Blog
What Is Compression Set? Why It Matters for Seal Design
Apr
A rubber seal that looks perfect on day one can be leaking by year two. In most cases, the failure was predictable; and preventable; if compression set had been properly specified at the design stage. This guide explains what compression set is, how it is measured, how values differ by compound, and what it means for the service life of your seals and gaskets.
What Is Compression Set?
When a rubber seal is installed, it is compressed between two mating surfaces. That compression deforms the seal cross-section, and the resulting elastic reaction force; the rubber pushing back; is what creates the sealing pressure. The seal works because the rubber wants to return to its original shape and cannot, so it pushes against the surfaces instead.
The problem is that rubber is not a perfect elastic material. Under sustained compressive load, especially at elevated temperature, the polymer chains in the rubber gradually relax and rearrange into a partially compressed configuration. When the compressive load is eventually released; or when the mating surfaces move apart slightly due to thermal cycling or vibration; the rubber does not fully return to its original thickness. Some permanent deformation remains. That permanent deformation is compression set.
Compression set is the permanent reduction in thickness that a rubber specimen retains after being compressed for a defined time at a defined temperature and then released. It is expressed as a percentage of the original deflection. A compression set of 0% means the rubber recovered fully to its original thickness; perfectly elastic, no permanent deformation. A compression set of 100% means the rubber did not recover at all; it remained at the compressed thickness permanently. In real sealing applications, all rubber materials fall somewhere between these extremes, and the question is always: how much permanent deformation is acceptable before the seal loses enough contact force to allow leakage?
The practical consequence of compression set is straightforward. A static o-ring seal in a hydraulic fitting starts with a specific gland fill percentage; the ratio of the o-ring cross-section area to the gland cavity area. The difference between the uncompressed and compressed cross-section height creates the contact stress against the mating surfaces that seals the fluid. As compression set accumulates over months and years of service, the effective cross-section height of the o-ring decreases. Contact stress falls. At some point; which depends on the system pressure, the surface finish of the mating faces, and how much cross-section height has been lost; the contact stress drops below what is required to maintain a seal, and the fitting begins to seep. The rubber has not cracked or degraded chemically. It has simply flattened.
What Causes Compression Set?
Compression set has two underlying mechanisms that operate simultaneously, and understanding both is important for specifying materials correctly.
Physical Stress Relaxation
When rubber is deformed, the polymer chain network is stretched and strained. Over time, even at room temperature, the chains physically rearrange to reduce the stored elastic energy; the material “creeps” toward a lower-stress configuration. This process is purely mechanical and is why even a perfectly formulated seal at room temperature will accumulate some compression set over a long enough time. It is accelerated by temperature because elevated temperature gives the polymer chains more kinetic energy to rearrange. This is why compression set testing is conducted at elevated temperatures; the test is designed to replicate accelerated real-world stress relaxation in the time available for a lab test.
Chemical Crosslink Degradation
Rubber elasticity depends on the network of crosslinks formed during vulcanization. These crosslinks act like molecular springs; they are what causes the material to return to its original shape after deformation. In service, heat, oxygen, ozone, and fluid exposure all gradually attack and break crosslinks. As crosslinks degrade, the material loses elastic memory. The portion of the compression that the remaining crosslinks cannot recover from increases. This component of compression set is irreversible and compounds over the life of the seal. It is the dominant mechanism in long-term high-temperature seal applications and is why FKM and silicone; with more chemically stable crosslink systems; outperform NBR and neoprene in high-temperature sustained-compression applications.
How Compression Set Is Measured: ASTM D395
Compression set is standardized under ASTM D395, Standard Test Methods for Rubber Property: Compression Set. The standard defines three test methods, of which Method B is by far the most widely specified and the one you will encounter on virtually every material data sheet.
ASTM D395 Method B: Constant Deflection
The specimen (a cylindrical button, 29 mm diameter by 12.5 mm thick, Type 1) is placed between two parallel steel plates separated by steel spacer bars. The spacers are sized to compress the specimen to exactly 75% of its original thickness, a 25% deflection. The assembly is placed in a circulating air oven at the specified test temperature and held for the specified test duration. Standard conditions are 22 hours at 100°C for general-purpose testing, though application-specific conditions vary widely (70 hours at 150°C for automotive engine seal compounds, for example). After the test duration, the fixture is removed from the oven, the specimen is released from the plates, and it is allowed to recover at room temperature for exactly 30 minutes. The thickness is then measured and the compression set calculated.
tᵣ = recovered specimen thickness after release and 30-min recovery
tₛ = spacer bar thickness (= t₀ × 0.75)
Compression set is the permanent deformation (orange) that remains after the rubber recovers. Lower is better.
A Worked Example
A 50 Shore A NBR compound button is tested at 22 hours / 100°C. Original thickness t₀ = 12.5 mm. Spacer thickness tₛ = 9.375 mm (75% of original). After testing and 30-minute recovery, measured thickness tᵣ = 11.0 mm.
That 48% means nearly half of the deflection imposed during the test was not recovered. In a gland designed for 25% squeeze on an o-ring cross-section, a 48% compression set means the effective squeeze would approach zero; the seal would lose sealing force and fail. This is why compression set is one of the first properties a seal engineer reviews when evaluating a compound for a long-term static application.
Method A vs. Method B
Method A applies a constant force rather than a constant deflection. It is less commonly specified but relevant when the critical service condition is a defined load rather than a defined displacement; certain gasket applications under defined bolt loads, for example. Method B is the default for virtually all elastomeric seal material specifications unless a specific reason to use Method A is stated.
Compression Set by Elastomer: What to Expect
Compression set performance varies considerably across elastomer families, and within each family it varies by compound formulation, crosslink density, and cure system. The values below are representative ranges for properly formulated seal-grade compounds tested at ASTM D395 Method B, 22 hours / 100°C unless otherwise noted. Individual compounds will differ; always request compound-specific data for critical applications.
| Elastomer | Typical CS (%) | Test Conditions | Notes for Seal Engineers |
|---|---|---|---|
| FKM | 15-25% | 70 hrs / 200°C | Best high-temperature compression set resistance of any standard elastomer. The correct choice for static seals in aggressive chemical and sustained high-temperature service. Values at 200°C test conditions demonstrate genuine long-term performance where other elastomers would show dramatically higher set. |
| Silicone (VMQ) | 15-30% | 22 hrs / 175°C | Excellent elevated-temperature compression set resistance. Widely used for static seals at 150-200°C where FKM’s chemical resistance is not required. Low modulus means lower initial sealing force per unit squeeze, so gland design must account for this. Platinum-cured silicone achieves better compression set than peroxide-cured grades. |
| EPDM (Peroxide-Cured) | 20-35% | 22 hrs / 100°C | Significantly better compression set than sulfur-cured EPDM at equivalent temperature. The correct specification for steam gaskets, hot water seals, and potable water applications where long-term sealing force retention is required. Peroxide cure forms thermally stable C-C crosslinks rather than the sulfidic crosslinks of sulfur cure. |
| HNBR | 25-40% | 70 hrs / 150°C | Substantially better hot-oil compression set resistance than standard NBR. The correct upgrade for o-rings and seals in hot oil service above 250°F where NBR’s crosslinks degrade rapidly. Maintains sealing force at elevated temperature where NBR would harden and crack rather than set. |
| NBR (Nitrile) | 30-50% | 22 hrs / 100°C | Acceptable for moderate-temperature petroleum oil sealing where the service temperature stays below 250°F. At elevated temperatures, NBR’s compression set deteriorates rapidly as polysulfidic crosslinks revert under heat. An NBR o-ring that performs acceptably at 150°F may have dangerously high compression set at 230°F after extended service. |
| Neoprene (CR) | 35-55% | 22 hrs / 100°C | Moderate compression set for general industrial sealing where the primary driver is balanced oil and weather resistance rather than long-term sealing force retention. Not recommended for static seals requiring multi-year service life in demanding environments without periodic re-torquing of fasteners. |
| EPDM (Sulfur-Cured) | 40-65% | 22 hrs / 100°C | Standard catalog gasket sheet grade. Adequate for low-pressure, moderate-temperature applications with regular gasket replacement. Higher compression set than peroxide-cured grades makes this unsuitable for applications requiring multi-year sealing force retention without mechanical re-torquing. Often found in distributor sheet stock but misapplied in steam and hot water sealing programs. |
| Natural Rubber | 20-45% | 22 hrs / 70°C | Good compression set at moderate temperatures, but degrades significantly above 100°C as natural crosslinks are attacked by heat and oxygen. Appropriate for indoor static sealing applications at ambient temperature. Not appropriate for any sustained elevated-temperature sealing application despite low values at room-temperature test conditions. |
Why Compression Set Matters for Seal Design
Static vs. Dynamic Seals: Different Requirements
Static seals (o-rings in face seal grooves, flat gaskets between flanges, plug seals in ported fittings) are compressed once at installation and remain under that compression throughout their service life. The contact force that creates the seal is entirely dependent on the elastic recovery force of the rubber. As compression set accumulates over months and years, the contact force decreases continuously. This makes compression set the dominant material selection criterion for any static seal expected to hold pressure without re-torquing for extended service intervals.
The critical design question for a static seal is: after the expected service interval at the expected service temperature, will the remaining contact force; reduced by accumulated compression set; still exceed the minimum sealing pressure required to hold the system pressure? Working backward from this question, the engineer can calculate the maximum acceptable compression set for a given gland design, which then drives compound selection.
Dynamic seals (reciprocating rod seals, rotating shaft seals, oscillating pivot seals) operate differently. The seal cross-section is repeatedly compressed and partially released as the mating surface moves. Compression set in a dynamic seal causes the seal to “creep” into a shape that more permanently conforms to its average compressed position. This can actually reduce friction and the tendency toward “stick-slip” in reciprocating rod seals, making moderate compression set less damaging in dynamic applications than in static ones.
However, dynamic seals also face abrasion and fatigue loading that static seals do not, and a high-compression-set compound may also have inferior fatigue and tear resistance in dynamic service. The optimal dynamic seal compound balances compression set against abrasion resistance, modulus, and fatigue life rather than minimizing compression set in isolation.
Thermal Cycling and Compression Set
The most demanding condition for a static seal is not constant high temperature but repeated thermal cycling. A seal that sees 150°F during operation and ambient temperature during shutdown cycles between a state where the rubber is relatively stiff (cold) and one where it is more compliant (hot). At elevated temperature, compression set accumulates faster. When the system cools, the rubber’s modulus increases; it becomes stiffer; but the cross-section has already relaxed to a smaller height during the hot period. The result is a seal that is stiffer but thinner: higher modulus per unit of remaining thickness, but less total contact force because the cross-section is smaller than it was at installation.
After enough thermal cycles, a seal that passed all initial leak checks can fail because the accumulated compression set from the hot periods has permanently reduced the cross-section height below what is required to maintain sealing contact force at cold conditions. This failure mode is common in engine coolant seals, steam system gaskets, and outdoor process equipment seals that cycle with the environment. The solution is either a low-compression-set compound for the service temperature or an increased initial squeeze; more gland compression at installation to provide reserve sealing force for the accumulated set.
Bolt Load and Gasket Seating
For flat gaskets in bolted flanges, compression set interacts directly with bolt relaxation over time. At installation, the bolt load compresses the gasket and creates the sealing interface. Over time, both the gasket creeps (compression set) and the bolt slowly relaxes under sustained tensile load; a phenomenon called bolt relaxation or bolt creep. Both reduce the residual clamp force on the gasket. If the combined loss of clamp force from bolt relaxation and gasket compression set reduces the residual contact stress below the minimum required sealing stress, the joint leaks.
This is why high-performance bolted flange joints in process piping and pressure vessels are designed with initial bolt load that provides significant reserve; the initial sealing stress must be high enough that after both bolt relaxation and gasket compression set have occurred over the expected service interval, the residual contact stress still exceeds the system pressure at the leakage interface. Gasket compression set data, specifically long-duration creep relaxation data (not just short-term ASTM D395 values), is an essential input to this calculation.
How to Specify Compression Set Correctly
A compression set limit in a material specification is only meaningful when it is accompanied by the test conditions; the method, temperature, and duration. Specifying “compression set less than 30%” without specifying the test conditions is incomplete and will be interpreted differently by different suppliers.
- Identify the maximum service temperature of the seal in its installed environment, not just the ambient temperature. An o-ring in a hydraulic line sees the fluid temperature, which may be significantly higher than the ambient environment around the equipment.
- Select the test temperature to match the maximum service temperature. Testing at a lower temperature than the seal will actually experience underestimates the compression set the seal will develop in service. The test temperature should equal or exceed the maximum service temperature.
- Select the test duration based on the expected service interval. A 22-hour test is a screening test. A 70-hour or 168-hour test gives more meaningful data for seals expected to hold pressure for months or years without replacement. For very long service intervals, request long-term creep relaxation data, not just standard ASTM D395 results.
- Set the maximum acceptable value based on the gland design. Calculate the minimum residual contact force required at the end of the service interval, work backward to the maximum allowable cross-section height loss, and convert that to a compression set limit as a percentage of the initial deflection. Do not simply specify an arbitrary percentage copied from a previous specification without verifying it applies to the new design.
- Require Method B (constant deflection) unless there is a specific reason to use Method A. This is the standard that most material data sheet values are reported under, making comparison meaningful.
- Request lot-by-lot certification for critical applications. Compression set can vary significantly between compound batches depending on mixing, cure time, and cure temperature. A material specification should state whether lot testing or qualification testing per defined intervals is required.
Three Common Compression Set Specification Mistakes
1. Using Room-Temperature Data for High-Temperature Applications
Many elastomers show very good compression set values at room temperature or at 70°C; values that look acceptable on a data sheet. The same compound tested at the actual service temperature (150°C, 175°C, 200°C) shows dramatically higher set values that would immediately disqualify the material. Always test at the application temperature. A material that performs well at 70°C may be entirely unsuitable at 150°C, and this difference is often the root cause of seal failures in high-temperature equipment that “worked fine” during initial qualification testing at ambient conditions.
2. Accepting Catalog EPDM Sheet for Steam or Long-Service Sealing
Commercial EPDM gasket sheet from distributors is almost universally sulfur-cured; the cure system that is faster and less expensive to process. Sulfur-cured EPDM has compression set values 40-80% higher than peroxide-cured grades under equivalent test conditions. For steam gaskets, hot water seals, or any application requiring multi-year gasketed joint integrity without retorquing, specifying sulfur-cured catalog EPDM will result in gasket creep and joint leakage within a fraction of the expected service interval. The fix is simple; specify peroxide-cured EPDM; but it requires knowing the distinction exists. Most procurement teams do not, and many suppliers do not surface it unless asked.
3. Ignoring Compression Set for Dynamic Applications
While dynamic seals are less sensitive to compression set than static seals, ignoring it entirely creates a different problem: a high-compression-set compound in a dynamic seal will gradually “flow” toward its most compressed configuration and away from its least compressed configuration. In a reciprocating rod seal, this means the seal progressively loses its designed interference on the rod and gains excess material on the back side of the groove. The result is reduced sealing force on the rod and eventual leakage; a failure that looks like a worn seal but is actually a set seal. Specifying a compression set limit; less demanding than for static seals, but present; prevents this failure mode.
Key Takeaways
- Compression set measures permanent deformation after sustained compression. Lower values indicate better elastic recovery and longer seal life.
- The ASTM D395 Method B formula is: CS (%) = [(t₀ – tᵣ) / (t₀ – tₛ)] × 100. Test conditions must always be stated alongside the result.
- Two mechanisms drive compression set: physical stress relaxation (mechanical) and crosslink degradation (chemical). Both are accelerated by temperature.
- FKM and silicone offer the best high-temperature compression set resistance. EPDM peroxide-cured is the correct choice for steam and hot water. NBR is acceptable at moderate temperatures but degrades rapidly above 250°F.
- Cure system matters for EPDM: peroxide-cured grades have 40-80% better compression set than sulfur-cured catalog sheet at equivalent test conditions.
- Static seals are more sensitive to compression set than dynamic seals. Specify strictly for static seals; relax limits moderately for dynamic applications.
- Thermal cycling accelerates failure. Design reserve sealing force to account for accumulated compression set across the expected number of thermal cycles, not just sustained high temperature.
- Always specify test conditions (ASTM D395 Method B, temperature, duration) alongside the maximum acceptable compression set percentage. An isolated percentage without conditions is not a meaningful specification.
Related Technical Resources
Need a Low Compression Set Seal Compound?
Wayne Rubber manufactures custom o-rings, gaskets, and seals to compound specification. Provide the service temperature, fluid contact, service interval, and any compression set limit and Wayne Rubber’s engineering team will recommend the correct compound and produce test data to confirm performance.