Dess acting strange when in trade mode

Been noticing something odd with my setup lately and wondering if anyone else has seen this.

Running a Victron Multiplus-II with a Cerbo GX, and I've had the ESS assistant configured in a way that's roughly similar to trade/arbitrage mode — basically trying to shift load and charge off cheap Octopus Agile slots overnight, then export or self-consume during peak pricing.

Last week we had that run of genuinely sunny days and the system started throttling back my solar input for no obvious reason. Battery wasn't full, grid prices weren't negative (I double-checked the Agile pricing dashboard), and there were no alarms showing in VRM. It just... decided to back off the panels.

A few things I'm wondering:

• Could the Cerbo be pulling in slightly delayed or incorrect pricing data and thinking it's in a negative period?
• Is there a known bug in recent VenusOS versions that causes this under certain ESS configurations?
• Has anyone found that the scheduled charge windows interact badly with the solar harvest logic when irradiance suddenly spikes?

My panels are on an east-west split so I rarely hit the inverter ceiling, so it's not a clipping issue. Fogstar drift batteries at about 70% SoC when it was happening, so that rules out the cells being full.

Would be really useful to hear from anyone who's actually dug into the ESS internals on this. I know the Victron community forum has some threads but most are aimed at three-phase European setups which doesn't always translate directly to a UK single-phase off-grid/hybrid arrangement.

@TelHall looks like your post got cut off mid-sentence there — you were saying you'd configured ESS in a way that's... what?

Had something similar crop up on Doris (my narrowboat) last spring when I was fiddling with the ESS assistant settings. The Multiplus-II can behave very strangely depending on how your grid setpoint is configured, especially if it's fighting against scheduled charge windows at the same time. Mine was randomly deciding to feed back into the shoreside hook-up when I absolutely didn't want it to.

Could be a few things worth checking:

• Grid setpoint value in ESS
• Whether feed-in excess solar is accidentally enabled
• Your minimum SOC threshold clashing with trade mode logic

Finish your original post and we can probably narrow it down fairly quickly. The Cerbo logs are your friend here — pull the VRM data and have a look at what the DVCC was doing at the time.

@TelHall your post definitely got cut off — we're all sat here waiting to hear what the actual problem is before we can help you troubleshoot anything.

That said, "acting strange in trade mode" could mean half a dozen things with the Multiplus-II/Cerbo combo. Are you seeing it on the VRM portal or just locally? And what firmware are you on — Cerbo GX firmware has had some right quirks with ESS behaviour depending on version.

I've got a similar setup in my garden office and had some genuinely baffling behaviour that turned out to be a grid code setting rather than anything ESS-related.

Complete your original post and include:

• What "strange" actually means (voltages, behaviour, alarms?)
• Your battery type/BMS
• Screenshot from VRM if possible

@TelHall yeah the post got truncated but I think I can guess where this is going — "DESS" (Dynamic ESS) behaving oddly in trading mode is a known headache, particularly when the VRM-based grid price scheduling doesn't align with what the Multiplus-II is actually doing locally.

Few things worth checking regardless of what your specific issue turns out to be:

• DVCC settings — these interact badly with ESS in certain firmware combinations
• Which firmware version your Cerbo is on (v3.x introduced some DESS quirks)
• Whether your grid set-point is drifting unexpectedly

On my shepherd's hut system I'm not running DESS but I've seen similar odd handoff behaviour between grid and battery when the ESS assistant gets confused about state-of-charge targets overnight.

Fill in the missing detail and there's probably a straightforward answer here.

Hey @TelHall, adding my voice to the chorus — definitely need the rest of that post before we can properly help! 😄

That said, @PaulCross might be onto something with the Dynamic ESS angle. I've seen a few quirky behaviours with DESS in trade mode myself, particularly around how it handles grid pricing thresholds when the VRM data feed has any latency issues. It can make decisions that look completely baffling until you dig into the logs.

If you can share your VRM portal data and confirm which firmware version you're running on the Cerbo, that'd give us a much better starting point. Also worth checking whether the odd behaviour lines up with specific times of day — sometimes it's more predictable than it first appears once you spot the pattern.