Building a Uniform Electromagnet

[tare3]

I am looking to build a electromagnet. I wanted to build a Helmholtz coil because it has a relatively constant magnetic field. But unfortunately, I can't build one large enough at a reasonable cost to get the desired strength.

I am looking now to build a C frame magnet because I can get a strength of at least 0.1 T. I just don't know how uniform their magnetic field strengths are. I need a 3 cm by 3 cm by 3 cm approximately constant magnetic field. I read some other post but they didn't really deal with the idea of maximizing uniformity over a desired area. Anything relating to the shape or size of the core would be great. Also if you have any pointer, suggestions, calculations or anything that could possibly help.

I've done my research on a Helmholtz coil so i have a idea of what i need and can do but i can't find much about a C frame magnet.

 

[IRstuff]

How "approximately constant?"

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[tare3]

As close to 100% for the entire 3x3x3cm block i can get. I realize that its not going to be perfect but the reason i tried the helmholtz first is because it was known for being almost 99% for the length of the radius of the coil. I just need a new design that is as constant and uniform as possible. I'm think i am going to have to make a soft iron core to get a .3 T strength that i would want, but i don't know how to make it almost constant throughout the field. I saw that you can focus the field by lofting the core to a smaller point but that doesn't really tell me about how uniform it is.

 

[IRstuff]

see:

http://physicsx.pr.erau.edu/HelmholtzCoils/index.html

and

http://www.phy.cuhk.edu.hk/sure/comments_2009/ltfrep.pdf

According to those pages, the field of a Helmholtz coil is only uniform over about 20% of the diameter. Since one would not expect a magnet to be that different, we can expect that you would need a diameter of 15 cm to even come close to being sufficiently uniform.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[tare3]

I had viewed the first sight you linked to before and misinterpreted the scale the first time. You are correct that it is 20% uniform over its entire radius(distance between coils) but its almost completely uniform throughout that 20% which was what I was looking for. I've actually seen a website that said it was most uniform under 10% of the radius.

But that still leave me with a problem because i cannot make a strong enough magnetic field based on the specifications. I simulated a couple different gauges of wire with different currents and different radius as well as done some calculations with other professors. With certain limit, I've found that the strongest ideal field i can get is .028 Tesla. It would be difficult because i either need to make it big to fit more larger gauge wire for higher current and/or have a obscene amount of turns. This is no where near the intensity i need. So that is why I'm looking for a new design. I was recommend that the high permeability of soft iron would make it easier to achieve my goal, which is why i'm curious about the C frame design.

 

[electricpete]

With C-shaped soft iron core (excited by turns of wire wrapped around) and a gap between the faces of the C, you can get a fairly uniform field in the gap between the faces.

Let's say for simplicity the C has a cross section of dxd and the gap distance is g. You get a more uniform field when d>g (the higher the ratio d/g, the more uniform). If your area of interest is 3cm*3cm*3cm then of course you need g>=3. How much bigger? If the thing you're studying doens't affect the field, then I'd say make g=3 (to maximize d/g). But if the thing you're studying affects the field slightly (such as mu differs from mu0, or current induced during motion), then you may want to make g something larger than 3

=====================================
(2B)+(2B)' ?

 

[RyreInc]

A Helmholtz coil has two rings at +/- 26.565 degrees (atan(r/2r)). If you extend these rings +/- 26.565 degrees you have a spherical shape with holes at each end, similar to a bowed out solenoid. The field produced is more uniform than a standard Helmholtz and can easily be made much stronger since there is much more room for wire.

Check out this presentation I previously made for more details.

 

[tare3]

Are you saying to bow out each individual coil of the Helmholtz solenoid or bow them both so they make a circle if you know what i mean.

coil1 coil2 co__il1 co__il2
/ \ / \ / \
/ \
-----> vs ---------->
\ /
\ / \__/ \__/
coil1 coil2 coil1 coil2
The illustration kind of the frame that the coil would be wrapped around. I wasn't sure which one you meant. I think you meant the first. (i hope that illustration makes since) The arrow is the magnetic field. I realize they should be rounded better. Also you said ~40% longer, you can make the coil without affecting the field. I read somewhere that the bisected area of the coil should not be more that 1/10 of the radius in any dimension. I think that is a little extreme, but I want to make sure if this idea is going to work. I was thinking that as long as the distance from the center of each ring was equal to the average radius i would be fine(hopeful). So i could make it as wide as will fit?

 

[RyreInc]

I would draw it more like this:

coil1 / ^ \ coil 1
/ | \
\ | /
coil2 \ | / coil 2

field points up

The center of each coil needs to be 26.565 degrees from horizontal (i.e. two coils separated by one radius as in a standard Helmholtz). Whether each coil spans 26.565 +/- 1 degrees (typical Helmholtz) or 26.565 +/- 26.565 degrees doesn't really matter.

But really, it's all one coil--just like a solenoid, but with curved walls.

A solenoid with straight walls would need to be about 40% longer to achieve the same magnitude uniformity in a given area.

I believe you could make the coil as thick as you want, as long as these other requirements are met. This is equivalent to nesting coils inside one another.