One of 3 legs from the solar panels to the DC Disconnect has no voltage

[Adamk]

Hello, one of our inverters (sunpower SP-5200) that I haven't checked in a while is giving an error. The system is about 17 years old, I checked the manual and it is a low input voltage issue. I opened the DC Disconnect (SBCD 4-40) and started probing. There are 3 wires coming out of the solar panels (SPR-220) One of the wires has no voltage but the other 2 do. I have a second inverter in that system that has voltage on all 3 wires and is working fine.

What would be the next steps for troubleshooting? Does anyone recommend someone in Southern California that can come out and help diagnose and repair the issue at a reasonable cost?

End goal for that system is to disconnect from the grid add battery backup but will be posting soon on how to accomplish that

 

[FilterGuy]

Hello, one of our inverters (sunpower SP-5200) that I haven't checked in a while is giving an error. The system is about 17 years old, I checked the manual and it is a low input voltage issue. I opened the DC Disconnect (SBCD 4-40) and started probing. There are 3 wires coming out of the solar panels (SPR-220) One of the wires has no voltage but the other 2 do. I have a second inverter in that system that has voltage on all 3 wires and is working fine.

What would be the next steps for troubleshooting? Does anyone recommend someone in Southern California that can come out and help diagnose and repair the issue at a reasonable cost?

End goal for that system is to disconnect from the grid add battery backup but will be posting soon on how to accomplish that

Could you provide some kind of diagram of the system? I am struggling to understand what the '3 legs' are. Typically, there is a positive, negative, and ground.... and I would not expect to see voltage on the ground. (what are the colors of the wires?)

 

[FilterGuy]

BTW: Please be very careful around that system.

I have a second inverter in that system that has voltage on all 3 wires and is working fine.

If one of those wires is ground, it means there is a ground fault. The pv voltage can be quite high so a ground fault can be very dangerous.

 

[Adamk]

Could you provide some kind of diagram of the system? I am struggling to understand what the '3 legs' are. Typically, there is a positive, negative, and ground.... and I would not expect to see voltage on the ground. (what are the colors of the wires?)

Hello, I am attaching an image of the DC Disconnect panel diagram. There are 3 black wires coming in to the disconnect from the panels. The green lines indicate voltage and the red line is the one that has no voltage. The blue circle is what I was using for my neutral on my digital meter. The other DC Disconnect has voltage on all 3 legs. **Note: there is no 4th wire coming into the disconnect on either panel

 

[Adamk]

Hello, I am attaching an image of the DC Disconnect panel diagram. There are 3 black wires coming in to the disconnect from the panels. The green lines indicate voltage and the red line is the one that has no voltage. The blue circle is what I was using for my neutral on my digital meter. The other DC Disconnect has voltage on all 3 legs. **Note: there is no 4th wire coming into the disconnect on either panel

There was no schematic of the system provided by the installers. There are 2 5k inverters and I think 10k of panels on the roof so it's essentially 2 separate systems, 2 inverters, 2 disconnects, 2 breakers.

 

[FilterGuy]

OK. I understand now. The disconnect is also a combiner box. Now your description makes more sense to me.

Side note: Using the term 'PV Neutral' is interesting. It makes me wonder if the Sunnyboy system grounds the negative of the PV. (That used to be a lot more common than it is now).

Could you please tell us what voltage you are reading between PV Neutral and each of the PV Positive inputs?

 

[FilterGuy]

OK. I understand now. The disconnect is also a combiner box. Now your description makes more sense to me.

Side note: Using the term 'PV Neutral' is interesting. It makes me wonder if the Sunnyboy system grounds the negative of the PV. (That used to be a lot more common than it is now).

Could you please tell us what voltage you are reading between PV Neutral and each of the PV Positive inputs?

Please do the measurement on both systems so we can compare between the one not working and the one that is working.

 

[Adamk]

OK. I understand now. The disconnect is also a combiner box. Now your description makes more sense to me.

Side note: Using the term 'PV Neutral' is interesting. It makes me wonder if the Sunnyboy system grounds the negative of the PV. (That used to be a lot more common than it is now).

Could you please tell us what voltage you are reading between PV Neutral and each of the PV Positive inputs?

on the working one I measured a little over 300V on each input. The non working unit had the same voltage on the 2 working inputs and 0 on the third one

 

[FilterGuy]

on the working one I measured a little over 300V on each input. The non working unit had the same voltage on the 2 working inputs and 0 on the third one

a low input voltage issue.

Hmmmm,

1) Check the fuse on the line showing 0V. If the voltage is zero at the input of one input and not the others, there must be a disconnection somewhere. The most likely place for the disconnect is the fuse.
NOTE: If the fuse is blown, there is a good chance it will immediately blow when you replace it. If it does blow, there is a very good chance there is a short someplace on the bad string.

2. Please measure the voltage between the PV Neutral and PV Positive output. This will help us understand if the low-voltage warning is a problem in the inverter-disconnect or after the inverter-disconnect.

 

[Adamk]

Hmmmm,

1) Check the fuse on the line showing 0V. If the voltage is zero at the input of one input and not the others, there must be a disconnection somewhere. The most likely place for the disconnect is the fuse.
NOTE: If the fuse is blown, there is a good chance it will immediately blow when you replace it. If it does blow, there is a very good chance there is a short someplace on the bad string.

2. Please measure the voltage between the PV Neutral and PV Positive output. This will help us understand if the low-voltage warning is a problem in the inverter-disconnect or after the inverter-disconnect.

I tried to look yesterday for a fuse but did not see one in the building. The input wires are coming from outside where the panels are through a conduit. I'll check again and see if I can trace the conduit back to another box outside on the roof.

Assuming I do find a fuse, what is the procedure to replace it since there is active voltage on the line? Is there a special tool that prevents shock?

Yesterday I flipped the disconnect on and measured the voltage at the PV Hot Output and there was zero volts there as well. The inverter is also reporting very low input voltage between 0-16V

Thank you

 

[FilterGuy]

I tried to look yesterday for a fuse but did not see one in the building.

I looked for a manual but could not find one.

My guess is the input lugs are the fuse holders.



The holder usually will rotate out to both disconnect and expose the fuse:




A less common style is a pull-out cartridge

Yesterday I flipped the disconnect on and measured the voltage at the PV Hot Output and there was zero volts there as well.

OK.... the most likely situation is that you have either two or three blown fuses. Either the two that show voltage or all 3. My guess is that it is the two that are still showing voltage.

The question then becomes: Why did they blow? That is hard to say. The obvious culprit is the string with no voltage.... but it could be a few different things on that string.

• A +to- short in one of the panels. This seems unlikely. A short in the panel would present as a lower voltage on the string, not zero voltage.
• An open circuit someplace along the string. This is a possibility, but it does not explain why the other fuses blew out.
• A short from Positive to Ground. This seems like a reasonable possibility.... but still does not completely describe the symptoms. I would expect a short from positive to ground to blow the fuse on that string but not the other fuses. However, if the fuses are not properly sized, it could explain things.

 

[FilterGuy]

Question: Do you have the spec sheet for the panels? The fuses should be 1.55x the Isc of the panel.

If you find the fuses are blown, leave the fuse out of the string with 0 volts and replace the other two fuses. My guess is the inverter will start working.... but at a reduced power output.

Assuming I am correct, it would prove there is a problem with the string with zero voltage.

 

[Adamk]

Question: Do you have the spec sheet for the panels? The fuses should be 1.55x the Isc of the panel.

If you find the fuses are blown, leave the fuse out of the string with 0 volts and replace the other two fuses. My guess is the inverter will start working.... but at a reduced power output.

Assuming I am correct, it would prove there is a problem with the string with zero voltage.

Thank you and yes the lugs do look like that, I will check the fuses tomorrow but the logic doesn't seem right to me. Since these 3 wires are coming directly from the panels and I am checking voltage before the fuses wouldn't I still see voltage on those terminals even if the fuses were blown?

I had shut off the DC and AC on that system about a week ago and I unfortunately didn't check if they were working before shutting them off. Maybe the fuse blew when I turned them back on yesterday? Considering they are 17 years old they might have been weak and with the sudden surge they blew?

The panels are sunpower spr-220

ISC is 5.95 Amps does this mean I need 5.95 x1.55 in total or is it 5.95 x # of panels on the string x 1.55?

 

[FilterGuy]

Thank you and yes the lugs do look like that, I will check the fuses tomorrow but the logic doesn't seem right to me. Since these 3 wires are coming directly from the panels and I am checking voltage before the fuses wouldn't I still see voltage on those terminals even if the fuses were blown?

Yes, the voltage on the input to the fuse would still be visible. However, post #10 indicated the output had no voltage. That indicates there is an 'open' someplace between the input and output.... the most logical place for the 'open' is a blown fuse.

I had shut off the DC and AC on that system about a week ago and I unfortunately didn't check if they were working before shutting them off. Maybe the fuse blew when I turned them back on yesterday? Considering they are 17 years old they might have been weak and with the sudden surge they blew?

Anything is possible, but I can't think of a mechanism that would cause what you are seeing. If the unit had the error before you shut it off, it seems very unlikely switching it on/off caused the problem.

The panels are sunpower spr-220

ISC is 5.95 Amps does this mean I need 5.95 x1.55 in total or is it 5.95 x # of panels on the string x 1.55?

On a single string, the current remains constant (Voltage adds). Consequently, the fuse should be 5.95 x 1.55 = 9.22. Round that up to 10A.

If you want to learn more about PV fusing and where the 1.55x comes from, you might want to review this:

Fusing & Wire Sizing guidelines for solar panels

To get the paper, click on the orange'' button at the top of the screen. This resource is intended to help the user understand when they need to add Over Current Protection Devices (OCPDs - Fuses or Breakers) to the solar panel array and what...

diysolarforum.com

 

[Adamk]

Yes, the voltage on the input to the fuse would still be visible. However, post #10 indicated the output had no voltage. That indicates there is an 'open' someplace between the input and output.... the most logical place for the 'open' is a blown fuse.

Anything is possible, but I can't think of a mechanism that would cause what you are seeing. If the unit had the error before you shut it off, it seems very unlikely switching it on/off caused the problem.

On a single string, the current remains constant (Voltage adds). Consequently, the fuse should be 5.95 x 1.55 = 9.22. Round that up to 10A.

If you want to learn more about PV fusing and where the 1.55x comes from, you might want to review this:

Fusing & Wire Sizing guidelines for solar panels

To get the paper, click on the orange'' button at the top of the screen. This resource is intended to help the user understand when they need to add Over Current Protection Devices (OCPDs - Fuses or Breakers) to the solar panel array and what...

diysolarforum.com

Thank you, so in regards to the fuses, if there is 300V on leg 1 and 2 and zero on leg 3 on the input side, we should at least see 300v on the output when the Disconnect is flipped on assuming those fuses are still good right?

If there is a short or open on leg 3 before it comes into the disconnect and I install new fuses, are there any concerns of further damage by flipping the disconnect back on to see if the fuses were the problem?

Could a temporary workaround be to leave the fuse out of the input with 0 voltage and run the system with the 2 working legs only?

 

[FilterGuy]

I just re-read an earlier post and I think I misunderstood the measurements that were taken.

This is a diagram of what is there (I think)

The blue circle is what I was using for my neutral on my digital meter.

I had thought the measurement was from the ground.... so I was misinterpreting what is happening.

The measurments were:
N-1 = ~300V
N-2 = ~300V
N-3 = ~0V

You are correct. The voltage would show regardless of the state of the fuse.

That tells us string 1 and 2 are OK but string 3 has a problem. This tells us nothing about the fuses

 

[Adamk]

that diagram is correct.

 

[FilterGuy]

Yesterday I flipped the disconnect on and measured the voltage at the PV Hot Output and there was zero volts there as well. The inverter is also reporting very low input voltage between 0-16V

If I understand, the measurement was from Neutral to Hot and it came out as zero.



THis tells us that either fuse 1 & 2 are blown or string 3 is completely shorted and pulling the output to zero.

Once we find out the state of the fuses we will know which it is.

 

[Adamk]

Ok here are the results using your diagram above

With disconnect off
input of fuse leg 1 = 364V - output of fuse leg 1 = 364V - 0 Ohms across fuse
input of fuse leg 2 = 364V - output of fuse leg 2 = 364V - 0 Ohms across fuse
input of fuse leg 3 = 0V - output of fuse leg 3 = 0V - 0 Ohms across fuse
0 Volts across H and N on the output

With Disconnect on
input of fuse leg 1 = 364V - output of fuse leg 1 = 364V - 0 Ohms across fuse
input of fuse leg 2 = 364V - output of fuse leg 2 = 364V - 0 Ohms across fuse
input of fuse leg 3 = 0V - output of fuse leg 3 = 0V - 0 Ohms across fuse
-11 Volts across H and N on the output

 

[FilterGuy]

Ok here are the results using your diagram above

With disconnect off
input of fuse leg 1 = 364V - output of fuse leg 1 = 364V - 0 Ohms across fuse
input of fuse leg 2 = 364V - output of fuse leg 2 = 364V - 0 Ohms across fuse
input of fuse leg 3 = 0V - output of fuse leg 3 = 0V - 0 Ohms across fuse
0 Volts across H and N on the output

With Disconnect on
input of fuse leg 1 = 364V - output of fuse leg 1 = 364V - 0 Ohms across fuse
input of fuse leg 2 = 364V - output of fuse leg 2 = 364V - 0 Ohms across fuse
input of fuse leg 3 = 0V - output of fuse leg 3 = 0V - 0 Ohms across fuse
-11 Volts across H and N on the output

How strange!



That doesn't even seem possible!! I am clearly missing something important.

With the disconnect on I would expect the voltage at 'H' to be the same as the voltage at the fuse output. If the voltage at H was zero, I would just assume there is a problem with the disconnect switch, but to have it go negative just doesn't make sense. Now you have my curiosity up... I love a good mystery.

Try disconnecting the inverter from 'H' and redo the measurements. It will help us narrow down where things are getting wonky.

Also, did you verify the inputs go to fuse holders? What happens if you remove fuse 3?

 

[Adamk]

Took a few more measurements and now i'm confused

Sorry, I didn't pay attention the first time but the voltage at leg 1 and 2 is -360V I confirmed the negative voltage on the working inverters disconnect as well. I also confirmed it wasn't the meter reading wrong by checking it against a power supply I have. So maybe they wired it in backwards? Or maybe it's supposed to have a negative voltage?

The readout on the working inverter shows this.

PV input 352V
Total output -19431 kW-Hr

I think I also forgot to measure the voltage at the fuse with the disconnect powered on last time.

Here are the probe results
with fuse 3 and 4 removed and disconnect switched off I'm getting -360 at both ends of the 1 and 2 fuses. With it switched on I'm getting -11v across N and H also at the fuses 1 and 2

with fuses 1,3,4 removed and disconnect on, I am seeing the same as above -11v everywhere after the fuse and -11v at the input of fuse 2
with fuses 2,3,4 removed and disconnect on, I am seeing the same as above -11v everywhere after the fuse and -11v at the input of fuse 1

I am attaching a picture of the disconnect if it helps

 

[fromport]

Sorry, I didn't pay attention the first time but the voltage at leg 1 and 2 is -360V I confirmed the negative voltage on the working inverters disconnect as well. I also confirmed it wasn't the meter reading wrong by checking it against a power supply I have. So maybe they wired it in backwards? Or maybe it's supposed to have a negative voltage?

As you can see they put red tape on the 3 wires indicating those are the PLUS leads, connected to the "PV NEUTRAL INPUT" block from the schematic you posted before.
Just reverse the red/black in the diagram between inverter & disconnect/combiner of @FilterGuy
They fused & switched the negative wires.
Not usual , but the ideas behind is still the same.

Here are the probe results
with fuse 3 and 4 removed and disconnect switched off I'm getting -360 at both ends of the 1 and 2 fuses. With it switched on I'm getting -11v across N and H also at the fuses 1 and 2

My bet is on a bad inverter.
Pulling all the power to almost 0 volt.
Do you happen to have a clamp on amp meter by any chance ?

 

[Adamk]

As you can see they put red tape on the 3 wires indicating those are the PLUS leads, connected to the "PV NEUTRAL INPUT" block from the schematic you posted before.
Just reverse the red/black in the diagram between inverter & disconnect/combiner of @FilterGuy
They fused & switched the negative wires.
Not usual , but the ideas behind is still the same.


My bet is on a bad inverter.
Pulling all the power to almost 0 volt.
Do you happen to have a clamp on amp meter by any chance ?

yes I have one. What should I test?

Also, not sure if it matters but AC was disconnected from the inverter while I was performing the tests.

 

[fromport]

yes I have one. What should I test?

Also, not sure if it matters but AC was disconnected from the inverter while I was performing the tests.

how much current is running through the 2 wires that do show voltage with the disconnect open, after you close the disconnect (switch it to on).

The 3rd string not showing any voltage is complete different issue, you probably have had only 2/3 of the PV power for a while.
Currently I would be mainly interested to see why the inverter complaints about low input.

 

[FilterGuy]

As you can see they put red tape on the 3 wires indicating those are the PLUS leads, connected to the "PV NEUTRAL INPUT" block from the schematic you posted before.

Good Catch.

In some cases, they ground the positive side of a DC circuit. This was more common for PV circuits in the past when most of the SCCs were fully isolated from the AC. The other interesting point is that in the NEC, white is reserved for the grounded conductor (Neutral). I would not be at all surprised if the voltage between earth-ground and the 'neutral' was near zero on that circuit.