Been noticing something a bit odd with my setup lately and wondering if anyone else has clocked this.
On days where I've got a decent solar forecast — say the Victron VRM is predicting a solid charge cycle — the consumption estimates seem to behave differently depending on whether I'm actually in the charging window or just looking ahead at the forecast.
Specifically: when the battery is actively charging, it looks like my system is attributing general household loads to the PV source in the forecast view. But then when you look at the actual data afterwards, those same loads were clearly being pulled from the grid during that period. So the forecast was painting a rosier picture than reality.
My setup is a fairly typical garden office arrangement — Victron Multiplus-II, couple of Fogstar Drift 100Ah lithiums, and a 600W array on the south-facing roof. Nothing exotic.
I first noticed it on a bright April day when I had a big load running (not EV-scale, but a decent draw) while the batteries were mid-charge. The forecast said "all good, PV covering it" — actuals told a different story.
My suspicion is the forecasting logic doesn't properly account for simultaneous high consumption during a charge cycle. It assumes PV covers consumption first, then dumps surplus into the battery — but real-world prioritisation doesn't always shake out that cleanly, especially when the BMS is doing its own thing.
Anyone else seen this pattern? Particularly curious if those of you with EV charging thrown into the mix have had wilder discrepancies — that must make the forecasting almost meaningless at times.
