If you have an air line to the surface, then running the enclosure at positive pressure to negate the force of water pressure at depth might be a starting point.

I would also reconsider whether a linear actuator is the right approach. A hydraulic piston external to the housing where your motor resides reduces the complexity of any waterproofing, and might be a better approach if you’re dealing with significant depths/pressures.

If the travel distance of the rod is fairly limited, you can use a corrugated boot.

Otherwise, maybe a gland seal with positive air pressure on the dry side?

Would depend on how deep underwater and the pressure and so forth, but maybe look at how steam engine piston rods are sealed - an o-ring or a wiper seal would probably do the job

Look at the gland end of any hydraulic cylinder.

If you need a shaft to slide through the wall of your housing, Google “shaft seal” or “oil seal.” The seal usually presses into a hole in your housing, so you will want the housing to be metal rather than plastic.

https://www.mcmaster.com/products/shaft-seals/

A dynamic seal like this is going to be difficult to implement. The shaft will need to be very smooth and carefully aligned to the housing. Some small amount of water is likely to get past the seal.

I’m with @Longpork3@lemmy.nz. Consider other ways to actuate your claw that don’t require moving a part through the housing wall. I’m addition to pneumatic or hydraulic actuators outaide your housing, here’s an idea using magnets and screws:

https://org.osu.edu/osu-uwrt/2022/04/11/underwater-claw-development/

Consider a hydraulic actuator that already works underwater?

Thanks for all your responses! I should have mentioned that pneumatics and hydraulics are off the table at this time.

Right now we’re thinking of manufacturing a flexible seal that can travel with the rod while maintaining a seal with the housing. That might be over engineering a bit, but we’ll be doing some testing this week to see how it goes!

I’m late to the party, but I’m not sure why nobody mentioned the keyword “stuffing box”. That’s what this exact part is called in maritime operations, and as the name suggests it was traditionally a box packed with something to slow down the movement of water. You still have to bail on the other end, though, because it’s not perfect.

However, since it’s just a linear actuator it’d be simpler to go with an accordion-type covering. At shallow depths you could use a normal plastic one, if it needs to go deep metal ones exist. That’s much simpler to deal with than an actual sliding bearing.