EG4 3000EHV-48V Solar Suddenly not Being Recognized

[coppergenie]

Hello!

I have a 3000EHV-48V AIO inverter that's been connected to a series solar array with disconnect switch, all working properly for months. Today suddenly the AIO was not recognizing any power from the solar array, but still works with battery.

There is a warning on the inside cover of the AIO that says not to connect PV power while the AIO is running, which I am sure I may have done a few times because this label is on the inside cover, and I forgot about it. What effect might this have?

Notes:
- I have cycled the AIO and battery
- No grid power connected
- LifePower4 48V 100Ah battery powers the AIO fine
- Typical maximum input is: 180 V, 5 A, 900 W
- Panel specs: 4 in series, P_nom = 345 W, I_sc = 9.6 A, V_oc = 46.6 V
- No open circuits detected along the PV series loop (only checked visually)
- Screw terminals are tight for PV input on AIO
- No faults or error codes show up on the AIO display
- No new odors or sounds near the AIO
- Support request sent to EG4, still waiting

Any ideas to what may be wrong and how I can troubleshoot? I have a prong and clamp multimeter.

 

[coppergenie]

I am especially desperate because it happens the day before a 3-day holiday weekend. So I'll be powerless (lol) and not able to use most businesses including tech support til Monday. So any insight on this is very welcome.

 

[BarracudaBob]

Take voltage readings with a volt meter. YOu have a short to ground, open, high resistance or some other electrical related issue.

You need to be aboer 120VDC and under 420VDC for it to work properly

 

[Mattb4]

Hello!

I have a 3000EHV-48V AIO inverter that's been connected to a series solar array with disconnect switch, all working properly for months. Today suddenly the AIO was not recognizing any power from the solar array, but still works with battery.

There is a warning on the inside cover of the AIO that says not to connect PV power while the AIO is running, which I am sure I may have done a few times because this label is on the inside cover, and I forgot about it. What effect might this have?

Notes:
- I have cycled the AIO and battery
- No grid power connected
- LifePower4 48V 100Ah battery powers the AIO fine
- Typical maximum input is: 180 V, 5 A, 900 W
- Panel specs: 4 in series, P_nom = 345 W, I_sc = 9.6 A, V_oc = 46.6 V
- No open circuits detected along the PV series loop (only checked visually)
- Screw terminals are tight for PV input on AIO
- No faults or error codes show up on the AIO display
- No new odors or sounds near the AIO
- Support request sent to EG4, still waiting

Any ideas to what may be wrong and how I can troubleshoot? I have a prong and clamp multimeter.

Considering that every day PV disconnects after dark and reconnects with daylight this warning label is likely not worded correctly. It is not sufficient to visually inspect for opens. I would start doing what Barracuda Bob suggest and take voltage readings. If you have good voltage on PV wires when AIO is not connected, but it drops to zero once you connect, that is a sign of a bad connection or panel. Check all MC4's and especially check any inline MC4 fuse holders.

 

[K7TH]

I have the same inverter, and I have routinely connected my solar at 207v when the unit is running. However, IMHO that warning is for YOUR fingers and maybe causing sparks. But I have seen similar issues with my mid nite 150, and every time something like NO PV, Resting, or ARC fault was due to bad connections on the PV, combiner, MCBs in the line. Don't forget to check for good connections on the AIO since I have found that the EG4 3K connections are lame! I think you will find the issue and report back with your findings. Happy 4th everyone!

 

[BarracudaBob]

I got blasted with no PV connected to the MPPT so be warned that there is voltage coming from the inverter. I had an issue a while ago that something was disconnected so been there done that.

 

[K7TH]

I got blasted with no PV connected to the MPPT so be warned that there is voltage coming from the inverter. I had an issue a while ago that something was disconnected so been there done that.

True due to the caps I believe. The midnite 150 will drop the PV down from 150 to 14.5v when the disconnect is engaged, so I always power down when doing any maintenance of ANY kind. Just yesterday at work, we had to do a 'flea drain' to upgrade firmware on a file server. I have not heard that term or ran into that issue for many years....Happy Fourth everyone.

 

[kc130up]

I have the same inverter and you can definitely connect/disconnect the PV while the inverter is running. I believe there is a relay I've heard click when the PV drops below 50v or so.

As was stated above, check the voltage on the solar string. Something may have came loose.

A few months ago, i only had 5 panels connected to the EG4-3K and was right at 130v. One panel went bad and i was down to 105v and it wouldnt produce any power.

 

[coppergenie]

Considering that every day PV disconnects after dark and reconnects with daylight this warning label is likely not worded correctly. It is not sufficient to visually inspect for opens. I would start doing what Barracuda Bob suggest and take voltage readings. If you have good voltage on PV wires when AIO is not connected, but it drops to zero once you connect, that is a sign of a bad connection or panel. Check all MC4's and especially check any inline MC4 fuse holders.

Thank you. Could you point me to a guide on this? For context, there is no way to check voltage on my system (before the AIO) without disconnecting components. I'm getting no voltage reading at the AIO PV terminal.

My main struggle is knowing what I should be reading for different test conditions and what the readings would indicate about the component or system.

 

[Mattb4]

You listed 4 in series panels with Voc of 46.6v Thus at your AIO you should have up to 186.4vDC. If you have a PV array disconnect (breaker or switch) between your array and the AIO you can cut the circuit and measure if that voltage is present on the PV side. If it is, but it goes to zero, when you turn the breaker back on you know that there is a problem with a connection or panel*.

Your MC4's are also a point where you can disconnect (being careful though they should not be under load if your AIO is not working and the disconnect is open) and measure voltage from individual panels.

* It is possible your SCC is toasted but I would not think that as likely.

 

[K7TH]

Also, on MC4 connectors the heat can expand them or cause a situation where they disconnect with the next panel or hop. Also, some of these MC4 connectors do NOT mate well if you buy from one, two, or three providers.....ask me why I know...... Anyway, my bet it that your controller is good and you need to get out there and trace each an every connection until you reach the controller......its a PITA but you will learn a lot from the exercise. Better now than in the middle of a storm or outage.

 

[coppergenie]

You listed 4 in series panels with Voc of 46.6v Thus at your AIO you should have up to 186.4vDC. If you have a PV array disconnect (breaker or switch) between your array and the AIO you can cut the circuit and measure if that voltage is present on the PV side. If it is, but it goes to zero, when you turn the breaker back on you know that there is a problem with a connection or panel*.

Thank you. I have checked the following:
- each panel has about 39.5 V across
- series string has 159.1 V across
- voltage across whole system tested on positive side of panel string is 158.3 V (DMM placed in series, AIO off, but disconnect switch was closed; not sure if this is bad practice. Must still be something connecting the two PV sides inside the AIO even while off.)

So the whole loop has continuity. I don't know what else to try with my knowledge. You mentioned some indicator of the voltsge going to zero, but I didn't understand. Could you clarify that if it's still applicable?

 

[kc130up]

You have 159v with all 4 panels in series, but 0v at the AIO pv terminals? Or you have 159v from the panels with the AIO disconnected?

 

[coppergenie]

You have 159v with all 4 panels in series, but 0v at the AIO pv terminals?

Yes, 159 V with DMM in parallel. 0 V with DMM in parallel (didn't know it was parallel at the time) at the AIO PV terminals. Though these were not tested at the same time. Do those two readings contradict each other?

Or you have 159v from the panels with the AIO disconnected?

AIO is switched off. AIO PV terminals are still connected.

 

[kc130up]

Try disconnecting the PV terminals from the EG4 and test the string (assuming you don't have a breaker/disconnect to test at).

I have a PV breaker so when I needed to test I would turn the breaker off, disconnect the MC4's, turn the breaker on, and test at the MC4 output of the breaker.




Having the EG4 off, doesn't mean a short/wiring issue isnt still present.

 

[JWLV]

Check your connections. Over time with the PV wires heating and cooling every day/night, the connection for the pv wires at the inverter could have come loose. This has happened to me once. So now I check it every few months to make sure they are nice & tight.

 

[coppergenie]

Try disconnecting the PV terminals from the EG4 and test the string (assuming you don't have a breaker/disconnect to test at).

Yes, I have a disconnect switch like yours in the photo.

Testing at the switch I get continuity, 0.5 A, but no voltage. Current reading way low. I guess it's time to take all my shrink tubing off and check every connection because it seems like a loose or corroded connection somewhere.

 

[kc130up]

You're 95% there. Bad connection or bad panel. Hopefully its just a loose connection!

 

[K7TH]

Yes, 159 V with DMM in parallel. 0 V with DMM in parallel (didn't know it was parallel at the time) at the AIO PV terminals. Though these were not tested at the same time. Do those two readings contradict each other?

AIO is switched off. AIO PV terminals are still connected.

Hum, going to 0 volts under load in my mind means something is opening up (open circuit) or eating the power (short circuit or corrosion / burnt terminal some place. You need to remove the AIO PV connections and test the panel facing side of the disconnect as well as the output side of the disconnect to see IF the voltage is crossing the disconnect. This would at least tell you the disconnect probably is working. You can always take a pig tail for +/- using 10awg and alligator clips to bypass the disconnect. Also, one other option that I have ran into is that UNDER LOAD, one bypass diode could be opening on a PV panel. IF you can you may need to INPECT each panel's wiring box. I had two panels with lifted solder creating a low/zero voltage and causing a now power situation. Here is SE Michigan my panels run 140 to -20F so there is that and time will eventually cause a fatigue in the system.....somewhere!

 

[coppergenie]

Here is SE Michigan my panels run 140 to -20F so there is that and time will eventually cause a fatigue in the system.....somewhere!

Thank you for the insight! I have done all of the following so far:
- Checked voltage across each panel (good results)
- Switched out every panel (I have a spare), testing the system with each switchout (no change)
- Across the PV disconnect switch: OL readings when open; negligible resistance readings when closed
- Across the PV side (four panels and 500 ft of AWG 14 stranded; not including the disconnect switch): solid continuity beep, ZERO OHMS with no variance (even the dc switch showed 0.3 ohms when closed), <0.1 A, <0.1 V. It's cloudy, but still I expect voltage reading to be normal. On top of negligible voltage, I'm wondering what could possibly be causing the zero ohms reading (no the DMM leads were not touching).
- Plus anything else I have already mentioned

Since I'm reading no significant voltage across the PV load side, I assume there is a problem here, like you mention. But since I switched out all panels with no change, and I get the right voltage across the panel string, I'm assuming it's some problem in the wires or connectors, I just don't understand what the readings indicate (continuity, 0 ohm, <0.1 A, <0.1 V).

Every joint is crimped, no solder. And I don't expect it to be a loose crimp problem, because I adjusted them all several times throughout this testing and not once has the AIO displayed anything. The only thing I can think of now is to replace some of the MC4s.

 

[kc130up]

What voltage do you get across all the panels right before the wires connect to the inverter?

From here ----- [ panel ] ------ [ panel ] ------ [ panel ] ------ [ panel ] ------ [ panel ] ------ [ panel ] ------ to here