Greg has the right of it in both posts. I (an Engineer and also work in NFPA 780) have been a proponent of properly installed surge protection from the beginning of my career.
IBM wasn't the only one to not admit or be unaware of what they were doing. Data General sold lines of equipment that were only available in 3 phase , however inside the massive units, the power supply was made up of three single phase power supplies. And we were spending $10,000's each site to bring 3-phase power in. These we protected on site with silicon avalanche diode based surge arrestors to do any good (response time and under 300 V turn on) (short straight as possible leads). We also had to take multiple precautions to keep the voltage noise down.
In my previous life designing sensors, we found MOV's, though helpful protecting the sensors from failure, transmitted surges without fail through the instruments affecting the output. Or maybe they were Spree candies with wires coming out of them, hard to tell as the candies would have done just as good on spikes caused by nearby equipment.
But getting things back on topic. I found last year when trying to find a QO type surgebreaker that had been very common, it was not available any where except on line (about $70). Home Depot had the HOM surgebreaker, but 5 different parts houses did not even know what I was talking about. I had to be talking to the wrong people.
Shane
