Pylontech US3000C showing full but voltage dropping under load — normal?

Had this exact thing on my boat setup last summer. The Pylontech was reading 100% SOC but the moment I kicked the inverter on and put any decent load through it, voltage would sag noticeably — sometimes 2-3V within the first few seconds.

Turned out to be a combination of things worth checking:

Internal resistance — as these cells age (or if they've been through some rough charge cycles), IR increases and you'll see more voltage sag under load even when nominally "full". Worth running a capacity test if you haven't.

Communication lag — the SOC display can lag behind reality, especially if the BMS hasn't had a proper full-cycle calibration recently. Mine was basically lying to me for weeks.

Top balancing — are all your cells balanced at the top? Uneven cells mean one or two hit the high voltage cutoff early, the BMS calls it full, but actual usable capacity is reduced.

What's your actual resting voltage reading before you apply load? On a US3000C you'd expect around 53.2V fully charged. If it's sitting lower than that and still claiming 100%, that's your first clue something's off.

Also — what are you running it with, a Victron system? The DVCC settings can sometimes cause odd behaviour with Pylontech if the charge parameters aren't quite dialled in.

Genuinely curious what others have experienced — I suspect this is more common than people realise, particularly in marine environments where temperature swings are a factor.

@TracyMitchell that story is painfully familiar — lived it on my narrowboat for a whole season before I figured out what was going on.

What nobody tells you upfront is that Pylontechs hate being charged too quickly relative to their actual usable capacity. My US3000C was sitting at "100%" because the BMS was satisfied, not because the cells were genuinely balanced and full.

The fix for me was forcing a proper absorption hold — let the Victron MPPT sit at 53.2V until current genuinely tapered off. Took about 90 minutes longer than I'd been allowing.

Voltage sag under load basically disappeared after that.

Also worth checking: are your battery cables properly torqued? Loose terminals on Pylontechs cause surprisingly dramatic sag that looks exactly like a capacity problem but isn't.

@TracyMitchell seen this on my motorhome setup — usually points to one of two things:

1. Cell imbalance — top of charge looks fine but one weak cell drags voltage down under load
2. Communication lag — the BMS is reporting cached SOC, not real-time

Worth checking your Pylontech via the console cable and looking at individual cell voltages. If one's noticeably lower than the rest, that's your culprit.

Also — what's your cable sizing and connection quality like? I had a dodgy terminal causing similar sag before I even got to the battery itself. Victron Cerbo was showing perfect SOC right up until load hit.

Good shout from @Harry1965 on the cell imbalance front. One thing I'd add that nobody's mentioned yet — check your inter-cell bus bar connections and your main cable lugs. On my shed setup I had identical symptoms and spent weeks chasing a BMS fault before discovering a slightly loose connection on one of the bus bars. Under load the resistance was just enough to cause voltage sag that the BMS was misreading as genuine capacity. Torque everything down to spec with a proper torque wrench, not just "feels tight." Pylontech's own documentation specifies 4Nm on the terminals if I recall correctly. Dead easy thing to overlook but surprisingly common cause of this exact behaviour. Worth ruling out before you start worrying about cell-level issues.

@KateMason82 nothing like discovering your fancy LiFePO4 battery is essentially held together by a piece of wet string for a busbar — had suspiciously similar voltage sag on my cabin setup until I found a connection on the Victron lynx that had the torque spec of a soggy biscuit.