Waterproofing servos

[Bob Gato]

Very understandable Nick... Although the cheap little throwaway servos are like laboratory mice... If you can make one work underwater you probably can make any server work on the water.

[Monahan Steam Models]

OK crazy idea #47...
what if you drilled and tapped a port in the servo case(anywhere) and screwed in a tiny pressure port (borrowed from the nitro RC airplane world)-then ran a short dead end silicon tube or a tube made from another stretchy expanding material like latex surgical. Who says the accumulator has to be part of the servo- it could be more bourdon tube like?

Nope, not crazy at all and very doable. I’ve seen crazier things in the world happen over the past week.

I’m pausing my own waterproofing experiments on these knock off sub-micro SG90 servos for a few reasons. First the quality of the servo is not worth the time and work invested to improve it. Second, the build that I was using these servos in can utilize a slightly larger, higher quality manufactured micro servo that draw less amps. Third reason and again this relates to the build I’m using these in, the work and time that is needed to upgrade a $2 servo to be dependable and waterproof in a ‘in the wet’ environment far outweigh the ease of just installing them in a WTC.

[Bob Gato]

I am in the "Zone" Randy!

[RanSan]

OK crazy idea #47...
what if you drilled and tapped a port in the servo case(anywhere) and screwed in a tiny pressure port (borrowed from the nitro RC airplane world)-then ran a short dead end silicon tube or a tube made from another stretchy expanding material like latex surgical. Who says the accumulator has to be part of the servo- it could be more bourdon tube like?

Funny you should say that Bob. That’s sort of how the thrusters on the Atlantis tourist subs work.

[Bob Gato]

OK crazy idea #47...
what if you drilled and tapped a port in the servo case(anywhere) and screwed in a tiny pressure port (borrowed from the nitro RC airplane world)-then ran a short dead end silicon tube or a tube made from another stretchy expanding material like latex surgical. Who says the accumulator has to be part of the servo- it could be more bourdon tube like?

[Monahan Steam Models]

It never ends. Huh, Nick!

David
Fellow Traveler down these rocky roads of advancement

That’s the truth right there! The only way forward is to stay in the fight.

[redboat219]

Thanks Sam.

[Sam Victory]

Sam, do you have a link to Chinese RC sub sites?

潜艇模型讨论区-5iMX.com 我爱模型 玩家论坛 ——专业遥控模型和无人机玩家论坛（玩模型就上我爱模型，创始于2003年）

潜艇-台湾遥控模型-RCTW

Here, I haven't paid attention for a long time, because rc submarines are very rare in China and few people want to participate in the hobby. The second link is to a forum in Taiwan province, and the first link is to a forum in mainland China. I often worry that this hobby will disappear in my generation.

V

[redboat219]

Sam, do you have a link to Chinese RC sub sites?

[Sam Victory]

Try this, I know a guy who runs most of his submarines with the servo submerged in water and has been running them for years. His method is this and simple: first open the servo housing and glue the control circuit board of the rudder with 704 glue (this is a silicone rubber used to seal the circuit board and make electronic components waterproof, like your ESC). Then lube the servo gears, liberally. Then cover the rudder housing and seal the edges of the housing with 704 glue. I know this makes the servo housing look ugly because it's covered in glue, but it proved to be effective. Finally, mount your servo upside down (rocker arms facing down) I think this allows the lube to naturally build up at the gears by gravity and lubricates the gears nicely.

Kafuter 45g Silicone Industrial Adhesive K 704 705 704B RTV Silicone Rubber White balck Transparent Glue|Silicone Sealant| - AliExpress

[He Who Shall Not Be Named]

Steve,

Yes that is a possibility. I am still unsure of why exactly the diaphragm failed in the first place. The servo has been sitting on the bench untouched, in a climate controlled room since it was assembled. The day before it was fine. It was either Murphy, the damn cats or something else that caused the failure. Really the diaphragm should have been more securely fixed and sealed into place by a mechanical means.

Nick

Sounds like the adhesive used to bond the soft rubber disc to the hard plastic body failed -- likely chemical attack from the oil. Find an adhesives chemistry that is inert to the oil. Or change the oil type.

It never ends. Huh, Nick!

David
Fellow Traveler down these rocky roads of advancement

[Monahan Steam Models]

Could you put the diaphragm on the inside of the case?

Steve,

Yes that is a possibility. I am still unsure of why exactly the diaphragm failed in the first place. The servo has been sitting on the bench untouched, in a climate controlled room since it was assembled. The day before it was fine. It was either Murphy, the damn cats or something else that caused the failure. Really the diaphragm should have been more securely fixed and sealed into place by a mechanical means.

Nick

[SubDude]

Could you put the diaphragm on the inside of the case?

[Monahan Steam Models]

First major failure. The waterproofed servo has been left sitting in the bench for the last 20 days. Today when I looked at it, the equalizer diaphragm had popped off of its RTV seal and the mineral oil had leaked out.



Unsure what caused this failure, but it’s not looking good.

Nick

[Monahan Steam Models]

It has been 10 days since I finished the build and testing of the sg90 servo and it has not leaked a drop of mineral oil. Just plugged it into the servo tester to run it through some paces and it still works the same as before. At this point I’m happy with the results but the slightly higher current draw than stock is still a concern.

I tested a non knock off, original sg90 servo and found that this unmodified servo current draw was half of what I saw with the servo batch I used in these tests. Seems there might be a manufacturing quality issue with the knock off servos. There was no significant difference between the two servo types or their components upon visual inspection.

That is what I have to report so far.

Nick