SOFAR 8kW PV Installation: Opinions, Durability, Failures & Experiences with Inverters

SOFAR 4.4 KTLX G3 - PvConnectError at 555 V / (0.0 A), has anyone encountered this problem?

Servus, after upgrading my 2023 home installation I have moved the installation from the house to the allotment and am having a problem with the SOFAR 4.4 KTLX G3 inverter and am looking for the opinion of someone more experienced before I report the equipment for service.
Symptoms - On the display all the time: PvConnectError, PV1: 555 V - 0.0 A
I attach a photo of the display with the message.
That is:
-the voltage from the panels is (~555 V)
-inverter sees string
but is not drawing current (0.0 A)
What I have already checked:
Polarity correct (+ meter to red, - to black → +555 V)
String working (with load flows >8 A)
MC4 plugs correctly made (+ red / - black, markings match)
AC voltages OK; L1: 240 V, L2: 239 V and L3: 239 V
No other errors on display (only PvConnectError)
Installation without grounded poles (floating)
I carried out an additional test.
I connected directly to the inverter (from a neighbour), an independent string of ~344 V DC: separately to PV1 and separately to PV2. Identical result: after initialising the inverter, the following reappears: PvConnectError, PV1: 344 V 0.0 A (i.e. only the voltage changes)
What is strange. Inverter: sees PV voltage (both 555 V and 344 V), does not draw any current (0 A), on both MPPTs (PV1 and PV2) behaves identically. It looks as if: the MPPT is not switched on at all!
Questions:
1. Has anyone encountered this symptom with the SOFAR G3?
2. Can the message "PvConnectError" with correct voltage mean:
-pV / MPPT input failure?
-a problem with the DC relay?
-a software / start-up lockout?
3. Is there anything else worth checking before service?
Summary:
- two different strings → same effect
- two MPPT → same effect
- correct AC and DC voltages
- no current on PV side
It all looks like a problem on the inverter side, but I prefer to verify before reporting.
I would appreciate any concrete suggestion!

@zybinek you can still check the + and - to PE to see if it exceeds 30 VDC, although it would be best to measure the insulation resistance with a special meter, but if it comes out measuring OK then something is wrong with the inverter.

And what does the manual say? AI ask, maybe he has already encountered such an error.

As suggested, I asked AI - Copilot 365.
Below is his response:

"I have checked the topic more extensively and it looks quite clear:

the inverter sees the PV voltage (344-555 V),
but does not draw any current (0.0 A),
both MPPTs behave identically,
second, independent string - exactly the same effect.

That is, in practice, the PV installation falls off.
The PvConnectError in this situation looks like a failure to switch on the DC path - the inverter "sees" the string, but the MPPT does not start.
From the suggestions of colleagues I will still check:

+/- voltages relative to PE (whether there is an isolation problem)

On the other hand, with these symptoms it looks most like:
→ damage to the PV input / MPPT section of the inverter."

I wonder how this problem is seen by practicing installers?

Check also this PE what your colleague here suggested >>21909160
Because if there is an error here, the inverter won't start either.

As suggested by ja_pizgam I checked with a METEX 6030 meter on the 600 v DC range the "+" and "-" voltages relative to PE, with the "-" of the meter caught behind PE.
At first, relative to the "-" string (black lead) a voltage above -500V appeared and started to drop rapidly to about -92V and stabilised at this level and when I moved the meter's plus lead to the "+" string (red lead) a voltage above +500V DC appeared briefly and after a while started to drop to about + 91V.
What should I do in this situation to reduce it by at least 3 times?

Well then, either you have damaged wire insulation somewhere, or there is a strong induction of voltage in these wires. All you have to do now is measure the insulation resistance with a special meter.

In my hyd it was throwing a similar PV error, the software has bolted despite upgrading to different versions. Currently there is one string and it goes in "parallel" mode even though it should be independent

zybinek wrote:

What should I do in this situation to reduce it by at least 3 times?

Look through the mc4 connectors to see if any are lying in the water, check the wires.
Is it PE you have sure? maybe it is not contacting and something is induced from the phases in PE? maybe one of the panels has leakage?
Something must be on the connections since it was running and after moving it stopped.

Added after 2 [minutes]:

Where do you have it installed anyway? city, region.

Thank you for all the hints!

The String 555V consists of 2 tables of 250 Wp panels. One table has 9 panels and the other has 8 panels.
The installation with these panels was commissioned in 2016 near Legionowo. So the panels and cables have at least 10 years of use behind them, while the SOFAR 4.4 KTLX-G3 inverter has a warranty term as shown in the picture. I got it for a small amount from my brother, who was upgrading his installation and had to replace the inverter with a hybrid one. The tables are 10-15 m from the inverter and the power cables are buried 50-60 cm underground.

DJ MHz wrote - " Actually there is one string and it goes in 'parallel' mode even though it should be independent " maybe there is actually something wrong in the settings?

Or does anyone, based on this information, have any more ideas on what to do about this faff?

Problem solved.
The cause of the error PvConnectError was an incorrectly set PV input mode on the inverter.
Parallel mode was set and the installation was not symmetrical (one string connected to one MPPT.
After changing the setting in the menu:
Settings -> Input modes -> password 0001 -> mode "independent" the inverter started to work correctly and draw current from the panels.
The installation of the panels was operational and the problem lay solely in the inverter configuration.
Thank you for all the hints that helped me solve the problem!

Under 0715 you also have various interesting settings

I was in standalone mode with the units connected independently – I physically switched them to parallel mode to split the currents, changed the setting to parallel (but didn’t reset the inverter), which meant the settings didn’t take effect.

I can’t remember what the error was, but the result was that output didn’t exceed 200–300 W (even though it should have been running at 8 kW at the time) – the string voltages were a bit odd too.

P.S. If anyone has 3.3, 4.4, 5.5 or 6.6 models for sale or exchange (I have 11ktlx and 20ktlx g3), please email