Poor Module Performance

[posplayr]

This is exactly what happened with my array. It took my installer 2-3 days to get my SolarEdge monitoring account set up after the install. I noticed right away that one panel was underperforming by a significant amount--each panel did about 2.27-2.42kWh per day, but this one panel did about 0.87kWh. In the morning, the SolarEdge inverter would show 16/16 modules online, but once it got warmer, it would show 15/16 modules online. This went on for 3 days--I noticed this toward the end of a Friday so they weren't able to come out until the following Monday.

It turned out to be a faulty SE optimizer. The installer swapped it out. The system has been performing well ever since. The installer weren't sure what to do at first because this was the first time they've had a faulty SE optimizer (of course it would happen to me).

We both learned a great deal from this issue. I showed the office folks how to issue site API keys so that folks can do their monitoring on PVOutput. I'll post a thread to describe my install in detail once I get my Rainforest Eagle set up.

Kind of poor fault isolation if you have to remove a panel to rule out a panel to identify the optimizer as being at fault. Things do break, but the should be internal diagnostics in the optimizer to fault detect and isolate. If faulty optimizers are without fault detection and fault isolation (to the module) it is not designed very well and leave it at that.

[pure3d2]

Kind of poor fault isolation if you have to remove a panel to rule out a panel to identify the optimizer as being at fault. Things do break, but the should be internal diagnostics in the optimizer to fault detect and isolate. If faulty optimizers are without fault detection and fault isolation (to the module) it is not designed very well and leave it at that.

They didn't have to remove the panel. They just took the optimizer out and tested it to determine that it was faulty. My apologies for the confusion.

[posplayr]

They didn't have to remove the panel. They just took the optimizer out and tested it to determine that it was faulty. My apologies for the confusion.

With the OP SolareEdge suggested an open circuit test of the panel to isolate the panels performance from the array and more specifically its respective optimizer. In effect they are fault isolating their optimizer. In your case, they could also not fault isolate as well (as a built in capability) and apparently could only isolate to the optimizer by a standalone module checker. The really worst part of this is that the SolarEdge equipment doesn't seem to be able to fault detect relying on the homeowner to monitor and detect abnormalities.

In the diagnostics world, Fault detection is knowing in there is a problem to some certainty, and Fault Isolation is being able to to identify where the problem is. In an ideal, world I for one would expect the system to fault detect(and provide a warning) if there is a portion of the array operating low and fault isolate to whether it is a specific panel or optimizer.

[sensij]

With the OP SolareEdge suggested an open circuit test of the panel to isolate the panels performance from the array and more specifically its respective optimizer. In effect they are fault isolating their optimizer. In your case, they could also not fault isolate as well (as a built in capability) and apparently could only isolate to the optimizer by a standalone module checker. The really worst part of this is that the SolarEdge equipment doesn't seem to be able to fault detect relying on the homeowner to monitor and detect abnormalities.

In the diagnostics world, Fault detection is knowing in there is a problem to some certainty, and Fault Isolation is being able to to identify where the problem is. In an ideal, world I for one would expect the system to fault detect(and provide a warning) if there is a portion of the array operating low and fault isolate to whether it is a specific panel or optimizer.

No reason to throw stones at SolarEdge, no inverters have the features you are describing.

[posplayr]

No reason to throw stones at SolarEdge, no inverters have the features you are describing.

I suspect if no inverters have that feature it is owing to an immaturity of the market and not technical impossibilities. My observations are technical in nature and my opinions are personal yet based on experience in the field of diagnostics. I'm not sure I'm "throwing stones" as you suggest, just making observations none of which are incorrect. Considering this is the third Solaredge system I have seen posted about in the last 2 weeks with aberrant reporting it is not a completely irrelevant comment or topic.

[sensij]

Well, for what it is worth, i agree that the number of reporting errors cropping up is starting to become noticeable. I don't think I've actually seen a system yet that truly fails to generate power.. It seems to me that the problems have been limited to the feedback, making people think that a panel isn't performing (even though it is), with the only fix being to replace the optimizer.

I'd love to see data from an actual nonproducing system, if one is ever confirmed bad.

[posplayr]

Well, for what it is worth, i agree that the number of reporting errors cropping up is starting to become noticeable. I don't think I've actually seen a system yet that truly fails to generate power.. It seems to me that the problems have been limited to the feedback, making people think that a panel isn't performing (even though it is), with the only fix being to replace the optimizer.

I'd love to see data from an actual nonproducing system, if one is ever confirmed bad.

Ask pure3d2; he says he just had a tech come and confirm a bad module.

[sensij]

Ask pure3d2; he says he just had a tech come and confirm a bad module.

It sounds like his optimizer was replaced because it wasn't communicating. i have already seen two other cases where that condition can occur but the optimizer is still putting out full power. When the system switched from 16/16 reporting to something less, did the inverter output show a step drop? Is the inverter output higher now that the optimizer has been replaced? My guess, based on what I've seen recently, is no, but I'd love to be proven wrong.

[SunEagle]

It sounds like his optimizer was replaced because it wasn't communicating. i have already seen two other cases where that condition can occur but the optimizer is still putting out full power. When the system switched from 16/16 reporting to something less, did the inverter output show a step drop? Is the inverter output higher now that the optimizer has been replaced? My guess, based on what I've seen recently, is no, but I'd love to be proven wrong.

Normal steps taken by most "trouble shooters". Get as much info as possible to determine the true root cause. Too little data and too many assumptions leads to jumping to the wrong conclusion.

[ecoalex]

Loss Not As Severe

My loss isn't that severe, the 2 modules under perform about 100 watts in the morning, but recover about 2 Pm to normal production.

[ecoalex]

Improvement

Solaredge support emailed me yesterday they had changed settings in the #2 string inverter.

We have clouds this morning, in viewing the generation of string #2, I see the generation of the suspect modules #4, #17 are now equal to their neighbors.Was this due to Solaredge's inverter changes, or the clouds this morning? Time will tell.

[pure3d2]

It sounds like his optimizer was replaced because it wasn't communicating. i have already seen two other cases where that condition can occur but the optimizer is still putting out full power. When the system switched from 16/16 reporting to something less, did the inverter output show a step drop? Is the inverter output higher now that the optimizer has been replaced? My guess, based on what I've seen recently, is no, but I'd love to be proven wrong.

When it was reporting 15/16, it was both from the SE monitoring portal and from the inverter's LCD screen. During my peak production period, I was usually seeing 4.1-4.8kW of instantaneous power production (on the inverter LCD screen) when all 16 panels were producing (optimizers functioning). Once it began reporting 15/16, my production would hover around 3.7-4.5kW. It mostly stayed in the sub 4.0kW range.

After the optimizer was replaced, the inverter's LCD screen reported a higher output (back to the 4.1-4.8kW range and sometimes hitting 5.1kW).

What I don't know is, if I was still getting full power despite what the inverter LCD screen was reporting while that one optimizer was "down." I don't know if SE measures based on what the optimizers tell the inverter or if the inverter measures the actual power that it's delivering to the AC power panel. If it's the former, then we can't put much faith into the information being reported since that's assuming that the optimizers are always telling the truth.

sensij, I'm always delighted to read your posts, especially the ones in that HX_Guy's mega install nightmare thread.