Battery Pack Sizing with Standard Modules: A Step-by-Step Guide
Short answer: To size a stationary battery pack with standard modules, convert your usable energy requirement to nominal capacity using your target depth of discharge, check that your chosen series string voltage falls within the inverter's DC voltage window, verify the C-rate at peak power, then repeat the string in parallel to reach capacity. Standard module formats make each step discrete rather than continuous.
Sizing a stationary battery energy storage system (BESS) is not a single calculation — it is a sequence of constraints that must be satisfied in a fixed order. Skipping one or confusing nominal capacity with usable energy is the most common source of undersized packs that trip on under-voltage during discharge. This article walks through the full process with explicit assumptions and a worked numerical example, using standard prismatic battery modules as the building block.
Step 1: Usable energy versus nominal capacity
The first number a project specifies is usable energy: the kilowatt-hours the system must deliver to the AC bus in normal operation. That figure is not the same as the nominal energy stored in the cells.
Two losses sit between them:
- Depth of discharge (DoD): cycling cells between 0 % and 100 % state of charge (SoC) accelerates degradation. Stationary LFP modules are commonly specified at 80 % DoD, which is also the depth at which their datasheets state cycle life — 8,000 cycles and upward for a purpose-built ESS module. NCM is operated somewhat shallower, commonly 70–80 %.
- Round-trip efficiency: a realistic AC-to-AC round-trip efficiency for a modern BESS with a well-matched inverter is 88–94 %. For DC-side sizing (cell and module selection), the relevant figure is the DC energy that must be stored, not the AC energy.
Worked example — project assumptions:
| Parameter | Value | Note |
|---|---|---|
| Required usable AC energy | 100 kWh | Project specification |
| Target DoD | 80 % | LFP, matching the datasheet cycle-life basis |
| DC-side efficiency factor | 0.96 | Inverter + cabling losses |
| Required nominal DC energy | 100 ÷ (0.80 × 0.96) | = 130.2 kWh |
The pack must therefore hold at least 130.2 kWh nominally to deliver 100 kWh to the AC bus under the stated assumptions. Every assumption here is explicit and adjustable; change the DoD or efficiency and the nominal requirement shifts accordingly.
Step 2: Power requirement and C-rate
C-rate is the ratio of continuous power to nominal energy capacity. A 1C discharge means the pack is fully discharged in one hour. Standard stationary BESS applications typically run at 0.25C to 0.5C continuous; peak values up to 1C are common for frequency response.
For the worked example, assume a required continuous AC power of 50 kW.
At the DC side (accounting for inverter efficiency of, say, 96 %):
DC power required = 50 kW ÷ 0.96 ≈ 52 kW
With 130.2 kWh nominal:
C-rate = 52 kW ÷ 130.2 kWh ≈ 0.40C
Stationary LFP modules are commonly rated for 1C continuous discharge with a short-duration peak of up to 3C, so 0.40C sits well inside the envelope. This is the point to verify against the specific module datasheet — browse the cell and module catalogue for confirmed ratings.
Step 3: The inverter DC voltage window — the hard constraint
The DC voltage window of the inverter is the single hardest boundary in the system. A string that falls below the minimum MPPT voltage at low SoC will be disconnected by the inverter. A string that exceeds the maximum input voltage will damage the inverter.
Typical DC voltage windows for commercial stationary inverters (not specific to any brand):
| Inverter class | Minimum DC voltage (V) | Maximum DC voltage (V) |
|---|---|---|
| Small C&I (up to ~100 kW) | 200–350 | 800–1000 |
| Medium C&I (100–500 kW) | 500–600 | 1000–1500 |
| Utility-scale | 600–900 | 1500 |
These are realistic industry ranges, not guaranteed specifications. Always confirm with the inverter manufacturer.
The worked example uses a purpose-built stationary LFP module: 1P16S, 100 Ah, 5.12 kWh. An LFP cell has a nominal voltage of 3.2 V and an operating range of approximately 2.5–3.65 V per cell, so sixteen cells in series give:
- Nominal voltage: 16 × 3.2 V = 51.2 V
- Minimum voltage (at 0 % SoC): 16 × 2.5 V = 40.0 V
- Maximum voltage (at 100 % SoC): 16 × 3.65 V = 58.4 V
That is a different building block from the automotive formats, and the difference matters for sizing. A VDA 355 module measures approximately 355 × 151 × 108 mm and was standardised through the German automotive industry association (VDA) for mechanical interchangeability in vehicle packs; MEB 590 is the larger platform counterpart. Both turn up in stationary projects, second-life in particular, but they are specified for vehicle duty — typically 1,200–2,000 cycles at 80 % DoD against 8,000 or more for a purpose-built ESS module. Check the cycle life at the DoD you intend to use before treating an automotive module as a stationary building block.
For the worked example, assume the inverter has a DC window of 350–800 V. Both bounds have to hold in the worst case: the string must clear the inverter floor when it is empty, and stay under the ceiling when it is full. The number of modules in series per string is therefore bounded:
Minimum modules in series: 350 V ÷ 40.0 V (module voltage at 0 % SoC) ≈ 9 modules (rounds up) Maximum modules in series: 800 V ÷ 58.4 V (module voltage at 100 % SoC) ≈ 13 modules (rounds down)
Taking the top of that range, 13 modules in series:
- Nominal string voltage: 13 × 51.2 V = 665.6 V
- String minimum voltage: 13 × 40.0 V = 520 V (above 350 V minimum ✓)
- String maximum voltage: 13 × 58.4 V = 759.2 V (below 800 V maximum ✓)
Step 4: Module energy and string capacity
A 1P16S LFP module with 100 Ah cells holds:
Module nominal energy = 51.2 V × 100 Ah = 5.12 kWh
One 13-module string:
String nominal energy = 13 × 5.12 kWh = 66.6 kWh
The project requires 130.2 kWh nominal. Dividing by string energy:
Strings in parallel = 130.2 kWh ÷ 66.6 kWh ≈ 1.96 → round up to 2 strings
Two parallel strings of 13 modules = 26 modules total, nominal energy:
26 × 5.12 kWh = 133.1 kWh
Usable energy at 80 % DoD, 96 % DC efficiency: 133.1 × 0.80 × 0.96 ≈ 102.2 kWh — just above the 100 kWh requirement, without a single module bought that the project does not use.
Step 5: Why standardised formats make the configuration discrete
With a freely chosen cell format, voltage and capacity can be tuned continuously. Standard module formats — an automotive envelope such as VDA 355 or MEB 590, or a fixed ESS module — introduce quantisation: string voltage and energy only change in fixed steps. This is not a limitation; it is a practical advantage.
- The mechanical envelope, cooling interface and busbar layout are already defined.
- Multiple cell suppliers can fill the same module format, reducing single-source risk.
- Second-life automotive modules entering the stationary market share the same standardised geometry, a market the International Energy Agency tracks in its battery supply chain reporting.
For integrators, this means that configuration iterations are rapid: adjusting the series count by one module shifts string voltage by 51.2 V (for the 1P16S LFP module above), and the effect on the inverter voltage window is immediately calculable. See battery modules from Avantis Energy for the available formats and confirmed cell capacities.
Configuration comparison
The table below compares three configurations for the same 100 kWh usable energy requirement, all using 1P16S LFP modules with 100 Ah cells (5.12 kWh/module), at 80 % DoD and 96 % DC efficiency.
| Configuration | Modules in series | Parallel strings | Total modules | Nominal energy (kWh) | Usable energy (kWh) | Nominal string voltage (V) |
|---|---|---|---|---|---|---|
| A | 9 | 3 | 27 | 138.2 | 106.2 | 460.8 |
| B | 13 | 2 | 26 | 133.1 | 102.2 | 665.6 |
| C | 17 | 2 | 34 | 174.1 | 133.7 | 870.4 |
Configuration A stays inside a 350–600 V window but needs one module more than B for the same job. Configuration B is the tightest fit to the target. Configuration C reaches 992.8 V with its strings full, so it needs a 1000 V class inverter — and it delivers a third more energy than the project asked for. That overshoot is the price of quantisation when the voltage window pushes the series count up, and it is the argument for choosing the configuration against the inverter window rather than against the energy target.
For projects where NCM chemistry is preferred — higher energy density per module at the cost of a narrower operating SoC window — see the discussion of NCM versus LFP battery chemistry and its effect on sizing assumptions.
For automotive-format modules, the larger MEB 590 in particular, the same method applies with different per-module voltage and capacity figures — and with a cycle life specified for vehicle duty rather than daily stationary cycling. Details are in the article on the next-generation MEB platform battery module.
Compliance with IEC 62619 (safety requirements for stationary lithium cells and batteries) applies at the system level and is a parallel workstream to electrical sizing, not a substitute for it. See IEC 62619 for the current edition of that standard.
For a broader view of when stationary storage projects are financially viable, the C&I energy storage cost-effectiveness analysis covers the economic side of the same decision.
Frequently asked questions
What is the difference between nominal energy and usable energy in a battery pack?
Nominal energy is the total energy stored in the cells at full charge. Usable energy is what the system can deliver after applying depth of discharge (DoD) and accounting for efficiency losses. For an 80 % DoD LFP system, usable energy is approximately 77 % of nominal DC energy after inverter losses.
How do I determine how many modules to connect in series?
Divide the inverter's minimum DC voltage by the module voltage at 0 % SoC and round up: the string must stay above the inverter floor when it is empty. Divide the inverter's maximum DC voltage by the module voltage at 100 % SoC and round down: the string must stay under the ceiling when it is full. Choose a series count that keeps string voltage inside that window across the full SoC range.
What C-rate should I design for in a stationary BESS?
Most commercial and industrial stationary applications run at 0.25C to 0.5C continuous discharge. Peak demand response may require up to 1C for short durations. Stationary LFP modules are commonly rated for 1C continuous with a short-duration peak of up to 3C, but confirm the rated C-rate on the specific module datasheet before finalising the design.
Why does a standardised module format simplify pack sizing?
A standardised format — such as VDA 355 or MEB 590 — fixes the module voltage, capacity and mechanical envelope. This makes string voltage and energy change in defined steps rather than continuously, so each design iteration is a rapid recalculation rather than a cell-level redesign. It also allows modules from different cell suppliers to be substituted without changing the pack architecture.
Does the sizing method differ for LFP versus NCM modules?
The method is identical, but the input parameters differ. NCM cells have a higher nominal voltage (approximately 3.6–3.7 V versus 3.2 V for LFP) and a wider voltage swing between empty and full, which affects both the series count calculation and the DoD assumption. NCM systems are typically operated at a narrower SoC window — commonly 70–80 % DoD — to preserve cycle life.
Discuss your configuration
If your project constraints — inverter voltage window, peak power, usable energy target — do not map cleanly onto the configurations above, discuss your pack requirements with Avantis Energy. Standard module formats and confirmed cell specifications are the starting point for every sizing conversation.