Well, 5.4 is quite different from 19.5 kWp… if he feeds 4 kWh into the grid, he gets… oh, great… €4.00 in feed-in payments…
and if the battery is already full by 10.00 a.m., everything has to be curtailed..
and the probably additional 50–80 kWh go unused.
But let’s leave it at that… it’s not really part of the topic… but I have a small system like that too… because it’s simply cheaper to buy some electricity for around three months in winter than to put a huge system on the roof.
He has 3 MPII-5000 units, so of his maximum generation of 19.1 kW, he can already feed 13.8 kW into the grid. He therefore only needs to charge his storage battery when his PV system is producing more than 13.8 kW. At night, however, he has no solar generation at all, and that is when he empties his 20 kWh storage battery again.
I don’t see any solar generation being wasted there.
Unfortunately, I haven’t been able to test it in a real system yet either, to see how it behaves with a Fronius and AUX 1.
Do you mean a grid setpoint of -4000 W?
I could imagine that this wouldn’t really work; it would probably be better to test it with a grid setpoint of 0 W and actual surplus feed-in.
Section 4.3.14 of the ESS manual refers to zero feed-in or the feed-in limit set on the GX device (Settings → ESS → Grid feed-in); this isn’t really related to the “disable Feed-In” signal sent via AUX 1 from the MultiPlus.
Hello,
since I’m in the same utility area (Waldviertel), my system is set up in roughly the same way, and I’m also waiting for approval 
I can say that if it’s configured the way they want, the Fronius goes into standby.
So unfortunately Dennibu and Ferdi are right—the AUX1 does not pass the signal on to the Cerbo. So there’s no alternative but to use the version with contact multiplication. Well, it does have certain advantages ….
No, I don’t mean the grid setpoint; that is set to 0 watts anyway, but rather the maximum feed-in.
Well, that is (unfortunately) complete nonsense—the 19.1 kW is the maximum yield that the PV modules could achieve with perpendicular irradiation; in Central Europe (my system is located at 48° 20’ north latitude), this value can never even theoretically be reached—but that is not the point of the discussion here. Strictly speaking, the peak value for each module should be specified in relation to the latitude of the installation site.
Realistically, the modules produce a maximum of 12–15 kW around midday, although I have never actually seen this output, because in summer, with a cloudless sky, the batteries (21 kWh capacity, with SOCmin set to 20% and normally charged only to 95% SOC) are full by around 10:00. My overnight consumption in summer is around 4 kWh, so the battery is unlikely to be empty then; in winter, things look quite different … Apart from that, I am only allowed to feed 4 kW into the grid. And that’s why HolgerBle is unfortunately right: in summer, probably 50–80 kWh will be lost …
As it appears, the Fronius only goes into standby when you apply the correct signal (open or closed) to the appropriately configured IO pin. I don’t know whether this is the same for all Fronius units. So if, like me, you have a Victron system and a Fronius, you have no alternative but to implement contact multiplication. As soon as I have implemented this on my system, I will post the procedure here.
PS: Circuit diagrams removed because they were incorrect!
Didn’t read it properly ??? He may only feed 4 kW into the grid…. not 14 kW.
You seem to have misunderstood something. He never needs to feed more than 4 kW into the grid. He just needs to make sure that his battery is empty before the midday peak. Then all the surplus from his PV system goes into the battery, while he still feeds only 4 kW into the grid. After three hours, the midday peak is over, and he empties his battery again through self-consumption and feeding 4 kW into the grid.
Then please read two posts further up ;O))))
And it’s more about pointing out that many people build a far too large (and therefore very expensive) system instead of buying some electricity in winter… and being satisfied with 20–30% self-sufficiency … and saving a lot of money.
So, I carried on testing today in glorious sunshine. And it doesn’t look too bad. I have no idea why, but as soon as I switch on the switch (only on the Fronius, standing in for the WLV), nothing is fed into the grid any more, while the consumers continue to be supplied. Although the documentation I have from Fronius is more than sparse — hundreds of pages on monitoring and WARNING — DANGER — BLABLABLA, but hardly any useful technical information. Also be careful when wiring the PINS — the manual says PIN 3, but it apparently actually means PIN 4 … So far, I have only tested the devices independently of one another; now I’ll carry out the wiring and cabling for the contact multiplication and test what happens then … I’ll post the results here afterwards.
@all
Here’s an explanation of the topic of WLV (active power specification) — as there seem to be a great many different interpretations here of what this means and how it is to be implemented…
+++ Applies ONLY to Austria +++
The grid implementation regulations of MOST grid operators in Austria have required the following since the beginning of 2025 (this also applies to modifications/extensions) for EVERY parallel-operation system with a capacity greater than 800 W:
The output power (not the feed-in power) of the entire parallel-operation system must be capable of being reduced to 0 W via a potential-free switching contact of the grid operator (load-switching device)!
(I.e. when this contact is closed, the output power must be ZERO AND all loads must once again be supplied from the GRID (= grid consumption!))
+++++
Solution using ONLY Fronius — very simple: connect the switching contact between V+ and IN8 (for example); configure the settings under I/O power management accordingly
+++++
Solution using ONLY Victronenergy MultiPlus-II:
Connect the switching contact to AUX1 of the MP-II MASTER (L1); GRID code: Austria
Here, the AC output power of the MULTI-PLUS-II system is regulated to ZERO; third-party inverters (whether primary or secondary) are NOT regulated as well. DC charge controllers (e.g. MPPTs) can still charge the batteries, but nothing is inverted to the AC side 
+++++
In mixed systems (e.g. Victron + Fronius), the ONLY option at present is a discrete setup with SIGNAL MULTIPLICATION and separate routing to EACH inverter in order to comply with the grid requirements under the WLV.
(Please take into account the voltage drop in long signal cables, relay power requirements, etc.)
+++++
I hope everything is clear now
I ONLY wanted to know from MATTHIAS LANGE @M_Lange whether forwarding the WLV SIGNAL from AUX1 via MODbus to the Fronius inverters is planned (to reduce the output to ZERO)?
THANK YOU
Oh, come on… When are we finally going to get reasonably consistent rules across Europe?
That would certainly be desirable, but the rules in the US are even less consistent than in Europe. And unfortunately, unfortunately, we are still a very long way from a United States of Europe.
Thanks for the information, Thomas. A very clear and easy-to-understand explanation. It isn’t stated so clearly in any manual, neither by Victron nor, especially, by Fronius.
You could perhaps test it beforehand with a simple switch. With my Symo 5.0-3-M, it’s V+ and IN5.
For both: simply use the utility’s switching contact and a DC power source (power supply, max. 48 V DC) to control a relay. The relay must have at least two suitable switching contacts (changeover contacts in my case).
However, there is one thing I cannot accept as it stands:
A few months ago, I spoke to an employee of EVN who only mentioned zero feed-in. That would be fine by me — after all, no one wants the grid to collapse. I’ll ask again, because I cannot accept being forced to import electricity from the grid.
Well, yes … the energy suppliers do want to sell electricity after all … so they set requirements like that … (which actually NO ONE can understand)
Addendum Fronius Symo 5.0-3-M configuration – active power setpoint (WLV)
Warning: The PIN numbers in the input pattern of the EVU editor, I1 to I8, are NOT
the same as those in the IO assignment: IO assignment = input pattern PIN no. + 1
or input pattern PIN no. = IO assignment - 1
According to the manual, rules that are not required should be deactivated (deleted)
The following settings are configured by default:
In principle, these settings could be used; however, lines 2 and 3 should always be deactivated.
Alternatively, you could use the following settings, for example (or as Thomas has just suggested):
As you can see, I4 is shown here in the input pattern. However, PIN 5 is shown in the IO assignment. PIN 5 must also be connected on the orange plug!
The corresponding IO assignment then looks like this:
That’s my information/interpretation too.
The NAP must be regulated to 0 feed-in, and theoretically you can still produce PV energy for your own consumption and use it directly; it just mustn’t feed anything into the grid any more (small fluctuations when the load changes should be okay, I think).
It might be necessary to switch to “regulation of the individual phases” so that if 1,000 W is being drawn on L1, 1,000 W isn’t fed into L2+L3, but instead each phase is regulated separately to 0 feed-in.
Edit: I’m referring to Germany.
I have just read through the parallel operation requirements for Lower Austria and Upper Austria, and unfortunately Thomas is right: these two federal states require the system to be switched off. Not Burgenland and Styria, for example. There, zero feed-in is permitted.
As you can see, even the individual federal states cannot agree …
PV Austria is also opposed to a complete shutdown, as well as to requiring a specific technical solution for implementing the cabling, connections, etc. Personally, I believe this is not the final word on the matter.
Of course, reducing feed-in to zero when the grid is overloaded is necessary, and I also assume that this will genuinely only be done in extreme cases, rather than being a requirement to draw power from the grid. Another point is that the system owner will (almost certainly) also have to pay a fee for the ripple-control receiver.
With these requirements, I would not build a PV system today … or else I would opt for a completely off-grid solution from the outset …
Oh dear, you can’t write things like that here 
. The term „off-grid system“ is hotly debated and stretched pretty far… In reality, an electrician deals with it too, and they have to gain experience somehow.
If it takes a total of three hours of bureaucratic hell and a bit of modification work, that’s still better than cobbling together a system at the end so that you can at least convince yourself you don’t have to register it. A bit more motivation! You’re nearly there.
Even three hours of bureaucracy have long since stopped being enough. The grid operator has to be willing as well. The current situation here is that the system has to be fully controllable and capable of being switched off, even with zero export. There are entire fields full of systems that aren’t controllable, but when it comes to private individuals, you can’t get anywhere. I gave up after six months. The switch has now cost me less money and hassle than having to beg them to connect the system to the grid.
Right, I’ve tried out a whole lot of things now (a kind person helped me a great deal with it) …
So, fundamentally, this works with the WLV as Thomas explained, BUT:
The settings in the utility editor need to be changed slightly: According to the (Victron) manual, IO control must be specified as the first priority in the priority settings. In the IO control’s input pattern, “Contact closed” (blue) should be selected for the 0% control rule, and “Contact open” (white) for the 100% control rule. If both rules are enabled, the following happens: When the 100% rule is active, the Fronius produces whatever the sun happens to be providing at that moment—in other words, 100%—and when the batteries for the MP2s are full and household consumption is currently low, the Fronius feeds flat out into the grid, even if the Cerbo is shouting “curtail”. The Fronius simply doesn’t care what comes in via Modbus; it stubbornly continues to follow the IO rule—the contact is open and the rule is active, so it operates at 100%. Normally, this wouldn’t cause any further problems. However, if the utility (as is unfortunately common nowadays) specifies an export limit of, say, 4.0 kW, and this has also been entered correctly on the Cerbo, things look different. The Fronius completely ignores the Cerbo because IO control has the highest priority.
However, if you don’t activate the 100% rule (or don’t program it at all), things look different: The contact is open, so the Fronius can’t do anything with it (because it has no rule for an open contact), so it checks what priority number 2 is, and Modbus is back in control.
Conclusion: The Victron system does what it’s supposed to do; however, if the 100% rule is activated, the Fronius simply doesn’t care.
