I guess I'll have to wind my own inductor, again

Greetings:

What are the intended uses for these things:

formatting link

I tried hooking a pair of 27uH (they are also rated at 16uH at 9.8A), with a pair of 8.2uF caps as a differential LCR filter at 100kHz, to a load of 2.55ohms as the output filter stage for an Apex SA-60 PWM amplifer.

Idling with zero volts out the coils get warm. But when I put 7Vrms at

4kHz into the 2.5ohms, they get almost smokin' and whine as well as the output being substantially attenuated. Simulation indicates only about 5A peak through the things with these numbers, and the output amplitude shouldn't experience much attenuation even up to 10kHz.

If these things can't even tolerate 4kHz at half their rated current, what are they good for? Why do companies like JW Miller market products that are not specified completely? Perhaps the only thing they are really good for is EMI chokes.

I guess I'll be making a gapped core inductor on a good HF core like the

3C90 I bought a while back. Hope ETD29 is big enough so I don't have to order more...

Good day!

--
_______________________________________________________________________
Christopher R. Carlen
 Click to see the full signature
Reply to
Chris Carlen
Loading thread data ...

Sounds like powdered iron junk. If you want to stick with a torroid, use the Magnetics, Inc "Kool-Mu" material. It's almost as good as permalloy powder, but a lot cheaper. I think Micrometals has an equivalent, now. I was burning paint off powdered-iron cores, and that fixed it.

John

Reply to
John Larkin

Greetings Mr. Carlen. What are they good for? They are good for making DC-DC converters as is indicated on the link you provided. The table showing the current rating of your device is specifically a DC current rating, and is usually close to the saturation limit of the core material. If these are used in a switch mode powersupply where they handle mostly DC, but with a small AC ripple current riding on top of the DC current, then they shouldn't get so hot even when handling rated current. In a typical switch mode powersupply operating at full load in continuous conduction mode the RMS ripple current might be say 20% or something of the DC current flowing through the inductor.

I don't understand quite what you are doing, but it sounds like you are using them in an AC application. Indeed powdered iron is not a good choice for this type of application. Gapped ferrite is much better suited when making inductors designed to handle substantial AC current at anything but the lowest of frequencies.

Reply to
Fritz Schlunder

How large is the 100kHz component of current? According to the spec, the wire diameter is .053 (probably 16 guage), but the skin depth at 100kHz is .01 inches.

Reply to
The Phantom

Careful. Ferrites saturate rapidly. You might need an air gap.

Check out Epcos's Ferrite magnetic design tool ( downloadable from epcos.com ).

It does all the calculations for you. You want the DC bias tab btw.

Graham

Reply to
Pooh Bear

Not that much, about 2.1A p-p to 2.6A p-p riding on the lower frequency ripple current, at the conditions described. A little more on average when idling, 2.6A p-p. If anything, this would warrant slightly hotter cores when idling.

JW Magnetics has confirmed they are for EMI.

Thanks for the input.

Good day!

--
_______________________________________________________________________
Christopher R. Carlen
 Click to see the full signature
Reply to
Chris Carlen

Yes, JW Miller says they are EMI. It says "DC-DC converter, EMI filter applications" but then says "low core loss".

Oh well.

Thanks for the tip about Kool-Mu. I might like to use toroid, to avoid alot of fringing fields from my spacer-gapped cores wandering about my chassis, which will also have some more sensitive analog stuff.

But I might be able to test things with my ETD29-3C90 cores anyways...

Thanks for the input.

Good day!

--
_______________________________________________________________________
Christopher R. Carlen
 Click to see the full signature
Reply to
Chris Carlen

Yes, it was a test, just to see if I'd get lucky.

My slowly growing experience with this sort of thing is indicating when hands-on inductor design is called for.

This is such a case.

Good day!

--
_______________________________________________________________________
Christopher R. Carlen
 Click to see the full signature
Reply to
Chris Carlen

Thanks for the link.

--
_______________________________________________________________________
Christopher R. Carlen
 Click to see the full signature
Reply to
Chris Carlen

What I've learned from your story here is that one shouldn't expect to use a given inductor in an AC application unless the manufacturer gives you a hysteresis loss curve!

Reply to
Joel Kolstad

Where is "epcos.com" anyway?

I vaguely recall attempting that URL a long time ago and also came up with nothing.

Oh, here's a link, from Eastern Components:

formatting link

Good day!

--
_______________________________________________________________________
Christopher R. Carlen
 Click to see the full signature
Reply to
Chris Carlen

Micrometals

formatting link
have a nice little free downloadable application that works out those losses and then presents you with various options you can wind on their cores. You can trade DC R - power losses etc.

Graham

Reply to
Pooh Bear

Hi Chris,

John Larkin hit the nail on the head, as usual.

in a dc-dc app, most of the current is DC, with a small amount of ripple

- 10%-20% usually. So iron powder is ok. Hell, old AC drives I have worked on used DC bus chokes made from powder coated 1mm steel plate (bugger all AC flux so who cares about losses). I have set numerous Si-steel and powdered iron chokes on fire by uising them as ac output filters in drives :)

What colour is the core? from that you can work out the material (most likely its a micrometals, perhaps yellow/white?) and from their look at the loss curves. Their are some fairly good powdered iron cores, but if you dont need quite as much Bsat, Kool-mu winds hands down every time.

Cheers Terry

Reply to
Terry Given

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.