Laminating Mu Metal Core

A related good question: why do you think you want to do this yourself? Doing it with a stripwound (torus) or cut core will certainly give better performance!

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

There's a guy who never needed mu metal to really be mu metal. ;)

Just dropping it on the floor reduces its mu, let alone winding it round a mandrel or die cutting it.

It takes an 1100 C hydrogen anneal to fix it.

Cheers

Phil Hobbs

In the context of a transformer (presumably for low frequencies), a stripwound core can certainly compete with mu-metal (stripwound allows amorphous alloys with some stellar properties).

A mu-metal toroid core is possible, by punching donut shaped sections and stacking; It might be possible to etch the shapes instead, and skip re-annealing, Stacked toroid cores are rarely done, because it wastes material.

Russian dolls

NT

** Winding a small toriodal core is not near as easy as a bobbin core type.

** Sounds weird - bet you have never seen one.

** It is *rarely* done co it does not work.

The strip material used in toroidal transformer cores is "grain oriented" along the direction of the strip. When wound, this aids greatly in reducing magnetising losses and increases primary inductance to the same end.

Annular sections cannot be grain oriented in the same way.

.... Phil

** Should be better worded:

Applying fine gauge windings onto a small, toriodal core is not near as easy as applying the same onto a plastic bobbin for use with regular shape cores.

.... Phil

Blah blah blah, you're just repeating the properties of mu metal that everyone else blindly quoted -- and not at all talking about /why mu metal might need to be mu metal/! C'mon, Phil! ;-)

Premade mu metal cores are available (with better properties than you'll get from a hand-made custom assembly, at a teeny fraction of the price!).

Premade cores are available, but whether they have enough winding and cross sectional areas, and a small enough outline to satisfy the OP, we'll apparently never know.. he never mentioned *why* he thinks he wants this, just that he's set on doing it...

Nanocrystalline materials also get very high (20-100k mu_r).

And if you really need mu metal to be mu metal, you'll probably consider supermalloy (mu_r up to 1e6!) instead of regular permalloy.

There are very few applications where you need extreme mu, and not also a lot of flux. 10k mu_r makes quite excellent common mode chokes, 5k mu_r makes excellent pulse transformers, and 1k mu_r makes excellent power transformers. (To go even further: 100 mu_r is far more than needed to make fine inductors in molded shapes like rods and barrels.)

The most common application, where you don't need much flux, and you do need extreme mu_r, is also a custom formed shape: magnetic shields. After making a one of those, hydrogen annealing is mandatory -- at least, if you want to use a minimum amount of material. Even here, though, there is a practical limit to mu_r: to be able to withstand the flux due to Earth's magnetic field, without saturation, you need a certain minimum amount of material. And that cross section will get you quite a lot of attenuation, so you don't need to have outrageous permeability to get there. The remaining applications, that do require extreme shielding, are confined to very few instrumentation and physical uses, with budgets big enough to handle the process.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

limit to mu_r: to be

Perpendicular B is continuous at a material interface, and a high-mu material bends the external B so that it's almost perpendicular to the surface. In other words, H drops, B doesn't jump.

Super permalloy is used in very thin films to make the pole pieces in magnetic recording heads, and they don't saturate in Earth's field.

The advantage of mu metal over super permalloy is basically its ductility--it's fairly soft stuff, so you can cut it and bend it easily, though you have to anneal it again afterwards.

Cheers

Phil Hobbs

At the surface, yeah, but all those field lines collect together and run down the sides, then spread out again and exit.

Draw a square, with one side parallel to ambient \vec{B}. The flux entering the "top" gets concentrated down the lengthwise sides. |B| ~= 0 inside the square, and B in the sides is B_amb * top area / side cross section.

Right? :)

How thin and how wide?

In usual ambient conditions, you get a ~10,000 advantage on area, which is pretty good.

Good point. More expensive, too.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

ntering

square, and B in the sides is B_amb * top area / side cross section.

The saturation B of mu metal or permalloy is around 10 kilogauss (1 tesla), as usual. The Earth's field is about half a gauss. Takes a lot of geometry to make a dent in that ratio.

Cheers

Phil Hobbs