Solar question

[Scillypete]

On Tuesday early in the morning we had a fairly intense thunderstorm, later that day I sailed to Mousehole, everything was working but the solar wasn’t charging. On Wednesday I investigated further, I have two sets of 2x100w panels in parallel these are fed to a junction which then goes to the mppt charge controller. There was zero voltage from either set of panels to the junction. I checked the wiring right through from the panels mounted on the doghouse until I found a section blown out, fuses were all good, cut out the blown wire and made a joint, voltage measured at 25v in good sunshine so that set was now providing charge. Went through the wiring on the A frame set and the only fault was a possible dubious connection on the mc4 so remade it and voila voltage restored.

now the real question is could a close by flash of lightning cause a solar power surge which caused weak spots to blow even before hitting the fuses. Some of those lightning flashes can be extremely bright would that cause a sudden electric pulse from a solar panel. I am trying to work out why both sets of panels would have failed at the same time and struggling to come up with an answer.

 

[Tim Good]

Interesting…. When you say you “found a section blown out” do you mean melted? Like as though your cable gauge was insufficient for the amps running through it?

 

[William_H]

Interesting question but no I don't have an answer. My home solar PV system has 5 amp fuses in PV to inverter wiring both pos and negative. Despite max current capability of PV array about 3 amps one fuse has blown twice now. So similar question.
A lightning strike can involve large currents (large magnetic fields )which can then induce current in to any wiring nearby. The solar PV panels have in effect a long wire trail between banks of cells which could pick up excess current.
Re the idea that a lightning flash might induce the panel to generate more current than normal, I am a bit skeptical. A lightning flash can seem very bright in a dark sky but full sunshine is very bright also. Need some real light level figures here. Certainly a solar panel can respond quickly to a flash generating a pulse of current. Small panels have been used I think as flash detectors for slave camera flash or my welding helmet uses PV panels to detect weld flash and darken glass very rapidly.
ol'will

 

[sarabande]

A reasonably good, if rather general article on lightning and solar panels

Lightning And Solar Panels: What You Need To Know - DiscoverSolarPower.com

Although they are dealing with landside panels, there is a recommendation for earthing and lightning arrestors about halfway through.

I can't remember reading about the effect of lightning on boat solar panels anywhere else .

 

[William_H]

That link Discover solar power.com is interesting but has many really bad statements. How about the photo of the electrician with multimeter inspecting centre of PV panel. The meter is misleading. Being all glass covered there is no way you can make a meter connection.
Lightning arrestor will be in the form of a high resistance spike or brush that tends to discharge atmospheric charge build up on a continual basis so discouraging a lightning strike. The other more common form is a very low resistance conductor from a high point down to a good earth. This will attract lightning to itself so diverting high current strike from valuable assets below and around. So found on church steeples.
On a sail boat it will inevitably have a low resistance ali mast as built in lightning conductor. What you must do is ensure that the base of the mast has a similar low resistance connection to the sea. Any electrical resistance will encourage current to heat the conductor and when it fuses cause voltage to arc across the gap. So wet GRP will just boil the moisture and destroy GRP just like a tree explodes with lightning strike. No do not connect stay wires to sea. Stainless steel has a higher resistance to current and will melt with not so much current compared to ali. ol'will

 

[Scillypete]

Interesting…. When you say you “found a section blown out” do you mean melted? Like as though your cable gauge was insufficient for the amps running through it?

Just on one cable and just the positive side, it’s probably almost a five metre run of tinned cable that was newly fitted 18 months ago and luckily for me the damaged part was in a locker space so I could see it and get at it and not in the cable run conduit.

 

[Tim Good]

Just on one cable and just the positive side, it’s probably almost a five metre run of tinned cable that was newly fitted 18 months ago and luckily for me the damaged part was in a locker space so I could see it and get at it and not in the cable run conduit.

Sorry you don’t mention how it’s damaged? Do you mean it melted due to current overload? It wasn’t at or just before a joint?

 

[Travelling Westerly]

Any chance that an induced EMF/transient voltage spike on the cable/conductor occurred during the nearby lightning?
I'm no expert but seem to recall some weapon engineers going on about it some time back hence my very vague knowledge about it...
I'm sure there are some electrical or RF experts around who can debunk this theory if it's wrong

 

[Scillypete]

I should have taken a picture at the time but I just wanted to get it charging again and only questioned the why afterwards.

Sorry you don’t mention how it’s damaged? Do you mean it melted due to current overload? It wasn’t at or just before a joint?

it was roughly about the middle section of an unjointed cable the positive side looked as though it had burst over about a 3/4 inch section with I suppose could be a melted part of the outer black cable cover, the actual remaining wire just crumbled, it’s a two core tinned cable and not that old same section wiring as from the panels to the mc4 connection that the panels come with. I’m wondering A; was it a damaged or flawed cable, or B; did it get damaged whilst installing it and pulling it through the cable ducting which was already fairly congested and thus cause a weakness within the wire

still more baffling is the why both sets of independently wired panels failed in similar fashion at the same time.

 

[Scillypete]

Any chance that an induced EMF/transient voltage spike on the cable/conductor occurred during the nearby lightning?
I'm no expert but seem to recall some weapon engineers going on about it some time back hence my very vague knowledge about it...
I'm sure there are some electrical or RF experts around who can debunk this theory if it's wrong

no other electronic devices or components failed to work so I can pretty much rule out a lightning strike I would think

 

[Travelling Westerly]

no other electronic devices or components failed to work so I can pretty much rule out a lightning strike I would think

I didn't say it was a direct hit to you, nearby I said.
I don't pretend to understand it fully and would refer to an RF expert. It often affects coiled wires etc.

 

[Scillypete]

Just for a bit more clarity the boat has been in regular use and we had been off for several days then spent the night at home with the thunderstorm in the morning before setting off for Mousehole, solar was good before but not after. I have a battery monitor plus meters for the solar so I can definitively identify when it went awry

 

[William_H]

On my little boat I had a battery cable running under the carpet adjacent to the side of the bunk.(for lights and radio) Had a failure of power and could not figure power in one end of wire and not at the other end. Finally found a point where apparently the plastic insulation had been holed allowing moisture in. About 3/4 inch of insulation was bulging. Wire inside gone to dust. On another occasion I had a trailer lights cable 5 wires in outer sheath. One wire failed. I eventually tracked it down where in the factory they had just twisted a new wire onto old and fed in to sheath. So definitely factory induced failure. So your wire could have failed from damage perhaps induced by lightning current or just coincidental to storm. ol'will