Thank you for your response. I am in the process of testing shaft seals and am considering the exact issues you bring up (seal drag and sealing ability with side force on the shaft). I think I have a decent solution, but time will tell.
Thanks for pointing out the steerable ROV light application and your target depth. I’m taking notes…
Regarding cost, it’s going to depend on what people are looking for, so it’s good to get your input. I’m considering two different “tiers” of servos:
Lower cost (maybe $200?):
Motor and gear train: brushed motor, torque and speed similar to a $30 R/C servo (which would be more than enough for your ROV light application), suitable for days of continuous operation
Materials: made from materials that are fine for short-term immersion in seawater (anodized aluminum, stainless steel, plastic).
Seals: fine for short-term immersion up to 200 meters. I think to go deeper than that, or for months instead of days, you’d want the servo to be oil-filled, which would be offered for a minimal additional cost.
Cable: the cable would be potted into the enclosure, and may not have a hermetic seal inside of it (meaning that if you got a cut in the cable jacket, the water could migrate into the servo, but with the proper cable jacket material, this risk would be very low).
Higher cost (maybe $500?):
Motor and gear train: brushless motor, torque and speed similar to a $150 R/C servo, suitable for months of continuous operation.
Materials: made from materials that will allow almost indefinite exposure to seawater (engineering plastics, titanium).
Seals: same as above, would be fine for immersion up to 200 meters without oil filling, but going deeper would require oil filling (for a minimal extra cost).
Cable: the cable would at a minimum contain a hermetic seal to protect against water intrusion if the cable jacket is compromised. A better option (which would probably add another $75-$100) is to have a true depth-rated connector and cable system, similar to what you’d see on industrial actuators and instruments.
All servos would be tested under pressure before being shipped. Obviously, I will never be able to match the cost of a mass-produced, imported servo. Rather, my focus will be on quality, reliability, and customer support.
Any further thoughts you have would be greatly appreciated…