Hi everyone,

This may get a bit technical, so if you're not into this area, just skip this post. I've always had an interest in electricity. Back in my younger days I did quite a bit of wiring, and I still keep my hand in electrical and HVAC diagnosis and repair! My firsthand experience with electrical surges occurred last summer. As a result of windstorms, our office was without power for about 24 hours. Power was then restored for about 30 minutes. I was able to come to our office during this time, and noted that several circuit breakers were tripped. I tried to reset them, but they would instantly trip again. Shortly after this, however, the power went off again, and stayed off for about 24 more hours.

When the power came back on, I tried to reset the circuit breakers, and they reset normally. I tried to turn computers on, several would not power on. When I unplugged them from this surge protector strip, and plugged them directly into the wall, they powered on normally. We did not lose a single computer.

However, of 7 APC surge protector strips, five were damaged. On one or two, the "loss of protection" light was illuminated. The rest appeared normal. However, when I physically opened the cases, electronic components were obviously badly damaged, and there was evidence of smoke and melted plastic inside. All of our battery backups were fried as well, these were APC as well.

I believe that, during the brief time power was restored, it was at a significant overvoltage. These surge protectors burned up internally, protecting (successfully) the computers.

I find the mechanism of this surge protector interesting. It appears that, regardless of brand or cost, the basic protection in all of them is pretty much the same. There is a small solid-state device, connected from both the hot and the neutral line, running to the ground. At normal voltages, this will not conduct electricity. But when the voltage rises above threshold, it will conduct power from that line to the ground, basically providing an intentional short circuit, try to keep the overvoltage from reaching the computer. In our case, I believe that the current to ground exceeded 20 amps, and tripped the circuit breaker.

I have several conclusions from this episode. First, someone asked how many surges can be absorbed. Unfortunately, there appears to be absolutely no way of knowing this. My surge protectors were obviously fried. However, several of them did not indicate a problem by the indicator lights. I am planning to replace ours every time there's indication of a significant overvoltage, and I've arbitrarily pick three years as replacement interval.

Secondly, all of our critical or expensive electronic equipment is plugged into a surge protector, including printers. I don't really understand the printer maker's recommendation not to use a surge strip, I suspect their concern is that a low quality surge strip may have undersized wiring, and cause a voltage drop to the printer, or that numerous items may be plugged into it, again in a poorly designed surge protector this could cause problems. But one printer plugged into a surge strip, plugged into a wall outlet, can't possibly cause a problem, and may save the equipment during an overvoltage. It did save ours!

Third, I believe that battery backup (UPS) are very fragile when it comes to significant overvoltages. All of ours fried. Since last summer's episode, I have been plugging the UPS into a surge strip, and then into the wall outlet. Important: the surge protector goes between the battery backup and the wall outlet, not between the battery backup and the computer!

Finally, as also mentioned above, test your UPs. I just pull the UPS plug out of the wall, see if the computer will stay on, and give it 15 or 20 min. to make sure battery life is sufficient. Last year Sandeep indicated concern about this method, he was concerned about loss of grounding when you pull the plug. But when you pull the plug, you are disconnected from power, so you don't need grounding. Sandeep mentioned that there would be a possibility of a stray current coming in through an ethernet cable or so forth. For our applications, I don't think that is a concern.

Hope this helps someone!

Gene

Gene Nallin MD solo family practice with one PA Cumberland, Md