Micro Short Circuiting on Solar Panels

[FNQGU]

Soooo....

Has anyone had this happen on their solar panels before? The squiggly lines across the solar cells. It doesn't look good, but do you know what it is and what causes it?

I have had it suggested that this is a micro short circuit, which sort of makes sense, given it looks vaguely like a lightning strike, but in reality I have no idea.

There is a bit of water on the panels too, so ignore that. I had just washed everything after a trip when I noticed the weird lines.

8182481823

[pollenface]

All of my solar panels end up with those appearances ... they still work enough to keep my happy :)

[PeeBee]

Can you detect a loss of output? Never seen anything like this before either.

[Bigcol]

how old are the panels?
looks to me like when you crack a circuit board, but not break it completely

do you have a gap between the roof and the bottom of the panels?

[FNQGU]

The panels are about 3 years old. Set up with a gap between roof and panel of about an inch. I've been suspicious of a drop in output but haven't measured it as yet. I think that will be today's job. I was doing a fair amount of driving recently and not relying on the panels so didn't notice it as much as I might if I was camped up. Normally though, the panels will have me fully re-charged by say 10am if in good sunlight. On a couple of days I noticed it was still charging at 11am and through till nearly lunch time, but didn't think much of it as I was running the fridge as a freezer and overnight the battery would have had a lower SOC. Will take a closer look at output today.

[Cuppa]

Have re-posted on a friend's forum where I hope to obtain an experienced solar opinion. Will get back to you if/when I hear. Haven't see this before myself.

[Cuppa]

Ben, 2 responses so far. They refer to the issue as a 'solar worm. Uncertain as to cause, but pics look virtually identical to yours.

One from a friend who reported same concerns as you back in 2017. Measuring showed no drop in output at the time. He reports that 3 years on the panels are still working fine & producing the same rated output. These were bought from Low Energy developments & produced by TN Solar. Co with the following label on them.

81825

One example of a number to show the 'worm' on his panels

81826


Second from another friend who recently bought a job lot of cheap used panels ($14 each for 250w 24v panels). He commented "About half had that 'worm' on them. Some really bad. Testing them in bright sunlight showed very little drop in performance at all".

[FNQGU]

Thanks Cuppa, much appreciated. The 'worm' seems to be only on two cells so far on one of the two panels. Another bit of feedback I received says that the panels could be on the way out. It will be interesting to measure the output and see if there is any difference to the ratings.

[Cuppa]

Thanks Cuppa, much appreciated. The 'worm' seems to be only on two cells so far on one of the two panels. Another bit of feedback I received says that the panels could be on the way out. It will be interesting to measure the output and see if there is any difference to the ratings.

Given that it will be difficult to replicate the conditions under which the rating is made, it probably makes sense to just measure the relative output of both panels to compare.

[Cuppa]

And another response, this from the chap I'd been hoping would reply. He runs a business building & installing lithium batteries & their associated solar systems. I had not mentioned any possibility of 'arcing'. This is his response to seeing your photos.

The wiggly lines are just what they look like, an arc between the grid wires in those modules. If you look real close, you can see the bubble of copper built up each side if the arc line. If the arc repeats too often in the same place, those tiny threads actually get burnt through and module loss starts to become panel output loss. The output voltage open circuit will be the same as the specs on the back of the panel say, but the short circuit current starts to drop off the more damage there is in any module.

The cause, generally poor wiring practices, either panels connected in parallel with under size cable and each panel added in to this under size wire. eg. 6mm auto cable can carry the output of a 150w 12v panel, not 2 or 3 panels all paralleled into the same single run of 6mm auto cable.

The problem starts with the panel getting hot because it can't get all its output through the wire because of a resistance build up, the weakest modules get the hottest in the panel that suffers with the most resistance, generally the panels further down the string. Add a larger capacity panel to the same parallel string and real problems start to emerge.

When the solar controller determines the battery is full or needs to drop to a low voltage such as the step from boost or bulk charge down to float charge. The only way it can do that is to stop accepting any output/input from the solar. The solar array voltage now climbs to open circuit voltage, the hottest modules have the lowest output so therefore the lowest push back against current coming in rather than current going out, so the module becomes a heater and finally an arc runs across the module trying to reduce the current and voltage, a short circuit. This path is now slightly burnt so a carbon track is laid, this makes the path for an arc to travel down easier than the first time, and so the arc continues to burn across the module until it fails.

This is why blocking diodes were included in all earlier built solar panels. When domestic roof top solar started, panels were connected in a single series string to build the voltage and not the current. Bypass diodes remained but the blocking diode became redundant resulting in a cost saving for the panel manufacturer and the buyer being none the wiser.

The arc across the panel is the end result, same thing happens when more than 1 series string is connected into the MPPT controller, the high output from one string reverse flows into the string with a shaded panel, then the same process as mentioned about happens.

My reading of your situation then Ben, combining the info from all sources, is that the panel is on the way out, but it could take potentially a long time to fail & let you down. If feasible fitting a blocking diode inside it's rear junction box, may well prevent any further deterioration, as might increasing the thickness of copper in the wire between panels & to the the regulator.

[FNQGU]

Cuppa - wow! Thanks mate. That is a great explanation and gives me something to work on.

I haven't yet measured and compared the output of the panels and will have to replace my multi-meter first (cheap nasty one fell apart yesterday). The wiring may also be on the small side and the panel with the 'worming' is the second panel of the two in the string. The cells that are damaged are the ones furthest away from the controller also.

I have no idea what a 'blocking diode' is, but no doubt my local Jaycar guy will.

By the sounds of it, this sort of thing could happen to any and all solar panels without the blocking diodes? eg. an expensive panel might also have the same problem as these eBay ones.

[Cuppa]

A blocking diode as I understand it is an ‘electrical one way valve’ to stop back flow. I’m only guessing that fitting one may help. However heavier cabling is probably the better solution I’d think. Blocking diodes will be cheap, & I don’t think doing both would do any harm.

I would hope that any 12v panels made for mobile use would still have blocking diodes fitted, but who knows? I do recall in my early RV days folk having issues with batteries discharging through the panels at night & blocking diodes being the solution.... before they became standard fitments.

Easiest way to check output of panels, if you can get to the individual positive/negative cables is with a dc clamp meter. They can be had fairly cheaply, or come as a combined multi meter/clamp meter. Need to check they are suitable for dc though. Some are only good for ac.