Blog
O-Ring Extrusion Failure: Causes, Prevention & Gland Design
Jun
O-ring extrusion is the most common failure mode in hydraulic and high-pressure sealing systems, and the most preventable. The physics is straightforward: above a certain pressure threshold, for a given clearance gap and compound hardness, the o-ring will be forced into the gap and damaged. The threshold is calculable. The prevention measures are well-defined. And yet extrusion failures continue to appear in hydraulic cylinders, pneumatic actuators, valve bodies, and high-pressure fittings in predictable patterns that indicate the same design and specification errors being made repeatedly. This guide provides the quantitative engineering reference that the standard “reduce the clearance gap and add backup rings” advice omits: the actual numbers, the selection criteria, and the forensic identification of what went wrong after a failure occurs.
The Mechanism: How O-Ring Extrusion Occurs
When an o-ring is installed in a gland and the system is pressurized, the hydraulic or pneumatic pressure acts on the entire exposed surface of the o-ring. The pressure load attempts to move the o-ring in the direction of lower pressure. In a static face seal with no clearance gap between the mating surfaces, the o-ring has nowhere to move and the seal holds indefinitely at essentially any pressure the hardware can contain.
In a radial seal (a piston seal or rod seal), there is necessarily a diametral clearance gap between the piston and the bore, or between the rod and the rod guide. This gap exists because two concentric cylindrical surfaces cannot be manufactured to zero tolerance and must be able to move relative to each other. The clearance gap is where extrusion begins: the pressurized o-ring sees a low-resistance path at the gap and the elastomer flows into it.
The initial extrusion is not catastrophic. A small amount of elastomer extruding into the clearance gap during each pressurization cycle creates a thin fin of material at the gap edge. As the pressure pulses, the clearance gap opens and closes slightly due to hardware deflection under load. The extruded fin is alternately pinched between the mating hardware surfaces and released. Over many pressure cycles, the pinching action tears small pieces from the fin, creating the characteristic “nibbling” damage pattern visible on the downstream edge of a failed o-ring.
At higher pressures or larger clearance gaps, the extrusion is more aggressive. The o-ring can partially roll into the gap, creating a larger extruded zone that is then pinched more severely. In extreme cases of very high pressure with a large gap and soft compound, the o-ring can extrude completely into the gap in a single pressurization event, tearing the cross-section and causing immediate gross leakage rather than the gradual nibbling failure of moderate extrusion.
The Clearance Gap, Pressure, and Hardness Relationship
The extrusion resistance of an o-ring is determined by the interaction of three variables: the diametral clearance gap, the system pressure, and the compound hardness (Shore A). These three variables define the extrusion envelope: the combination of gap and pressure that a given compound hardness can resist without extruding. The Parker O-Ring Handbook and Eriks O-Ring Handbook both publish extrusion pressure-clearance charts for standard elastomers. The values below are derived from these standard references and represent the maximum allowable clearance gap (total diametral clearance, both sides) for different pressure and hardness combinations.
| System pressure | 70 Shore A compound | 80 Shore A compound | 90 Shore A compound | With PTFE backup ring |
|---|---|---|---|---|
| Up to 500 psi (34 bar) | 0.008″ (0.20mm) | 0.010″ (0.25mm) | 0.013″ (0.33mm) | Gap limit effectively eliminated |
| 500-1,000 psi (34-69 bar) | 0.005″ (0.13mm) | 0.007″ (0.18mm) | 0.010″ (0.25mm) | Gap limit effectively eliminated |
| 1,000-1,500 psi (69-103 bar) | 0.003″ (0.08mm) | 0.005″ (0.13mm) | 0.008″ (0.20mm) | Gap limit effectively eliminated |
| 1,500-3,000 psi (103-207 bar) | Not recommended without backup ring | 0.003″ (0.08mm) | 0.005″ (0.13mm) | Gap limit effectively eliminated |
| 3,000-5,000 psi (207-345 bar) | Not recommended without backup ring | Not recommended without backup ring | 0.003″ (0.08mm) | Gap limit effectively eliminated |
| Above 5,000 psi (345 bar) | Backup ring required regardless of hardness | Backup ring required regardless of hardness | Backup ring required regardless of hardness | Spring-energized PTFE seal preferred above 10,000 psi |
The gap values in this table are total diametral clearance, meaning the sum of the gap on both sides of the o-ring (or the difference between the bore ID and the piston OD). For a piston seal application with a bore diameter of 2.000″ and a piston diameter of 1.995″, the diametral clearance is 0.005″. Whether this is within the safe range depends on the system pressure and the compound hardness selected, as the table shows.
Tolerance Stack: Why the Manufactured Clearance Gap Is Not the Worst-Case Gap
The clearance gap analysis must account for the worst-case combination of manufacturing tolerances on the bore and piston diameters, not just the nominal clearance. A specification that calls for a 2.000″ bore and a 1.995″ piston (nominal diametral clearance 0.005″) with ±0.001″ tolerances on each can produce assemblies where the bore is 2.001″ and the piston is 1.994″, a diametral clearance of 0.007″. Whether this worst-case clearance is within the safe range for the operating pressure and compound hardness must be checked, not the nominal clearance alone.
The calculation is straightforward:
Gap_max = (2.000 + 0.001) – (1.995 – 0.001) = 2.001 – 1.994 = 0.007″
Check 0.007″ against the pressure and hardness table for the application.
This worst-case analysis is frequently skipped during design and only discovered after extrusion failures occur in field assemblies that happen to be at the tolerance extremes. In high-pressure hydraulic cylinders or actuators where the o-ring is the critical sealing element, worst-case clearance gap analysis should be a required step in gland design review.
The tolerance stack problem is compounded by hardware wear in dynamic applications. A hydraulic cylinder piston that starts with a nominal 0.005″ clearance gap will, after thousands of reciprocating cycles with contaminated hydraulic fluid, develop measurable bore and piston surface wear that increases the effective clearance gap. An o-ring specification that is at the margin of the extrusion limit at assembly may be well outside it after 100,000 operating cycles. For high-cycle dynamic applications, the extrusion gap analysis should use a wear-adjusted clearance estimate rather than the as-manufactured nominal.
Temperature Effects on Extrusion Resistance
The clearance gap limits in the table apply at room temperature. Elevated temperature reduces the effective Shore A hardness of rubber compounds, shifting every compound toward a lower extrusion resistance category. A 70 Shore A NBR compound that is within the safe clearance range at 70°F may be effectively equivalent to a 50-55 Shore A compound at 200°F, placing it outside the safe range for the same clearance gap and pressure.
The temperature effect on Shore A hardness varies by compound family. NBR loses hardness relatively rapidly with temperature above 150°F. FKM maintains hardness better at elevated temperature than NBR, which is one reason FKM is preferred for high-temperature hydraulic sealing even where its chemical resistance is not required. HNBR falls between NBR and FKM in high-temperature hardness retention.
For applications where the operating temperature exceeds 150°F, adjust the clearance gap analysis by using an effective hardness 5-10 Shore A points lower than the room-temperature compound specification. If a compound is specified as 70 Shore A for a 200°F application, evaluate the clearance gap limit using the 70 Shore A row at the operating pressure, then confirm with the compound supplier that the compound maintains at least that effective hardness at the operating temperature. If it does not, either increase the compound hardness specification or reduce the clearance gap to compensate.
Spiral Failure: The Dynamic Seal Failure Mode That Mimics Extrusion
Spiral failure, also called twist failure or cork-screw failure, is a dynamic seal failure mode that is frequently confused with extrusion damage because the two can occur together and produce similar-looking damaged o-rings. Understanding the distinction is important because the corrective actions for spiral failure are different from those for extrusion.
The spiral failure mechanism
In a reciprocating rod seal or piston seal, the o-ring must translate axially with the rod or piston while maintaining radial contact with both the gland bore and the piston/rod surface. The o-ring does this by rolling slightly as the rod moves, maintaining its circular cross-section through a combination of rolling and sliding contact.
Spiral failure occurs when the friction forces on the inner diameter (ID) and outer diameter (OD) of the o-ring become unbalanced during a reciprocating stroke. If the friction on the ID is higher than on the OD (or vice versa), the o-ring attempts to roll at different rates on the two surfaces. Instead of rolling as a complete torus, it twists progressively with each stroke until the cross-section is permanently deformed into a helical shape. The helically deformed o-ring no longer seals the full circumference of the gland, and leakage begins.
Visual identification of spiral failure
A spirally failed o-ring shows a regular diagonal cut or crease pattern around the circumference when the o-ring is removed from the gland and inspected at room temperature. The diagonal marks are typically spaced at intervals related to the ID and OD circumferences and the twist rate during operation. The damage pattern differs from extrusion nibbling, which produces irregular bite marks on the downstream edge of the o-ring concentrated at one or two points around the circumference.
Conditions that cause spiral failure
Short stroke length. Reciprocating seals that operate over very short stroke lengths (less than approximately 1.5 times the o-ring cross-section diameter) do not complete enough of a full roll cycle per stroke to distribute twist evenly. The o-ring accumulates twist in one direction with each stroke rather than rolling and reversing. Short stroke hydraulic cylinders and short-travel pneumatic actuators are the most common spiral failure location for this reason.
High friction differential. If the friction on the ID contact surface differs significantly from the friction on the OD contact surface (due to different surface finishes, different lubrication conditions, or asymmetric gland proportions), the roll rate differential that drives spiral twist increases. A rod with a smooth polished surface and a gland bore with a rougher finish creates this condition.
Improper o-ring cross-section selection. A larger cross-section o-ring in a given stroke length is more likely to spiral because the ratio of cross-section diameter to stroke length is higher. For very short-stroke applications, a smaller cross-section o-ring with the same ID may reduce spiral failure risk by reducing the roll arc per stroke.
Inadequate lubrication. Spiral failure is more common in dry or poorly lubricated applications because differential friction increases when lubrication is absent. Pneumatic seals, which operate in dry air with minimal lubrication, are more susceptible to spiral failure than hydraulic seals operating in oil. Ensuring adequate lubrication of both the ID and OD contact surfaces at installation reduces spiral failure risk.
Forensic Identification: What the Failed O-Ring Tells You
A failed o-ring removed from its gland contains diagnostic information about the root cause of the failure. The following identifies the most common damage patterns and their specific causes, allowing the engineer to identify the correct corrective action rather than replacing with the same specification and experiencing the same failure.
What it looks like: Small irregular pieces removed from the low-pressure edge of the o-ring, concentrated at one to three locations around the circumference. The remainder of the o-ring cross-section is undamaged. Root cause: Localized clearance gap larger than the extrusion limit, often due to bore out-of-roundness or eccentric assembly. The gap is larger at specific angular positions than the nominal clearance calculation would suggest. Corrective action: Verify bore roundness and assembly concentricity before assuming a clearance gap problem. If bore roundness is within tolerance, add a backup ring or increase compound hardness.
What it looks like: Small pieces removed from the downstream edge of the o-ring uniformly around the full circumference, with the upstream face undamaged. The o-ring is thinner on the low-pressure side. Root cause: The diametral clearance gap systematically exceeds the extrusion limit for the operating pressure and compound hardness. This is the classic extrusion failure pattern. Corrective action: Evaluate the clearance gap against the pressure/hardness table. Either tighten the hardware clearance to within the limit, increase compound hardness, or add a PTFE backup ring on the downstream side.
What it looks like: The o-ring cross-section is significantly flattened. Material has extruded on both the high-pressure and low-pressure sides of the gland. The o-ring may have frilled or feathered edges on both parting faces. Root cause: The o-ring has swelled thermally or chemically to the point where the gland fill exceeds 85-90%, and the increased volume has no room to go except into the clearance gaps and the parting line. This is thermal/chemical swell extrusion, not pressure extrusion. Corrective action: Check chemical compatibility of the compound with the fluid. Check fluid temperature against the compound’s service limit. If either is marginal, change the compound. The gland dimensions may need to be revised to allow for the thermal expansion of the compound at operating temperature.
What it looks like: The o-ring does not lie flat when removed. Diagonal surface creases are visible, spaced at regular intervals around the circumference. The cross-section may appear intact with no material removed. Root cause: Spiral failure from unbalanced friction in a reciprocating dynamic seal. Corrective action: Reduce stroke-to-cross-section ratio by using a smaller cross-section o-ring, improve lubrication uniformity on both ID and OD contact surfaces, or switch to a D-ring or quad-ring profile that resists twist by design.
What it looks like: The o-ring is hard and brittle. Cracks appear on the outer surface. The cross-section has not been removed by extrusion and the material has zero elastic recovery. Root cause: The o-ring has been operating above its maximum continuous service temperature, causing crosslink degradation and hardening. This is a thermal failure that looks like an extrusion failure in the field because the hardened seal is no longer sealing, but the failure mechanism and corrective action are completely different. Corrective action: Confirm the fluid temperature at the seal location. If it exceeds the compound’s rated service temperature, upgrade to a higher-temperature compound: HNBR for hot oil above 250°F, FKM above 300°F.
Backup Ring Selection: When to Use One and Which Type
Backup rings are the primary engineering solution for applications where the clearance gap cannot be tightened sufficiently to prevent extrusion. A backup ring is a harder, non-elastomeric ring installed in the gland on the low-pressure side of the o-ring. Its function is to bridge the clearance gap, preventing the o-ring elastomer from extruding into the gap while allowing the o-ring to maintain its sealing function against the pressurized fluid.
The standard backup ring material. PTFE is chemically inert to virtually all hydraulic fluids, fuels, and solvents. It has low friction against metal surfaces, which is important for dynamic seals where the backup ring must slide with the rod or piston. PTFE backup rings are available in solid continuous form for static seals and in spiral-cut or bias-cut form for installation over pistons or into gland bores without disassembly of the hardware. The spiral-cut form has a gap that should be oriented away from the pressure source to minimize leakage through the cut.
Nylon (polyamide) backup rings are stiffer than PTFE and less compressible, which makes them effective at bridging larger clearance gaps. They are preferred where the clearance gap varies significantly due to hardware tolerance or wear, because nylon’s higher modulus prevents it from being displaced into the gap under pressure. The limitation is chemical resistance: nylon is attacked by strong acids and bases and has limited resistance to some hydraulic fluids. Confirm fluid compatibility before specifying nylon backup rings in chemical process applications.
For unidirectional pressure (pressure from one side only): install one backup ring on the low-pressure side of the o-ring. The o-ring faces the pressure and the backup ring prevents it from extruding into the downstream gap. For bidirectional pressure (pressure can come from either side): install one backup ring on each side of the o-ring. The two-backup-ring configuration prevents extrusion regardless of pressure direction. Both configurations require widening the gland groove to accommodate the backup ring(s) in addition to the o-ring. Confirm total groove width against the o-ring and backup ring supplier’s combined dimensional requirements.
Design Review Checklist for Extrusion Prevention
At the design stage
1. Calculate worst-case diametral clearance gap using maximum bore tolerance and minimum piston/rod tolerance. Do not use nominal clearance for extrusion analysis.
2. Check worst-case gap against the pressure-hardness table for the compound specified. If the gap exceeds the safe limit, either tighten the tolerance specification, increase compound hardness, or add a backup ring.
3. Confirm compound hardness at operating temperature with the compound supplier. A 70 Shore A compound at room temperature may be effectively 58-62 Shore A at 200°F operating temperature.
4. Verify gland dimensions against AS568 or ISO 3601 standard gland tables for the o-ring cross-section and application type (static face seal, piston seal, rod seal). Use the Gland Calculator to confirm squeeze, gland fill, and groove dimensions.
For dynamic applications
5. Evaluate spiral failure risk for reciprocating seals with stroke lengths less than 1.5× the o-ring cross-section diameter. Consider smaller cross-section, D-ring profile, or quad-ring profile for short-stroke applications.
6. Specify surface finish for rod and bore contact surfaces. Standard hydraulic seal surface finish: 8-16 µin Ra for dynamic contact surfaces, 16-32 µin Ra for static contact surfaces. Inadequate surface finish increases friction differential that drives spiral failure.
7. Specify lubrication at installation for dynamic seals. Lightly coat the o-ring, rod, and bore with compatible lubricant (system fluid for hydraulic applications) during installation to ensure uniform friction from first stroke.
8. Evaluate backup ring requirement before finalizing the gland design. Adding a backup ring requires widening the gland. This change is far less expensive to make during design than after tooling is completed.