What would be the way to protect the panels... or a chained reaction... or forget those ! ... to avoid the house burning top to bottom ?



Solar panel damaged by -
- factory defect,
- impact of stone, limb,...
- rain ingress,
-whatever else.




-Pictures borrowed from the web-

My roof is making 4200W right now...

 

Ouch! I sure don't remember anything like that in the National Electrical Code handbook but I'm a bit dated on it. I would think there should be some kind of protective interrupt on each panel. Out of my bailiwick, sorry. That would be a LOT of inline fuses to do each panel.

 

Well, seems the fuses will not even notice arcing inside/within a panel, no matter how many fuses be fitted in the chain of panels.

 

Is that one series string?

 

I have two series of six panels of 40VDC 400W each on the roof yielding two 'channels' of 240VDC feeding two input ports into the inverter, grid tied.

(-)---------------1---2---3---4---5---6------------------(+)240VDC 2400W 10A
(-)---------------1---2---3---4---5---6------------------(+)240VDC 2400W 10A

I have to travel for a month and sheeeting my pants to leave it in operation, nobody home. My roof below, installed 3 months ago :

 

Innocent question: are the panels unprotected for impacts? Would you lose much efficiency through 1/8" polycarbonate? It would buy you a lot of protection. I can't think how to limit damage causing internal shortings, other than prevent the impacts.

 

For impact prevention it may work. If a 2m x 1m sheet is available at less? than the cost of a panel
Aluminium foil under panels to block/catch falling sparks could work also for either internal failures and impacts. Cannot think of a circuit to detect such arcing failure and trigger something (what? -and- how?) to stop a chain action
I would not know where to look for aluminium foil that big.

 

I presume that both strings feed into a single MPPT. Is there a diode in series with each string, rated at >240V ?

 

Innocent question: are the panels unprotected for impacts? Would you lose much efficiency through 1/8" polycarbonate? It would buy you a lot of protection. I can't think how to limit damage causing internal shortings, other than prevent the impacts.

Adding an impact shield between the cell and sunshine costs about 8% of the light to cover both front and back surface reflections and then you have take the transmission of the medium (polycarbonate) into account as well.

 

I'm considering a solar installation and hadn't thought about that. I just assumed that the government would require impact resistance.

The National Renewable Energy Lab had 1 panel out of more than 3000 damaged in an unusually severe hailstorm in 2017.

https://www.energy.gov/eere/articles/hail-no-national-labs-solar-panels-survive-severe-storm

 

video camera to keep and eye on it first sign of trouble shut it down.
Or each cell monitored for short probably be cheaper to just stay with the power company.

 

Correction: Camera to call fire department. Shutting down may not solve arcing.
Correction: Arcing is not from a short but an open.


dl324 : I believe the pictures in post #1 are not from impacts but arcing from open circuits. The glass is tempered, good to a certain point. Yes, failures are not that frequent. Seems arcing comes mostly from factory flaws

No diodes outside the inverter. Two separate MPPT ports take the two strings into the inverter.

 

I'm thinking of this from a physics perspective...

Roofs without solar panels don't catch fire from exposure to the sun because the sun's energy is spread out evenly across the whole roof. The pictures indicate the concentration of the sun's energy in one spot, like a giant magnifying glass would do. So power from adjacent panels must be flowing (coming back to that) into these shorted panels, concentrating the sun's energy from the entire panel array, into this shorted section. It seems to me, this could be prevented by putting diodes in series with each series string, so that current can't flow laterally to the other series strings.

(-)---------------1---2---3---4---5---6------------>|------(+)240VDC 2400W 10A
(-)---------------1---2---3---4---5---6------------>|------(+)240VDC 2400W 10A

Now about the flowing part. Most cordless tool battery packs are made up of several lithium-something batteries in series, like your panels are. It is common to develop a bad cell in these packs. If a cell is shorted, this is not a problem with the battery sitting on the table. It only becomes a problem when current is flowing; when charging or discharging. Then, current through the bad battery (which is now effectively an in-line shunt resistor) causes it to heat up like a heating element. This is why tool batteries nowdays usually have several pins connected internally in the battery, to a few spots in the series of batteries, to make sure the charge is balanced and no bad cells.

So, as long as you're not using the panels (disconnect open), I - think? - you should be fine (but maybe there is still some danger from adjacent cells within an individual panel causing problems; I don't know enough to say.

I suppose you could do like the cordless tool manufacturers and tap a signal wire into each junction between the series panels, feed to a microcontroller, to monitor individual panel health. Maybe it can send you a text. You know, to let you know your house is about to burn down. That way you're not surprised by the pile of ashes when you get home. Or if you want go really advanced, also put a relay between each panel and let your microcontroller control the relays to isolate suspected bad panels so your house stays where you left it.

 

PLEASE ! What do you mean by "short/shorted section/shorted panels" ?

I suspect a similar damage in diesel/gasoline powered generators... If a flying screw/debris/loose/failed solder something finds its way into spinning windings and it "hits the fan" there is no fuse placed anywhere that would stop the electric diarrhea.

 

Well, seems the fuses will not even notice arcing inside/within a panel, no matter how many fuses be fitted in the chain of panels.

The best protection against fires or even just heat and smoke from failures within a solar panel, or even a generator system, is physical separation. It works quite well if done correctly. The down-sides are cost and space. The other option is adequate barriers that would halt the spread of smoke, heat, and fire. Those are mostly cost constrained.
An adequate protection and fault isolation scheme can certainly reduce the area of influence caused by almost any kind of failure, and that would probably include quite a few fuses. And the scheme will also require a great deal of insight and understanding to be both adequate and economical.

 

PLEASE ! What do you mean by "short/shorted section/shorted panels" ?

... effectively an in-line shunt resistor) causes it to heat up like a heating element.

 

"Short" is actually meaning a connection that happened that causes the supply current to bypass the intended load, thus "shortening" the intended path. Probably it is not the best descriptive term, but it is simple, easy, and conveys close to the intended meaning to most people most of the time. But certainly it is not a detailed description of the fault.
In the description of a solar panel fault it is intending to mean an unintended connection across either a single cell, several cells, or across the whole panel. The result would be a current flow limited only by the source resistance, usually much more than the design current.

 

If one of the tiny traces or wires in the solar cell/panel conductive paths gets open, by fatigue, flaw, bad design, overcurrent somehow, the diarrhea hits the fan.

Pick any...
---->

---->

---->

---->

---->

----> https://www.youtube.com/watch?v=Jtd8gThGo0M

Keep in mind a blown fuse or blown path still presents a gap that spark can jump across. Even better if there is conductive/carbon residues deposited from when it blew. THAT is the core of the thing. Has nothing to do with vulgar street talk "shorted".

I may need to investigate if my inverter has built-in AFCI... and if such animal would be of any use.
----> https://www.elprocus.com/afci-arc-fault-circuit-interrupter-working-principle/

 

Keep in mind a blown fuse or blown path still presents a gap that spark can jump across. Even better if there is conductive/carbon residues deposited from when it blew. THAT is the core of the thing. Has nothing to do with vulgar street talk "shorted".

Oh, well, sorry for attempting to help you using words you don't like and choose not to understand.

Keep in mind all these DC arcing demonstrations require current to be flowing for the arc to occur. Do you have any videos to show what happens when you cause an open circuit in the middle of what was already an open circuit?

Just disconnect the array.

 

Microinverters.