Victron Lynx BMS (non NG) ignores SoC reset settings

Been down a similar rabbit hole recently with my garden office build, so this caught my attention.

I'm running a Lynx BMS (older non-NG) with a set of Fogstar Drift cells rather than a native Victron battery bank. The SoC reset thresholds — specifically the voltage-based full charge detection — seem to behave oddly when you're not using a supported BMS-integrated pack. It'll occasionally refuse to reset to 100% even after a proper absorption cycle, leaving the MPPT and inverter making decisions based on a stale SoC figure.

What I think is happening:

• The Lynx BMS expects a charge current tail signal that only native Victron batteries (or properly integrated third-party packs via VE.CAN) provide
• Without that handshake, the voltage threshold alone isn't consistently triggering the reset
• The workaround I've been using is setting a tighter absorption voltage in VEConfigure and manually nudging the SoC via VRM when it drifts badly — not ideal, obviously

Has anyone found a cleaner fix? I've seen suggestions about setting the [BMS] SoC reset voltage parameter directly in the Lynx config via MPPT, but documentation for the non-NG version is genuinely sparse. The Victron community forum has some threads but most reference the NG hardware.

Would be particularly interested if anyone's got this working reliably with non-Victron LiFePO₄ — Fogstar, EG4, or similar. Feels like a gap in Victron's documentation that punishes anyone not buying a full Victron ecosystem, which isn't always realistic at UK pricing.

@EcoFlow_Queen the Lynx BMS (non-NG) is notoriously fussy about SoC resets when paired with non-native batteries — it essentially wants to "see" a full absorption cycle completed to its own satisfaction before it'll trust the reset.

With Fogstar Drift cells specifically, the voltage thresholds don't always align cleanly with what the Lynx expects at 100%. Worth checking your charge voltage is hitting the BMS's configured full-charge detection voltage precisely — even 0.1V short and it'll stubbornly refuse to reset.

Also, are you using VE.Can or just monitoring via Cerbo? I found on my boat setup that the SoC sync behaves differently depending on which comms path is active. VE.Direct vs CAN makes a real difference to how the BMS reports state upstream.

Have a look in VictronConnect under the BMS settings — there's a "SoC on restart" parameter that's easy to overlook.

@EcoFlow_Queen the non-NG Lynx BMS uses a charged voltage detection threshold internally — if your Fogstar Drift cells hit absorption voltage but the BMS doesn't see the tail current drop below ~4% of capacity, it flatly refuses to reset SoC to 100%.

Check VEConfigure → Battery settings → "Charged detection" and ensure your tail current value matches what the Fogstar cells actually pull at end-of-charge (typically quite low on LFP, around 0.5–1% C).

Classic gotcha: the non-NG firmware also has a known quirk where it ignores the reset if the charge source disconnects before the tail current window completes — so a properly sized solar array that clouds over at 99% will leave you perpetually "almost full" like my dignity after that last wiring inspection. 🔌

Ran into exactly this with my motorhome build last year. The non-NG Lynx BMS has a tail current threshold alongside the charged voltage detection @OldSailor mentioned — both conditions need to be satisfied simultaneously for it to register a full SoC reset.

With Fogstar Drift cells, the absorption tail current drops off very slowly compared to what Victron expects from their own batteries. So the BMS sees voltage hit the threshold, but tail current is still too high — no reset triggered.

Worth checking in VictronConnect under BMS settings:

• Charged detection current — try bumping this up
• Charged voltage — should be slightly below your actual top-of-charge voltage

Took me three weeks of head-scratching before I cracked it. The manual glosses over the interaction between those two parameters almost criminally.

@LiFePO4Nerd nails the tail current piece — but the bit everyone misses is that both conditions must be satisfied simultaneously: voltage threshold AND tail current drop, within the same detection window; miss either and the BMS just shrugs and carries on drifting. With Fogstar Drift cells absorbing slowly at the top, your tail current stays stubbornly high right when the voltage briefly kisses the threshold — classic non-NG trap. Try nudging your charger's absorption voltage down 0.05–0.1V so the cells reach tail current before the detection window closes; sorted mine on the boat that way.

Good shout from @OldSailor and @LiFePO4Nerd on the dual-condition thing.

One extra bit worth checking with the Fogstar Drift cells specifically — their absorption phase can be a bit drawn out compared to factory Victron batteries, so the tail current drop-off sometimes happens after the charged voltage window closes. Net result: BMS never sees both conditions met simultaneously and just... doesn't reset.

Had something similar with my garden office setup. Fixed it by slightly extending the "detection time" parameter in VEConfigure — gives the system a longer window to catch that tail current falling whilst still above threshold.

Worth also double-checking your charger's absorption timeout isn't cutting things short before tail current hits your target figure. Victron MPPT + Fogstar combo can need a bit of fettling to play nicely with the non-NG BMS defaults.