Securing magnets

[Murlin]

I suddenly had visions of my magnets flying off and killing something or someone....

One thought came to mind.

I wonder what would haappen If I ground a shank the mags and pinned them into a small pocket?

Would this cause feild disruptions?

Perhaps the the safty factor would outwiegh a small disruption in the fields down on the bottom side of the magnet.

Murlin teh research, research.....

[SparWeb]

Scroll down to "Jensen Genny"

or

http://www.fieldlines.com/story/2006/6/8/24353/80984

[Murlin]

Most excellent......thak you sir, I did a search but came up empty.

I hate to remove some of the nickle protection, but if I encapuslate, them in some 0f the epoxy about half way up, it should add back whatever protection I ground off...

Hopefully if you flood cooled them as you ground them and didn't shock them or get them hot, it would not have any adverse affects on the strength...

Hopefully.....

Murlin

[DanG]

The chemical composition of a neodymium magnet is Nd2Fe14B, which is two atoms of neodymium, fourteen iron atoms, and one boron atom. Clumps of this have about the same strength as theatre popcorn as compared to most other iron alloys.

Manufactuors 'sinter' Nd2Fe14B from a fine powder form to a solid via compacting in a die & fusing it with heat and pressure. Although the sintered magnets are solid, they a very similar to a ceramic and so easily broken & chipped. Sintered neo magnets require coatings like nickle or epoxy to keep oxygen in the air from rapidly corroding the composition.

Anyway, to answer your question, even if the magnet was cast in your suggested shape at the factory just natural expansion and contraction from heating cycles would most likely cause fractures around the 'keyed' area - the magnet would still fly off but leave behind the dovetail portion....

-

[Murlin]

Ahh well that why we ask questions....

I had considered thermal expansion... The pockets would have enough clearance and the gap would be taken up with a coushion made up of the epoxy so it would not have any stresses built up as it contracts or expands.  The undercut would just keep the magnets from actually falling out if they loose the magnatism and or the epoxy around them loosens up.

So from what you are saying, even recessing them in a square pocket, like the above post, would cause the same effect if there wasnt enough clearance to account for thermal expansion.

Hrmmmm....

Perhaps a compression shim between the shank and the side of the steel pocket....

Murlin

[Flux]

Yes DanG is right, If you are worried drill 2 holes outside each magnet position and fit 2 roll pins for the magnet to sit against when it is thrown out. In the perfect world they would be stainless but you will not notice the difference.

The less damage you do to the plating the better even if you cover with epoxy. Corrosion can be a serious problem unless you are in a dry environment.

Flux

[Countryboy]

If you read the stories about the Dans brakedrum alternators, you will find that they machined a groove in the face of the brake drum for the magnets to sit in.  The lip of the groove helps hold the magnets in place.

Or, do as Flux suggested and use roll pins for the magnet to butt up against.

There has also been some discussion about machining a metal ring that fits over the outside diameter of the magnets to secure them.

[Murlin]

Ok here is my take on this subject:

I have concerns where if a problem occurs, and the machine starts to over heat, thermal expansion will cause the machine to tear itself apart. And with the curent approach, all we can do is to just sit back and watch our investment disinigrate before our eyes.

Well that is not acceptible to me. Especially when I have the foresight to perhaps stop it from happening.

I know for a fact that steel will expand at .0014 per inch

@ 400 degrees F.

This means IF(and if being the optium word) our machines have a problem, the diffrence between the thermal expansion of the steel and that of the bonding epoxy that holds our magnets in place is enough to cause a metldown of the whole thing.

On the smaller machines it may not be enough to cause much problem but the bigger the diameter of the rotors get, the equasion will reach 100% failure, sooner or later.  Steel rotors @ 20" in dia will expand + .032"@ 400 degreesa F , and the epoxy will not move much at all, and seperation of the two materials WILL AND HAS occured. It has been documented here in the pages of this Fourm on even the 10' machines.

So I try to expect this, and come up with a solution to combat this failure, so we can just replace the defective parts and maintain structural integrity of the entire unit.

We know that if everything works right, there is not much of a problem.   But I know from experience that nothing EVER goes as planned and sooner or later something fails.

From what you guys say , magnets are like a huge metalic spounge.   Very porous and will soak up moisture and oxodize and breakdown, very brittle.  

This is why we plate them, to seal the magnets and protect from the elements stopping this breakdown,(oxidation) from happening.  Knowing this, we can come up with a solution for the problems that arise in working with this material.

I would change the current geometry of the magnets to have a 15 degree angle on the sides of magnets with a small flat(about 1/8") the remaining distance to the back.

Then I would machine a strait walled pocket in the rotors to occomidate the width of the magnets, allowing for a certain amount of thermal expansion to occure without causing the magnets to shatter against the sides of the pocket.

Then I would machine some phenolic keepers, with 15 degrees on each side, so they would hold down 2 sides of two magnets, and bolt them down with two flat head SS cap screws between the gap of each two magnets.

Problem solved.......

I will want to get magnets of this geometry.  

Can I have mine replated and re magnitized if I grind the in?

Or would it be cheaper to just order them to a specified geometry from the start?

I would supply prints to the specified dimensions......

Murlin

[Flux]

It is possible to machine magnets without demagnetising them, but difficult.

I doubt that you will find anyone to plate them or remagnetise them. Better pay the earth for specials from a supplier.

I really can't see why you are bothered about 400 degF. You may keep lumps of sintered metal in place at that temperature but they will not be magnets.

flux

[dinges]

I seem to remember that Murlin has (or has access to) EDM (electrical discharge machining). A sink-EDM machine could be the way to make holes in them. The bevels or whatever could probably be grinded. Not the average DIY stuff, but doable.

As far as ordering magnets with the right geometry goes; you also have to take into account the dimensional accuracy of the magnets, as they are delivered from the factory. I don't know from memory the dimensional tolerances, but they were huge, IIRC.

And then there's still that coating issue, which is important for long-term reliability. I'd think that epoxy would be good enough (besides, epoxy is one of the coatings you can get from the magnet-factory; it's just that most of us use nickel-coated magnets.)

BTW, I like the way you think about gennies. I too don't like putting in a lot of work, time & money if I'm not sure the thing will run for years without too much hassle.

Peter.

[Murlin]

I have beeb grinding the permant mag chuck on my surface grinder for years, and it has never lost any strength yet

I know that if you lost the stator and the genney started running away with itself, the mags would probably have demagnatized by then. And at some point you would need to try to shut it down.  this could require that you get fairley close.

This could happen in a matter of a few minutes.   Induction coils heat up metal glowing red hot in seconds.  If one of my kids or anyone for that matter, got killed with this thing, I would not be able to live with myself.

I think I will just settle for modifying these and then order some specials, when and if they fail.  They will be extremely easy to replace.

And as far as tolerances go, you must fit each one.....no big deal to me, been working that way for years

Murlin teh cautious.....

[ghurd]

Now might be good time to remind everyone of the dangers associated with grinding or machining neos.

Poison gases.  Toxic fires.  Explosive combustion.

G-

[dinges]

Hmmm...

Why don't they teach these things in school? Thanks for the head-up, will definitely look deeper into it before continuing.

Peter.

[Murlin]

Thank you ghurd, I will be sure and take all the proper precautions.  I usally hold my breath and have a fan blowing out the door when I grind hazardous materials.  But I do have a good Dual Cartridge Respirator Assembly if needed.

I plan on using flood coolant and I will certainly stand behind a shield now that you have said that.

Correction about the permenant mag chuck.  The mags on that are on the inside and the the lead and steel plates are what gets gound in.

I have ground carbide and ceramic inserts with my diomand wheels.

But if the magnets cannot be ground, I will find out real quick and have to think about another solution....

One other possible solution would be to seperate the eopxy into 4 sections by machining grove through the resin.

This way it couldnt be sheered up by thermal expansion. as easily

Might be the easiest and best solution.

Murlin

[Murlin]

I did find this info out there and so the only thing I would have to worry about is sealing them back up to protect them from oxidation.....perhaps painting them with an epoxy paint would seal them up pretty good and help them to last longer...

Machining and tolerances

Neodymium (we like to call it "Neo") is easily ground,

although coolants must be used in the process to avoid

spontaneous combustion of powder. Grinding should be

completed before coating or plating is applied to protect

the magnets from corrosion. Standard tolerances for Neo

magnets are +/-.005" for ground dimensions.

[oztules]

Useful information Murlin for those trying to squeeze the last tolerances out of their motor conversions.(make the magnet ring oversize and grind it down to size).

Also in the Zubbly style, fully sealed motors should have a propensity to not corrode the magnets if the coating is slightly questionable.

..........oztules