I have used three small solar panels from scrapped garden ornaments to charge 4 NiCd AAA cells in an ultrasonic cat-scarer. There is a diode in -line to stop the batteries discharging back through the panels.
This unit has been working fine for over 3 months but had stopped recently. On investigation, one of the panels seems to have changed its polarity,and provide a reversed voltage when lit, so counteracting the voltage from the other two panels.
I've gone through "Am I going mad?" and "Did I measure correctly?" and the three panels are still soldered together in series so it's a definite oddity.

Does anybdy have any ideas or experience of this sort of thing at all?

 

How many Solar Panels in series (3) ?
And what is Voltage, Amps, Watts at Max Power?

What is the voltage on each panel under strong sunlight?
If one Panel is Open then will the voltage appear to be reverse voltage on that panel when connected to a load?
DIsconnect all PV Panels from any load and then verify reverse voltage.

 

I was testing disconnected from any load, and testing individual panels as well as the whole set.
The other two panels each provide about 1V under an 11W CFL in an anglepoise. The reversed one (which is a different model) provides about -2V. The whole string supplies around -10mV.
Feeding output direct to a multimeter, the other two provide 100mA and 50mA, and the reversed one about -100mA.

In other words, yes, I was measuring correctly and the reversed cell is working reasonably well as a solar panel, just in reverse, after months of forwards working!

 

I think mvas hit the nail on the head. If there is *any* current in the loop - any load- then the two cells that are producing voltage will add and show up as a reverse voltage on an open panel.

 

Just reverse reversed cell & you should be back in business? I've never seen a reversed cell , but many of my little stash of cells have no polarity markings. Wild guess: Reversed cell was origionally reverse connected with a temporary short which finally opened??

 

I did consider just reversing the cell now, but it really was working the right way round before! It's not just a case of the marking being reversed and me not noticiing first time around. The 3 cells between them were charging the 4.8V of NiCd battery with this cell doing its bit. Even if it was shorted internally, the other 2 cells would not have provided enough voltage to keep the NiCds charged.
So I'm really intrigued as to how this could happen, and I'm also wary of using the reversed cell any way at all - it seems very dodgy to me.
I'm not really asking here to work out what to do, I'm asking what the *** happened!

 

Test each cell for current output independantly - totally disconnected from the rest of the circuit. Still think you will find the 'reversed' panel actually an open circuit.