O-ring testing standards explained: how to read service life from a material certificate

An O-ring's material certificate is not a filing-cabinet formality — it is a prediction of seal life written in the language of laboratory standards. When you order a ring in FKM or NBR, the supplier will give you the material code per ISO 1629 / ASTM D1418, but the real question is this: will this seal last three years in my application, or fail in three months? The answer is hidden in four groups of measurements — compression set, heat ageing, volume change after immersion in the service medium, and hardness. This article is for the buyer and the maintenance engineer holding a certificate who needs to know which number flags which risk, and where the limits are beyond which a figure is just marketing. We give numbers only where they are verifiable from the standards; where a standard prescribes no specific value, we offer qualitative guidance rather than an invented figure.

Compression set (ISO 815 vs ASTM D395): how much sealing force remains

Compression set — the permanent deformation left after a material is squeezed — is the single most important number for a static seal. An O-ring seals because it is compressed in its groove and its own elasticity pushes back against the mating faces. If the material loses that elasticity under long, hot compression and stays permanently flattened, the sealing force disappears and the ring begins to leak with no visible damage at all. ISO 815 measures exactly this: a specimen is compressed to a constant deflection — normally 25 % — held compressed at a chosen temperature for a set time, then released, and after recovery the standard records what percentage of the original deflection remains permanently deformed. A result of 0 % means perfect recovery; 100 % means fully lost elasticity. ISO 815-1 covers ambient and elevated temperatures, while ISO 815-2 covers low temperatures.

When reading a certificate, always check the conditions under which the value was measured, because without them the number is meaningless. Short tests — typically 22 hours at elevated temperature — serve mainly as a check on the state of cure and the quality of the compound; long tests, often around 1000 hours, take account of the effect of ageing and are used to predict real sealing service life. The American equivalent, ASTM D395, has Method A (constant force) and Method B (constant deflection, typically 25 % for 22 hours); it is Method B that aligns most closely with ISO 815. Beware of comparing numbers directly across standards: although the methods are related, the way the specimen is released and the reference basis differ, so never swap an ISO 815 value one-for-one with an ASTM D395 value without checking the method. In practice: a lower compression set means a longer static-seal life, which is exactly why a compression-set failure points you toward FKM or FFKM, materials with markedly better recovery than standard NBR.

Heat ageing (ISO 188): what happens after years in the heat

No measurement on the day of delivery tells you how the material will fare after two years in a hot environment — which is why accelerated ageing per ISO 188 exists. A specimen is exposed to hot air in an oven for a defined period, then its properties are compared with an unaged specimen. The standard recommends measuring tensile strength, stress at a given elongation and elongation at break per ISO 37, together with hardness per ISO 48-2. ISO 188 has several methods that differ in the type of oven and the airflow (from slow laminar flow to high-speed turbulent flow with specimen rotation). The principle of acceleration is simple: each additional roughly 10 °C approximately doubles the rate of thermal degradation, so a few days in the oven simulate months or years of real service.

For the buyer, the direction of change is what matters. Most rubbers harden and embrittle with age — elongation at break falls and hardness rises, which on a ring ends in radial cracks and lost elasticity, exactly as in thermal degradation in service. Some compounds, by contrast, soften. A certificate showing a small change in elongation and a small rise in hardness after ageing at your operating temperature is a strong signal of long life; a large drop in elongation is a warning that the material is under-rated for that temperature. For a heat failure, the fix is a material with a higher ceiling — FKM to about 200 °C, VMQ silicone with its wide range for static high-temperature applications, and at the extreme FFKM with a continuous ceiling around 260 °C.

Volume change in the medium (ASTM D471 / ISO 1817): swell as a warning

Chemical incompatibility is the most common cause of premature O-ring failure, and it is measured by immersing a specimen in the service fluid per ASTM D471, or its international equivalent ISO 1817. The specimen is immersed in oil, fuel, solvent or chemical for a defined time — typically 24, 72 or 168 hours — and then the percentage change in volume, mass and hardness is measured. ASTM D471 also works with reference fluids (the IRM 901 to 903 reference oils and reference fuels) so that results are comparable between laboratories. Volume swell is a direct indicator of chemical compatibility: the more a material swells in the medium, the faster it loses its sealing ability.

To grade the numbers we use the proven five-tier scale from the reference handbooks, the same scale built into our chemical-compatibility matrix. Volume swell up to about 5 % (and a hardness change up to about 10 %) is excellent and suitable for long-term service; 5 to 15 % is good and acceptable for normal duty; 15 to 30 % is marginal and fit only for static or short-term use; above 30 % swell, or rapid degradation, means the material is unusable and must be avoided. Mild swell is not always bad: a small volume change can seal a static groove more tightly. Heavy swell, however, softens the ring, overloads it in the groove and accelerates abrasion, so when designing a groove always allow a volume reserve and keep the groove fill ratio below 90 % so a swollen ring does not exceed its capacity.

Hardness (ISO 48 / Shore A): the right method for a round cross-section

Hardness is the property buyers know best — it is given in Shore A points, and most O-rings sit roughly in the 40 to 90 Shore A range. NBR and EPDM both span the full 40 to 90 Shore A band, FKM is typically 60 to 90, and VMQ silicone runs from a soft 30 upward. Hardness governs how the ring behaves under pressure: a softer ring conforms better to an imperfect surface and seals at low pressure, while a harder ring resists extrusion into a gap better at high pressure. For a high-pressure application with a large gap, the right answer is higher hardness or a PTFE back-up ring, not a softer material.

The catch is which method the hardness was measured by, because on the round cross-section of an O-ring it makes a difference. ISO 48-4 (and the American ASTM D2240) describes durometer measurement — Shore — on flat specimens. For rounded, non-planar bodies, that is, precisely for O-rings, ISO 48-2 is the relevant standard, measuring hardness in IRHD (International Rubber Hardness Degrees), with the American counterpart ASTM D1415. IRHD uses a defined force and time on a thicker specimen and gives tighter, more repeatable control in the laboratory, whereas Shore A is faster and more widespread. If a certificate states an O-ring's hardness measured on a flat body by the Shore method, treat it as indicative — for the round cross-section itself, the methodologically correct value is IRHD per ISO 48-2.

Low temperature and brittleness (ASTM D746 / ISO 812): when the seal stiffens

The lower end of the temperature range is verified by a brittleness test. ASTM D746 (and the international ISO 812) determines the brittleness temperature — the lowest temperature at which the material does not yet fracture on impact; more precisely, the temperature at which half the specimens fail by brittle fracture under defined impact conditions. For classic vulcanised rubber, ASTM D2137 is often more appropriate. Read the value together with the lower limit of the material's continuous range from the datasheet: NBR has a lower limit around −30 °C, FKM around −20 °C, while EPDM and polyurethane reach roughly −40 °C and VMQ silicone as far as −60 °C. Below the brittleness temperature even an otherwise suitable material loses its elasticity, stiffens and cracks on motion or impact — which is why, for low-temperature applications, the brittleness temperature is one of the figures worth requesting from the certificate.

How to read these numbers when buying

The four groups of measurements together give the full picture: compression set per ISO 815 tells you how much sealing force remains after time under squeeze; ISO 188 shows how the material withstands years in the heat; ASTM D471 / ISO 1817 reveals whether the medium will swell or break down the ring; and hardness per ISO 48, in the correct method, governs behaviour under pressure. When reading a certificate, always verify the test temperature and time — a value without conditions has no evidential force — and compare only values measured by the same method and standard. If any of these measurements is weak at your operating temperature and medium, that is an early warning of failure, before you have even fitted the ring. When you are unsure which material will hold up under your conditions, use our configurator and chemical-compatibility matrix, or write to us — a material verified in advance beats a premature failure in service.