Author Topic: Magnet Fixation (Read 4618 times)

fritzblitz · « **on:** December 11, 2009, 01:46:43 PM »

A real good two component adhesive will hold the magnets in place quite securely. To make sure it really gets a sit, I roughened the surface of the disc on the lathe (I was lucky someone let me use his).

The second device I use is a roll-pin (split pin, spring pin).

First I use it as a means of fixation and secondly it serves me to define the position of the magnets.

First I make a circular pitch of the disk with dividers to minimize mistakes by inaccuracy, after that I work with a gauge to space the magnets, as you can see on the pictures. I drill three holes with help of the gauge, then I shift it, one of the holes is my reference point for the next three holes etc., and I always compare my bearings with the circular pitch, a part of a millimeter deviance might lead to a lot of millimetres in the end! I put some pieces of wood on the disk to show how I put the magnets.

In the end. I put in the roll-pins and glue the magnets into position.

And give them a finish with a two-component laquer.

gotwind2 · « **Reply #1 on:** December 11, 2009, 02:32:17 PM »

I like the split pin retainer idea fritzblitz.

Well greased or even heavily lacquered to avoid corrossion maybe.

The non 'potted' or resined rotor would also act as a cooling 'fan' for the stator.

I've witnessed this method succesfully before.

Ben.

Dave B · « **Reply #2 on:** December 11, 2009, 05:43:19 PM »

Way to go, I hope others will take your advice also to use a quality adhesive to adhere the magnets to the rotor. You pins give added security, great idea.

A solid chunk of what ever to completely cast the magnets will crack when the rotors flex and the temperatures change to extremes. We've seen it too many times, water gets in and under the magnets then poof, failure and then the "I wonder what can be done or why did this happen" questions follow.

Here is a good answer, true that casting and bands might look nice and is marketable but, if you don't have the stuff to do that right you will wish you had kept it simple. Prepare the surface properly and Glue the magnets on with a quality adhesive meant for the job. Done, finished no worries. Happy to see you posted this. Dave B.

Ungrounded Lightning Rod · « **Reply #3 on:** December 11, 2009, 08:48:26 PM »

Nice.

I'd drive the pins in uniformly and to be about half the height of the magnets to minimize and equalize "field-shorting" while retaining enough support to keep the magnets from trying to somersault over them.

But I'm a purist. B-) Higher pins will only short a LITTLE of the field and "somersaulting" forces will be tiny compared to your adhesive (and directly fighting the magnetic attraction.

With those pins you might not really NEED adhesive - except maybe to cushion the magnets as you attach them to keep them from going "CLANK"!. B-)

Flux · « **Reply #4 on:** December 12, 2009, 12:52:07 AM »

Yes this is a good idea. I have used 2 pins on the outsides of big heavy ceramic magnets , I didn't bother with the side pins.

I have never been worried by keeping magnets on, my worry is corrosion and I have always potted them to try to avoid this, but I am coming round to Dave B's belief that large resin blocks are prone to cracking. Certainly polyester blocks of resin give no long term weather protection at all and as yet I have no evidence of whether cast epoxy blocks work in the long term.

I have not found epoxy paint satisfactory on some other things I tried it on but if enough coats are applied it might work.

My problems have been magnet corrosion, magnets don't come off and stators don't burn out so I think I will concentrate on weather protection and keep a keen watch out for which ideas work. In many areas it is dry enough for this not even to be an issue so perhaps there is not a lot of experience about.

If you live in an area where corrosion is not an issue then fixing magnets in this way is an excellent idea.

A good adhesive alone holds them but when corrosion starts the nickel coat separates from the magnet and you may end up with just the nickel coating stuck on. Those pins give you certainty and the flux loss is very small. On my machine I used brass but roll pins are simpler.

Flux

jlt · « **Reply #5 on:** December 12, 2009, 01:01:10 AM »

That should work okay but you still need a band around the out side. I use a automotive windshield urethane to hold mine in place.It's a lot ease yer than potting them and if you need to remove them you can cut out the urethane with a stainless steel knife. if you use epoxy they are stuck forever .

jlt · « **Reply #6 on:** December 12, 2009, 01:06:55 AM »

I had another look at your pictures and see that you have a pin on the outside edge to keep your mag's in place .

captain nodge · « **Reply #7 on:** December 12, 2009, 03:18:55 AM »

Small thought, you can buy or even cast your own sacrificial anodes very small ones on the outside of roter, would these help to alieviate corrosion probs?

Nice bit of engineering there fritzblitz, did you find it difficult matching the drill size for the roll pins?

nodge

bj · « **Reply #8 on:** December 12, 2009, 05:57:45 AM »

Great idea, I will use it on my next one. Thanks very much for the post.

bzrqmy · « **Reply #9 on:** December 12, 2009, 06:08:47 AM »

Nice work. I used the magnets with a single hole in the center and pressed a roll pin in that after using epoxy like you did. Then use the TSC Tractor paint as a weather seal. Time will tell. You jig is genius! I like the fact that after you get the first hole drilled, you can use it to center punch all the rest of your holes to ensure even spacing. I will be copying this for my next rotor. Don't know that your manufacturing capabilities are, but I could see a plastic version of this being a good E-bay seller.

Again, nice work. This is why I love this forum. I know the Dans' have been taking a lot of heat about the performance of the board lately, which seems to be resolved. I think a bit of acknowledgement need to go to the guys that run the forum. Godd job to the admins as well!!!

youmanskids · « **Reply #10 on:** December 12, 2009, 07:02:39 PM »

wouldnt it be eaiser to buy mags that come w center holes and then just drill and tap each one with a center mount screw ? (i know most mags dont have a center hole, but if available you would get the same mechanical fixation)... does the central screw change the mag field to any extent? just a thought from a newbie.

ghurd · « **Reply #11 on:** December 12, 2009, 07:50:58 PM »

It does weaken the field some. Not a lot. None if the screws are SS.

Last I saw, the neos with countersunk holes were pretty pricey.

I like the pins, though I'd use brass or SS.

G-

Flux · « **Reply #12 on:** December 13, 2009, 01:24:13 AM »

Magnets with holes in them are very convenient for motor conversions especially where you want to make a pole of more than one magnet, the repulsion is a problem and screws would help greatly until the glue sets. Also the radial machines have much more tendency to throw the magnets off.

I find it strange that we never see issues of magnets coming off on motor conversions but people are scared silly about it happening on axials where the force needed to hold them is tiny.

If you can get magnets at the same price with screw holes then fine but in a damp climate I would still rather use the roll pins than have another corrosion point and invariably tightening screws will damage the coating round the countersink even if the manufacturer successfully manages to coat the inside of a hole with epoxy or nickel plate.

Flux

neilho · « **Reply #13 on:** December 13, 2009, 08:28:36 AM »

Your rotor looks good. I'm curious as to the specific epoxy system used. Seems to me epoxy is a good system- heat applied to the back side of the magnet rotor can soften the epoxy enough to remove individual magnets.

Neil

angus · « **Reply #14 on:** December 13, 2009, 09:16:32 AM »

Again excellent idea but curious to know why the side pins?

I am guessing that the rotor is to spin in a clock wise direction and the pins are there to counter any side thrust both kinetic and electrical.

However what about the times when the rotor needs to be braked (in an overspeed situation)-would not this create a side thrust that will tend to move the magnets away from the side pins and would not side pins be required on the other side of the magnets to stabilize this

Regards.

Angus (The older I get the better I was)

Ungrounded Lightning Rod · « **Reply #15 on:** December 14, 2009, 05:24:09 PM »

Why does he need a band? The outer roll pin applies the centripetal force (especially if it's a minuscule tad past the magnet center line relative to the two beside-the-magnet pins.)

I'd be tempted to go with Flux' suggestion of two roll pins on the outside and one between the magnets, to provide more support. But fritzblitz' two-betwen-one-outside approach should provide more than adequate support and better alignment in the presence of small inaccuracies in pin location.

When installing the magnets I'd glue the first one down, then (once it's set) work from there around in the direction of the first magnet's no-pin side. That way the attraction of the new magnet for its previously mounted neighbor should hold the new one firmly against the side pins.

Ungrounded Lightning Rod · « **Reply #16 on:** December 14, 2009, 05:26:56 PM »

Why not just electroplate the whole rotor with zinc? B-) (It's called "anodization".)

It would be interesting to see if the mag field affects the distribution of the plating.

Ungrounded Lightning Rod · « **Reply #17 on:** December 14, 2009, 05:28:22 PM »

I'd start four separate quarter-circle runs so any errors in the jig wouldn't accumulate unduly.

Ungrounded Lightning Rod · « **Reply #18 on:** December 14, 2009, 05:34:57 PM »

Re axial vs. radial for magnet drift:

In an axial the centrifugal effect is fighting the magnet's attraction for the surface it's mounted on (which is far from canceled by its attraction for the surface on the other side of the gap.) The side force is just the drag against the field from the current in the stator coils, which is much smaller.

In a radial both the stator drag and the centrifugal effect are at right angles to the magnet's attraction for the surface. Without some stop or mounting screw the magnet is only held in place by friction and/or the glue.

Ungrounded Lightning Rod · « **Reply #19 on:** December 14, 2009, 05:38:58 PM »

As I read it they're there to position the magnet until the glue sets and they're on one side to avoid the critical tolerances you'd get if they were on both sides.

If they're on the side that pushes the magnets during generation it's a bonus. Meanwhile the glue can handle the (rare to nonexistent) case when there's a magnetic force on them trying to accelerate the rotor.

Braking puts the force in the same direction as generation (because generation IS braking) and you'd only need support on the other side if you are motoring.

scoraigwind · « **Reply #20 on:** December 15, 2009, 02:02:19 PM »

I must be missing something because I never do have magnets coming off the rotor. It's very simple to make a casting that keeps the magnets on. You can reinforce it with fibreglass or stainless steel wire or bands - they all seem to work - it's a non problem for me.

I do have a really big issue with magnet corrosion. That problem is very much alive and bothering me in 2009. I don't see the mechanical fixings approach helping with this corrosion issue at all.

For me the issue of magnets flying off got solved around 1999 or 2000 when I started to cast them in resin. The issue of corrosion of the magnets and the rotor disk is the live one.

To be honest I am about ready to go back to using Ferrite magnets I am so sick of seeing expensive Neos trashing themselves from within. For me the issue with protecting the magnet rotor is what sort of surface finish will prevent corrosion, and how to choose magnets that will not corrode from within. Preventing them flying off is furthest from my mind. It simply doesn't happen.

fritzblitz · « **Reply #21 on:** December 16, 2009, 01:10:12 PM »

Well, that is a lot of resonance to this issue, thanks for the discussion, there are really many interesting points coming up.

Well, there were practical points that made me fix the magnets this way. I hoped to save me a lot of work.

In the first place, making a magnet position jig and the whole handling seemed so laborious to me. I hate cyanid acrylate, by gluing the magnets into the "jig" it seems to be keen to glue my fingers rather than anything I intended to glue. So why not position them with roll pins? And glue with a two-compound glue?

And the thought of another mould and all that reeking polyester and itching glass fibre! Having been in touch with boats I know polyester to be hygroscopic, for this the yearly check for "blisters", indicating osmosis, all the recoating etc. So if the coating is the crucial thing, why not coat the magnets directly?

That is why we made the second windmill this way ( my diary on march 20th, 2006, PMG generator flying in Spain). Result after these years: My generator needs repairing, the casting has cracks, the magnets start to corrode. My friend's machine with pin fixed magnets built at the same time (same area) still runs undisturbed, no rust, no cracks, no dislocations.

Ungrounded Lightning Rod · « **Reply #22 on:** December 16, 2009, 03:46:31 PM »

perhaps the stresses from the (successfully holding the magnets) potting is what's CAUSING the corrosion, by grabbing and pulling on the magnets' plating until it breaks enough to let water in.

fabricator · « **Reply #23 on:** December 16, 2009, 04:10:34 PM »

I have to wonder if potting is causing most of the corrosion problems, especially if you use anything that is at all hygroscopic, the potting material will collect and hold water in close contact with the plates and magnets, if you are in an especially corrosive environment, like anywhere near salt water, the effects would only be accelerated.

Dave B · « **Reply #24 on:** December 18, 2009, 03:52:24 PM »

Again, great to hear of your success this way with the rotors, I support your process 100%.

Steel rotors flex and expand and contract with temperature. If your (embedded or cast) single chunk of material does not flex and expand and contract at the very same rate as the rotors it will crack. Perfect conditions then for trapped water and corrosion.

Properly prepare the surfaces and adhere the magnets with a quality adhesive. Paint the rotors with a qualty enamel or epoxy paint, done, finished no worries. Look them over and touch up the paint during your normal inspections.

Sorry it may not look as cool as embedded rotors with banding which is proving successful if correctly done. I am greatly in favor of KISS for the rotors, function over form.

Dave B.

ghurd · « **Reply #25 on:** December 18, 2009, 05:16:36 PM »

"In a radial both the stator drag and the centrifugal effect are at right angles to the magnet's attraction for the surface"

Yes, but at the same time?

I can't get it reconciled. And I just took an aspirin as a preventative measure.

Is a 48V machine, pumping 100A at 100RPM, more likely to throw a magnet at 49V than 48V?

(Yes, but at the same time?) Maybe you meant at the same time?

RHR, LHR, FE BH curve frequency stuff, etc.

And I just took another aspirin as an additional preventative measure.

G-

News:

Author Topic: Magnet Fixation (Read 4618 times)