Right-sizing a photovoltaic system is the single most impactful design decision you can make. An oversized array wastes capital on unharvested energy during inverter clipping hours. An undersized array leaves capacity on the table — and revenue with it. The optimal point lives where marginal energy yield gain equals marginal equipment cost, and finding it requires a systematic methodology.

For commercial & industrial (C&I) and utility-scale projects, the sizing calculation layers multiple interdependent constraints: inverter loading ratio (ILR), C-rate matching with battery storage, DC-to-AC ratio limits from the inverter manufacturer, and interconnection agreements that cap AC export. This guide walks through each constraint and shows how to optimize for maximum project IRR.

What You'll Learn

Inverter Loading Ratio and Clipping Economics

The inverter loading ratio (ILR) — also called the DC-to-AC ratio — is the ratio of the PV array's DC nameplate capacity to the inverter's AC output capacity. An ILR of 1.30 means 130 kWDC of modules behind a 100 kWAC inverter.

Higher ILR means more energy capture during low-light hours (morning, evening, winter, overcast days) at the cost of clipping during peak irradiance hours. The economic question is: at what ILR does the incremental annual energy yield (kWh/yr) no longer justify the additional module cost ($/WDC)?

For a typical utility-scale project in the southwestern US with single-axis tracking:

  • ILR 1.20: ~1.5% annual clipping loss, low capital cost
  • ILR 1.35: ~3.5% clipping loss, moderate additional module cost
  • ILR 1.50: ~6.5% clipping loss, high module cost, minimal yield gain over 1.35

The optimal ILR typically falls between 1.25 and 1.45 for single-axis tracker projects, and between 1.10 and 1.25 for fixed-tilt or rooftop C&I systems where space and racking costs are different constraints.

ILR vs clipping loss and annual yield chart showing optimal ILR zone between 1.25 and 1.45

Key insight: Clipping loss below 3% is almost always economic to accept. The marginal cost of adding more modules to push ILR from 1.20 to 1.35 is typically recovered within 2-4 years through the additional energy harvested during non-peak hours.

C-Rate Matching with BESS

When a PV system is paired with battery energy storage, the sizing relationship becomes more complex. The battery's C-rate — maximum charge/discharge power relative to its energy capacity — must align with the PV array's power profile.

Consider a 100 MWAC PV farm paired with a 50 MW / 200 MWh BESS. The battery charges at 0.25C (50 MW into 200 MWh). The PV array at ILR 1.30 produces 130 MWDC but is capped at 100 MWAC by the inverter. During the solar noon ramp, the array can sustain 95-100 MWAC for roughly 4 hours. The battery can capture 50 MW of that — but the remaining 45-50 MW must be exported to the grid or curtailed.

To avoid curtailment while fully charging the battery, the designer has three levers:

  • Increase battery power (MW): Higher C-rate captures more PV output but costs more in power electronics
  • Reduce ILR: A lower DC-to-AC ratio reduces peak power at the cost of total energy
  • Accept curtailment: Let some energy go during the peak hours if it's economically optimal

Our detailed BESS degradation modeling guide explains how C-rate directly affects battery aging — a higher C-rate for charging accelerates cycle aging and reduces SOH over the project life. This means the C-rate matching decision also feeds back into battery augmentation costs and replacement timing.

DC-to-AC Ratio Limits and Degradation

Inverter manufacturers specify maximum DC-to-AC ratios based on the inverter's input voltage and current limits. A typical central inverter might allow ILR up to 1.50, while string inverters often cap at 1.35. These limits exist because:

  • Input current limit: The inverter can only accept a maximum DC current per MPPT channel
  • Voltage window: The string voltage must stay within the MPPT range at all temperatures
  • Thermal management: Sustained high DC input during clipping generates heat that must be dissipated

Module degradation also affects ILR over time. As modules degrade (typically 0.4-0.7%/year for monocrystalline silicon), the effective ILR drops. A system starting at ILR 1.35 might operate at ILR 1.20 by year 10 and ILR 1.05 by year 20. The optimal initial ILR should account for this — a higher starting ILR compensates for degradation and maintains inverter utilization later in life.

Interconnection and AC Export Limits

Interconnection agreements often cap the maximum AC export at a fixed value — typically 90-105% of the inverter nameplate. If you have a 100 MWAC interconnection limit, installing 120 MWAC of inverters is not an option. Instead, the DC array must be sized to maximize energy within that 100 MWAC export cap.

This scenario favors high ILR (1.40-1.60) since the AC export limit is the binding constraint, and the goal is to maximize energy delivered within that constraint. The module cost is traded against the additional energy during shoulder hours, and clipping losses can reach 8-10% before the marginal benefit disappears.

In markets with time-of-delivery (TOD) tariffs or time-varying pricing, the value of energy during different hours changes the calculus. High ILR systems deliver proportionally more energy during morning and evening shoulders — which are often higher-value periods in markets like California (CAISO duck curve) or Hawaii (HECO).

IRR-Driven Sizing Optimization

The project's internal rate of return (IRR) — not annual energy yield — should be the optimization target. A project optimized for maximum MWh/yr will have a different ILR than one optimized for maximum IRR, because the cost of adding modules eventually exceeds the revenue from the extra energy.

The IRR optimization requires running hourly simulations across multiple ILR scenarios and computing project financials for each:

  • Revenue: Hourly energy yield × PPA price (or market price) for each year, accounting for degradation
  • Cost: Additional module cost + additional racking + additional land area + additional BOS
  • OPEX: O&M scales roughly with total nameplate, not AC capacity

For a typical 100 MWAC utility project in Texas (ERCOT North, 2026 pricing):

  • ILR 1.20: Project IRR = 8.7%
  • ILR 1.35: Project IRR = 9.4%
  • ILR 1.50: Project IRR = 9.1%

The 1.35 ILR case wins because the ~3.5% additional energy yield over the 1.20 case outweighs the module cost, while the 1.50 case sees diminishing returns that don't justify the incremental capital.

Real-World Example: 50 MW Utility Project

To illustrate the process, consider a 50 MWAC utility-scale PV project in California's Central Valley with single-axis tracking and a 20-year PPA at $38/MWh.

Baseline (ILR 1.30):

  • 65 MWDC array, 50 MWAC inverters
  • Annual yield: 114,500 MWh/yr (first year, P50)
  • Clipping loss: 2.8%
  • Installed cost: $66.3M ($1.02/WDC)
  • Project IRR: 8.9%

Optimized (ILR 1.40):

  • 70 MWDC array, 50 MWAC inverters
  • Annual yield: 119,200 MWh/yr (first year, P50)
  • Clipping loss: 4.1%
  • Incremental cost: ~$5.1M added module, racking, and land
  • Project IRR: 9.3%

The IRR improves by 40 basis points. Over a 25-year project life, the additional 4,700 MWh/yr at $38/MWh generates roughly $3.6M in additional revenue — a solid return on the $5.1M investment when factoring in the time value of money and tax benefits.

How Energy Optima Handles Sizing

Energy Optima automates the PV system sizing optimization process. The platform simulates hourly energy production across any ILR range (1.00 to 2.00), using site-specific TMY data, module performance models (single-diode or PVsyst-equivalent), and inverter efficiency curves. The results are fed directly into the financial model to compute NPV, IRR, LCOE, and payback period for each sizing scenario.

The platform supports:

  • Automated ILR sweep with energy yield and financial output for each ratio
  • BESS co-simulation with C-rate matching and degradation modeling using manufacturer-specific 3D interpolation
  • Interconnection limit constraints with optimal DC sizing under export caps
  • Time-of-delivery tariff modeling for high-ILR value analysis
  • Degradation-adjusted ILR — the system accounts for module and inverter degradation over the project life

The result: you find the exact ILR that maximizes IRR for your specific site, equipment, and market conditions — without running 50 manual PVsyst simulations.