Coupled Toroidal Chokes

[Syd Rumpo]

I want to put some modulation onto a DC power line.

I used a suitably rated COTS toroidal DC choke and was going to hand
wind another winding to introduce the modulation. But that was tedious
given the number of turns I'd need for a reasonable ratio.

Instead, I took another similar toroid and wound some wire through both
cores, several turns and connected the two ends. Viola, a large fiddle.

It works, well enough, but how to analyse or simulate this? How many
turns? It seems to stop mattering after a few, but I'm not sure how to
do the maths on this. What are the pitfalls?

Cheers
--
Syd

[Tim Wescott]

So, to save yourself having to wind wire through one toroid, you wound
wire through two, and that's good. Uh, OK.

You have created two transformers, connected back to back. The more
turns you wound, the higher the inductance of the coupling windings.
Presumably you're working in voltage mode (sorry I'm not stating this
well -- I haven't had breakfast yet) which is why more turns on the
coupling windings are better up to a point, then the effect tapers off
dramatically.

Figure out the inductance/turn on the coils, count turns in your coupling
coils, and do the analysis accordingly.

Of course, those few turns on _one_ toroid, being driven directly at low
impedance, will do.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

[Syd Rumpo]

Thanks, that's clarified things a bit. It's much easier to wind 25
turns through two toroids than about 200 through one! Also the
isolation is better having more layers of insulation to get through.

Driving at a low impedance doesn't fit with the rest of the circuit,
unless of course I used a transformer...

Cheers
--
Syd

[josephkk]

On Fri, 28 Sep 2012 12:21:55 +0100, Syd Rumpo <use...@nononono.co.uk>
wrote:

It almost sounds like you want a telephone hybrid. Bi-directional voice
(and faster for some applications) in the presence of DC and ringing
voltages.

HTH

?-)