Parallelization of two Victron BESS (3 Quattro + Cerbo GX) using a DEIF Controller?

Hello,
My questions are about the compatibility between Victron equipment and the DEIF system.
For now, I have one BESS (batteries + solar panel) system operating in three-phase mode (with 3 Quattro).
But I want to increased power of my installation, by adding another BESS on the same architecture.
In the end I want two independent BESS systems, each with its own VE.Bus and its own Cerbo GX. (I need both systems to remain completely independent, and for several technical reasons I cannot merge all six Quattros into a single VE.Bus.)

To gather the power of the two BESS I have to merge their output. So to do this I want to use DEIF controller. I see that DEIF controller and Victron are compatible.
But I’m not sure if the Deif controller can synchronise and parallelize two separate three-phase inverter systems, each managed by its own Cerbo GX. I’m not sure that all…
If I can’t use DEIF controller in this way, do you have other ideas about the synchronization and parallelization of 2 different Victron BESS ?

Thanks for your help,

JMN

No, it won’t. You also don’t mention which DEIF controller you are thinking of, and they make many.
If you really need two independent systems, but synchronised, then you have to run one of those systems as a grid connected ESS ( the smaller) connected to a microgrid managed by the other inverter group. This can also have an AC input for backup.
Rules for this is that the master group MUST be at least twice the size (kVA rating) of the connected system, otherwise you are likely to encounter stability problems. The smaller system does not have to be run in ESS mode. This depends on what power flows you want to achieve. If you need this system to run a single larger load than the master group can manage, then the smaller group has to be set to ESS to manage the back feed of power.

At first, thanks for you answer !

That’s what I thought…

I wanted to use the AGC150, as I understand, it’s the only compatible model (maybe with the ASC150 because they are very similar but I didn’t find information about it)

If I’m am undestand, you propose this solution :

With this solution, both ESS units are synchronized, but the ESS1 will deliver its power first (compared to the other one which will be probably in power assist mode) as a result, ESS1 will probably discharge completely before ESS2. At this moment if we need a to power a big load, only ESS2 will be able to discharge because ESS1 will already be empty.

In fact, I need a balanced discharging of the two ESS, in the same time. if ESS1 is 50%, ESS2 should also be around 50%.

I also need a perfect synchronization because I need to be able to deliver 30kVA x 2 = 60kVA (30kVA by ESS)

Did I correctly understand what you proposed?

I don’t understand this :

Thanks for your time !

JNM

Yes, that topology is correct. ‘ESS1’ does not need to be in ESS mode, it operates as the microgrid mater. Loads can also be applied to the AC out of ‘ESS1’; this would normally be the case. ESS2 can be forced to back feed power to The ‘ESS1’ loads by setting the grid setpoint. Yuo also need to select a grid code in ESS2 that allows export from DC - most of the grid codes do not.
Re sizing of inverters.
Technically the system can work if the ESS2 inverters are the same as the ‘ESS1’ inverters, this obeys the Victron 1:1 rule. However, this can cause stability issues with the voltage. It will be more stable with 2:1 so the ESS2 inverters do not cause so much perturbation of the ESS1 output if loads / sources suddenly switch. There are other posts on this forum discussing this type of Microgrid.

If you really need this level of power,then you would be better looking at a parallel inverter setup for the master, capable of delivering the full continuous power - be sure to look at the derating levels.
Synchronisation of the inverters is carried out over the VE-bus {this is why generator synchronisers won’t work} BUT AC coupled ESS systems are looking for a stable AC reference. This can be disturbed by single large loads switchinng on and off.

Since the other day, I found an idea to synchronize at least two Victron BESS (maybe more).
I remind you of my constraints :

  • I cannot use the VE.Bus to synchronize my different BESS
  • Need the combined power of several BESS operating at the same time

To explain the schematics quickly, I want to synchronize my BESS thanks to a common three phase source, the BESS needs to be synchronized with the AC1_in, so the two BESS will be synchronized on the same three phase → the AC_Out of the two BESS will also be synchronized, and I can merge the two out
I add 2 synch check relay before the merge, just un case…

In theory, I think this system will work, but in practice, I’m not sure…
I’d appreciate a second opinion

The problem with this is the power of the three phase source will need to be at least equal to that of the sum of the BESS systems attached to it. This sort of defeats the topology you originally appeared to want.

Technically, you should operate the 2 inverter systems as a parallel set, and whilst the Victron recommendation/ instruction is that the battery banks are also paralleled, as long as the negatives are tied together, then the system will operates until the first battery discharges.

Looking back to your original aim, the only way to achieve this is with the following topology:

Sorry but I wasn’t clear enough, in my idea, the three phase source is not a power source.
Only few watts are produced here (around 50w for each ESS connected for the grid setpoint)
The meaning of the three phase source is only to provide a common “three phase source”. 99% of the power needs to be produced in the paralel ESS.

I didn’t specify this, sorry for it.

In this case, is your opinion about my schematics remain the same than in your previous message ?

this is not how the Quatro’s work. You can’t synchronise 2 systems by feeding them a token power to AC in - yes they will synchronise, but then they will try to transfer the load to the source - which will instantly collapse due to insufficient capacity. Even inb power assist mode, the inverter only produces up to 50% of the load power (I think).90 - 99% is not possible.