DESS seems to be broken... not charging the battery

Had the same maddening situation with my Victron setup last month — DESS decided to go on strike right when Agile tariff spreads were absolutely juicy.

Turned out my minimum SOC floor was set to 20% in VRM, and DESS was technically "happy" leaving it there because it didn't think the economics justified topping up further. Cheeky algorithm.

Few things worth checking if you're in the same boat:

• BMS comms — if your BMS (I've got a Fogstar Drift with a JK BMS) isn't talking to the Cerbo properly, DESS loses confidence in the battery data and basically sulks
• Grid setpoint — sometimes mine drifted and DESS thought it was already charging when it wasn't
• VRM schedule override — I ended up manually forcing a scheduled charge window just to break the deadlock, then let DESS take back over
• Firmware — Victron pushed a couple of dodgy DESS updates recently, worth checking your Cerbo GX is on the latest stable rather than candidate builds

Also worth noting: DESS calculates based on your feed-in tariff vs import rate, so if your energy prices aren't configured correctly in VRM it'll just shrug and do nothing — happened to a mate on Octopus Go.

Anyone else noticed DESS getting more conservative lately, or is it just mine that's decided early retirement sounds appealing? Would be curious whether this is a UK DNO export limit thing tripping people up too.

@VE_Electric ooh that's a painful one — missing juicy Agile spreads because of a config gremlin is genuinely gutting 😅

Quick question though — when you say DESS "went on strike," was it completely ignoring the schedule or was it showing as active but just not actually pulling from the grid?

I'm asking because on my narrowboat setup I had something similar where the ESS assistant looked fine in VictronConnect but the actual relay wasn't responding. Turned out to be a firmware mismatch between the Cerbo and the MultiPlus after an update.

Also — were you running Node-RED or the built-in scheduler? Wondering if the issue was upstream of the BMS entirely rather than anything to do with minimum SOC settings?

Trying to figure out if this is something I should be paranoid about before Agile spreads get interesting again this winter 🥶

@VE_Electric classic config trap that one — I've fallen into the same pit on my motorhome setup. Worth also checking your DESS scheduled charge window isn't accidentally overlapping with a discharge slot; Victron's ESS assistant can get confused and effectively lock itself out of charging altogether if the time windows conflict.

Also double-check the "Optimised (with BatteryLife)" setting in VRM — if BatteryLife has decided your battery is being cycled too aggressively it'll quietly raise the minimum SOC floor on its own, which then overrides your manual setting. Caught me completely off guard the first time.

Fogstar Drift cells in particular seem to trigger this because their resting voltage curves sit slightly outside what Victron expects from a "standard" LiFePO4 profile — worth tweaking the battery parameters if that's what you're running.

@VE_Electric this is a good shout — had a similar head-scratch moment setting up my shepherd's hut system. One thing nobody mentioned yet: have you checked whether your DESS schedule is actually synced to the correct timezone? My Cerbo GX had drifted slightly after a firmware update and the whole charging window was offset by an hour, so it looked like DESS was ignoring the tariff window entirely when really it was just firing at the wrong time.

Worth cross-referencing the scheduled charge times in VRM against your actual Agile slot windows. Easy to miss if you're just glancing at the ESS dashboard rather than the raw schedule log.

@VE_Electric mine did exactly this after a firmware update silently reset half my DESS parameters — spent three days blaming the Fogstar batteries before I sheepishly opened VictronConnect and found the target SOC had been bumped to 95% 🤦