Watch O-Ring -- Will it survive 50m water depth? 

One thing I might try is lubricating the O-ring to make sure it seats in it's final position without getting pinched or torqued against the surfaces as they rotate and compress into place. I am sure the Swiss have a very special lubricant for this, but the industrial solution is NyoGel 760G. This is a silica based grease that is dielectric and typically used for mil-spec connectors, but SureFire did a bunch of testing years ago and found it is ideal for O-ring seals on flashlight bodies.

You can also use a flat plate over the o ring and use the screw to push the plate down just like the vostok amphibia does it.

I'd vote for moving the o-ring away for a face seal based on my limited experience with seals. Break all edges as always and if possible. We used Parker Super O-Lube which, given the amount in the bottle, is expensive. The Parker O-ring design handbook can also be found for free pretty easily online if you want to have at least 2 references on designs (and most go to when they hear they have to seal).

The face seal is common on many watches from various manufacturers. I've cut 100's of 1000's of O-ring grooves in end caps and in pistons for hydraulic cylinders. We always did it on a lathe and used grooving inserts with a round tip to better match the O-ring cross sectional shape. We always stretched the O-rings to install them and we coated them with hydraulic oil prior to assembly. I would keep them dry for inside a watch! While you can buy tiny end mills for cutting groves with a mill I'm not so sure how cost effective they are. A little hand profiling on an old end mill with a grinder might be cheaper and faster for a prototype.

+1 on lubing the o-ring with silicone for production, but for testing i question it. The first option that comes to mind is completely removing the threads from the equation. You would do that by trepanning a groove away from the threads, and then having the mating part just seal up against that groove. That's going to require more of delta in the diameter, since now you have to not only accommodate the o-ring, but a flat inside of the o-ring as well. The other option I can think of, requires more depth, but does not require additional width. This would be to have the thread at the end of the boss, an unthreaded section, and then the o-ring section. On the mating piece with the recess, the bottom of the recessed wall would be threaded, above that would be pla8n, and above that would be enlarged for the o-ring.

Happy Next Year John.

You might find that using gasket lube helps with the sealing. The O Ring tends to drag against the faces compressing it, the silicone gasket lube helps it slip and compress. As always, interesting world you're wandering into. If you are not a member of the NAWCC, you might want to join for library access. Steffan Pahlow has written a hard to get and often only available from Swiss publishers book on case making, it might be helpful and when I was still a member they had copies for lending. George Daniels books may offer some insight on case making. One of the lessons I learned in engineering is to avoid getting trapped in trying to solve simple looking but complicated problems that others have already worked out :-) Happy New Year!

I think the mill marks on the o-ring seats are making a high pressure seal more difficult. Maybe fly-cut the seats with a custom tool of the right diameter so you get (hopefully) a mirror finish on the seats, or at least mill marks that are concentric with the o-ring. Or maybe use some custom rotating piece that's close to the full diameter and polishing compound.

50 meters is 5 bars, not 50 bars. I suggest lubing the oring with silicone grease. I remember orings can be ordered with different rubber hardness. Purhaps a slightly harder rubber type will help at higher pressure.

Happy New Year John. interesting video. I would try to avoid pulling the o-ring over the threads. The threads are simply drawn as two overlapping parts in Fusion 360. But in reality there is an empty space when the thread ends. This could reduce the pressure on the o-ring. Just guess. Some lube isn't wrong either. Thanks for the video.

Great video 👍 John If I had to understand this issue I would purchase a watch like a Seiko with the appropriate water resistance rating .Then I would using high magnification inspect cut profile of O ring seat and lid seat shape profile and measure o ring size ratio to case fitment size.

I would make it a straight thread port seal. used in hydraulic fillings instead of a channel for the seal its a angled chamfer. and geese the seals.

Face seal would be my approach. Also, I couldn’t tell on the last test if the bubble grew because it leaked or if that was just the bubble already there expanding after the pressure drop.

50 bar= 500m H2O water hight, 5bar= 50m of water.

You might not be having a leak. From what i remember a couple bubbles running off isn't an indication that there is a leak, there's just a little bit of air stuck in between things that escapes, rather if you were having a continuous stream of bubbles that means a leak.

Hi, 50m should be 5 bars not 50: 1bar is more or less 10m. 5 Kg/cm^2 on the scale is 5 bars wich is correct. I don't get why you compress the air outside first, because in real life you bring the watch with atmospheric pressure and submerge it gradually with external pressure building up. If you compress the air to 5 bar, you are compressing the tiny air pocket left by the compressed o-ring to 5 bar too outside the inner chamber (The tiny anular air pocket i'm talking about, is in the upper left corner of the square at the 12:20 time). Then you submerge and nothing happens cause the pressures are equal. My personal opinion is that when you release the pressure, the bubbles you see is just the compressed air in that small pocket that is expanding. As mentioned by Greg Koenig you need to lubricate the o-ring with silicon oil to avoid twisting or warping, only then you are sure the oring is working as designed. Happy new year!

Hi why dont you share this PDF with us?

What about pressurized chamber(using something similar to a chemical getter in radio tubes just the opposite and also not using induction to activate it). Might enjoy this movie though: Le Temps d'Anna (TV Movie 2016)

Use an isolated groove and make sure the oring is supposed to be an ID or OD?

Not making cookie cutter watches for shure.

50 bar is 500 meters, not 50

Yep, face seal and o-ring grease. I use to make scientific instruments and deploy them to a depth greater than this that's how we did it. We had some other designs to but this should get you the results your after. Apply the grease and wipe most of it so it's just a thin film. Make sure you faces are clean and there are no scratches, and nothing is stuck to the o-ring or surface face after the grease is applied. Check you o-ring by feel, I use to hold them and run my finger tips over them to feel any anomalies. Be careful you're not damaging the surfaces or o-ring with your metal tweezers.

Your second option is how most case back seals work.