Broken panels or broken inverter? Help!

Hi folks,

Yesterday I noticed my production was 30% down from the usual. I have 3 strings of panels (3x330w each). Each string has the 3 panels in series, to get to around 110V, and then they all connect in parallel to my Axpert Inverter.

After some quick diagnostics, I found out one of the strings was not giving any power to the inverter. (I realized this by disconnecting one by one, and noticing that this one did not make any difference being unplugged or not).

I spent most of my day trying to check the cables (some of them underground), and making sure there was no connectivity issue. I tested connectivity with my multimeter and everything seemed to be ok (I did not want to dig the underground cables...)

I checked the voltage of this particular string, and it was ok at around 110V.

Today I tried something else, and this is when things begin to get weird.

I disconnected completely the 2 good strings. I measured the voltage of the bad string before connecting to fuse box+inverter. 110V as expected.

Once I plug things on, the voltage drops to about 50-60V, and the inverter is not able to "connect" the charge (it does this "click" sounds every couple of seconds).

I'm lost trying to understand what this might be. Please let me know if you have any ideas. I really do miss those precious 30% now in winter time :(

It sounds like once it's underload, one of those panels stop working.
So the next step world be to test each solar panel with a load to see which one of them fail.
Do you have any 24Vdc to 30Vdc load around the house?

The fact that the inverter worked with other solar panels strings make me think that the problem is not with in the inverter.  

S Bengi wrote:It sounds like once it's underload, one of those panels stop working.
So the next step world be to test each solar panel with a load to see which one of them fail.

Not sure how I can test each panel individually, because it wont give enough voltage for the inverter to start the charge.

Do you have any 24Vdc to 30Vdc load around the house?

Everything everything is 220v. Not sure what you meant

Do you have an electrical multimeter?  As in voltmeter and especially amp meter?  In this case a clamp on meter would be ideal.  

What I am hypothesizing is that one panel is producing sufficient voltage but for some reason insufficient amps.  A clamp on amp meter could answer this hypothesis quickly.

Just a thought,

Eric

Do you have an incandescent lightbulb we can probably use that as a load

Eric Hanson wrote:Do you have an electrical multimeter?  As in voltmeter and especially amp meter?  In this case a clamp on meter would be ideal.  

What I am hypothesizing is that one panel is producing sufficient voltage but for some reason insufficient amps.  A clamp on amp meter could answer this hypothesis quickly.

Just a thought,

Eric

I have a clamp meter but it says AC all the time and I cant seem to make it switch to DC. Any other way to test the amps?

Is there a way to unplug the panel and test for overall wattage?  Maybe run some lights directly from the panel to get an idea if it is producing full amperage?  My thinking is that it could be producing adequate voltage but tiny amperage and thus low wattage.  Maybe this would work?

Also, I like S Bengi’s idea.

Eric Hanson wrote:Is there a way to unplug the panel and test for overall wattage?  Maybe run some lights directly from the panel to get an idea if it is producing full amperage?  My thinking is that it could be producing adequate voltage but tiny amperage and thus low wattage.  Maybe this would work?

These are 48V panels... I dont think I have anything that I can plug to it directly... :/

Can you wire up an incandescent light bulb?  You could check it on the good panels and then see if the bad panel fails to produce and light at all.

Nuno Donato wrote:
These are 48V panels... I dont think I have anything that I can plug to it directly... :/

330W solar panels are usually only 30V once you connect them to anything. They are usually use to charge 24V batteries.
Fully charged batteries are usually more than just 24V, and the solar panel needs to be a few volts above that.
330W/11A = 30V

The light bulb in your fridge might work, esp if you have a spare one around the house. Do you have an aquarium heater, that would work too.

I
It sounds like a textbook case of a bad panel. It might also be an arcing connection at an mc4 connection which you get sometimes when the installer makes his own homerun cable ends. If it is a bad connection with the breaker off disconnect each connection in the string and reconnect. If it's a bad panel, on a string of 3 panels you could probably test for the fault if it's a 24 v system, maybe on  48 volt as well. You will need a 2 lengths of mc4 cable to do it though or be able to move a panel closer. With the breaker off, Simply connect a string of 2 panels at a time leaving one out and see if the controller produces power from it. Alternate which one you leave out until you find the culprit. You can also short a single panels mc4 connections to itself but you would need a clamp on DC ammeter for that...
Trouble shooting a bad string sucks. Good luck
Cheers, David

Thanks everyone for the suggestions. This is a 48V system with 48V batteries.

Since my multimeter clamp is AC only I'm out of luck testing that.
But I did run my own cables to the MC4, so I'm going to check those out first.

While a bit riskier, you can try adding one of the suspicious solar panels to a good string.

1. Disconnect all strings from the inverter
2. Connect only 1 of the know good string to the inverter
3. Next add one of the suspicious solar panels to the working good string
4. Check to see if power production goes up of it stop production
5. If power production goes up, repeat the step with another one of the suspicious solar panel until you find the one that is broken.

Update!
With only the bad string connected, I took out one of the panels and I think I got lucky right on the first one! I tried to connect the string with 2 panels to the inverter and it began charging! So i guess I found out the culprit.

Is it worth to open the panel's "box" and look into some possible issues I can fix? Or is it good to replace? (its only 2 years old :( )

Does this confirm the problem is the string and not the panel?

Eric Hanson wrote:Does this confirm the problem is the string and not the panel?

Well, for me the problem is the panel, once I took it out, the rest seems to be working

since I got down from 9 to 8 panels, I wonder if its worth to reconfigure my setup
from: (3x300) + (3x+300) + (3x300)
to: (2x300) + (2x300) + (2x300) + (2x300)

I need to get new cables for that and re-do some wiring... is it worth it?

Nuno Donato wrote:Update!
With only the bad string connected, I took out one of the panels and I think I got lucky right on the first one! I tried to connect the string with 2 panels to the inverter and it began charging! So i guess I found out the culprit.

Is it worth to open the panel's "box" and look into some possible issues I can fix? Or is it good to replace? (its only 2 years old :( )

it could still be the string or the panel at this point... Try rehooking the panel and see if it's still charges. If it does it's the string if it does not more likely the panel. Also check the factory mc4 on the bad panel...