Sources for small hydraulic actuators

I'm building a fairly large AUV (robot submarine, essentially) and have encountered difficult problems moving from simple test models to
building a real hull; most of which centers around sealing the hull piercings. In brief, for a small submarine, it's a difficult problem to seal rotating shaft egress points - the techniques used on boats and real submarines allow some leakage, which is managed by bilge pumps - but a small vessel doesn't have enough buoyancy margin to allow taking on water. For other reasons, I don't want to pressurize the hull to match ambient. My current designs use electric propulsion and regular stepper motors for the rudders/planes, and that just isn't going to work long-term.
Analysis of the tradeoffs involved leads me to believe that hydraulic actuators are the best solution to get power out of the pressure hull into the props and control surfaces, and I'm looking for resources for primarily low-cost, secondarily small-size hydraulics. The pressures involved are not high by normal hydraulics standards; a couple of hundred PSIG would be quite adequate. I'm looking for:
* pump * at least 2 double-acting cylinders, ideally bore ~60mm, stroke 150mm. * at least 1 hydraulic motor * solenoid valves for the above * hopefully an off-the-shelf manifold where I can add a pressure sensor, rather than having to machine my own manifold
Cylinders and motor should be, preferably, stainless steel.
I would like these to be either DIRT cheap or totally standardized products, so that I can be assured of buying at least one set of backup parts a little ways down the line (probably Dec of next year). I hope I can get all the above for under $500, though I realize this probably isn't realistic.
The problem is not mainly one of size (I can resize the hull, and add batteries, within a reasonable limit). The problem is more one of standardized hose fittings and price :(
My other option is to go with waterproof DC motors, but these are so exotically expensive (at least, the ones I've found) that I want to avoid that route - besides, hydraulic pipes are easier to run through bulkheads than electric cables. But if someone knows of a reasonably priced source of both stepper (preferably 4-phase unipolar) and regular DC motors that are waterproof to either 5 or 45 meters [two different project phases] I'd be interested to hear about it. Freshwater operation, for the moment.
Any thoughts would be greatly appreciated.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Have you considered using Oil Compensated Brushless DC motors. I work for Woods Hole Oceanographic as well as SeaBotix and have done many ROV designs using them and have found them to be very reliable and virtually free of leak issues providing you have a good positive compensation supply. You can run brushed motors in mineral oil for a relatively short period of time. Problem is they tend to collect brush debris and eventually short out. If your budget is tight a cheap and easy supply of watertight motors are bilge pump motors. Many of the hobbiest ROVs are using them. You could also contact SeaBotix and get thrusters from them but I'm not sure what the price would be. Hydraulics are not very efficient, are heavy, have high acoustic noise and are messy. If you plan on controlling your AUV with acoustic commands from the surface the noise will definitely be an issue and if you plan on running the AUV in a drinking water tank etc the potential hydralic oil contamination will be an issue.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

designs
can
bilge
price
running
There are suppliers around that sell self lubricating parts which don`t require oil and the hydraulics could be powered by the surrounding water itself. Once pressure has been stored in a hydraulic acumulator (Cylinder with a spring at one end), this can be used as a type of mechanical Battery (silent too). I wonder if even ordinary cheap small electric motors could be adapted to work actually in the water itself (where the Brush debris is constantly washed away)? I`d imagine something like streamlining the armature against water resistance by filling in between the windings and replacing the bearings with self-lubricating bushes... ----------------------------------------------------------------------- Ashley Clarke -------------------------------------------------------
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

They work without modification, but there are big issues once you go over to salt water in particular. Motors of the size I'm using are not cheap (currently I'm using a bunch of motors that were originally intended for some kind of aerospace application, and I happened to get them surplus at a good price).
What I'm thinking of when I say waterproof motor is something where the permanent magnets are in the rotor and the electromagnets are in the stator, and the stator is entirely sealed. Plus all bearings are rated for underwater use. This would make it a synchronous DC motor of the type used in, say, floppy drives.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Why not drop the rotars / motors all together and use hydraulic jet propulsion? An aperture / intake sucks the water in on an even keel and another one (or two...) blows it out again after passing through internal bellows. Isn`t this how a lot of sea creatures get around anyway? ----------------------------------------------------------------------- Ashley Clarke -------------------------------------------------------
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

This still doesn't solve the steering problem (still need steppers or something to aim the jet exhaust), but I'll grant you it's an interesting idea. Bear in mind that in order to avoid the exact same problem of getting wires or shafts or pipes through the pressure hull, this bellows has to be able to withstand the FULL pressure differential from sea level atmospheric to 45 meters underwater, without rupturing and filling the hull with water, and that the bellows pumping mechanism has to be able to work against that head of pressure too!
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Looking at ROV's - they use small, brushless motors, that are oil filled. It seems one approach would be to make part of your hull, outside the pressure hull, filled with oil and with a portion of the compartment with a flexible rubber divider between the oil and the outside water. That would compensate for any small compression that would take place from trapped air. Your seals between motors and actuators, and the outside water only need to keep the two separate. Wires of course need to seal between pressure hull and motor compartment, but there are many ways to do that ( such as don't use wire, use bolts with seals or rod encapsulated in epoxy), that are much easier than a rotating shaft.
mgg
--
sig goes here

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The Bellows/ pumping mechanism could be fitted to the exterior with a bit of protecton from a cage. Steering could be acheived by manouevring flexible pipe/s from the stem. I am not an expert on deep water machinery but surely if the drive mechanism was submerged in its own seperate (exterior) casing the forces of pressure are around the components which still provide the same amount of work in proportion to ambient pressure. The completed Sub` using these techniques could be totally covered in a rubber jacket and have the appearance of a Sea mammal! ----------------------------------------------------------------------- Ashley Clarke -------------------------------------------------------
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Fri, 22 Oct 2004 01:05:27 GMT, "Lewin A.R.W. Edwards"

I have little knowledge in this area, but it seems that if you use two jets horizontally separated, having output on the left one will propel the thing forward and to the right, likewise the right one would send it forward and to the left. Turn them both on for straight ahead. Have them send it down overall and have a third one under them for up. If you only have on/off control of their outputs you can "pwm" them to do propulsion and steering.

----- http://mindspring.com/~benbradley
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Have you considered looking at adapting aquarium pumps? They function in the exact manner you describe, are waterproof, and cheap. Plus, they're easy to maintain. (http://www.marinedepot.com/md_viewItem.asp?idproduct=TA3160 -- that pump is a brute at 1000GPH, smaller volume == lower price.)
Unfortunately it doesn't resolve the issues with the bulkhead leaking...
How deep is this unit intended to go?
Steve

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Take a look at the specs page, and see the maximum lift distance. The meatiest pump they sell can only lift 13.5' which is not enough for even my short-range target of 3m. Like I said, the obvious problem with the pump method is that either the pump has to be able to push against the full head of pressure outside the hull, or you recreate the problem of having to move the wires through a bulkhead (plus in that second case you need the pump to be totally waterproof). Oh, and those pumps are 115VAC, which means I have to deal with inverter losses too.
There are probably pumps used in some industrial application which would run off 24VDC and meet the pressure criteria (or I could adapt a pump with a different motor). Not sure about this scheme though, I get the feeling control is more difficult. I'll look at it, though - no solution too crazy.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Perhaps look for 'borehole' pumps - some are designed to work a long way down ?
Incidentally, I would guess that inverter (electrical) losses would be minute compared with any hydraulic system - small ones in particular are terrible. At the sort of pressures you are talking about, getting electrical connections through through a pressure hull is not a problem, although the fewer the better.
As others have said, your cheapest and simplest option is to use simple dc motors, running flooded with oil or distilled water. They will not last for ever, but you will be able to replace them several times for a fraction of the cost of other solutions !
Lastly, the kind of pump used for central heating circulation is often magnetically coupled, so perhaps you could hack one of those.
Dave

Posted Via Nuthinbutnews.Com Premium Usenet Newsgroup Services ---------------------------------------------------------- ** SPEED ** RETENTION ** COMPLETION ** ANONYMITY ** ---------------------------------------------------------- http://www.nuthinbutnews.com
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Have you looked at submersable well pumps? They are made to be in water and their wiring is sealed against water and pressure. If you were to mount it in a well passing along the bottom or center of the RUV the outlet pipe would be your thruster. Some are available in 12dvc IIRC and some can lift water 100+ feet.
All you would need to do was connect the power through water tight connectors.
Also have you checked the web sites of the RUV competions teams? Perhaps they have parts lists and design problem discussions and troble shooting tips. IIRC their are several competions for remote underwater vehicles for high school and coldge students.
Charles L.

electrical
for
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
For prop; The aquarium pumps have an impeller attached to a rotating cylindrical magnet with a hole down the center for a ceramic shaft - this assembly is completely submerged. The stator - a coil usually driven by 120V @60Hz stays dry on the outside of the plastic housing around the rotor - and essentially 'outside' of the water system. Adapting this to the sub - imagine the impeller is replaced by a propeller mounted to the magnet - the coil stator stays dry inside the submarine - all you would need to ensure is that the hull material at the interface between the rotor & stator isn't ferromagnetic. It's simple to rewind this coil for lower voltage/higher current etc. Since the propeller is completely outside the sub, all it does is move water - at whatever depth - lift specs are irrelevant - or am I missing something.
Kevin
p.s. 13.5' = 4.1m

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hi Bob,
Thanks for the reply, some interesting material there.

This works OK for the propulsion, but not the control surfaces. I don't know of such a thing as a waterproof stepper. Is there such a thing?

How watertight are these, though? Can they survive a 45 meter swim? Cost of replacement? MTBF? I realize you might not have those numbers to hand, but a ballpark would be nice. The simpler phase of this operation is rated to 3m with a margin to 5m to allow for momentary loss of control, different strata of water, etc. The harder phase is rated to 30m with a margin to 45m (the reason for the considerably wider margin is that the "big" project will run faster, and will also have dynamic depth control, and a big margin is nice to allow for control problems). IOW it has to run reliably at 30m, and it has to be able to emergency surface, at least, after going down to 45m.

Err... compared to what? One of the main reasons real submarines use hydraulics for their control surfaces is because they are quieter than electrical actuators, or so the US Navy training manuals would have me believe :). Or were you referring to the main propulsion motor?

No. "A"UV - autonomous. Once surfaced it can be guided at short range via radio, but when it's submerged it's on its own intelligence entirely.

I specified freshwater mainly to distinguish it from a deep-sea operation. This device will be operated in a lake which is not used for either drinking or irrigation. It's so dirty and polluted that the fish order in Evian.
I have looked at buying one of those divers' sled type things - I'm not sure of the correct word for them, but they're a thruster for SCUBA divers - but they are fantastically expensive.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I was refering to the propulsion motors. Steppers don't have brushes so oil compensating is definately an option.

Do a search on the MATE ROV competitions and look for someone using them. I haven't used them myself but know several of those teams use them and I suspect they would be reasonably reliable to 5 meters. I'm more familiar with 6000 meter vehicles.

Errr.... compared to the DC Brushless propulsion motors. Big difference between your "large" AUV and a Navy submarine.

Hope you can afford to kiss it goodbye when it goes astray (which it will). All the AUVs I've worked with have an acoustic control backup system to get them back on the surface when they go bad by dropping a weight. How do you navigate without using acoustic transponders? Do you have an inertial guidance system? Dead reckoning? If your task is to do a sonar survey then noise may well be an issue.

I take it this is a one-off project and not something you hope to take to production. Have you looked at trolling motors? Again, I have no experience with them but some people do use them for ROV/AUV projects.
Woods Hole is developing a hybrid AUV/ROV which is full ocean depth rated and will only have a single mode fiber connection to the surface for video and control. It will run automously if the fiber breaks. James Cameron did something very similar on his Jake and Elmo ROVs.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The device has basic inertial navigation (i.e. it's not accurate and I don't trust it), but it's designed to do the "accurate" navigation using GPS on the surface, then dive and gather a stripe of info (not sonar, but video/temp/pressure/vehicle engineering info using a camcorder to gather the images, and writing the telemetry data onto the audio track of the videotape). The second-phase project uses computer-controlled cameras with motion detection to wait for and then watch "objects" (i.e. fish) that swim into view. That's why it requires dynamic depth control.
There is a recovery system built in which releases a styrofoam marker buoy tethered by 1000 feet of fishing line, also releases a ballast weight, and starts an acoustic beacon. The recovery system trips if the depth exceeds a threshold, or if the main battery goes offline or undervoltage, and also if a couple of other conditions occur (watchdog bite on the main micro, for instance).

Right.
Thanks a lot for your suggestions, it's given me several avenues to explore.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 21 Oct 2004 01:35:31 GMT, "Lewin A.R.W. Edwards"

Expensive but not way out of line. Firebox in the UK has them on for about $200-$300 US Maybe one of these would work as main propulsion and then you just have to worry about a rudder type mechanism.
http://www.firebox.com/index.html?dir=firebox&action=product&pidA3
Cheers V
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Lewin A.R.W. Edwards wrote:

How can this be? Worst case couldn't you run wires inside conduits which look very much like hydraulic pipes?
Mitch Berkson
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Sooner or later, the wire has to exit the pipe. Labyrinths help to seal it, but they're a weak point. There are connectors that go across a bulkhead, but they're REALLY expensive rated for these depths (like, $200 per side). You also then hit the situation where there is a big pressure differential across the wire's insulation.
The reason hydraulics are so attractive is because you just drill a hole and screw a double-ended through-bulkhead barb, and optionally weld or solder the joint. You can run the hydraulics with a static pressure that matches the surroundings, and there is no pressure differential between the hose and water. This mandates a well-sealed hydraulic system, but that's easy to achieve (since it is a design goal not to spill oil anyway :)
At least, this is how I've analyzed it on paper. I'm not experienced with hydraulics in general, though I've used some small pneumatic actuators many years ago.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.