An o-ring groove is sized to squeeze the ring a set amount: for a 0.139 in (3.53 mm) cross-section Apple Rubber gives a dynamic (piston or rod) gland depth of 0.119-0.123 in with a 0.185 in groove (9-17% squeeze), a static radial gland of 0.110-0.116 in (14-23%) and a static face gland of 0.100-0.108 in (20-30%) with a 0.195 in groove. The tables below give the same figures for every AS568 cross-section from 0.020 to 0.275 in, from Apple Rubber's Seal Design Guide; SAE AS5857 (static) and AS4716 (dynamic) standardise glands too, and other seal makers publish their own tables, so pick one system and use it consistently. All dimensions are in inches; ring sizes by dash number are on the AS568 O-Ring Size Chart, and the Unit Converter converts them to millimetres.
Dynamic (reciprocating) glands: piston and rod seals
| O-ring CS (in) | Gland depth (in) | Squeeze (in) | Squeeze (%) | Diametral clearance max. (in) | Groove width, no back-up (in) | Groove width, 1 back-up (in) | Groove width, 2 back-ups (in) |
|---|---|---|---|---|---|---|---|
| .040 | .031/.033 | .004/.012 | 11-28 | .004 | .063 | — | — |
| .050 | .039/.041 | .006/.014 | 13-26 | .004 | .073 | — | — |
| .060 | .047/.049 | .008/.016 | 14-25 | .004 | .084 | — | — |
| .070 | .055/.057 | .010/.018 | 15-25 | .004 | .095 | .150 | .208 |
| .103 | .087/.090 | .010/.019 | 10-18 | .005 | .145 | .187 | .249 |
| .139 | .119/.123 | .012/.024 | 9-17 | .006 | .185 | .222 | .301 |
| .210 | .183/.188 | .017/.032 | 8.5-15 | .006 | .285 | .338 | .428 |
| .275 | .234/.240 | .029/.047 | 10.5-17 | .007 | .375 | .440 | .579 |
Static glands: radial (bore or plug) and axial (face) seals
| O-ring CS (in) | Radial gland depth (in) | Radial squeeze (%) | Axial (face) gland depth (in) | Axial squeeze (%) | Diametral clearance max. (in) | Groove width, no back-up (in) | Groove width, 1 back-up (in) | Groove width, 2 back-ups (in) |
|---|---|---|---|---|---|---|---|---|
| .020 | .013-.014 | 22-41 | .013-.014 | 22-41 | .002 | .035 | — | — |
| .030 | .020-.022 | 19-39 | .020-.022 | 19-39 | .003 | .045 | — | — |
| .040 | .027-.030 | 19-37 | .027-.030 | 19-37 | .003 | .060 | — | — |
| .050 | .035-.039 | 17-34 | .034-.038 | 19-36 | .004 | .075 | — | — |
| .060 | .042-.047 | 18-33 | .042-.046 | 19-33 | .004 | .090 | — | — |
| .070 | .050-.055 | 18-32 | .049-.054 | 19-33 | .004 | .105 | .150 | .208 |
| .103 | .080-.086 | 14-25 | .075-.081 | 19-29 | .005 | .146 | .182 | .244 |
| .139 | .110-.116 | 14-23 | .100-.108 | 20-30 | .006 | .195 | .217 | .296 |
| .210 | .170-.176 | 14-21 | .155-.165 | 20-28 | .006 | .280 | .333 | .423 |
| .275 | .225-.235 | 13-20 | .205-.215 | 20-27 | .007 | .350 | .435 | .574 |
Groove radius and eccentricity
| O-ring CS (in) | Dynamic squeeze (in) | Dynamic groove radius (in) | Dynamic eccentricity max. (in) | Static squeeze radial / axial (in) | Static groove radius (in) | Static eccentricity max. (in) |
|---|---|---|---|---|---|---|
| .020 | — | — | — | .004-.009 / .004-.009 | — | .0015 |
| .030 | — | — | — | .005-.013 / .005-.013 | — | .0015 |
| .040 | .004/.012 | .005-.008 | .002 | .007-.016 / .007-.016 | .005-.008 | .002 |
| .050 | .006/.014 | .005-.008 | .002 | .008-.018 / .009-.019 | .005-.008 | .002 |
| .060 | .008/.016 | .005-.008 | .002 | .010-.021 / .011-.021 | .005-.008 | .002 |
| .070 | .010/.018 | .005-.015 | .002 | .012-.023 / .013-.024 | .005-.015 | .002 |
| .103 | .010/.019 | .005-.020 | .003 | .014-.026 / .019-.031 | .005-.020 | .003 |
| .139 | .012/.024 | .005-.030 | .004 | .019-.033 / .027-.043 | .005-.030 | .004 |
| .210 | .017/.032 | .005-.050 | .006 | .029-.045 / .040-.060 | .005-.050 | .006 |
| .275 | .029/.047 | .005-.060 | .008 | .034-.056 / .054-.076 | .005-.060 | .008 |
How to read the gland tables
- Gland depth vs groove depth. Groove depth is what you machine into one part. Gland depth is groove depth plus the diametral clearance between the parts, and it is the gland depth that sets the squeeze (Apple Rubber, section 3). For a face seal the clearance is zero, so the two are the same.
- Groove width is ±0.005 in in both tables. "Diametral clearance max." is the largest total gap between bore and piston (or rod and bore) the table allows.
- Static radial values assume an application pressure limit of 1,500 psi; above that, check extrusion (below) or ask the maker.
- 0.020 in section: Apple Rubber recommends ordering the ring with a tighter ±0.002 in section tolerance.
- Squeeze % is the range from the smallest ring in the deepest gland to the largest ring in the shallowest gland, using the AS568 section tolerances (for example 0.139 ± 0.004 in a 0.119-0.123 in gland gives 0.012-0.024 in, 9-17%).
- Different makers, different numbers. These are Apple Rubber's general guidelines ("a compilation of data from various sources"), not values copied from SAE AS5857 or AS4716. The old version of this page mixed tables from several sources; it now uses one.
Design rules from the same guide
| Squeeze | Roughly 10-40% for static seals and no more than 30% for dynamic seals; less squeeze lowers friction in a dynamic seal but can leak at low pressure. |
|---|---|
| Groove fill | The maximum o-ring volume should not exceed 90% of the minimum gland void (worst-case tolerances). For a 45° crush seal, keep the ring volume within 95% of the gland void. |
| Containment | At least 75% of the cross-section should sit inside the groove so the ring does not roll or extrude out. |
| Stretch | 1-5% on the ID once installed, 2% ideal; not more than 100% during installation. Stretch % = (groove diameter ÷ o-ring ID − 1) × 100. |
| Groove walls | Rectangular, straight-sided grooves are best; 5° sloping sides are acceptable up to 1,500 psi. Break all sharp corners by at least 0.005 in. |
| Static finish | 32 µin RMS preferred (64-128 µin can be tolerated); 32 µin for liquids, 16 µin for vacuum and gases. |
| Dynamic finish | 10-20 µin on the surface the seal slides on; a highly polished surface is not wanted because it will not hold lubricant. Honed, burnished or hard-chrome surfaces are best; aluminium and brass are generally not suitable for dynamic seals. |
| Rotary shafts | Shaft 16 µin RMS and within 0.0005 in TIR, gland 32 µin RMS; no stretch over the shaft; squeeze as little as 0.002 in by using a ring with an OD about 5% larger than the gland; not recommended below −40 °F or above +250 °F. |
| Ports | Do not pass a reciprocating o-ring over ports: it nibbles and wears. |
Extrusion and back-up rings
- Whether a ring extrudes depends on the pressure, the diametral clearance gap and the ring hardness (Apple Rubber, Illustration 5.2). Worked example from the guide: 70 Shore A at 1,000 psi with a 0.016 in gap extrudes; 80 Shore A at the same pressure and gap is acceptable.
- Ways to stop extrusion: a harder ring, a back-up ring to close the gap, a smaller clearance, or lower system pressure.
- Two back-up rings (one on each side) are preferred, so the gap is blocked whichever way the pressure acts and the ring cannot be fitted the wrong way round. With only one, put it on the low-pressure side.
- Removed from the old page: a fixed rule of "back-up rings above 1,500 psi dynamic or 3,000 psi static". No source states it as a rule; use the pressure-clearance-hardness check above or the maker's extrusion chart.
Compound temperature ranges (standard compounds)
| Material (Apple code) | Standard compound range (°F) | Standard compound range (°C) | Hardness range (Shore A) |
|---|---|---|---|
| Nitrile, NBR / Buna-N (BN) | −40 to +257 | −40 to +125 | 40 to 90 |
| Hydrogenated nitrile, HNBR (HN, ZT) | −30 to +300 (dry heat) | −34 to +149 | 50 to 90 |
| Fluorocarbon, FKM / Viton (VT) | −13 to +446 | −25 to +230 | 45 to 90 |
| Ethylene propylene, EPDM (EP) | −76 to +275 | −60 to +135 | 40 to 95 |
| Silicone, VMQ (SL) | −85 to +400 | −65 to +204 | 5 to 80 |
| Perfluoroelastomer, FFKM (KA) | −13 to +600 | −25 to +316 | 65 to 90 |
| Millable polyurethane, AU / EU (MP) | −30 to +175 | −34 to +79 | 40 to 90 |
Compound limits that matter for gland design
- Nitrile: the most widely used and economical seal elastomer, for oil-resistant uses of all types; attacked by small amounts of ozone.
- FKM: high temperature with wide chemical resistance and low outgassing (good vacuum seals); not for ketones, amines, low-molecular-weight esters and ethers, Skydrol fluids, or where good low-temperature flexibility is needed.
- EPDM: weather, brake-fluid and cooling-system use; no resistance to hydrocarbon fluids.
- Silicone: static seals at extreme temperatures; low abrasion and tear resistance and high friction rule it out of many dynamic seals, and it is highly permeable to gases.
- FFKM: resists virtually all chemicals but can swell in some fluorinated solvents and must not meet molten or gaseous alkali metals.
- Polyurethane: high hydraulic pressures and wear; softens and loses strength at elevated temperature unless specially compounded.
- Temperatures are Apple Rubber's standard-compound ranges; special compounds go further (e.g. nitrile to −67 °F, silicone to −148 °F). °C values are converted on this page and rounded. Check chemical compatibility with the maker before choosing.
Corrections to the previous version
- Gland depths and widths replaced with one maker's tables (Apple Rubber). Example: 0.070 in static face depth was 0.050-0.057 in, now 0.049-0.054 in (axial); dynamic widths are now single ±0.005 in values (0.095 in for 0.070 in section).
- Squeeze bands per seal type ("20-30% face, 15-25% radial, 8-16% dynamic, 0-8% rotary") replaced with the per-section ranges in the tables.
- The "AS568 common sizes" list had wrong inside diameters (e.g. -022 was 1.114 in, actually 0.989 in; -226 was 2.239 in, actually 1.984 in; -326 was 2.000 in, actually 1.600 in). The full, checked list is now on the AS568 O-Ring Size Chart.
- Removed as unsourced: "groove fill 60-85%", "max 3% OD compression", the back-up-ring pressure thresholds, the JIS B 2401 examples and the vacuum squeeze rule. Compound temperature ranges corrected to the source (e.g. EPDM was −60 to +300 °F, now −76 to +275 °F).
Sources
- Apple Rubber Products, Inc., Seal Design Guide, 2020 (www.applerubber.com/src/pdf/seal-design-guide.pdf). Used for: AS568 sizes and tolerances (section 9), gland design tables (section 4, p. 20), stretch, squeeze, surface finish, extrusion and material temperature ranges.
- Pages used in Apple Rubber's guide: section 3 (stretch, gland vs groove depth), section 4 p. 20 (Table A dynamic and the static gland table, surface finish, machining, rotary), section 5 (extrusion limit, back-up rings), section 6 (material temperature ranges and limitations).
- SAE International, AS5857 Gland Design, O-Ring and Other Seals for Static Applications and AS4716 Gland Design, O-Ring and Other Elastomeric Seals (paywalled). Named for reference only; no values on this page come from them.
Frequently asked questions
How deep should an o-ring groove be?
Deep enough to squeeze the ring by the right amount. For a 0.139 in (3.53 mm, -2xx) ring Apple Rubber gives a gland depth of 0.119 to 0.123 in for a dynamic piston or rod seal (9-17% squeeze), 0.110 to 0.116 in for a static radial seal (14-23%) and 0.100 to 0.108 in for a static face seal (20-30%). Gland depth includes the diametral clearance; the machined groove is shallower by that gap.
How wide should an o-ring groove be?
For a 0.139 in ring with no back-up ring, Apple Rubber gives 0.185 in (±0.005) for a dynamic gland and 0.195 in for a static gland. One back-up ring needs 0.222 in (dynamic) or 0.217 in (static); two back-ups need 0.301 in or 0.296 in.
How much should an o-ring be squeezed?
Apple Rubber's general guideline is roughly 10-40% for static seals and no more than 30% for dynamic seals. Its gland tables work out to 8.5-28% for dynamic glands (depending on section) and 13-41% for static glands; face (axial) glands run tighter than radial ones, e.g. 20-30% against 14-23% for a 0.139 in ring.
How much can you stretch an o-ring?
In the installed groove, 1 to 5% on the inside diameter, with 2% ideal; more than 5% thins the section and speeds ageing. During installation over a piston, do not stretch it more than 100%, and stretch it evenly. On a rotary shaft use no stretch at all (Apple Rubber).
When do you need a back-up ring?
When the pressure and the clearance gap would extrude the ring for its hardness. In Apple Rubber's example a 70 Shore A ring at 1,000 psi with a 0.016 in diametral clearance falls in the extrusion zone, while an 80 Shore A ring at the same pressure and gap does not. The fixes are a harder ring, a back-up ring, a smaller gap or lower pressure; two back-ups (one each side) are preferred, and a single back-up goes on the low-pressure side.