Magnetic Levitation Bearings

It won't work. You can build a magnetic thrust bearing, but not using a PM-PM magnet configuration. The reason is Earnshaw's theorem which, in effect, states unequivocally that there is no possible stable static levitating configuration of permanent magnets. In practical terms, this means your system will move until it hits a physical constraint (the edge of a cone for example.)

As I said earlier, you can build a magnetic thrust bearing. To make it work, one side of it needs to be controlled - in effect modulating the repulsion force to keep the rotor at a fixed distance from the stator. It's a tricky control problem in a large dynamic system subject to stochastic (random) loads - in your case, the wind. A good mechanical bearing would provide very low friction. Magnetic bearings aren't going to add much to overall system efficiency. In this case, simpler is better.

I don't think I understand maybe you can help me with this, and I can't argue that it is simpler and cheaper to use standard bearings.

But I guess I need some more education. I have seen how a spinning top can levitate on two magnets as long as the top is spinning. But if you have multiple magnets configured all the way around a cone with there being two cones and magnets on each cone. Such that the weight pushes the top cone down inside the bottom cone. With that configuration shouldn't you always have a magnet pushing against a magnet. Now I can see how you have to keep the top cone from popping out but I was figuring on having two such bearings, one on top of the wings and one on the bottom. Shouldn't there always be a magnet pushing against a magnet?

I have seen something like this once before. It was on Mythbusters and they were testing anti-gravity machines. They had something just like this with the two cones. And they were very surprised that the inside or top cone was levitating inside the bottom cone. Of course they proved that it wasn't anti-gravity but they had still made a crude magnetic bearing at least it appeared they had.

I know that what I'm talking about wouldn't be cheap since it would requre a lot of magnets but it still seems like it would work.

The spinning top system you refer to is called a Levitron. It works because the top is spinning (so it's not a static system.) However, the downfall from a practical point of view is that the potential energy "well" that the top sits in is tiny. The total system must be precisely level, the weight of the top must be exactly right to within a fraction of a gram, and the spin speed must be within a narrow range. Any disturbance will bump the top out of it's potential energy well and the thing smacks down. I know, I own one. They're a lot of fun to play with and they fascinate your friends.

My point with all this is that the Levitron is far too finnicky to be a practical system, and its finnickyness (not really a word?) is due to the fundamental physics involved. There are some papers out there about how they work but the math involved gets really hairy.

Dullin, Holger R; Easton Robert W. "Stability of Levitrons" (PDF)

Simon, Martin D; Heflinger, Lee O. "Spin Stabilized Magnetic Levitation," American Journal of Physics, April 1997

"You can build a magnetic thrust bearing, but not using a PM-PM magnet configuration."

Have you never seen one PM ring magnet floating over another on a post? If that isn't a thrust bearing, I don't know what is. It does require the post to keep the floating magnet from falling off, but so do virtually all other kinds of thrust bearings. It should be obvious that there is a limit to the load that can be supported, depending on the strength of the magnets.

Of course I have. I love the things. However, the system is unstable. It will always be up against a mechanical stop. In practical terms, this means you'll never be able to keep it from wobbling, even if you constrain it.

Another issue you'll run into is that even if you build a constrained PM-PM configuration, you have no control over the stiffness of the bearing.

Given the practical problems involved, and the small amount of efficiency gain you'll get over a mechanical system, the mechanical system is definitely the more practical solution.

My 2 cents...

There are ways to use permanent magnets, actively controlled electromagnets, and eddy current repulsion, to produce low friction bearings. HOWEVER, there is usually a very strong reason for adding this kind of cost and complexity to a particular system. For some high speed systems (200,000+ RPM) low friction magnetic bearings could perform better and outlast mechanical bearings. For your application, I don't see any significant benefit to using magnetic bearings. The added cost and complexity will probably hurt the main project. That would be unfortunate.

I highly recommend you proceed with your current project using good quality mechanical bearings. You should also continue researching magnetic bearings (and maybe experimenting with them as a SIDE project) to fully understand their pros and cons. Best wishes on your wind project.

The machine shop I worked in used auto and truck front spindles for winch construction. A spindle like that would do for your use, too. The great thing was availability of replacement bearings at the local auto store.

Since you are making the alternator too, I just want to kick out this offhand idea. The maglev trains lift the train, and also keep it from rubbing in the turns. I think they use electromagnets at right angles to each other to control the train. They also push it along.

Now here's the weird part. What if you maglev the VAWT up to running position, and when it gets turning good it somehow generates juice right back thru the same electromagnets for a net gain of output? It would be like regenerative braking.

Don't slam me, guys, I'm just a hopeless Laithwaite fan.

Hi,

good idea to play with magnetic bearings,

but first finish your turbine plus generator with conventional bearings (ball-bearings may be more complicated than you estimate!)

So the above suggestion to use an existing bearing and shaft assembly is the best idea to have a fast success.

If playing with magnetic bearings:

- forget about total magnetic suspension (5 degrees of freedom),

- think about either to support the axial load by a magnetic assembly and allow for radial friction-type bearings or about a lower ball bearing to support axial load and radial load and an upper magnetic bearing to give radial load capacity and stiffness.

The latter one has proven fantastic success in the Pfeiffer-Vacuum company turbo-molecular pumps.

The axial magnetic bearing:

a. primitive: two flat magnets opposed to each other

b.: advanced: a center arrangemant like a. and an outer assembly of two flat rings also opposed to each other but reversed direction, plus two steel plates to back the two pieces of each.

The radial bearing primitive:(with minor axial thrust if not oriented perfectly): two rings radially magnetised, one on axis one in housing.

Radial advanced: 4 magnet rings axially at a distance of 1 ring-thickness, two at stator, two at rotor, steel sleeves to support magnets and to guide magnetic flux.

Ball bearing to be commented later.

RHABE

Check out this link could be helpfull.

http://www.techbriefs.com/component/content/article/5424

I downloaded the file, it looks good. If I can insert it, I will. The Glenn Research Center seems only interested in the levitation properties of this item. They mention induced current in the coil, but don't say how much. It sure looks like this could be both a bearing and a generator. They only show a radial bearing, but a thrust bearing should not be too difficult. The original drawing of the Halback array could be put in a circle for thrust, from what I see.

The document is an Adobe Reader 8 pdf file. If anyone knows how to insert these files here, please let me know. Everytime I want to do something, I gotta download another darn utility!

Go back one screen from the PDF

Right click & left click "copy link location"

Write your comment

I usually write something like Follow This Link

Highlight it

Left click the hyperlink icon [4th from the left on the CR4 editor toolbar] & paste in the link

That's assuming the PDF originated on the internet.

Is there a way to post actual PDF documents? (That would be wonderful, as I can convert virtually anything to a PDF)

Copy & paste? If the PDF's are your's

or a file sharing site....

But the offers tend to change, making free solutions only temporary

I've never been able to paste anything but text in CR4. That could be because I'm using a Mac... I've always converted things to JPGs, and CR4 accepts those.

Nice list of file sharing... Thanks!

Dick

I tried (and it worked) to paste a text from word saved in RTF format available on word.

Re-read my post! No problem with text; the problem is large detailed graphics. I'd love to be able to insert PDF files like I can with JPGs.

Dick

follow this link: http://www.defensetechbriefs.com/component/content/article/5424-lew-18239-1?start=1

This is a test, hope it works. I downloaded the file, It would be easier to just have the file here, other users may not have the utility or access to NASA tech briefs. I had to go thru a registering and e-mail reply hoop to get in.

Hate to be a pain, but if I learn this right, I can post stuff as I find it.

No problem always interesting to get all the bits working.....

Which that link doesn't for me

http://www.defensetechbriefs.com/component/

content/article/5424-lew-18239-1?start=1

There you go. When I clicked on the 'link' in your reply, it took me right to the spot. I will try to insert a pdf file from my documents, they have made good progress with the Halback levitation device, got a model turbofan going. I e-mailed Dennis J Eichenberg with the LDF project at Glenn Research, got a response. They have axial and radial levitation bearings. They have rotary power to the fan, efficiently. Now we just have to see if the unit can be used for wind power, he says yes.

Google tm-2006-214481 for the turbofan. It has drawings, math, graphs, etc.

I think listing and googling may be the simplest solution for my linking problem.

Like I told DK a few posts back, free file sharing is available online:

http://www.google.com/search?hl=en&q=free+file+sharing&sourceid=navclient-ff&rlz=1B3GGGL_enUS338US338&ie=UTF-8&aq=t&oq=free+file

& you would be able to post links directed to whichever site.

Garthh, like you said in reply #18, a link doesn't always work. If you are not registered at that site, you can't get in. Thanks for the advice, I'm not going to use PtoP just yet. I will see how Google referrals work. When I had Limewire, it wanted to raid my computer for everything in it!

Is there a way to use the google resource without making everyone who want's to see the file a collaborator?

Wow, I really don't understand your question, sorry. When I check something on Google, I see more than one source of the information in question. Something can start on NASA Tech Briefs, but end up available from other sources, even you-tube! So if I list a document, I will Google it first to make sure it is available from more than one source.

As I do this reply, the 'insert hyperlink' icon is greyed, not available. 'Insert /edit image' is available, but won't insert the pdf from my documents.

So I highlighted a word for spell-check and the 'insert hyperlink' icon becomes available.... I think I am getting closer to the solution. God, I'm stupid.

Yes if you highlight a word you can turn it into a hyperlink

I have a google account, using google docs

I can save a 10mb PDF

like this parts manual for a chainsaw I'm working on

You can check your links, after you click preview & before you submit

I'm answering my own question here...

I just never bothered to figure out what I could do with it...

There are limits on the size of individual documents, I can't quite get a handle on the total storage available.....

I have a couple thousand pics on Picasa web album, another google app.

Here.