all but tolerances and can usually be reported together; a third-point 30 mm ISO result or an EN 843-1 20 mm result is not the same test, because the stressed volume differs and the Weibull size effect changes the measured strength.

If a specification cites more than one of these, a fixture with interchangeable roller sets and spans that reach 20, 30, 40 and 80 mm covers all three.

Equipment for ASTM C1161

  • Fixtures: fully articulating four-point and three-point bend fixtures in 2.5 kN and 10 kN capacities with 2.25, 4.5 and 9 mm roller sets (and 4.5 to 5 mm sets for ISO 14704), scaled span adjustment from 10 to 80 mm, quarter-point and third-point loading heads, and high-temperature versions for C1211, all in your machine's adapter.

  • Frames: a 5 kN benchtop universal testing machine with a 1 or 2 kN load cell and 0.2 to 1.0 mm/min speeds; the benchtop range to 10 kN for configuration C and for sharing the frame with plastics and composite flexure.

  • Deflection: optional; a video extensometer gives mid-span deflection and modulus without touching the specimen.

  • Environment: environmental chambers and furnaces for C1211 elevated-temperature flexure.

  • Related fixtures on the same frame: C1499 ring-on-ring equibiaxial flexure for discs, C158 glass flexure, D790 and D6272 for plastics, D7264 for composites, and C1341 for ceramic matrix composites.

Our earlier article on ASTM C1161 flexural strength of advanced ceramics covers the fixture in more detail, and the 3-point vs 4-point bend article explains the choice of loading.

ASTM C1161 FAQ

What is ASTM C1161?

The ASTM standard test method for flexural strength of advanced ceramics at ambient temperature. A rectangular bar is broken in four-point (quarter-point) or three-point bending on rollers, and the break load is converted to the outer-fibre stress at fracture.

What are the ASTM C1161 specimen sizes?

Configuration A: 1.5 × 2 × 25 mm on a 20 mm span; configuration B: 3 × 4 × 45 mm on a 40 mm span; configuration C: 6 × 8 × 90 mm on an 80 mm span. The load span is half the support span in four-point loading. B is the standard specimen.

What is the test speed for ASTM C1161?

0.2 mm/min for configuration A, 0.5 mm/min for B and 1.0 mm/min for C, chosen to give an outer-fibre strain rate of about 1 × 10⁻⁴ per second and failure in roughly 3 to 30 seconds.

What roller diameters does ASTM C1161 use?

About 1.5 times the specimen depth: 2.0 to 2.5 mm for configuration A, 4.5 mm for B and 9.0 mm for C, with the roller length at least three times the specimen width. Rollers must roll freely and be hardened steel (40 HRC or more) or ceramic.

How is flexural strength calculated in ASTM C1161?

Four-point: S = 3PL/(4bd²); three-point: S = 3PL/(2bd²), where P is the break load, L the support span, b the width and d the depth. For a B bar breaking at 600 N on a 40 mm span, S = 3 × 600 × 40 / (4 × 4 × 9) = 500 MPa.

When is a fully articulating fixture required?

Whenever the specimen's loaded faces cannot be made parallel within 0.015 mm (0.03 mm for configuration C), which covers as-fired, heat-treated, oxidised and coated specimens. It is permitted for machined specimens too; a semi-articulating fixture is only adequate for flat, parallel, ground bars.

How many specimens does ASTM C1161 require?

At least 10 for a mean strength and at least 30 when a Weibull modulus is to be determined per ASTM C1239, with more if fractography shows more than one flaw population.

What is the difference between three-point and four-point results?

Four-point loading stresses the whole region between the inner rollers uniformly, so it samples more material and gives a lower, more representative strength; three-point loading stresses only the line under the centre roller and gives values the standard describes as likely to be much higher. Report the configuration and span with every result, for example S(4,40).

What are the ASTM C1161 edge and machining requirements?

Longitudinal diamond grinding in coarse (150 grit), intermediate (240–320 grit) and finish (400–600 grit) steps with limited depth per pass, and a 45° chamfer of 0.12 ± 0.03 mm or a radius of 0.15 ± 0.05 mm on the long edges. Transverse grinding is not allowed.

When should a set of ASTM C1161 results be rejected?

If about 10 % or more of the specimens break outside the inner load span, or if the primary fractures occur within 0.5 mm of an inner roller, the fixture is misaligned or not articulating and the set is rejected until the fixture is corrected.

What load cell do I need for ASTM C1161?

Loads are small: a configuration B bar at 500 MPa breaks at about 600 N and a C bar at 1,000 MPa at about 4.8 kN. A 1 to 5 kN load cell on a 2.5 to 10 kN frame covers all three configurations.

What is the difference between ASTM C1161 and ISO 14704?

Both use 3 × 4 mm bars on a 40 mm quarter-point span at 0.5 mm/min with 10 or 30 specimens, but ISO 14704 also allows a 30 mm third-point span, uses a ± 0.2 mm cross-section tolerance instead of ± 0.13 mm, specifies roller hardness by specimen strength, and has no A or C sizes. EN 843-1 defines 20 mm and 40 mm span geometries.

Need a fixture, grips or a complete system for ASTM C1161? Contact Universal Grip with your specimen details, capacity and machine connection and we will quote the right setup.

ISO equivalent: see our ISO 14704 ceramic flexural strength guide.

bottom of page