Design Guide › Seals
Seals design rules
36 of the guide’s 1374 rules carry the seals tag.
Rules for vacuum seals: O-ring gland dimensions and surface finish, gasket grooves, insulator sleeves on feedthroughs, and double-gasket flanges with pump-out for hard-to-seal joints.
Each rule keeps its formula where the source gives one, a verbatim quote, a page-level
citation, and a stable identifier (dg-NNNN) that resolves here and on the
all-in-one guide.
To combine this tag with another (rules carrying both), use the filterable view: /design-guide/?domain=seals and add a second chip. Related domains, by how often they share a rule with this one: Vacuum (26), Fabrication (18), Materials (4), Vacuum chamber (2), Beam measurement (1).
Verify before use. Every rule here is a source extract in the vocabulary of the editorial methodology — faithful to its cited page, not an independently validated engineering requirement. Re-read any rule that drives a real design decision at the cited page before committing metal, money, or high voltage to it.
-
Seal flanges with a gasket in a machined groove, gasket ~50 percent thicker than groove depth; 1/4-in gaskets suffice for even the largest seals; use neoprene (low vapor pressure, grease-tolerant) and lay a thin copper-foil strip half-over the gasket where RF current must cross the joint.
gasket thickness ~ 1.5x groove depth; 1/4-in section adequate for largest flangesSource, quote & tabletop applicability
about 50 per cent thicker than the depth of the groove to allow for compression ... 1/4-in. gaskets have proved adequate for even the largest seals ... Conductivity for rf currents through such a seal can be assured by half-covering the gasket with a thin copper-foil strip.
Livingston & Blewett, Particle Accelerators (1962) — p. 199-201
Tabletop: Directly usable rules for the reference machine's lid and port seals; the copper-foil RF bridge over elastomer joints prevents mysterious Q loss and local heating.
-
Insulate the dee support stem by slipping a glass (pyrex) sleeve completely over it from the dee edge to at least 2 inches beyond the vacuum seal.
insulating sleeve extends >= 2 in beyond the sealSource, quote & tabletop applicability
slipping a 1/4 in. pyrex tube completely over the 3/16 in. copper dee support rod from the dee edge to at least 2 inches beyond the seal
Wouters, General Recommendations for Design of Small Cyclotrons — UCRL-476 (1949) — p. 7
Tabletop: Directly applicable at 1.3 kV and above: give the stem a continuous insulating sleeve with generous creepage past the grounded feedthrough.
-
Machine face-seal grooves for vacuum to the Parker chart: for a 1/8 in. (0.139) cross-section ring use gland depth 0.101-0.107, squeeze 20-30%, vacuum groove width 0.158-0.164, groove radius 0.010-0.025.
W=.139+/-.004: L=.101-.107, squeeze .028-.042 (20-30%), G(vacuum)=.158-.164, R=.010-.025; W=.210: L=.152-.162, G=.239-.244; W=.275: L=.201-.211, G=.309-.314Source, quote & tabletop applicability
201 through 284 / 1/8 / .139 +/-.004 / .101 to .107 / .028 to .042 / 20 to 30 / .177 to .187 / .158 to .164 / .010 to .025
Parker Hannifin, O-Ring Handbook — Design Chart 4-3: O-Ring Face Seal Glands — p. 1
Tabletop: Directly hands the mill the numbers for every lid and port groove on the next machine's chamber; note the vacuum groove width is narrower than the liquid-service column.
-
Finish O-ring sealing faces to 16 RMS for vacuum and gas service (32 RMS is acceptable only for liquids), with groove sidewalls at 63 RMS and a 0-5 degree sidewall angle.
sealing face 16 RMS (vacuum/gas), 32 RMS (liquid); groove walls 63 RMS; sidewall taper 0-5 deg; break corners approx .005 radSource, quote & tabletop applicability
Surface finish X: 32 for liquids, 16 for vacuum and gases. Finishes are RMS values.
Parker Hannifin, O-Ring Handbook — Design Chart 4-3: O-Ring Face Seal Glands — p. 1
Tabletop: A fly-cut or turned finish on the chamber lid seat should be specified/checked to 16 RMS; a rougher face is a common reason a 1e-6 Torr system stalls in the 1e-5s.
-
Locate a face-seal groove by the diameter the pressure pushes the ring toward: for internal (outward) pressure dimension groove OD = mean O-ring OD; for external pressure (vacuum chambers) dimension groove ID = mean O-ring ID, tolerance +1% of ID but not more than +0.060.
external pressure (vacuum): H_i = mean O-ring ID, tol +1% ID (max +0.060); internal pressure: H_o = mean O-ring OD, tol -1% OD (max -0.060)Source, quote & tabletop applicability
For External Pressure (inward pressure direction) dimension the groove by its inside diameter (Hi) and width: (H)i = Mean I.D. of O-ring
Parker Hannifin, O-Ring Handbook — Design Chart 4-3: O-Ring Face Seal Glands — p. 1
Tabletop: For an evacuated chamber atmospheric pressure pushes the ring inward, so the groove ID (not OD) is the controlled dimension when laying out the lid groove.
-
Use a double-gasket seal with a pump-out connection between gaskets on large or troublesome flanges so tightness can be tested quickly and a leak can be pumped away in service.
Source, quote & tabletop applicability
A double-gasket seal is frequently used with a pump-out connection to the space between gaskets. This arrangement makes it possible to test the seal for vacuum-tightness quickly and with certainty.
Livingston & Blewett, Particle Accelerators (1962) — p. 200
Tabletop: Worth adopting on a next machine's main lid: a guard-vacuum groove turns the worst leak hunt into a valve twist.
-
For static vacuum seals use face or dovetail grooves with heavy squeeze; increasing squeeze from 15% to 50% reduces helium leak rate dramatically, and above ~30% squeeze vacuum grease adds little further benefit.
squeeze 15% -> 30% -> 50% gives steeply decreasing He leak rate; grease benefit large at 15%, small at 30%, undetectable at 50%Source, quote & tabletop applicability
increasing the squeeze reduced the leak rate dramatically... at 50% squeeze the beneficial effect of the grease was not detectable.
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 3-4
Tabletop: Justifies cutting a next machine's grooves at the deep end of the squeeze range (25-30%) rather than greasing the rings harder; grease is a crutch for light squeeze.
-
Estimate O-ring permeation leak rate with L = 0.7*F*D*P*Q*(1-S)^2, where F is gas permeability of the elastomer, D ring ID in inches, P differential in psi, Q a squeeze/lubrication factor (~1.35 dry at 20% squeeze), S fractional squeeze.
L(std cc/s) = 0.7 F D P Q (1-S)^2; F in std cc-cm/(cm^2 s bar), D in inches, P in psiSource, quote & tabletop applicability
L = .7FDPQ(1-S)2 where: L = Approximate leak rate of the seal, std. cc/sec.
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 5
Tabletop: Lets the builder compute the permeation floor of their 10-inch Viton lid seal and check whether O-ring permeation, not leaks, sets their ultimate pressure.
-
Avoid tool marks perpendicular to the O-ring sealing line; the ideal vacuum-flange finish has a circular lay (concentric with the ring), since a radial scratch is a built-in leak path.
Source, quote & tabletop applicability
care being taken to insure that there are no machine or tool marks perpendicular to the seal... The ideal surface finish for any vacuum seal flange has a circular lay
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 5
Tabletop: Face the lid seat on a lathe (concentric lay) rather than fly-cutting or hand-sanding radially; never sand a groove crosswise to remove a blemish.
-
Vacuum weight loss at 1e-6 Torr over two weeks is under 0.2% for butyl (0.18%), fluorocarbon (0.07-0.09%) and low-loss silicone, versus 1-3.5% for nitrile - avoid nitrile near optics, insulators, or RF surfaces.
% weight loss, 336 h @ ~1e-6 Torr, 21 C: butyl 0.18, neoprene 0.13, fluorocarbon 0.07-0.09, silicone 0.03-0.31, EPDM 0.39-0.92, nitrile 1.06-3.45, polyurethane 1.29Source, quote & tabletop applicability
Vacuum Level: Approximately 1 x 10-6 torr ... Butyl .18 ... Nitrile 1.06 ... Nitrile 3.45 ... Fluorocarbon .07
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 6
Tabletop: At exactly the reference machine's operating pressure: cheap Buna-N rings will slowly deposit oily film on feedthrough insulators and dee stems; Viton's 0.1%-class loss is why it is worth the money.
-
Pick low-permeability elastomers for vacuum: butyl is best (He permeability 6.5e-8 std cc-cm/cm2-s-bar), Viton fluorocarbon is close (12.7e-8) and adds 205 C capability, while silicone is ~37x worse (238e-8) and should be avoided as a vacuum seal.
He permeability x1e-8 std cc-cm/cm2-s-bar @77F: butyl 6.5, neoprene 6.5, nitrile 8.0, fluorocarbon 12.7, EPDM 19.7, fluorosilicone 143, silicone 238Source, quote & tabletop applicability
Butyl 6.5 @ 77F ... Fluorocarbon 12.7 @ 77F ... Silicone 238.0 @ 77F
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 7
Tabletop: Confirms Viton is a sound choice at 1e-6 Torr; if helium leak checking becomes routine, remember He walks through silicone and fluorosilicone.
-
Compress Viton O-rings 15-20% of chord diameter (Kalrez max 12%); aim for initial contact pressure of at least 13 kg/cm2 for 60-75 Shore gaskets - a 3.2 mm ring at 75 Shore develops about 2.7 kg per cm of ring length.
compression 15-20% (Viton), <=12% (Kalrez); min contact pressure ~13 kg/cm2; seal force ~2.7 kg/cm for 0.318 cm ring @75 ShoreSource, quote & tabletop applicability
O-rings are typically compressed 15-20% of their diameter... the general criterion for high vacuum sealing to be a minimum initial contact pressure of 13 kg/cm2
O'Hanlon, A User's Guide to Vacuum Technology, 3rd ed. (2003) — p. 337-338
Tabletop: The 2.7 kg/cm figure sizes the lid bolting: a 10-inch-circumference seal needs on the order of 200+ kg of clamping just for the ring, before atmospheric load helps.
-
An unbaked Viton O-ring outgasses ~1e-3 Pa-m/s initially; a 4-h 150 C vacuum bake plus 12 h pumping drops it to 4e-7 Pa-m/s (2500x), but re-exposure to air reloads it with water.
Viton: 1e-3 Pa-m/s unbaked -> 4e-7 Pa-m/s after 4 h @150C + 12 h pumping; solvent washing is ineffectiveSource, quote & tabletop applicability
An unbaked Viton O-ring will have an initial outgassing rate of 10-3 Pa-m/s... After a 4-h bake at 150C and 12 h of pumping, this value is reduced to 4x10-7 Pa-m/s.
O'Hanlon, A User's Guide to Vacuum Technology, 3rd ed. (2003) — p. 340
Tabletop: Pre-baking the Viton rings in a small vacuum oven before assembly is one of the cheapest order-of-magnitude improvements available to a diffusion-pumped 1e-6 Torr system.
-
Do not grease static elastomer seals: grease traps gas pockets that release as pressure bursts; if a scratched main-door flange forces it, apply the thinnest possible film with a lint-free cloth, and always wear gloves since finger oils have high vapor pressure.
Source, quote & tabletop applicability
Grease is not needed to make a static seal between an elastomer and a metal surface. It will cause pressure bursts as trapped gas pockets are released.
O'Hanlon, A User's Guide to Vacuum Technology, 3rd ed. (2003) — p. 340
Tabletop: Counters the amateur habit of greasing everything; on a clean 16 RMS seat with proper squeeze, dry Viton seals better and cleaner.
-
Helium permeates a typical Viton O-ring in about 20 minutes, so during MSLD leak checking of an elastomer-sealed system the He background creeps up and won't fall until the gaskets degas - take a break rather than chase phantom leaks, and never leak check during bakeout.
He permeation time through Viton gasket ~20 min at room temperature; much faster hotSource, quote & tabletop applicability
The permeation time is about 20 min for a typical Viton O-ring... First, do not attempt to leak check the system during baking.
O'Hanlon, A User's Guide to Vacuum Technology, 3rd ed. (2003) — p. 469-470
Tabletop: On an all-Viton chamber, spray briefly and wait; a slowly rising He signal minutes after spraying is gasket permeation, not a leak at the last joint sprayed.
-
Elastomer-sealed flange systems (ANSI/ISO/KF) are realistically good to ~1e-6 Torr (rated 1e-8) and limited to ~150 C bakes; if a joint must ever be baked hotter or hold UHV, design in a metal seal (Conflat copper 300+ C) from the start.
elastomer flanges: rated 1e-8 Torr, better suited to 1e-6 Torr, 150 C max; metal seals (CF/VATSEAL) bakeable to 300 CSource, quote & tabletop applicability
Vacuum rated to 1 x 10-8 Torr (better suited to 1 x 10-6 Torr). Temperature rating is dependent on which elastomer o-ring is used (usually 150C)
Bertolini, Accelerator Vacuum and Mechanical Engineering — USPAS course, UCRL-MI-201847 (2004) — p. 443-446
Tabletop: Matches the reference machine's observed 1e-6 range with Viton - it is the natural floor of an all-elastomer system; a CF port or two on a next machine (gauge, RGA) buys headroom cheaply.
-
O-ring seal design for accelerator vacuum: prefer face seals, use as heavy a squeeze as possible, consider lubrication only when heavy squeeze is impossible, expect heavy flange construction to react the squeeze, and use two O-rings with a guard vacuum between them to drastically cut permeation.
guard vacuum example: 760 Torr across 1st ring reduced to 1e-2 Torr across 2nd ring (DARHT-II: 15 mTorr guard, 5e-8 Torr design pressure)Source, quote & tabletop applicability
Face-type o-ring seals are recommended. Use as heavy a squeeze as possible... Two o-rings in series can drastically reduce permeation.
Bertolini, Accelerator Vacuum and Mechanical Engineering — USPAS course, UCRL-MI-201847 (2004) — p. 71-74
Tabletop: If a next machine targets below ~5e-7 Torr with elastomers, a double O-ring lid groove pumped by the existing roughing pump (guard at ~15 mTorr) removes the permeation floor for the cost of one extra groove and a hose barb.
-
Vacuum-weld discipline: make seam welds continuous and only on the atmosphere side, and stagger-weld internal bracing, so trapped volumes (virtual leaks) cannot form.
Source, quote & tabletop applicability
Seam welds are continuous, with welding only on the atmosphere side. The internal braces are stagger welded to keep virtual leaks at a minimum.
Ramler & Parker, The Argonne 60-Inch Cyclotron — ANL-5907 (1959) — p. 7
Tabletop: Standard practice worth enforcing on any welded chamber or fitting on a next machine: no vacuum-side seal welds, no closed pockets.
-
Match the chamber to the magnet: a 2.54 cm thick aluminium ring of 9.9 cm outer / 8.5 cm inner radius with 0.65 cm lids, ten KF-16 ports epoxied in at equal angles, and a Viton O-ring groove in the lids gives a workable 2e-6 Torr tabletop chamber.
wall ring 2.54 cm thick, r_out 9.9 cm, r_in 8.5 cm; lids 0.65 cm; 10 x KF-16; Viton O-ring; base 2e-6 TorrSource, quote & tabletop applicability
A 2.54 cm thick ring with an outer radius of 9.9 cm and an inner radius of 8.5 cm was milled from 6061 T6 aluminium ... Ports were made in the chamber wall using ten KF-16 flanges, which were secured using Hysol Loctite 1C vacuum epoxy
Tabletop: A complete, copyable chamber spec for an 8-inch-pole machine, including the epoxy-in-flange trick that avoids welding.
-
Seal large flanges with a continuous square-section rubber gasket in a groove sized so the metal faces land metal-to-metal; the metal contact gives alignment and the gasket cannot be over-crushed.
groove volume >= gasket volume; metal-to-metal closureSource, quote & tabletop applicability
continuous square rubber gaskets located in grooves in the faceplates of sufficient cross section to accommodate the entire gasket under pressure. The resulting metal-to-metal contact ... has proved very satisfactory.
Livingston & Boch, The Oak Ridge 86-Inch Cyclotron — ORNL-1196, OSTI 4357145 (1952) — p. 41
Tabletop: Directly applicable to a next machine's chamber lids and faceplates; grooved captive gaskets beat flat sheet gaskets for repeatable sealing and alignment.
-
For vacuum service pick elastomers on three axes - low gas permeability (butyl best, fluorocarbon good, silicone/fluorosilicone worst), low vacuum weight loss, and good compression-set resistance - and use up to 40% squeeze with a correspondingly wider groove.
vacuum squeeze up to 40% with increased groove width; postcure Viton/silicone compounds to drive off volatiles before serviceSource, quote & tabletop applicability
Employing a seal squeeze of up to 40% inhibits media flow through the seal... because of the decreased groove depth, increased groove width is essential.
Apple Rubber Products, Seal Design Guide — p. 84
Tabletop: Endorses going beyond Parker's 30% on critical static vacuum joints if the groove is widened to take the displaced volume; 'postcuring' is the vendor name for the pre-bake O'Hanlon recommends.
-
In a group of bolts, earlier-tightened bolts relax as later ones compress the joint (elastic interaction, creep of loaded surfaces), which can nearly eliminate their tension - tighten flange bolt circles in a cross pattern and in multiple passes, re-checking the first bolts.
Source, quote & tabletop applicability
As we tighten the rest of the bolts the joint is further compressed, and the previously tightened bolts tend to relax and lose some of their preload. In some cases, this can virtually eliminate our bolt tension.
Fastenal, Technical Reference Guide, Rev. 9 (2005) — p. 26
Tabletop: On the next machine's lid, a single-pass tightening leaves the first-torqued sector under-clamped and is a classic cause of an O-ring leak that 'moves' with each reassembly.
-
When machining an extra-shallow gland to get heavy squeeze, widen the groove enough to accommodate the full O-ring volume, or the ring will be crushed instead of sealed.
Source, quote & tabletop applicability
when an extra-shallow gland is desired in order to increase the squeeze, it must be made wide enough to accomodate the full O-ring volume.
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 1-2
Tabletop: If a next machine's grooves are cut for 30% squeeze, groove cross-section area must still exceed ring cross-section area - check fill before cutting.
-
Use dovetail grooves only where the ring must be retained (vertical faces, lids that open); they cost more to machine, need the sharp-corner radius R held closely, and Parker sizes them for less squeeze (16-27%) with metal-to-metal flange contact.
Dovetail: 66 deg walls; W=.139 -> L=.111-.113, G=.113-.117, squeeze 20%, R=.010; radius R is critical - too small damages ring, too large causes extrusionSource, quote & tabletop applicability
Radius R is CRITICAL. Insufficient radius will potentially cause damage to the O-ring during installation, while excessive radius may contribute to extrusion.
Parker Hannifin, O-Ring Vacuum Sealing, Catalog 5705B (1998) — p. 10
Tabletop: Worth it for a hinged or frequently-removed lid where the ring falls out during assembly; otherwise plain rectangular face grooves are cheaper and more tolerant.
-
Avoid welding lids onto a thin flat vacuum chamber: weld shrinkage warped the whole frame; grinding off the weld and sealing with a flat Viton gasket fixed it - prefer demountable elastomer seals for flat chambers.
Source, quote & tabletop applicability
after the welding, the bottom plate contracted so much that it bent the whole frame out of shape... seal the bottom plate against the frame using a flat Viton ring.
Baumgartner, The Cyclotron Kids' 2 MeV Proton Cyclotron — Cyclotrons 2013, WE1PB05 (2013) — p. 2-3
Tabletop: A fabrication trap the builder can skip entirely: o-ring/flat-gasket both lids, weld nothing flat and thin.
-
Stretch a groove-mounted O-ring 1-5% on its ID (2% ideal); more than 5% stretch thins the cross-section, accelerates aging, and loses seal compression.
O-ring ID = groove diameter / (1 + stretch), stretch 0.01-0.05, ideal 0.02; CS reduction ~ f(% stretch)Source, quote & tabletop applicability
This stretch should be between 1%-5% with 2% as the ideal in most applications. A stretch greater than 5% is not recommended.
Apple Rubber Products, Seal Design Guide — p. 11
Tabletop: When picking the AS-568 size for a non-standard groove (dee-stem feedthrough, viewport), size so the ring sits at ~2% stretch rather than swimming or straining.
-
Never let the O-ring volume exceed the gland volume (crush seals excepted, where fill should still stay under 95% of the gland void) - thermal expansion or swell with a 100% -filled gland destroys the seal or the hardware.
V_oring(max, incl. tolerances) < V_gland(min); crush seals: V_oring <= 0.95 * V_glandSource, quote & tabletop applicability
The maximum volume of the O-ring should never surpass the minimum volume of the gland... the O-ring volume does not exceed 95% of the gland void.
Apple Rubber Products, Seal Design Guide — p. 14-16
Tabletop: Check fill arithmetic including worst-case ring tolerance before machining; Viton heated by RF or magnet proximity expands ~16e-5/C and needs that free volume.
-
Static gland sealing faces tolerate 64-128 RMS but 32 RMS is preferred (16 RMS for vacuum/gas); compress static seal cross-sections 10-40% and dynamic seals only 10-30%.
static faces: 32 RMS preferred (64-128 tolerable), 16 RMS vacuum/gas; static squeeze 10-40%, dynamic 10-30%Source, quote & tabletop applicability
a finish of 32 micro-inches RMS is preferred... Static seal cross sections are generally compressed from 10% to 40%, whereas dynamic seals are from 10% to only 30%.
Apple Rubber Products, Seal Design Guide — p. 19, 61
Tabletop: For a rotating or sliding shaft feedthrough (target manipulator), back off to <=30% squeeze and a 16-32 RMS shaft finish, or friction will shred the ring.
-
Handle O-rings like precision parts: clean the gland of all debris, lightly coat the ring with a compatible lubricant (never a lubricant of the same chemistry as the ring - like dissolves like), cover threads/sharp edges with tape during installation, and remove twists.
Source, quote & tabletop applicability
Do not use a lubricant composed of the same material as the O-ring because 'like' will dissolve 'like.' For example, a silicone lubricant should not be used with a silicone O-ring.
Apple Rubber Products, Seal Design Guide — p. 20, 109
Tabletop: Silicone vacuum grease on Viton is fine; silicone grease on a silicone ring (or hydrocarbon grease on Buna-N) slowly destroys it.
-
To run a standard round O-ring in a non-round (rectangular) face-seal groove, keep every inside corner radius at least 3x the O-ring cross-section diameter and match ring centerline length to groove centerline length.
inside corner radius >= 3 * O-ring CS diameter; O-ring ID = (groove CL length / 3.14) - O-ring CSSource, quote & tabletop applicability
In order to use a standard round O-ring, the inside corner radius of the groove should not be less than three times (3X) the O-ring cross-section diameter.
Apple Rubber Products, Seal Design Guide — p. 84
Tabletop: Directly applicable to a racetrack or rectangular lid/port on a next machine: a 1/8 in. cord ring needs >=3/8 in. corner radii or it will bunch and leak at the corners.
-
In a well-welded chamber the residual leaks are the gasketed joints, and the specific failure mode is non-uniform gasket thickness - use uniform gasket stock and suspect gaskets before welds.
Source, quote & tabletop applicability
Only two leaks were detected, and these were in the gasket seals. They were believed to be due to non-uniform thickness of gasket material.
Creutz, Design and Construction of Synchro-Cyclotron — NYO-780 (1950) — p. 24
Tabletop: Directly applicable leak-hunting prior - on a next machine's chamber, check O-ring/gasket joints first and control gasket thickness uniformity.
-
Build beam probes as RF-shielded copper fingers entering through a Wilson seal so radial depth is adjustable under vacuum - Berkeley added a whole second diagnostic probe this way without breaking vacuum architecture.
Source, quote & tabletop applicability
An auxiliary copper probe, shielded for RF pickup, was introduced into the main vacuum tank through a Wilson seal on the port near the ion source ... made adjustable as to its radial depth
Yeater, 184″ Cyclotron: Synchroscope Beam Pictures on Two Probes — MDDC-987 (1947) — p. 3
Tabletop: Exactly the right probe pattern for the reference machine's chamber: an O-ring/Wilson-sealed sliding shaft with grounded shield tube; unshielded probes near a 9 MHz dee read RF pickup, not beam.
-
Weld all direct vacuum connections; make demountable joints as welding-neck flanges with DOUBLE O-ring grooves in a standard flat-face flange and a pump-out port between the two gaskets (permits leak checking the joint and guarding the inner seal).
Source, quote & tabletop applicability
welding neck flanges with double O-ring gasket grooves machined in a standard flat face flange and provided with a pump-out connection between the two gaskets.
Foss et al., Cyclotron Component Design Technical Reports — TID-454 (1952) — p. 141
Tabletop: The double-O-ring-with-interspace-pumpout trick is worth stealing for any large troublesome amateur flange (chamber lids especially) - the interspace can be sniffed for leak location or held at rough vacuum to null permeation across the inner ring.
-
Pass a rotating shaft into vacuum with a ferrofluidic seal: ferrite-loaded low-vapor-pressure liquid held in ~0.005 in. radial gaps by magnetic fields, sealing a full atmosphere with no sliding contact.
ferrofluidic rotary seal; ~0.005 in. radial gap; holds 1 atm differential; commercial item (Ferrofluidics Corp., 1971)Source, quote & tabletop applicability
a "Ferrofluidic" vacuum seal having ~ 0.005 in. radial gaps in which a ferrite loaded low vapor pressure liquid is held by magnetic fields
Tabletop: Transferable component class, already commercial in 1971 and cheap today — the clean answer any time an accelerator mechanism (chopper, rotating target, variable capacitor) needs rotary motion through the chamber wall.
-
Motion feedthroughs are a seal failure class of their own: chevron-stack elastomer seals on the source's radial and azimuthal drives were unreliable and short-lived; the durable replacement was two simple 1/4-inch cross-section O-rings, each in a polished, close-tolerance machined adapter ring properly fitted to the housing.
Source, quote & tabletop applicability
The Buna N, chevron shaped vacuum seals between the cyclotron accelerator tank and the radial and azimuthal drive mechanisms ... were unreliable and displayed a short life expectancy.
Tabletop: For sliding/rotating shafts into the chamber, dual O-rings in polished glands (surface finish and land tolerance doing the real work) outlast fancier stack packings; seal geometry matters less than the finish it rides on.
-
Double gaskets with a pumped interspace, everywhere possible: UW used "Berkeley standard double groove gaskets with pump outs ... on all joints" of the tank, and double gaskets with pump-out space wherever possible throughout the 4800-liter envelope; single stuffing-box gaskets only around water lines, ion-gauge tubes, RF-loop insulators, viewports. Materials: Garlock style 8474 hycar rubber, molded hycar sandwich gaskets, standard O-rings; chevron seals on moving shafts (O-rings on small ones). The leak record vindicates the ranking: leaks appeared in two stuffing-box seals and one flat gasket (rubber failure) — none in the double-gasket joints.
Source, quote & tabletop applicability
Double gaskets with a pump-out space between the two gaskets are used wherever possible for making seals at the joints in the system.
Tabletop: Same guard-vacuum doctrine as ornl-1196/anl-5907 — on a tabletop machine the pumped interspace is most worth its complexity on the largest lid and on any joint that gets opened often; the 1951 materials list maps to modern Buna-N/Viton.