19 hours ago, Coulomb said:

This is great data; I bet even Voltronic R&D don't have 6 strings of PV in various weather conditions!

Just so I can be clear on what is needed, what do you mean by fluctuations in the lower range? Do you mean when the PV voltage is say 120-170 V, the resets are setting the voltage to max so often that you end up losing solar energy production?

I worry that the 175 V threshold is specific to a particular string configuration, and may not be appropriate for longer or shorter strings of panels / total cells. Your suggested values, all quite low values, seem to me to have a better change of working for most panel configurations.

 "We have insulated the engine compartment of a car with a squeaking cail belt."

Please remove the setting of the voltage from the firmware.
Removing the SOLAR cable from the SPS is the real solution!

Edited January 4, 20233 yr by Volker

2 hours ago, Volker said:

Attention - Attention - Attention

When the SOLAR cable is disconnected at the SPS, EVERYTHING changes!

The two blue curves are the inverter with the cable disconnected!
Red is the original firmware, otherwise all on version 656.05.

I will also disconnect the cable to see how it behaves without it. What would be the disadvantages with a disconnected cable? I use all 3 sources of energy from solar/grid/battery, not necessarily in that order....when there is enough sun, I often turn off the grid remotely through a contactor commanded by Shelly EM and stay on the battery...I wouldn't wants to stay without electricity in the house .. 

From what I see from the graph you posted, there is a huge difference compared to the production with the cable removed and those without the cable removed This is basically how it should behave in low light!

L.E. : here is the production of the  Solax and Aexpert

Solax 0,50hWh

 

Aexpert 0,5 kWh

Edited January 4, 20233 yr by ClauBebe

2 hours ago, Volker said:

Please remove the setting of the voltage from the firmware.

My impression is that factory firmware with the Solar cable unplugged was still stuck at 90 V some of the time.

So what parts need to stay, and what parts go?

@Georg594, any thoughts?

57 minutes ago, ClauBebe said:

What would be the disadvantages with a disconnected cable?

There are some downsides; see Georg's original thread. I forget the details. I think you can't operate batteryless with that plug removed, for example.

Sorry, I thought everyone was disconnecting their Solar cables already.

As far as I know, even though that cable is labelled Solar and goes to the solar charge controller, the power on that connector is what is called HFPW+ or HFPWR, for High Frequency PoWeR. Basically a 100 kHz square wave that when rectified produces 12 VDC, and it comes from the main power supply. This main power supply can get its power from AC-in, or the battery. I'm pretty sure that it can get it from solar too. But I'd love to have that confirmed.

 

Edited January 4, 20233 yr by Coulomb

5 hours ago, Coulomb said:

Sorry, I thought everyone was disconnecting their Solar cables already.

I didn't even know about the part with the disconnected "SOLAR" cable until you posted...

I will also disconnect the cable to see how it behaves without it

anyway, without hardware changes the inverter is "waw" with QFW656.05 compared to OFW56.02 that I had
Never in conditions of rain and fog, Aexpert didn't manage to produce more than the Solax, although the arrangement of the panels is more advantageous for Aexpert

it is the second dark day when Aexpert manages to produce more than Solax

I hope to find time tonight to open the inverter and disconnect the cable "SOLAR", because it powers the whole house and lasts until I disassemble it and reassemble it  

First I disconnected the SOLAR and GRID cables on the SPC board.
1. The inverter does not start when grid is switched on.
2. Starting via the button is possible.
3. The inverter then switches itself complete off again!!

Information: I have a parallel operation of 6 inverters. All inverters must be either in power out or standby mode, otherwise I see a CAN bus error (F80).

I have therefore reconnected GRID and only SOLOR is disconnected!

I disconnected the SOLAR cable at the output of the MPPT board. I wanted to avoid a cable with 400V that was not connected at other end. The cable was secured with a cable tie (see picture).

I have modified 3 inverters. One inverter has the original firmware, so I will see tomorrow if the 90V error occurs.

(And anyone who does not feel confident should have it done by someone with appropriate experience.)

On 2023/01/04 at 7:24 PM, ClauBebe said:

I didn't even know about the part with the disconnected "SOLAR" cable until you posted...

I will also disconnect the cable to see how it behaves without it

anyway, without hardware changes the inverter is "waw" with QFW656.05 compared to OFW56.02 that I had
Never in conditions of rain and fog, Aexpert didn't manage to produce more than the Solax, although the arrangement of the panels is more advantageous for Aexpert

it is the second dark day when Aexpert manages to produce more than Solax

I hope to find time tonight to open the inverter and disconnect the cable "SOLAR", because it powers the whole house and lasts until I disassemble it and reassemble it  

Green: original condition
Blue: SOLAR cable disconnected

On 2023/01/04 at 10:30 PM, Coulomb said:

As far as I know, even though that cable is labelled Solar and goes to the solar charge controller, the power on that connector is what is called HFPW+ or HFPWR, for High Frequency PoWeR. Basically a 100 kHz square wave that when rectified produces 12 VDC, and it comes from the main power supply.

I'm having doubts about this now. Looking at some photos, CN5 (where the cable in question connects to the SCC/MPPT board) seems to be connected to BUS-, the thick black wire that runs off to the main board. That's the output of the MPPT. Though there seem to be medium current diodes (3-6 A) near either end of the cable in question. So confusing.

Edit: It might be useful to know if these are ultra-fast diodes (often with part numbers starting with UF), or slow, DC diodes. Outlined in orange:

Edited January 6, 20233 yr by Coulomb

9 hours ago, Coulomb said:

I'm having doubts about this now. Looking at some photos, CN5 (where the cable in question connects to the SCC/MPPT board) seems to be connected to BUS-, the thick black wire that runs off to the main board. That's the output of the MPPT. Though there seem to be medium current diodes (3-6 A) near either end of the cable in question. So confusing.

Edit: It might be useful to know if these are ultra-fast diodes (often with part numbers starting with UF), or slow, DC diodes. Outlined in orange:

I have not found any information on this in the service manual.
https://powerforum.co.za/applications/core/interface/file/attachment.php?id=30207&key=980058f2d3c190a1a81f178547c8fc75

I can measure the bus connection and see if I can read something on the components...

 

On 2023/01/04 at 2:30 PM, Coulomb said:

There are some downsides; see Georg's original thread. I forget the details. I think you can't operate batteryless with that plug removed, for example.

Sorry, I thought everyone was disconnecting their Solar cables already.

As far as I know, even though that cable is labelled Solar and goes to the solar charge controller, the power on that connector is what is called HFPW+ or HFPWR, for High Frequency PoWeR. Basically a 100 kHz square wave that when rectified produces 12 VDC, and it comes from the main power supply. This main power supply can get its power from AC-in, or the battery. I'm pretty sure that it can get it from solar too. But I'd love to have that confirmed.

 

Just went through this amazing thread post by post. I just have to say that I am amazed at the wealth of information and learning curve on my end. Thank you for all the contributions. 

@Coulomb can you point us to that "thread" with a link, i was not able to find it.

4 hours ago, mafiose13 said:

can you point us to that "thread" with a link, i was not able to find it.

look here

Coulomb

if it helps, I have also put today's data here in the .cvc file, including the one with bus voltage

sunny day in the morning and then with the sun hiding among the clouds and clear at sunset

 

Load power-data-2023-01-06 22_50_30.csv Bus voltage-data-2023-01-06 22_50_21.csv PV power-data-2023-01-06 22_50_05.csv MPPT 2 Power-data-2023-01-06 22_49_56.csv MPPT 1 Power-data-2023-01-06 22_49_47.csv MPPT 2 Current-data-2023-01-06 22_49_35.csv MPPT 1 Current-data-2023-01-06 22_49_24.csv MPPT 2 Voltage-data-2023-01-06 22_48_45.csv MPPT 1 Voltage-data-2023-01-06 22_48_27.csv

On 2023/01/06 at 6:44 AM, Volker said:

I have therefore reconnected GRID and only SOLAR is disconnected!

Yes, I can't see any advantage in disconnecting the GRID connection (230 VAC) from the utility power supply. @Georg594, why do you disconnect yours? Just because the other circuit on that same board is apparently causing trouble?

On 2023/01/06 at 6:44 AM, Volker said:

I disconnected the SOLAR cable at the output of the MPPT board. I wanted to avoid a cable with 400V that was not connected at other end. The cable was secured with a cable tie (see picture).

Can you see that this cable in fact has BUS+ and BUS- on it? I guess that makes some sense. And my idea that this connector has HFPW+ on it is looking very unlikely. The fact that it is a 3-pin connector with the middle one unused suggests that there is high voltage, and they are increasing creepage distance this way; that would only be needed if the cable carried high voltage between the two conductors.

The reason for disconnect the AC plug is when you disconnect Solar it pulls constantly around 60 watts from grid , in every situation and every mode (usb,sub,sbu) .

To originally avoid this there is a priority logic (in hardware)on the help supply so solar prio1 , ac prio2.

 

So if you don't disconnect both the grid meter shows around 1.5kwh per day.

 

 

Are you shure with bus - and bus + ?

Mine has 2 diodes ( pv1+ diode + pv2+ diode , the other wire is pv -.

 

I testet it on mine and then the inverter was not able to switch of because of the information from the 4 pin plug on the help pcb is : stay on .

Forgot to say the test was to supply the help pcb from dc bus directly 

2 hours ago, Georg594 said:

Are you sure with bus - and bus + ?

No, not at all.

Quote

Mine has 2 diodes ( pv1+ diode + pv2+ diode , the other wire is pv -.

Ah, that makes more sense. I'm guessing that PV- is commoned to BUS-. So it only takes power from whichever PV input has the highest voltage at any instant. It must be confusing to the MPPT logic, especially in low light conditions, to suddenly have 40-60 W of load come or go. This could well be the core of the problem. Say the algorithm decides to lower the voltage, but it's now 20 W below the maximum power point. It measures the PV power, and finds that it has gone up. Not because this is a higher power point, but because the power supply's 40 W is now being supplied by the other PV panels on the other MPPT port! So it rewards that bad behaviour by lowering the PV voltage again, reducing power further.

But that would mean that this is exclusively a MAX problem, with its dual high voltage SCCs. Or at least a lot worse on the MAX models, especially if the number of panels is roughly equal on each MPPT. As they often but not always would be. Perhaps you owners could keep a lookout for that behaviour?

I imagine that the power supply's load would be very roughly constant. So it might be possible to take this effect into consideration. Taking out the plug from the Solar power supply means that the power for the main power supply has to take a longer, less efficient route: through the SCC(s), via the DC bus to the DC-DC converter. That's 2 conversion steps,  potentially going up then down in voltage, compared to just one step for the Solar power supply. Add one more step for the output of the main power supply, so then it's 3 steps (PV -> bus -> battery -> output) versus 2 (PV -> battery -> output). I assume that the solar charge controller board's gate drivers and logic run off the 48 or 60 V (battery) bus. 

4 hours ago, Coulomb said:

Yes, I can't see any advantage in disconnecting the GRID connection (230 VAC) from the utility power supply. @Georg594, why do you disconnect yours? Just because the other circuit on that same board is apparently causing trouble?

Can you see that this cable in fact has BUS+ and BUS- on it? I guess that makes some sense. And my idea that this connector has HFPW+ on it is looking very unlikely. The fact that it is a 3-pin connector with the middle one unused suggests that there is high voltage, and they are increasing creepage distance this way; that would only be needed if the cable carried high voltage between the two conductors.

I disconnected Grid because I wanted to see if the current consumption drops: 4 inverters switched on consume approx. 350 W without load.
I also think it is a connection to the bus. I measured 0 Ohm between MINUS-SOLAR and MINUS-BUS.
The inverters are mounted on the wall, it is difficult to measure when holding the cover with one hand. I should probably take an inverter out of parallel operation and run it on the table in the workshop....

23 minutes ago, Volker said:

I also think it is a connection to the bus. I measured 0 Ohm between MINUS-SOLAR and MINUS-BUS.

But that would also be the case for my latest presumed arrangement:

Perhaps you could verify by finding about 0.4 V drop from PV1+ or PV2+ to Solar+.

1 hour ago, Coulomb said:

But that would mean that this is exclusively a MAX problem, with its dual high voltage SCCs. Or at least a lot worse on the MAX models, especially if the number of panels is roughly equal on each MPPT. As they often but not always would be. Perhaps you owners could keep a lookout for that behaviour?

I had wondered several times that the MPPTs influence each other. But only with exactly one inverter!

Then I connected the two strings to another inverter. Again, the same effect.

Yes, exactly these two strings have 10 modules. The influence only occurred with a uniformly overcast sky. Not in sunshine, because the fields are aligned very differently! Here, the two MPPTs took turns driving the load.

The effect disappeared after disconnecting the SOLAR cable.
You have found the explanation.

 

 

4 hours ago, Georg594 said:

Forgot to say the test was to supply the help pcb from dc bus directly 

Have you connected the SOLAR cable behind the two diodes?

Unfortunately, I still don't know why the inverter switches off when I disconnect SOLAR and GRID from the PLC. Minimising the consumption would be good...

Even bigger for me is the new problem: When I switch to SBU mode, Grid remains connected and supplies power to the output.
I use MultiSIBControl (http://multisibcontrol.net/) and Remodedesktop, so I still have to check if it is only a problem of the display (hope it). Now go to the inverter room....

"Grüße aus dem Norden"

I cut off the two small wire from mppt board and adapt them to the two big outputs from the mppt pcb .

 

But this let the inverter run constantly and like coulomb said triple energy conversion losses .

 

 

Yes, I had suspected that. I really want to reduce the current consumption from the grid.
Therefore I will insert a diode between Bus+ and Solar+. (I am in favour of the minimal intervention in the original hardware...)

23 hours ago, Volker said:

Therefore I will insert a diode between Bus+ and Solar+.

I assume that you know what you're doing; it seems there are plenty of ways to get this wrong.

It had no success.
I soldered a diode between Bus+ and Solar+ and disconnected Gid from the PLC; SOLAR is connected. The inverter can be started with and without the grid using the front button. But then it goes into error state F9 (Bus soft start failed).

(I will probably disconnect the inverters on the grid side using an installation contactor and only connect them at night when the battery needs to be recharged).

Can someone give me some lessons? The inverters also run without PV and grid when only the battery is connected. Why does the inverter need this power supply unit (PLC) if the batteries are not discharged? Does anyone have an explanation?

I will disconnect the grid from the PLC again, maybe I just need to switch on all 3 phases online fast enough! What do you think?

Here are the diagrams with SOLAR cable is disconnected on all inverters.

Inverter 1 00056.06 (Georg)

Inverter 2 00656.05 (latest version)

Inverter 3 00056.05 (not used)

Inverter 4 00056.05 (original)

Inverter 5 00665.05 (latest version)

Inverter 6 00056.06 (not used)

 

1. 656.05 and original

The 90V error in the original software (red): At the beginning (<90V), the original software has an advantage for a short time. Our setting of the voltage is disadvantageous here. We should not do anything below 89.5 V.

Then comes the 90 V clamping. The patched firmware has an advantage - but we change the voltage value too much. Is it enough if we don't set hard but only add 10 V?

The original firmware also tends to fall back to 90V!