I have a Lynx Shunt + MultiPlus-II 48/10000/140-100. It is hooked up to a rather large battery bank. There is no comms (i.e. no CANbus, RS485 etc) between the BMS of the battery and the Victron setup (the system infers charge levels via voltage / shunt settings only). Controlling the lot is an Ekrano GX
This is a grid-connected application with PV. I get power cheaply between 2100 → 0000, so during this time I run all my loads (hot water heating, dishwasher, EV charging, whatever). I have also set up a scheduled charge to take advantage of this:
This has been working great except I am running into an irritating issue. Once the battery hits 100% and the smart shunt syncs, the battery begins discharging into my loads, and the grid draw is maintained close to 0w. I expected changing the mode of operation from Optimized (without BatteryLife) which is what I normally run it as to Keep batteries charged would do the job, but it doesn’t have the desired effect. I can also do this which achieves the desired outcome:
Because setting it below 100% was the only way I could get it to stop discharging.
Edit, just to clarify, the screenshot below shows what I normally have my scheduled charging set to; but in this state, if it hits 100% BEFORE 00:00, it will begin discharging to power loads.
Setting it to below 100% and then setting “self-consumption above limit” to “PV” only is the only way I could get it to stop discharging my battery into my loads.
“Keep battery charged” did not in fact keep the batteries charged.
The scheduler behaviour was improved over time to allow discharging. Haven’t used it in ages.
iirc the scheduler considers the schedule complete if either the duration elapses or if the SOC is reached, it then reverts to self consumption.
You would need to choose PV- only for self-consumption so the battery is not used during this time.
Way back before peak shaving etc, it would just hang on until the time elapsed which didn’t suit a lot of use-cases.
I have not had it discharge during a window set, this is why i find this interesting.
The only thing i could think of was another control that was overriding the schedule charge in some way. Like a peak shave or an input current limit etc.
I have several systems where it charges up in the window then will only discharge after the window ends. So haven’t had an issue with the feature. Am enjoying it actually i used to use a node red flow to do something similar
Just tested this myself.
With self-consumption set to PV only and an active schedule where the battery is above the target SOC, the system does not discharge battery, it just uses solar.
I drop solar, grid is used.
I change it to PV and battery, the battery takes over the load.
The problem is when you set the SoC for scheduled charging to 100%, the options to choose PV or PV+Battery disappear. I think I can live with it - now that I know this is to be expected.
For now, I’m using a couple of automations in Home Assistant.
First, I have a pair of automations - If SoC exceeds 80% during the day (i.e. sun above horizon) - it will set ESS Max Charge Current to 0 so that all excess solar is sent back to the grid. If SoC drops below 70% that change is reversed so the battery can charge.
Then, another automation which:
At 20:59:
If SoC <= 75% it turns On the ESS Charge Schedule & sets ESS Max Charge Current to 140 (so the system can charge at full rate from the grid)
If SoC > 75% it turns off the ESS Charge Schedule & sets the ESS Max Inverter Power Limit to 0 (so no charging from grid + no battery discharge)
At 00:01, the above changes are reversed:
If ESS Charge Schedule is On, it turns it off & changes ESS Max Charge Current to 140 - To get the system it ready for tomorrow to charge from PV
If ESS Charge Schedule wasn’t On, it sets the ESS Max Inverter Power Limit to -1 so that the battery can start discharging to power loads through the night.
Now my question is, what happens if the power goes out (Grid failure) when the ESS Max Inverter Power Limit is set to 0? Will this make my loads go dark as the discharge limit is set to 0?
Tested the >75% condition last night, and it all seemed to work as expected. Any critiques of the above? Happy to leave it as is for now unless anyone can spot any glaring issues…
If you have it working as you want it its fine.
If it is not behaving as expected and its not a vanilla installation then the automations have to be looked at first.
The other way to make feedback happen is freeze the soc value at 2% behind its current value (if battery life it aon) so that excess feedback happens.
Then all you do is raise the value in your charge window to 100%.
