Victron 3 phase Multiplus-II 6k5, 46 kWh DIY (Gelderland, Netherlands)

Anyone else ditched the BMS SOC readings entirely in favour of a Victron SmartShunt, or is it just me treating my JK like a very expensive on/off switch?

Running a similar setup on my boat — Fogstar Drift cells, Victron Multiplus-II, and the JK BMS doing its protection duties whilst the SmartShunt does the actual thinking. The JK SOC wanders off like a confused sheep the moment you look at it funny, but the SmartShunt stays remarkably honest once you've nailed the charge efficiency factor.

Curious whether the Netherlands setup above is seeing the same drift issues I had early doors — mine was genuinely reading 67% whilst the SmartShunt was calmly reporting 41%. Not ideal when you're trying to work out if you've got enough grunt to run the kettle without embarrassing yourself.

Few things that sorted it for me:

• Proper tail current calibration on the SmartShunt — took three full cycles to settle
• Ignoring the JK app SOC completely (out of sight, out of mind)
• Setting Victron DVCC so the SmartShunt is the trusted source throughout the system

The 13-metre cable run to the panel is interesting too — anyone factored in voltage drop at that distance when sizing their DC cabling? On the boat I'm paranoid about every metre of run, but a garage install with 6k5 of inverter pushing serious current could get spicy if the cable spec is marginal.

What BMS are others pairing with Victron kit and actually trusting for SOC, rather than just using as a glorified fuse?

@BorderWanderer that's exactly how I ended up running things on my boat! The JK does its job protecting the cells but I genuinely trust the SmartShunt far more for day-to-day SOC accuracy.

One thing I've been curious about though — do you find the SmartShunt's charged voltage threshold needs tweaking when the BMS occasionally nudges things slightly? I've had a couple of instances where my SmartShunt thought it was at 100% a bit earlier than expected and I'm wondering if my absorption settings need revisiting.

Also, did you go through Fogstar for your cells or source them elsewhere? I'm about to spec out a second bank for the shepherd's hut I'm building and trying to decide whether to stick with the same supplier for consistency.

@BorderWanderer my JK BMS has basically become the world's most overqualified bouncer — just standing there looking intimidating while the SmartShunt actually runs the guest list.

@BorderWanderer absolutely the right approach in my view. The SmartShunt's coulomb counting is simply more accurate for day-to-day SOC tracking than BMS voltage-based estimation, particularly at the flat part of the LiFePO4 discharge curve where voltage tells you almost nothing useful.

What I'd add — make sure your SmartShunt's charge efficiency factor and Peukert exponent are properly calibrated for your specific cells. Fogstar Drift cells in my motorhome setup needed a few full cycles before the SmartShunt settled into consistently accurate readings. The defaults aren't always optimal out of the box.

The JK handling cell-level protection whilst Victron's DVCC manages the actual charge parameters via VE.Bus is a genuinely elegant division of labour. Best of both worlds really.

Have you integrated your JK into Venus OS via the RS485 adapter? Worth doing — gives you cell-level voltage visibility without relying on the BMS SOC figure.

Ha, @BorderWanderer — guilty as charged here too! My JK gets its moment of glory during charge/discharge limits and cell balancing, but for actual SOC I trust the SmartShunt completely. One thing worth mentioning that nobody's touched on yet: make sure your SmartShunt's charged voltage threshold and tail current settings are dialled in properly for LiFePO4, otherwise it'll never properly synchronise and you'll see drift creep in over weeks. I spent ages scratching my head wondering why mine was reading 94% at what was clearly a full charge before realising the defaults were set for lead-acid. Once I sorted those parameters it's been rock solid for months. The Victron ecosystem talking to itself via VE.Smart networking makes the whole thing genuinely satisfying to monitor. What capacity are your Fogstar cells?

Really interesting thread — I've been wrestling with the same question on my motorhome build and for emergency backup at home.

One thing worth flagging that nobody's mentioned: synchronisation points matter a lot if you're relying purely on the SmartShunt. When cells hit top or bottom balance, that's your golden opportunity to let the BMS reset the shunt's reference. Without occasional sync events, coulomb counting drift can get surprisingly bad over months, especially if you're not regularly cycling the full SOC range.

Do any of you with JK setups have the BMS talking to Victron via CAN or UART? I've been looking at whether the JK's own SOC could act as a cross-check rather than being ignored completely — curious if anyone's actually got that working reliably with a Multiplus-II, or whether the integration is still a bit temperamental?