Have I got a magnetic challenge for you! 5

[Greavis]

Hello,

I'm working on a private project and I have (what seems to be) a simple requirement but I've run into some troubles. First, let me explain the requirement and then I'll explain what I've done and why (I believe) it isn't working.

### The requirement ###

Given a normal magnet with a surrounding field as shown in Figure A (if it doesn't show below go here ->

http://www.egretendeavors.com/shield_q.gif)

the requirement is to apply a shield to one side of the magnet such that the field does not extend past it and is, thus, "flattened" on that side. See Figure B.

Here is the trick (and the key requirement): It is very important that the width of the shield (w) be as low as possible. Other pieces of metal will come very close to the magnet and would, normally, be very impacted by its magnetic field. But, because of the shield, the req is to have the other pieces of metal not be impacted at all (as if there wasn't a magnet right next to it). More than anything that is the biggest requirement (hence the minimum shield width).

The magnet materials and shield materials are flexible. Meaning I am able to use whatever will work to flatten the field on one side with a minimum shield width.

### What I've done so far ###

I'm currently testing with Neodymium block magnets that have these specs:
* Pull Force: 20.07 lbs
* Surface Field: 5755 Gauss
* Brmax: 12,600 Gauss
* BHmax: 38 MGOe

For details of the magnet go here ->

http://www.kjmagnetics.com/proddetail.asp?prod=B88X0

I'm also using some magnetic foil (0.01 inch thick) shielding with these specs:
* Magnetic saturation of 21400 Gauss
* Maximum permeability of 4000 (I assume Gauss)

In order to achieve the desired goal I've had to apply 20+ layers of the foil to one side of the magnet (usually around the same size of the magnet but I've also done layers extending past the magnet). The problem is that by the time I've added 20+ layers the total width of the shield is about 1 inch (because of bends in the foil). The magnetic field is weak at that point -- not because of the shielding -- but, I fear, because of the forced distance from the magnet (i.e. The field is inherently weak at that distance and it has very little to do with the shielding).

So, because of the width of the shield, my method defeats the primary requirement (i.e. of keeping the shield width low).

Even though I'm new to the magnetic world I would have thought that my shielding would have dampened the magnetic field more than it does. Any ideas of what I am doing wrong? What are the key factors to achieve this requirement? Am I using the wrong materials? For the foil (shield) I thought magnetic saturation was key but should it have a higher permeability instead? Are there certain key shapes of the foil that need to be used (e.g. "The foil should extend past the magnet at a distance twice it's length", etc...)?

I'm relatively new to the magnet world so any help is greatly appreciated.

Thank you!
- Greavis

 

[electricpete]

Two ideas.. not sure if they're allowed
1 - Make your shield taller extending above an dbelow the magent (is it allowed?)
2 - Better yet - make your shield wrap around the top and the bottom of the magnet like a backwards C.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

The units of permeability are not gauss.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[IRstuff]

I'm not sure what the problem really is.

Every single modern harddrive comes with a two-pole head actuator that has opposing poles mounted on a flat piece of steel. There's almost zero field leakage to the other side of the steel plate, which is only about 0.1" thick.

Using a mumetal shield would probably work even better. Most mumetal shields I've seen are quite thin.

TTFN

FAQ731-376

 

[Greavis]

So, electricpete, in short it seems the shape of the shield is the issue. Yes, wrapping it in a "c" shape is allowed; as long as it still behaves like a normal magnet on the opposite side of the shield.

IRStuff, perhaps this would apply as well since harddrives are completely enclosed within their shield (at least I would assume so). Also, if what you say is true, harddrive shielding is 10x thicker than what I'm using now. Perhaps I need thicker material or, in theory, only 10 layers of the foil I'm using. The foil has the obvious advantage of being able to shape/cut it.

Thank you, both, for your replies.

 

[prex]

My first point is that, as you probably already know, there is no magnetically insulating material, just like there are electrically insulating materials. So what you represent in Fig.B (the flux pattern) is simply impossible: if you put a non ferromagnetic material as your shield the flux lines will stay unchanged, and there is no material that will repel the flux lines like you represent them.
The only thing you can do, and you are indeed doing, is to use a ferromagnetic material as your shield. However, contrary to what you expect, this will attract the flux lines and will cause a major part of them to pass through the shield (the fraction of captured flux depending of course on the thickness, permeability and saturation of the shield).
The consequence of this is that your magnet will lose most of its power, so now the question becomes: what is this magnet for and how strong you need it to be?
To say the same in other words: fully shielding a magnet is easy, just enclose it in a ferromagnetic box, but of course this makes it useless!

prex

http://www.xcalcs.com

: Online tools for structural design

http://www.megamag.it

: Magnetic brakes for fun rides

http://www.levitans.com

: Air bearing pads

 

[IRstuff]

Harddrive chassis' are either forged or cast aluminum.

el is used as a cover, but more for rigidity than for magnetic shielding. The flux shorting bars are only slightly wider that the magnets themselves.


TTFN

FAQ731-376

 

[Greavis]

Prex, yes I understand your points clearly and, yes, the diagram should be more clear as well; the lines should be going through the shield, not repelled by it. My intent was to apply a shield to attract the magnetic field away from it's normal path and pull it in close to the magnet. The strength of the magnet or the materials used (for either the magnet or shield) are flexible as long as the magnetic field on one side of the magnet is pulled in close (as possible) while the other side is "normal".

My understanding was redirecting the magnetic field wouldn't cause the magnet to loose it's strength -- I'll have to look into that further -- but, even so, some loss is OK as long as the above requirement is satisfied.

Thanks for your post.

 

[prex]

To be clearer, not only the lines on one side of the magnet will be slurped, but almost all of them. And, as I recalled above, you'll be free from the magnetic field on the outer surface of the shield, but the stregth of the magnet will be heavily affected. You should assess this point before going on with the geometry set up.

prex

http://www.xcalcs.com

: Online tools for structural design

http://www.megamag.it

: Magnetic brakes for fun rides

http://www.levitans.com

: Air bearing pads

 

[sreid]

Look and see if a Halbach Array might help you with your problem.

http://en.wikipedia.org/wiki/Halbach_array

 

[Greavis]

Prex, your points (again) are well taken -- some weakening is OK as long as one side is pulled in very close. Also, I updated the image as well.

So, now that we are fully clear about the situation, do you have any ideas about why my previous solution didn't work? Do you believe it to be just the shape of the shield or am I, in general, doing it correctly but the material needs to change (do I need a higher permeability or saturation or both?)?

These are the areas where I lack the experience. To be honest I haven't had a chance to try the backwards "C" configuration yet (as mentioned by electricpete) -- and I plan to -- but my intuition tells me that if 20+ layers on one side doesn't do it then I have a hard time see how fewer layers wrapped around three sides will (but what do I know?!).

In short, I'm just looking for some more practical tips on how to achieve the desired result without spending tons of money on wasted foil patterns that don't work (or foil materials that won't do the job) if a little up-front knowledge can point me in the right direction.

Thanks to everyone for their contributions.

 

[IRstuff]

You need a thicker shield. You have to allow room for the field lines to bend.

TTFN

FAQ731-376

 

[Greavis]

That's what I'm talkin' about, IRstuff, thank you!

 

[sreid]

Try Magnetic Shield Corp. (mu-metal)

http://www.magnetic-shield.com/index.html

They have material that is useful in high magnetic fields (such as you have). Their evauation kit would be usesful if you have some way to measure the effectiveness of the shielding.

 

[prex]

As long as the section of the shield is a few sq mm, it won't saturate and the field on its surface will be substantially zero (though I don't know what is zero for you).
To wrap the shield on three sides of the magnet (but why not all around?) will help, but won't change too much if you are interested at a zero field only in the middle of shield's outer face.
I have however to insist on my main point: you need to define what you have to do with the magnet. What you get with the magnet + shield arrangement is something similar to a single fairly long magnet with distant N & S poles (anf this is normally not what you seek in a magnet arrangement), except that the flux lines will extend much shorter beyond the pole faces as they are bent by the shield. Is really this what you need?

prex

http://www.xcalcs.com

: Online tools for structural design

http://www.megamag.it

: Magnetic brakes for fun rides

http://www.levitans.com

: Air bearing pads

 

[Greavis]

Hello Prex,

"Zero" for me would be if I put a piece of metal right on (or next to) the shield it won't be effected by the magnet at all. Whether that is really zero or some small value I don't know but that is the desired behavior.

Why not wrap it all the way around? Well here is the theory. The shield needs to "hide" one whole side of the magnet -- not at just one point -- but, at the same time, still be a functional magnet (even if that means at a loss of strength). If I only wrap the shield around like a backwards "C" then the field lines would still be strong on the opposite side of the shield and, thus, so would the magent as a whole (or at least stronger I should say). That was the idea as I understood it. Again, I'm new to this so I could be missing some obvious things (which is why I'm posting ).

If wrapping the whole magnet in a shield is what I have to do to block one side (and the magnet still has some life left) then that's what I'll do.

So, in essence, the first priority is to block one, whole side of the magnet.

The second priority is to maximize the magnet's strength.

Naturally, if blocking one side "kills" the magnet altogether then that defeats the purpose; if I have to simply reduce the strength on one side to keep the magnet alive then that will have to be the solution.

Does that help answer your questions?

 

[Greavis]

sreid, yes I saw them as well. I was hoping to find out something like "you need high permeability" or "high saturation foil with those magnets" so I could focus in on the foil and the design rather than spending $200 on a 6 pieces of foil where only one (hopefully) will suit me.

But I appreciate the link, though, thanks.

 

[sreid]

High permiability is what is usually required. Like for shielding magnetic read heads from low level 60 cycle fields. You need a shield which is going to "Short Circuit" a high intensity field so you'll need the high saturation material.

You might start by simply trying cold rolled steel. Cheap and has a high saturation.

 

[IRstuff]

You still haven't explained what function you're trying to achieve. Please avoid trying to describe what your notional solution is, as this may not be the optimum solution.

From what you've discussed so far, I see no difference between what you have and the moving coil head actuators used on a hard drive. You have two magnets, side-by-side, flat, about the size of a 2032 battery. One has N up, the other S up. They are both mounted on a flat substrate that provides both structural and shielding functions. I suggest that you spend $40 or so and buy and hard drive and some Torx drivers and have at it. Near as I can tell, the actuator magnet structure duplicates your drawing, except that the exposed face away from the shield has N and S poles perpendicular to the open face.

TTFN

FAQ731-376

 

[Greavis]

IRstuff, as much as I'd like to discuss it all in detail I legally can't; not without NDA's, etc... which is why I've tried to describe the req's as in much details as possible.

I'll look into the hard drive example as well (I never knew they were like that)... thanks again.