Restrict battery for household use only

[Palermo (B.Seesink)]

Hi, I have a question about my setup. Because I experience some strange behavior in my opinion. I also think it is a small change in settings (at least I hope so)

I have installed 3 Victron’s (Multiplus-II 48_3000_35-32) for a 3-phase home system. Cerbo is connected to the victron and the battery. The cables of the 3-phase are connected to the AC-in port of the multiplus. ESS is installed. And actually everything seems to work. During the day the battery charges to 94%. But when the sun goes down, it changes The battery delivers 1000W (sometimes more) back to the grid until 40% of the battery is reached. But that is configured in the cerbo (minimum SOC (unless the grid fails) is set to 40%).

In the meter cupboard I have the Victron VM-3P75CT that communicates with the Cerbo via UTP network.

What should I configure to prevent the battery from discharging at night.

I hope someone can help me.

Greetings
Bavo

[dognose]

@Palermo

back to the grid until 40%

Back to the grid or to your AC consumers?

Your screenshot shows regular consumers on AC IN?

If you only want to have power provided to AC OUT, you can change the Selfconsumption Mode of ESS from “All Loads” to “Critical Loads”.

[Palermo (B.Seesink)]

Hi dognose,

@dognose

Back to the grid or to your AC consumers?

Yes, For clarity. I want it to AC customers in other words house usage.
And indeed the screenshot AC loads is the AC-In port of the Multiplus-II time was around 10 pm

I have nothing connected to the AC-Out1 and the AC-Out2.

[dognose]

@Palermo

And indeed the screenshot AC loads is the AC-In port of the Multiplus-II time was around 10 pm

Ok, but wheres the Issue then?

If there are 1324 Watts of consumers, the battery will discharge by around that amount - isn’t that, what you want by saying:

@Palermo

I want it to AC customers in other words house usage.

[Palermo (B.Seesink)]

@dognose

If there are 1324 Watts of consumers, the battery will discharge by around that amount - isn’t that, what you want by saying:

Yes, I understand the confusion, Sorry
I have a P1 meter in the meter cupboard. This meter is connected to the grid counter.
This shows that 629 watts and sometimes 1300 watts are returned to the Grid. The screenshot is in Dutch. But it shows what I mean.

To clarify, L1 and L2 shows a green curved bar. This means power to the grid.
Why is it doing that. May be a setting or is It strange behavior?

[dognose]

@Palermo

Why is it doing that. May be a setting or is It strange behavior?

You are maybe a bit confused, because you have enabled Multiphase-Mode: Sum of all phases:

The Inverters will balance themselfes in a way, so that:

• Each inverter pulls the same from the DC Side
• Sum over all Phases is “0”

The later may be totally fine with L1=200; L2=200; L3=-400 (Sum is 0)

(Your L3 shows -2 amps, but no negative power?)

Usually this is fine - except your provider and/or personal favour asks for all phases to be 0.

Then you can enable the “Individual Phase regulation”. It comes with some drawbacks like uneven load between the multis and may even lead to situations where one multi is charging to DC, and another one is pulling from dc.

(Hence that option is usually only required and favoured when technically required)

[DuivertNL (Marc)]

Hi im from the netherlands too, your graphs are completly normal with multiphase regulation

Your net use from grid is zero that way, just as dognose stated

So no worries. Should leave it as is, whats your grid setpoint?

ESS will always try to stay at the grid setpoint but balance on grid, so sometimes its feeding back a little more, sometimes it uses a little more

If you want to stop from discharging completly at night you could use scheduled charging in the ESS settings

[Palermo (B.Seesink)]

Indeed I have “Total of all phases” selected. Your explanation is clear to me I think.

@DuivertNL My “Grid Setpoint” is set to 20W

If I look at the screenshot which I just made. (21:00/(9PM))

Sorry for the large screenshot. I’ve placed 2 browsers side by side.
In this case:

And the battery is not even Discharging.
The sum of all phases is 79W, is this acceptable with the Grid Setpoint (20)?
According to the P1 meter this is 1941W pulling from the grid.

Also the numbers are not matching Maybe L1 does due to time variance.
But the L2 and L3 not even close.

Maybe I’m overthinking this

[dognose]

@Palermo

The sum of all phases is 79W, is this acceptable with the Grid Setpoint (20)?

That depends on a huge amount of factors, and it will be “somewhere around 0”, but may also go away for some seconds, until the MP / Batterie has adjusted it’s power. So, 79W doesn’t look problematic in the first place, at the end of the day, you may end with something like 0.3 kWh beeing pulled from grid.

@Palermo

According to the P1 meter this is 1941W pulling from the grid.

Then there is something different happening there as well. Usually the maths here are the following:

• Grid Values are measured by the meter.
• The MP measures it’s own AC-IN-Values
• The difference of these two values MUSST be AC-Load.

So, no matter what: The Grid-Readout on the cerbo and grid meter SHOULD be the same. If some phases are swapped, this would lead to weird AC-Loads, but you should be able to tell that from the grid values exactly, which is not the case here.

These half-round circles are from your grid-providers meter, do I understand that right?
Is there something between that meter and the victron meter?

If not, either of the two devices seem to have an issue. You would need an external energy meter to figure out who’'s right and who’s wrong.

If you could draw a schematic how everything is connected that would be very helpful.

Just to make sure, in the ESS-Settings, you have selected “Metering Method: External Meter”?

@Palermo

I have the Victron VM-3P75CT

Dunno about this meter, but for a lot of meters of this kind, you have to take care, that you connect the sensors that attach to the cable to exactly the port (on the meter) that is calibrated for this unique sensor.
Check the manual and/or paperwork you received along with the device, if there is such a constraint as well. (Then the ports are calibrated to the very specific sensor and do not deliver correct results when one of the other two sensors are connectd.)

Since L1 seems fine, you may have swapped the sensors for L2/L3 even if attached to the correct phase.

[Palermo (B.Seesink)]

First of all, thank you for your help so far. I really appreciate it.

I live in the Netherlands. Here we have a meter with a P1 port. This meter is owned by the energy grid operator. I can’t change anything in it, this meter is disconnected. The energy companies read this meter via transmitters.
the information from the P1 port is the same information that the energy companies receive.
This P1 port is connected to an RBPI with software on it that reads this meter. This has been on my system for years and also matches the data that the energy company uses for the bill.

I have the Victron VM-3P75CT energy meter in the meter cupboard with the CT terminals on the 3 phases that come from the grid operator. And the phases that go to the energy meter are connected to the distribution blocks in the cupboard.
In the Cerbo ESS configuration, the Grit metering is set to “External Meter”
I would expect that the Grid values ​​should be around the same. For L1 that is indeed the case. Even when I look now, that is indeed the case. I suspect that I have to do something with the L2 and L3 terminals. Maybe they are being disrupted by each other!?

I am working on making a drawing. I do have one, but I will make it digitally, that is a bit clearer.

In the Netherlands we have a company that maintains the power grid in the streets and a company with which we have a contract. For a contract we can choose between multiple energy companies.
I do not know how that is arranged in other countries.

All devices have been updated to the latest firmware version before they are installed.

[dognose]

@Palermo

I would expect that the Grid values ​​should be around the same. For L1 that is indeed the case. Even when I look now, that is indeed the case.

If there is nothing in between - they should be very close, some timing difference, tho - but overall same readings.

@Palermo

I suspect that I have to do something with the L2 and L3 terminals. Maybe they are being disrupted by each other!?

I’ve just checked the manual, and indeed, it is as assumed:

155987-Victron_VM-3P75CT_Energy_Meter-pdf-en.pdf

Page 4: The transformers are marked with an indicator showing which input port they belong to. The devices are calibrated at the factory, and accuracy will decrease if the current transformers are not matched to the correct input.

So, double check if the sensor you are using on PHASE 2 is also marked to belong to the CT-2 input, and not the sensor that is supposed to go to CT-3 of the meter. (repeat for every phase, ofc)

Also double check the v-sense wires of the meter. Measuring the voltage of a wrong phase may lead to a wrong power-computation along with the measured current.

[Palermo (B.Seesink)]

@dognose
The end is in sight
I took another good look at the phase clamps. And indeed. they were not connected properly. I have now set them correctly.
Very good tip. Thanks

But it was still not finished. Phase 2 and Phase 3 are connected the other way around in the VM-3P75CT. I can only fix this when the power is completely off. I will do this this weekend.
The value of the VM-3P75CT and the P1 meter are almost the same.

Then I noticed that (somehow) Dynamic ESS was activated. What I understood is that this is only possible with a dynamic contract. And I don’t have that (yet). So I don’t trade with power. I have turned this part off.
The result is that from that moment on the battery was used for use in the house.

Thanks a lot for the help.
Absolutely fantastic