ESS CVL wie einstellen um Abregelung zu starten

Been wrestling with this exact scenario on my static caravan setup recently, so thought it'd be worth opening it up for discussion here.

Running a Multiplus-II 48/3000 with a SmartSolar MPPT 150/35 and a Cerbo GX managing the whole lot. Battery bank is Fogstar Drift 48V cells configured in the VEConfigure ESS assistant.

The bit I keep second-guessing is the Charge Voltage Limit (CVL) parameter in the ESS settings — specifically when it actually triggers the MPPT to back off and start curtailing charge current. My understanding is:

• CVL is broadcast over the VE.Bus/DVCC network as the upper voltage ceiling
• Once battery voltage hits CVL, the MPPT should ramp down to maintain that ceiling rather than push past it
• But in practice I'm seeing the MPPT occasionally overshoot by 0.2–0.3V before it reacts

Is anyone else seeing this lag? I'm wondering if it's a DVCC polling interval issue or whether my CVL is simply set too close to the absorption voltage, leaving no headroom for the system to respond cleanly.

Also curious — for those running LiFePO4 without a proper BMS-to-Cerbo comms link (i.e., no CANbus, just manual DVCC settings), how are you determining your CVL value? I've been setting it 0.2V below the BMS high-voltage cutoff, which feels conservative but I've no real benchmark to compare against.

Would be genuinely useful to compare configs. What DVCC and ESS charge parameters are others using on similar Victron setups?

Hey @Cerbo_Geek, interesting one! Just to flag though - your thread title's in German (CVL = Charge Voltage Limit) which might explain why it hasn't had much traction yet on an English-speaking forum. Worth editing perhaps?

On the actual topic though, with ESS on a Multiplus-II you'll want to look at your DVCC settings in VenusOS - specifically the CVL being broadcast from your BMS (or set manually if you're running a non-managed battery). The MPPT will throttle back once that ceiling's hit via the shared voltage sense.

What battery chemistry are you running? That'll determine where you want to set the CVL to trigger curtailment properly. LiFePO4 behaves quite differently to lead-acid here. Also worth checking your ESS assistant version is current - there were some quirks in earlier builds around this. 🙂

@Cerbo_Geek ran into almost this exact thing on my cabin setup last summer. The CVL from the BMS via DVCC is supposed to throttle the MPPT smoothly but in practice I found the SmartSolar was a bit slow to react and you'd get brief overvoltage spikes before it caught up.

What sorted it for me was tightening the absorption voltage in the MPPT profile to sit just below the CVL ceiling — maybe 0.2V headroom — so the charger was already easing off before the BMS even needed to intervene.

Also worth checking your DVCC "Shared Voltage Sense" is actually enabled in the Cerbo settings, not just assumed. Caught me out for weeks that one.

What BMS are you running? Makes a difference to how aggressively it pulls the CVL down.

@Cerbo_Geek the bit that catches people out here is the MPPT response lag — even with DVCC and CVL set correctly, the SmartSolar needs a moment to ramp down once the Multiplus signals it's approaching absorption. If your BMS is sending a hard CVL drop too aggressively, the MPPT briefly overshoots before ESS can throttle it.

Worth checking in VRM whether the CVL steps down gradually or drops sharply — I saw exactly this on my 150/35 paired with a Fogstar Drift 48V. Smoothing it out on the BMS side (if yours supports charge current limits alongside CVL) sorted the ghost overcharge alarms immediately.

Also confirm DVCC > Shared Voltage Sense is active, otherwise the Multiplus and MPPT are working off slightly different voltage references.

What @T6Solar said about MPPT lag is real — had exactly this on the narrowboat last winter. My Fogstar Drift 200Ah cells were hitting absorption before the SmartSolar 100/30 had properly wound back its charge current. The boat's BMS was screaming CVL at the Cerbo but the MPPT just kept pushing for a few seconds longer than comfortable.

What sorted it for me was tweaking the DVCC maximum charge current slightly lower than the BMS's actual limit — gave the system a bit of breathing room before the cells hit that ceiling. Think of it like leaving a stopping distance on the motorway rather than tailgating the car ahead.

Worth checking your VRM logs too — the CVL response timeline is all there if you know where to look.

Good thread, been through this myself with a similar setup.

One thing worth adding that nobody's mentioned yet — check your DVCC Maximum Charge Current setting in addition to the CVL. I found that even when the CVL was pulling the MPPT back correctly, having the max charge current set too generously meant the system was still pushing harder than the cells wanted to see at top-of-charge. Tightening that up gave me noticeably smoother absorption behaviour.

Also worth logging your VE.Bus state in VRM over a few charge cycles before tweaking anything — gives you a proper picture of where the transitions are happening rather than guessing. @Ozzy8's point about the Fogstar cells is interesting too; I've noticed LiFePO4 packs vary quite a bit in how quickly they communicate back through the BMS during that final push. 📊