O-ring Extrusion, Compression Set and Chemical Swell Failure: Root Cause and Corrective Action

Overview #

When we conduct failure analysis on returned O-rings from fluid power systems, the single most common finding is not wrong material grade — it is wrong hardness specification combined with inadequate groove geometry. Procurement teams routinely specify Shore A 70 as a default without checking the system pressure, gap clearance, or fluid compatibility, and then attribute the resulting extrusion failures to “supplier quality.” The material is usually fine. The specification is wrong. Understanding the six primary O-ring failure modes — extrusion, compression set, chemical swell, abrasion, installation damage, and hardness mismatch — and their measurable thresholds is the difference between a seal that lasts 10,000 hours and one that fails in 400.

Failure Mode Identification: Visual Diagnosis Before Root Cause #

Before any corrective action, the returned seal must be examined correctly. Most failure analysis errors happen at this stage — teams identify the symptom (leakage) without identifying the mechanism (extrusion vs. compression set vs. chemical attack), and the corrective action targets the wrong variable.

The six failure modes produce visually distinct signatures:

The most important rule in failure analysis: do not assume the failure mode from the leak location alone. A seal leaking at the static face can be failing by compression set, chemical swell, or installation damage — three completely different corrective actions.

For O-rings and static seals sourced from China, we always request the failed seal back from the customer before issuing a replacement recommendation. The visual evidence on the returned part is more diagnostic than any COA.

Extrusion Failure: Clearance Gap, Pressure, and Hardness Thresholds #

Extrusion is the most common failure mode in hydraulic and pneumatic systems operating above 100 bar. The O-ring is forced into the diametral clearance gap between mating hardware components, and the extruded material is sheared off during pressure cycling. The result is progressive material loss, leakage, and eventual seal destruction.

The governing relationship is between system pressure, diametral clearance gap, and O-ring hardness. ASTM International D2000 and ISO Standards 3601-1 both provide clearance gap limits as a function of pressure and durometer — but the specific values are what most procurement teams never look up.

For a Shore A 70 NBR O-ring:
– Maximum recommended clearance gap at 70 bar: 0.10 mm (one-sided)
– Maximum recommended clearance gap at 140 bar: 0.05 mm (one-sided)
– At 210 bar with Shore A 70: extrusion risk is high regardless of gap — upgrade to Shore A 90 is required

For a Shore A 90 NBR O-ring:
– Maximum recommended clearance gap at 210 bar: 0.10 mm (one-sided)
– Usable to approximately 350 bar with backup rings installed

The corrective action for extrusion is almost never “use a better O-ring.” It is one of three things: reduce the clearance gap by tightening hardware tolerances, increase O-ring hardness to Shore A 90, or install PTFE or nylon backup rings on the low-pressure side of the groove. Backup rings are the correct solution when hardware rework is not feasible.

Most procurement teams over-specify tensile strength and under-specify the parameter that actually matters in extrusion applications: hardness at operating temperature. An NBR compound rated Shore A 70 at 23°C may measure Shore A 58–62 at 100°C — and at that softened state, the clearance gap tolerance drops significantly. Always request hardness-at-temperature data, not just ambient hardness, when specifying seals for elevated-temperature hydraulic systems.

Compression Set Failure: Test Method, Threshold, and Thermal Root Cause #

Compression set is the permanent deformation retained by an elastomer after sustained compressive load. It is expressed as a percentage: 0% means full recovery, 100% means no recovery. In practice, any O-ring with compression set above 25% in a static application will begin to show leakage as the sealing force drops below the minimum required contact stress.

The standard test is ASTM International D395 Method B: the specimen is compressed to 25% of its original thickness, held at a specified temperature for 70 hours, released, and measured after 30 minutes of recovery. The result is the percentage of original deflection that was not recovered.

In our qualification program, we apply the following pass/fail thresholds for static sealing applications:

• NBR (70 Shore A): compression set ≤ 20% after 70h at 100°C per ASTM D395B
• FKM (75 Shore A): compression set ≤ 15% after 70h at 175°C per ASTM D395B
• EPDM (70 Shore A): compression set ≤ 20% after 70h at 125°C per ASTM D395B
• Silicone (50 Shore A): compression set ≤ 30% after 70h at 175°C per ASTM D395B

We reject batches where Shore A hardness deviates more than ±3 points from the specified grade, and we treat any compression set result above these thresholds as a disqualifying finding — not a conditional pass.

The root causes of excessive compression set are: (1) thermal degradation from sustained exposure above the compound’s rated temperature, (2) over-compression from incorrect groove depth — a groove that produces more than 30% squeeze on a static O-ring accelerates compression set significantly, and (3) wrong compound, specifically using a general-purpose NBR in a fluid or temperature environment that requires FKM or HNBR.

In our supplier qualification program, we have seen suppliers pass initial sample approval with compression set values of 18% and then deliver production batches testing at 28–32%. The trigger is almost always a raw material substitution at the compounder level — a lower-grade carbon black or a different plasticizer system — something that a standard COA listing only hardness and tensile strength will not catch. Incoming compression set spot-testing on production batches is not optional for critical sealing applications.

Chemical Swell and Fluid Compatibility: Measurable Limits and Material Selection #

Chemical swell occurs when the elastomer absorbs the process fluid, causing volume increase, softening, and loss of mechanical properties. Unlike extrusion or compression set, chemical swell is not a specification error — it is a material selection error, and it is entirely preventable.

Volume swell is measured per ASTM International D471: the specimen is immersed in the test fluid at a specified temperature for 70 hours, and volume change is measured. Acceptable limits for static sealing applications:

• Volume swell ≤ 10%: acceptable for most static applications
• Volume swell 10–20%: marginal; may be acceptable with reduced groove fill
• Volume swell > 20%: disqualifying for static seals; the O-ring will extrude from the groove under compression

The most common chemical compatibility errors we see in fluid power applications:

NBR in phosphate ester hydraulic fluids (e.g., Skydrol, Fyrquel): NBR swells catastrophically — volume increase of 40–80% is typical. The correct material is EPDM or butyl rubber. This is a well-documented incompatibility that still appears regularly in procurement specifications because the buyer specified “NBR hydraulic seal” without checking the fluid type.

FKM in ketone-based fluids or low-molecular-weight esters: FKM, despite its chemical resistance reputation, is attacked by MEK, acetone, and some brake fluids. Volume swell of 30–50% has been measured. The correct material for ketone service is EPDM or PTFE-encapsulated seals.

EPDM in petroleum-based oils: EPDM has essentially no resistance to petroleum hydrocarbons. Volume swell exceeds 100% in mineral oil within 24 hours. This is the most common installation error when EPDM seals from water/steam systems are accidentally used in oil-lubricated equipment.

For hydraulic and pneumatic seals in multi-fluid systems, we always recommend requesting a fluid compatibility matrix from the supplier — not just a generic “compatible with hydraulic oil” statement. The specific fluid formulation, operating temperature, and concentration all affect the swell result.

Most Western buyers do not realize that the GB/T China Standards governing elastomer compound testing in China allow immersion test durations and temperatures that differ from ISO Standards 1817 — the Chinese standard may use a shorter immersion period or a lower test temperature, producing a more favorable swell result on the COA. When sourcing from China, always specify the test method explicitly: “ASTM D471, 70h, [fluid name], [temperature]°C” — not just “fluid compatibility test.”

Abrasion, Installation Damage, and Hardness Mismatch: The Preventable Failures #

Abrasion in static O-ring applications is almost always caused by one of two things: reciprocating motion in what was specified as a static application, or contaminated fluid with particulate above 25 microns. For genuinely static applications, abrasion failure indicates the seal is experiencing micro-motion — thermal cycling, pressure pulsation, or vibration — that was not accounted for in the design. The corrective action is either a dynamic seal design (lip seal or U-cup) or a harder compound (Shore A 80–90) with a finer surface finish on the mating hardware (Ra ≤ 0.8 µm for static, Ra ≤ 0.4 µm for dynamic).

Installation damage is the failure mode most often misattributed to material quality. Spiral cuts — a helical cut pattern running around the O-ring cross-section — are caused by the O-ring rotating and sliding simultaneously during installation over a shaft or through a bore. Single-plane cuts are caused by sharp edges on ports, threads, or groove corners. Both are installation failures, not material failures. The corrective action is: chamfer all leading edges to 15–20°, use a proper installation cone or sleeve, and apply the correct lubricant — compatible with both the elastomer and the process fluid. Using petroleum-based grease on an EPDM seal causes immediate chemical attack.

Hardness mismatch is the specification error we see most frequently in Chinese-sourced O-ring procurement. The default specification of Shore A 70 is appropriate for low-to-medium pressure pneumatic and hydraulic applications up to approximately 70 bar with correct groove geometry. It is not appropriate for high-pressure hydraulic systems, high-temperature applications where the compound softens significantly, or applications with large clearance gaps. The corrective action is straightforward: match hardness to the pressure-gap-temperature combination using the clearance gap charts in ISO Standards 3601-4 or the ASTM International D2000 line call system.

Honestly, the specification that procurement teams most often get wrong is not the material grade — it is the hardness grade. A Shore A 70 FKM O-ring in a 200 bar hydraulic system will extrude just as reliably as a Shore A 70 NBR O-ring. The material upgrade to FKM solved the chemical compatibility problem but left the mechanical failure mode intact.

Practical Guidance for Buyers #

When sourcing O-rings from China for fluid power applications, the first specification to request from suppliers is not tensile strength — it is compression set per ASTM International D395 Method B at your operating temperature, with a pass threshold of ≤20% for NBR and ≤15% for FKM. Tensile strength is easy to meet and easy to report accurately; compression set requires proper test equipment and cannot be estimated from hardness alone.

The most common sourcing mistake is accepting a COA that lists only hardness, tensile strength, and elongation — the three parameters that are cheapest to test and least predictive of in-service performance. We have seen batches pass on all three parameters and fail in service within 400 hours because compression set was never tested. Request three consecutive batch COAs before committing to volume, and include compression set and volume swell (per ASTM D471 in your specific process fluid) as mandatory line items.

Before committing to a volume order, require a qualification sample set of at least 25 pieces from a production batch — not from a dedicated sample run — and conduct incoming inspection including Shore A hardness (reject if ±3 points from spec), dimensional check per ISO Standards 3601-1 tolerance class N, and compression set testing. If the supplier cannot provide production-batch samples for qualification, that is itself a disqualifying finding.

Frequently Asked Questions #

Q1: What compression set value should I use as a pass/fail threshold for static O-ring seals?

A: For static applications, reject any batch where compression set exceeds 20% for NBR or 25% for silicone after 70 hours at operating temperature per ASTM International D395 Method B. Above these thresholds, long-term sealing force is insufficient.

Q2: How do I select between NBR, FKM, and EPDM for a hydraulic system?

A: The decision is driven by fluid type and temperature, not by price. NBR covers petroleum-based hydraulic oils to 100°C continuous. FKM is required above 150°C or for synthetic ester and fuel-contact applications. EPDM is mandatory for phosphate ester fluids and water/steam — but will fail immediately in any petroleum oil. Check volume swell per ASTM International D471 in your specific fluid before finalizing material selection.

Q3: What causes spiral cut failure and how do I prevent it?

A: This is where most installation failures originate. Spiral cuts are caused by simultaneous rotation and axial sliding during installation — the O-ring rolls instead of sliding cleanly into the groove. The fix is a 15–20° lead chamfer on all hardware edges, a proper installation sleeve, and a compatible lubricant. It has nothing to do with O-ring material quality.

Q4: What certifications and test documentation should I require from a Chinese O-ring supplier before volume purchase?

A: Require a material compound certification traceable to a specific batch number, compression set results per ASTM International D395B or ISO Standards 815, dimensional inspection report per ISO 3601-1 tolerance class N, and — for food, pharmaceutical, or potable water applications — FDA Guidelines 21 CFR 177.2600 compliance or NSF International 61 certification. A generic “RoHS compliant” statement is not a substitute for compound-specific test data.

Q5: Is a higher Shore A hardness always better for high-pressure applications?

A: No. Shore A 90 reduces extrusion risk at high pressure but increases the risk of groove damage, hardware fretting, and seal leakage at low pressure where the harder compound cannot conform to surface irregularities. Match hardness to the specific pressure-gap combination using the clearance gap tables in ISO Standards 3601-4 — do not default to the hardest available compound.

Published by sinoraw.com Technical Team | Request a sourcing consultation