Active EMI filter

[jashugan]

Hello,
Did you ever use Active EMI filter for reducing noise instead of the Classic passive filter? If you did, can you share your experience?

[Benta]

I think you need to be more specific.

[jashugan]

I mean, to reduce EMI there are two different ways, passive and active filters. I always used passive filters, both for easiness, and because of the low power projects I developed until now. But for the latest project, I face the need to reduce a lot EMI, and I don't have the space to reduce. Of course I could work with transformer and the rest of the switching power supply, but I was thinking with an active EMI filter, like this one:

or this

I have no experience in use of such filters, so I wonder if any of you had used these active filters before. Did they work? Were they effective, in term of space and cost?

[jwet]

They really work to the extent that you can get through your EMI test regime especially in automotive apps. TI made a big splash with these at CPEC and IEEE Conferences a few years back.  They had lots of data showing before and after kind of results for CISPR 25 tests- you can dig this stuff up.  What I think they really allow you to do is to make decent EMI filters without so much copper and iron- ie smaller and maybe cheaper.

CISPR 25/5 goes down to 150 KHz and at low frequencies, its really hard to get rid of conducted EMI without big reactances.  The bain of automotive engineering these days is AM radios- 500 - 2000 Khz.  The other one is hefty USB power delivery to the console.

I sat through some seminars by TI and Tyco and the idea is that they're really effective at lower frequencies and at higher frequencies, passives take over.  Depending on where your problem is, it might be worthwhile. 

Corcom and other passive filter companies made some blocks like these for general use in the 1990's (industrial/non automotive) but they stopped.  I worked for Maxim and we looked at them as a new product area but passed.  The corocm parts came and went.   This usually means that there were better ways of getting there.  Today, in automotive, with the size constraints and gnarly problems, they may find a home.  The only thing that I would be nervous about would be to design them in, TI and the market overcome the problems other ways and they disappear leaving a hole in your product.

Good luck.

[nimish]

They work. And yeah you can reduce emi filter size and weight considerably. It's usually not worth it unless weight and size are at a premium and you have also done stuff like upping switching frequency, interleaving and the like. This research presentation has the state of the art. Not cheap but as it points out after moving to higher frequencies and WBG it's the next bottleneck in increasing density.

https://www.psma.com/sites/default/files/uploads/files/Active_EMI_Filters_for_High_Power_Density_Converters_Balaji_Narayanasamy_Tesla.pdf

[Smokey]

This controller has the active EMI filter built in:
https://www.ti.com/lit/ds/symlink/lm25149.pdf

[jashugan]

Wow, really impressive!

[jashugan]

Thanks for you support, I will try to do some experiments, and I will report them.