3
\$\begingroup\$

I have an application where I am producing fairly high short duration pulses into a load, 50A and 100us. I am looking for a common mode inductor to help reduce my EMC emissions.

Because these pulses are short and infrequent then from an rms heating point of view this inductor can probably be rated at 5A or less.

Being common mode then the fluxes generated by the outgoing and returning currents will tend to cancel which is the point of a common mode inductor. There must be a point however, where due to slight differences in the coupling between the two windings and the core the choke can saturate and lose inductance.

Does anybody have any experience of this and to what extent you can push the pulsed current beyond the continuous rms rating without loss of inductance.

Are certain types of design better than others in this regard, I would expect toroidal cores to be better than most but I have no evidence to support this. Finally is there any published data available?

\$\endgroup\$
5
  • \$\begingroup\$ If you can't find this kind of detail in the datasheet of a balun or common mode choke, talk to the manufacturer. \$\endgroup\$ – Olin Lathrop Jun 7 '16 at 15:23
  • \$\begingroup\$ @OlinLathrop This seems not to be a common requirement, I have spoken to one manufacturer who can find no data for this but is arranging some testing to find out. \$\endgroup\$ – Warren Hill Jun 7 '16 at 15:28
  • \$\begingroup\$ An 'inductor' won't stop a 50A 100 uS current from radiating if your wiring is a good antenna. It's possible to radiate RF from differential current, not just from common-mode current (and the inductor will have no effect on the outbound and return currents). \$\endgroup\$ – Whit3rd Jun 8 '16 at 7:47
  • \$\begingroup\$ @Whit3rd My problem is with radiated emissions and my reason for thinking a common mode choke may help is I have wrapped three turns of my lead to the load around a large ferrite at the equipment end and this reduces my emissions. \$\endgroup\$ – Warren Hill Jun 8 '16 at 8:46
  • 1
    \$\begingroup\$ It may be that a clamp filter (two C sections that is assembled around the cable) will be as effective as multiple turns around a ferrite rod. It's worth a try (solved my problem once, at an emissions test) with the smallest available item that will enclose the cable. Remember to try this on ALL the cables, including AC input and control cables; the one with the scary power levels isn't the only or even the main suspect. \$\endgroup\$ – Whit3rd Jun 9 '16 at 3:52
2
\$\begingroup\$

If you are winding your own, you can use bifilar wire which when not twisted on a toroid is pretty nearly perfect. Otherwise you can expect to see the effect of a fraction of a turn; if you use one turn at your maximum current to specify your toroid you will be safe. Wind in bifilar fashion in any case for the best results. If you wind one winding over the other the results will not be as good.

You can measure the results directly with a third (temporary) coil of fine wire, short out the output and drive the input with some AC.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks, I'm hoping to use an off the shelf part for this but a custom design is not totally out of the question, though it would probably be more expensive. I've up voted your answer for now but will wait to see if I get any other replies before deciding which to accept. \$\endgroup\$ – Warren Hill Jun 7 '16 at 18:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.