Variable peak shaving and Dynamic ESS

Anyone else wrestling with getting dynamic ESS to play nicely with peak shaving on a Victron setup?

Running a Multiplus-II 3kVA at the static caravan with a Fogstar 10kWh lithium pack and a couple of Victron SmartSolar MPPTs — all tied together through VRM. Works a treat most of the time, but I'm trying to nail down variable peak shaving limits rather than one fixed threshold.

At the moment I've got peak shaving capped at roughly 13A/3kW through VRM, which stops me hammering the grid connection (dodgy supply out here in the sticks). Problem is, that limit feels too conservative on a sunny afternoon when the batteries are rammed and too aggressive on a gloomy January morning when the pack is sat at 20%.

What I actually want is for the peak shaving threshold to flex based on state of charge — so it's more aggressive when the battery is healthy and backs off when reserves are low. Is that achievable through the dynamic ESS settings, or am I dreaming?

I've had a poke around in the Node-RED flows on the Cerbo and suspect the answer lives in there somewhere, but my programming skills peaked around 1987 on a BBC Micro.

Has anyone actually got dynamic peak shaving limits working in a UK setup — ideally without needing a computer science degree? Any VRM schedules, Node-RED templates, or even just a sensible workaround would be brilliant.

Hey @RetiredNurse61, yes this caught me out too! The key thing to understand is that Dynamic ESS and peak shaving are essentially competing for control of the same setpoint, so you can't really run both simultaneously the way the menus suggest you might.

What worked for me was ditching Dynamic ESS entirely and just using the ESS scheduled charging with a fixed grid setpoint instead. Set your grid feed-in limit appropriately and let the BMS handle the rest.

One thing worth checking — are you on the latest Venus OS firmware? There were some fairly significant ESS logic changes around v3.3 that affected how the Multiplus-II handles the AC input current limit during peak periods.

What's your current grid setpoint set to, and are you seeing the battery discharge when you don't want it to? That'll help narrow things down. 🙂

@RetiredNurse61 had exactly this headache on the narrowboat last season. The bit that nobody tells you upfront is that Dynamic ESS essentially takes over the grid setpoint logic, so your manual peak shaving threshold becomes a passenger rather than the driver.

What finally made sense for me was treating them as a hierarchy — Dynamic ESS sets the broader strategy based on tariff pricing, and peak shaving operates within that envelope, not alongside it.

Worth checking your VRM portal under Advanced settings whether you've got "BatteryLife" still active, because that quietly undermines the whole thing by reserving capacity the peak shaving algorithm was counting on.

@LiFePO4Pro makes a fair point above, but I'd add — with a Fogstar pack specifically, make sure your min SoC floor is configured in both places. Victron doesn't always sync those automatically and that inconsistency alone can make the behaviour look completely random.

Great topic — this is something I've spent far too many evenings down the rabbit hole on with my own cabin setup.

The bit worth adding: Dynamic ESS pulls grid price data via the VRM portal, so your schedule_charge and peak shaving thresholds need to coexist carefully in ESS Assistant settings. If your grid setpoint is fighting the Dynamic ESS target SOC, you'll get the system constantly hunting — chasing its own tail essentially.

Key things to check:

• Minimum SOC — set conservatively or Dynamic ESS ignores your peak shaving window entirely
• Grid setpoint (typically 50W) — too aggressive and you'll see unnecessary grid draw during peak hours
• Make sure your VRM Remote Console is actually syncing current tariff data — Octopus Agile users especially, this drops out silently sometimes

Worth cross-referencing the Victron Community forums too — there's a cracking thread on dbus-mqtt overrides that gives you much finer control.

Been through this exact faff with my own Multiplus-II setup. One thing worth checking that nobody's mentioned — if you're running Dynamic ESS via VRM, make sure your BatteryLife setting isn't fighting against it. The two can work against each other, with BatteryLife holding the SOC floor higher than DESS expects and causing weird grid draw patterns at peak times.

Also worth disabling DESS temporarily and running plain peak shaving first, just to confirm your grid setpoint is actually working correctly in Node-RED or VEConfigure. Eliminates variables before you layer complexity on top.

@RetiredNurse61 what firmware version are you on? There were some DESS improvements in recent Venus OS large builds that sorted a few edge cases around dynamic grid setpoints. Worth checking you're not on something older.

Ran into this exact mess last winter on my static. The thing that sorted it for me was nailing down the grid setpoint in VE.Configure — mine was set way too conservatively and the ESS kept second-guessing itself whenever the kettle kicked in.

Drop it to around -50W and Dynamic ESS suddenly becomes a lot less twitchy. Took me an embarrassing number of evenings to figure that out 😅

Also worth checking your BMS communication is solid — my Fogstar pack was occasionally going quiet over VE.Bus and the Multiplus would just shrug and do whatever it fancied with the grid, completely ignoring the peak shaving schedule.

VRM portal logs are your friend here — painful to read but they'll show you exactly when it's going rogue.