DVCC and battery cell voltage balancing - 16S Sunwoda batteries not achieving balance

Been wrestling with this one on my cabin setup for a few weeks now and curious if others have hit the same wall.

Running a 48V bank with 16S cells and DVCC enabled through my Cerbo GX — the BMS is talking to the system fine, all comms look healthy, but the cells just aren't getting the opportunity to properly balance out. Top-end cells are sitting noticeably higher than the laggards and I can't seem to coax the system into holding at absorption long enough to let things settle.

From what I've read on the Victron docs, DVCC is supposed to hand charge control over to the BMS entirely — which sounds great in theory, but in practice it means your BMS dictates when to stop, and if it's cutting off before weaker cells have caught up... you're a bit stuck.

Few things I'm wondering:

• Has anyone managed to tweak balancing behaviour via the BMS settings directly rather than the Victron side?
• Is there a case for temporarily disabling DVCC to force a manual absorption cycle and let passive balancing catch up?
• Anyone running Sunwoda cells specifically who's cracked this?

My instinct is the cells need a few deliberate full cycles with a bit more headroom at the top end, but I'm wary of pushing anything without a better understanding of what's going on under the hood.

Would be good to hear from anyone who's gone through this — especially with larger 16S configurations. Feels like one of those things where the documentation only gets you so far and real-world experience fills the gaps.

@LochChild had almost identical grief last winter on the boat. 16S Sunwoda pack, Cerbo GX, DVCC switched on — cells sitting stubbornly unbalanced at the top end no matter what I tried.

The thing that cracked it for me was checking whether the BMS was actually asserting a CVL (Charge Voltage Limit) low enough to let the cells sit at top-of-charge long enough to balance. My BMS was cutting charge requests too early — Victron obediently stopped pushing, but the balancers never got their window.

Bumped the absorption time in the BMS parameters and set a slightly lower CVL ceiling so the pack absorbed more gradually. Gave the active balancer time to actually work.

What BMS are you running alongside the Sunwodas? That detail changes everything here — some communicate CVL properly via CAN, others are dumb as posts.

Classic 16S drama — my van build taught me that DVCC's "Shared Voltage Sense" can quietly fight your BMS if the Cerbo's firmware isn't current, so check Remote Console > Settings > Firmware before anything else because Victron silently fixed a 16S balancing edge case in a recent release that had me pulling my hair out for a fortnight.

Good shout from @LazyRanger on the voltage sense conflict — worth adding that your charge voltage ceiling in DVCC needs to account for genuine top balancing headroom. If your absorption voltage is set even slightly conservative, the upper cells never get the chance to bleed off properly through natural balancing. I had a similar situation with a 16S pack and found bumping the charge voltage up by about 0.05V per cell made a noticeable difference, whilst keeping the high voltage disconnect in the BMS as the real safety net. Also worth checking whether your BMS has active or passive balancing — Sunwodas are typically passive, so they genuinely need sustained time sat at top of charge to sort themselves out. Don't rush the absorption phase.

Great thread to land on as a new member, @LochChild — welcome to the forum, you've clearly come in with a proper technical question rather than just "why isn't my solar working" 😄

One thing nobody's mentioned yet: check your cell capacity spread. If individual cells have drifted apart in actual capacity (not just voltage), no amount of DVCC tuning will achieve tight balance — the BMS is just chasing its tail. Worth doing a slow top-balance outside the system if you can safely disconnect the pack.

Also, what firmware is your Cerbo running? There were some meaningful DVCC improvements in recent VenusOS releases that specifically helped multi-cell balancing behaviour.

Had this exact headache in my static caravan — turned out my Sunwoda cells were "balanced" according to the BMS but the Cerbo was still throttling charge current before they could properly top-balance, because DVCC's absorption voltage was set a touch low for 16S geometry (3.45V × 16 = 55.2V, not the default 54.4V most people leave it at). Bump that ceiling up fractionally and give it a proper full cycle or three — mine sorted itself out without touching anything else. Worth double-checking your BMS cell overvoltage protection threshold isn't closer to that ceiling than you'd like though, or you'll be having a very different conversation with your Victron kit. 🔧

My Fogstar Drift cells pulled the same stunt — BMS swearing blind everything was fine while one rogue cell was quietly living its best life at 3.62V whilst the rest lounged around at 3.48V 😅

Worth checking whether your DVCC SVS (Shared Voltage Sense) is actually enabled or just claiming to be — mine showed ticked in VRM but the Cerbo wasn't actually pushing the corrected voltage through to the MPPT until I toggled it off and back on again.

Also, at 16S you really want your absorption ceiling set to exactly 58.4V — anything vague like 58.0V and the tail current never triggers properly, so the balancer in the BMS barely gets a look-in.

@LochChild on my boat I solved the suspiciously cheerful BMS problem by enabling SVS (Shared Voltage Sense) alongside DVCC — the Cerbo then cross-references what the BMS claims against what the Victron SmartShunt actually measures, and suddenly the lies become very obvious very quickly.