Electromagnetic Pulse

Re: Electromagnetic Pulse

Here is a pretty reasonable looking EMP Primer on Wiki (yea--I know--but Wiki on technical subjects can be pretty good--at least as good as an encyclopedia).

Basically, you will have different frequencies that will travel different distances and such... Small stand-a-lone electronics (flash memory sticks) would require very high frequencies--which do not travel far.

Devices that are attached to utility power can be badly affected by low frequency pluses that energize wires that are miles long--and can focus energy at connection points (i.e., utility power and computer connections).

Off-Grid solar is small(er) so, I would guess, are not going to be as affected by far away sources.

I have done ESD testing (electrostatic discharge; smaller version of EMP) -- basically using a capacitor, high voltage, and a fast high voltage relay to do air and contact testing (simulate human body and equipment static discharges).

This was almost 30 years ago when I started doing it... And from what I have seen:

1. for spark discharges, lower voltage discharges (below ionization voltage of air--700 volts or less--if I recall correctly) generally affected nearby equipment less than a 7,000-15,000 volt discharge. The high edge rate of lower voltage discharges caused higher frequencies which "got into" the computers easier.
2. We also did contact discharges (same test rig, with a high voltage relay to allow current injection with high edge rates that was not affected by ionization voltage issues). Very nasty.
3. As equipment had to meet better EMI regulations and susceptibility tests (pushed originally by European Regulations)--the better the equipment was constructed to limit radio emissions, the better it passed susceptibility testing.
4. Every computer that passed EMI (radiation) testing--always passed the 5 or so susceptibility testing (ESD, conduction from AC "shower of sparks" simulating Florescent Fixture noises, surge testing--electric motor/inductive testing, etc.) with ease (I never had a failure).

So--my two cents worth--following good EMI design techniques (metal boxes, with screws, shielding fingers, shielded cables, Ferrite Blocks, no long wires in loops with no shielding, lots of twisted pairs, etc.) would also do pretty well against outside EMP...

I have seen testing (back in the late 1970's), a simple air gap ESD test would take out computers around the lab in a 100' radius (including the device under test). When I was working in, for example, telephone switch rooms for the phone company--we could not use electronic flash as its EMP would take down the computers. In Alberta Canada, they would have problems with dry air and down coats in winter--guy wearing a down coat would just walk down the isle and crash everything as he passed.

Just a few fixes (improved internal grounding, and such)--and the DUT would pass with flying colors.

When wiring your DC (panels, DC wiring, etc.)--don't make big loops of your + and - cables. Instead, always run them as matched pairs (tied together, same lengths, etc.). And if practical, give them twists (i.e., CAT 5 computer cable has a twist every 1-2 inches).

Now, the interesting thing about off-grid systems--from what I remember our host (WindSun / Admin) answered years ago--it was the AC side of the inverters that usually failed from a lightning strike (classic high energy EMP).

So--if you want things to survive--having a spare inverter may be a good thing to have in your metal spares chest.

-Bill

I should add that putting your wiring in metal conduit (properly bonded to the metal breaker/wiring boxes) is a good thing too.

Solar Guppy has done a fare amount of writing here on his lightning protection/grounding philosophy... And it is quite different from NEC / typical recommendations--but I can see quite a bit to suggest following his methods.

An issue is that "mixing" philosophies in one installation can cause its own problems.

In the end, treat EMP like radio waves (a broad range from low to high frequencies) and avoid issues with making loops of wires (antenna) and bonding gaps (long conduit that gets to within an inch or so of an inverter's wiring box, without a solid metal braid or equivalent connection), and such.

Earth Grounding--I am not so sure that this matters a whole bunch... The bigger help is to make sure that there is a good, heavy gauge (stranded or "flat" braided wire is even better) that attaches to all of your metal equipment cases into one big common electrical "ground" plain.

For safety, you need to dissapate the energy from a direct lightning strike somewhere--so a good (set of) ground rod(s) and a single electrical ground point where everything is connected in a "star pattern" is not a bad thing.

Here is a good thread (with some links) about Lightning and Grounding...