Magnetic Coupling of Mecahnical Energy

The soft irons can not touch the two pulleys, there has to be gap (air or some ftrictionless material like teflon)

There will be lower magnetic field through the soft iron due to the increased reluctance. And this will restrict the power transfer.

For strong enough magnets, money will be an obstacle.

Time lags and "cogging" are likely to introduce instability, such as skipping and other chaotic behavior.

"Canned" magnetic drive pumps, on the other hand, are successful; but their basic geometry is different, and the magnets don't "jump" from pole to pole.

Thank you for your constructive comments. What I have suggested is just for BRAIN STORMING. In fact I thought over comemnt made by a Guest earlier - and I felt - we have hundreds of horsepower motors at high speeds too. They have airgaps too. So that is really not an issue. That is right - we have to keep total reluctance of the path low. So we have to keep lenght (effectively airgap) small, keep area (of contact) large. F = K m1*m2/(d^2) which is standard equation for force of attraction/ repulsion between two magnets. So strenght of two magnets, smaller size of air gap and larger area of coupling forces. In my diagram I have shown only a two couplers- whereas a belt is in contact with many teeth of a gear- so that is another way of improving coupling.

I do look forward implementing this at low cost- as you have mentioned cost may be another aspect - which should be borne in mind.

This sort of thing is sometimes done to couple a motor and pump impeller, to make a fully sealed pump.

1. What do you mean by "far"? For magnetic fields to transfer significant power, "far" would be a few millimeters.

2. When you compare this to a belt, you are implying pulleys, meaning two parallel shafts separated by a distance greater than the sum of the pulley radii. Magnetic forces follow the inverse square law, so any magnetic interaction between the two disks will be in the region where the two disks almost touch, NOT between the tops and bottoms of the disks.

3. For this coupling to work, the two disks must be almost touching all around the periphery, and their shafts must be very close to a common axis. Any misalignment will increase the gap, and weaken the coupling.

4. It would be possible to transfer energy from one disk to two others, if one of the two had a hollow shaft with the powered disk shaft inside the hollow one. More than two will NOT work, due to the large gaps.

5. I believe introducing a magnetic fluid ( a slurry) would cause more fluid friction losses than it would improve coupling. It would certainly cause wear problems.

6. There are no teeth! Only magnetic poles.

7. You don't want repelling forces, only attractive ones.

8. Advantages: Total seal is possible between energy source and output (Magnetic stirrer.) Limitations: Very weak. For parallel widely separated shafts, as I believe you envision, it simply will not work.

• Thank you for your very elaborate, constructive comments.
  
• I am trying to develop applications involving magnetics. Hence this kind of ideas keep cropping up.
• I am looking for belt drive replacement applications.
• Hence distance between pulleys is fairly large- non contact type.
  

• In belt drives, you need the two pulleys to be in same plain and shafts have to be parallel. To transfer mechanical energy to other plains or non parallel shafts involves far more complex arrangements. So magnetic circuit based coupling should be preferable in such applications.
• In motors - you have two air gaps between stator and motor in the magnetic circuit path and you can have motors with hundreds of HP. All that I am suggesting is two wheels or pulleys with two air gaps with a low reluctance magnetic path - which is called soft iron or something which does not get permanently magnetized (low remnance).
• I have seen magnetically coupled applications for vacuum pumps and chemical stirrers etc where the stirrer and magnet are just separated by a thin wall.
• Magnetic poles act like discrete teeth. In fact you can mold teeth to any shape and magnetise in any direction you need.
  

These thoughts are only to guide discussions towards finding magnetic coupling applications.

"Begin with the end in mind." -- Stephen Covey

After reading again, I see that I missed the soft iron bars, a critical part of your construction. I thought the black lines represented forces acting between the tops and bottoms of the wheels - sorry.

Here is something that might be worth trying:

Notice that there are TWO soft Iron pieces separated by brass at top and bottom (think about a magnetic vice). Two are required so the system 'knows' which direction to rotate. As I've indicated, it should be possible to have the two disks of different sizes, although I'm not at all sure that would be an advantage. As I see it, the number of poles must be the same on both disks, in order for them to remain synchronized, and for the forces to result in a true force couple on the output disk. This means that it could NOT be used to change speed of shaft rotation, as a pulley system can.

Thank you for primafacie confirming that it is worth trying out.

You have added two soft iron pieces - my own guess is may be more. But in terms of basic concept - you are right - to couple North to south and south to North from Wheel 1 to wheel 2.

Biggest advantage is - when the two shafts are not aligned and off set in 3 dimension. Further there is no constraint on power or speed.

Further I want you to think over the constraint of 1:!. There is no need. You can really have N1/N2 ratio like with any gears or belt drive.

Hi,

A.: this is equivalent to a 1-phase synchronous motor drive.

So direction of revolution (left or right) will be undefined.

You will need either 2 or 3 phases.

B.: There is no closed magnetic circuit in the coupling from wheel 1 to wheel 2, thus a very bad magnetic effectivity. There would be a need to use 2 magnets as one "tooth", one north, the other one south-oriented. And a soft magnetic structure inside the wheels to couple these 2 magnets at their back. This together with 2 soft iron couplers at each side will give a quite good synchronous coupling.

C.: Individual magnets are not a necessity, you may use 1 central magnet (magnetised axially) in each disk and two pole-pieces at the sides with soft-magnetic fingers. This arrangement will be much cheaper and also smooth in operation as attaching and detaching of a tooth is generating impulse forces.

Make a trial, may prove good.

RHABE

ALL - Your observations are well taken.

The magnetic paths need to be completed from one wheel to the other through the couplers and again within each wheel.

As regards unceratinity in direction of rotation - note this is similar to a belt drive. So one wheel is driven (from some source) and the other has to simply follow- whichever direction.

Further I also wanted to bring another analogy with belt drive - there will be slippage if the load is more. For light loads, slippage will be nearly negligible.

So will this magnetic oupling be equivalent to a timed or toothed belt or a simple smooth belt - where slippage could be high.

Hi,

"As regards uncertainty in direction of rotation - note this is similar to a belt drive"

This is similar to a synchronous motor that has a phase lag from input to output if loaded and stops totally if overloaded.

"As regards uncertainty in direction of rotation - note this is similar to a belt drive"

There is no slip if load is within acceptable limit.

DKwarner is correct with his sketch, only the distance between the two soft magnetic bars shall be a bit bigger. The 2 parts can be displaced radially or axially or at 2 sides of the arrangement.

Your arrangement reinvents the "electric transmission synchro" that was in use since 1916 and is still in use but with a servo amplifier added.

Get a book on synchros and resolvers for deeper insights. This solution works without permanent magnets.

RHABE

I did an experiement in synchro transmission while in college 40 years ago.

I am trying to find novel solutions based purely on applications of magnets and magnetic materials and wish to avoid electric curernts - which calls for a source of power.

So I trying to work withi such boundry conditions. It is nice to note everyone has contributed well in this brain stroming.

Something similar, see eddy current,

http://www.roymech.co.uk/Useful_Tables/Drive/Clutches.html

Thank you for the link. However the eddy current coupling is very different.

These are definitely examples of magnetic coupling. I was trying to take this further to an application where the the driving and driven shafts are offset in 3 dimension - where belt drives cannot work at all easily.

I think that face-to-face disks with north poles on one disk, and south poles on the other disk, (with an air gap) would be so much more effective. Additionaly, I do not see the design you mention even working.

With your design, if you visualize the magnetic field between driving pole 1, and driven pole 2 through the ferrous bar, as being a rubber band, then as the driving rotor rotates, it will tend to stretch the rubber band. This will cause the driven rotor to want to center on the iron bar, and not move from it. (imho)

have you built any prototypes? any proof of concept at all? maybe i'm not understanding it correctly.

Chris

I agree that face-to face disks having a common axis but separated shafts would be more effective, but you would need alternating poles on both disks, with the N poles of one disk aligned with the S poles of the other, and vice-versa. That may be what you intended to say...

I agree with your second paragraph - that is why I suggested two soft-iron pole pieces on each side.

Neither of us really know what the OP is trying to accomplish, but he did indicate trying to make a magnetic equivalent of a belt and pulleys, where the two shafts are separated by a significant distance. I think it can be done, but only for relatively low torque applications, and I highly doubt if it will transfer energy efficiently.

Hi DKW,

thanks for your message.. this is how I'm seeing the disks.. and I'm pretty sure that if the driving disk turns, that the driven disk will have a coupling with it, and that coupling will be a function of distance and field density. (please excuse the orientation issues of the drawings.. but I think we are clear on face to face.)

If one magnet attracts north to south, there is no reason that many magnets won't behave exactly the same. hopefully we are saying the same thing

Chris

The problem with that setup is the lack of closed loops of magnetic field. There must be only very small air gaps in every entire loop. As you show it, the N poles of the left disk must be all on the left side of each magnet, and their fields must connect to the S poles of the magnets is the right hand disk. This can only happen through the metal in the central portion of the disks, and through the very large air gap outside the disks. neither of these paths exert any torque when one disk is rotated with respect to the other. Even the poles facing each other will exert very little torque, since the next pole around the disk is of the same polarity.

If the poles on each disk alternate NSNSNS..., then the two disks will automatically align themselves with one disk rotated one pole spacing, so all magnets are attracting N-S. When one disk is rotated there will be an attraction in one direction and a repulsion in the other, so the disks will follow each other. Think about your rubber bands! Here the only air gap is the spacing between the two disks (one at each pole), so the attractions can be quite large, and there will be considerably larger torque possible.

There must be an appropriate number of poles; it must be an even number (for every magnet to have two poles).

1. You were responding to Mr Christ in which he had shown 2 discs with only South poles on one side and only North poles on one side Electrostatic charges alone can remain only positive or only negative, but not poles of a magnet. Any way let us overlook that small oversight.
2. So now you suggested having number of N & S poles alternatively and coupling them together - which you also will work - as you only confirm it. This is for a specific case of two shafts being aligned- with minor variations/ deviations.Next push the logic a little further -
3. why have so many N & S have only one large N & large S on each disc. This can be done by having a larger bulkier N & S so that pole strengths are same and effect is same.
4. Now go little further and separate the two shafts by say 20 cms and introduce a soft iron coupler as suggested earlier - so that total of two air-gaps is the same as single air-gap before. Effect and drive power will remain same !!
5. Push argument still further and shift the two shafts which are 20 cms away and don't even have them aligned. They can be in any plain at any angle. Will this reduce power output or coupling?
6. Now we still have 1:! coupling. What is the harm if we have N1 number of pole sets on one wheel and N2 number of pole sets on another wheel. This now brings me to the boundary conditions I have specified in my first poser !!!
7. Innovation is in trying to meet the boundary conditions specified
   

I am surprised you guys responding from US at this hour - it is 2 PM and it must be nearly 3 or 4 AM now there !!

"why have so many N & S have only one large N & large S on each disc. This can be done by having a larger bulkier N & S so that pole strengths are same and effect is same."

Several reasons:

1. In a face-to-face arrangement, a 2-pole arrangement would work (at very low torque), but in a side-by side arrangement, the second wheel would not 'know' which direction to rotate.

2. Torque is force times distance. The force is mostly exerted at the poles. you want those poles as close as possible to the outside rim of each disc. Your side-by-side arrangement does have a slight advantage in this respect, as the poles would presumably be on the rim.

3. That inverse-square law! When your first wheel rotates, it is effectively switching poles on those soft iron pole pieces (the first wheel is a generator, and the second wheel is a motor). When the poles switch, they must now attract the next adjacent pole. The farther away that pole is, the weaker will be the attraction, and twice as far away means 4 times weaker attraction, etc.

"Now go little further and separate the two shafts by say 20 cms and introduce a soft iron coupler as suggested earlier - so that total of two air-gaps is the same as single air-gap before. Effect and drive power will remain same !"

This means the air gaps in your arrangement must be half the separation for each of the two gaps. This in turn means you are going to have to manufacture the unit with higher precision, which costs more. Motors are always made with the smallest gaps consistent with the fabrication methods used; higher precision always costs more.

"Push argument still further and shift the two shafts which are 20 cms away and don't even have them aligned. They can be in any plain at any angle. Will this reduce power output or coupling?"

Once again, yes, the shafts could be made perpendicular to any 2 planes you wish, but fabrication costs would be high for anything other than parallel or perpendicular shafts.

No, it was 9:44 PM when I posted that last set of comments. Now this time it is strange; I couldn't sleep, so I got the computer - I commonly work in bed on my laptop, as I am now at 2:30 AM.

• Thank you so much for that very elaborate explanation.
• When I meant reduce number of poles to say 2, the pole faces also will have to be increased. I suppose an implementing engineer will take into consideration various issues you have mentioned - regarding increasing number of poles, placing them closer to the rim to increase torque.
• Now you brought in issue of higher air-gap. I was also looking at application of magnetic fluids in tweeter speakers That has been on my mind - to fill space with magnetic fluids - without enforcing too tight tolerances on machining. Any views on this?
• Further using this arrangement to get N1 / N2 ratio between wheels - will there be any problem? I do not think so.
  

You're welcome!

Again, with only two poles, there are no tangential forces, so no torque. If the motor is in its stable position (N stator to S rotor and S stator to N Rotor), and you swap poles, it will now be producing repulsive forces instead of attractive (N stator to N rotor and S stator to S rotor), but those forces pass directly through the center of the rotor. There is nothing to cause rotation. Rotor inertia might keep it going after a manual start, but it would rotate equally well in either direction. And I reiterate that inverse square law - it is so far to the next pole that the forces will be very weak. Within reason, the more poles you have, the greater torque you will have.

I am not aware of the use of magnetic fluids in tweeters. Generally, fluids are viscous and slow things down - not what you want in a tweeter. I did Google it, and found nothing on the first page of hits.

After thinking about it a little more, I believe there may be some possible N1:N2 ratios that would work for speed change, but I think both rotors must have N & S poles 180° apart, plus intermediate pairs of poles, so the possible ratios are limited.

On some previous post you asked whether this system would act like a smooth or a cogged belt. Definitely cogged!

Pl see US Patent 4957 644. Subjet is Magnetically controllable couplings containing ferrofluids.I hope you have registered with Freepatents and get full patent doc.

http://www.roymech.co.uk/Useful_Tables/Drive/Clutches.html

You can also see some experiements in You Tube on magnetic fluids or ferro fluids. I ma really looking forward some very interestng applications for this. Speaker tweeters are supposed to be using magfluids for better heat dissipation and better coupling.