Panel heat dissipation with failed MLPE / microinverter

[daklein]

Conditions were just right to notice this today. Panel with failing microinverter melts the snow off, the rest didn't melt. Normally I clear the snow but didn't today since the very thin snow would just melt off _if_ the sun came out much.

The array has old m210/m190 microinverters, and I've been skeptical of the panel on the right, from looking at the monitoring. It works fine some days, like yesterday. Over past 7 days it's only 10-20% lower output than the rest. Cloudy day output today was only 25Wh with zero output in the middle of the day, vs. 100-125Wh for the other panels.



Failing microinverter output, last two days, melted snow on the right:

Good microinverter output, last two days, no melted snow on the left:

I suppose this means the failing microinverter is short circuiting the panel and dissipating the heat in the panel? How does that work with the panel I-V curve, there's no power if the panel voltage is shorted. But maybe it's the panel internal resistance * short circuit current squared, which wouldn't be zero? (Talesun 260w, courtesy of DC-Solar!)

 

[sunworshipper]

The non-functioning panel generates 20% more heat (energy that is not converted to photo current) than the other panels, which may translate to a couple Celsius higher temperature, just enough to melt the dusting of snow.

 

[timselectric]

Short circuit current is maximum current.
This would create enough heat to melt a dusting of snow.

 

[sunworshipper]

Short circuit current is maximum current.
This would create enough heat to melt a dusting of snow.

the Joule heating by Isc would be occurring outside of the panels, in the wires and not contribute to the panels temperature. If the panels terminals are open, the photo electrons would recombine within the cells and add to the overall heat accumulation in the cells. A simple test would be to short one of the other, know to be good panels and see if it melts snow under similar conditions.

edited for grammar

 

[timselectric]

the Joule heating by Isc would be occurring outside of the panels, in the wires and not contribute to the panels temperature. If the panels terminals are open, the photo electrons would recombine within the cells and add to the overall heat accumulation in the cells. A simple test would be to short one of the other, know to be good panels and see if it melts snow under similar conditions.

edited for grammar

Current flowing through a circuit, will create heat in every part of the circuit.
A short circuit, produces maximum current and maximum heating.
And yes, it would be easy to verify on the next panel over.

 

[sunworshipper]

Current flowing through a circuit, will create heat in every part of the circuit.
A short circuit, produces maximum current and maximum heating.

Yes, the current creates heat throughout the circuit but not uniformly. It's high where the resistance is high so that I2R is maximized.

Let's ignore short circuit scenario for the moment. If a panel is connected to a MPPT controller and a battery, the battery would receive all the harvested PV energy, with very little or nothing wasted as heat in the panel (I'm ignoring the energy consumed by the MPPT electronics). However, by your argument, the harvested energy would be dissipated in the panel/cells since the current is also flowing through the cells. We can't have the battery receive the harvested energy *and* the panel dissipate a similar amount of energy.

 

[daklein]

The circuit is a set of resistances, and a voltage source (the cells). The wires and connectors should be pretty low resistance, and not much voltage drop. The cells and bussbars in the panel have some amount of internal resistance also. If the panel is driving a load like a battery directly, or a mppt controller, the load would have most all of the voltage drop or resistance or impedance, and the panel & wiring would have not much voltage drop or loss. If the microinverter is failed, it could be failed open or shorted. In this case, it must be shorting the PV inputs when it's failed or turning itself off. So it looks like the total resistance of the wires, connectors and shorted microinverter, all together are small enough that the panel internal resistance is the primary load.

 

[PVGeezer]

First quiz question: Name a device that gets colder when you turn it on. And no not a fridge because only the inside gets colder.

Answer: a solar module

2nd question: Name a second device that gets colder when you turn it on.

Answer: ....

I'll wait...

And a module that is shorted (operating at Isc) will be the same temp overall as one operating of Voc. The one at Isc will have some cells hotter and some colder though, because some cells will be forward biased and some reverse biased.

 

[daklein]

And a module that is shorted (operating at Isc) will be the same temp overall as one operating of Voc). The one at Isc will have some cells hotter and some colder though, because some cells will be forward biased and some reverse biased.

I have seen light frost melted selectively by one or two individual cells on a panel.