Section 9 EEG control with VE.Bus firmware version 555 and later

Other languages German community
, ,
1

Hello everyone,
we are the happy owners of a three-phase system with three MultiPlus 3000 units, two MPPTs and a Fronius Primo on AC-Out. Three Pylontech US5000 batteries are charged by the system.
The system is currently configured off-grid and provides hot water and supplies selected circuits in the house — there is no grid connection whatsoever.

Nevertheless, we would like to convert it to a grid-connected setup, as there can be significant excess energy in summer, while things can occasionally become tight in winter. I have now come across the requirements under §14a and §9 EEG. An EM24 is installed between the grid feed-in point and the house distribution board. In other words, the EM24 can also measure other household loads on the grid side.

§14a — in other words, switching off the charger — works perfectly. The control line is connected to AUX2, and grid charging can no longer take place. So far, so good.

The tests for the §9 EEG configuration have not been quite as encouraging so far. When AUX1 is switched, no more energy is fed into the grid, thereby complying with the statutory requirements. However, loads connected on the AC-In side are NO LONGER supplied by the PV system at all, because not only the feed-in power to the grid is switched off, but also the power within the house. Only loads on AC-Out continue to be supplied, provided that PV output and battery SOC allow it.
At least the Fronius, which is connected via Modbus/TCP, is controlled accordingly.

In short: at least as far as the grid operator is concerned, this is compliant, and the feed-in can be reliably switched off using an RSE or control box. Unfortunately, this also affects the loads on AC-In — the benefit of now being able to supply these via a grid-connected setup would disappear in the event of throttling or shutdown, and only AC-Out loads would continue to be supplied.

If we switch the ESS to zero feed-in, that also works perfectly and the grid power is quickly regulated to 0 watts. Here too, however: when AUX1 is switched, all consumers on the AC-In side are then supplied only by grid power.

Is there any setting that makes “Disable-Feed-In” mean only “actual feed-in”? Or is a new update planned for this in the near future? Or will I have no choice but to connect the entire house to AC-Out?

2

@nick22

I’ve just read a comment by @M_Lange in another thread saying that AUX1 is in fact always set to zero for AC_in, meaning that grid-connected loads aren’t taken into account. I hope I remember that correctly, as I can’t find the thread anymore.

I’ve connected all the loads to AC_out — or split them across the two outputs — but I have three MP2 5Ks, which can pass through 50 A. So, of course, it regulates against that.

The MP2 3k should be able to pass through 32 A. With a 35 A SLS and 32 A circuit breakers, that should work for you too, as long as you don’t have any large loads. However, 63 A domestic connections or SLS devices are standard.

It’s important to have a changeover switch so that, in the event of maintenance, the house can be switched directly to the grid.

Good luck!

3

hi,
the multis don’t know your grid meter; they only know their own output power on AC-in, and when feed-in is disabled, that is simply set to the setpoint of 0.

bye

4

I just read a comment by @M_Lange in another thread saying that with AUX1, AC_in is indeed always set to zero, meaning that grid-connected loads are not taken into account.

Yes, unfortunately that’s the way it is. Under the current requirements of the German Grid Code, triggering AUX1 means that power is drawn exclusively from the grid. It doesn’t matter whether the batteries are full or not, or whether the load is connected to AC-IN or AC-OUT.

5

I have zero feed-in, so it doesn’t affect me. But does the feed-in really have to be adjustable all the way down to zero? Wasn’t it also possible to throttle it to 30 or 60%? And wasn’t the AUX input specifically programmable for this with an assistant? I remember reading something about this exact topic, but I didn’t look into it any further because it doesn’t affect me.

6

That’s rather odd. Curtailing via Aux1 works only rather poorly (see the post: “Ripple control receiver…”) — presumably because the power reduction must also be guaranteed without the GX and grid meter. However, zero export can be configured in the software on the GX, with this setting protected at the user password level.

7

I suppose so. But since I waived any EEG feed-in remuneration when I registered, as a prerequisite for registering the system as a self-consumption installation (zero export), and since I have an IMSys installed rather than any kind of meter that runs backwards, it would be completely out of the question for me to configure my system for export now. So I don’t really understand your post.

8

That’s one way of looking at it—you’re not wrong.

9

By the way, I see it similarly when it comes to dimming under Section 14a or similar provisions. Via the IMSys, the DSO will be able to check whether any dimming command (if one is ever issued) is being complied with. Anyone who fails to comply will have to expect a possible contractual penalty—perhaps repayment of the discounts or something similar. At least, that’s what I assume. So I see no reason to hide this behind some kind of access restriction. I would view curtailments similarly. As so often in Germany, the drawback is the coverage of IMSys. I’ve had one for two years now.

10

Many thanks for your contributions!
Unfortunately, I was afraid this might happen. Hopefully something a bit more appealing will come along soon.

Is it just me, or do manufacturers always seem to have to implement these special requirements specifically for Germany? :grinning_face_with_smiling_eyes: But there’s no getting around it: if you want to feed electricity into the grid, you have to suffer. Or else no grid connection or zero export …

11

From day one, I planned and operated my system as a zero-export system. It’s working brilliantly, and I’m grateful for that decision almost every day. I’m anything but a conspiracy theorist, and I work with some DSOs on a daily basis, but for the time being I’ll definitely be sticking with zero export.
Let’s see what the future brings.

12

Our utility providers are a bit more radical about it … Zero export isn’t enough .. it has to be shut down.. so that you have to buy your electricity again ;O)))
Unless you disconnect from the grid and have everything connected to ACout.

13

This requires a legal basis. Are you from Germany???

14

Yes, from D.. There are already reports of the utility provider shutting everything off, leaving the customer to buy electricity.

15

In all the cases I’m aware of, the systems are technically unable to curtail the feed-in separately, but can only curtail generation. The legal situation is clear: feed-in must be curtailed. That is precisely why it is so important that Victron ensures this via the Aux input. Incidentally, this has absolutely nothing to do with a self-consumption system (zero export).

16

I’ve just looked at Section 9 of the EEG again. It only ever refers to feed-in power. That’s why I don’t understand the entire discussion you’ve initiated. If it is technically feasible, the system will be configured so that, when the curtailment signal is received, the required curtailment is actually ensured. As described above, with the MP2s via the Aux input. Done. This will have to be ensured in future and must also be ensured via the IMSys.

A self-consumption system (zero export) is a completely different matter. First of all, the applicant waives any remuneration under the EEG. Therefore, every operator already has an interest in preventing electricity from flowing back into the grid. What must be ensured is that any flow back into the grid does not exceed the technically necessary control losses. There is therefore no need to curtail anything at all.

These matters are clearly regulated in both the EEG and the EnWG. If a DNO is not aware of the legislation that applies to them, I would send them exactly that legislation. Highlighting the relevant passages with a marker often helps.

17

Apparently you haven’t understood it correctly … if you switch off the MP II via OFF … then no production takes place at all.
And if everything runs via ACin … then you have to buy electricity, since the MP II is no longer allowed to do anything.
If the loads run via ACout .. then it’s different. But not everyone has it set up that way ;O)))

18

Exactly—that’s what Holger is saying.
Curtailing via AUX must be required by law, and that isn’t the problem either.
The unfortunate thing is that grid-parallel loads on the AC input will then draw power from the grid with the current firmware.
I understood the legal situation in the same way: feeding power into the grid is to be prevented—the consumption within the house remains permitted during shutdown/curtailment by the grid operator.

19

hello,
this won’t change even with newer firmware, because the Multi doesn’t know the power at the grid connection, and there is also no meter that can be connected directly to the VE.Bus.

if your grid operator accepts throttling via the control system, you can connect the signal to the GX and control it with Node-RED.

bye

20

The legislator does not stipulate whether the Multi must be curtailed via Aux or by some other means. It only stipulates that it must be curtailed. However, some Bavarian DSOs are only happy with this if the inverter is switched off completely and the operators have to purchase their own consumption from the grid. In terms of the overall balance, it makes little difference to the amount of electricity whether less is fed into the grid or more is drawn from it instead.