Is this panel a problem?

[Heataholic]

So my install went well enough and the system has been running for a little over two months now. I have a 10 panel array facing 195 @ 18 degree tilt, not sure how accurate the tilt is since it seems low to me but that is what the installers put in. I have one panel that is always producing a little lower that the other 9 on the string, about 9 watts average on a 230 watt peak for the neighboring panels.

Attached is the generation from today but it mirrors most days and I'm talking about 1.2.10 (bottom right). Is this an issue or am I looking at it too closely? I haven't asked the installers to take a look at it yet, should I?

[max2k]

So my install went well enough and the system has been running for a little over two months now. I have a 10 panel array facing 195 @ 18 degree tilt, not sure how accurate the tilt is since it seems low to me but that is what the installers put in. I have one panel that is always producing a little lower that the other 9 on the string, about 9 watts average on a 230 watt peak for the neighboring panels.

Attached is the generation from today but it mirrors most days and I'm talking about 1.2.10 (bottom right). Is this an issue or am I looking at it too closely? I haven't asked the installers to take a look at it yet, should I?

it is within 5-7W from panels next to it, I think it is normal. Is it partially shaded by anything?

[Heataholic]

No shade on the entire roof all day.

[ButchDeal]

If you scroll through the day does it get close to the others at times?
It is fairly close and pv modules are generally rated with some variance in production capability.
it could also be a but of dirt in that module as well.

[SunEagle]

The 7 to 10watt difference seems to carry through out the day so it is possible there is a wiring issue externally or internal defect causing the panel to be less productive then the others.

While it may be not easy I would swap the position of the panel in question with one of the other panels to see if the problem moves or stays in position 1.2.10.

[ButchDeal]

The 7 to 10watt difference seems to carry through out the day so it is possible there is a wiring issue externally or internal defect causing the panel to be less productive then the others.

While it may be not easy I would swap the position of the panel in question with one of the other panels to see if the problem moves or stays in position 1.2.10.

well I would look at the voltages etc before even thinking of moving a module for this. it is generally within 5 watts of the neighbors. Most modules come with a tolerance of -0, +5w ratting

The layout alone makes me suspect that there is an obstruction or shadow causing object to the south east of that module.
Other wise why was the bottom row shifted to the west?
why were to more modules not installed to the east on the bottom row?

[peakbagger]

FYI Modules are built and then sorted for output at the factory. Its not unusual to see three or four models of the same physical size panel that differ only in wattage. This is especially so with amorphous panels. The installer may have been short one panel of a particular wattage and substituted one with lower rated wattage. This may have been inadvertent as its easy to mix them up and the only way to know is look up the data sheet for each panel. The panel will look identical it just may have had a lower rating from the factory. Some folks have claimed that paying for the high wattage panel of a particulate model panel is probably a waste as they tend to average out in the long run but in theory the guarantee is tied to the original panel rating so the degradation should be the same percent decline for each panel.

[SunEagle]

well I would look at the voltages etc before even thinking of moving a module for this. it is generally within 5 watts of the neighbors. Most modules come with a tolerance of -0, +5w ratting

The layout alone makes me suspect that there is an obstruction or shadow causing object to the south east of that module.
Other wise why was the bottom row shifted to the west?
why were to more modules not installed to the east on the bottom row?

Good points. Either that panel is performing on the low side of it nameplate wattage (which could also be a mfg defect) or as I suspect there is a wiring issue that may be affecting it's performance.

I am not sure about a shade issue since the panel under performs all day but you could be right that there is some type of obstruction blocking part of it. But according to the OP there isn't any shade on the panel.

Could the lowest corner of that panel be too low to get the full sun light? Hard to determine without doing more hands on research of the array and installation.

[littleharbor]

Could the lowest corner of that panel be too low to get the full sun light? Hard to determine without doing more hands on research of the array and installation.

If so, the problem may get worse as the days get shorter and sun rides lower in the sky.

[sensij]

Looking at voltage and current for that panel, as Butch said, will offer more clues, but shade doesn't look likely. I have a couple panels with that much variation in my system too (pic later), at least some of which is explained by temperature variation due to differences in airflow on different parts of the roof. If that corner of the OP's roof is more sheltered from the breeze, or is less effective at rejecting heat for a variety of other reasons, the panel could be identical to the others and still underperform.

[bcroe]

The difference is pretty much in the range of mfg tolerance; I would note it and see if it gets
worse. One weak cell could be responsible; that might be revealed by taking non contact
thermal readings of the front of the panels after some time running at full power. Worked here.
Bruce Roe

[SunEagle]

If so, the problem may get worse as the days get shorter and sun rides lower in the sky.

You are correct. The highest difference of 10 watts from it neighboring panel was during the sweet spot of the day.

[sensij]

You are correct. The highest difference of 10 watts from it neighboring panel was during the sweet spot of the day.

Please be careful to not read more into the data than what is there. SolarEdge doesn't report the power produced by each panel instantaneously, it relies on random timing that can lead to differences of 10-20 min from one panel to the next. During the mid-morning and afternoon, this difference in timing can make it appear like one panel is over or under producing, but really, it is just reporting power at a slightly different time of day when more or less irradiance was present.

During the middle of the day, irradiance rate of change is lowest, and panel power should more or less converge. However, the middle of the day is when performance variation due to temperature differences is likely to be most visible.

Looking at daily or weekly totals can offer insight into performance too. Below is the energy produced in my system from the last week... none of the days was truly clear, there were morning clouds to varying degrees as is typical this time of year. In this view, I have a low of 81.33 kWh to a high of 88.45 kWh over the week, a difference of around 8%. However, I know that the panel that got 88 kWh is out on the end of the roof, gets great airflow, and doesn't even have an optimizer underneath of it to generate heat or impede airflow since I wired the string in a way that tried to minimize the overall physical length of the circuit (the panel next to it, 1.2.13, has two optimizers under it). I also have QC data from the pallet that would lead me to expect some performances differences from one panel to the next, but haven't bothered to correlate the panel serial numbers to the actual installed location.

To some extent, there might also be calibration differences in the optimizers. If you add up the optimizer energy, it never equals the energy reported by the inverter, or the energy I measure with a revenue grade meter on the inverter's output. Some of that is related to the inverter efficiency, but there isn't much reason to believe that the accuracy of the optimizers is better than a couple percent.

Again, the OP's system could be perfectly healthy, and even if one panel out of 10 truly is under-performing by 4%, the net impact on the system is only 0.4%... unless it gets worse, the cost of messing with it is probably greater than the value of any additional energy to be generated. production.JPG

[Mike90250]

Or, it could be poor calibration on a reporting module

[SunEagle]

Please be careful to not read more into the data than what is there. SolarEdge doesn't report the power produced by each panel instantaneously, it relies on random timing that can lead to differences of 10-20 min from one panel to the next. During the mid-morning and afternoon, this difference in timing can make it appear like one panel is over or under producing, but really, it is just reporting power at a slightly different time of day when more or less irradiance was present.

During the middle of the day, irradiance rate of change is lowest, and panel power should more or less converge. However, the middle of the day is when performance variation due to temperature differences is likely to be most visible.

Looking at daily or weekly totals can offer insight into performance too. Below is the energy produced in my system from the last week... none of the days was truly clear, there were morning clouds to varying degrees as is typical this time of year. In this view, I have a low of 81.33 kWh to a high of 88.45 kWh over the week, a difference of around 8%. However, I know that the panel that got 88 kWh is out on the end of the roof, gets great airflow, and doesn't even have an optimizer underneath of it to generate heat or impede airflow since I wired the string in a way that tried to minimize the overall physical length of the circuit the panel next to it, 1.2.13, has two optimizers under it). I also have QC data from the pallet that would lead me to expect some performances differences from one panel to the next, but haven't bothered to correlate the panel serial numbers to the actual installed location.

To some extent, there might also be calibration differences in the optimizers. If you add up the optimizer energy, it never equals the energy reported by the inverter, or the energy I measure with a revenue grade meter on the inverter's output. Some of that is related to the inverter efficiency, but there isn't much reason to believe that the accuracy of the optimizers is better than a couple percent.

Again, the OP's system could be perfectly healthy, and even if one panel out of 10 truly is under-performing by 4%, the net impact on the system is only 0.4%... unless it gets worse, the cost of messing with it is probably greater than the value of any additional energy to be generated.

Ah. I just learned something concerning SolarEdge data collection. Thanks for pointing that out.