My Solaredge system dead. How to figure if it's an optimizer or inverter that's bad?

[sensij]

But, the 8 AWG has no neutral wire. It's just 2+1. Again, why does the inverter have a neutral at all? Is it because the internal electronics that draw the piddly 2.5watts run on 120V and thus use one hot leg and the neutral? And, if that is the reason, would it not be acceptable to have the two hot legs be 8 AWG and the neutral be 10 AWG?

705.95(B) is the applicable section of code (2014 NEC).

(B) Neutral Conductor for Instrumentation, Voltage
Detection or Phase Detection. A conductor used solely for
instrumentation, voltage detection, or phase detection and
connected to a single-phase or 3-phase utility-interactive
inverter, shall be permitted to be sized at less than the
ampacity of the other current-carrying conductors and shall
be sized equal to or larger than the equipment grounding
conductor.

For EGC sizing:

EGC.JPG

[bcroe]

My inverters have a neutral so that the voltage from neutral to either hot lead can be monitored; no
significant current flows in this lead. Check code, you may allowed to use smaller wire there. Bruce Roe

[kny]

Just because the responders on this thread have been so helpful and informative to this point, I'll take it one step further.

Why does the SE-6000 inverter use a neutral wire?

And, here's where I'm going with that. I may have an easy solution to my undersized wiring issue. Directly on the other side of the wall of the inverter is a junction box with 8 AWG wiring for an a/c blower. Inverter on exterior of wall, junction box for blower on interior of wall, virtually same location. The blower presumably has 8 AWG wiring because you could add a heating element, which I do not have and will never want. Right now I've got 8 AWG wiring for a blower and condensate pump that together never draw more than 300 watts and could easily be served by 14 AWG let alone 10 AWG. It would be little work to swap the blower and solar wiring and replace the solar 30 amp breaker with a 40 amp breaker. Wham bam, wiring/breaker problem solved. With the new P400 optimizers installed the grounding problem will be solved. If the inverter is toast (still unknown?) the inverter is replaced, and all my system problems are solved. I actually have a proper, fully compliant system, and I could even add some panels to the strings in the future, upgrade to a SE-7600 and still be good on the wiring.

But, the 8 AWG has no neutral wire. It's just 2+1. Again, why does the inverter have a neutral at all? Is it because the internal electronics that draw the piddly 2.5watts run on 120V and thus use one hot leg and the neutral? And, if that is the reason, would it not be acceptable to have the two hot legs be 8 AWG and the neutral be 10 AWG?

[sensij]

Green is module voltage. Red is optimizer voltage. System gave up the ghost in ballpark of 1:30-1:45pm, at which point the inverter shows 6.68kw power.

I have metal seam roof and S-5 clamps. No roof penetrations, thankfully.

Thanks for posting the data! At ~30 V on the optimizer, I guess that is from the 12 panel string. It looks to me like the optimizer has the panel voltage pegged against an upper limit in the low 50's... normal mppt behavior would show something like a U-shape to the curve as the panel warms up during the day and the Vmp drops in response to the shift in temperature. The jump after the error may be the panel's open circuit (or at least, high resistance) voltage... mine did something similar when it had a problem. See below (my colors are opposite yours).

Day in which I crashed the string by incorrectly shutting it down for service: Failure.JPG




Healthy behavior around that same time (I have a 12 panel string, too). The mess at the end of the day is how it reacts to shade from a nearby palm tree. Note the dip in the panel voltage mid-day. healthy.JPG

[FFE]

Kny, Next time you look at the display on your inverter you might want to look for the following when you scroll through the screens.
PWR CTRL: ******

If you have that screen you might want to read what that means in the inverter manual.

[kny]

OP, would you be willing to post the optimizer data Butch is asking about? I'd love to learn how they behave in a completely mismatched installation. It is hard to say whether this or the AC terminations, or some combination is responsible for your problems, but hopefully a fresh start will get you the system you want. They should be able to fix the physical string vs logical string problem in the portal when the new optimizers go in, too.

Do you have attic access? With the mistakes so far, it wouldn't be surprising if not every lag bolt hit its rafter true. Might be worth a spot check of a few, from the inside.

Green is module voltage. Red is optimizer voltage. System gave up the ghost in ballpark of 1:30-1:45pm, at which point the inverter shows 6.68kw power.

I have metal seam roof and S-5 clamps. No roof penetrations, thankfully.

[sensij]

When you look at the solaredge monitoring you can graph the pv module voltage and see where they were running, I would imagine pretty close to max optimizer voltage of 48v, the optimizers might have been clipping as well ( not sure there as this is way outside solaredge allowed installs)

OP, would you be willing to post the optimizer data Butch is asking about? I'd love to learn how they behave in a completely mismatched installation. It is hard to say whether this or the AC terminations, or some combination is responsible for your problems, but hopefully a fresh start will get you the system you want. They should be able to fix the physical string vs logical string problem in the portal when the new optimizers go in, too.

Do you have attic access? With the mistakes so far, it wouldn't be surprising if not every lag bolt hit its rafter true. Might be worth a spot check of a few, from the inside.

[kny]

Yes that is p300 which is very wrong further the bolt in the photo does not look like the ironridge micro inverter kit I linked to.
You should have the p370 optimally, or p400 will work. The p300 and p320 have max operating voltage of 48v the E327 has VMPP of 54.7V.
When you look at the solaredge monitoring you can graph the pv module voltage and see where they were running, I would imagine pretty close to max optimizer voltage of 48v, the optimizers might have been clipping as well ( not sure there as this is way outside solaredge allowed installs)
Have you checked that you got th E327s too

equipment changes are sometimes necessary but should be like or better equipment and sertainly appropriate equipment

The module voltage was in the low 50s until everything died, then shot up to upper 50s as high as 62 for the rest of the day.

[kny]

I do have the correct E20-327 panels. The racking rails are Unirac. Definitely no star washer, grounding plate, or grounding wire/lug present. Good news is that as they now must come and replace all optimizers I can ensure they ground the system.

I spoke with the installer minutes ago. I did a lot of talking and he did a lot of listening. I'm trying to believe "innocent mistake" but between the wiring/breaker, grounding, and improper equipment I am having increasing difficulty giving them the benefit of the doubt.

[ButchDeal]

Can I assume SLD-P300 PO model numbers means P-300 optimizers? Because that's what I've got despite an invoice showing P400s.

I'm now going from mildly pissed off to severely pissed off. I've got Sunpower E-327 panels and they should not be mated with P300s.

Yes that is p300 which is very wrong further the bolt in the photo does not look like the ironridge micro inverter kit I linked to.
You should have the p370 optimally, or p400 will work. The p300 and p320 have max operating voltage of 48v the E327 has VMPP of 54.7V.
When you look at the solaredge monitoring you can graph the pv module voltage and see where they were running, I would imagine pretty close to max optimizer voltage of 48v, the optimizers might have been clipping as well ( not sure there as this is way outside solaredge allowed installs)
Have you checked that you got th E327s too

equipment changes are sometimes necessary but should be like or better equipment and sertainly appropriate equipment

[ButchDeal]

The MI-BHW microinverter grounding kit, which is the kit you linked, does not include a washer with sharp points. The bolt head has small ridges to grab the optimizer or micro-inverter, and the T has ridges that would grab the slot in the rail. I wasn't sure if that alone would be enough to provide ground, so I sandwiched the star washer that SolarEdge provides as well, as shown in SolarEdge's application note. It would be nice if SolarEdge would update their note to include the XR10/XR100/XR1000 rails and the newest mounting hardware, which would really confirm that the star washer is not needed for that combination.

I'll check for continuity on my black rails without the star washer, relying on just the features of the MI-BHW kit. Basically, if what you are saying is right, if the OP's installer used the right hardware, the stars may not even be needed and the installation may be OK in this respect.

Agreed it is vague and the document said it works with Enphase for grounding but not specifically solaredge.

[kny]

Can I assume SLD-P300 PO model numbers means P-300 optimizers? Because that's what I've got despite an invoice showing P400s.

I'm now going from mildly pissed off to severely pissed off. I've got Sunpower E-327 panels and they should not be mated with P300s.

[sensij]

You should get the washers from the racking company. They are part of the rail grounding system. Iron ridge has kits for micro inverter installation that comes with two in each kit.
if the rails are ironridge and they used the ironridge micro inverter mounting kits then the star washers are not needed as the newer kits have a washer with sharp points to do the same thing.
http://files.ironridge.com/pitched-r...ion_Manual.pdf

BTW micro inverters need to be grounded the same as the optimizers.

The MI-BHW microinverter grounding kit, which is the kit you linked, does not include a washer with sharp points. The bolt head has small ridges to grab the optimizer or micro-inverter, and the T has ridges that would grab the slot in the rail. I wasn't sure if that alone would be enough to provide ground, so I sandwiched the star washer that SolarEdge provides as well, as shown in SolarEdge's application note. It would be nice if SolarEdge would update their note to include the XR10/XR100/XR1000 rails and the newest mounting hardware, which would really confirm that the star washer is not needed for that combination.

I'll check for continuity on my black rails without the star washer, relying on just the features of the MI-BHW kit. Basically, if what you are saying is right, if the OP's installer used the right hardware, the stars may not even be needed and the installation may be OK in this respect.

[ButchDeal]

If the star washer is left out, there probably is still enough contact between the optimizer and rail to "ground" it, but it may just not be good enough contact in all environments to meet UL, driving the need for the washer.

that depends on the rails and corrosion. Black rails would be unlikely to have any ground without the washer.

[ButchDeal]

The lack of grounding of the optimizers, on the other hand, has me legitimately pissed off. There is no excuse for it. Pages 12/13 of the PO install guide make very clear the two means of grounding the POs, and neither were done. That is unacceptable. This did not save the install company time or money. The star washers come with the POs; they simply didn't know what they were for and didn't use them and threw them away.

When the inevitable happens and the install company comes out to swap out a new inverter and/or optimizer(s) I want to have 22 star washers to hand to them and demand they install. Does anyone know whether any old star washer with teeth that I can pick up at HD or Lowes will do? Is there anything special to these piercing star washers that come with SolarEdge power optimizers?

You should get the washers from the racking company. They are part of the rail grounding system. Iron ridge has kits for micro inverter installation that comes with two in each kit.
if the rails are ironridge and they used the ironridge micro inverter mounting kits then the star washers are not needed as the newer kits have a washer with sharp points to do the same thing.

http://files.ironridge.com/pitched-roof-mounting/resources/brochures/IronRidge_Flush_Mount_Installation_Manual.pdf

BTW micro inverters need to be grounded the same as the optimizers.