Anyone else struggling to get accurate SOC readings with a cheap shunt meter on LiFePO4?

I've got a 200Ah LiFePO4 battery (one of the Topband ones) hooked up to a 40A Renogy MPPT and a 1000W inverter in my van build. Been using a Bayite 100A shunt meter to track state of charge, but the readings are all over the place. It'll show 80% and then the BMS cuts out under load — obviously nowhere near as full as it thinks.

I'm pretty sure the issue is that the shunt isn't being calibrated properly for lithium chemistry. The voltage-based SOC curve for LiFePO4 is so flat between about 20% and 90% that voltage alone is basically useless, and I'm not convinced the Bayite is doing proper coulomb counting either. I've tried resetting it after a full charge cycle but it still drifts badly after a few days.

Has anyone actually got reliable SOC tracking on a budget, or is it worth just biting the bullet and getting a Victron BMV-712? I've seen them going for around £90 second-hand on eBay which feels steep when the whole shunt meter cost me £18, but if it genuinely works properly I suppose it pays for itself in peace of mind. Curious what others are running.

Paul1999 replied:

@Roger1983 yeah, this is a really common frustration with LiFePO4 specifically. The problem is that the voltage curve on LFP is incredibly flat through most of its range, so any shunt meter relying heavily on voltage for SOC estimation just falls apart. The Bayite units aren't bad for what they cost, but they need a proper calibration cycle to work well — fully charge, let it sit for an hour, then zero the meter. Also worth checking your shunt connection quality; even a bit of resistance there throws the current measurement off significantly over time.

One thing that genuinely helped me was setting the Peukert exponent correctly for LFP (closer to 1.05 rather than the default which is usually set for lead-acid). Most people overlook that setting entirely. Won't cure everything but makes a noticeable difference to accumulated error.

DalesOffGrid replied:

@Roger1983 the flat voltage curve on LiFePO4 makes your cheap shunt about as useful as a chocolate teapot — current tracking drift is the real killer, so splash out on a Victron BMV-712 and suddenly your SOC goes from "educated guess" to "actually reliable." Learned this the hard way when my garden office Fogstar battery supposedly had 60% left and then faceplanted at 3am.

FrostyTinker replied:

@Roger1983 Something worth checking that nobody's mentioned yet — make sure your shunt is properly calibrated for your actual battery capacity. Most of these budget units ship with default settings that assume lead-acid charge efficiency (around 85%), but LiFePO4 is closer to 98-99%. If your meter's Peukert exponent and efficiency factor aren't adjusted, the coulomb counting drifts badly over time regardless of how flat the voltage curve is.

Also, the Bayite needs a genuine full charge cycle to "sync" its zero point. Get your Topband to 100% via the MPPT, hold it there briefly, then reset the meter. Doing this regularly keeps the drift manageable. It won't be perfect, but you'll get usable readings in the 20-80% range where it matters most for day-to-day use.