How did you get into electronics/ engineering and when did you start?

At nine years old, I build a crystal radio and it worked. I was fascinated! That led to my ham radio license at 14, and ultimately a BSEE from the University of Nebraska. My first job was working on VHF/UHF radios at Collins Radio. Two years later, I moved to Sperry Univac Defense Systems, where I got introduced to EMI.

How did you get into EMI?

A few weeks after joining Univac, I was asked to join a new group doing EMI design work. It seems having “radio” twice on my resume doomed me. Frankly, I wasn’t sure I wanted to do it, so I opted for a six month review. After six months, I was hooked, and have been involved with EMI ever since.

While at Univac, a co-worker (Bill Kimmel, my business partner) and I started consulting on a part time basis. We both had a lot of military EMI/TEMPEST experience, but to avoid conflicts, we decided to pursue newly emerging EMI problems in the commercial markets. We eventually went full time in 1987 (the day the market crashed) and have been designing, training, and troubleshooting EMI issues ever since that fateful day. We are now active across a wide range of industries.

What are your favorite hardware tools that you use?

For emissions, a spectrum analyzer with antennas or probes. I like “sniffer” probes for checking out circuit boards, and current probes for checking cables. For susceptibility, an ESD gun and a hand held VHF/UHF radio. The former can also be used to simulate some power disturbances too. The latter can be used to check for gross levels of RF suscpetibility.

What are your favorite software tools that you use?

I do very little with software. Most of my computations are “back of the envelope” calculations. As a consulting engineer, I don’t have the luxury of doing in-depth analysis — my customers need immediate fixes today, not highly precise answers six months in the future.

For an engineer that deals with EMC compliance issues regularly, what lab tools/resources would help him most?

For emissions, a spectrum analyzer, antennas (for radiated emissions), and a LISN (line impedance stabilization network). Supplement with clamp on RF current probes, RF sniffer probes (magnetic loops preferred), and a big assortment of clamp on ferrites. Copper tape and aluminum foil are also useful resources.

For susceptibility, an ESD gun and some way to force RF failures. In the latter case, a hand held VHF/UHF radio can give gross information. For more details, you’ll need a signal generator/amplifier/antenna, and a shielded room or TEM/GTEM cell. The TEM/GTEM is practical only if your devices are relatively small. Same resources as above – ferrites, copper tape, and aluminum foil for starters.

One of the best tools, however, is the grey matter between your ears. Think like a medical doctor. If the design is new, how can you innoculate it against possible EMI diseases? If troubleshooting an existing problem, how can you first diagnose the condition so you can write a prescription?

What is the hardest/trickiest bug you have ever fixed?

Although I deal with a lot of problems, the trickiest was probably a prototype production welder that used 15 kW of RF to start the arc. The arc itself could be several inches long. As soon the arc started, all the electronic controls immediately shut down.

Upon initial inspection, I was not sure the problem could be fixed. But thanks to aluminum foil, hose clamps, and an undying faith in EMI shielding theory, I got it working. It still amazes me, though.

Do you have any tips in correlating EMC measurements between different labs?

In my experience, the differences between commercial EMC labs are small. If there are big differences (i.e. greater than a few dB for emissions), try to find out why before continuing. The differences are usually due to the system under test. Look for different hardware (including cables), different peripherals, and even different software. No change is too small to be insignificant.

On the other hand, correlation between a commerical lab and the engineering lab can vary widely. Rather than focus on exact correlation, establish a baseline before making any changes. Then look for relative improvements compared to the baseline. When the improvement looks good, you are ready to go back to the commerical lab.

What is your approach to solving an EMC emissions issue?

For conducted emissions, I focus on the power interface. A quick test is to add a power line filter external to the box. A spectrum analyzer and cable current probes can help determine coupling modes — common mode or differential mode.

For radiated emissions, I use a simple model of “hidden antennas/hidden transmitters.” Tools needed are a spectrum analyzer, antenna, sniffer probes, and cable current probes. (If in the engineering lab, the antenna can even be home made, as you are only looking for relative improvements.)

Here are some steps to isolate “hidden antennas”:

• Remove all the cables except the power. If the unit still fails, wrap the unit (if in a metal box) in aluminum foil. If it still fails, the power line is an antenna, and go to work on it. (Try a ferrite on the power cable.) If the unit passes, remove the foil and hunt for the leaks. (Try copper tape and gaskets on seams, cover large openings, etc.)

• Once the unit works with just the power, add cables one at a time and observe the results. If one fails, either fix it or put it aside and continue. After the offending cables are isolated, work on them. (Try ferrites or add on filter connectors.)

Here are some steps to isolate “hidden tranmsitters”:

• Probe the circuit boards and internal cables for hot spots with a magnetic field (loop) sniffer probe. Apply fixes such as power decoupling, filters (ferrties or RC filters on clock lines) to the biggest sources. Try bulk ferrites on internal cables. Retest for radiated emissions.

• Note that the highest “transmitters” may not result in the highest emissions, so if the problem still persists, try more fixes on lower level “hidden transmitters.”

Finally, don’t be too scientific, and only apply one fix at a time. Rather, keep adding them until you achieve success. Emissions problems are often like fixing a leaky boat — apply only one patch at a time and you’ll never fix the overall problem.

What is on your bookshelf?

I probably have over 100 EMI/Signal Integrity books on my bookshelf. If I could only have four, however, they would be:

“EDN Designer’s Guide to EMC” written by Bill Kimmel and me. Written as a beginner’s guide for non-EMC designers, with simple explanations and recommendations.
“Electromagnetic Compatibility Enginering” by Henry Ott. Written at an intermediate level for working engineers, with clear expanations.
“Introduction to Electromagnetic Compatibility” by Clayton Paul. Written at an advanced level, yet still very readable.
“High Speed Digital Design” by Howard Johnson. Although focused on signal integrity, almost everything he suggests applies to EMI/EMC as well.

Do you have any tricks up your sleeve?

I’m a big advocate of EMI planning. That means reviewing designs early, and if necessary, performing “precompliance” tests. Waiting until the end of a project to validate the EMI design can be very costly and difficult to solve.

At the circuit board level, pay attention to EMI-critical circuits (clocks, resets, voltage regulators, analog, and I/O). At the system level, pay attention to the enclosures and shielded cable terminations (no pigtails!) I often joke that I put two kids through college by repeatedly solving this last problem.

What has been your favorite project?

That is a tough one, because after all these years, there have been so many. Working on medical devices has been particularly satisfying, knowing that my efforts may have helped save lives.

But one project stands out — an experimental satellite I worked back in 2001. With a wide range of potential problems and extreme environmental constraints, it was very challenging. Launched several years ago, it is still working (beyond the original life estimate.) So when I gaze into the night sky, it is kind of neat to know some of my work may be flying above me.

Do you have any note-worthy engineering experiences?

The most note-worthy occurred when I was still an engineering student. While home on a visit, I was asked to hook up a new air conditioner. It was a very hot day, and after running the 220 volt wiring through a fiberglass filled attic, I was not in the best of moods. Upon opening the access panel, I was annoyed to find small wiring that was not even color coded correctly. But what the heck — just connect black to black, red to red, and the white to yellow. Later (after smoke poured out of the unit), I realized I had connected the 220 volts to the 24 volt control wiring.

The worst part was explaining what I had done to the HVAC contractor in our small town. I’m sure everyone eventually heard about that one. But the lesson was a good one — when dealing with power wiring, I still never assume anything and always double check everything.

What are you currently working on?

Last year we decided to increase our web presence. We completely remodeled our web site, which now includes a blog, an on-line store, and numerous free resources (bibliography of EMI publications, back issues of our 20 years of newsletters, etc.) More features are planned.

To enhance our EMI training, we’ve added webinars. These augment the public EMI design courses we have co-hosted with Tektronix for 20 years. We’re exploring other ways to expand EMI awareness within the design community as well. Please visit our web site for details.

On a personal basis, last year I launched a blog on consulting. When my own son asked about consulting, I decided it was time to share what I’ve learned. I’ve now spent over half my career as a consulting engineer, and it has been most interesting. If you visit, be sure to click on the special welcome for geeks.

What direction do you see your business heading in the next few years?

We expect to see EMI problems increase. These problems are driven by advances in the state of the art — higher clock speeds, faster edge rates, increased circuit density, and lower power levels. These trends will affect both EMI emissions and immunity across the electronics spectrum, from simple embedded controllers to the highest speed computers.

What challenges do you foresee in our industry?

I heartily agree with my business partner Bill Kimmel — the biggest challenge is to remain technologically competitive in the world wide marketplace. Finally, I think it is a great time in world history to be an engineer.