04
Chapter Four

System Sizing & Production Estimation

This is where your energy data, site assessment, and goals converge into an actual system specification. How many panels do you need? How big should the inverter be? Where exactly do the panels go on your roof? The answers come from a design process that balances energy production targets against physical constraints, code requirements, and budget.

We handle the detailed engineering for our clients, but understanding the fundamentals helps you make informed decisions and evaluate whether a design makes sense. This chapter walks you through the sizing calculation, the factors that affect real-world production, and the basics of array layout and string design.

The sizing calculation

At its simplest, system sizing is a division problem: how much energy do you need, divided by how much energy a kilowatt of solar produces at your location. The result is your target system size in kilowatts (kW).

The basic formula

System Size (kW) = Annual kWh Target ÷ Annual kWh per kW installed

Example: 10,000 kWh ÷ 1,400 kWh/kW = 7.14 kW system

Let’s break that down:

  • Annual kWh target — this comes from Chapter 2. For a 100% offset, it’s your total annual electricity consumption. Adjust up if you’re planning for future loads (EV, heat pump) or down if you’re targeting a partial offset.
  • Annual kWh per kW installed — this is your site’s specific yield, measured in kWh/kWp. It depends on your location, roof orientation, tilt, shading, and system losses. In the US, this number typically ranges from about 1,100 kWh/kW in the cloudier northern states to 1,800+ kWh/kW in the desert Southwest.

Quick reference: PVWatts

The NREL PVWatts Calculator is a free tool that estimates annual production for a given system size, location, tilt, and azimuth. It’s a solid starting point for ballpark sizing. Enter your address, plug in a system size, and it gives you estimated annual kWh output. We use more detailed modeling tools for final designs, but PVWatts is useful for early-stage sanity checks.

This formula gives you a starting point, not a final answer. The real design process accounts for specific yield more precisely using your actual site data, adjusts for system losses, and then fits the result to real panel counts and inverter capacities. A 7.14 kW calculation might become a 7.2 kW system (18 panels × 400 W) or a 7.6 kW system (19 panels × 400 W) depending on your panel choice, available roof space, and design constraints.

Factors that affect real-world production

A solar panel’s nameplate wattage is measured under Standard Test Conditions (STC): 1,000 W/m² irradiance, 25°C cell temperature, AM 1.5 spectrum. Real rooftops aren’t laboratories. Several factors cause your system to produce less than the theoretical maximum:

FactorTypical LossNotes
Temperature5–12%Panels lose efficiency as they heat up. Hot climates see larger losses. Temperature coefficient is listed on the panel datasheet.
Soiling1–5%Dust, pollen, bird droppings, and other debris. Rain does a decent job of cleaning tilted panels; flat or very low-tilt arrays accumulate more.
Shading0–25%+Highly site-specific. Even partial shading can cause significant losses on string inverter systems. MLPE mitigates this.
Inverter efficiency2–4%The DC-to-AC conversion isn’t 100% efficient. Modern inverters are typically 96–98% efficient at their rated operating point.
Wiring losses0.5–2%Resistance in DC and AC wiring. Proper wire sizing minimizes this. Longer runs (like ground-mount systems) have higher losses.
Module mismatch0.5–2%Slight manufacturing variation between panels on the same string. MLPE eliminates this loss.
Degradation~0.4–0.5%/yearPanels slowly lose output over time. Most panels are warrantied to produce at least 80–85% of rated power after 25–30 years.
Snow / ice0–10%Snow-covered panels produce nothing until they shed. Tilted panels clear faster. Varies hugely by climate.

When you stack all of these together, a typical system produces somewhere between 75–85% of its theoretical maximum annual output. This is captured in a metric called the performance ratio. Production modeling software accounts for each of these factors individually based on your site’s specific conditions — that’s why a proper production estimate is more useful than a back-of-envelope calculation.

DC-to-AC ratio (inverter sizing)

Your solar panels produce DC power; the inverter converts it to AC. In most residential systems, the total DC nameplate wattage of the panels is intentionally larger than the AC rating of the inverter. This is called the DC-to-AC ratio (also called the inverter loading ratio).

For example, 8 kW of panels paired with a 6 kW inverter gives a DC:AC ratio of 1.33. This might seem counterintuitive — why install more panels than the inverter can handle?

Why oversize the array relative to the inverter

  • Panels rarely hit their nameplate rating — real-world conditions (temperature, soiling, angle of incidence) mean your panels almost never produce 100% of STC power simultaneously. A 1.2–1.3 DC:AC ratio captures more energy during morning, evening, and non-ideal conditions without significant clipping at peak.
  • Clipping losses are small — the amount of energy lost during the brief periods when panel output exceeds inverter capacity is typically 1–3% annually. The energy gained during all other hours more than compensates.
  • Cost efficiency — panels are cheap relative to inverters. Adding extra panels to an existing inverter is often more cost-effective than upgrading to a larger inverter.

Typical DC:AC ratios for residential systems range from 1.1 to 1.35. The optimal ratio depends on your specific location, orientation, and inverter characteristics. Higher ratios make sense in cloudier climates or on east/west-facing roofs where peak output is naturally lower. In very sunny locations with south-facing roofs, a lower ratio minimizes clipping.

Inverter manufacturers specify a maximum DC input power and voltage — you can’t exceed those limits regardless of the ratio. We calculate the optimal DC:AC ratio as part of the system design.

String sizing basics

If your system uses a string inverter (as opposed to microinverters), the panels are wired in series to form “strings.” Each string must be designed to stay within the inverter’s voltage and current windows under all expected conditions. This is one of the most critical engineering tasks in system design.

Maximum Voltage

Panel voltage increases in cold temperatures. The string’s open-circuit voltage (Voc) at the coldest expected temperature at your site must not exceed the inverter’s maximum input voltage — typically 500 V or 600 V for residential inverters. Exceeding this can damage the inverter and is a code violation.

Minimum Voltage

Panel voltage decreases in hot temperatures. The string’s operating voltage (Vmp) at the hottest expected temperature must stay above the inverter’s minimum MPPT voltage. If it drops below, the inverter can’t track the panels’ maximum power point and you lose production.

MPPT Range

The inverter’s MPPT (maximum power point tracking) input has a voltage window. Your string voltage needs to fall within this window under all operating conditions throughout the year. Most modern string inverters have multiple MPPT inputs, which allows you to run strings of different lengths or orientations.

Current Limits

Each MPPT input has a maximum current rating. The short-circuit current (Isc) of your string — adjusted for temperature and irradiance — must not exceed this. If you’re running parallel strings on the same MPPT, their combined current must stay within limits.

String sizing uses temperature-corrected voltage and current values based on the historical temperature extremes at your location (ASHRAE design temperatures). This is not guesswork — it’s a calculation using the panel’s temperature coefficients from its datasheet and the inverter’s input specifications. NEC 690.7 governs maximum voltage calculations for PV systems.

If you’re using microinverters, string sizing doesn’t apply in the traditional sense — each panel has its own inverter, and the AC output of each microinverter is combined on a branch circuit. The design constraints shift to branch circuit sizing and total system output limits instead.

Array layout and design considerations

Once you know the system size and inverter configuration, the next step is fitting the panels onto your roof (or ground-mount area) in a layout that satisfies production goals, code requirements, and practical installation constraints.

A good array layout accounts for:

  • Fire code setbacks and access pathways — covered in Chapter 3. The layout must leave required clear zones along the ridge, eaves, hips, and valleys per your AHJ’s adopted fire code.
  • Obstructions — panels must be placed around vents, chimneys, skylights, and other roof features with adequate clearance for both shade avoidance and maintenance access.
  • Panel orientation — panels can be mounted in landscape or portrait orientation. The choice affects racking layout, row spacing, and how many panels fit in a given area. Racking manufacturers specify which orientations their systems support.
  • Racking attachment points — roof attachments need to land on rafters or structural members, not just sheathing. The panel layout needs to align with the racking system’s span tables and the underlying roof framing.
  • Conduit routing — the layout should consider where conduit will run from the array to the inverter and from the inverter to the electrical panel. Shorter, more direct conduit runs mean lower material cost, less voltage drop, and a cleaner installation.
  • Aesthetics — panel placement should look intentional, not random. Aligned rows, consistent spacing, and clean edges make a difference in curb appeal and can matter for HOA approval.

The design deliverable for the layout is a roof plan (or site plan for ground mounts) drawn to scale, showing panel positions, setbacks, access pathways, conduit routes, and equipment locations. This is part of the plan set used for permitting and serves as the installation guide for your roofer and electrician.

The single-line diagram

Every solar permit application requires a single-line diagram (SLD) — a schematic that shows the electrical architecture of the entire system from panels to point of interconnection. It’s not a wiring diagram with every conductor; it’s a simplified representation showing the major components and how they connect.

A typical residential SLD includes:

  • PV modules with quantity, wattage, voltage, and current ratings
  • String configuration (how many panels per string, how many strings per MPPT)
  • DC disconnect (if required), rapid shutdown equipment, and any combiners
  • Inverter make, model, and ratings
  • AC disconnect, production meter (if required by utility), and overcurrent protection
  • Battery system and associated equipment (if applicable)
  • Wire sizes, conduit sizes, and overcurrent protection ratings for each segment
  • Grounding electrode conductor and equipment grounding details
  • Point of interconnection at the main electrical panel, including breaker size and busbar rating

The SLD is a critical part of the plan set we produce for our clients. Your inspector will check the installed system against this diagram, so it needs to be accurate and detailed. We’ll cover the electrical design elements in more detail in Chapter 6.

Common sizing mistakes to avoid

These are the errors we see most often when homeowners try to size their own systems without engineering support:

  • Using monthly averages instead of annual totals. Monthly usage varies significantly. A system sized for your July bill will be far too small for December, and vice versa. Always size based on 12 months of data.
  • Ignoring setbacks and usable area. Counting every panel that physically fits on the roof without accounting for fire code setbacks, access pathways, and obstructions. The usable area is always less than the total area.
  • Not accounting for system losses. Using the panel’s nameplate wattage and multiplying by sun hours without derating for temperature, soiling, inverter efficiency, and other losses. This overestimates production by 15–25%.
  • Significantly oversizing beyond net metering limits. If your utility caps net-metered systems at 100% of historical usage, designing for 130% means you’re paying for panels whose production you won’t get fair value for. Plan for future loads, but verify what your utility allows first.
  • Mismatching string length to inverter specs. Incorrect string sizing can result in a system that exceeds the inverter’s maximum voltage (dangerous and a code violation), drops below the MPPT window (lost production), or doesn’t start up in cold weather when you need it.

Every one of these mistakes is avoidable with proper engineering. That’s the value of getting the design right before you spend money on equipment — the design phase is where expensive errors are cheapest to fix.

Ready to size your system?

Bring your 12 months of usage data and your site details to a consultation. We’ll run the numbers, model production for your specific roof, and deliver a design that’s sized right — not too big, not too small, and built to code from day one.

Book a Consultation