question

Vicstix avatar image

Why does the production of my Fronius drop drastically until it is restarted in an AC Coupled system with a Victron Inverter

I installed an 8kVA Victron Quattro Inverter, with Venus GX, 13.8kWH BYD battery, 250V 100A Smart solar MPPT Charge Controller, 3500W PV panels for the Charge controller.

Alao installed is a 5kW Fronius Primo Inverter with 6,300W PV panels


The system works fine and I could get as much as 5000W on the Fronius; but the production of the Fronius drops drastically almost daily; it could go as low as 190w (even when there's ample sunlight) and stays that way until the Fronius is restarted (then it starts producing normally again). Also when the batteries are full, the production of the Fronius drops and the loads start taking from the battery.

What could be responsible for this, and how can this be corrected?

Take note also that the panels are not shaded, and the production of the Charge Controller's panels stay as expected.

The system is entirely off grid.

The assistant used is the PV assistant.


I'll greatly appreciate any help I can get from @guy Stewart and from anyone. Thanks.

victron productsFroniusAssistants
8 comments
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @Vicstix

Without more information this cannot be answered...what is AC-frequency when this happens, battery voltage, charge state, etc.

can you post some VRM information with this visible?

Also, what firmware are you running?

@ Daniel Boekel, thanks so much for your response.

The firmware version on the Victron Quattro Inverter was 2655430.vfd; and I upgraded it to 2655465.vff.

I have included an image showing the display on the Fronius Inverter. I have also included VRM Remote Console screenshots and an image of the Color control's display. The battery voltage and state of charge are shown in the screenshots. The AC frequency when the power of the Fronius started dropping was 51.2Hz, then it became 49.9Hz after a few minutes.


Also when the batteries are full, the production of the Fronius drops and the loads start taking from the battery.

If you have an off-grid system and the battery is full, the Fronius is supposed to reduce its output (all the way down to 0W and this zero feed-in is actually what one wants and why Victron-Fronius is a good duo), because where should the energy go else?

Even if you have loads, you cannot assume the Fronius will be regulated "just in time" by some GX device to exactly match them and leave the battery alone.


The system, however, should be designed so the Fronius would cover loads "just in time" by itself.


So I do see another issue with your system. You seem to have a DC/AC hybrid - a configuration we build on a regular basis - and from the energy flow screenshot it looks like the loads are mostly covered by the DC system ("PV charger"). This is quite wrong.

Usually, you want the Fronius connected to AC-Out2 ("critical loads"), where it covers with its output the load and only the excess energy is used via charger to load the battery.

In such a system, the DC PV charger is supposed to load the battery only/mostly. As is, the picture suggests you are servicing your loads with the MPPT -> Inverter efficiency of roughly 91% (98% * 93%), when the Fronius could cover this with a 98% efficiency and you could load your batteries with 98%-99% efficiency from the MPPT directly.

@PetaJoule, thanks a lot for this. The zero-feed-in explains why the PV inverter would not produce when the battery is full; but I still need to know why it would not produce even when the battery is not full.

Yes, I have a DC/AC Hybrid system.

Are you saying that the Fronius should be connected to AC Out 2? Also, you referred to AC Out 2 "critical loads", did you mean "non-critical loads"? because I've always known AC Out 1 to be for critical loads, while AC Out 2 caters for non critical loads.

Yes, Fronius should be connected to AC-Out 1, which is "Critical Loads" (i.e. output that remains on when inverting). I mixed that up, so yeah AC-Out 2 is not used at all.

So if Fronius and loads are connected to AC-Out 1 , loads will preferably take power from the Fronius and only the excess will be used to charge the battery, which is what you want. The screenshot from the GX suggests otherwise, but in the other comment you confirmed you have set it up this way.

I'm not familiar with the BYD battery, could be that it has some sort of charge limit kicking in, although I'd hope that not being the case.

So my biggest issue is really with what I see on the GX screenshot. Fronius is throttled, MPPT not(?), it looks like loads are powered from MPPT -> Inverter chain (which is suboptimal) and only a small excess is used to charge battery (I assume it is charged as I see no - in front of the 201W)

Daniel asked you about the firmwares (of GX, of Quattro are most interesting, but while you are at it, check the MPPT too) - I would like to know these too.

ESS is still a construction site, so newer seems better. ;-)

A little more information about the system: The SOC does not affect the PV Inverter's harvest and the frequency stays pretty constant. I have attached screenshots. Thanks

Is your system connected like this or not?

"PV Inverter" being the Fronius. (and please discard the blue arrow on top going into the MP)


1576755535818.png (222.6 KiB)

@PetaJoule, that is exactly how the system was set up. The coupling was done on AC Out 1. AC Out 2 wasn't used at all.

1 Answer
Daniël Boekel (Victron Energy Staff) avatar image

Hi @Vicstix

I don't actually see any power being taken from the battery on your screenshots?

please do remember that because regulating the AC coupled PV takes a couple of seconds to ramp up / down, so having some charging / discharging is normal, especially when off-grid.

1 comment Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

@Daniël Boekel (Victron Energy Staff), thanks so much. I would take another screenshot when the sky isn't so clear and more loads are put on so you can see when power is being taken from the battery