I have a problem with our 3-phase system. Briefly about the setup:
3* MP2 5000 (ESS with feed-in)
1* MPPT RS450/200 (10 kwp)
1* 15kWh Seplos V1 battery
1* Hoymiles HM-800
1* Hoymiles HM-600
1* Hoymiles HMS-1600T
Load and microinverter are connected to AC-out1, grid is connected via ACin.
At irregular intervals (currently once a day) the entire power supply in the house fails. This usually happens in the evening. There are usually no large consumers running at this time.
No error is reported in the VRM portal.
The following error message is then displayed in the detail log: VE.Bus Error 1: Device is switched off because one of the other phases in the system has switched off (in logfile time around 18:45)
I can also see that it increases the frequency significantly. The micro inverter also reports a frequency overrun at this point (see screenshots).
This problem only occurs in the summer months. I have not yet been able to observe it in winter.
I have already checked the wiring (DC / AC / communication).
I hope you have an idea what the problem could be. I am attaching the log and a few screenshots.
Why does the MP have to disconnect the grid to adjust the frequency on the AC-Out side?
The Battery seems to look good - there are no Errors or other indicators.
No, here are the Firmware Versionen i used:
MultiPlus-II 48/5000/70-48: v552 (2623552)
SmartSolar MPPT RS 450/200: v1.22
Cerbo GX: 3.60 (updated today)
Because it has to synchronize with the grid to feedback. So it would usually be at grid frequency.
They only frequency shift when offgrid.
To control when ongrid then communicating with the sunspec protocol is done with those that are compatible.
There has been alot of background work on it recently to open up the system to other PV inverters, that is what the new Firmware v3.6 does.
As a further thought, if it can frequency shift the grid then you have a very high resistance connection.
Can you see the hoymiles production on the GX?
Do yourself a favour and switch off auto updates on the GX.
Third party modifications on new GX code can be a challenge as they rarely maintain the mods to keep up with changes and can create problems.
Typically the advice is to try run the system without mods to see if it stabilises.
There are some points in it that I can’t explain.
The failure occurred on 2025-06-10 21:35:01 (i.e. at a time when there was no longer any significant solar yield, so it can’t have anything to do with a frequency shift to regulate the microinverters down).
Nevertheless, the MP pulls up the frequency at AC-out1 to 52.9 Hz → why does it do this? At the same time, a frequency of 50.1 Hz is present at AC-in.
It can’t frequency shift the grid side. You system is not powerful enough to do that.
It especially wont be able to if it is disconnected.
It is disconnected because it does not want to be synchronizating to the suddne up and down in frequency tha you are seeing in the logs (it is viewed as instable)
The problem is what ever is on the grid or the grid itself.
