Performance of damaged panels

[adaviel]

How resilient are commercial panels to damage? If, say, one cell is cracked in a 50-cell panel, do you get no power, or 98% power or does it depend on the design ?
Are commercial panels typically 100% series-connected (50 cells in series), or series-parallel (25 in series in parallel with 25 in series, or 10//10//10//10//10 etc.)
If one cell is cracked across, do you get no power from it or does it depend on the crack ?
How granular is the monitoring on a typical commercial installation ? Would you get monitoring per-panel, or within a panel, or just for the complete installation ?

[SunEagle]

Hello adaviel and welcome to Solar Panel Talk.

The way a pv panel works depends on the number of cells and how they are wired. Each manufacturer may wire the cells differently. Although I would guess that any cell that is cracked would no longer work and any of the cells wired in series with the cracked on will probably also not work.

Monitoring depends on the manufacture of the panels and inverter although most do not offer individual monitoring unless you have included micro inverters with each panel.

[inetdog]

How resilient are commercial panels to damage? If, say, one cell is cracked in a 50-cell panel, do you get no power, or 98% power or does it depend on the design ?
Are commercial panels typically 100% series-connected (50 cells in series), or series-parallel (25 in series in parallel with 25 in series, or 10//10//10//10//10 etc.)
If one cell is cracked across, do you get no power from it or does it depend on the crack ?
How granular is the monitoring on a typical commercial installation ? Would you get monitoring per-panel, or within a panel, or just for the complete installation ?

A commercial panel will almost always contain what are called bypass diodes in parallel with two or more series wired groups of cells making up the panel. If a 60 cell panel is divided into three groups of 20 cells each you will normally see three diodes inside the junction box of the panel. (Some newer panels may actually integrate them inside the panel itself.)
If one cell in one group gets cracked and opens up you will lose the output of the 20 cell group it belongs to.
You may still read Voc as normal if the cell is not completely open, but when you put on a load you will find that Imp is normal and Vmp is 2/3 of normal.
A 50 cell panel would probably be divided into two groups, so you would get 1/2 voltage at best. And if one cell in each group of 25 is damaged you would get essentially nothing out.

[bcroe]

How resilient are commercial panels to damage? If, say, one cell is cracked in a 50-cell panel, do you get no power, or 98% power or does it depend on the design ?
Are commercial panels typically 100% series-connected (50 cells in series), or series-parallel (25 in series in parallel with 25 in series, or 10//10//10//10//10 etc.)
If one cell is cracked across, do you get no power from it or does it depend on the crack ?
How granular is the monitoring on a typical commercial installation ? Would you get monitoring per-panel, or within a panel, or just for the complete installation ?

Commercial panels are amazingly tough. I have run some 3 million panel hours here without a single failure. If I did manage
to crack a panel, I'd write it off. Observations have been that once the seal is broken, semiconductor devices die in a fairly short time.

A typical 60 cell panel is divided into 3 groups connected in series. Failure of a group to putout adequate current under heavy
load will cause its voltage to drop out and current from other groups to flow through bypass diodes instead.
The intent of this function is to allow partial operation of a panel with some temporary shade, not to deal with faults.

I only monitor an array often enough to see if any faults have developed in the last 24 hours. Its possible to monitor individual
panels every few minutes; that system is a lot more likely to fail than the panels.

The new faults I've seen relate to wiring and other equipment. A couple no name panels I picked up cheaply were initially bad
and had to be replaced, because they couldn't keep up with the others. Monitoring might help locate them, but you don't need
it if you understand some things about electrical circuits. Bruce Roe

[adaviel]

A 50 cell panel would probably be divided into two groups, so you would get 1/2 voltage at best. And if one cell in each group of 25 is damaged you would get essentially nothing out.

Thanks.
I was thinking of something like a solar-powered car. If someone drops a wrench on a cell and cracks it, taking out 25 seems unacceptable. I guess one could use bypass zeners in smaller groups, or perhaps per-cell. Maybe it's not that much an issue anyway; I have a broken cell that seems to still work (not in a panel though), and it seems unlikely a break would go across the tabbing wire, so it might require two separate breaks either side of the wire to completely disable a cell and thus a string.