Common Mode Transformer Saturation

"D from BC"

** Will he ever learn to read and not make wrong assumptions ??

** There is no spec for "saturation" on that page.

The amp ratings are based on temp rise alone.

..... Phil

ooops. I didn't check the fine print was at the bottom of the graph.. Yup ..The current spec is based on a heat rise. And I think that's a rise from the DCR from DC.

And you're right, that sheet has no saturation data.. The inductance spec is open circuit. There is no inductance listed at Irms max. Ex.. At 7Arms passing through the transformer the inductance is ...unknown.. But at least I know how warm it'll get if the ambient temp is 85C.

Can I assume the CM transformer gets too hot first (from Idc^2*DCR) before saturation? I think the saturation data is not listed cause it's not relevant. In common mode there's opposing magnetization of the core and I think that takes more current compared to differential mode. It's the heat from wire resistance and core temp that set the limits.. Not really the core saturation. So I am guessing that the inductance at bias (all the way up to Irmsmax) will be not too far from the open circuit inductance spec.

Am I far off ?? :(

"D from BC" "Phil Allison"

** It will never saturate if used as intended.

** Just a bit.

** If the core gets very hot - like 150C, inductance will fall off a bit.

Curie point temp and all that.

example:

Hello,

For the differential case (so equal current, but opposite direction), the net field in the core material is close to zero. Therefore current is limited by ohmic losses in the two windings.

The common mode current (that is the sum of I(1-2) + I(4-3) ) causes a net magnetic field inside the core. When you look the intended application, they mention "EMC filters". In most cases common mode levels generated by equipment are far below the saturation current of the CM coil. So therefore they do not specify the CM current rating.

Depending on you application, you may run into problems. When the CM coil is subjected to CM pulses as used in surge immunity testing, the V*t product may result in high common mode current. Saturation may also occur when one of the windings carry (part of) DC supply current. I'm currently working on a power application where this effect will happen when no measures are taken.

You can measure the CM inductance versus CM current behavior yourself when you have two same pieces. Put them in series; supply current through the series circuit (decouple your power supply for the small signal AC frequency). Both coils now carry equal DC current. You can use one of the 2 windings (and leave open the unused), or parallel the two windings.

At the connection of the 2 same-type coils you can inject an AC signal via a coupling capacitor (and series resistor). This AC current distributes equally through the inductors. Gradually increase the current and look to the AC voltage at the interconnection. Above a certain DC current, the AC signal at the interconnection will drop because of core saturation.

In most cases you want a high (CMinductance)/(leakage inductance) ratio. This requires closed magnetic circuits (no air gap) with short magnetic field paths. This, however, results in low CM current handling capability.

Best regards,

Wim PA3DJS

in case of PM, don't forget to remove abc.

I was kind of wondering the same thing. I was going to test it.

I have a nice 3" high-mu toroid with some 6AWG hookup wire wrapped around it, for common mode duty. I wound it in the "bunch of turns up the left side, bunch of turns up the right side in the opposite direction" style, so it will have quite a lot of leakage inductance. Thing is, if each side saturates from the ~100A peaks I'll be drawing from it, that will quite excellently defeat the purpose of having it.

Since the windings oppose, it should indeed act like two rod cores. The question then becomes, what is the effective airgap, and its corresponding saturation ampturnsage?

I was going to test it by hooking up a little buck converter to the line end of the choke, shorting the load end. So I've got maybe 100A drawn through at piss-all voltage. I've got two ends (the buck converter looks like a voltage source), so I can play with it in common mode as any old inductor and watch the B-H curve (well, I-t curve) while varying amps. I'll need to set up a big row of IRFZ46N's though, there's maybe 200A peak I'd want to look at. Don't have any way to measure that much DC, either... maybe I should settle for a smaller scale test...

Of course, if it were wound bifilar, this would be a trivial question. But that doesn't have much leakage inductance, so you'd definitely need a differential mode choke for extra filtering. (I should probably have one anyway, since I'm running a rather low frequency...)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"D from BC"  wrote in message 
news:jin8h5thv83hokfs8s25pkkqm7ub93nf5t@4ax.com...
> Is the datasheet spec for common mode transformers for the common mode
> current for saturation?? Or is that the differential mode current for
> saturation?
> Of maybe it\'s the max current through one coil..
> Datasheet example:
> http://www.cooperbussmann.com/pdf/69eca15b-d981-4cf9-b53a-c878f70dbd24.pdf
>
> Is the differential mode current for saturation much higher than the
> common mode current for saturation?
> I suppose when a common mode transformer gets a differential current a
> field and antifield has to form in the core and I suppose this might
> virtually split up the core into something that probably acts like 2
> rod cores.
>

example:

Thanks :)