ISO 814
Rubber-to-Metal Adhesion, Two-Plate Method

ISO 814 Two-Plate Adhesion Test Fixture for Rubber to Metal by Universal Grip

Downloads: ISO 814 Two-Plate Adhesion Test Fixture for Rubber to Metal – Universal Grip Data Sheet (PDF)

Written by Chris Thomas, Universal Grip Company · Reviewed by Frank Bertini · Updated October 27, 2026

ISO 814 is the international test method for measuring how strongly a vulcanized or thermoplastic rubber is bonded to metal. A short rubber cylinder is moulded and bonded between two parallel metal plates, and the assembly is pulled apart in tension at 90° to the bonded faces. The maximum force divided by the bonded cross-sectional area gives the adhesion value in pascals (in practice MPa), and the location of the failure is recorded with a standard set of symbols.

The method is used by compounders, bonding-agent (primer and cover cement) suppliers and manufacturers of bonded parts such as engine and suspension mounts to develop compounds, qualify adhesive systems and control the moulding process. Universal Grip builds a dedicated ISO 814 two-plate adhesion test fixture with split holders, thumb screws and self-aligning clevis housings, described in the equipment section below.

In this guide: What ISO 814 Measures · Current Edition, Scope and History · ISO 814 Test Piece: Dimensions, Preparation and Number · Conditioning and Time Between Vulcanization and Test · Test Machine, Fixture and Load Cell Requirements · ISO 814 Test Procedure Step by Step · Calculating ISO 814 Adhesion Strength · Failure Symbols and Interpreting Results · Common ISO 814 Errors and How to Avoid Them · ISO 814 vs ASTM D429 and Related Standards · What to Include in the ISO 814 Test Report · ISO 814 Equipment from Universal Grip · FAQ

What ISO 814 Measures

The principle of ISO 814 is simple: measure the force needed to rupture a standard-sized unit of rubber bonded to two parallel metal plates, with the force applied at 90° to the bonded surfaces. Because the rubber disc is thin and fully bonded on both faces, the bond line sees a nominally uniform tensile (normal) stress rather than the concentrated line load of a peel test.

The test therefore answers two questions at once:

How much tensile stress the bonded assembly carries before it fails, reported as the adhesion value in pascals (normally expressed in MPa).
Where it fails: in the rubber itself, at one of the adhesive or primer interfaces, or in the metal. This failure location is often more informative than the number.

A bond that fails entirely in the rubber (symbol R) is generally the target, because it means the bond is stronger than the rubber. In that case the measured value is a lower bound on the true bond strength, not the bond strength itself. Interfacial failures point to surface preparation, adhesive selection, film thickness, contamination or cure problems.

Note: ISO 814 is a laboratory method. Its scope states that it applies primarily to test pieces prepared in the laboratory under standard conditions, to provide data for compound development and control of manufacturing methods. It is not a test of finished bonded products, whose geometry and stress state are different.

Current Edition, Scope and History

The current version is ISO 814:2017, the sixth edition, prepared by ISO/TC 45 (Rubber and rubber products), Subcommittee SC 2 (Testing and analysis). It was last reviewed and confirmed by ISO in 2022, so it remains the edition to cite. The full title is "Rubber, vulcanized or thermoplastic — Determination of adhesion to metal — Two-plate method".

ISO 814:2017 (6th edition): minor revision of the 2011 edition. It updated the normative references (Clause 2) and the failure symbols (Clause 9.2).
ISO 814:2011 (5th edition): replaced by the 2017 edition.
ISO 814:2007: title extended to vulcanized or thermoplastic rubber. It already specified a 25 mm/min ± 5 mm/min separation rate, at least three test pieces and reporting of each individual result with the percentage of each failure type (some older national versions used 50 mm/min and five test pieces).
ISO 814:1996: titled "Rubber, vulcanized — Determination of adhesion to metal — Two-plate method", before thermoplastic rubber was added to the title.

The 2017 edition references three other standards that you need alongside it: ISO 5893 (constant-rate-of-traverse test machines), ISO 18899:2013 (calibration of rubber test equipment) and ISO 23529 (preparation and conditioning of rubber test pieces). The calibration schedule is given in Annex A.

Note: If a method summary you are reading still quotes a 50 mm/min pull speed, five test pieces, or only the R, RC, CP and M failure codes, it is based on an older national standard or on ASTM D429, not on the current ISO 814.

ISO 814 Test Piece: Dimensions, Preparation and Number

The standard test piece is a rubber cylinder bonded on both circular faces to metal plates. The dimensions specified in ISO 814:2017 are:

Rubber thickness: 3 mm ± 0.1 mm.
Rubber diameter: 35 mm to 40 mm, known to the nearest 0.1 mm.
Metal plates: two plates of equal diameter, about 0.1 mm smaller in diameter than the rubber cylinder.
Plate thickness: at least 9 mm.
Plate material: preferably rolled carbon steel; other metals may be used as long as the essential dimensions are kept.
Rubber overhang: the mould is designed so the rubber projects about 0.05 mm beyond the plate edges, which prevents the sharp metal edge from initiating a tear.
Measurement of dimensions: in accordance with ISO 23529.
Number of test pieces: at least three.

Preparing bonded test pieces

Preparation is where most ISO 814 scatter originates, so the standard is specific about good practice:

Prepare and treat the metal surfaces according to the adhesive system being investigated (for example degreasing, grit blasting, priming and applying the cover cement).
Cut the rubber discs with a circular die, sized so that only a limited amount of flash forms during moulding, and treat the rubber bonding surface as the method under investigation requires.
Keep exposed bonding surfaces free of dust, moisture and foreign matter, and do not touch them by hand during assembly.
Mould the assembly by heating under pressure for a set time at a controlled temperature, both chosen to suit the compound and adhesive system.
Remove the test pieces from the mould carefully and avoid stressing the bonded surfaces until they have cooled.

The plates also need a feature that lets the test fixture grip them, such as a groove or thread on the back face or rim. ISO 814 shows a fixture only as an example, so the exact gripping geometry is set by the laboratory and must match the fixture.

Conditioning and Time Between Vulcanization and Test

Condition the test pieces for at least 16 h immediately before testing.
Conditioning temperature: 23 °C ± 2 °C or 27 °C ± 2 °C. Use the same temperature throughout any series of tests that will be compared.
Interval between vulcanization and testing: as specified in ISO 23529.

Testing at a temperature other than the conditioning temperature is a common requirement for automotive mounts. If you test hot or cold, fit an environmental chamber around the fixture, allow the bonded test piece to reach temperature through its 9 mm or thicker plates before loading, and report the test temperature separately from the conditioning temperature.

Test Machine, Fixture and Load Cell Requirements

Test machine

ISO 814 calls for a tensile testing machine that meets ISO 5893, with force measurement to class 2 accuracy and a moving grip that travels at 25 mm/min ± 5 mm/min. Inertialess force measurement (an electronic load cell, or an optical system) is preferred; pendulum-type dynamometers are discouraged because friction and inertia in the mechanism bias the reading. Any modern electromechanical universal testing machine with a strain-gauge load cell meets this preference.

Fixtures and alignment

The fixtures must hold the test piece and keep the applied load accurately centred, so that the tension is distributed evenly over the bonded cross-section. This is the critical requirement. A 3 mm rubber layer between two stiff plates is very sensitive to bending: any angular misalignment between the machine axis and the bond line turns part of the bond into a peel edge, which starts failure early at one side and lowers the result.

Good two-plate fixtures therefore use pinned or spherical (self-aligning) connections at both ends, so the test piece can rotate to align itself with the load line as force builds, and grip the plates in a way that does not pre-load the bond during clamping.

Sizing the load cell

Estimate the peak force before choosing a load cell. Force equals stress times bonded area: a 40 mm diameter test piece (1,257 mm²) that reaches 8 MPa needs about 10 kN, while a 35 mm test piece (962 mm²) at 5 MPa needs about 4.8 kN. Choose a load cell so typical peaks fall well within its verified range, and confirm the fixture and machine capacity cover the strongest bonds you expect.

ISO 814 Test Procedure Step by Step

1. Confirm the test pieces have been conditioned for at least 16 h at 23 °C ± 2 °C or 27 °C ± 2 °C and that the vulcanization-to-test interval complies with ISO 23529.
2. Measure and record the rubber diameter to the nearest 0.1 mm (and check the 3 mm ± 0.1 mm rubber thickness).
3. Set the crosshead speed to 25 mm/min (tolerance ± 5 mm/min) and zero the force channel with the fixtures hanging free.
4. Mount the test piece in the fixtures and centre it carefully so the tension will be uniform across the cross-section. Do not apply bending or twisting while clamping.
5. Remove slack without pre-loading the bond, then start the test.
6. Separate the grips at constant rate until the test piece breaks, and record the maximum force.
7. Remove the halves and examine both fracture faces. Assign the failure symbol(s) and estimate the percentage of the bonded area for each.
8. Repeat for the remaining test pieces (at least three in total).

Capturing the full force-displacement curve with testing software is worth doing even though ISO 814 only requires the maximum force. A curve that drops in steps, or a peak that occurs at a much lower extension than its neighbors, often signals misalignment or a partial interfacial failure.

Calculating ISO 814 Adhesion Strength

The adhesion value is the maximum force divided by the cross-sectional area of the test piece, expressed in pascals:

Bonded area: A = π × d² / 4, where d is the measured diameter in mm.
Adhesion value: σ = F_max / A. With F in newtons and A in mm², σ is in N/mm², which equals MPa.

Worked example

Three test pieces of one compound and adhesive system have a rubber diameter of 38.0 mm. The bonded area is π × 38.0² / 4 = 1,134.1 mm². The maximum forces recorded are 6.12 kN, 6.45 kN and 5.98 kN.

Test piece 1: 6,120 N ÷ 1,134.1 mm² = 5.40 MPa, failure R-100 (all in the rubber).
Test piece 2: 6,450 N ÷ 1,134.1 mm² = 5.69 MPa, failure R-100.
Test piece 3: 5,980 N ÷ 1,134.1 mm² = 5.27 MPa, failure R-80 / RC-20.
Arithmetic mean, if your own specification asks for one: (5.40 + 5.69 + 5.27) ÷ 3 = 5.45 MPa.

ISO 814 asks for the result of each test piece together with the percentage of each failure type, so report the three values and their failure codes rather than only an average. Here the first two pieces show the bond outperforming the rubber; the third shows 20% failure at the rubber-to-adhesive interface, which deserves a look at cover-cement application or mould contamination.

Note: Which diameter? The plates are about 0.1 mm smaller than the rubber. On a 38.0 mm test piece, using 37.9 mm instead of 38.0 mm changes the area by about 0.5% and raises each result by about 0.03 MPa. State which dimension you use and use it consistently.

Failure Symbols and Interpreting Results

The 2017 edition updated the failure symbols so they cover both primer-plus-adhesive systems and direct bonding. Record every symbol that applies and the estimated percentage of the bonded area for each:

R: failure in the rubber.
RC: failure at the rubber–adhesive interface.
CP: failure at the adhesive–primer interface (when a primer is used).
PS: failure at the primer–substrate interface (when a primer is used).
CS: failure at the adhesive–substrate interface (when no primer is used).
D: failure at the rubber–substrate interface in direct bonding (no adhesive).
S: failure in the substrate.

Reading the results

R-100 at a high value: the bond exceeds the cohesive strength of the rubber. Further bonding work will not raise the number; the compound sets the limit.
RC failures: poor wetting or reaction between rubber and cover cement, scorched or bloomed rubber surface, or a cover cement poorly matched to the compound.
CP failures: incompatible primer and cover cement, or incorrect drying time between coats.
PS or CS failures: metal preparation problems such as residual oil, oxidation after blasting, or excessive time between blasting and priming.
Low values with mixed modes and wide scatter: often alignment or test-piece preparation, rather than the adhesive system.

Values from ISO 814 are only comparable when compound, adhesive system, metal, plate preparation, cure, conditioning and test temperature are the same. Use the method for ranking and control, not as a design allowable for a production part.

Common ISO 814 Errors and How to Avoid Them

Misalignment: rigid, non-aligning grips or plates clamped off-axis cause one-sided peel and low results. Use self-aligning, pinned connections and check that both plate faces are parallel.
Pre-loading during clamping: tightening a fixture against a bonded test piece can crack the bond edge before the test starts. Use holders that engage the plates without loading the rubber.
Wrong speed: 25 mm/min ± 5 mm/min is required. Rubber is rate sensitive, so 50 mm/min (an older value) gives higher forces that are not comparable.
Wrong geometry: ASTM D429 Method A test pieces (3.2 mm rubber, 39.9 mm diameter) are not ISO 814 test pieces (3 mm, 35–40 mm). Do not mix data sets.
Torn edges: if the rubber does not overhang the plates slightly, the sharp metal edge starts a tear. Check mould design and flash.
Short conditioning: testing sooner than 16 h after conditioning starts, or without respecting the ISO 23529 interval after vulcanization, adds scatter.
Reporting only an average: the standard expects each result and the failure-type percentages.

ISO 814 vs ASTM D429 and Related Standards

The closest ASTM equivalent is ASTM D429 Method A, which also bonds a rubber cylinder between two parallel metal plates and pulls it apart in tension. The two methods are similar in principle but not interchangeable:

Rubber thickness: ISO 814 uses 3 mm ± 0.1 mm; ASTM D429 Method A uses 3.2 mm ± 0.1 mm.
Diameter: ISO 814 allows 35 mm to 40 mm; ASTM D429 Method A specifies an end area of 1,250 mm² (39.9 mm diameter), with a smaller alternative size whose results should not be compared with the standard size.
Plates: ISO 814 requires at least 9 mm thickness; ASTM D429 Method A requires at least 9.5 mm.
Speed: ISO 814 runs at 25 mm/min ± 5 mm/min; ASTM D429 Method A runs at 0.40 mm/s ± 0.04 mm/s (24 mm/min).
Failure codes: ISO 814:2017 uses R, RC, CP, PS, CS, D and S; ASTM D429 uses R, RC, CP and M.

ASTM D429 also contains peel methods. In the ISO system, the 90° peel test of rubber bonded to a single metal plate is ISO 813, the one-plate method. Use ISO 814 when you want a uniform tensile bond stress; use ISO 813 when the application is peel-dominated or you want to see how a bond resists crack growth. The cohesive tensile strength of the rubber itself, which caps an R-mode result, is measured with ISO 37.

What to Include in the ISO 814 Test Report

A reference to ISO 814 (ISO 814:2017).
Test piece details: vulcanization time, temperature and date, compound and adhesive system, and the metal if not steel.
Conditioning time and temperature, and the test temperature.
Number of test pieces tested.
The adhesion value of each test piece and the percentage of each failure type.
Any unusual observations and any operations not specified in the standard.
The date of the test.

ISO 814 Equipment from Universal Grip

ISO 814 Adhesion Test Fixture (part ref. UGC-ISO814-ATF)

Our two-plate adhesion test fixture is a tensile fixture built for rubber bonded between two metal plates. Split holders with thumb screws engage the grooved end plates of the test piece, and pinned clevis housings with self-aligning seats pull the assembly apart along its axis so the bond sees uniform tension.

Test method: two-plate tensile adhesion, with axial tension perpendicular to the bond.
Specimens: rubber disc bonded between two metal plates, per ISO 814.
Load capacity: 10 kN standard; higher capacities on request.
Material: hardened steel holders and housings.
Gripping: split cups that slide onto grooved test piece plates and are secured with thumb screws, without tools.
Alignment: self-aligning seats in the clevis housings keep the bond in pure axial tension.
Machine connection: pin connections at both ends, made to suit your frame.
Customization: holder geometry is made to suit your test piece plates.

When ordering, specify the test piece plate diameter and groove geometry, the required capacity and your test machine connection. Typical applications include rubber-to-metal bonded components, bonding agent qualification, engine and suspension mounts, and adhesive and primer development.

Machines, adapters and environmental testing

The fixture runs on electromechanical universal testing machines, and a benchtop frame is often sufficient at 10 kN. Where the pin connection on your frame differs, see our adapters. For more rubber methods see our elastomer testing page, and for bonding agents our adhesive testing range.

Tell us your plate drawing and frame model and contact us for a quote.

ISO 814 FAQ

What is ISO 814?

ISO 814 is the ISO test method for determining the adhesion of vulcanized or thermoplastic rubber to metal using the two-plate method. A 3 mm thick rubber cylinder bonded between two metal plates is pulled apart at 90° to the bond, and the maximum force divided by the bonded area gives the adhesion value in pascals. The failure location is recorded with standard symbols.

What is the current edition of ISO 814?

The current edition is ISO 814:2017, the sixth edition, which ISO last reviewed and confirmed in 2022. It replaced ISO 814:2011 as a minor revision that updated the normative references and the failure symbols.

What are the ISO 814 test piece dimensions?

The rubber cylinder is 3 mm ± 0.1 mm thick and 35 mm to 40 mm in diameter, known to the nearest 0.1 mm. It is bonded between two metal plates, preferably rolled carbon steel, at least 9 mm thick and about 0.1 mm smaller in diameter than the rubber, so the rubber overhangs the plate edges by about 0.05 mm.

What test speed does ISO 814 use?

The grips separate at a constant 25 mm/min ± 5 mm/min until the test piece breaks. Older national versions used 50 mm/min, and results at that rate are not comparable because rubber is rate sensitive.

How many test pieces are required by ISO 814?

At least three test pieces are tested. The standard asks for the result of each test piece and the percentage of each failure type to be reported, rather than only an average.

How is ISO 814 adhesion strength calculated?

Divide the maximum force by the cross-sectional area of the test piece, A = π × d² / 4. For example, 6,120 N on a 38.0 mm diameter test piece (1,134.1 mm²) gives 5.40 MPa.

What do the failure codes R, RC, CP, PS, CS, D and S mean?

R is failure in the rubber; RC at the rubber–adhesive interface; CP at the adhesive–primer interface; PS at the primer–substrate interface; CS at the adhesive–substrate interface when no primer is used; D at the rubber–substrate interface in direct bonding; and S in the substrate. Report each code that applies with its estimated percentage of the bonded area.

Is ISO 814 the same as ASTM D429?

Not exactly. ISO 814 is closest to ASTM D429 Method A, but the ASTM method uses a 3.2 mm thick, 39.9 mm diameter rubber cylinder, 9.5 mm minimum plates, 0.40 mm/s and different failure codes. See our ASTM D429 guide for that method.

What is the difference between ISO 814 and ISO 813?

ISO 814 pulls a rubber disc bonded between two plates in direct tension, giving a stress in MPa. ISO 813 is the one-plate method, a 90° peel test of a rubber strip bonded to a single metal plate, giving a peel force per unit width. Both are used for rubber-to-metal bond development.

How should ISO 814 test pieces be conditioned?

Condition them for at least 16 h immediately before testing at 23 °C ± 2 °C or 27 °C ± 2 °C, using the same temperature for any series that will be compared. The interval between vulcanization and testing follows ISO 23529.

What fixture do I need for ISO 814?

You need a fixture that grips each metal plate and keeps the load accurately centred on the bond, ideally with self-aligning pinned connections. Universal Grip's ISO 814 adhesion test fixture uses split holders with thumb screws on grooved plates and self-aligning clevis housings, 10 kN standard. Contact us with your plate geometry for a quote.

Need a fixture, grips or a complete test system for ISO 814? Contact Universal Grip with your material, specimen size and machine connection and we will quote the right setup.

Literature

Pricing and Availability

ISO 814 Rubber-to-Metal Adhesion, Two-Plate Method