Tape wound toroid for 40kHz

The Magnetics 2016 28-page Tape-Wound-Cores catalog has loss curves to 100kHz. Yessir, they look pretty lossy. But if you're only making a CT, etc., maybe it'd be OK.

A Rogowski coil might work here; it has no magnetic core. 5" ID is big for a magnetic-core toroid.

Ferrite toroids come that large, but you didn't mention how closely the fit needs to be...

VAC or Metglas/HMG have, or can make, cores that big, with reasonable losses. They aren't cheap ($200?).

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

Thank you.

Thanks. That price wouldn't be a problem, if it works.

I don't yet know the exact dimensions, that's outside my control but it would be immutable.

Just musing, instead of a circular toroid, what about, say, eight or more shortish cylindrical inductors wound on ferrite rod and joined in series to make an octagonal (or more) toroid approximation? Part of me thinks of that a gapped core, but I'm not sure.

Cheers

The main idea of a toroid core is to provide a contiguous path for the magnetic flux around the conductor carrying the current. The octagon would work if you can join the corners without any air gaps. Remember that an air gap represents an addition to the magnetic resistance mu times the core material.

The flux from the big conductor wouldn't thread the cores much--too much reluctance from the big gaps. You could use four rods in a square, with some nice thin glue layers.

Cheers

Phil Hobbs

ed in

of me

Grind down 1/2 way on the ends and stack 'em like Lincoln logs.

George H.

If a current transformer winding plus burden resistance were zero, there'd be negligiible losses because the core magnetization would be infinitesimal.

So use two current transformer secondary windings, one for feedback current and the second for sensing. An op amp senses through one winding, drives feedback current through the second winding,

Like a Rogowski coil, it won't be great at lower frequencies; there's DC offset issues, but manageable ones.

People do make true DCCTs (I did one myself, once) but they are a big deal. Next best is a Hall-gap sensor with a feedback winding.

You can buy Rowgowskis

and it wouldn't be terrible to make a high-frequency version.

or just use a negative resistance with a single winding. either way GBW is a thing to be avoided.

yeah, high frequencies could be a challenge too.

Silicon iron i think becomes more lossy above 1KC..

I guess you have never seen POWER toroids..

Yes, but carefully designed audio power transformers using thin silcon steel laminations work well up to tens of kHz. So the OP's idea may be doable.

Metallic glass ribbon can go even higher, but it isn't cheap.

Yes, slide on one winding bobbin, and assemble a pair of Ferroxcube U cores. If your center conductor is too big for that, make one up with series I sections.

What are you trying to do?

If it's a CT, use a Rogowski coil. If not, tell us about your frequencies and current sizes, etc.

I get 4GHz of bandwidth with an arrangement like that, so I think 40KHz should be possible. You're making a current transformer, right? 40kHz is a bit low for ferrite, but a bit high for silicon iron. Tape-wound nano-crystalline toroids should work.

Keep impedances low.

Jeroen Belleman

I've seen them and used them. And I repeat that a wound toroid with a

You can buy a Rogowski that looks like a piece of rope that you wrap around a giant conductor.

Metglas is amazing. u up to 1e6, treated right.

But a 5" ID wound core will be a big deal.

I wonder how much primary current is expected.