Car emissions

The Case of the Missing EMI Engineer

Despite a fancy EMI lab, a company’s newest product design can’t pass an emissions standard.

Download this article as a .PDF

The case I describe here perfectly illustrates what can happen when management fails to manage the fallout of a decision.

A small company in Detroit that made anti-locking breaking systems called me in with a frantic request: Their current design was failing the EMI emissions standard for the automotive industry by a factor of over 1,000X and they needed me to troubleshoot the problem.

Pretty ridiculous, huh?

The engineers were dumbfounded, as this had not been an issue with previous designs. When I arrived, they quickly hustled me down to their in-house EMI testing lab which they were eager to show off

For an engineer like me, the test lab was a dream—I estimated there was at least a million-dollar’s worth of the latest equipment there.

Before I began, one of the younger engineers asked me if I could show them how to use the spectrum analyzer. Now I was dumbfounded!

I agreed, but I couldn’t resist asking the question, “This is a really nice lab you have here. Do you have an in-house EMC guy?”

“We had a guy that ran the lab, in fact he put this all together, but management fired him,” was the response.

Now baffled, I showed them how to use the spectrum analyzer and then turned to the problem at hand. It was like watching World War III. The design was failing miserably. The engineers went on and on about how they didn’t understand, that it was a similar design to one they had used for years and had never had a problem with it. Something was very wrong.

So, I asked to see the circuit schematics for the power supply for the previous and current designs.

The switch-mode power supply is usually a major source of EMI in motor control applications. Most power supply designs use a field effect transistor and a gate drive resistor, which forms a low-pass RC filter, to slow things down. (The term for this “slow down” feature is slew rate.)

The first thing I noticed was that the earlier design incorporated a 10-ohm resistor, while the current design did not.

Adding it did the trick, all the noise disappeared and the design met the spec. A simple fix, to an ugly situation.

I couldn’t help but ask whether this was something the former EMC engineer had reviewed. They told me he was already gone by the time they were ready to review the redesign.

Daryl Gerke is a founding partner of Kimmel Gerke Associates, Ltd. The firm specializes in EMC consulting and training.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish