House Consumption shows 0 W after adding Huawei SUN2000 to Victron ESS

Bit of a head-scratcher I wanted to throw out to the community, because I've been going round in circles on this one.

Fitted a second inverter to a customer's site last month — they already had a decent Victron ESS setup running with a Quattro, Cerbo GX, ET112 grid meter, and a Pylontech stack. All singing, all dancing. We bolted on a Huawei SUN2000 4kW to grab some extra generation from a south-facing roof section that wasn't being used.

The moment the SUN2000 came online, House Consumption dropped to 0W on the Cerbo dashboard. Everything else looked fine — battery SOC updating, grid reading present — but that consumption figure just flatlined. Completely useless for monitoring.

My suspicion is the ET112 is getting confused about the direction or magnitude of power flow now that there are two inverters feeding the bus simultaneously. The Cerbo seems to be doing its maths wrong somewhere, possibly treating the SUN2000 generation as consumption offset rather than a separate source.

Has anyone else mixed Huawei SUN2000s into an existing Victron ESS alongside another AC-coupled inverter? I know Fronius plays nicely with Victron's frequency-shift control — not sure Huawei is quite as well-behaved in that regard, and I wonder if that's muddying the waters somehow.

Currently thinking the fix might involve:

• Moving the ET112 position
• Adding a second meter specifically for the SUN2000
• Or rethinking how the SUN2000 is presented to the ESS entirely

Would love to hear if anyone's navigated this. UK installs, import/export metering complications welcome too — MPAN

@KevLee this is almost certainly a CT clamp orientation or placement issue rather than anything exotic. When you add a third-party inverter like the SUN2000 alongside Victron ESS, the grid meter (assuming you're using a Victron ET340 or similar) needs to see net grid flow — if the Huawei's output is being measured on the wrong side of the CT, it'll cancel out your load reading entirely.

Few things to check:

• CT clamp arrow pointing toward grid (not load)
• SUN2000 AC output connecting before the Victron meter measurement point
• Whether you've set the ESS assistant to "External current sensor" correctly

Had something adjacent on the narrowboat when I added a second MPPT — the Cerbo GX was showing nonsense until I sorted the measurement topology. Not identical but same root cause logic.

What's your grid meter model and where exactly is the SUN2000 tied in?

@KevLee worth checking whether the Huawei's own meter is interfering with the Victron's grid sensing. SUN2000 units have an internal DTSU666 meter that can conflict if both systems are trying to be the "master" energy meter on the same grid connection.

Had something loosely similar when I was sorting EV charging integration at my static — two systems both thinking they're in control of grid measurement gives you exactly this kind of phantom 0W reading.

In Victron's ESS assistant settings, check your grid meter source is explicitly set and not defaulting to a confused state. If you're running VenusOS, also worth looking at the MQTT/Modbus data to see what the Victron thinks it's measuring vs what the SUN2000 is reporting — the discrepancy usually tells you where the breakdown is.

@Spud79 and @MuddyTrekker have both touched on the likely culprits already, so won't repeat that. One thing I'd add — check whether the Victron is actually seeing the Huawei as a "load" rather than generation. Had something similar in my shepherd's hut setup (smaller scale obviously) where the MPPT and a secondary inverter were fighting over what counted as consumption.

In the Victron portal, worth looking at the raw AC input/output figures rather than just the dashboard "House Consumption" tile — that number is calculated, not measured directly. If the maths is wrong due to the SUN2000 pushing power in an unexpected direction, it'll just show zero rather than a negative.

Check your VRM data logs too — if it briefly shows correct values then drops to 0W, that points to a comms/polling conflict rather than wiring.

@KevLee had almost identical bother when I added a second AC source to my own setup. The thing that caught me out was the Modbus device instance — if both inverters are on the same Venus OS device list with conflicting IDs, the ESS logic can get confused about which power readings to trust and just zeroes out the house load calculation entirely.

Worth jumping into Venus OS → Settings → Services → Modbus TCP and checking for any duplicate instances. Also double-check the ESS assistant version — older versions genuinely don't handle multiple AC coupled sources cleanly.

Victron's VRM portal should show you separate feeds if everything's registered correctly. If you're seeing only one device showing up there, that's your smoking gun.