Inverter only producing half power

[JerryA]

To follow up ... Technician came out again. I mentioned to him that the system seemed to have two modes. Producing normal power and sometimes producing half power. He told me he found a loose connection where the fuses were on the roof. I am thinking that one of the two strings was intermittent. Was never failing when the technicians were present. Time will tell if this fixed the problem.

[nwdiver]

Finally got the inverter replaced. Still under producing so I guess people here were right about the pannels.

How many panels and what type in a string? Which new inverter did you get?

A friends inverter began to suffer as the panels aged because the MPP band was 350-480. The voltage of the panels dipped a bit as they aged and on warmer days the inverter was significantly outside it's MPP band causing a ~20% drop in performance.

[JerryA]

Finally got the inverter replaced. Still under producing so I guess people here were right about the pannels.

[JerryA]

Thank you! These are great suggestions.

[jflorey2]

I did narrow it down. The problem is with the inverter. That was what I said to begin with. Whatever problem might happen in the panels it would result in different currents in the two parallel arrays. No one has suggested a single thing I might do or check on the inverter.

OK, so some other things to try -

Disconnect one string. If power falls by 50% (and strings are equal size) then it's likely a problem that's affecting the entire array, like fouling or catastrophic degradation. If it remains at the same power, you are likely seeing the inverter doing power limiting due to hitting a thermal or current limit (or more accurately THINKING it is hitting a thermal or current limit.)

Now disconnect the other. If one array causes no change when disconnected, but the other causes a huge change, then that's the problem - an issue on the inverter's terminal block, or in the wiring between the inverter and the array.

Monitor array voltage, then disconnect the AC from the grid and reconnect it. It will take around 2 minutes for the inverter to turn back on, depending on when it was made. If both array voltages droop a fair amount (say 10% or so) then the inverter is accurately finding the peak power point, and the problem is on the array side. If it remains at almost the same voltage (within a few percent) then it's not accurately tracking MPPT and it's the inverter.

BTW the Fronius only has one MPPT channel; the three string inputs are physically connected together internally. So it can't be that one of the "two units" failed.

BTW2: "If the array was bad then the voltage input to the inverter would be sagging under load. Didn't see that happening. " - Voltage sagging means everything is working right; the inverter is hunting for the MPP voltage, finding it and then operating there.

Mandatory warning - unless you are VERY experienced working with HVDC, do NOT try to mess with the PV side of things unless there's a technician/electrician there you to help you! HVDC is much more dangerous than AC, and will kill you very quickly unless you know exactly what you are doing.

[JerryA]

I did narrow it down. The problem is with the inverter. That was what I said to begin with. Whatever problem might happen in the panels it would result in different currents in the two parallel arrays. No one has suggested a single thing I might do or check on the inverter.

[jflorey2]

I see now I should have insisted it was a problem with the panels. Then you guys would have told me how to troubleshoot the inverter.

Well, the problem is one of the two. Start narrowing it down!

[JerryA]

I see now I should have insisted it was a problem with the panels. Then you guys would have told me how to troubleshoot the inverter.

[jflorey2]

I have a couple that do that. With 3 bypass diode sections, they show good open ckt voltage. But in a string,
under best sun the voltage drops to 2/3 or 1/3 of normal. It happens because one or 2 sections cannot
maintain full current under best sun. So the rest of the string forces them into bypass.

They have no problem with open circuit voltage, and a short circuit test demonstrates the good section(s)
while the weak sections go into bypass.

I have a couple of these, culled from 4 strings of no name Evil bay panels. Still trying to find a good use
for them, either run stand alone for some minor function, or set up a pair with weak sections shorted, to
replace one good panel. Bruce Roe

Interesting. Thanks for the info.

[bcroe]

Have you ever seen a panel give a good Voc and a good Isc - then collapse when a 'normal' load is placed on it? I never have. I can't even imagine how it would happen - a very high resistance connection somewhere? But that would give you a low Isc as well.

I have a couple that do that. With 3 bypass diode sections, they show good open ckt voltage. But in a string,
under best sun the voltage drops to 2/3 or 1/3 of normal. It happens because one or 2 sections cannot
maintain full current under best sun. So the rest of the string forces them into bypass.

They have no problem with open circuit voltage, and a short circuit test demonstrates the good section(s)
while the weak sections go into bypass.

I have a couple of these, culled from 4 strings of no name Evil bay panels. Still trying to find a good use
for them, either run stand alone for some minor function, or set up a pair with weak sections shorted, to
replace one good panel. Bruce Roe

[littleharbor]

The bad BP panels had a problem with the leads lacking in strain relief They could test ok Voc and Isc but once connected and warm in the sun they would start failing, some scorching from the back of the j-box . Some worse than others, some charred, some even so hot as to shatter the glass.

[jflorey2]

My bad on the wording. Thing is Isc. isn't going to give you a working panel's results. Just like open circuit voltage , in that you can get full Voc. from a bad panel, put a load on it and the voltage can collapse if it is a bad panel. I saw lots of this with the bad BP Solar products.

Have you ever seen a panel give a good Voc and a good Isc - then collapse when a 'normal' load is placed on it? I never have. I can't even imagine how it would happen - a very high resistance connection somewhere? But that would give you a low Isc as well.

[littleharbor]

Isc is tested under maximum load (short circuit.) For panel tests you just connect the two leads together and use a clamp on ammeter.

My bad on the wording. Thing is Isc. isn't going to give you a working panel's results. Just like open circuit voltage , in that you can get full Voc. from a bad panel, put a load on it and the voltage can collapse if it is a bad panel. I saw lots of this with the bad BP Solar products.

[Ampster]

The suspense is killing me. I hope we hear the final outcome.

I am happy I have Solaredge and Enphase on my installations because in both cases I can see panel level production and ought to he able to see if my Solaredge inverters are performing. One is 300 miles away but monitoring allows me to set alerts.

[jflorey2]

[LEFT][COLOR=#000000][FONT=Helvetica][SIZE=13px]The thing is there's no load in panels when checking either Voc. or Isc.

Isc is tested under maximum load (short circuit.) For panel tests you just connect the two leads together and use a clamp on ammeter.