The PCS sits between the battery and the AC grid. The DC voltage window on the battery side determines the IGBT stage, the bus capacitor, the DC cable cross-section, and the round-trip efficiency at the terminal. There is no single right answer; the answer depends on what the rest of the system looks like.
Three common choices for a 1 MW PCS in 2026: 600 V DC low-voltage, 1000 V DC medium-voltage, and 1500 V DC high-voltage. Each is a valid option for a different site layout.
600 V DC low-voltage
Best when the battery is lithium iron phosphate (LiFePO4) and the cell string is in the 150–180 cell range. The cell cabinets from a major LFP supplier will deliver a nominal 614 V DC, which sits inside the 600 V cabinet’s input window. Round-trip efficiency at full load is 96.5%. DC cable cross-section at 1 MW is 95 mm² copper or 120 mm² aluminium, which is manageable in a standard container.
Two things to watch: the DC bus bar layout has to be planned with the battery cabinet layout in mind, and a 600 V system may need an extra step-up transformer to reach the AC grid voltage, which costs both capex and 0.5% efficiency.
1000 V DC medium-voltage
Best when the battery chemistry tolerates a higher string voltage (high-nickel NMC, or a flow battery where the cell voltage scales). The cabinet sits inside the 1000 V window with margin. Round-trip efficiency is 97.0%. DC cable cross-section drops to 50² copper, which fits through standard cable glands without custom glands.
This is the default we ship for 1 MW industrial cabinets in 2026. The reason is not the electrical — a 1000 V and 1500 V cabinet have nearly identical efficiency — the reason is the supply chain: 1000 V IGBT modules are stocked by three manufacturers with lead times of six weeks. 1500 V modules are stocked by one, with lead times of twelve to sixteen weeks.
1500 V DC high-voltage
Best when the PCS is feeding a solar array directly (DC-coupled PV+storage) or when the AC grid is at 11 kV or higher with a low step-up ratio. Round-trip efficiency is 97.1%. DC cable cross-section drops to 35 mm² copper, which is a real saving on large sites.
The catch is the battery side. LiFePO4 at 1 C rate and 1500 V DC requires a cell string of roughly 400 cells. Most cell cabinets do not stack that high on a single string without a midpoint tap, and adding the midpoint tap introduces a fault-mode that the BMS has to handle. For a site where the battery cabinet is already specified, 1500 V often forces the integrator to choose between longer string length and BMS complexity.
Numbers side by side
| Voltage | Round-trip | DC cable (Cu) | IGBT lead time | Cell string (LFP) |
|---|---|---|---|---|
| 600 V | 96.5% | 95 mm² | 4 weeks | 150–180 cells |
| 1000 V | 97.0% | 50 mm² | 6 weeks | 250–300 cells |
| 1500 V | 97.1% | 35 mm² | 12–16 weeks | 350–420 cells (with midpoint tap) |
How to choose, in three questions
- What cell chemistry and cell cabinet configuration is the integrator already committed to?
- What is the AC grid voltage, and how many step-up stages does the design tolerate?
- What is the supply-chain reliability you need — can you accept 12–16 weeks on a single IGBT supplier?
Pick the voltage that lets you answer all three with the lowest combined capex, opex, and supply-chain risk. For most 1 MW industrial sites today, the answer is 1000 V DC, but the answer is not universal.
For 1 MW-class PCS cabinets in 600 V, 1000 V, and 1500 V DC variants, contact: 58298175@qq.com / +86-181-5931-0123. We will send a one-page comparison sheet with the requested cell cabinet configuration against the requested grid code.
