Here is a new one to understand.
After having set my Multi so that it will ignore AC in, unless otherwise instructed, today I discoverd that it is allowing 97W to pass through God knows where to🤔
How is that possible? How can I find out where this energy is going?
Are you able to confirm it is connected to grid?
Is the 97W self consumption maybe showing on the input?
Can you share a screen shot of what you are seeing?
Here is what the Cerbo shows
And here is my grid sensor
The grid on-off contactor is powered by the inverters so it can’t be that
Strangely enough, if disconnect AC-in from the slave Multi, the reading drops from 97w to 64w
This is the system in charge mode
Obvioustly I use the Cerbo generator start for the grid AC-in
I wonder if there is some form of feedback or other creating the confusion, the fact is that if I switch off the Grid-in contactor, which is only feeding the two inverters AC-in, the abnormal reading disappears showing 0 as it has always done before.
Are you really switching off BOTH AC In ?
What happens if you disconnect the mains connection or pull the fuse ? does the value change ?
Both AC come from the same contactor so when that is on the off position there is no power to both inverters.
I even tried to physically disconnect the wire and again the reading went to 0, as soon as I reconnected it it came back to 97.
It’s as if something was in between the AC-in and whatever disconnects by ignoring the AC-in.
Perhaps that internal relay?
Maybe Victron uses the AC-in as a feed to the relay that connects it to the internal circuitry?
you switch the Phase AND Neutral off ?
If you don’t switch off the internal relays, a current could really flow between the multis as compensation
yes it’s a DP DT Contactor
I’m not sure I understand what you mean here 
What internal relays? Both inverters have the same assistants. Do you mean between the two inverters? Used by what? This 97w must be absorbed by some device, it can’t just disappear in thin air, can it ?
Simply disconnect ONE AC In to rule this out …
When I did that physically the system halted returning the following message:
No AC on slave
but only for testing !!!
I can try again, perhaps now that the slave has the “ignore AC-in” it might not halt the all system
You see, I did not even want to connect AC to the slave since I was advised that only the master can be used for charging but when the error appeared I was forced to.
I still have no evidence that the slave is taking any part in charging but I can check with an ammeter
Normally, your internal relays should automatically disconnect the connection to the AC In terminals … if you ignore the AC in …
There should therefore be no connection from the master to the slave on the AC In side … But we have to start somewhere
Switch off the mains and disconnect an AC In …
Which internal relays? We’ve always discussed that AC flows automatically as soon as it’s available so if there is a relay to stop it from reaching the circuits it must be of an NC type which only opens under request by “ignore AC in” correct ?
The connection exists externally because when I did not have it I was getting the error previously mentioned.
I agree, the internal relays, whatever they are, should disconnect both ACin under the “ignore ACin” state unless the relay is after some other circuit.
so these are the relays working in conjunction with "Ignore AC in’?
By the schematic, they would be NO and activated by the current sensor
Yes … NO-Relais… They close if the ESS Relais Test is positiv … “Klicker klicker” 
Again to the net measurement above
Where exactly did you measure this small power ???
I read the 97w in my Grid meter as soon as I connect the AC to the inverters. I have temporarily introduced a way to manually interrupt when not charging otherwise I would unnecessarily be drawing that power from the grid even when I’m running off-grid
why don’t I try what you suggested first by attempting to remove AC into the Slave ?
I can do that in the morning and report
