080405-1917 EST USA
I second ultramegabob's suggestion in post 40. I have had one of these for 40 years at my home, and I also have some secondary protection nearer some equipment. In that time I have had one failure in a TV set that was probably lightning related.
About 25 years ago I had a computer directly connected via RS232 with a plotter. There was 30 ft or so of RS232 cable between the equipment. A lightning storm wiped out RS232 components at both ends. That is one reason that today I make and sell RS232 isolators. These also are a great solution to ground path noise problems, and provide high baud rate capability at long cable lengths.
Some side information. I check the ground path current individually on four different devices:
Hewlett-Packard 5Si laser printer, 10 MA.
An older computer, 3 MA.
A UPS power supply with nothing plugged in, 2 MA.
An 8 ft Slimline with a noise filter, 3 MA.
Since the input filters on these machines probably include capacitors from the power leads to the chassis and therefore to the cord ground pin I can expect the the signals from each to be phase correlated with the AC line and in turn should directly add to each other. If all four were connected to the same circuit, then the sum would be about 18 MA.
.