Update on my home stormwater pump backup using a 48V MP2 with 5kWH LFP battery, no PV

[KiwiME (Paul)]

In summary I’ve built a Multiplus-II plus battery system to provide backup power for a critical stormwater pump drawing about 1.1kW for periodic 1 minute pumpouts. I’d posted a couple of times before for comments and suggestions, links here (1 and 2).

The big question was ‘will it start the pump’ and I provided a oscilloscope graph of the startup current demand as a reference. In the end there was really no firm basis to decide on a suitable MP-II size as it seems the under-one-second inrush performance of these units remains undocumented. I ended up going with the smaller 48/3000VA MP-II system as it best matched the 5kWh (100Ah) LFP DAH Solar battery that I had chosen, the least expensive I could find on our local (NZ) market. Going with a 5000VA MP2 would not have helped as the battery 100A max output was the limiting factor. My ‘Plan B’ if needed, was to install a smaller pump.

The system is built up far enough for a ‘wet’ test and (as it’s not wired to the house yet) I was able to simply plug it into my generator/ATS inlet in time for today’s rain. The long-awaited news is that it worked perfectly with no errors. The battery was at around 60% SoC (it’s not calibrated by the MP2 yet) and the power conversion to AC was done with an efficiency of 91%. I’m losing 12.5W in the DC wiring which I think is fine, using the minimalist 16mm2 wire that came with the battery. I have a 125A fuse and 63A CB in the DC path.

Needless to say that I’m very pleased that it works and can look forward to planning the next steps.

Noting that the standby battery drain into the MP2 is 1A, with an anticipated 50% DoD (70% to 20%) I’m looking at about 2 days of backup, not including any energy used for pumping. Chances are that’s enough since my local grid is very reliable and the risk of power loss during rainfall is already quite low.

For reference, pump start-up on grid power, dry:

[KiwiME (Paul)]

Looking through the LFP battery BMS settings I can develop a sense of why this combination works, noting that I took a chance on this configuration, with the only “Plan B” of changing to to a smaller pump. Going to the next larger battery (10kWh) and inverter (5kVA) was not practical.

The advertised battery spec did not include the valuable information of momentary current limits v.s. time. I’ve plotted the timeline of a theoretical startup combining battery allowed output and the variable (documented) inverter limits and efficiency at those various power levels. In short, I’m just sneaking under the system limits. I’m not able to take into account the inverter’s stored capacitance which must play a role in those first few milliseconds.

So, the red line is the pump starting power just barely avoiding the system limits in black. I’ve assumed 48V nominal from the battery and 230VAC out from the inverter.