Selecting the Right DC Cable Size for Solar PV Arrays: A Technical Guide for Engineers

What Is This Calculation and Why It Matters

Selecting the correct DC cable size for a solar photovoltaic (PV) array is a foundational engineering decision that directly impacts system safety, efficiency, reliability, and compliance. Unlike AC systems—where voltage drop, thermal derating, and fault current considerations are well-documented in mainstream electrical codes—DC PV circuits present unique challenges: continuous unidirectional current flow, absence of zero-crossings (affecting arc sustainability), elevated operating temperatures due to rooftop mounting, and long service lifetimes (25+ years) under variable environmental stress.

An undersized cable leads to excessive resistive losses (I²R heating), unacceptable voltage drop across the string or combiner run, thermal degradation of insulation, accelerated conductor oxidation, and—in worst cases—fire hazards from sustained overheating or arcing faults. Oversizing, while seemingly conservative, incurs unnecessary material cost, installation complexity (e.g., conduit fill issues), and reduced flexibility during routing. Moreover, improper sizing violates key international and national standards—including IEC 60364-5-52, IEC 60287-1-1, and NEC Article 690—exposing designers and installers to liability and rejection during commissioning inspections.

The core objective of the DC cable sizing calculation is twofold: (1) ensure the conductor’s ampacity exceeds the maximum continuous current (including overcurrent protection margins), and (2) limit the steady-state voltage drop to ≤2–3% of nominal system voltage—typically 2% for critical commercial or utility-scale arrays per best-practice guidelines (e.g., IEEE 1547-2018 Annex D). Both criteria must be satisfied simultaneously; satisfying only one renders the design non-compliant.

Theory and Formula Walkthrough

The DC cable sizing process involves two parallel but interdependent calculations: ampacity verification and voltage drop verification. The larger of the two resulting cross-sectional areas (in mm²) governs the final selection.

1. Current Calculation

For DC PV circuits, power factor (PF) is irrelevant—DC has no phase angle. However, many calculators retain PF as an input to support hybrid or inverter-input (AC-side) calculations. For pure DC string or array wiring, set PF = 1.0. The design current is:

[ I_{\text{design}} = \frac{P_{\text{max}}}{V_{\text{dc}}} ]

Where:

• $P_{\text{max}}$ = Maximum continuous DC power (W), typically the STC-rated array power × 1.25 safety factor (per NEC 690.8(A)(3) for continuous load);
• $V_{\text{dc}}$ = Nominal DC system voltage (V), usually the open-circuit voltage ($V_{\text{oc}}$) at lowest expected ambient temperature (to account for voltage rise in cold conditions) — not the MPP voltage.

⚠️ Critical note: Using $V_{\text{mp}}$ instead of $V_{\text{oc}}$ (temperature-corrected) for current calculation risks undersizing cables for worst-case voltage—and thus current—scenarios. Per NEC 690.8(A)(1), ampacity must be based on the maximum circuit current, defined as 125% of the rated short-circuit current ($I_{\text{sc}}$) of the array.

2. Voltage Drop Calculation (DC)

For a single-conductor run (one-way length $L$), the total loop length is $2L$ (out-and-back). The voltage drop is:

[ \Delta V = 2 \cdot \rho \cdot \frac{L \cdot I_{\text{design}}}{A} ]

Expressed as a percentage:

[ %,\Delta V = \left( \frac{2 \cdot \rho \cdot L \cdot I_{\text{design}}}{A \cdot V_{\text{dc}}} \right) \times 100 ]

Where:

• $\rho$ = Resistivity of copper = 0.0172 Ω·mm²/m (at 20°C); for aluminum, use 0.0283 Ω·mm²/m;
• $L$ = One-way cable length (m) — not total circuit length;
• $I_{\text{design}}$ = Design current (A);
• $A$ = Cross-sectional area (mm²);
• $V_{\text{dc}}$ = Nominal DC system voltage (V).

Rearranging to solve for minimum required area $A_{\text{VD}}$:

[ A_{\text{VD}} = \frac{2 \cdot \rho \cdot L \cdot I_{\text{design}} \cdot 100}{%,\Delta V_{\text{max}} \cdot V_{\text{dc}}} ]

3. Ampacity-Based Sizing (IEC & NEC)

Per IEC 60287-1-1 Clause 4, the permissible current-carrying capacity $I_z$ depends on:

• Conductor material and cross-section;
• Insulation type (e.g., XLPE, EPR) and temperature rating (e.g., 90°C);
• Installation method (e.g., free air, buried, in conduit, bundled);
• Ambient temperature (derating applied if >30°C);
• Grouping factors (if ≥3 circuits in same conduit or tray).

NEC Table 310.16 provides base ampacities; NEC 690.8(A)(2) mandates that conductors be sized to carry not less than 125% of the maximum current (i.e., $I_{\text{sc}} \times 1.25$), and not less than 125% of the inverter’s maximum continuous output current for inverter output circuits.

Thus, minimum ampacity-based area $A_{\text{amp}}$ is determined by referencing standardized tables (e.g., IEC 60502-2 or NEC Table 310.16) after applying all applicable correction and grouping factors.

The final recommended cable size is:

[ A_{\text{req}} = \max\left(A_{\text{VD}},, A_{\text{amp}}\right) ]

Standard Requirements (Cited Clauses)

Compliance is non-negotiable—and context-dependent:

• IEC 60364-5-52 (Low-voltage electrical installations — Selection and erection of electrical equipment — Wiring systems): Section 525 mandates that “the cross-sectional area of conductors shall be selected so that the voltage drop does not impair the proper operation of equipment.” While it does not prescribe a universal % limit, Annex B recommends ≤3% for final subcircuits and ≤5% for distribution circuits. For PV-specific applications, IEC 62548:2016 (Photovoltaic arrays — Design requirements) explicitly recommends ≤1.5% for module-level wiring and ≤2% for array-to-inverter runs.

• IEC 60287-1-1 Clause 4: Specifies the generalized equation for current rating under steady-state conditions, incorporating conductor resistance, thermal resistances of insulation and surrounding medium, and ambient temperature. It requires explicit consideration of solar radiation absorption on rooftop-mounted cables—a frequently overlooked factor that can raise surface temperature by 15–25°C above ambient.

• NEC 2020 Article 690 (Solar Photovoltaic Systems):

  • §690.8(A)(1): Defines maximum circuit current as 125% of $I_{\text{sc}}$ (STC) — not nameplate power divided by voltage.
  • §690.8(A)(2): Requires conductors to be rated ≥125% of maximum circuit current.
  • §690.8(B)(2): Mandates overcurrent protection sized ≥125% of max circuit current (but ≤156% for certain fuse types).
  • §690.31(E): Requires PV source and output circuits to be installed in raceways, cables, or busways rated for sunlight resistance (e.g., USE-2, PV Wire, or RHH/RHW-2 with sunlight-resistance marking).

Failure to reference these clauses—not just general ampacity tables—constitutes non-compliance, even if voltage drop appears acceptable.

Common Mistakes and How to Avoid Them

1. Using MPP voltage instead of temperature-corrected $V_{\text{oc}}$

   • Risk: Underestimates worst-case voltage and overestimates current, leading to undersized conductors.
   • Fix: Calculate $V_{\text{oc, corrected}} = V_{\text{oc, STC}} \times \left[1 + \alpha_V \times (T_{\text{min}} - 25^\circ\text{C})\right]$, where $\alpha_V$ is the temperature coefficient of $V_{\text{oc}}$ (typically −0.3%/°C for crystalline Si). Use this $V_{\text{oc}}$ value to determine minimum conductor insulation voltage rating (e.g., 1.5× $V_{\text{oc, corrected}}$ per IEC 62548).

2. Ignoring ambient and solar heat gain derating

   • Risk: A 6 mm² Cu cable rated 50 A in free air at 30°C may derate to <35 A on a black rooftop at 55°C ambient + 20°C solar gain.
   • Fix: Apply NEC Table 310.15(B)(3)(c) (sunlight exposure adder: +15°C) and Table 310.15(B)(2)(a) (ambient temperature correction). For IEC, use IEC 60287-2-1 for soil/air thermal resistivity and IEC 60287-3-2 for solar radiation.

3. Neglecting DC arc fault behavior

   • Risk: Small conductors (<6 mm²) increase arc impedance instability, raising risk of sustained series arcs (per UL 1699B).
   • Fix: Follow NEC 690.11 (arc-fault circuit interruption) and prefer ≥6 mm² for strings >150 V; verify compatibility with AFCI devices.

4. Assuming ‘acceptable voltage drop’ is purely economic

   • Risk: 2% drop at 1000 V DC = 20 V loss; at 400 V DC = 8 V loss—but the power loss scales with $I^2R$. Lower voltage systems demand disproportionately larger cables.
   • Fix: Always compute absolute power loss: $P_{\text{loss}} = I_{\text{design}}^2 \times R_{\text{cable}}$. Compare against ROI thresholds (e.g., <0.5% of annual energy yield).

5. Omitting termination temperature ratings

   • Risk: A 90°C-rated cable terminated on a 75°C-rated breaker or fuse block forces derating to 75°C ampacity—often reducing usable current by 15–20%.
   • Fix: Match conductor insulation rating to all connected equipment temperature ratings; consult NEC 110.14(C).

Worked Example with Realistic Numbers

Scenario: A commercial rooftop PV array with 20 × 400 W bifacial modules (8 kW STC), configured in two 10-module strings. Each string has $V_{\text{oc}} = 48.5,\text{V}$, $I_{\text{sc}} = 12.2,\text{A}$ (STC). System voltage = 485 V DC (10 × 48.5 V). Lowest recorded ambient = −10°C. Cable run from array to inverter: 50 m one-way. Acceptable voltage drop = 2%. Conductor: Cu, XLPE, installed on rooftop tray (exposed to sun), ambient = 40°C.

Step 1: Determine design current

• Max circuit current = $I_{\text{sc}} \times 1.25 = 12.2 \times 1.25 = 15.25,\text{A}$
• Temperature-corrected $V_{\text{oc}}$: $\alpha_V = -0.30%/^\circ\text{C} \Rightarrow V_{\text{oc, min}} = 48.5 \times [1 + 0.003 \times (-10 - 25)] = 48.5 \times 0.895 = 43.4,\text{V}$ → String $V_{\text{oc}} = 10 \times 43.4 = 434,\text{V}$
• But for current calculation, we still use $I_{\text{sc}} \times 1.25$ — voltage affects insulation rating, not ampacity.

Step 2: Voltage drop–based area

• $I = 15.25,\text{A},, L = 50,\text{m},, V = 485,\text{V},, %\Delta V = 2%,, \rho = 0.0172$
• $A_{\text{VD}} = \frac{2 \times 0.0172 \times 50 \times 15.25 \times 100}{2 \times 485} = \frac{262.3}{970} = 0.27,\text{mm}^2$
• This seems tiny—but remember: this is only the voltage drop criterion. Ampacity dominates.

Step 3: Ampacity-based area

• Base ampacity for 4 mm² Cu XLPE (90°C) in free air = 51 A (IEC 60502-2)
• Derating: Ambient 40°C → factor = 0.88 (IEC 60287-2-0); Sun exposure → +15°C effective → use 55°C curve → factor ≈ 0.75; Bundling (2 strings) → factor = 0.82
• Total derating = $0.75 \times 0.82 = 0.615$
• Adjusted ampacity = $51 \times 0.615 = 31.4,\text{A} > 15.25,\text{A}$ → 4 mm² appears sufficient
• But NEC Table 310.16 (75°C column, THWN-2): 4 mm² = 35 A → derated for 55°C ambient: 0.58 → 20.3 A > 15.25 A ✓
• However, NEC 690.31(E) requires PV Wire or USE-2: 4 mm² PV Wire (90°C) = 55 A → derated to ~34 A → still adequate.

Step 4: Final selection

• $A_{\text{VD}} = 0.27,\text{mm}^2$, $A_{\text{amp}} \approx 4,\text{mm}^2$
• Recommended cable size = 4 mm² Cu (AWG 11)
• Verify voltage drop: $%\Delta V = \frac{2 \times 0.0172 \times 50 \times 15.25}{4 \times 485} \times 100 = 0.67%$ → well within 2%.

Conclusion: Despite low calculated $A_{\text{VD}}$, ampacity and mechanical robustness (minimum 6 AWG often specified for field durability) drive selection. In practice, many engineers specify 6 mm² (AWG 10) for 8 kW arrays to accommodate future expansion, simplify termination, and improve fault current handling—demonstrating that engineering judgment extends beyond formulaic outputs.

This guide synthesizes IEC, NEC, and industry best practices. Always perform site-specific thermal modeling, validate with manufacturer datasheets, and engage qualified PV system designers for final approval.