Dynamic PV String Voltage Analysis under Historical Meteorological Conditions.

MaxVoc Calculator performs dynamic PV string voltage analysis using real historical meteorological data and site-specific engineering conditions.

Instead of relying only on static worst-case assumptions, the methodology evaluates the actual coincidence between irradiance and temperature across up to 15 years of hourly climatic records to calculate the maximum number of modules per string.

Based on the simulation principles described in IEC 62548-1:2023 Annex F.1.1 Method a) — and informed by more than 20 years of international experience in utility-scale solar, applied physics, engineering mathematics and the analytical capability that large historical datasets now make possible — the platform extends the workflow into a fully dynamic engineering assessment using: historical meteorological datasets · irradiance transposition · ambient and cell temperature modelling · module I-V curves and Voc-GII logarithmic relationships · bifacial rear irradiance correction · thermal operating models (NMOT / NOCT) · degradation and manufacturing factors · project-specific design parameters · configurable engineering safety margins.

The workflow integrates:

Module characterisation — I-V curves and Voc-GII logarithmic relationships
Historical meteorological datasets — NASA POWER hourly records (up to 15 years)
Irradiance transposition — GHI → GII for SAT, fixed tilt, E/W and rooftop systems
Bifacial rear irradiance correction
Hourly Voc assessment under operational and non-operational thermal conditions
CSV engineering export — full hourly records, voltage matrices, inputs and methodology summary
Indicative inverter MPPT voltage assessment
DC BOS cost impact analysis for optimised string sizing

The result is a transparent engineering workflow for evaluating the maximum PV system voltage, maximum number of modules per string, operational voltage distributions, and indicative inverter operating ranges — based on the real historical operating environment and conditions of the selected location.

On this page

The problem The solution · 6 steps Special cases Worked example Cost Impact ↗

The problem

Traditional design methods calculate maximum string voltage using minimum expected temperature and full STC irradiance — a combination that rarely occurs in practice.

Most national PV standards — IEC 62548-1, AS/NZS 5033, NEC 690.7, UNE-HD 60364-7-712 and their regional guidelines — derive maximum string voltage from minimum expected ambient temperature, corrected for the temperature coefficient of Voc. The formula is straightforward, consistent and universally applied.

The limitation is a physical one. Minimum ambient temperatures typically occur at night or at first light/nightfall hours, during winter, when incident irradiance is low. At low irradiance, module Voc is logarithmically reduced relative to its STC value. The static method does not account for this — it applies full STC irradiance at minimum temperature, which is rarely the actual worst-case condition. Critical situations where irradiation is significant (above 300 W/m²) and temperature is still low need to be evaluated diligently.

In solar PV projects, specially utility-scale, this conservatism directly increases string count, DC BOS quantities, combiner boxes and isolators count, cable lengths, installation complexity and commissioning.

What the standards assume

Voc at minimum ambient temperature with full STC irradiance (1000 W/m²)

→ Historical records indicate that these conditions rarely coincide in practice

What actually happens

Minimum ambient temperatures typically occur near dawn, when GII is low

→ Voc is logarithmically reduced at low irradiance

The solution

IEC 62548-1:2023 Annex F.1.1 already recognises the limitation — and defines historical simulation as a primary normative method.

IEC 62548-1:2023 Annex F.1.1 defines three normative methodologies for assessing maximum PV string voltage under low-temperature operating conditions.

IEC 62548-1:2023 Annex F.1.1 — Normative Calculation Pathways

Method a) — Simulation using historical site data, low annual temperatures and low irradiance conditions
This methodology evaluates PV string voltage using historical meteorological datasets and irradiance-temperature coincidence. MaxVoc Calculator is based on the engineering principles described in this methodology, extending the concept into a more detailed dynamic workflow using historical hourly meteorological datasets, irradiance transposition, thermal modelling and Voc-GII relationships.

Method b) — ASHRAE Extreme Annual Mean Minimum Design Dry Bulb Temperature
This methodology uses conservative minimum design temperatures derived from ASHRAE climatic datasets together with simplified irradiance correction approaches defined by the standard. It remains a widely used engineering approach across the PV industry, although it does not evaluate the full dynamic coincidence between irradiance and temperature conditions over time.

Method c) — Applicable national regulations
This pathway allows the use of national electrical standards, regulations or authority-specific methodologies applicable within a given jurisdiction. In practice, many national methodologies remain highly conservative and are often based on static worst-case assumptions rather than dynamic climatic assessment.

The standard states in NOTE 2:

"Use of the absolute minimum temperature for a site may be too conservative, because it does not take into account the actual cell temperature and voltage at low irradiance conditions."

MaxVoc implements Method a) — record by record, across up to 15 years of hourly data. For each record:

GII is calculated from GHI using the transposition model
Cell temperature is estimated from ambient temperature and GII
Voc is calculated from the module equations characterised in Step 01
String voltage = Voc × number of modules
Result: pass or fail against the specified system voltage limit

Project engineers remain responsible for compliance with all applicable local standards, regulations and authority requirements regardless of the methodology used.

Every assumption is stated. Every parameter is sourced. The output is a calculated maximum module count per string — not a conservative estimate, but a result based on the actual historical operating envelope of the site.

Step 01

I-V Curves & Voc-GII

Logarithmic Voc equations from datasheet parameters. Any crystalline Si module.

Step 02

Meteorological Data

Up to 15 years of real hourly GHI and temperature from NASA POWER. Locations covered by NASA POWER datasets.

Step 03

Transposition Factor

GHI → GII for your array geometry. SAT trackers, fixed tilt, rooftop, E/W, backtracking.

Step 04

Bifaciality

Rear-face GII correction. Bifaciality factor and view factor model. Optional.

Step 05

Max Voltage Analysis

Modules per string calculated through hourly dataset evaluation. The output is a calculated maximum series module count.

Step 06

Calculation Export

Full CSV export — hourly voltage records, Voc matrices, all inputs, string count results and complete methodology summary. The primary engineering artifact for transparency, audit and independent verification.

Special cases

Where dynamic analysis is most useful.

Dynamic analysis is most useful where string design is constrained close to the system voltage limit — and where each additional module per string has a measurable impact on BOS quantities, plant architecture or yield.

⚡ 2P Trackers

Portrait dual-row trackers constrain module count per string to specific multiples. Gaining one or two additional modules per string directly reduces string count and DC BOS quantities across the entire plant.

▦ Prefabricated structures — East West Mount

High-density prefab structures with fixed module counts per block. If an East West Mount block is engineered around 30 modules per string but the static method limits you to 28, the result is lost capacity and unnecessary wiring complexity — repeated across every block in the plant.

◱ Geometric Constraints

Irregular parcels, terrain slopes and cable routing requirements that force non-standard string configurations.

⚙ Inverter Block Sizing

MPPT input voltage ranges and string-to-inverter ratio optimisation that benefit from maximum series modules.

Engineering Export

What the CSV export contains

Full hourly records for every year of meteorological data — irradiance, temperature, transposition factor, cell temperature, Voc non-operational and Voc operational for each record. Plus inputs summary, voltage matrix and methodology notes.

Date	Hour	GHI (W/m²)	GII (W/m²)	Tamb (°C)	Tcell (°C)	Voc_notop (V)	Voc_op (V)
27-Feb-2023	08:00	187	267	6.9	9.2	48.91	47.83
12-Jan-2021	08:00	98	141	6.8	8.6	48.74	47.68
03-Jan-2023	07:00	52	98	7.1	8.1	48.52	47.49
08-Jan-2021	14:00	31	52	7.2	7.8	48.31	47.28
... and 87,644 more hourly records analysed

Sample data — Madrid, Spain · LONGi LR8-66HYD-650M · SAT tracker · 2014–2023 · Values per module (single string)

Worked example — String count comparison

Static: 28 modules/string → Dynamic: 30 modules/string

476 fewer strings · Indicative total saving — 100 MWdc reference project

Terms of Use & Intellectual Property

MaxVoc Calculator is free to use for legitimate engineering analysis. For professional use, outputs should be reviewed by a qualified engineer before use in formal design, procurement, authority submissions or investment decisions. The platform design, workflow architecture, calculation methodology, module database and engineering explanations are the intellectual property of MaxVoc Calculator. Unauthorised reproduction, redistribution, rebranding or commercial reuse is prohibited. This tool is provided as-is without warranty; the authors accept no liability for engineering decisions made on the basis of its outputs. Use of this platform constitutes acceptance of these terms.

📂 Drop .PAN file here or click to browse

PVSyst .pan format. Electrical parameters auto-imported and locked. Thermal values remain editable.

Electrical parameters

Voc (V)

Isc (A)

Vmp (V)

Imp (A)

Temperature coefficients

β Voc (%/°C)

α Isc (%/°C)

Module power (Wp)

Thermal & degradation

NOCT (°C)

NMOT (°C)

LID (% of Voc)

Pmax tolerance (%)

I-V Curves to Voc-GII

The Logarithmic Voc–GII Model ▸

Plotting Voc values extracted from I-V curves across a range of temperatures and irradiance levels reveals a fundamental physical property of crystalline silicon modules:

Voc does not scale linearly with irradiance — it responds logarithmically.
Doubling GII from 50 to 100 W/m² increases Voc by a small fixed increment. Doubling again from 500 to 1000 W/m² increases it by the same small increment. This means the Voc at low irradiance is significantly lower than STC values suggest — and that is exactly the condition that occurs when minimum temperature is reached in practice.

The conventional temperature-correction formula applies the full STC Voc regardless of actual irradiance. In real conditions, minimum ambient temperature occurs near dawn in winter, when GII is very low. The module Voc at that moment is substantially lower than at STC — a reduction the static method ignores entirely.

Equation format — one per temperature step

          Voc  =  m  ·  ln(GII)  +  b
        

Parameter	Definition
Voc	Open-circuit voltage (V) — the output. One value for every GII × temperature combination.
m	Slope — how much Voc rises per doubling of irradiance. Slightly smaller at higher temperatures because Voc becomes less sensitive to irradiance as the cell warms.
ln(GII)	Natural logarithm of irradiance. This is what makes the relationship logarithmic — a 10× increase in GII adds ln(10) ≈ 2.3 to the argument, not 10×.
b	Intercept — baseline Voc extrapolated to 1 W/m² (where ln(1) = 0). Drops linearly with temperature, driven by βVoc. This is the temperature-sensitive component.

Fitted equations — your module (8 equations · 1 per temperature step)

          — run Generate I-V Curves to see your module's equations —
        

Both m and b vary linearly with cell temperature across the −10°C to +25°C range. This is a key insight: it means that once the 8 per-temperature equations are fitted, m and b can be computed for any intermediate temperature — no interpolation gaps, no lookup tables. The result is a master equation: Voc(GII, Tcell) = mCalc(Tcell) × ln(GII) + bCalc(Tcell). In Step 05, this equation is applied to every hourly meteorological record — actual GII and actual ambient temperature — to compute the real string voltage at that moment.

⚡

No curves generated

Select a module preset or enter parameters, then click Generate.

01 I-V Curves — all temperatures overlaid

Temp:

GII:

The right graph is built from the end points of the I-V curves on the left · each coloured dot is one Voc value read at I = 0 A · lower temperature → higher Voc

I-V Curves

Voc extracted from I-V curves

03 Voc Table — extracted from I-V curves

Each row of this table is one Voc value extracted from the I-V curves at a specific temperature and irradiance. These values are then fitted to the logarithmic equation:
Voc = m · ln(GII) + b
m — how much Voc rises per doubling of irradiance. Slightly smaller at higher temperatures because Voc becomes less sensitive to irradiance as the cell warms.
b — baseline Voc at 1 W/m² (where ln(1) = 0). Drops linearly with temperature, driven by βVoc. This is what makes the curve shift downward as the module gets warmer.
Both m and b vary linearly with temperature, generating temperature-dependent Voc–GII equations across the full analysed operating range.

04 Master Equations — Voc for Any GII and Temperature

Fitting m and b from the 8 per-temperature equations reveals that both vary linearly with cell temperature. This gives two master linear equations — one for m, one for b — that together predict Voc at any irradiance and cell temperature without lookup tables or interpolation gaps. The result is a continuous, physically-grounded voltage model that can be evaluated at every hourly record in the meteorological dataset.

Master equations

mCalc(Tcell) — slope of the Voc-GII logarithmic curve at a given cell temperature. Slightly smaller at higher temperatures: as the module warms, Voc becomes less sensitive to changes in irradiance. This coefficient is fitted linearly from the 8 per-temperature slopes.
bCalc(Tcell) — baseline voltage at 1 W/m², where ln(1) = 0. Drops linearly with temperature — the primary temperature-driven component, governed by βVoc. This is what shifts the entire curve downward as temperature rises.
Voc(GII, Tcell) — the calculated open-circuit voltage for any irradiance and temperature combination. In Step 05, this equation is evaluated for every hourly meteorological record in the dataset — actual GII and actual ambient temperature — to determine whether the resulting string voltage exceeds the system voltage limit at any point in the historical dataset.
ln(GII) — Voc responds logarithmically to irradiance. A doubling of GII — say from 50 to 100 W/m², or from 500 to 1000 W/m² — adds the same fixed voltage increment. This means Voc at 100 W/m² is only modestly lower than at 1000 W/m²; the static method's assumption of full STC irradiance at minimum temperature is therefore systematically conservative.

05 Step Completed 01 — General Summary

Location

Latitude

Longitude

Dataset period

Years of data

Historical validation

Cross-check NASA POWER against ERA5 long-term record (Open-Meteo archive, 1940–present).

Meteorological Data — NASA POWER · MERRA-2

Hourly GHI and T2m · MERRA-2 reanalysis · ~0.5° × 0.625° · downloaded via NASA API

🌍

No data loaded

Enter coordinates and click Download & Analyse.

Dataset — from Step 02

Run Step 02 first.

Array geometry

Structure type

Max rotation (°)

Typical: 55°

Axis azimuth (°N=0)

0 = true N-S

Backtracking

Enable backtrackingAvoids inter-row shading at low sun angles

GCR — Ground Coverage Ratio

Utility-scale typical: 0.35–0.45

Tilt (°)

Azimuth (°N=0)

No backtracking — fixed structures do not track the sun.

Roof pitch (°)

Flush-mounted = roof pitch angle

Roof azimuth (°N=0)

Direction roof faces

No backtracking. Albedo below applies to roof surface reflection — typically lower than ground (use 0.10–0.15 for dark tiles).

Transposition settings

Min GHI for TF (W/m²)

Below this → TF = 1.0

Tilt (°)

Same angle for both East and West faces. Typical: 10–15°

Ridge azimuth (°N=0)

Direction the ridge runs. 0° = N-S ridge (faces exactly E and W). East face = ridge+90°, West face = ridge−90°.

GII and TF are computed independently for East and West faces. Both are retained for Step 05. East strings typically govern worst-case Voc in cold climates — coldest temperatures coincide with morning sun on the East face.

Transposition Factor — GHI → GII

The module receives GII (Global Incident Irradiance on the plane of array), not GHI. TF = GII / GHI is calculated for every hourly record using solar geometry and the Erbs decomposition model. Supported structures: single-axis tracker (SAT) with optional backtracking, fixed-tilt ground mount, and rooftop. The worst case — highest TF coinciding with lowest temperature — defines the critical Voc condition passed to Step 04.

☀

No transposition calculated

Complete Step 02 first, then configure array geometry and run.

03 TF Heatmaps — Average and Maximum (Hour × Month)

Average TF

Maximum TF

04 Coldest Daytime Records — GHI > 0

Date	Hour	GHI	GII	TF	Tamb (°C)

05 GII Calculation — Method & Traceability

GII = GHI × TF — step by step 1. GHI — Global Horizontal Irradiance from NASA POWER (W/m²). Measured on a horizontal surface.
2. Erbs decomposition — GHI is split into DNI (direct normal) and DHI (diffuse horizontal) using the clearness index k_t = GHI / GHI_{extraterrestrial}.
3. Solar geometry — Sun position (altitude, azimuth) is computed for each hour from latitude, longitude, and day of year.
4. Tracker/tilt geometry — Angle of Incidence (AOI) on the module plane is computed from sun position and structure parameters.
5. GII = DNI·cos(AOI) + DHI·(1+cos(tilt))/2 + GHI·albedo·(1−cos(tilt))/2
6. TF = GII / GHI — stored per record for full traceability.
→ Voc(GII, Tcell) = mCalc(Tcell) × ln(GII) + bCalc(Tcell) — applied in Step 05 using this GII.

TF Mode for Step 05: Actual hourly GII (geometric calculation — most accurate) GHI × TF_max (conservative — reproducible manually) GHI × TF_avg (average — intermediate)

Critical Records — GHI · TF · GII · Tamb (coldest 20 daytime instances)

Date	Hour	GHI (W/m²)	TF applied	GII (W/m²)	Tamb (°C)	Tcell NOCT (°C)	Voc/module (V)

06 Step Completed 03 — General Summary

Site & Array — from Steps 02 & 03

Complete Steps 02 and 03 first.

Ground albedo

Complete Step 02 to load NASA POWER albedo

Albedo value

Ground mount typical: 0.15–0.25 · Sand/gravel: 0.25–0.40 · Rooftop: 0.10–0.15

Bifacial rear irradiance

Apply bifacial correction Skip bifacial correction intentionally

Bifaciality factor (BF)

Ratio of rear-face to front-face Voc. Typical range: 0.65–0.80. From module datasheet.

Rear irradiance fraction (% of GHI)

What to enter:
Fraction of GHI reaching the module rear face.
· SAT tracker, GCR 0.3: 10–15%
· SAT tracker, GCR 0.4: 7–12%
· Fixed tilt, GCR 0.3: 5–10%
· Conservative default: 10%
Formula: GII_rear = (fraction/100) × GHI

Bifaciality — Effective GII Correction

Bifacial modules generate additional current from rear-face irradiance reflected off the ground or roof surface. This increases the effective GII seen by the cells — and therefore increases Voc. This step quantifies that additional irradiance hour by hour and produces a corrected GII_eff = GII_front + BF × GII_rear for every record.

☀

Bifaciality not yet calculated

Configure bifaciality parameters and click Calculate.
If your module is monofacial, click Skip.

>

Input Summary — Steps 01 → 04

Run Steps 01–03 first.

System Voltage Limit

Maximum system voltage

Calculation Scenario

Include LID correctionReduces Isc by LID% — marginal effect on string voltage

Include Pmax toleranceIncreases Voc by tol% — conservative

Transposition Factor (GHI → GII)

Using actual GII from solar geometry — most accurate, full traceability

Dynamic Voc Analysis

Dynamic Voc is evaluated across all hourly records in the historical dataset. The result is the maximum modules per string that comply with the selected voltage limit and modelling assumptions.

01 Conclusion — Maximum Modules per String

Maximum series modules per string

—

02 Method Comparison

IEC 62548-1:2023 Annex F.1.1 dynamic method vs conventional static formula (AS/NZS 5033 · NEC 690.7 · IEC 62548-1 Method b).

03 Indicative Vmpp Check — Inverter MPPT Range

Indicative estimate only. Estimated string Vmpp is calculated from:
Vmpp_STC × modules/string × temperature correction
This is a simplified engineering approximation for inverter MPPT sanity checking only. Not a dynamic Vmpp–GII model.
Do not use Voc operational values for MPPT validation.

Modules per string (M)

MPPT range Min V Max V Dashed lines show the MPPT window. Green bars = within MPPT range.

04 Full Input Summary

05 Voltage Matrices — All Scenarios

GII (W/m²) × Tamb (°C). Thermal correction applied internally. Red = exceeds limit · Amber = within 2%.

M =

Voltage concepts used in this analysis

Voc non-operational (NOCT): Open-circuit voltage when the module is connected but no current is flowing. Uses the NOCT thermal model. This is the conservative voltage for system limit compliance checks (e.g. 1500 Vdc).
Voc operational (NMOT): Open-circuit voltage immediately after opening a circuit that was previously operating, with current flowing. Uses the NMOT thermal model which reflects active operating conditions. This is still an open-circuit voltage — it is not Vmpp.
Vmpp: The voltage at the maximum power point of the I-V curve. It is a different electrical quantity and requires a separate Vmpp–GII–temperature model. Vmpp is not calculated by this tool's logarithmic Voc model and should not be inferred from Voc operational.

Voc — Not Operational · NOCT

GII (W/m²) ↓ · Tamb (°C) →

String voltage (V)

Instances exceeding limit (count)

String voltage vs GII

Voc Operational · NMOT

GII (W/m²) ↓ · Tamb (°C) →

String voltage (V)

Instances exceeding limit (count)

String voltage vs GII

Colour = temperature step (blue = −10°C → green/red = +25°C) Below limit Within 2% of limit Exceeds limit

About the author

JJ Ferrandis

CPEng Electrical · Engineers Australia · Sydney, Australia

Engineer working across utility-scale solar PV, BESS and renewable energy infrastructure for nearly 20 years internationally.

LinkedIn → Get in touch ☕ Buy me a coffee →

Over the years I have worked across engineering, project delivery, EPC contracting, manufacturers, development and commercial strategy, participating in projects ranging from distributed PV systems to large utility-scale renewable energy infrastructure.

A big part of my career has been spent at the intersection between:

· engineering

· construction

· commercial reality

· contractual risk

· and practical project delivery

which probably explains why I tend to approach engineering problems from both a technical and real-world perspective.

One thing that always fascinated me in solar engineering was how often critical design decisions still relied on simplified assumptions, even when huge amounts of climatic and operational data were already available. That curiosity eventually became MaxVoc Calculator. Not as a commercial product, but as an independent engineering project built in personal time to explore whether PV string voltage analysis could become:

· more transparent

· more data-driven

· easier to understand

· and better aligned with real operating conditions

The broader vision behind the platform is to help solar projects become more competitive while remaining:

· fully compliant with local and international standards

· aligned with OEM specifications and warranties

· acceptable to customers, lenders and insurers

· and ultimately technically robust and fully bankable

In essence, the objective is simple: help contribute, even in a small way, to a renewable energy industry that becomes progressively more rigorous, competitive and technically transparent. Hopefully one day methodologies like this will become more common within industry standards and engineering workflows themselves. Optimistic? Yes. Unrealistic? I do not think so.

Outside the software itself, my professional background includes:

· utility-scale solar PV and BESS

· HV substations and grid connection

· EPC delivery and contracting strategy

· performance guarantees

· technical-commercial risk allocation

· DC electrical design

· engineering due diligence and project execution

The platform remains free to use and continues to evolve through engineering feedback, industry discussion and ongoing refinement.

From the field

Construction phase site visit — New Zealand

Construction phase — New Zealand

East West Mount system

Tracker and fixed-tilt array — Australia

Engineering background

Utility-scale solar PV BESS HV substations EPC contracting DC string voltage analysis System electrical design IEC 62548 / AS5033 / NEC 690 Computational methodology Contractual risk allocation Grid connection Performance guarantees Data-driven engineering workflows

About this tool

This methodology has been developed as a computational implementation of IEC 62548-1:2023 Annex F.1.1 Method a), using historical meteorological datasets and logarithmic Voc modelling.

The I-V curve and Voc modelling approach used in MaxVoc Calculator is grounded in the same underlying semiconductor physics and PV modelling principles used by recognised industry frameworks, including the Sandia PV Array Performance Model (SAPM, King et al., SAND2004-3535), the De Soto single-diode model, NREL's System Advisor Model (SAM) and the open-source pvlib-python library.

Rather than requiring specialised outdoor characterisation coefficients available only for a subset of historical modules, MaxVoc Calculator uses standard manufacturer datasheet parameters. This allows the methodology to be applied to a much wider range of commercially available PV modules without loss of physical rigour.

Free platform. Transparent calculations. Engineering-driven methodology.

Feedback, constructive discussion and technical review are always welcome.

LinkedIn → Get in touch ☕ Buy me a coffee →

Support this project — buy me a coffee

Buy me a coffee ☕

This tool is free and will remain free. If it has saved you time or improved your design, a coffee is always welcome.

☕ Buy me a coffee

Terms of Use & Intellectual Property

MaxVoc Calculator is free to use for legitimate engineering analysis. All outputs must be independently reviewed by a qualified engineer before use in design, procurement, authority submissions or investment decisions. The platform design, workflow architecture, calculation methodology, module database and engineering explanations are the intellectual property of MaxVoc Calculator. Unauthorised reproduction, redistribution, rebranding or commercial reuse is prohibited. This tool is provided as-is without warranty; the authors accept no liability for engineering decisions made on the basis of its outputs. Use of this platform constitutes acceptance of these terms.

MaxVoc Calculator · Reference

Definitions, Terms and Acronyms

Reference terminology for the dynamic Voc methodology. All terms are used consistently throughout the calculation steps and align with IEC 62548-1:2023, AS/NZS 5033:2021, NEC 690 and equivalent national standards.

References & Technical Sources

Standards

AS/NZS 5033:2021
IEC 62548-1:2023
IEC 61215
IEC 61853-2
UNE-HD 60364-7-712

Meteorological Datasets

NASA POWER (MERRA-2)
Open-Meteo ERA5

Industry References

King et al. (2004) — Sandia PV Array Performance Model (SAPM), SAND2004-3535
De Soto et al. (2006) — Improvement and validation of a model for photovoltaic array performance
NREL System Advisor Model (SAM)
pvlib-python — open-source PV modelling library
PVsyst documentation

Terms of Use & Intellectual Property

MaxVoc Calculator is free to use for legitimate engineering analysis. All outputs must be independently reviewed by a qualified engineer before use in design, procurement, authority submissions or investment decisions. The platform design, workflow architecture, calculation methodology, module database and engineering explanations are the intellectual property of MaxVoc Calculator. Unauthorised reproduction, redistribution, rebranding or commercial reuse is prohibited. This tool is provided as-is without warranty; the authors accept no liability for engineering decisions made on the basis of its outputs. Use of this platform constitutes acceptance of these terms.

How to use this tool

Instructions & User Guide

MaxVoc follows a sequential six-step workflow — each step feeds into the next.

Mandatory: Steps 01, 02, 03
Optional: Step 04 (Bifaciality — skip for monofacial modules)
Output: Step 05 calculates the maximum series module count · Step 06 estimates DC BOS cost impact

STEP 01

I-V Curves & Voc-GII

Concept — how I-V curves become Voc values

The right graph is built from the end points of the I-V curves on the left. Each I-V curve ends at Voc where current = 0 A. At lower temperature, the curve extends further right — higher Voc. Those end points, repeated across multiple irradiance levels, form the Voc-GII relationship shown in Step 01.

What it does

Builds the mathematical Voc model for your module from datasheet parameters. Generates I-V curves across temperature and irradiance, extracts Voc values, and fits logarithmic equations of the form Voc(GII, T) = m(T)·ln(GII) + b(T).

I-V curves are constructed using the two-point scaling method derived from the single-diode equation — the same physical basis as the Sandia PV Array Performance Model (King et al., SAND2004-3535) and the De Soto five-parameter model used in NREL's SAM and PVLib. Isc scales linearly with irradiance and temperature (via αIsc); Voc scales logarithmically with irradiance and linearly with temperature (via βVoc and the thermal voltage Ns·kT/q). This is the standard approach for Voc evaluation from datasheet parameters when full outdoor characterisation data is unavailable.

MaxVoc Calculator uses standard manufacturer datasheet parameters rather than specialised outdoor characterisation coefficients. This allows the methodology to be applied to the full range of commercially available PV modules, including current-generation products for which laboratory-derived model coefficients do not exist in public databases.

Inputs required

Voc STC (V), Isc STC (A), temperature coefficient βVoc (%/°C), αIsc (%/°C), NOCT (°C), LID (%), Pmax positive tolerance (%). All values from the module datasheet. PAN files (PVSyst format) can be uploaded to auto-fill parameters.

Notes

LID: typically 2% for monocrystalline Si, 0% for bifacial N-type TOPCon. Applying LID reduces the calculated Voc and increases the allowable module count — it is the less conservative assumption for voltage compliance. Omitting LID gives a higher Voc and is more conservative.
Pmax tolerance: enter as a percentage (e.g. 3 for +3%). Applied conservatively as a full Voc uplift: Voc_eff = Voc_STC × (1 + tol/100). In practice the Voc component is lower — using the full % is a conservative upper bound
If your datasheet states Voc tolerance explicitly (e.g. ±3%), use that value directly in the same field
If the datasheet shows a Voc range, use the higher value
Built-in presets available for common modules

STEP 02

Meteorological Data

What it does

Downloads up to 15 years of hourly GHI and ambient temperature (T2m) from NASA POWER (MERRA-2 reanalysis) for your project coordinates. Identifies the coldest daytime instances, which govern worst-case Voc.

Inputs required

Latitude and longitude in decimal degrees (negative values for South and West). Number of years: 10 years is standard. 15 years is recommended for cold-climate or high-altitude sites.

Notes

Requires an internet connection
NASA POWER returns hourly averages — instantaneous temperatures within an hour may be lower than reported
Use the Historical Validation (ERA5) to cross-check the dataset minimum against the absolute site record

STEP 03

Transposition Factor

What it does

Converts GHI (Global Horizontal Irradiance) to GII (Global Incident Irradiance on the plane of array) for your specific array geometry, hour by hour across the full dataset. Uses solar geometry and the Erbs decomposition model.

Structure options

Single-axis tracker (SAT) with optional backtracking and GCR. Fixed tilt. Rooftop. East-West back-to-back (East West Mount, 2P). Azimuth: 0° = North, 180° = South. The tool auto-suggests the optimal value based on hemisphere.

Notes

For E/W structures, GII is computed independently for East and West faces
East strings typically govern worst-case Voc in cold climates — coldest temperatures coincide with morning sun on the East face

STEP 04

Bifaciality Optional

What it does

Corrects GII for rear-face irradiance in bifacial modules. GII_eff = GII_front + BF × GII_rear. A higher effective GII increases Voc, so skipping this step for bifacial modules is non-conservative.

Inputs

Bifaciality factor BF (from datasheet, typically 0.65 to 0.80). Rear irradiance model: View factor (recommended, uses albedo from Step 02) or Fixed fraction (percentage of GHI).

Skip if

Module is monofacial. Use the "Skip (monofacial module)" button to pass GII_front directly to Step 05 without rear correction.

STEP 05

Max Voltage Analysis

What it does

Evaluates Voc for every hourly record in the dataset using the module equations from Step 01 and the GII values from Steps 03 and 04. Applies the NOCT thermal model to estimate cell temperature. Finds the worst-case Voc and calculates the maximum number of modules per string that does not exceed the specified voltage limit within the analysed dataset.

Inputs

System voltage limit: 1500 V (IEC / NEC utility scale) or 1000 V (AS/NZS 5033 · NEC 690 residential). LID correction toggle. Max TF toggle (conservative or average transposition factor).

Output

Maximum modules per string
Worst-case Voc — NOCT and NMOT conditions
Method comparison vs conventional static formula (AS/NZS 5033 · NEC 690.7 · IEC)
Voc vs GII charts
Voltage matrices
Full input summary
String count impact estimate

STEP 06

Cost Impact

What it does

Estimates the DC BOS cost saving enabled by the dynamic method relative to the conventional static result. Compares the number of strings required under each approach and quantifies the difference in cable, fuse and combiner quantities.

Inputs

Project DC capacity (MWp). Module power (Wp, auto-filled from Step 01). Dynamic M (auto-filled from Step 05). Static M from the conventional method (entered manually). DC cable cost (AUD/m), average cable run (m/string), fuse pair cost (AUD).

Output

Strings saved
Estimated DC BOS saving — broken down by cable and fuse cost
Indicative total saving
Results are illustrative — depend on actual project topology and procurement

Key considerations

Azimuth convention

0° = North · 90° = East · 180° = South · 270° = West
Southern Hemisphere optimal: 0° (North-facing)
Northern Hemisphere optimal: 180° (South-facing)
The tool auto-sets the default based on latitude

NASA POWER data

MERRA-2 reanalysis · ~55 km grid resolution · hourly averages · 2001–2023
Not station data — spatial and temporal averaging means instantaneous extremes may not be fully captured
Cross-check with ERA5 via the Step 02 Historical Validation

Conservative vs realistic

Most conservative: LID OFF + Max TF ON — highest Voc, lowest module count
Less conservative (sensitivity): LID ON + Average TF — do not use for formal design without project design basis support

Formal engineering use

If used for design approval, grid connection, certification or authority submission — the calculation must be independently reviewed
Sign-off by a qualified engineer is required where mandated by the applicable jurisdiction

Terms of Use & Intellectual Property

MaxVoc Calculator is free to use for legitimate engineering analysis. All outputs must be independently reviewed by a qualified engineer before use in design, procurement, authority submissions or investment decisions. The platform design, workflow architecture, calculation methodology, module database and engineering explanations are the intellectual property of MaxVoc Calculator. Unauthorised reproduction, redistribution, rebranding or commercial reuse is prohibited. This tool is provided as-is without warranty; the authors accept no liability for engineering decisions made on the basis of its outputs. Use of this platform constitutes acceptance of these terms.

Methodology

Assumptions & Limitations

This page documents the main modelling assumptions, scope boundaries and review considerations for MaxVoc Calculator. These should be understood before using the output for design, compliance or approval workflows.

Intended Use

This tool is intended for preliminary design, engineering assessment, comparative analysis and IEC 62548 Annex F.1.1 engineering workflows.

The tool provides a computational engineering assessment. It does not replace project-specific engineering review, authority approval, manufacturer review or formal compliance certification where these are required.

Engineering Review Status

Validation status

Internal engineering checks completed. Independent third-party certification has not been completed.

Benchmarking status

Benchmarked internally against conventional static methods and reference calculations.

Professional use

Outputs should be reviewed by a qualified engineer before use in formal design, procurement, compliance or authority submissions.

Modelling Assumptions

Meteorological data

NASA POWER / MERRA-2

~55 km grid resolution — not station data
Hourly averages only — sub-hourly transients not captured
Dataset extends to approximately 2023
Future climate extremes are not represented
ERA5 cross-check is an independent estimate but also a reanalysis product

Thermal model

NOCT / NMOT steady-state

Formula (non-operational, open circuit): T_cell = T_amb + (NOCT − 20) / 800 × GII
Formula (operational, at MPP): T_cell = T_amb + (NMOT − 20) / 800 × GII
Steady-state approximation per IEC 61215 — thermal lag, wind speed and mounting configuration not modelled
NOCT is used for Voc non-operational (conservative, no current flowing — for voltage limit compliance)
NMOT is used for Voc operational (current was flowing before circuit opened — secondary engineering output, not Vmpp)

Module model

Single-diode / crystalline Si

Applicable to crystalline silicon: monocrystalline, polycrystalline, TOPCon, HJT
Thin-film modules (CdTe, CIGS) are supported when correct datasheet parameters are entered. Temperature coefficients in thin-film may exhibit greater irradiance dependence than the constant-βVoc approximation assumes — results are valid for engineering estimation
Pmax positive tolerance applied as a full Voc uplift: Voc_eff = Voc_STC × (1 + tol/100) — conservative; consistent with IEC 62548-1:2023 and AS/NZS 5033:2021
LID applied as a static Voc reduction to the characterised equations

Transposition model

Erbs decomposition / isotropic sky

Erbs empirical decomposition model
Row-to-row shading not modelled — transposition applied at cell plane without inter-row corrections
High TF values can occur at low irradiance (dawn geometry) — do not imply high incident irradiance
SAT backtracking uses an idealised algorithm — actual tracker performance may differ

Bifaciality

Step 04, simplified screening

Simplified view factor model: GII_rear = albedo × GHI × view_factor, or fixed fraction of GHI
First-order screening only — not a full bifacial energy simulation
Significant Voc differences between monofacial and bifacial configurations are possible — review per project

Scope and approval

IEC 62548-1:2023 Annex F.1.1

Provides a computational framework for IEC 62548-1:2023 Annex F.1.1 Method a)
Provides a transparent, traceable computational basis for IEC 62548-1:2023 Annex F.1.1 Method a) compliance workflows. Project-specific authority approval, manufacturer review and formal certification remain separate processes
Where formal third-party validation is required for a specific jurisdiction or project, that remains a separate engagement
Where a signed engineering report is required by local standards, that requirement exists independently of this tool

Key Limitations

Spatial resolution

NASA POWER grid resolution is approximately 55 km. Results may not represent localised microclimatic conditions such as valley frost, coastal effects or urban heat islands.

Temporal resolution

Hourly records only. Sub-hourly temperature transients and irradiance spikes are not captured. Cold spells under one hour are not resolved.

Voc tolerance

Pmax positive tolerance is available as an optional input in Step 01. Default is 0% (not applied). Where applied, it is modelled as a full Voc uplift — conservative upper bound.

Dataset period

The methodology evaluates historical data within the selected period (2001 to 2023). Results represent the analysed period only. Future climate extremes beyond the historical dataset are not represented.

Technology scope

Crystalline silicon only (mono, poly, TOPCon, HJT). Thin-film modules (CdTe, CIGS) are supported when correct datasheet parameters are used. Temperature coefficients may exhibit greater irradiance dependence than assumed by a constant βVoc — results remain valid for engineering estimation.

Shading not modelled

Row-to-row, near-field and far-field shading not modelled. The transposition model operates at cell plane level. Projects with significant shading exposure require additional analysis.

Normative basis

This tool implements IEC 62548-1:2023 Annex F.1.1 Method a): simulation using historical site data, considering low annual temperatures and low irradiance conditions. This is one of three normative calculation paths defined by the standard.

Compliance with local electrical standards (AS/NZS 5033, NEC 690, or national equivalent) remains the responsibility of the project engineer.

Future development may include implementation of the Sandia module temperature model (King et al., SAND2004-3535), which incorporates wind-speed effects and module mounting characteristics. The current version uses the NOCT and NMOT thermal models, consistent with the level of information typically available from manufacturer datasheets.

Terms of Use & Intellectual Property

MaxVoc Calculator is free to use for legitimate engineering analysis. All outputs must be independently reviewed by a qualified engineer before use in design, procurement, authority submissions or investment decisions. The platform design, workflow architecture, calculation methodology, module database and engineering explanations are the intellectual property of MaxVoc Calculator. Unauthorised reproduction, redistribution, rebranding or commercial reuse is prohibited. This tool is provided as-is without warranty; the authors accept no liability for engineering decisions made on the basis of its outputs. Use of this platform constitutes acceptance of these terms.

Financial Impact — Worked Example

Static method: 28 modules per string.

Dynamic method result: 30 modules per string.

100 MWdc project · 500 Wp modules · 1500 V system voltage · SAT tracker · 6mm² DC cable, 100 m circuit full length

String Count Impact

Method	Modules/string	Total modules	Strings required
Static (conventional — AS/NZS 5033 · NEC 690.7 · IEC)	28	200,000	7,143
Dynamic (IEC 62548-1:2023 Annex F.1.1)	30	200,000	6,667
Reduction	+2 mod/str	—	476 fewer (6.7%)

Indicative DC BOS Component Breakdown

Component	Basis	Unit reference	Indicative impact
DC cable reduction	6mm² cable, 100m homerun	A$9.5/m	A$452k
String fuse reduction	Both legs, 20A	A$35/string	A$17k
DC combiner box reduction	29 boxes eliminated (16 str/box)	A$1,800/box	A$52k
Installation impact	Labour, termination, testing, commissioning	A$120/string	A$57k
Total indicative DC BOS impact			A$578k

Cable unit costs are indicative reference values only: 4mm² ~A$6.0/m, 6mm² ~A$9.5/m, 10mm² ~A$15.0/m round-trip installed. Apply project-specific rates for commercial assessments.

Strings eliminated

476

6.7% fewer DC strings · same DC capacity

DC BOS impact

A$578k

Indicative, 100 MWdc reference project

Scales linearly

+2

Extra modules per string without exceeding 1500 V

Numbers are illustrative. Actual impact depends on project topology, procurement and site conditions. Run the full analysis — Step 06 generates a project-specific sensitivity model using your module and meteorological data.

Terms of Use & Intellectual Property

MaxVoc Calculator is free to use for legitimate engineering analysis. All outputs must be independently reviewed by a qualified engineer before use in design, procurement, authority submissions or investment decisions. The platform design, workflow architecture, calculation methodology, module database and engineering explanations are the intellectual property of MaxVoc Calculator. Unauthorised reproduction, redistribution, rebranding or commercial reuse is prohibited. This tool is provided as-is without warranty; the authors accept no liability for engineering decisions made on the basis of its outputs. Use of this platform constitutes acceptance of these terms.

Step 06 — Cost Impact

DC BOS Reduction Sensitivity Model

Enter assumptions to estimate the indicative DC BOS impact of the string count difference between the dynamic and static methods. All values are indicative and should be verified against project-specific procurement and installation data.

Module Wattage

—

Dynamic M (Step 05)

—

Static M

enter below

Currency & Project Parameters

Currency

Project DC capacity (MWp)

Module power (Wp)

Dynamic M (modules/string)

Static M (modules/string)

DC BOS Assumptions

DC string cable size

Average string homerun length (m)

Round-trip positive and negative DC conductors from string termination to combiner box or inverter

DC topology

Fuse configuration

Fuse rating

Fuse cost per string (installed)

Total per string including fuse holders — auto-updated by rating and configuration

Include installation impact

Enter Static M (modules/string) to calculate the string count impact.

Step 07 — Operating Condition Review

Manufacturer Operating Condition Review Request

Generate a structured technical request to submit to equipment manufacturers. The request communicates the assessed operating voltage envelope and asks whether it affects product applicability, warranty conditions or published operating limits. The engineering assessment remains the responsibility of the project designer.

Equipment Type

Select the equipment category for which you are preparing the review request. Each category generates equipment-appropriate framing.

Manufacturer & Equipment Details

Manufacturer name

Equipment model / series (optional)

Rated DC voltage (V) (from datasheet)

Designer / organisation

Engineering Assessment Summary auto-filled from Steps 01–05

Complete Step 05 to populate.

📋

Review request will appear here

Select equipment type, confirm details, then click Generate Review Request.

Welcome to MaxVoc Calculator

Enter your email to download

Enter your details to download

MaxVoc Calculator is designed for desktop use

Dynamic PV String Voltage Analysis under Historical Meteorological Conditions.

Traditional design methods calculate maximum string voltage using minimum expected temperature and full STC irradiance — a combination that rarely occurs in practice.

IEC 62548-1:2023 Annex F.1.1 already recognises the limitation — and defines historical simulation as a primary normative method.

Where dynamic analysis is most useful.

JJ Ferrandis

Definitions, Terms and Acronyms

Instructions & User Guide

Assumptions & Limitations

DC BOS Reduction Sensitivity Model

Manufacturer Operating Condition Review Request