On 2022/12/09 at 4:23 PM, Georg594 said:

 

 

INTRO:

Long ago i had the theory that Internal Supply of the inverter will cause currents on both mppts inputs which will interference with mppt regulation.

 

NOW IT'S PROVEN:

The simple solution is to pull of the plug on the supply pcb which is marked with "SOLAR".

The downside:   1. Loosing of Batteryless Operation .

                            2. Inverter will not start from PV inputs when its switched off by the button to charge the battery.

MPPT works now with a minimum current of 0.0002A    really   0.2mA   (first time i cant believe the measurement)

 

FIRMWARE Modification is still needed but just only a littlebit .....not so much as in the past.

 

@Coulomb

Only the subtraction value needs to be zero, no longer need to set the minimum MPPT voltage higher than 90V or reset mppt code.
Maybe we can even lower this for maximum compatibility.(eg 60v like in the twin models)

 

 

 

Here a picture of the disconnected plug (on the upper right)

The safest way is to completely disconnect your inverter from any powersource before pull of the plug .

 

RESULT :

 

I supply PV 2 input with a isolation tester which provide 250V

MPPT point found by the firmware at 168V and 0,000487A 

 

WOW this is 0,082 watts 😅

 

MAYBE IT CAN USE THE MOONLIGHT FOR SOLAR PRODUCTION??? I'll check

 

Green: original condition
Blue: SOLAR cable disconnected

5 hours ago, Volker said:

Green: original condition
Blue: SOLAR cable disconnected

Wow.

I wish I could understand what causes such dramatic changes, almost always for the better.

10 hours ago, Coulomb said:

Wow.

I wish I could understand what causes such dramatic changes, almost always for the better.

The improvement is really dramatic: all three converted inverters produce more output in low light conditions.

The inverter with the original software hangs at 90V in the morning. But it doesn't seem to be a bug, but a feature. I can't say more yet.

BUT: I can no longer switch the inverters to battery mode (SBU)!

I guess it's more complicated than just disconnecting a cable! I must continue testing...

Has anyone tried this hardware modification with any other Axpert lines that seem to suffer from the same mppt bug?

On my axpert 8kw all modes work normal, ( solar & AC pulled off).

 

But what i note was a charging from grid to battery when solar power is low , the measurements are so wrong of my axpert that it thought there is 100 or 200 w solar power available and because the dc bus is shared for mppt & charger the power comes from grid to the battery.

 

The power moving in the dc bus is just a firmware calculation & there is no measurement of current flow, just a logical firmware decision....

 

....if it measures 200w solar power input it will take this power out with the charger from the dc bus but the power in real comes from grid .

 

The measurements are sometimes way off reality.....real was 9 watt but axpert shows 71w 

 

 

The " help" supply is using around 40 watts from solar when connected and pulls down the voltage so that mppt point cannot be used .

(that does not mean that the 40 watts are not consumed)

 

If you calculate the 40 watts in general and the higher voltage because of mppt point is now always found by firmware the diagram makes also sence to me.

 

@ Coulomb

Where do you see dramatic increase in power ? 

 

Did i miss something?

Biggest incress in watt is from 38watt to 137watt

2 hours ago, Georg594 said:

On my axpert 8kw all modes work normal, ( solar & AC pulled off).

 

But what i note was a charging from grid to battery when solar power is low , the measurements are so wrong of my axpert that it thought there is 100 or 200 w solar power available and because the dc bus is shared for mppt & charger the power comes from grid to the battery.

 

The power moving in the dc bus is just a firmware calculation & there is no measurement of current flow, just a logical firmware decision....

 

....if it measures 200w solar power input it will take this power out with the charger from the dc bus but the power in real comes from grid .

 

The measurements are sometimes way off reality.....real was 9 watt but axpert shows 71w 

 

 

I assume that it is only one inverter compared to Volker which has a parallel operation of 6 inverters, probably from here the differences in behavior from them

https://powerforum.co.za/topic/15092-axpert-max-ii-10kw/page/7/#:~:text=complete off again!!-,Information%3A I have a parallel operation of 6 inverters. All inverters must be either in power out or standby mode%2C otherwise I see a CAN bus error (F80).,-I have therefore

"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 only one inverter installed in single phase, but unfortunately I didn't have time to open it and remove the "SOLAR" plug to see what happens

Anyway, for now the best firmware that exists and is tested by me for Aexpert Max II 10 Kw is QFW 656.05 made by Coulomb  (I don't underestimate the work done by Georg594 either!)

 

Edited January 6, 20233 yr by ClauBebe

10 hours ago, Georg594 said:

@ Coulomb

Where do you see dramatic increase in power ? 

Well, like in Volker's graph here. In one case in the late afternoon, from about 37 W to over 150 W. OK, it's only ~115 W in absolute terms, but the ratio is over 4x. Psychologically dramatic. Granted, not really all that dramatic.

For me, the technical findings are already "dramatic" and I am also happy about the power generated. Here in the north of Germany there is little sun and light in winter. In southern Germany, for example, solar radiation is already 30% higher...
Of course, the 300 Ws (!) from the moon at midnight (yes, really!) are not that much. 😊

I am very happy that through your and Georg's work I can now explain the influence of the two MPPTs and the "morning and evening gap". The 90V problem in phase 2 with the original firmware still exists, so the adjustment of the firmware is also necessary (diagrams follow).

Greetings, could I find out what detail is that overlined by yellow square? (Photo 4) As I know it's a part of Help Supply

There is something written like "F1". Probably it may help
 

 

Edited January 31, 20233 yr by Juvenate

On 2022/12/10 at 2:28 AM, Coulomb said:

Heh. The measurement is in hundredths of an amp (10 mA units), and the tracker stops as soon as it finds an increase in PV power. So it just happened to stop at 168 V when it the first bit of noise caused the measured power to increase from zero to just over zero. The half milliamp of current draw that you measured is likely due to the voltage measurement resistors, not the boost controller actually pushing power to the DC bus.

This is however a great find. Perhaps they can fix it in future by moving where the current measurement is done. Maybe DIYers could even do the mod ourselves, and leave that connector in place.

[ Edit: On further thought, it may be that the power, which is in the form of spikey square waves, is inducing noise into the current measurement. (I'm guessing that this connector supplies HFPWR, high frequency power, from the main power supply). The fix might be as simple as adding a capacitor somewhere. ]

So with original firmware, it still gets stuck at 90 V even with the plug out? And the subtract zero completely fixes the problem, as long as the plug is removed?

Well done doing what Voltronic's engineers haven't been able to do for many years! 👍

Greetings sir, don't you know how can I find out what detail is on photo? (overlined by yellow square) It's placed on help supply chip

Edited January 31, 20233 yr by Juvenate

6 hours ago, Juvenate said:

Greetings sir, don't you know how can I find out what detail is on photo? (overlined by yellow square) It's placed on help supply chip

3A 250VAC Pico Fuse. This is fast acting PCB fuse. From LittleFuse manufacturer. 

Hi

I'm joining this forum in the hope that someone will respond and maybe you have already tested it. I have a PV array with 3 Voltronic Power Axpert MAX 8000-48-230 inverters (they should be identical to yours) the original fw in the inverters was 72.03 now I uploaded version 72.07 I couldn't find a newer one on official sources. Do you have a solution for the "90V at low light" problem? I searched the forums and found FW 90.06 but the one I understood is not ideal. The links don't work with other versions...

thanks for any help or answer

Hi

I'm joining this forum in the hope that someone will respond and maybe you have already tested it. I have a PV array with 3 Voltronic Power Axpert MAX 8000-48-230 inverters (they should be identical to yours) the original fw in the inverters was 72.03 now I uploaded version 72.07 I couldn't find a newer one on official sources. Do you have a solution for the "90V at low light" problem? I searched the forums and found FW 90.06 but the one I understood is not ideal. The links don't work with other versions...

thanks for any help or answer