The Case of the Noisy Nuclear Power Plant

Oct. 6, 2018
False alarms cause a near management-meltdown, but the source is troublingly nebulous.

Prior to the 1996 establishment of EMC Directives in Europe, the prime objective of commercial regulations was to protect nearby radio and television receivers. The EU EMC Directive also addressed susceptibility.

You can imagine the havoc noisy environments wreaked on electronic equipment operating before immunity to RF, ESD, and power disturbances became a mandatory requirement.

Early on in my consulting days, I was called in to investigate just such a situation at a nuclear power plant, of all places a nuclear plant is not where you would want to experience interference problems! About once a month, a sensor would trigger an alarm indicating a radiation leak.

This was big stuff for a nuclear facility. When an alarm like this tripped, they would need to shut operations down and isolate the sensor. If the operators could not confirm it was a false alarm in 30 minutes or less, the entire plant would need to be taken offline. It would take months before operations could resume.

Even though in every case engineers were able to determine that the alarm was indeed false, it still took weeks to process all the paperwork necessary to convince the nuclear regulatory commission of that fact.

Management was growing weary.

After about a year of this nonsense and hoping to resolve the issue once and for all, plant management decided to bring me in. I was a little spooked to visit the plant, given the tight security and the armed guards behind glass plates, in spite of the fact that they were unfailingly polite. What I did know for sure was that I didn’t want to be there, but was glad to see they were so serious about security.

The sensor in question was mounted on a smokestack, and we had to walk out onto a huge catwalk. I had to then lean over the rail to get at the box, which was a foot or two away.

My game plan was to try and force a failure. I knew from past experience that a handheld radio was a very likely culprit.

I asked my chaperone whether the facility had a night watchman and if he carried a handheld radio when he walked the property. If so, I wanted to get my hands on that radio. When the answer was affirmative, the smell of rat permeated the space.

First thing I did was turn the radio on and off a couple of times, in close proximity to the sensor. We then went back down to the control room to check whether or not the operators saw any glitches.

The operators confirmed that they had seen a glitch—at exactly the time we keyed the radio. At this point, it was fairly easy to work out a hypothesis: The night watchman goes up on the catwalk to talk a look and thinks to himself, “Hmm, this looks like a pretty good place for me to take a smoke.” At the same time, he’s thinking he better call in to tell people where he’s at.

As for a fix, there were two simple options, and I planned to use both. First, I had brought in some of these cheap ferrite beads to wrap along the cable to filter out the RFI.  Second, I posted a big sign reading: “Do not use a radio within 20 feet of this area.”

The “fix” was both. The ferrites AND the sign. The ferrites probably solved the problem — the sign reminded everyone of the problem, so hopefully it would not happen somewhere else. As far as I know they never had any more problems.

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

Sponsored Recommendations


To join the conversation, and become an exclusive member of Electronic Design, create an account today!