6 x 20kV Multiplus = 2 per phase :: Does this make sense? Or do I go with an HV system rather

Been mulling this over for a while now and curious what others think, particularly those running larger setups.

I'm looking at scaling up significantly beyond my current shepherd's hut system and the maths keeps pointing me toward either 6 x Multiplus-II 5000VA (2 per phase) in a three-phase arrangement, or jumping to an HV battery system with something like a Deye hybrid inverter.

The Victron route feels safer — I know the ecosystem, the monitoring through VRM is brilliant, and the community support is second to none. But stacking six inverters feels like a lot of potential failure points, and the complexity of commissioning a three-phase parallel system isn't trivial.

The HV Deye route is more elegant on paper — fewer boxes, cleaner wiring, purpose-built for scale. But I'm less familiar with the long-term reliability picture, particularly for an off-grid cabin install where I can't exactly ring a local engineer at short notice.

A few questions I'm wrestling with:

• Has anyone here actually run 2 x Multiplus per phase off-grid (not grid-tied) long term? Any gotchas?
• Is Deye's HV ecosystem mature enough to trust for a primary off-grid system in the UK yet?
• How are people handling battery bank sizing at this scale — Fogstar Drift cells, or going full rack-mount LiFePO4?

My gut says stay Victron, but the head says the HV route is the more sensible engineering choice at this power level.

Would love to hear from anyone who's been through a similar decision — especially if you've gone one way and regretted it.

@Gill1982 the post got cut off but I can guess where you're going — 2× Multiplus-II per phase in a 3-phase stack, yes?

Ran something similar in a test config before settling on my current shepherd's hut setup. The parallel-per-phase approach works well with Victron's ESS assistant but you're essentially doubling your single points of failure and your firmware sync headaches.

Key consideration nobody mentions: phase balancing becomes critical with uneven loads. Victron's 3-phase compensation handles it, but only within limits.

The HV battery argument (Pylontech Force H2 territory) becomes compelling above ~30kWh usable — fewer strings, lower cable losses, tidier BMS comms.

What's your actual load profile and peak draw? That'll determine whether the Multiplus stack genuinely outperforms an HV alternative or whether you're over-engineering it.

@Gill1982 assuming @HighlandExplorer has guessed correctly —

Worth clarifying which Multiplus-II variant you mean. The Multiplus-II 48/5000 is the common one for stacking; you'd parallel two per phase giving you roughly 10kVA per phase, 30kVA total.

Few things to consider:

• All six units need to be identical firmware versions before commissioning — Victron are strict on this
• You'll need a Cerbo GX coordinating the whole stack
• Battery bank sizing becomes significant at that scale; Pylontech or similar 48V HV-compatible stacks work well

Regarding HV (high-voltage battery systems with something like a Fronius or SMA Sunny Boy Storage) — that's a fundamentally different architecture. Makes more sense if you're AC-coupled with existing solar already installed.

What's your primary use case? The shepherd's hut background suggests off-grid rather than grid-tied, which changes the answer considerably.

@RetiredPlumber has a point worth finishing — the 48/5000, 48/8000, or 48/10000 variants change the maths considerably before anyone can give sensible advice.

That said, the "2 per phase" approach using the Multiplus-II with Victron's three-phase stacking is a well-trodden path. I went down a similar rabbit hole when I was speccing out my tiny house system, though single-phase in my case. The key thing nobody mentions early enough: your battery bus needs to handle the aggregate current from all six units simultaneously, which at 48V gets eye-watering very quickly.

HV systems sidestep that precisely because the amperage drops dramatically at higher voltages. Worth pricing up a proper HV stack against the copper and busbars alone before assuming the Multiplus route is cheaper.

What's your anticipated peak load across all three phases?

@Gill1982 what's the actual load you're trying to hit? Because stacking 6 Multiplus-IIs is a serious chunk of kit and money before you've even looked at the battery bank needed to back it up.

Also — what's your grid connection situation? If you've got decent grid access, a 3-phase Quattro setup might actually make more sense than paralleling Multis. And if you're genuinely off-grid with loads that justify 6 inverters, I'd be asking whether HV battery systems like the Pylontech Force-H2 are worth the conversation with a proper installer.

What's this actually powering? Shepherd's hut to this is quite a jump.