Author Topic: industrial waterproof miniature servos (Read 9388 times)

Admin · « **on:** January 17, 2007, 01:06:57 PM »

anyone know of a place selling industrial strength servos, preferably small-ish and waterproof to a depth of 150m+ ?

cost and torque doesnt matter . . .

JonHylands · « **Reply #1 on:** January 17, 2007, 02:06:09 PM »

Ahhh, the holy grail of small AUVs...

Don't hold your breath. I've spent years looking for stuff like that.

Course, I haven't looked in the last couple, so things may have changed, but I wouldn't bet on it. Easiest thing to do it build a simple linear actuator, and seal it in a tube with the electronics. Use a shaft seal like this:

http://www.modelluboot.de/Zubehoer/Dichtungen.html#WDR

That's what I have on MicroSeeker, and it works pretty well.

What size constraints do you have?

- Jon

Militoy · « **Reply #2 on:** January 17, 2007, 02:34:33 PM »

Instead of a shaft seal - as long as cost isn't a major issue - you could seal the entire actuator into a tube or can, and transfer the motion using a welded stainless-steel bellows. That would eliminate any issues with sealing actuator shafts - kind of like the new non-breaching periscopes on subs. Either a waterproof connector can be used for control, or a hermetic glass-to-metal header. I use the SS bellows for control of oil expansion (ullage) in high-voltage transformers, sometimes at higher pressures than you'll see at 150M depth. Your biggest challenge if you take this approach will be to find a servo or actuator that can expand the bellows and its (probably dry nitrogen) air charge against the pressure at your max depth, while also moving the work load. At shallower depths of course, it just gets easier.

EDIT - FYI: Here's a link to one (good) mfr. The edge-welded types cost more, but are much more compact and flexible.
http://www.alloybellows.com/

Admin · « **Reply #3 on:** January 17, 2007, 02:56:12 PM »

ok I should clarify a little more . . .
so this is basically a continuation from this post:
http://www.societyofrobots.com/robotforum/index.php?topic=105.0

The original solution in that post worked. But now that it works for ~20 feet, they (the Navy) want it to work for deep water . . . meaning hobby servos just wont cut it. I was hoping to just scale up my original design/solution.

To summarize, I will need to do a large quantity of these - not just a one off. It has to be guaranteed reliable, professional, and with a short development time. So Im looking for a quick off the shelf solution for mass production . . .

I think I will have to tell them its a no-go project, based on the constraints and whats on the market . . . Whoever markets the first waterproof servos will make lots of money, for sure . . .

Militoy · « **Reply #4 on:** January 17, 2007, 04:46:38 PM »

Out of curiosity - how much torque do you need for the servos in this application? Rotary motion required, or linear throw? Degrees of motion (if rotary) or length (if linear)? Bidirectional motion, or just push or pull? A lot of questions, I know - but I thought I might take a quick crack at the problem - and maybe eventually add a useful item to my COTS catalog. I will say though - that to withstand over 220 psi while operating reliably underwater, they probably won't be cheap!

Admin · « **Reply #5 on:** January 17, 2007, 08:55:05 PM »

Quote

how much torque do you need for the servos in this application?

no less than 24oz-in (barely anything)

Quote

Rotary motion required, or linear throw?

the output needs to be linear (or near linear). on the current design, im using a linkage with rotating servos for near straight motion (rotation is small).

Quote

Degrees of motion (if rotary) or length (if linear)?

the servo doesnt rotate more than ~5 to ~15 degrees for rotary, or ~3mm for output linear motion. it needs to be relatively controlled (doesnt need to be perfect), but not binary motion, either.

Quote

Bidirectional motion, or just push or pull?

push/pull motion

Quote

they probably won't be cheap!

thats ok, im Navy funded - price doesnt matter

Militoy · « **Reply #6 on:** January 17, 2007, 10:04:52 PM »

Quote

thats ok, im Navy funded - price doesnt matter

Lookout - you'll give taxpayers the wrong idea! I know what you're saying, though - I supply equipment for Navy contracts regularly, and even though that market has become much more competitive and cost-sensitive than in years past, at least they are willing to pay a fair price for a reliable product. The day of the "$300 screwdriver" is long gone - but at least in a military contract, you won't be as likely to get your throat cut for a nickel.

I actually gave the problem a think-over on the way home today, and I think you may be right; There may well be a market for a line of underwater servos. After some thought though, I realized that my idea of a dry nitro-filled actuator isn't the way to go for deep-submersion applications. A fluid-filled actuator, like maybe filled with Diala-AX or Monsanto Coolanol, would resist crushing at depth much better than a gas-filled servo. I will dig into some of my "down-hole" application notes over the next few days. I'll PM you a SolidWorks sketch for you to wave around - and if anyone there shows serious interest, I would be willing to supply you with a loaner prototype unit to play with.

Admin · « **Reply #7 on:** January 18, 2007, 08:08:23 AM »

Quote

I'll PM you a SolidWorks sketch for you to wave around

I dont have SolidWorks, but I do have Autodesk . . .

What would you think the development time would be?

ballbreaker · « **Reply #8 on:** July 25, 2010, 02:14:47 AM »

Look here they have high torque and waterproof but i don't know about the size i haven't look closely.

http://www.traxxas.com/products/electric/rustler3705/trx_rustler3705_details.htm

voyager2 · « **Reply #9 on:** July 25, 2010, 04:03:52 AM »

Do you realize this thread is 3 years old?

Admin · « **Reply #10 on:** July 25, 2010, 07:28:16 AM »

Its 3 years old, be I'm still working on this project =P

I'm fairly sure when it says 'waterproof' that its actually only 'weather proof'. That said, I'll buy a few and test them out.

madsci1016 · « **Reply #11 on:** July 25, 2010, 07:51:00 PM »

I have a few of the waterproof servos. (didn't you see it in my Axon + Irobot tutorial video?) They survived my test or a few hours at the bottom of my full tube while continuously moving back and forth. That said, I don't think they would go to depth too well.

However, they seem to draw more then normal current for operation. it seems to be because of the o-ring they stuffed in to water-proof the shaft puts an awful lot of friction force on it.

Admin · « **Reply #12 on:** July 25, 2010, 09:08:56 PM »

Quote from: madsci1016 on July 25, 2010, 07:51:00 PM

However, they seem to draw more then normal current for operation. it seems to be because of the o-ring they stuffed in to water-proof the shaft puts an awful lot of friction force on it.

I custom waterproof my servos, and I do use an o-ring as well. I add a little silicon oil to the ring which in theory reduces the friction. Willing to add some oil and test your friction theory?

Admin · « **Reply #13 on:** August 02, 2010, 02:25:37 PM »

I just bought four of each water proof servo they had . . . almost $500 worth . . .

http://www.traxxas.com/products/accessories/trx_accessories_servos.htm

Towerhobbies.com had them all . . . but decent specs are non-existent unfortunately . . .

Anyway, I'll put them through some rough drowning and torque tests to see how waterproof and how reliable they actually are . . . if they pass, I'll use one for my next robot fish version at the end of this year.

If you don't hear back from me in a month or two, just remind me to post results here.

madsci1016 · « **Reply #14 on:** August 02, 2010, 02:48:45 PM »

Remember, they are water-proof, but not depth rated. So I'll bet more then a few feet deep and they will fail.

They were designed to be the steering servo for RC 4WD trucks that can go through puddles.

Razor Concepts · « **Reply #15 on:** August 02, 2010, 02:53:11 PM »

The company I'm interning at got some European servo company to custom manufacture waterproof servos, about the size of a hobby mini servo. Machined from aluminum, oil filled, and seals. The only real issue is the seal to the outside world from the horn, I don't know how well that servo is sealed as we just use the pressurized oil system (in the entire actuator compartment) as a just-in-case kind of thing.

madsci1016 · « **Reply #16 on:** August 02, 2010, 03:01:26 PM »

Is it a static oil pressure or dynamic pressure compensation?

Static oil pressure is asking for trouble. I would like to see a dynamic pressurized oil filled servo, but I wonder how much the windage loss increases due to the oil.

Admin · « **Reply #17 on:** August 02, 2010, 03:43:51 PM »

Quote from: Razor Concepts on August 02, 2010, 02:53:11 PM

The company I'm interning at got some European servo company to custom manufacture waterproof servos, about the size of a hobby mini servo. Machined from aluminum, oil filled, and seals. The only real issue is the seal to the outside world from the horn, I don't know how well that servo is sealed as we just use the pressurized oil system (in the entire actuator compartment) as a just-in-case kind of thing.

Any chance you can pass me some specs on them (dimensions, torque, weight, cost, company name, etc)? I have money.

I only need 10 feet depth, max. I'm pretty sure these servos I bought are 'waterproof', meaning just splash resistant. But I'd like to see by how much, and if some slight modifications can increase the depth.

Soeren · « **Reply #18 on:** August 02, 2010, 08:20:28 PM »

Hi,

If you're just after a 3mm linear motion, why not do this through either a flexible membrane, or as a "two part" magnetically transferred through a plastic housing of some kind. Feedback could be optical, Hall based, inductively coupled, or whatever.

Yes, I know you wanna buy it ready made, but the time you spend on it thus far...
Thinking out of the box and building something that negotiate the demand for absolutely watertight might be the "quick" solution, when you count the hours when you're bathing in oil and writing about it

If you make a working proto, I'm sure you can find a company willing to clone it into the number you need.

Admin · « **Reply #19 on:** September 07, 2010, 11:41:57 AM »

Quote from: ballbreaker on July 25, 2010, 02:14:47 AM

Look here they have high torque and waterproof but i don't know about the size i haven't look closely.

http://www.traxxas.com/products/electric/rustler3705/trx_rustler3705_details.htm

I just finished testing them.

I bought the 2075, 2056, 2080, and 2065. I put two of each in 1 foot of de-ionized water, and had them rotate +/-45 degrees every 0.9 seconds for 4.5 hours non-stop. No counter torque was applied to any of them. None of them got warm as the water kept them cool.

results:
A 2075 stopped working after 3.5 hours. It filled with water.

Two 2056 and one 2065 were acting funny - ie the control system wasn't doing what it is supposed to do, either overshooting or going too slow. They were all filled with water.

A 2075 and a 2080 were also filled with water, but still working fine.

A 2065 and a 2080 had *no* water inside and still working fine.

I'd say they are much more water resistant than other servos, but *not* waterproof as the packaging claims.

Oh, and they don't come with servo horns, so you'll need to buy Futaba-style horns separately to use the servos. Silly!

Upon close inspection, these servos aren't too special. They have finer tolerances on the connecting part seals, like where the wiring comes out and the output shaft, and an o-ring on the output shaft, but thats about it. I'll probably waterproof them SoR style and run the test again. I still have two more of each

Asellith · « **Reply #20 on:** September 10, 2010, 02:53:07 PM »

I feel your pain lately admin. Working on a project for a table top gaming product I'm going to try and sell. Everyone keeps telling me do this and do that and get mad at me when I say they will not work. I need a production model not making 10 or 15 of them I need to make 50 or a hundred in batches for sale. I also have a 40+ hour a week real job. So anything more then 15 secs of my time is not worth it per unit. After a month I got it down to gluing 1 part to another and tossing them in a bag then I'm done.

Also for your project I had an idea. It might be too complex but we are talking 10s of feet not hundreds right? Could you use a hydraulic system? for instance you have a pump and some servo operated valves the valves fill a rubber bag for lack of a better term and then drain when the servo cycles back. This keeps the electronics the same but also gives you a way to interact with the exterior of the unit from a water tight enclosure. 3mm of movement isn't much. The rubber booties might break down eventually and adding a hydraulic pump and storage might be interesting but its an idea.

Admin · « **Reply #21 on:** September 10, 2010, 05:50:29 PM »

Quote from: Asellith on September 10, 2010, 02:53:07 PM

I feel your pain lately admin. Working on a project for a table top gaming product I'm going to try and sell. Everyone keeps telling me do this and do that and get mad at me when I say they will not work. I need a production model not making 10 or 15 of them I need to make 50 or a hundred in batches for sale. I also have a 40+ hour a week real job. So anything more then 15 secs of my time is not worth it per unit. After a month I got it down to gluing 1 part to another and tossing them in a bag then I'm done.

Welcome to my world

And it doesn't get easier, because you only go further and harder with each future product.

As for not using servos for my robot fish - its too late, I've already zeroed in on the design. The next phase of the project is to scale the fish up to maybe 4x bigger, so now I'm looking into new servos that are more reliable and more efficient.

I just finished doing the SoR water-proofing method to these 'waterproof' servos. Just waiting for the RTV to cure and I'll test them out again. If none fail after say ~5 hours underwater, I'll say they're ready for future tests - loaded shaft (gear breakage) test, and efficiency testing.

Asellith · « **Reply #22 on:** September 13, 2010, 09:27:35 AM »

Quote from: Admin on September 10, 2010, 05:50:29 PM

Welcome to my world
And it doesn't get easier, because you only go further and harder with each future product.

I got smarter this time. I normally use ponoko for laser cut stuff and that only requires a little cleanup and tape removal then if retail sales it goes in a bag. My next project uses spoonflower to make gaming mats from printed fabric. I don't even touch those I just get cash for each yard sold. Although I think I found a hidden calling of mine. I do love me some DFM. Its kind of a unique challenge where fractions of a cent matter.

Admin · « **Reply #23 on:** September 14, 2010, 12:46:08 PM »

I'm running tests of the 2075 and 2056 (three of each), after waterproofing the electronics with RTV, using the same methods as above. I ran out of RTV so haven't yet tested the other servos.

I've had them 1ft underwater for 11.5 hours, and running non-stop for 9 of those hours (couldn't keep batteries charged long enough).

The three 2075 digital servos failed at ~3hrs, ~6hrs, and ~7hrs (powered run time). The first one somehow had its power lines shorted, and the other two just stopped moving. Opening them up, two were filled with water, and the third had no water (something else failed?).

The three analog 2056's failed after running ~9.5hrs, ~13.5hrs, and ~20 hours. The last servo was left unpowered underwater for an additional 18 hours in the middle of the test. Opening them up, one was filled with water, one had a few drops of water, and one had no water at all inside.

ps - I'll edit this post as more results are gathered.

Admin · « **Reply #24 on:** September 23, 2010, 02:27:38 PM »

I ran some 2065s and 2080s continuously under a foot of deionized water, using my waterproofing method.

The two 2065s failed after 19 hours and 35 hours respectively.
The two 2080s failed after 55 hours and 57 hours respectively.

I noticed the ball-bearings rusted up and broke in some of the servos I've tested, apparently not stainless steel . . .

Anyway, I've decided that in terms of torque I really need the 2075 for my application . . . unfortunately its the most likely to fail. I've thought up a new and improved waterproofing technique, so I'll buy a few more and test again to see what happens.

Admin · « **Reply #25 on:** December 16, 2010, 07:15:16 PM »

Good news, I've invented the new ultimate servo waterproofing method. It's cheaper, easier, cleaner, and faster than the other methods.

the test:
I waterproofed two 2075 servos and two 2056 servos.

results:
A 2075 failed after 23 hours. Upon opening it up for inspection, no water had entered the motor, leading me to believe it was a faulty/shoddy servo.

The other 2075, and both 2056's, lasted for 225+ hours! They never failed, so I'm thinking they would have lasted much much longer if I didn't stop the test.

Anyway, when I have time, I'll document my new awesome waterproofing method

DJSprocket · « **Reply #26 on:** December 23, 2010, 06:38:01 AM »

Admin i love your thread so far. if you could drop me a line or update this thread with the water proofing technique you used that would be awesome.
DJS

TrickyNekro · « **Reply #27 on:** December 25, 2010, 07:20:01 PM »

I gave it some through.... And....
Why don't you use a hydraulic system...

You can start with some syringes and some surgical tubing...If you don't want to compensate for water pressure
compression use some liquid instead of air... At small pressures depths up to 30 - 40 meters, liquid compression
should be negligible.....

Diabetics' syringes should do... Also... With a hydraulic system you can adjust the compression rates which can be useful...

Fitting the servo to the syringe may be a pain... But it would be worth the while...

By the way... Have you ever opened a CD/DVD/BlueRay :p to see the head mechanism... Such a system can produce a very nice lineal motion, which you could fit to the hydraulic system...
They also come with nice encoders... although that means more coding for you...

Although salvaging CD mechanisms can't be military style, I'm sure you have you ways of getting such good stuff ;-)

Admin · « **Reply #28 on:** January 15, 2011, 08:54:14 AM »

Hitec is now selling "waterproof" servos. I guess that means I need to buy them and see how waterproof they actually are.

http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXZUR3
http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXZUR4

edit: here are the specs of the two, as the Hitec site doesn't have anything yet

Hitec HS-5646WP Waterproof, High Torque Digital Servo
• Speed (sec/60°): 0.26 (4.8V) / 0.22 (6.0V)
• Torque (Oz-in): 97.2 (4.8V) / 116.7 (6.0V)
• Size (Inches): 1.65 x 0.83 x 1.57
• Weight (oz): 2.05
• Gear type: Metal

This is the Hitec HS-5646WP Waterproof, High Torque Digital Servo. It features 97.2 oz-in torque @ 4.8 Volts and 116.7 oz-in Torque @ 6.0 Volts. The speed is 0.26 sec/60 degree (4.8 Volts) and 0.22 sec/60 degree (6.0 Volts).

Hitec HS-5086WP Waterproof Micro Servo
• Speed (sec/60°): 0.18 (4.8V) / 0.15 (6.0V)
• Torque (Oz-in): 38.9 (4.8V) / 44.4 (6.0V)
• Size (Inches): 1.22 x 0.60 x 1.22
• Weight (oz): 0.78
• Gear type: Metal

This is the Hitec HS-5086WP Waterproof Micro Servo. It features 38.9 oz-in torque @ 4.8 Volts and 44.4 oz-in Torque @ 6.0 Volts. The speed is 0.18 sec/60 degree (4.8 Volts) and 0.15 sec/60 degree (6.0 Volts).

lvmogensen · « **Reply #29 on:** January 31, 2011, 06:40:14 AM »

I also find the HITEC waterproof servo interesting.
I can inform you that they gave e a little more info on the waterproofing when I requested it.

Quote
Dear Lars,

It will have an IP68 rating but would not be able to withstand 100m without being modified for pressurization.

IP68 is up to 15 meters. The servo itself is based upon the HS-5645 but uses a different case and a waterproof bearing.

Thanks,

Shawn
/Quote

How is your progress on testing them?

News:

Author Topic: industrial waterproof miniature servos (Read 9388 times)

Related Topics