Feature request: Cerbo to allow BMS and battery monitor to be selected separately

Really glad someone else has raised this — it's something I've been wrestling with on my Vivaro build for a couple of months now.

My setup uses a JK BMS on a DIY 280Ah LiFePO4 pack (Fogstar cells), with a Victron Cerbo GX at the heart of everything. The JK does a decent job of cell balancing and protection, but I don't fully trust its coulomb counting for accurate SoC — particularly after a partial charge cycle or when temperatures drop in winter.

What I'd really like is to let the BMS handle the protective functions (cell over/undervoltage cutoff, cell balancing, high-temp protection) whilst letting the Cerbo defer to the SmartShunt for SoC reporting. At the moment, the Cerbo essentially forces you to pick one source as authoritative, which means you're either trusting your BMS's SoC figures or potentially losing some of the protection logic. Neither feels ideal.

A split configuration — BMS controls charging via DVCC, SmartShunt reports SoC — would be genuinely useful for anyone running DIY packs where the BMS is solid on protection but questionable on monitoring accuracy.

Has anyone found a workaround within the current Victron ecosystem? I've had a look through the VRM portal settings and the DVCC options, but nothing quite gets me there cleanly.

Would be worth raising formally on the Victron Community forum too — they do actually listen when enough people push for something. Anyone else running a similar hybrid approach with their Cerbo?

Same issue on my garden office build — Cerbo GX is brilliant but this one limitation is genuinely frustrating.

Running a Victron SmartShunt as my primary battery monitor because the SoC accuracy is far better than relying on the JK BMS figures alone, but the Cerbo insists on treating whichever device has the BMS role as the definitive source. You end up with the dashboard showing figures you don't fully trust.

@VivaroWanderer the workaround I've seen suggested is using dbus-spy to manually adjust device priorities, but that's not something most people should be poking around with — and it breaks on firmware updates anyway.

Victron's Feature Requests portal is probably the best route. If enough people upvote it they do actually act on these things — the DVCC improvements came from community pressure originally.

Same thing on my narrowboat. Ended up just leaving the Cerbo GX reading from the BMS and accepting the slightly dodgy SOC figures rather than fighting it.

Would be a proper useful addition though — the Victron SmartShunt gives far more reliable SOC tracking than the JK BMS does, but currently you're forced to pick one as the battery monitor source. Seems like an obvious separation to make given how many people are running third-party BMS units alongside Victron kit.

Hopefully Victron pick this up in a firmware update. They do seem to listen eventually.

Yeah this trips up a lot of people. The core issue is Cerbo picks one source for SOC — it can't blend a BMS feed for cell protection logic whilst pulling SOC from a dedicated shunt like the BMV-712 or SmartShunt.

Workaround I use on my garden office: set the SmartShunt as the battery monitor in Cerbo, but keep the JK BMS connected via RS485/Canbus so cell-level data still shows on VRM. You get accurate SOC and cell visibility without compromise.

@MuddyFisher don't just accept dodgy SOC — the shunt method above genuinely sorts it on a narrowboat setup too.

Upvoted the feature request though — Victron should allow independent selection. Would be cleaner than workarounds. Worth posting it on their community forum directly, they do actually listen.

Has anyone tried using a SmartShunt alongside the JK BMS and just pointing the Cerbo at the SmartShunt for SOC whilst keeping the BMS connected purely for cell-level protection data?

That's the setup I'm considering for my build — basically treating them as two separate jobs. SmartShunt handles the coulomb counting, JK handles cell balancing and protection trips.

My concern is whether the Cerbo would still trigger low-voltage alarms based on BMS cell data even if the SmartShunt is the designated battery monitor, or whether those two data streams stay properly independent. Does anyone know if the DVCC settings affect this?

@CE_Builds your point about a single SOC source makes sense architecturally, but surely Victron could allow alarms and protection logic to pull from a different source than SOC? Seems like a relatively clean separation to implement?