Maker Pro
Maker Pro

Skin effect a problem for inductor?

eem2am

Aug 3, 2009
429
Joined
Aug 3, 2009
Messages
429
Hello

Please could you tell if an off-the-shelf inductor will be ok for our requirement?

Eg will this inductor be ok….
2209-RC (47uH from Bourns.com)

http://www.bourns.com/data/global/pdfs/2200_series.pdf

It is the output inductor on a half-bridge smps.

It will see a Max current of following

Average current = 4 Amps.
RMS current = 4.4 Amps.
Peak current = 6.7 Amps

Ripple frequency = 111KHz.

The di is quite high (5.1Amps)
i.e. 5.1 Amps at the switching frequency (111KHz)

We need a high ripple in the inductor to improve our transient response.

Will an off-the shelf inductor with a single winding be ok, or will we need to custom build, with multiple strand conductors to mitigate skin effect?
 
Last edited by a moderator:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I think the most important detail is that the inductance at the rated current is only about 30uH.

A larger core, and biflar windings or stranded wire would probably improve things, but it all comes down to cost/benefit.

At 100khz, the skin depth is a little over 0.2mm in copper wire. So the ultimate would be to use 0.4mm stranded wire.

If the smps will be in continuous mode, there will be a fairly large DC component to the current through the inductor and skin effect will be less of an issue.
 

D_Hambley

Sep 1, 2011
11
Joined
Sep 1, 2011
Messages
11
Skin effect will not be a problem at 111kHz. Core loss may be high. I took a wild guess and assumed it was something similar to a MPP core. Working backwards from the Bourns data, I get something similar to a Mag-Inc 55206-A2 core with 26T, 21AWG. The core loss curves on the Mag-Inc website (http://www.mag-inc.com/products/powder-cores) shows about 2.4W core loss. The skin effect will make the effective resistance about 15% or 20% more (for the AC component of your current) so youv'e got only about 0.6W of copper loss.
Note - My core guess could be way off but, in general, with that amount of AC, core loss could be dominant.
 

eem2am

Aug 3, 2009
429
Joined
Aug 3, 2009
Messages
429
Thanks,

Though i'm mystified as to why people use multiple strand windings now, as you appear to confirm that skin effect is not that much of a problem.

For example, i worked at a television company, and they had three a strand secondary on their 16V(out), 33W, 70KHz, DCM 230VAC flyback.

So why were they bothering to use three strands i wonder...if skin effect isnt that bad?
 
Top