I have a 3 phase Multi RS setup with PV on L1 and L3. The BMS CVL is 53.8V, but I have seen the batteries happier at 53.6V (also specified in the datasheet as the recommended charge voltage).
I have seen that on some occasions the Multi RS ignore the DVCC limit of 53.6V and charges to the BMS CVL.
In the screenshots below, you can see that L1 is respecting the limit, but L3 not. Toggling PV in Victron Connect restores the required limits. I have seen this behaviour on both L1 and L3.
I had a similar issue and this is what I did for my setup that could help you as well
Since the Multi RS continues to ignore the GX-level DVCC limit, the most reliable fix for now is to lower the internal hardware limits on the Multi RS units themselves so they match your desired DVCC limit:
Connect to each Multi RS via VictronConnect either by using the Bluetooth app or the MK3-USB.
Go to Settings > Battery.
Set the Absorption and Float voltages to your preferred limit (e.g., 53.6V).
DVCC is a “top-down” command. If the command fails to sync, the Multi RS reverts to its internal settings. By setting the internal settings to 53.6V, even if it ignores the GX command, it will stop at 53.6V instead of the BMS limit.
I am sorry but this is not correct.
The system does not switch dynamically from external control via DVCC to the internal algo.
Please, some of your topics are leaning towards the quantity and not quality side of the content scale.
Please don’t guess or resort to the internet, it just causes confusion.
The enthusiasm is great, but accuracy is preferred.
Which is the time frame for these overshots? You don’t have a time reference on X-axis…
Although it’s only 200mV (0.3%) which is not so much and well within tolerances…
It seems that eventually Multi RS recovers…
Sorry that trend did not have a timescale, difficult to show with that historian.
The period between the peaks is about 90 minute. The duration spent above the DVCC limit in the first instance is 18 minutes and the second is 16 minutes before the MPPT backed off.
This is not a quick overshoot and it is a long time before the MPPT realised it overshot.
What I find peculiar is that one Multi RS with its MPPT knows it overshot and produces 0W, not even covering its own load on the HV DC bus. It is pulling all it power from the other Multi RS that is overshooting, all via the LV DC bus. Usually when the batteries are full, each Multi RS will cover its own AC loads from its own MPPT before reaching across the LV DC.
I maintain that this appears to be a bug, and requires at least some effort of investigation.
A systemic bug tends to be noticed and widely reported, more often than not it tends to be localised, as a community it is difficult to escalate a single issue.
What battery is this?
We tend to see that environments with DIY or unsupported batteries more frequently have issues.
If you feel you need formal support this needs to be driven through the support link or your distributor.
Try to see if all inverters have exact the same config.
Also play in DVCC with Shared voltage sense.
Also check the comms on the CAN bus, to be as good as possible.
DVCC sends update on every 3 seconds and sometimes if two consecutive settings are not the same they ignore the command.
Just shooting in the dark, as I don’t have a triphase system.
I think they still have some things to iron out.
The one is how they work out their total bank voltage.
Whats the CCL doing at the same time as the voltage shift?
Glad you have backed help. And a feesh set of eyes.
The bank was acting up prior to an RMA. Adaptive cell limits have been disabled and now CCL and DCL are constantly 600A. I am OK with that as the 3x Multi RS system is incapable of overwhelming the battery bank (granted that voltage limits are respected…)
