ASTM D790 - Plastics Flexural Strength Testing Equipment


Written by Chris Thomas, Universal Grip Company · Reviewed by Frank Bertini · Updated October 17, 2026
ASTM D790 is the standard test method for flexural properties of unreinforced and reinforced plastics and electrical insulating materials. A rectangular bar rests on two supports and is loaded at mid-span by a loading nose until it breaks or until the outer-fibre strain reaches 5 %; from the load and deflection come flexural strength, flexural stress at a given strain and flexural modulus. It is the most-run test on molded plastics after tensile per D638, because it uses the same molded bar as Izod and HDT, because it is quick, and because stiffness in bending is what most plastic parts are designed for.
This guide gives the specimen and span rules for every thickness band, the fixture dimensions, the two procedures and how to calculate the crosshead speed from the strain rate, the 5 % strain limit and why it is a strain and not a deflection, all four equations with a fully worked example, the three modulus definitions, compliance correction and the Type 1 and Type 2 deflection measurements, and a comparison with ISO 178, D6272 and D7264. It is current to D790-25 and ends with the fixtures and machines Universal Grip supplies for the test.
In this guide: What ASTM D790 Measures, and the 5 % Rule · Specimen Dimensions and Span by Thickness · The Fixture: Noses, Supports, Span and Alignment · Procedures A and B, and the Crosshead Speed Formula · Step-by-Step Procedure · Equations and a Complete Worked Example · ASTM D790 vs ISO 178, D6272 and D7264 · Common Errors · Equipment for ASTM D790 · FAQ
What ASTM D790 Measures, and the 5 % Rule
The method applies only to materials that break or yield within 5.0 % outer-fibre strain in three-point bending. If a specimen neither breaks nor yields by 5 % strain, D790 reports the flexural stress at 5 % strain and the modulus, but the proper method for such ductile materials is ASTM D6272, the four-point bend test, which spreads the maximum stress over the region between the loading noses. Several equipment pages describe the limit as 5 % deflection; it is 5 % strain, and on the standard bar the deflection at 5 % strain is 6.8 mm, a different number (the conversion is given below).
The properties are:
-
Flexural strength: the maximum flexural stress at or before break, for materials that break within 5 % strain.
-
Flexural yield strength: the stress at the first maximum of the load-deflection curve, for materials that yield within 5 % strain.
-
Flexural stress at 5 % strain (or another specified strain) for materials that do neither.
-
Flexural modulus: the ratio of stress to strain in the initial linear portion, as a tangent, secant or chord modulus.
Because the stress is calculated from elementary beam theory on the assumption of linear elastic behaviour and small deflections, flexural strength is an apparent value that depends on specimen size and span; it is used for comparison and specification, not as a design stress. The 3-point vs 4-point bend article explains when each geometry applies, and our bend testing applications page covers flexure across materials.
Specimen Dimensions and Span by Thickness
The specimen is a rectangular bar; the span-to-depth ratio, not the absolute size, is what the standard controls, and it is 16:1 (± 1) for all but a few cases.
-
Molded materials and sheet over 3.2 mm (1/8 in): the standard bar is 127 mm long × 12.7 mm wide × 3.2 mm thick (5 × 0.5 × 0.125 in), tested on a span of 16 × 3.2 = 51.2 mm (round to 51 mm is common; set to the measured depth). Thicker bars keep the 16:1 ratio and the length is span plus at least 10 % overhang at each end.
-
Sheet 1.6 to 3.2 mm (1/16 to 1/8 in): width 12.7 mm, span 16 × thickness, length at least span plus 20 %.
-
Sheet under 1.6 mm (1/16 in): specimen 50.8 mm long × 12.7 mm wide, tested flatwise on a 25.4 mm (1 in) span.
-
Laminated thermosetting materials and electrical insulating sheet: machined to 12.7 mm or 25.4 mm thick as specified, tested at 16:1.
-
Composites and other materials with low shear strength relative to tensile strength: span-to-depth ratios of 32:1 or 40:1 so that the bar fails in bending rather than interlaminar shear; for modulus of highly anisotropic laminates the standard's note recommends 60:1. For fibre-reinforced laminates the dedicated method is ASTM D7264, which uses 32:1 as standard.
Measurement. Measure width and depth to 0.03 mm (0.001 in) with a micrometer per ASTM D5947, three readings along the span region, and use the average. Depth enters the stress as d² and the modulus as d³: a 0.1 mm error on a 3.2 mm bar is a 6 % error in strength and a 9 % error in modulus, more than the whole precision budget of the test. Measure every specimen, not one per batch.
Preparation. Molded bars should be tested with the same face up each time (gate orientation and skin matter); machined specimens need smooth, parallel edges without saw marks, because a notch on the tension face starts the fracture. Condition per ASTM D618 (23 ± 2 °C, 50 ± 10 % RH for at least 40 hours) unless the material specification says otherwise, and test at least five specimens per condition; for anisotropic sheet, test five in each direction.
The Fixture: Noses, Supports, Span and Alignment
-
Loading nose and supports: cylindrical surfaces with a radius of 5.0 ± 0.1 mm, longer than the specimen width, and hard enough not to deform. Larger radii may be used on thick specimens to prevent indentation, and must be reported; smaller radii are not permitted because they indent the specimen and raise the apparent stress.
-
Span: set to 16 × the measured depth (or the ratio specified) within ± 1 %; a micrometer slide or scale and a gauge block check are the standard means. The supports must be parallel to each other and to the nose within a small fraction of a degree, or one edge of the bar loads first.
-
Nose position: centred between the supports to the same tolerance; an off-centre nose changes the moment arm and the stress calculation.
-
Specimen placement: centred on the supports, long axis perpendicular to the cylinders, with the overhang equal at each end.
-
Deflection measurement (Type 1 vs Type 2): Type 1 uses crosshead displacement; Type 2 uses a deflectometer or extensometer that measures the specimen's mid-span deflection directly. Type 1 is permitted provided the apparatus deflection (frame, load cell, fixture) is less than 1 % of the specimen deflection or is corrected for; D790-17 added Appendix X1 with a compliance-correction procedure and tightened the verification classes (Type 1 to Class B, Type 2 to Class B-2). For modulus on stiff materials, Type 2 or a corrected Type 1 is the only way to get an accurate value.
Universal Grip's three-point bend fixtures have 5 mm radius hardened noses and supports (with 10 mm and other radii available), a micrometer span adjustment with scale, a centring stop and mounting in your machine's adapter; the same base converts to four-point loading for D6272 and to the D7264 and C1161 configurations. A deflectometer that reads the specimen's underside at mid-span is available for Type 2 measurement.
Procedures A and B, and the Crosshead Speed Formula
D790 controls the rate of straining of the outer fibre, Z, not the crosshead speed directly, so the crosshead speed changes with span and depth. The two procedures are:
-
Procedure A: Z = 0.01 mm/mm/min. Used for modulus determination and for materials that break at small deflections. The default procedure.
-
Procedure B: Z = 0.10 mm/mm/min. Used for materials that do not break or yield in Procedure A within the 5 % strain limit; also permitted for strength when the specification allows. Modulus should still come from Procedure A.
The crosshead speed is R = Z L² ÷ (6 d), with L the span and d the depth in millimetres. For the standard bar (d = 3.2 mm, L = 51.2 mm): Procedure A gives R = 0.01 × 51.2² ÷ (6 × 3.2) = 26.2 ÷ 19.2 = 1.37 mm/min (most software and the equipment vendors round this to 1.4 mm/min); Procedure B gives 13.7 mm/min. For a 6.4 mm bar on a 102 mm span, Procedure A is 2.7 mm/min; for a 1.6 mm sheet specimen on a 25.6 mm span it is 0.68 mm/min. Set the speed per specimen batch from the measured depth, and record it. Note that at the standard size D790 Procedure A runs slower than ISO 178's 2 mm/min, not faster as is sometimes stated.
Deflection at 5 % strain. The test is stopped at break or at 5 % outer-fibre strain, whichever comes first. The deflection corresponding to a given strain is D = ε L² ÷ (6 d), so for the standard bar the 5 % limit is at D = 0.05 × 51.2² ÷ (6 × 3.2) = 6.8 mm of mid-span deflection. Use that value as the machine's end-of-test limit; a fixed deflection limit that is not recalculated for a different depth or span stops the test at the wrong strain.
Step-by-Step Procedure
-
1. Measure width and depth of each specimen (three readings, averaged) with a D5947 micrometer and record the conditioning.
-
2. Set the span to 16 × the measured depth (or the specified ratio), check it, and centre the loading nose.
-
3. Calculate the crosshead speed R = ZL²/6d for the procedure chosen and the 5 % strain deflection limit D = 0.05L²/6d.
-
4. Place the specimen centred on the supports with the designated face in tension (down), overhang equal at both ends.
-
5. Bring the nose into contact with a small preload just enough to seat the specimen (a few tenths of a newton on thin bars, a few newtons on thick ones), zero load and deflection.
-
6. Run at R, recording load and deflection (crosshead with compliance correction, or deflectometer) until break or until the deflection reaches the 5 % strain limit.
-
7. Record the maximum load (or the load at the first maximum for yielding materials), the load at 5 % strain if reached, and the initial slope of the load-deflection curve for modulus.
-
8. Calculate flexural strength or yield strength, stress at 5 % strain if relevant, flexural strain at break, and modulus for each of at least five specimens, and report mean and standard deviation.
Equations and a Complete Worked Example
-
Flexural stress: σ f = 3 P L ÷ (2 b d²), where P is the load, L the span, b the width and d the depth. In newtons and millimetres the result is MPa.
-
Large-deflection correction (for spans over 16:1 or deflections beyond about 10 % of span): σ f = (3 P L ÷ 2 b d²) × [1 + 6 (D ÷ L)² − 4 (d ÷ L)(D ÷ L)], where D is the mid-span deflection.
-
Flexural strain: ε f = 6 D d ÷ L².
-
Flexural modulus (tangent): E B = L³ m ÷ (4 b d³), where m is the slope of the tangent to the initial straight-line portion of the load-deflection curve, in N/mm.
-
Crosshead speed: R = Z L² ÷ (6 d).
Worked example. A molded polycarbonate bar measures b = 12.70 mm and d = 3.18 mm. Span L = 16 × 3.18 = 50.9 mm. Procedure A speed R = 0.01 × 50.9² ÷ (6 × 3.18) = 1.36 mm/min; the 5 % strain limit is D = 0.05 × 50.9² ÷ (6 × 3.18) = 6.79 mm. The bar yields at a first maximum of P = 118 N at a deflection of 5.9 mm and does not break before the 6.79 mm limit. Flexural yield strength σ f = 3 × 118 × 50.9 ÷ (2 × 12.70 × 3.18²) = 18,019 ÷ 256.9 = 70.1 MPa. Strain at yield ε f = 6 × 5.9 × 3.18 ÷ 50.9² = 112.6 ÷ 2,591 = 0.0435, or 4.3 %, inside the 5 % limit, so the result is a valid flexural yield strength. The initial slope of the corrected load-deflection curve is m = 46.0 N/mm, so E B = 50.9³ × 46.0 ÷ (4 × 12.70 × 3.18³) = 131,872 × 46.0 ÷ (50.8 × 32.16) = 6,066,100 ÷ 1,633.7 = 3,713 MPa, reported as 3.7 GPa. Five specimens averaging 69.8 MPa and 3.68 GPa with standard deviations of 0.9 MPa and 0.08 GPa are reported with the procedure (A), span-to-depth ratio (16:1), speed (1.36 mm/min) and deflection measurement type.
Modulus: tangent, secant and chord
D790-25 defines three ways to report modulus and requires that the one used be stated. The tangent modulus is the slope of the steepest initial straight portion after toe compensation, the traditional D790 value. The secant modulus is the slope of a line from the origin to a specified point on the curve, used for materials with no linear region. The chord modulus is the slope between two specified strain points, which is the ISO 178 approach (0.05 % to 0.25 % strain). Toe compensation removes the initial curved region caused by seating and fixture take-up; software does this by extrapolating the linear region back to zero load and shifting the deflection origin. On stiff materials the modulus is where compliance matters most: an uncorrected Type 1 (crosshead) measurement on a 10 GPa composite can read 20 % low because the frame and load cell deflection is counted as specimen deflection.
ASTM D790 vs ISO 178, D6272 and D7264
-
ISO 178: specimen 80 × 10 × 4 mm on a 64 mm span (16:1), loading nose radius 5 mm and support radius 5 mm (2 mm for thin specimens), speed 2 mm/min for modulus (then up to 100 mm/min to break on some materials), modulus as the chord between 0.05 % and 0.25 % strain, and the test continues to break or reports stress at 3.5 % strain. Results are similar to D790 but not comparable: the bar is thicker, the rate differs and the modulus definition differs.
-
ASTM D6272 (four-point): two loading noses at the third or quarter points, so the region between them sees uniform maximum stress; required for materials that do not break or yield within 5 % strain in D790, and with its own stress and modulus equations.
-
ASTM D7264: the flexure method for polymer matrix composites, with a standard 32:1 span, three- or four-point loading and its own specimen proportions; use it rather than D790 for laminates.
-
ASTM C1161: flexural strength of advanced ceramics, with A, B and C specimen sizes and fully articulating fixtures.
Common Errors
-
Span set from nominal thickness instead of measured depth, or not reset between batches.
-
Loading nose or support radii other than 5 mm (older fixtures often have 3.2 mm or 6.4 mm radii), or indentation on soft materials not addressed with larger radii.
-
Crosshead speed fixed (for example 1.3 mm/min for everything) instead of calculated from R = ZL²/6d.
-
Stopping at 5 % deflection or 5 mm, instead of the deflection corresponding to 5 % strain.
-
Reporting a flexural strength for a material that neither broke nor yielded within 5 % strain.
-
Modulus from uncorrected crosshead displacement on stiff materials.
-
Depth measured once, or with a vernier instead of a micrometer.
-
Specimen face orientation not controlled, or saw marks on the tension face.
-
Comparing D790 and ISO 178 values as if equivalent.
-
Citing an obsolete edition; D790-25 is current and the modulus definitions and verification classes have changed since D790-10.
Equipment for ASTM D790
-
Fixtures: three-point bend fixtures with 5 mm radius noses and supports, micrometer span adjustment and centring stop, convertible to four-point for D6272 and to D7264 and C1161 configurations; a small bend fixture for thin sheet on the 25.4 mm span; all in your machine's adapter.
-
Deflection measurement: a mid-span deflectometer for Type 2 measurement, or a video extensometer tracking the specimen's deflection directly.
-
Frames: flexural loads on plastics are usually under 1 kN and almost always under 5 kN, so a 2.5 kN or 5 kN benchtop universal testing machine with a 500 N to 2 kN load cell and speeds from 0.5 to 50 mm/min covers the method; a Mark-10 tensile tester with a bend fixture handles thin sheet and soft materials. Stiff glass-filled bars and thick laminates use a 10 kN or larger frame.
-
Environment: environmental chambers for flexural properties at service temperature.
-
Same frame: ASTM D638 tensile, D695 compression and D256 Izod on a separate pendulum round out the molded-bar property set.
See the plastics testing page for complete flexural and tensile systems, and the 3-point vs 4-point article for choosing between D790, D6272 and C1161.
ASTM D790 FAQ
What is ASTM D790?
The ASTM standard test method for flexural properties of plastics and electrical insulating materials. A rectangular bar is loaded in three-point bending on a 16:1 span until it breaks or reaches 5 % outer-fibre strain, giving flexural strength, flexural stress at a given strain and flexural modulus.
What is the standard ASTM D790 specimen size?
A molded bar 127 mm long, 12.7 mm wide and 3.2 mm thick (5 × 0.5 × 0.125 in), tested on a span of 16 times the depth, about 51 mm. Sheet under 1.6 mm uses a 50.8 × 12.7 mm specimen on a 25.4 mm span.
What is the span-to-depth ratio in ASTM D790?
16:1 (± 1) for most materials. Ratios of 32:1 or 40:1 are used for composites and other materials with low shear strength, and 60:1 is recommended for modulus of highly anisotropic laminates.
What is the test speed for ASTM D790?
It is calculated from the strain rate: R = ZL²/(6d), with Z = 0.01 mm/mm/min for Procedure A and 0.10 for Procedure B. For the standard 3.2 mm bar on a 51.2 mm span that is 1.37 mm/min (Procedure A) or 13.7 mm/min (Procedure B).
What is the difference between Procedure A and Procedure B?
Procedure A strains the outer fibre at 0.01 mm/mm/min and is used for modulus and for materials that break at small deflection. Procedure B strains at 0.10 mm/mm/min and is used for materials that do not break or yield within 5 % strain under Procedure A.
What does the 5 % strain limit in ASTM D790 mean?
The method applies only to materials that break or yield by 5 % outer-fibre strain. For the standard bar that corresponds to a mid-span deflection of 6.8 mm (D = 0.05L²/6d). Materials that pass 5 % without breaking or yielding should be tested by ASTM D6272.
How is flexural strength calculated in ASTM D790?
σ = 3PL/(2bd²), where P is the maximum load (or the load at yield), L the span, b the width and d the depth. In newtons and millimetres the result is in MPa.
How is flexural modulus calculated in ASTM D790?
E = L³m/(4bd³), where m is the slope of the initial linear portion of the load-deflection curve after toe compensation. D790-25 allows tangent, secant or chord modulus and requires that the type be reported.
What are the loading nose and support radii for ASTM D790?
5.0 ± 0.1 mm, with the cylinders longer than the specimen width. Larger radii may be used on thick specimens to avoid indentation and must be reported.
What is the difference between ASTM D790 and ISO 178?
ISO 178 uses an 80 × 10 × 4 mm bar on a 64 mm span at 2 mm/min, reports modulus as a chord between 0.05 and 0.25 % strain, and continues to break or 3.5 % strain. D790 uses a 3.2 mm bar at a strain-rate-based speed (1.37 mm/min) with a 5 % strain limit and a tangent modulus. Results are similar but not comparable.
How many specimens does ASTM D790 require?
At least five per condition, conditioned per ASTM D618 (23 ± 2 °C, 50 ± 10 % RH, at least 40 hours) unless the material specification states otherwise; five in each direction for anisotropic sheet.
What machine do I need for ASTM D790?
Any universal testing machine with ± 1 % force accuracy, speeds from about 0.5 to 50 mm/min, a three-point bend fixture with 5 mm radii and adjustable span, and either a deflectometer or compliance correction for crosshead deflection. Loads on plastics are usually under 1 kN, so a 2.5 or 5 kN benchtop frame with a 1 or 2 kN load cell is typical.
Need a fixture, grips or a complete system for ASTM D790? Contact Universal Grip with your specimen details, capacity and machine connection and we will quote the right setup.
