I’m experiencing a strange issue with my system (see video with “rasp”/“croak” noise), and I’d like to understand if anyone has encountered something similar or if there are better ways to monitor the situation.
System Configuration
MultiPlus 15kVA in ESS mode
SolarEdge inverter running in parallel (see image for schema)
ACOUT connected to the house
Monitoring via EASTRON, with measurements taken every 5 seconds
Issue Description
Occasionally (this has now happened three times), I notice that during full battery discharge, the MultiPlus enters a strange lockup state. The attached graph illustrates this behavior:
I’m writing to follow up on my previous post regarding the strange lockup behavior of my MultiPlus 15kVA with ACIN connected to the grid.
So far, I haven’t received any responses, and I’d like to share some additional information that might help shed some light on the problem.
Additional Observation from a Peer:
A friend of mine, who is using the exact same 15kVA model, has experienced the same issue. In his case, even after manually reactivating ACIN, the problem persists unless he waits for at least one hour before switching it back on. This delay seems to be necessary to restore normal operation, which suggests that there might be an underlying timing or state-reset issue within the system.
I’m writing once again regarding the unresolved issue with my MultiPlus 15000 , which appears to be a clear bug. Despite my previous posts detailing the problem, I have yet to receive any feedback or guidance from the community.
Key Points:
Confirmed by Multiple Users:
Not only have I experienced this anomaly, but a contact of mine using the same MultiPlus 15000 model has also encountered the same issue. In his case, the problem persists even after reactivating ACIN—unless he waits at least an hour before doing so.
Lack of Log or Error Indications:
The anomaly isn’t captured by the logs, nor does it generate any error or warning messages. This makes it extremely challenging to diagnose, monitor, or address using standard troubleshooting procedures.
Potentially Hazardous Voltage Fluctuations:
The issue results in dangerous and abrupt voltage swings in the home (ACOUT), which poses a significant risk. The repeated sudden and violent fluctuations could potentially damage connected equipment and create unsafe conditions.
Given the severity and potential hazards associated with this behavior, I kindly urge someone from Victron to take an interest in this matter. Any assistance, insights, or guidance on how to better monitor and resolve these issues would be greatly appreciated.
Thank you for your attention and prompt response to this critical safety concern.
Your graphs don’t show. But if this is happening while solar is producing, then it’s likely the location of your VM meter confusing things. Having it on grid only isn’t the data the Multi/GX wants to see.
As indicated by the arrow, the SolarEdge inverter was producing energy (yellow trace), while the Victron unit was charging the battery (blue, negative trace). Then, at one point, the Victron began drawing power from the grid (green trace) to ACIN (purple trace). Suddenly, the SolarEdge production stopped (probably it intercept a bad frequency or voltage), and the Victron started emitting an unusual noise—as you can see in the video—with the lights flickering on and off, reflected by the fluctuating purple trace.
This behavior is not correct. If SolarEdge stops producing, the ESS should automatically switch to battery power. In my previous post, even though the system was on battery, the Victron stopped using it (without reason) and instead drew power from the grid!
This is very strange behavior.
As just said, no logs from Cerbo, nor from batteries (Pytes V5a)
During battery discharge (blue line), at a specific point (as the consumption moved from 2 kWh to 3.5 kWh), the battery suddenly stops discharging—even though the SOC was around 60%, so there was no risk of overcharge. Instead, the MultiPlus starts drawing power from the grid, as seen by the negative purple trace (first red arrow), while the battery discharge (second red arrow) almost drops to zero.
At that moment, the unit begins emitting an unusual noise (as captured in the video, including the final “grat” sound) and the lights start pulsating. The purple trace fluctuates until I manually disconnect ACIN (third arrow). After waiting a while and reactivating ACIN, everything returns to normal, and the system resumes battery consumption (forth arrow).
As mentioned, I’m paired with another user who has the same MultiPlus 15000 issue—and he’s located in a completely different region of Italy and LOM enabled. This indicates that the problem isn’t due to local grid issues.
Any further insights or suggestions would be greatly appreciated.
Thank you in advance for your help.
Set up some monitoring in VRM with frequency and voltage. And DC ripple.
I have one ‘eh’ on your DC bus-> the 120mm² between the two bus bars is abit small (2x95 is significantly larger than 120mm²). The battery 70mm² is nice and equal to the 2x95mm².
What is your entire distance from battery bank to inverter?
Then on your AC runs what is the sized/length etc?
When this happens in an ess system, it usually points to high resistance in the feed lines.
@glfp , I would love to have a private conversation with you,
But don’t know how to private message from here. maybe a feature not yet available to me.
I’m using a 120 mm² cable for the battery connection, with a length of 2 meters. The batteries are located on the opposite side of this wall (in a warm zone ).
While I agree that 120 mm² is less than using two 95 mm² cables, I believe it should be sufficient for handling the 200A maximum power of the MultiPlus 15000—would you agree?
Distance from Multi and grid and Solaredge is 6 meters AC 10mm² cable.
I’ll try to extract the specific data from VRM that you mentioned.
However, I suspect that the sampling rate might be too low to capture all the nuances of the problem.
Do you have any suggestions on how to increase the sample frequency, or perhaps another tool we could use to obtain more detailed data on the voltage, frequency, and DC ripple?
Unfortunately on my personal monitoring tool (previous post figures) I forgot to record frequency info, so I do not have this kind of information
Other than adding it to your monitoring system and waiting again…
It does look like the input voltage and frequency takes a dive so brownout? (system tries to stay synchronous and cuts off?) . The PV inverter is agreeing with its shutdown state.
Seeing that the would make sense the sound is related to synchronisation problems.
DC rippple you can create a cutom graph on the vrm for.
It is physically measurable.
).