"MultiPlus-II 48/6500 drops AC-OUT when grid comes back, anyone else?

Q&A and troubleshooting
, , , ,
1

My name is Giuseppe Basilio Rudiger, from Chile. I did my first ESS installation as a backup system for an office, MultiPlus-II 48/6500, Cerbo GX, Pylontech Fidus 16 kWh battery, load around 450 W, no feed-in to the grid.

Since the beginning of the installation (July 13) the AC OUT 1 output has shut down about 7 times, at different periods of time, always close to a grid disturbance. It lasts about 2 minutes and the Cerbo reboots every time, and also there are records of clean grid outages that it transfers perfectly, even one of 2 hours.

I tried firmware v557 to v560, config with and without ESS, a voltage regulator upstream, battery firmware twice, ground relay measured and correct. And the most drastic thing was changing the inverter for another one from the same batch HQ2537 and the next day it failed the same, this time when the grid came back after a real outage.

My hypothesis is that the problem seems to be the BATCH or the model, and not my installation. This week I will install a MultiPlus-II 5k, to see how it behaves.

Has anyone with a 48/6500 from that batch seen the same with ground relay enabled and v560?

MultiPlus-II tested = (S/N (HQ2537GQ4VQ and HQ2537MEU7E))

I have data (my own logger at 1 second, separate from the Cerbo, that is why you see what VRM does not record), CSV, portal and events recorded. I can send by private message.

Regards

01_event7_grid-return-collapse
01_event7_grid-return-collapse1400×896 82.7 KB

“Grid outage, 4 minutes inverting fine, collapse when the grid comes back”

2

Sounds like your batteries are tripping in some of the incidents, indicator is the GX rebooting.

Are you using a particular grid code on the inverter?

3

I think the same about the DC bus dropping. I have my theory: in the VRM download logs the “Data partition free space” appears and it is only when the Cerbo GX boots. I tested it by switching the Cerbo GX off and on by hand and it appeared.

I checked the BMS and it shows no problems, even Pylontech staff checked it. In the 7 events the battery was at rest, close to 0 A. It only opened with grid present during the update, and nothing happened there. It seems it fails in the transition.

About the grid code, no feed-in = “None” in VEConfigure, that is “Other: not compliant to any grid code standard”. No feed-in to the grid, backup only. Transfer switch at 200/207/240/245 V, UPS function enabled, ground relay enabled.

There is this Victron notice: VE.Bus firmware v560/v562 - fixes for some 48V models (HQ2518 - HQ2611) - Victron Professional (Professional portal, needs login). It is from March 2026, it says the affected systems “would briefly shut down”, serial range HQ2518 to HQ2611, and that it was fixed with firmware v560. The serial numbers match both units I tested. I already had v560 installed and the last 3 events happened anyway.

Do you know if any grid code or setting changes how the MultiPlus-II reconnects to the grid? Or if you have had particular problems with this model?

4

I use grid code other. The behaviour between grid codes is very different each has a standard.
It is most interesting that it is on grid return.