Grid setpoint not kept when battery is more than 85% and filling ... normal behaviour?

Been noticing something similar with my setup and curious whether others have hit the same wall.

Running a single-phase Multiplus-II 3000 with a pair of Fogstar Drift 100Ah batteries (with their own BMS comms via DVCC), and whenever the SOC climbs above roughly 85-90%, the ESS grid setpoint basically gets ignored. The system just pulls whatever it wants from the grid to push the batteries to 100%, regardless of what I've set in VEConfigure or the GX portal.

I think what's happening is that the absorption/bulk phase essentially overrides the ESS logic once the BMS signals it wants a full top-up. Makes sense from a battery health perspective I suppose, but it's a bit annoying when you're trying to keep grid draw to a minimum during peak rate periods on an Octopus Agile tariff.

A few questions for the forum:

• Is this expected behaviour in ESS mode, or is there a setting I'm missing?
• Does it behave differently when the batteries are set to charge via scheduled charging rather than just free-running ESS?
• Anyone managed to tame this with a custom DVCC charge current limit?

I've had a poke around the Victron Community forums and there are some threads hinting this is "by design," but the explanations are a bit circular. Would be good to hear from people running Pylontech stacks especially, since those seem to come up a lot in relation to this quirk.

Happy to share screenshots of my CCGX data if that helps diagnose what's going on.

@RussGreen yeah this is basically Victron telling you "I'm fine thanks, stop fussing" 😄

When the battery gets above ~85% the Multi starts winding down the grid setpoint control because frankly it doesn't need to work as hard. The DVCC algorithm essentially throws its hands up and says "job's a good'un."

@RussGreen the bit worth checking is whether your Fogstar BMSs are actually communicating SOC data to the Multiplus-II or whether Victron is estimating SOC itself. If there's no proper BMS comms (via CAN or VE.Can), the Multiplus uses its own voltage-based SOC estimate which can drift significantly — meaning your 85% trigger might be firing at a very different actual state of charge than you think.

Worth installing a SmartShunt if you haven't already, and wiring the whole lot through VictronConnect/VRM so you can see what the MPII is actually reading vs what your BMS reports. Done this on my boat and the difference before/after proper monitoring was eye-opening.

Also double-check your Grid Setpoint and Feed-in excess solar settings in ESS — sometimes the assistant overrides the setpoint behaviour automatically above certain SOC thresholds by design.

@RussGreen welcome to the forum — good first post, straight to the point.

Worth asking: have you verified what your grid setpoint is actually configured to in VE.Configure? If it's set to, say, 50W and the system decides absorption/float priority overrides it, you can end up chasing your tail trying to figure out whether it's a BMS comms issue or simply intended behaviour.

Also curious — are you running ESS Assistant or just the standard Multiplus-II configuration? The behaviour can differ quite a bit depending on which path you've gone down.

@RayWatson81 raises a fair point about BMS comms. Do the Fogstar Drifts show up properly in VictronConnect, or are you relying purely on voltage-based SOC estimation? That distinction matters quite a bit for diagnosing this reliably.

@RussGreen to add to what @RayWatson81 is getting at — if your Fogstar Drift BMSs are communicating via a Cerbo GX or similar, check whether VEConfigure has "Keep batteries charged" or "Optimised (with BatteryLife)" selected rather than pure ESS. BatteryLife deliberately backs off grid setpoint control above your configured SOC limit to reduce cycle stress.

Also worth pulling up the ESS assistant logs in Venus OS — look for GridSetpointNotKept events and correlate the timestamps against SOC. On my narrowboat setup I had almost identical behaviour until I adjusted the minimum SOC and grid setpoint in the ESS menu rather than relying on defaults.

One more thing: confirm your DVCC settings are correct for the Fogstars — if CVL/CCL/DCL aren't being respected the Multiplus will make its own conservative decisions, which is probably what you're seeing above 85%.