Need help: Perm. Magnets and temp. changes

[BretMan]

Hello,

I'm looking into using permanent magnets for the suspension of a small mass in a device. It will be working in an ambient temperature of between +20F to +110F. I need the magnetic force to be stable. How will variations in temp. within that range effect the magnetic force of these magnets? Is there a better choice of magnet for this application?

Any guidance on this is much appreciated! Thank you.

Bret

Miami,FL USA

 

[MagnetDave]

What sort of magnet are you looking at? I imagine in this application you are not consider alnico or ferrite. (Alnico is not good at suspension applications and ferrite usually doesn't have the oomph)

For both NdFeB and SmCo this is a pretty mild temperature range. The NdFeB will show some degradation as it gets hotter. (depending on the grade - it's reversible temperature coefficient is about -.13%/C) SmCo, over this temperature range, will show virtually no change (Tc is about -.03%/C)

 

[BretMan]

Thanks MagnetDave! Really interesting and helpful.

More thoughts: I'm going to need to experiment but I don't want to be trying our things that won't work out in the end. What about machining? Are there any special drill bits or cutters best for defining the dimensions? Will it be a problem to make a magnet shaped like a tube or ring? These will be pretty small - 5mm max each (assuming I'll need 2 opposing ones for suspension purposes but only one needs to be tube shaped the other can be disc shaped.

I'm going to have to figure out what size and shape will be optimal and then order them to those specs. is there a spec for magnetic force and is there an instrument that reads the force to check it?

I have to figure out which magnet type will be best. The device is expected to also be used in high humidity situations. Also, this is cost sensitive so it needs to be low cost. What do you think?

Bret

Miami,FL USA

 

[MagnetDave]

Did you want a pony with that?

As far as machining: Magnets have to be ground. Period. They are intermetallics and machine like ceramics, only more brittle. The techniques for doing it are pretty specialized, and there only about four companies in the US that do it, and they usually have pretty sizeable minimums. (one of them is my employer, so I am being deliberately vague)

If you just want rings, that's a fairly common shape, and you can probably find that if you poke around on one of the websites that sells Chinese magnets or eBay.

Magnet grades are usually specified by energy product, as measured in MGOe. For instance N44 will have an energy product of 44 MGOe. MGOe is actually a mashup of the two characteristic that really define a magnet - remanance (which is the amount of magnetic flux it produces) and coercivity (the ability of the magnet to resist demagnetizing). Think of remanance like a car's top speed and coercivity like its torque.

Magnets are normally characterized either by a Hall probe (that measures the magnetic flux density at one point) or by a Helmholtz coil (that integrates the total flux). Again, this is pretty specialized stuff, but Hall probes are pretty cheap.

For the humidity I would either suggest samarium cobalt, or insist on a NdFeB magnet being nickel plated. NdFeB rusts pretty fast (the Nd is electrochemically similar to magnesium) so you need to protect it. Generally NdFeB magnets are significantly cheaper than SmCo (and have higher energy products, but lower coercivity), but both change based on market prices of the raw materials. For instance, right now the prices are closer because the Dysprosium that is added to NdFeB to improve the coercivity is extremely expensive. Two years ago it was cobalt driving SmCo through the roof.

Hope you enjoyed Magnets 101

 

[BretMan]

I guess I'm an appreciative person - with a bathroom too small for a pony.

The 101 info. has clarified and given me enough to begin sourcing what I need and at least ask the right questions. I'll be back then when I'm newly confused.

One last question regarding the China/Ebay magnets. From what I gathered and you pretty much confirmed, I can get what I need with the specs I need and I now must check the costs. Since the device these magnets are for will be acting to replace a metal spring for suspension purposes, I need the magnets to be at least as true to their original ratings specs as the metal spring. I'm assuming now they will even be better at it within the min./max. temps. mentioned. The device doesn't need to be ultra-precise, probably 2-4% divergence from specified specs will be acceptable. Do you think the Ebay magnets are reliably true to at least this 2-4% acceptable tolerance or should I just focus on higher end suppliers for this?

Bret

Miami,FL USA

 

[MagnetDave]

At the risk of tooting my own horn, to achieve that level of reliability you really need someone like me. This is the value we provide - we can give you EXACTLY what you want with tight quality controls, and take this very specialized area of engineering off your list of things you need to learn about. Even if we source the magnets from China, (which we will have better access to, yadda yadda yadda) we go through the product and guarantee what you're getting.

I try not to use this forum to promote the business - but hey, you asked, I answered.

 

[BretMan]

I don't see on this forum an option to send private messages and I would like follow up on your suggestion. Although, we do have a ways to go before anything solid materializes on the proprietary device we're developing. If there's a way to contact you directly without violating the forum rules, please let me know.

If not I'll just continue to post my questions as needed.

Bret

Miami,FL USA

 

[EdStainless]

Usually for instrument applications people choose to make the design a bit more involved but save other troubles.
If you build an assembly using a simple magnet shape and steel pole pieces you can get the field shape and strength without having to work with machining magnets.
You can also tune the actual field strength by reducing the field to a selected level. This is often done with shunts to "short out" some of the flux.
Dave does this for a living, talk to him.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[MagnetDave]

Bretman,

I am back from vacation, so let's see what we can do. At the moment I am still at our factory in Switzerland, so you will have to work with me on the time zone issues until I get back to the East Coast.

This link has a lot of information on the various types of Samarium Cobalt out there:

http://www.arnoldmagnetics.com/Content1.aspx?id=4210

That should help with your modeling.

When you get farther along, you can always work out nailing down a vendor. In the meantime, there is never anything wrong with having second and third opinions on what you're working on, so I would always encourage you to keep on posting here.

-Dave