title: O-Rings description: The Slocum hull and component O-rings: G2 single-seal vs G3 double-seal part numbers and sizes, the full spares kit, approved lubricants (Parker O-Lube vs Molykote and the community debate), inspection and replacement, sealing-surface scratches and the sanding controversy, lubricant dry-out, vacuum loss, and assembly tips.
O-Rings¶
O-rings are the glider's primary defense against flooding. Every hull joint, plug, and through-hull fitting relies on a correctly sized, clean, lightly lubricated O-ring seated in an undamaged groove against an undamaged sealing surface. Most Slocum leaks are not main-hull O-ring failures — they're bulkhead connectors, instrument seals installed with the wrong O-ring, or loose through-hull fittings — but the main-hull O-rings are what you inspect and replace most often, so getting their part numbers, lubricant, and handling right is foundational.
Source
Paraphrased from the Slocum G3 Glider Maintenance Manual (Rev. A, "O-ring Maintenance", "Main Hull Double O-ring Seals", "Main Hull O-Rings"), the Slocum Glider Operators Manual, the Teledyne Webb Research user forum, and the UG2 community Slack. Part numbers and prices change — always confirm the current part numbers with glidersupport@teledyne.com and defer to the official Teledyne documentation for your specific glider.
Single seal (G2) vs. double seal (G3)¶
A defining hardware difference between the platforms:
| G1 / G2 | G3 (and late/upgraded G2 with stiffening rings) | |
|---|---|---|
| Hull sealing surfaces | One O-ring | Two O-rings (dual seal) |
| Main hull O-ring P/N | G-024 (2-265-N674-70) | 304697 (2-264, BUNA) |
| O-ring size | 2-265, nitrile (NBR), 70 durometer | 2-264, BUNA, ~60–65 durometer |
| Why | Standard single seal | More sealing-surface area → reduced leak probability |
- The double seal lengthened the carbon-fiber hulls by 0.5" versus the G2, raising standard-config G3 displacement to ~57.7 L and lengthening the internal tie rod — G2 hulls will not fit a G3.
- Early G3s shipped with the G2-size ring (
2-265) and were notoriously hard to open and close; TWR later issued the correct smaller2-264double-seal ring. If a G3 is a "fight for your life" to assemble, check you're on the right O-ring first. - The G2 main hull O-ring (
2-265/N674-70) is nitrile, 70 durometer, ID 7.734", width 0.139". The vacuum/evacuation boss plug O-ring is also nitrile 70-durometer:3-904(G-030), ID 0.351", width 0.072".
Teledyne won't publish full seal specs
Operators have repeatedly asked TWR for G3 groove dimensions/tolerances (e.g. to build their own pressure-test end caps) and been declined. Identify the seal by part number, not by trying to reverse-measure the groove with shop calipers.
Part numbers — full spares kit¶
The complete O-ring spares kit is ASSY 4424 (a smaller kit, ASSY 4423, covers just the two most-used seals). Contents (part numbers are stable; prices are illustrative, ~2012):
| TWR P/N | Size / material | Where used |
|---|---|---|
| G-024 | 2-265 N674-70 | Main hull (G2 single seal) |
| 304697 | 2-264, BUNA, ~65D | Main hull (G3 double seal) |
| G-030 | 3-904 N674-70 | MS vacuum / evacuation boss plug (also G1 unpumped CTD piston seal) |
| B1-132 | 2-007 N674-70 | B1 |
| G-022 | 2-018 N674-70 | 7/8" |
| G-027 | 2-115 N674-70 | 7/8" tail tube |
| G-1214 | Parker 5-712 | — |
| G-1331 | 2-141 N674-70, NBR | — |
| G-1508 | 2-161 N674-70, NBR | — |
| G-607 | 5-105 N674-70 | — |
Order by part number and keep a kit aboard
Wing-rail weight O-rings (~size 010) and the small vacuum-plug / bullet-weight O-rings are the ones that dry out and crumble first — keep spares. A full O-ring spares kit is a standard item on deployment-spares lists.
Lubricant: what's approved, what the field actually uses¶
The official lubricant for main-hull O-rings and sealing surfaces is Parker O-Lube 884-4 (3135-LUBE / 3M-LUBE, "Parker Fibrous O-Lube 884-4", a petroleum naphthenic oil + barium soap), available from the TWR forum or by request from glider support. The maintenance manual procedure is simply: inspect, lubricate with Parker O-Lube, install, seat properly.
For electrical connectors the rules differ — don't cross them:
| Surface | Approved lubricant | Avoid |
|---|---|---|
| Main hull O-rings & sealing surfaces | Parker O-Lube 884-4 | — |
| Impulse connectors | Silicone lubricant | Any silicone spray containing acetone (breaks down the connector) |
| Subcon(n) connectors | Molykote 44 (Dow Corning) | — |
| Dummy / green plugs | O-Lube or silicone spray | — |
| Bellofram / pump rubber | Molykote 316 or 3M Silicone Lubricant | Silicone spray with acetone |
The Parker O-Lube dry-out problem (well known in the fleet)
A recurring community complaint: Parker O-Lube dries out — after a long mission, or even just sitting in a dry/air-conditioned shop, it turns into a crusty white gunk in the O-ring grooves that has to be scraped out, and is "completely non-lubricating." Notes from the community:
- Several operators prefer Molykote / Dow Corning High-Vacuum Grease (silicone) and report TWR has been "OK with it" — but it is silicone, which migrates onto everything and must never get on acoustic transducers.
- Some report Parker's basic O-Lube is hydrocarbon-based and not technically compatible with the glider O-ring material (slightly degrades it); they suggest Parker Super O-Lube (silicone) instead — though it's "so runny."
- Many stay on the standard Parker O-Lube purely to avoid doing anything non-standard that could be blamed for a leak, and simply replace dried O-rings rather than chase a better grease. A reasonable default.
Inspection & replacement¶
- Inspect every O-ring for cleanliness, nicks, slices, dents, and cracks before each deployment; inspect the sealing surfaces for scratches and defects under good light (a flashlight raking across the surface reveals scratches a fingertip misses).
- Replace as needed before every deployment. Many teams running single-use primary batteries replace all hull O-rings every time they open the glider (i.e., every mission). With rechargeable G3s that can recharge/download without opening, a common plan is to open once a year to inspect, replace hull O-rings, and test batteries.
- Cleanliness is everything. Foreign particles in the gland cause leaks and shorten O-ring life. Be "paranoid": gloves, hair retained, clean bench. The vast majority of well-handled hull O-rings simply don't leak.
- Assembly (from the Parker handbook): keep installed ID stretch under 5%, don't exceed ~25–50% elongation reaching the groove, and never twist the O-ring into place. Seat it fully before drawing the joint closed.
Small O-rings dry out and fall apart in storage
The smaller O-rings — vacuum plug, bullet weights, wing-rail weights — are prone to drying out, cracking, and crumbling while a glider sits on the shelf in a dry shop. Check them as part of every pre-deployment workup; some teams have moved wing-rail-weight O-rings to silicone to resist drying/cracking.
Sealing-surface scratches — the sanding debate¶
Carbon-fiber hull sealing surfaces do pick up fine scratches (from sharp edges around ballast bottles, sliding batteries in/out, corner bumpers on the mainboard). What to do about a scratch that crosses the O-ring band is genuinely contested in the community — present both views and decide deliberately:
Several experienced groups (e.g. Rutgers) wet-sand minor scratches out semi-regularly:
- Use fine paper, ~1500–2000 grit, wet.
- Sand a large area (roughly half the hull, scratch centered) to preserve roundness — don't just spot-sand.
- The "fingernail catch" test: if a dragged fingernail catches in it and it sits near the sealing surface, sand it; a few minutes to a few sessions per scratch.
- Prevention: wipe a thin layer of O-Lube on the hull when sliding batteries in/out as a protective film.
Pressure-seal practice says sanding an O-ring sealing surface by hand is a big risk:
- O-ring standards allow only ~0.003" of depression tolerance — and 0.003" is nearly invisible to the eye, so it's easy to remove too much and put the hull out of round.
- The advice: don't sand a hull yourself unless you have a lathe/mill that can hold the hull diameter and remove controlled tolerances.
- The "fingernail catchiness" test is subjective (depends on pressure, nail length, sensitivity).
Webb's position and a reality check
TWR's guidance: replace a hull if carbon fibers are exposed; minor paint/ finish chips (no damage felt in the grey/white material) just need touch-up (epoxy/siloxane paint, or even nail polish in a pinch). Teledyne has also provided some users written instructions on how to "polish" a sealing surface. Reality check from the field: one glider that had scratches polished per those instructions still wouldn't hold vacuum in the shop — so a visible scratch is not always the actual leak path. Diagnose before you sand.
Vacuum, leaks & testing¶
The internal vacuum is your standing leak indicator — always monitor it before launch (less vacuum than expected = a leak; positive pressure can mean dangerous gas accumulation). It fluctuates with temperature, so log temperature alongside vacuum.
- Never power a shallow glider without a vacuum.
- A reasonable "sealed" bar: vacuum stable for 1–2 weeks (recording temperature). Deploy a glider sealed only ~24 h only in an emergency.
- Slow loss over months is common even on a well-sealed glider on the shelf — not necessarily a real leak. But re-verify, and inspect/replace O-rings before redeploying after long storage. A brand-new factory-sealed, helium-tested G3 can still slowly bleed down on the bench; common real culprits found include loose through-hull fittings (e.g., a nose altimeter cable pass-through that needed a ¼ turn) as much as the hull O-rings themselves.
- Don't over-pull the vacuum to mask a marginal seal — excessively high vacuum can pull bubbles out of the ballast-engine oil.
- Leak testing is hard: vacuum-and-wait can take months to reveal a slow leak, and gliders that hold vacuum for months on the shelf sometimes leak as soon as they're in the water. Teams use sectional pressure monitors, pressure chambers (commercial or DIY McMaster-fitting builds), or helium leak detectors (accurate but expensive). No method is foolproof — cleanliness and inspection remain the best defense.
Assembling the double-seal G3¶
A practical pain point worth its own note — closing a G3 against two O-rings:
- Make sure you're on the correct
2-264G3 ring, not the larger G22-265. - Manually push the hull sections fully over the O-ring seals before relying on the tie rod to draw them together — tie-rod lengths vary (especially with stack-on bays like an AD2CP), and short ones won't "catch" until the sections are nearly home.
- Pulling a light vacuum during assembly helps seat the joint the last bit (many teams do this).
- Keep good PEEK hull-opening tools (they outlast the soft black ones but still round off); some teams make thin plastic wedges to crack a stubborn, long-stored hull — wedge away from the O-ring sealing surface, never into it.
Quick reference¶
| Item | Value |
|---|---|
| Main hull O-ring — G2 single seal | G-024 (2-265-N674-70, nitrile 70D) |
| Main hull O-ring — G3 double seal | 304697 (2-264, BUNA ~65D) |
| Vacuum / boss plug O-ring | G-030 (3-904 N674-70) |
| Full spares kit | ASSY 4424 (small: ASSY 4423) |
| Hull O-ring lubricant | Parker O-Lube 884-4 (3135-LUBE) |
| Impulse connectors | Silicone lubricant — no acetone |
| Subcon connectors | Molykote 44 |
| Bellofram / pump rubber | Molykote 316 / 3M Silicone — no acetone |
| Scratch sanding (if you must) | Wet, 1500–2000 grit, large area for roundness |
| "Sealed" criterion | Vacuum stable 1–2 weeks, temperature logged |
See also¶
- Tie Rod — what draws the O-ring joints closed, and why G3s can be hard to assemble.
- Pumps — Bellofram/oil-bladder rubber care and storage (related lubricant rules).
- Maintenance Schedule — where O-ring inspection fits in the service schedule.