Anyone else running Fogstar Drift cells in parallel banks? Balancing woes at low SOC

Been running a 2P4S setup with Fogstar Drift 280Ah cells in my narrowboat for about 8 months now. Generally brilliant, but I've started noticing one parallel group consistently reads about 15-20mV lower than the others when the bank drops below 20% SOC. Above that it's rock solid.

Using a Daly 4S 100A BMS and wondering if it's just a weak cell or something inherent to how the Daly handles balancing at lower voltages. Passive balancing at 3.4V threshold seems quite high for LiFePO4 — surely that's only active near the top end?

Has anyone swapped to a Seplos or JK BMS on a similar build and actually seen measurable improvement? The JK active balancer looks tempting but £££ vs a budget Daly replacement is a real consideration on a build that's already stretched.

BoatMartin | 847 posts | Narrowboat & Solar

@PanelGraham I had almost identical behaviour on my 2P4S setup last season. Worth checking your busbars and connections on that low group first — even slightly higher resistance can cause a consistent voltage drop that looks like a cell imbalance but isn't. I found a loose crimp was the culprit on mine.

If your connections are solid though, 15-20mV at low SOC is arguably less concerning than the same delta at high SOC — cells naturally spread more at the knees of the curve. Keep an eye on whether it's consistent across SOC ranges or only appearing below say 20%. That'll tell you whether you're looking at a genuine capacity mismatch or just normal curve behaviour.

What's your BMS reporting for individual cell temps? Cold spots on that group could also explain it on a narrowboat over winter.

CaddyWanderer | 312 posts | Campervan & Off-Grid

@PanelGraham Worth checking the cell-level internal resistance on that low group - 8 months in, a 15-20mV spread at low SOC could simply be one cell with slightly higher IR dragging its parallel partner down under load. Grab a proper IR meter if you can, rather than relying on BMS estimates.

Also worth torquing down your busbars on that group specifically. Loose connections can cause phantom voltage drops that look like a cell imbalance but aren't. Had exactly that on my camper build - drove me mad for weeks before I spotted a slightly loose M6 bolt.

What BMS are you running? Some units struggle with parallel configurations and the balancing current is just too weedy to catch up at lower SOC where cell curves are flatter.

Simon1988 | 1,203 posts | Narrowboat & Solar

@PanelGraham That 15-20mV delta is worth keeping an eye on but probably not panic stations yet. One thing I'd specifically check before anything else — are your busbars and inter-cell connections properly torqued? On my narrowboat the vibration from the engine gradually worked a couple of connections loose over several months, and the resulting resistance difference was causing almost exactly what you're describing. Gave everything a proper retorque with a calibrated torque wrench (8Nm on the Drift terminals from memory) and it sorted it immediately. Low SOC is where connection resistance really shows itself. Worth ruling that out before suspecting the cells themselves.

MI_OffGrid | 2,156 posts | Off-Grid Systems & Storage

@PanelGraham One thing I haven't seen mentioned yet — check your busbar connections between the parallel cells rather than assuming it's the cells themselves. A slightly high-resistance joint on one side of a parallel group will cause exactly this kind of persistent low reading, especially as you get towards the bottom of the charge curve where voltage differences become more meaningful.

Torque check everything to spec and give the contact surfaces a clean while you're at it. Copper oxidation on terminals is surprisingly common on narrowboats with the humidity fluctuations you get.

If the delta persists after that, then I'd start looking at cell-level IR differences as @CaddyWanderer suggests. But in my experience, mechanical connections are the culprit more often than people expect with parallel builds.

HazelPaddy | 634 posts | Solar & Batteries

@PanelGraham One thing nobody's mentioned — check your busbar connections between the parallel groups. Even a fraction of an ohm difference in contact resistance will cause exactly this kind of persistent voltage offset, especially as you approach the flat part of the LFP discharge curve where small resistance differences become more visible.

I had a similar issue with my own bank and it turned out one M6 bolt had only been finger-tight. Torqued everything properly to spec and the delta dropped from ~18mV to under 5mV within a few cycles.

Also worth confirming your BMS is measuring at cell level rather than group level — some cheaper units average across the parallel group and you could be masking a genuinely weak cell.

Curly38 | 847 posts | Cabin, Solar & EV Charging

Had almost identical grief with my cabin bank last winter. 15-20mV sounds minor until it isn't.

One thing nobody's touched on yet — have you actually load-tested those two suspect cells individually? I wasted three weeks chasing a "balancing issue" that turned out to be one cell with noticeably higher internal resistance. Showed up fine at rest, only misbehaved under load.

Grab a proper IR meter if you've not got one. The Fogstar Drifts are generally decent quality but cell-to-cell variation does happen.

Also worth logging whether the delta grows as SOC drops below 20% or stays roughly constant. Growing delta = cell issue. Flat delta throughout = almost certainly your connections as @HazelPaddy suggests.

Methodical elimination, mate. Don't guess.

FrostyTrekker | 1,203 posts | Off-Grid Living & Renewables

@PanelGraham Worth also looking at whether this is consistent across different temperatures. I found with my parallel bank that the low-reading group behaved itself in summer but started drifting noticeably once ambient dropped below about 10°C. If one cell in that parallel group has slightly higher internal resistance, it'll show up more dramatically when cold.

Try logging the delta over a few charge/discharge cycles at different temperatures — if the gap widens in the cold, you're likely looking at a weak cell rather than a connection issue.

Wayne1996 | 412 posts | Batteries & BMS

@PanelGraham Worth checking whether that group sits at the bottom of your voltage range throughout the discharge cycle or whether it only diverges as you approach the knee of the curve around 20% SOC. LiFePO4 cells get quite sensitive down there and small capacity differences that are invisible mid-range suddenly become obvious. If it's only diverging low, your cells are probably fine — just a slight capacity mismatch between the two in that group. A few top-balance cycles might tighten things up nicely before you write it off as a connection issue.

MariaJones | 234 posts | Solar, Static Caravan & Narrowboat

@PanelGraham 15-20mV on a narrowboat is practically begging you to check your busbar connections — mine looked fine until I actually torqued them properly and suddenly my "dodgy cell" became perfectly behaved, the little drama queen. Also worth logging it through a Victron Cerbo if you've got one; watching the divergence pattern over a full charge cycle told me more in an afternoon than eight months of guessing.