Our customer has had an ESS system installed for several months (Quattro 15kVA, 16 x Pylontech Batteries, 2 x 450/200 SmartSolars, Grid Series for failover, no energy meter). It was charging and discharging correctly for months, last week it stopped discharging the batteries below ~90%. Below is a graph showing it gradually reducing the battery discharge (there was plenty of solar and optimized without battery life).
The unusual behaviour initially appeared to be: During the day the PV generation exceeded the AC loads so the batteries remained at 100% and the SmartSolars reduced their output to match the AC loads, as DC Grid Feed-in had been disabled. Once the PV stopped generating, the battery would only discharge to ~90% and after a few days only 98%, and took all the AC power from the grid.
After updating all the firmware and tweaking any software parameters I could think of (enabling DC Grid Feed-in for example) we were able to get the system to discharge to 60% but in the middle of the night it stopped discharging and started drawing from the grid once again to recharge the batteries (not Sustain mode, ESS Mode is Optimized WITHOUT Battery life).
I asked the customer to disable the PV (isolated the SmartSolars) during the day and the battery did eventually discharge to 20%. However the 15kVA Quattro would only output about 6-7kW and the rest came from the grid, rather than the full 12kW output. The Pylon batteries have reported max temperature of 28 degrees so the internal room temperature won't be too high in the UK. I realise the Quattro could be a lot warmer, but the customer put a thermometer on top of the Quattro and never got more than 26 degrees.
When the PV is reactivated, for a short time it outputs full power and supplies the AC loads and excess feeds the batteries. However after a few minutes it reverts to limiting the PV and restricting the AC output of the Quattro. I've even had a situation where the PV appears to be charging the batteries but the Quattro is feeding all the AC loads from the grid.
After updating the Cerbo to v3, I was asked to update the MK3 in the Cerbo, which I did. This then flagged Error 14 Device Transmit Error which I cleared and the system worked correctly for a few minutes, then reverted to limiting PV and AC Out. Replacing the VE.Bus cable hasn't helped.
What seems to be happening is the system is limiting the solar output and even the Quattro output, some sort of logic error with the system. The batteries do not seem to be responsible. Disabling the solar generation allows the batteries to be discharged to their set floor, but I don't think the SmartSolars are responsible for this. My instinct makes me think the Cerbo or the Quattro are responsible.
I attempted to run the system without the grid connection, off-grid, without the solar working. The Quattro ramps up and supplies the loads but it almost immediately triggers the Temperature and Overload alarms with what the Quattro measures as 10kW AC load. However the Pylon batteries were reporting in excess of 12kW of DC discharge current. This seems odd as there shouldn't be 2kW loss in the Quattro, so I suspect there could be AC current measurement errors. There may also be temperature sensing errors.
I also noticed a strange voltage spike when switching to stand alone mode.
Grid feed-in is enabled, no export limitation set, Optimized Without Battery Life (set to 20% discharge level).
The Quattro is set to 100A AC current limit
ESS is G99, Pylontech recommended settings.
The Pylontech batteries have reported no alarms or warnings. Cells are always well balanced, all modules online, Discharge Current is 500A+ always from Parameters.
I've had a look at the Quattro derating document and I don't think the temperatures have been reached. How does the ESS software temperature de-rate the Quattro?
Has anyone come across this weird ESS logic before? It keeps changing so I think there is an issue with the Cerbo or Quattro which firmware hasn't resolved. Perhaps the Quattro Current Sensor, or the Temp Sensor, or VE.Bus coms.
I've put together some custom monitoring graphs showing the system switching to the grid despite there being battery and solar available.
I'm at a bit of a loss trying to get this system to keep functioning correctly. I can get the Quattro returned for testing but it is frankly easier to perform these tests while the customer has it set up and running, rather than build a new rig. I will eventually do this to rule out some variables but I was wondering if there were any suggestions from the community. Thanks, sorry its a long one.
