Question: Impact of EMP on Solar PV

[Tarrel]

Does anyone know the likely effect of an EMP / geomagnetic storm on an off-grid solar PV system? Would the panels be fried? How could one protect the charge controller? Any thoughts gratefully received.

[NorthernRaider]

Yup, panels are vulnerable as is the regulator and switch gear, its still just a big blooming circuit.

---

The only way to protect the CC is to put in insiode a faraday cage before the EMP hits, so what I'm saying is you will need spares of a really primative electromechanical type device if such a thing exists.

[Lightspeed]

(7 November 2012, 23:08)NorthernRaider Wrote: Yup, panels are vulnerable as is the regulator and switch gear, its still just a big blooming circuit.

---

The only way to protect the CC is to put in insiode a faraday cage before the EMP hits, so what I'm saying is you will need spares of a really primative electromechanical type device if such a thing exists.

I GUESS you'll be OK in a geomagnetiv / solar storm..... This is worth researching.

Full- on EMP as in Man Made, offensive wepon deplyment will take out both te cells and the controller depending on how close they are to the initial blast.

[Skean Dhude]

My understanding is a solar storm is just a low level EMP. Solar Panels are long circuits and I would guess that they are unprotected from solar flares. A bad one could therefore take them out.

This is the problem with solar flares. You never know the intensity of it until a few minutes before it hits.

[Paul]

Geo-magnetic storms will hurt systems connected to long wires.
They last for hours if not days.
Unplug at the panel and you're generally OK.

It's not enough to just unplug the controller.
It's the wire length that causes the problem.

[Skvez]

Let me preface this answer by saying I don't know for certain but I'll give you my 'feelings'.

As with all things related to EMP it depends on the strength of the EMP and the amount of coupling onto the device. There isn't a clean "safe / at risk" categorisation.
Transistors are damaged based on the amount of energy dissipated across the size of the PN junction in the transistor. Since solar panels are just really big (in terms of semi-conductors) PN junctions they are probably much more robust than other electronics.
I therefore 'feel' like the PV panels themselves are "low risk" from EMP.
The charge controller is probably "medium risk" as it's semiconductors are much smaller area and therefore more vulnerable, I don't consider it "high risk" as the runs of wiring connecting to it are relatively short and therefore won't pick up as much EMP field.

If I knew there was a solar flare scheduled to hit the earth I'd be disconnecting the charge controller and possibly putting it in a faraday cage for the duration. I'd probably not be trying to do the same with the PV units themselves because they are harder to get to (typically roof mounted) and physically bigger (harder to make a faraday cage for).

That said I should probably put a small PV panel in a faraday cage just in case.

[Paul]

Skvez (and other interested folk).

Probably teaching you all to suck eggs but it's relevant here.
The individual solar cells are just little diodes, like a LED.
Just backwards. To light a LED you put electricity in, where a PV cell uses light to make electricity.

Each PV cell is fine at MAKING electricity, it just don't like being FED too much electricity the wrong way round.
An EMP pulse or Geo-magnetic strike can do precisely that i.e. force feed something the wrong way round.

Each module in a PV array is usually protected with bypass diodes.
Those have a peak REVERSE voltage rating.
Simple thinking, if one array goes short, the diodes protect the rest of the panels by blowing fuses. These diodes can only take so much though.

Or, in plain speak, "Please don't squirt too much voltage the wrong way into me as the most I can take is":

1. 8A, 45 PIV (Peak Inverse Voltage) for 72 & 96 cell modules.
2. 12A, 200 PIV (Peak Inverse Voltage) for 96 (commercial use),
128 cell modules.

A typical set of figures from http://www.sunpowercorp.com

Reading Google articles on solar flairs, http://www.thespacereview.com/article/1549/2 is a good one.
It mentions a 1.5KT nuke frying diesel generator windings by breaking down the insulation resistance.

I used to test 40KVA Houchin diesel generator windings for a living using 1500 volts insulation test "zappers".
If they survived that particular test, the winding was deemed serviceable.

So I'm thinking a 1.5kt nuke must generate more than 1500 volts in one of its three pulses of energy.

Look above and 45 / 200 PIV figures look a bit sick against more than 1500 volts the wrong way round i.e. BANG (but quietly).

The suns flairs are WAY BIGGER than a piddling little 1.5Kt nuke.
Further away admittedly BUT still pack a considerable punch when they hit the earth.

A solar eruption hits in 3 ways (just like a nuke EMP does).
First comes electromagnetic radiation,
Then radiation, and finally the plasma from the sun.
This is the bit that screws up electrical systems.
1989, Canada (Quebec) was brought down by one.

Now if a PV panel gets hit by this 3 stage blast, PV cells or protection diodes may blow.
If the PV cells blow open circuit not a problem, if they blow as a short it's the same effect as dropping a spanner across a car battery (just not so spectacular).

Everything dies as it shorts out it's neighbour, and the next one, etc. A cascade failure.

Adding an earthing strap won't help as all your doing in grounding the casing.
So the only logical (but totally impractical thing to do) is to Faraday cage the panel.

In the long run it's just going to be better to simply unplug any long wires to stop induced voltage from the EMP / Solar flair feeding back into the panel or the controller and hope for the best.

[Skvez]

PV cells are similar to diodes but they're not "little", in semiconductor terms they're gigantic. Bigger is safer when it comes to diodes vs EMP.

I have never seen a PV array with a "bypass diode" for each junction of the array, they have output diodes to stop current flowing into the array (for instance if you have two arrays in parallel). This diode may offer some protection against an induced Voltage in the PV array wiring damaging the PV array itself. You could certainly replace this series diode with one of similar Amp rating but as high Peak Inverse Voltage as you can find.
A Silicon diode initially fails to short, then if enough current flows through the substrate it will melt open cct. Adding a fuse between the battery and the solar panel is unnecessary under normal operation but might help save your PV array in the case that an EMP destroys the protection diode and the shorts the battery through your PV array. On the other hand the PV array may be destroyed before the fuse had time to heat up and melt (still for the cost of a fuse it's probably worth trying, every little bit helps).

The EMP doesn't generate a Voltage, it emits a field that induces a Voltage based on the coupling between the field and the 'wire' it encounters.
Your generator is at risk because the windings are a very long wire all wrapped up into coils and the output of the generator is connected to potentially miles of load wiring.
In contrast the wiring connected to the solar cells is probably only tens of meters and so the induced Voltage could be order of magnitudes lower than that experienced by a generator beside it (the "1500V" you mention).

If the PV cells blow open or short your array is toast (I suppose you could break it up into individual cells in the hope that some of these survived and could be re-connected into a useful string).

[Paul]

PV cells usually have a surface mount bypass diode
Typically (1A 50Vrwm)

and the panels a 120% OL fuse with a panel bypass fuse.

You're right about the windings being long but for length the panels connect to an inverter, that connects to the house wiring, that to the grid. Hence my desire to isolate and cut the length of wiring.

[Skvez]

I've never seen individual bypass diodes on each cell as the diagram shows, they're certainly not present on the little cells I've taken apart but they might have been too cheap to include them. If they're on the big PV panels I have they're well hidden.

Certainly if your solar is connected to an inverter and the grid it's very vulnerable.
The first rule for surviving EMP is to disconnect yourself from the mains grid.
This is one of the reasons I prefer to run my PV system as a standalone 12V supply rather than hook it up to the grid through an inverter.