EMC book with emphasis on electromagnetics/physics fundamentals

[TimNJ]

Hello everyone,

Since graduating college four years ago, I've been working in AC-DC/DC-DC switch-mode power supply design. After my boss left the company two years ago, I was fortunate(?) enough to take over a big portion of the design role, which has been very interesting. But, this leaves me with very little guidance, expertise to look up to.

I feel like about 50% of my job is thinking about EMC. Like many people, I have the Henry Ott book, which has served as a great introduction and general reference, but has some limitations in depth of the material covered. I think of it on the same plane as The Art of Electronics: an excellent reference for practicing engineers, but perhaps not the best for fundamentals.

Yap, yap, yap...

Basically, I have a hard time visualizing various interactions between power supply (DUT) and the outside world, often in terms of interactions between the power supply and the distribution network in which it is connected to, parasitic return paths through earth/external reference plane (and what that means for radiated EMI, ESD, EFT performance, etc.), and interactions with the system it powers (w.r.t. common-mode currents, etc.).

I think my next step is to try formally modeling these types of systems with SPICE or at least on paper. I should say: I have done lots of modeling, but beyond a certain complexity, I think my models start falling apart. For example, making a reasonably accurate model of an small switch-mode transformer in SPICE?

Does anyone have any favorites? PDFs or other IEEE/whitepaper references are welcome too! I think I ultimately need to get down to the fundamentals.

Thanks,
Tim

[Conrad Hoffman]

EMC is (or was?) big business, so lots of seminars and things you can pay for. Not as many decent dedicated books that I know of. EMC is just plain old electronics, with a different viewpoint, so fundamentals, especially RF fundamentals, are something one should always go back and review. Heck, I find it useful to review Ohm's law every now and then. Old books are often better than new books and you can't go wrong with a copy of the various Terman works on your shelf. Almost everybody who came after copied his stuff, but usually not as well. https://en.wikipedia.org/wiki/Frederick_Terman 1) Electronic and Radio Engineering 2) Measurements in Radio Engineering 3) Fundamentals of Radio (don't know how useful this is).

Almost anything with titles like Antennas & Waves or Electromagnetic Field Theory should be useful. I find used bookstores an invaluable source of low cost texts. The best ones tend to be near university towns. I don't know where you are but New England has some excellent places like Old Number 6 Book Depot. Bequaert Old Books no longer has a shop, but deals entirely on the Internet, as do many shops. That's probably the best place to find good titles. I also like AbeBooks, owned by Amazon.

[2N3055]

What Conrad said...
Maybe check book  Mark I. Montrose - Printed Circuit Board Design Techniques for EMC Compliance_ A Handbook for Designers-Wiley-IEEE Press (2000)
Wouldn't say it has much more theory, but goes in a bit more details.. Also author lists the references.... That is a big list to browse through..

[jkostb]

If you really want go to the fundamentals of EMC then you need to study Electromagnetic wave propagation (i.e. classical Maxwell theory). Spice is not suitable as simulation tool for this purpose. With spice you can model only lumped element circuits and this is not suitable for modelling EFT, ESD etc.  Modelling and simulating these type of phenomena is quite complex and you need EM field solvers. There are several books which go deeper e.g. Clayton Introduction to electromagnetic compatibility, Bogdan Foundations of electromagnetic compatibility. For understanding emission problems it helps if you have some basic knowledge about antenna theory. The classical work for antenna theory is Antenna theory of Balanis. But I have to warn these books are quite mathematical.....

[TimNJ]

Thanks for the recommendations everyone. I will take a look at all of the books so far, read through their table of contents and references.

I'd say my biggest two issues are...

1.) trying to connect the dots between the low level theory and observed performance/phenomena
2.) finding the time to set up and run complex simulations in a commercial environment, where the reality is that we just need to make something that is good enough to sell. Obviously, some more advanced applications absolutely necessitate good simulation. In my case, not so much.

I think these are struggles most engineers go through. Not unique at all. And I don't expect anyone to be able to solve these problems for me. You can't.

Regarding connecting the dots, I really like to understand things on a deep level, but often cannot make it stick until I have a contextualized example of it in the real world. So, that's why I asked for recommendations per the title of this thread, in lieu of having a true mentor who can help contextualize and build upon the fundamentals with me.

I know SPICE is limited. I'm sure part of my failings in SPICE are trying to used a lumped model for something which really isn't. Any recommendations for starting with field solvers, FEM, etc? I don't expect to be competent any time soon, but seems like a powerful tool to have..

[jonpaul]

Bonjour, TimNJ: BRAVO for your efforts and SMPS apprentissage in the Power Electronics area.

I was an SMPS, HV and electronic ballast designer and consultant for the decades 1970s..1990s.

In that epoch, no simulations, design was by hand and then bench testing.

SMPS, filters, shielding and PCB layout is tightly interrelated. Choice of frequency and topology has a huge impact.

The magnetics designs (especially material, gaps, shields and winding) is critical in the SMPS transformers and the EMI filters.

Also essential: Understanding and separating Common Mode (CM) and Normal Mode (NM) or Differential Mode (DM) noise generation and filtering

The critical first step is to pin down the customer's REAL requirements.

There are many fine reference texts on EMC, regulatory compliance design and theory, besides the Henry OTT book.

If you can precise the power and voltage levels you work on, and equipment categories, (eg Avionics, Medical, Heavy Industrial, Power T&D) We are happy to  make specific suggestions.

Finally the IEEE, PESC, and other conferences  publish papers and proceedings on these topics.

Bon Chance,

Jon

[TimNJ]

Hi Jonpaul. Thank you for sharing your experience and advice.

I work primarily on AC-DC switch-mode power supplies between 10-1000W, universal line input 100-240VAC, usually 50/60Hz, sometimes 400Hz. Mostly “normal” industrial and medical stuff with 5-50VDC output.

I think my understanding of magnetics ~design~ is good enough at this point. Of course, I have a ton of room for improvement, but feel I can tackle the magnetics design challenges. But, modeling magnetics in terms of their EMC profile, not so much, just have some rules of thumb.

I would like to understand how to create models for CM/DM noise generators, predict the effects of inserting choke and ‘position x’, trying to help solve customer EMC issues (both conducted/radiated emissions + EFT, ESD, surge) from the power supply end, predict the effectiveness of various filters, etc.

Does that help? I’m really looking for something to help bridge the gap between theory and practice, and if had an SMPS emphasis, that would be ideal.

Thanks,
Tim

[jonpaul]

Rebonjour cher Tim:  I have a lot of material, and happy to assist you.

Those power ranges are just right, I had also done medical supplies, for lasers and op room illuminators.

The UL544 (?) Medical requirements drive a large and more complex filter due to very low L-N to earth leakages.

I am not good at SPICE or modeling, used only MicroCap, decades ago.

Was easier to just take a guess at best design, wind the inductor or transformer and test out.

For simulation of transients, the pros use very costly and dangerous ESD and IEEE 61000 transient generators, you can rent or go to a compliance test lab to use, you RARELY need to build or buy one.

HyPot is easier to test just a HV supply and current limit R.

We had many HyPot testers and a Danbridge 30 kV 1 uA tester.

I built several impulse generators, just a capacitive discharge with a spark gap into a series injection ferrite slug Tesla coil.

Some of the old Furman Sound protectors were designed with that simulator.

For some SMPS fun, check the old papers of Nathan SOKAL, old friend (RIP) inventor of Class E, aldo Dr Richard REDL and Dr Slobodan CUK.

Finally Dr Lee at Virginal Poly Tech had a power electronic group.

Will find some good ref books,

We can connect via PM or email,


Bon chance,

Jon

[TimNJ]

Thanks Jon Paul. If you have any book or literature recommendations, please feel free to leave them here under this thread! Maybe others will also be interested to see.

Where I work, we have all the equipment to test IEC61000 immunity, CISPR11/32 EMI, dielectric strength, etc. I’ve run all these tests many times. Somewhere in between recognizing a problem and testing a solution is where I tend to stumble. (I suppose many people do.) “What is the coupling path?” is a question I constantly find myself asking.

And of course, as you say, people have been working with these issues before computer modeling was practical, but now it seems like a good way to help build and reinforce some underlying concepts.

Thanks!

[jonpaul]

Tim: Coupling path is a combination of conductive, capacitive, inductive and radiated RF.

The exact determination is difficult and not really needed.

 Techniques for EMI and interference issues 

a/ Reduce root source of interference with soft switching,  resonance, class E, etc.

b/ CM and DM filters on mains and outputs

c/ Electrostatic shielding

d/ Careful grounding and PCB layout practice

e/  magnetics design for compliance,  Low external field,  interwinding Faraday shield

Book references: just a few of my favorites from my library: All are VERY old, look on Amazon or ebay:

Electromagnetic Compatibility Handbook: Violette/White/Violette

Power Line Filter Design... Mark Nave

Noise Reduction Techniques Henry Ott (latest ed)

Electromagnic Compossibility / Schlicke

Magnetic Components: Design and Applications
by  an old friend/ Steve Smith 


Just the ramblings of an old retired EE

Bon chance!

Jon

 

[jonpaul]

For latest conferences and papers in power electronics I highly recommend the PELS: IEEE Power Electronics Society
https://www.ieee-pels.org/

Previously called PESC: Power Electronics Specialist Conference

Also the Industrial Electronics  http://www.ieee-ies.org


Jon

[TimNJ]

Thank you again. I will look through the books you recommend. Some of them are very high cost on Amazon, but maybe cheaper earlier editions on Abebooks or eBay.

When it comes to fundamentals, I often find the older books (1950s - 1970s) have a way of writing which is refreshingly clear (straightforward) and often somewhat entertaining. Can‘t put a finger on what’s changed in English writing since then, but prefer the old style.

[nctnico]

Thanks for the recommendations everyone. I will take a look at all of the books so far, read through their table of contents and references.

I'd say my biggest two issues are...

1.) trying to connect the dots between the low level theory and observed performance/phenomena
2.) finding the time to set up and run complex simulations in a commercial environment, where the reality is that we just need to make something that is good enough to sell. Obviously, some more advanced applications absolutely necessitate good simulation. In my case, not so much.

What has served me well when dealing with EMC problems is to always remember that current flows in a circle. From there think which elements make that circle and subsequently how to break the circle.

[jonpaul]

Tim: over the ages, the frequency's go up and occasionally new topologies emerge, but theory and principles do not change.

Perhaps  could read them at a  good engineering school library.

Ebay and ham radio flea markets are the best sources,   Amazon sometimes  have used ones.

Abe Books is very costly

The books are well worth the cost if they help your skills and projects.

Finally the website of Dr Howard Johnson has a lot of good material although he is focused on high speed digital design.

http://www.sigcon.com/

Kind Regards,

Jon

[TimNJ]

Thanks for the recommendations everyone. I will take a look at all of the books so far, read through their table of contents and references.

I'd say my biggest two issues are...

1.) trying to connect the dots between the low level theory and observed performance/phenomena
2.) finding the time to set up and run complex simulations in a commercial environment, where the reality is that we just need to make something that is good enough to sell. Obviously, some more advanced applications absolutely necessitate good simulation. In my case, not so much.

What has served me well when dealing with EMC problems is to always remember that current flows in a circle. From there think which elements make that circle and subsequently how to break the circle.

This has helped me too. Though sometimes I wind up making it worse when I try to "break the circle", which some people might come back and say "go back to the fundamentals!", which of course is true, but, uhh....not everyone has 100% perfect understanding of fundamentals. 

Often I find myself second guessing myself as I'm explaining the idea of common-mode currents to co-workers, often in the vicinity of the EMI test bench. And then I point to the ceramic tile floor and wood-table and a bunch of otherwise "insulators" and say "yeah the current flows through there!". Then I get confused looks. Even for things I know are ~true~, i.e. things I've read 100x in literature,  it can be hard to convince myself that it can really be that way!

Not to make this a philosophy class.

[jonpaul]

Tim re CM coupling, the process of conversion of CM noise to DM noise is also very confusing.

Easy to demo CM on a diff amp or transformer coupled input, try it!


Jon

[temperance]

A book which served me well: Electromagnetic compatibility M.J. Coenen, J.J. Goedbloed

ISBN: 978-90-8572-034-8

You might also find this interesting: A Circuit Modeling Technique for the ISO 7637-3 Capacitive Coupling Clamp Test
https://biblio.ugent.be/publication/8548379

[cdev]

Suddenly there are a bunch of semi decent cheap EMC antennas available on ebay. I have bought a couple of them

Particularly the log-periodics work well for locating signals. Check them out!

You can use them for HDTV too, of course.

[TimNJ]

Suddenly there are a bunch of semi decent cheap EMC antennas available on ebay. I have bought a couple of them

Particularly the log-periodics work well for locating signals. Check them out!

You can use them for HDTV too, of course.

Thanks. Example link? Or name of example antenna?