Author Topic: Magnetic Couplers (Read 4906 times)

JavaMoose · « **on:** August 10, 2005, 09:25:58 PM »

Hello all,

It seems Hydro kinda fit this question the best. I built an Underwater ROV last month, and I am not in the process of making the thrusters for it. There are half a dozen ways to seal a shaft from an electric motor...but all of them have the potential to leak. So, that got me thinking of other options...

One option that seemed interesting was using magnets to make a coupling between the motor and the prop. This approach has two main benefits; (i) If the prop gets fouled (seaweed, etc) the motor can still spin, albeit with some drag and (ii) the motor can be sealed in a housing with no need for a risky/expensive shaft seal.

In looking on the web, I found a mfg that makes a stand to convert your bicycle into an exercise bike - and he uses 6 NdFeB (per side) to connect the bike to a viscous clutch. According to what they say on their site (grain of salt here) it can take 50hp of startup thrust before the mags will slip. If that is close to true, it may answer my next question:

Does a well-designed magnetic coupler using 4 or 6 NdFeB magnets (per side), with about a 1/4" gap, simply have the strength to push a prop through water (800 times more viscous than air) with decent speed/power?

Also, anyone who is interested can see my ROV project log: HERE

bob g · « **Reply #1 on:** August 10, 2005, 03:59:36 PM »

the short answer as i see it is no, it wont work, at least with any appreciable power output.

the longer answer is the why,

if you consider that folks on here are building 12 inch brake rotor mills, which have a fixed stator, and rotating rotor, which means no transmission of torque is allowed to the stator(non turning element)

the amount of power delivered from the prop and transmitted to the stator and mounts is likely one heck of alot less than what you need to drive a boat prop,especially at 1/4" air gap.

i think your only alternative is using a seal/shaft arrangment,

i really dont think you can transmit much over a few hp with this arrangment, and likely alot less than is needed.

but an interesting concept.

bob g

Phil Timmons · « **Reply #2 on:** August 10, 2005, 04:45:30 PM »

cool toy.

Just for another perspective, I think you should be able to make some sort of decent magnetic coupling. Not sure of all the possible down sides, yet. Anyway, I have a small pile of magnetic disks from old floppy drives and may play with some configurations.

For your toy, you do not need a hugh amount of power transfered anyway, do you?

How did you come up with a 1/4" gap? Is that the typical wall thickness of the PVC you are using?

Have you also considered hydraulic power transfer?

Shadow · « **Reply #3 on:** August 10, 2005, 05:19:37 PM »

Awesome piece of work! Very ingenious stuff! Be nice to see some footage of it in action.

JavaMoose · « **Reply #4 on:** August 10, 2005, 05:20:27 PM »

Yeah, I have some PMs from some old hard-drives - I figured before I fabbed something up to test I would ask.

Correct, given that it will be neutral boyant - it doesn't take a lot of power to get it moving with a decent speed. Also, since this isn't going into the ocean (just small lake and inner-harbors) it doesn't need to overcome any serious thrust.

I got the 1/4" from the thickness of the PVC wall, yes.

I had thought of doing a mini-hydraulic system (hydraulics are what the big-boy ROVs use) but I think it might add quite a bit of complexity. I need 1 motor, and hydraulic pump (maybe a power steering pump?) - those parts are easy and cheap. However, a hydraulic motor is not - they get big and expensive quick! There is also the issue of solonoid operated valves for the system - but it seems that they can be had cheap on eBay. So, I haven't ruled it out yet - but it is certain to be tricky.

windstuffnow · « **Reply #5 on:** August 10, 2005, 05:22:08 PM »

I'd say it's very do-able! I've seen some presurized stirling engines use this method to transfere the power while holding 2000 psi inside the case. You would want to make sure the magnets were spaced very close to each other otherwise you'll get some "slippage". It's easy to break a one on one magnet but when you get several closely spaced you'll find it's near impossible to get them to slip (lots of torque).

This of course depends on the magnets your using... lots of magnetic surface area.

Nice looking unit!

.

JavaMoose · « **Reply #6 on:** August 10, 2005, 05:26:31 PM »

Thanks! There are three videos done already...without thrusters it was just static - but some neat video (IMHO) none the less.

Here, save you the trouble of finding them:

JavaMoose · « **Reply #7 on:** August 10, 2005, 05:32:20 PM »

I was planning on using (from Wondermagnet, of course!) these 1/2" x 1/4" NdFeB discs. Seems like they would be a good size, without getting into the really big $20 each ones...

JavaMoose · « **Reply #8 on:** August 10, 2005, 05:54:10 PM »

I don't know, seems a bit different with a wind genny. There you are spinning magnets past coils (ie a generator) - sure there is some resistance from the field generated (cogging) - but you aren't matching up magnets on both side with the intention of getting them to lock.

Just in case anyone wants to see, here is the bicycle set-up that I had mentioned. It seems it is a viscous clutch-style for putting drag on the bike, and they say it operates up to 50mph (average 25mph). I would assume that the oil they are running this unit it, while maybe not as much drag as a prop submerged in water, must create a decent amount of drag.

Shadow · « **Reply #9 on:** August 10, 2005, 06:12:27 PM »

Forgot to add another comment earlier, what about a small submersible pump for thrusters?A couple 12 volt bilge pumps or similiar?

Phil Timmons · « **Reply #10 on:** August 10, 2005, 06:13:56 PM »

Thinking CHEAP -- hydraulic could be water, too? Like a little hand drill powered pump -- $10 or so at Homer Depot? Connect it with garden hose? I guess you could have solenoid valves on the rig if you wanted multiple "motors" (the little pumps used in reverse?). I have some washing machine solenoid valves to give you, if you want.

I looked at your tech-forum discussion -- someone mentioned submersible 12 V? pumps -- the output is the thrust . . .

Your control line is from the top side, right? So you have a hard line going down to the ROV, correct? And I guess you use a water line (garden hose?) from the top side, down to the ROV, so you could even have a pump on top side and let it exhaust as a thruster on the ROV?

JavaMoose · « **Reply #11 on:** August 10, 2005, 06:17:44 PM »

We had looked at those, and the one issue we found is that they are really only made to be in a few feet of water. With our tether length and the connector we are using - we are safe for at least 32ft (getting a new connector that is good to 100m as we speak) and can go up to 200ft straight down. Yes, our camera tube could handle that depth no problem...

So, I was worried that the bilge pumps just couldn't handle that depth...which is a shame - because they are inexpensive and fairly powerful little buggers!!

JavaMoose · « **Reply #12 on:** August 10, 2005, 06:21:45 PM »

True. Hey, I forgot about those little pumps, those things are neat (we have one up at the camp)! You have to be careful with the tether, make it too thick and it creates a lot of drag and you need more thrust to move - it also can throw off your balance (positive or negative versus neutral balance). So, I think running a garden hose might be a problem.

However, the idea of using them for thrust is an interesting one...

See my lower post about bilge pumps...

JYL · « **Reply #13 on:** August 10, 2005, 06:23:02 PM »

Hi,

It will work if the magnet are big enough for the torque you want to handle. Obviously, with only 4 or 6 magnets, so far apart, this can be only a few Newton (in-lb) if your magnet are too small/weak/cheap.

Actually, if you are going for 10 or 20 watts motors, if might be easy. If you are shooting for KW, low revolution... This might needs some more try and error.

If you can reduce the GAP between the two plates, it will be best. 1/4 Inch will be a problem with smaller/cheaper magnet. But, try it.

You can do a search, on Internet, about "Magnetic Clutch". You may find some examples since their are many possible configuration: Plate, shaft and tube, Eddy current type where only one plate has magnets, etc...

Actually, you can build your own Electric Motor with the same principle use for the Disk-Brake generator. (Magnet and Rotor can be put in water proof fiber glass resin) However, to work, the motor need 3 phase AC. So, it might not be a cheap project.

Chagrin · « **Reply #14 on:** August 10, 2005, 10:07:49 PM »

Google for "magnetic drive pump" and you'll find zillions of examples that show it will work.

Norm · « **Reply #15 on:** August 11, 2005, 05:35:45 AM »

You could use roller wave couplers a set of

rollers on a yoke opposing each other something

like the ends on a drive shaft universal...but

with a diaphram in between would work also...

but that's not what you asked, so ....yes!

magnets will work...also you could have a torque

switch that would operate an alarm or light if

the prop became fouled.

Have Fun!

( :>) Norm.

halfcrazy · « **Reply #16 on:** August 11, 2005, 10:27:38 AM »

i had a dc circulator that had magnetic drive forget the brand but there where 2 brands i think they where from backwoods solar both where magnetic drive the one i had pushed water thru 6 sets of 200 ft 1/2 radiant tubing so they must have some power?

Ungrounded Lightning Rod · « **Reply #17 on:** August 11, 2005, 01:25:07 PM »

Let's see:

A motor is a magnetic coupling between a virtual rotating magnet and one fixed to a shaft. You then plan to use this to drive real rotating magnets to spin a rotating shaft on the other side of the wall.

Why not eliminate the middleman?

Build a motor with the coils INSIDE your seal (on the stator laminates with the pole pieces up against the wall) and the rotor OUTSIDE. Spin the magnetic field eletrically. Then you're done.

All the standard electrically commutated motor design stuff applies.

You might also think about embedding the coils in epoxy/fiberglass material - like a genny stator or hull - and build a motor that will work despite being submerged, because it's thoroughly sealed against the surrounding water and its contaminating ions.

twidget · « **Reply #18 on:** August 11, 2005, 01:55:27 PM »

If you can get stong enough magents the magnetic couple will work. I used to work in a paper mill, where we used lots of nasty corrosive chmicals. We had a bunch of pumps that wre coupled to the motor with a magnetic couple, up to 5 Hp if I remember right. with real hard starting torque. they worked great.

Just my $0.02

Twidget

JavaMoose · « **Reply #19 on:** August 11, 2005, 04:53:58 PM »

Yeah, I had thoought about that too. I wonder how thin the side-walls of a tube with that design would need to be, though...

JavaMoose · « **Reply #20 on:** August 11, 2005, 04:55:38 PM »

Wow. I'm sure our motors, atm, will be WAY under 5hp.

OK, I am going to use some magnets I pulled from some junk hard-drives to build a prototype this weekend. We will see how it goes...

JavaMoose · « **Reply #21 on:** August 11, 2005, 05:02:07 PM »

Lots of good info from that search. I was thinking of two flat disks with magnets on them - but I found a pump mfg called Magnatex that had a really good technical drawing of how their magnetic coupling works.

Check it out:

hiker · « **Reply #22 on:** August 12, 2005, 12:38:39 AM »

years ago some cars had electric magnetic drive clucthes..after awhile thou the magnetic

powder they used in them would break down and cause slippage..that was back in the 50s

i belive...

Norm · « **Reply #23 on:** August 12, 2005, 02:34:24 AM »

Like a little hand drill powered pump -- $10 or so at Homer Depot?

Have you ever used one of these pumps...for

years I had wondered about these pumps...finally

I bought one from BigLots ...if the other pumps

work like this one did be sure and get your

money back...I was disappointed to the extreme!

What a bunch of crap it is!

I shoulda took it back...but it was only

$3.98...imagine the thousands of victims that

have said the same thing....

( :>( Norm.

kitno455 · « **Reply #24 on:** August 12, 2005, 04:15:50 PM »

i know this is on a much smaller scale, but i have a pet fountain that uses a small impeller direct drive to the motor. i dont know what you would call it, but its a bit like an induction motor, with a brittle ferrite armature, with a small hole thru its length, that runs on a stainless rod. no brushes, no bearings. all of this is in the water, in a 'pocket' in the side of the pump. the rest (stator) is epoxy dipped. it coggs quite a bit, and it really 'snaps' into place when you slide it down the rod, making me think it is permanent magnet with iron core stator.

seems like you should be able to do something similar with very small induction motor, provided you can find some way to protect the armature from corrosion (epoxy coated ferrite?)

you would also need some way to move your magnetic field in your coils, my unit is AC power.

allan

Phil Timmons · « **Reply #25 on:** August 13, 2005, 01:17:38 AM »

No I do not think I have.

Thanks for the info, Norm.

They are just total garbage, huh?

Do you think maybe you wound up with a just too cheap knock-off?

Dunno either way . . . but maybe like the cheap, cheap, Chinese fake knock-off power tools that are made to look like Makita, or DeWalt? I think something like "Ohio Forge" comes to mind.

But yeah, I think crappy stuff is made with the premise that the customer will just say "awww, forget it," and not get their money back.

JavaMoose · « **Reply #26 on:** August 15, 2005, 01:15:42 PM »

I think it depends on your expectations of them. We use one with an electric drill to pump drinking water from carried vessels to our larger tank. Works just fine for that, and is way easier than lugging a 25gal jug 8ft up to pour it into our holding tank (gravity fed faucet for freshwater).

News:

Author Topic: Magnetic Couplers (Read 4906 times)