Array of PMs on pole piece to= large PM?

[Todd619]

Is it possible to attach several permanent magnets to one side of piece of iron or other magnetically conductive material to get a highly uniform mag. field on the other side?

 

[prex]

It depends on what you mean by highly uniform.
The only way I know of obtaining a fairly uniform (=vector B constant in direction and value over a finite space) is the classical use of magnets: creating a gap in a horseshoe (=loop) configuration.

prex

http://www.xcalcs.com

Online tools for structural design

 

[Todd619]

Thanks for the input, let me explain a little more about what I have in mind. What I want to do is build a low power table top cyclotron as a hobby project using rare earth magnets instead of electromagnets. The largest disc shaped rare earth magnets I'm finding are about 4" in diameter which is not that large, and would be difficult and scary to work with. So I wanted to see if by placing several smaller ones on the back of an iron disc if I could produce a uniform field on the other side or will I end up with an inconsisted array of magnetic wells on the other side of the iron?

-Todd619

 

[Skogsgurra]

Putting them on the back of an iron disc is not a good idea since the field will be short-circuited by the disc itself. You probably need to use about the same magnetic configuration that ordinary cyclotrons use and put your RE magnets where the magnet coils and their cores usually are situated. You are right, large RE magnets are scary. I have hurt myself several times when I forgot about how powerful thay are. Take care.

 

[hacksaw]

are you using a xtal controlled exciter?

 

[sreid]

Large PM magnets are often built up using smaller magnets. I believe that an iron pole piece will smooth out the flux from an array of magnets on the back side. Getting a uniform magnetic field might best be accomplished by shaping the pole face. This would best be done using magnetic FEA.

 

[prex]

Todd619, you should tell us first what kind of dimensions are you speaking about (I mean the dimensions of the space where a uniform field is sought).
RE magnets as large as 100x100x25 (and possibly more) can be fabricated, and they can be used in arrays of different types. No problem if you put them in series (one on top of the other), a little more difficult to put them side by side (with the same field orientation), as they repel themselves (though not with a very high force).
You can create a fairly uniform field of some 0.5 T in a gap of 100x100x15 mm by using two magnets as above (or more smaller ones) and an iron to close the circuit (and to resist the attraction of many kN): is this close to what you have in mind?

prex

http://www.xcalcs.com

Online tools for structural design

 

[Todd619]

Thanks for the input everyone. Just from experimenting today with some ceramic magnets and a piece of steel it looks like I'm going to need to go with (2) large magnets instead of an array. I haven't figured it out yet but I would like to get hydrogen ions up to about 30 kev and I'm estimating that I would need RE magnets of at least 5" dia.
To Hacksaw; I was hoping to just use a tunable RF source and amplify it to the required power with off the shelf budget minded equipment, I'm not sure what an xtal controlled exciter is, but I'll look it up when I get a chance.

 

[EdStainless]

Most large PM assemblies are made from smaller magnets bonded together. If the magnets are rectangular, and similar is properties, you can't tell that it isn't a single magnet from the field.

One catch. The best adhesives are epoxies, and they require heated cure cycles. You will get some demagnetization at cure temp. How much depends on what PM material you are using and how hot you work.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[hacksaw]

the usual arrangement is a xtal controlled rf source with tunable magnet, suspect that swept freq approach is a bit more difficult in any event good luck, sounds like fun

 

[Todd619]

(to EdStainless)bonding the magnets sounds very doable, could I bond them to a backing matierial, instead of to eachother to really make sure there is no space between them? I would like to bond them to a backing anyway. Also does the shape (you said rectangular) make a difference or is it just a matter of not having a gap between the magnets? As far as the temp of the curing of the epoxy, maybe the Sm-Co PMs would be the best choice instead of the NdFeB ones, because they perform well at higher temperatures. I was under the impression that the loss in field strength due to heat above 180 degress with the NdFeB PMs was temporary, but if there is a permanent change in the magnet, it's something to consider.
These RE magnets if they are not coated seem to have a surface that would lend to very good bonding, anyone know what kind of tensile strength I could expect on the bond?

 

[EdStainless]

Now you are making me think hard, I have been out of the business since '89. Yes, we were making NeFe then.

Square or rect. is better so that you don't have gaps. They are a real bear to put together since they are magnetized. You need good clamps and fixtures because the repulsive forces are huge.
NeFe is nice because of the high B, you can get a lot of field. The demag that you get on slight heating isn't a permanent strucural change. But you would need to remagnetize in order to recover it. Mag'ing large assemblies is very tricky.
For high temp products (Alnico, and both 1:5 and 2:17 SmCo) we used ScotchWeld 2214. This is a 'B' staged epoxy with a fairly high cure temp (350F) and great strength. The epoxy and the bonds were stronger than the magnets. For working with NeFe we used a two part epoxy so that we could use low cure temp.
I would glue the magnets to a plate, and then to the pole pieces. The multiple bondlines don't help your magnetic field, but in these materials the impact is small.

How large of a magnet are you looking to build? I can look and see what I have around.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[Todd619]

Thanks for the info on the exact adhesives. Many of the magnet suppliers have segments that look like a ring that has been divided into pieces and look like they are meant to be fitted around a disc to create a larger disc assembly with no gaps. I would ultimately like to have a large disc shape (at least 5'-6" dia) so maybe that's the route I'll go, but I see no reason why a square assembly wouln't work also, so I appreciate the offer. I will need to research techniques of assembly if I'm going to do this myself, I have experience with epoxies in the composite industry, but keeping those magnets from snapping together will be a trick, can anyone recommend a source of info on how these assemblies are done or maybe better than that a company that could do it for me?
Also I'm thinking I'd like to bond them to a steel backing plate, Sreid said that that might smooth out the flux further and it would have the added advantages of being very strong and workable with a frame work to hold the two assemblies apart at a small distance and would also shield the field from everything on the other side. can a good bond be formed to steel with epoxy? I've seen transformer cores bonded in this way.