ROV Thruster Design Questions

You might have some fun researching Kort nozzles as a thruster design; I suspect they will be more efficient than pump designs. (Tugboat Annie would love this.)

If your 20 amps are distributed equally, 12V × 5A = 60W. But a search on "60 watt DC motors" turned up more irrelevant than useful stuff. (I didn't persist long, though.) I wonder if windshield wiper or antenna extension motors would suit; advantages would include availability and price.

Good luck with this competition. Please keep us posted.

thanks for the ideas.

We're actually looking at Rice nozzles for this thruster. These are specially contoured to be even more efficient at both forward and reverse thrust than Kort nozzles. We were planning to get these 3D printed on a MakerBot type of device, but we're open to other ideas and options.

I've tried to find 60W motors, but many of those are designed to operate at a higher voltage than we are allowed to used, or do not meet our criteria in some other way.

We've looked into windshield wiper motors and other ones like that, but they just dont have the power and speed we need. I found these motors (http://www.islproducts.com/component/virtuemart/4-motors-micro-dc/5-planetary-geared-motors). Does anyone have any experience with them?

Also, do you have any experience with the type of seals needed for the shaft on a project like this? Thanks for all of your help.

I'm not familiar with Rice nozzles, but what you say makes sense. One possibility for small propellers would be mixing props that are commonly used in barrels.

A lip seal is simple and economical, but has some friction as it grips around the shaft. A mechanical seal, as used in pumps, might be better in this respect, but is more complex and costly.

As for 12vdc motors, I don't yet know if the best idea is to find a motor of the "right" rpm already, or to use a higher-speed motor with gear reduction. If geared, planetary sounds good.

check CROUZET motors

crouzet.com

If you need 1 ask for a sample

For the seal I would suggest the V-seal (see above) put on the shaft water side. It has a lot of advantages with respect to the radial seals: tolerant for any deviation between shaft and bearing housing (of course in a limited frame but limits are by far >> than for radial seals).Have you thought to use pumps and ejectors ? Have at least a look at the possibility.I have not found the profile for the Rice device. I would like to compare to the other.Las but not least look at motors as MAXON which are very compact. If you do not find a motor for the power you may use 2 with a gear so that they both act on the shaft. For your power gears available of the shelf can be used. You can use gears cut from polyamid or other synthetic materials and small gears made from SS.

Thank you all so much for all of the input.

Here's a site with some information about the Rice nozzle profile. http://www.propellerpages.com/?c=nozzles&f=How_Nozzles_Work

It's a lot newer than the Kort nozzle, but it seems to offer significant improvements in thrust. We're trying to end up with a thruster nozzle similar to this one.

For the prop, we were thinking about just asking the company whose ROV is seen above (Videoray) for some of their props, lol. The alternative would be to find some sort of impeller designed for use in Rice nozzles, which may be more trouble than it's worth.

nickname, what do you mean by using pumps and ejectors? I've looked at Maxon and Crouznet motors, and although some would work, they are all a bit beyond our price range (especially when multiplied by the 4 motors we need, plus the 4 we need as backups). Also, a lot of the gearboxes I've looked at are way more expensive than the actual motors, which is a bit problematic. I'm looking at gear motors too though, and, as always, I am open to suggestions. The V-Seal looks more or less perfect, although I was wondering about how a seal like that has to be mounted. What would the diffrence between a v-seal and a mechanical seal be?

The circular lip seal is VERY sensitive to eccentricity. Without a spring it is not very efficient and Teflon has no own elasticity. The V-seal (which I used many times) is not any more sensitive to eccentricity, it can be easily assembled on the shaft and has its own elasticity, further more the pressure on the seal edge is dependent on the sealed pressure reducing the wear.

Nozzles are good for reducing tip losses and preventing damage, but they are only worthwhile if you have bad clearance issues limiting prop diameter. Almost all boats are designed for shallow draft, not efficiency. It is always better to move more water than to move it faster. See "Froude Number." Pushing on moving water is like sticking the back wheel of your bike on a treadmill. You have to use a higher gear, which gives less push for the power.

In Human Power boat racing, we use props between 30 and 50 cm diameter for 100 watts at 6kt. Lower speeds would want more diameter. BTW, a model aircraft prop running backwards only looses about 15% efficiency. If you still want a ring, you can cut the tips off a standard prop - be fussy about minimum clearance.

If you can pry them loose from the suppliers, you can get teflon lip seals, but the biggest gain in reducing friction is to remove the tension spring from the seal, and watch for wear, alignment, etc. Minimizing the shaft size helps too.

For hydrodynamic efficiency, you want the largest diameter propellers you can fit, and well-shaped foils. Fortunately, you can probably find something close to what you need from the APC model aircraft line for about fifteen bucks each. Bigger and slower would be better - why choose such a high shaft speed?

The extra challenge of using brushless motors would be very appropriate for a student project.

It would be interesting to hear what the objectives are: speed, maneuverability, combativeness, etc. This example might resemble "Robot Sumo", but maybe other goals are more important.

I have no idea what parameters you're constrained with but my 1st thought is why 4 motors? Why not one big motor, 4 solenoids and a positive displacement pump attached to the motor. Direct the power where you need it. Forget about shaft seal issues all together. With water jets you'll have less drag than all the guys with 4 props hanging in the water wherever they go.

Just out of curiosity why can't you use brushless motors?

The reason I ask is that if you go down the brushless path you can keep all the control digital and negate the need for things like the analogue control of current/voltage. You can also get rid of a whole lot of feedback provided you size the motors correctly.

These are students, I bet that's exactly why they want them to take a bit tougher road...to learn a few things

Another thing to consider with brushless motors is the ability to use an oil-filled pressure-balanced housing that helps mitigate pressure difference across the seal.

Brush-type motors do not operate well immersed in oil.

Some background information I should have included earlier: The ROV is about 20"x15"x12" or so. We are trying to maximize the speed of the ROV for the upcoming competition. In the MATE contest, ROVs must complete a series of tasks underwater in a timed setting. This year's competition was about shutting off a simulated oil leak. Details are avalible at http://www.materover.org/main/index.php?option=com_content&view=article&id=117&Itemid=171. Because of the time limits, getting a robot that is as powerful, quick and agile as possible is very important. Here's a pic of last year's robot, with the modified 1100 GPH Rule Bilge pumps acting as thrusters (we basically stripped it down to the waterproof motor cartridge and put a prop on it).

The hose in the background was part of one of the mission props. Using 4 motors (2 horizontal-forward/reverse and 2 vertical) lets us have a great degree of control over the robot. It can turn on a dime and also tilt quite well, which is great for picking things up off the floor. The yellow thing is a camera, and the thing in the front was our manipulator.

The primary issue with brushless motors are the ESCs needed to operate them; we can't really afford another $50 per motor. That and our electricican is adamently opposed to adding ESCs to the current system (it's currently just a PWM system), as it would increase the volume of the pressure housing on the ROV, which would displace more water, which would require us to add more weight to compensate for it, which slows down the ROV. Therefore, brushless motors are sort of out of the question for mostly practical reasons.

As for the seals, I'd like to use something that is relatively forgiving but will still seal nicely down to 20 meters or so (~66 feet). Would a V-Seal work for this

• "The primary issue with brushless motors are the ESCs needed to operate them; we can't really afford another $50 per motor. That and our electricican is adamently opposed to adding ESCs to the current system (it's currently just a PWM system), as it would increase the volume of the pressure housing on the ROV, which would displace more water, which would require us to add more weight to compensate for it, which slows down the ROV. Therefore, brushless motors are sort of out of the question for mostly practical reasons."

Brushless DC motors are in effect a type of stepper motor that are driven by energizing a series of coils one after the other. The faster you switch between coils the faster the motor turns and by reversing the sequence you reverse the direction of rotation.

The end result is that all you need to control a brushless DC motor is one power transistor per coil and that's it. Provided you have sized the motor correctly you don't need any feedback because you know how fast the motor is turning by the frequency you are stepping from one coil to the next.

Yes you can buy off the shelf brushless DC motor controllers but they are a doddle to build yourself.

I'm assuming that you have some sort of microcontroller based control system within your ROV so using brushless DC motors makes a hell of a lot more sense and would probably reduce the volume of the electronic package considerably as a couple of transistors take up bugger all space, particularly if you mount them on the outer case of the watertight enclosure so you in effect end up using the entire pool of water as a heat sink for them.

I'd suggest reading up on stepper motors and then reassessing the situation from there keeping in mind that brushless DC motors are just a type of stepper motor.

I haven't done this myself but if you make all the connections to your brushless motors watertight then you may be able to get away without seals and just let the motors flood with water. This would degrade their performance somewhat but on the flip side the water would act as a coolant on the coils and allow you to overdrive the motor thus regaining some if not all the lost performance not to mention doing away with the seals and associated losses. It's just a thought but it might be worth investing in a small brushless motor and giving it a go because it could save you a lot of headaches.

Two good sources for special motors are:

Plettenberg Motors and also Torcman Motors. Not too sure if they are going to fit into your budget..

If you are going to dive down to 20 meters I would also suggest you look at liquid-filled pressure compensated housings rather than a lip or mechanical seal. Will you be operating in fresh or sea water? This would also influence what you should use.

I've been working on a some what similar personal project. I started looking at radio controlled model web sites. You can get quite a few cheap brushed motors, but the problem with them is, they are all high speed. I've found much to my sorrow that low rpm means or $$$. So I decided to use the motors from cheap cordless drills, which come complete with gearboxes. You can get very cheap propellers online too. About $12 each of 70mm dia.

How well have the cordless drill motors worked? How did you seal them?

Brushless motors seem to be a nice option, based on what everyone has been saying. Does anyone have any suggestions?

If it were my project I would definitely be going down the brushless/stepper motor direction for the following reasons:

• They interface with microcontrollers very well and the interface circuitry is fairly simple.
• If sized correctly they don't require feedback because you know how fast they are rotating by how quickly you are stepping from one coil to the next. They also come in a wide range of sizes from those that drive the analogue hands in digital watches to ones that can power a car.
• You have very precise control over them. Every time you step from on coil to the next the motor will rotate through a precise and predetermined angle. You can then control how far something moves just by counting how many steps you have put the motor through.
• Being brushless eliminates the brushes and commutator which are one of the most failure prone parts in a motor.
• Sealing the motors if required at all is going to be much simpler.

There's probably a whole host of other things that I can't think of at the moment but stepper/brushless motors are cropping up more and more particularly when you have digital control going on which is now just about all the time. Actually you can probably get hold of some motors to experiment with by pulling apart some old printers. Both the carriage and paper drive motors are nearly always stepper motors for the reasons I have detailed above.

That's my opinion but then again I'm a stepper/brushless motor advocate so I'm biased towards using them but the final decision is up to you. However, I definitely would recommend getting hold of a couple of old motors and having a play with them before making the final decision.

On my project all my electronics are installed in a water tight container (it all needs to be self contained, no umbilical). I have not had a chance to try out the motors yet as I'm still waiting for the stuffing box to be made. On a plus the cordless drill motors do come with a ready made gearbox and clutch.

What are you using for propellers if I may ask?

Are your controls analog or digital?

You may be interested in this. It may save you some time building it from scratch.

Four blade ducted propeller

Some sort of stepper/brushless system really seems to be a practical solution for this system.

Robin, for the past few years we have been using model boat props (some sort of 2 bladed octura model). these have worked pretty well, but we need to use something designed for use in a nozzle with our Rice nozzles for the next year. I think we'll still go with a diy approach. Might I ask, what sort of project are you working on? What depth are you operating at?

It's a diver thrust assist pack that fits onto the divers bottle. Not sure yet on the max depth but probably not more them 40m. I'm making it because I can't use my dive scooter and camera at the same time and with the drag of my camera any current stronger then 1 knot make it very hard to remain on station. The link to the website I posted all so have 4 bladed ducted props.

Okay, that makes sense. As a diver, I can relate to how difficult it can be to multitask underwater, especially in current. What sort of battery system are you going to use? Are drill motors powerful enough for what you need? If not, I'd suggest looking at electric scooter motors, which should be more than powerful enough in almost any current.

Hi, do u still need help in the ROV thruster design? I had been designing & producing ROV thrusters. Let me know if you need my advice.

We've more or less dealt with the thruster issue for this year, but I'd still love to hear what you've got to say about the topic for future reference.

If you're interested, we ended up using modified bilge pumps again this year, but out fitted them with 60mm Kort props and custom 3D printed Rice nozzles, which have made a significant improvement in performance.

We've been maintaining a website at tachyonrobotics.blogspot.com, if you want some pictures of how we've executed it or videos of our robot in action.

Hi, since it's just a school project, I'm glad bilge pumps works well for you. However, if you intend to build an ROV for work purposes, then bilge pumps will not work in deeper depth and you will need motor to run the thruster. Kort props are excellent and they works great for most ROVs.

I had some other classified projects dealing with thrusters with military involvement, therefore couldn't talk much over it.