Engineering CalculatorsBusbar Temperature & Ampacity Calculator
Model a single tapered busbar (with axial heat conduction between sections of different cross-section) or a stack of parallel bars, then apply a steady load, a short-circuit fault, or a multi-step drive-cycle profile. Every formula used is shown below with your numbers substituted in.
| Section | Area (mm²) | Current density (A/mm²) | Rac (µΩ) | Loss (W) | Temp (°C) |
|---|---|---|---|---|---|
| Section 1 | 1,500 | 0.67 | 5.2 | 5.2 | 46.6 |
| Section 2 | 800 | 1.25 | 19 | 19 | 50.4 |
| Total | 2,300 | — | — | 24.2 | — |
Conduction between sections: Section 1→Section 2 -2.8 W
New to IEC 60287 / 60865-1? Read the guide: Busbar Heating: Ampacity to IEC 60287 and Faults to IEC 60865-1 — a plain-language explainer of the standard behind this calculator.
Current is entered as RMS (root-mean-square) throughout — continuous, fault, and every load-profile step — and used directly as I in P = I²·Rac, the standard formula for average AC (or DC) power dissipation in a resistor; no separate peak/RMS conversion is applied or needed, since RMS is defined precisely so that equation gives the correct heating effect. Steady-state and load-profile heating are solved with a nodal thermal network: each section (or the stacked-bar bundle, as one lumped node) generates I²R heat, exchanges heat with neighbouring sections by axial conduction (generalised fin equation with internal heat generation, discretised and solved with the Thomas algorithm), and loses heat to ambient by natural convection + radiation, in series with any coating's conduction resistance (t/(k·A), same generalised-fin-style discretisation). Skin effect uses the IEC 60287-1-1 formula. Load profiles are marched forward in time with backward-Euler integration (unconditionally stable) using each material's density and specific heat for thermal capacitance. Short-circuit heating uses the IEC 60865-1 adiabatic method per section (K = 226 A·√s/mm² copper, 148 aluminium; β = 234.5°C copper, 228°C aluminium) — conduction between sections is neglected for faults since it is slow relative to typical fault durations. Multi-bar bundles reduce exposed surface area on faces that face a narrow gap; proximity effect on resistance and PWM switching-ripple heating are not modelled. A section with "Apply conduction" ticked loses heat through an additional parallel path — thermal interface material + metallic section, same t/(k·A) conduction idiom as the coating — applied to exactly one face (width×length) of that section, which is correspondingly removed from its air-exposed area. The metallic section's outer face is taken to be directly at the coolant's inlet temperature (no separate coolant-film resistance term is modelled). Flow rate and coolant medium feed one exact, separate calculation (an energy-balance coolant temperature rise) that is informational only and not fed back into the busbar temperature result, which always uses the fixed inlet temperature as the coolant-side sink, the same way ambient air is always treated as a fixed-temperature reservoir. A coating or overmould is a conduction resistance t/(k·A) in series with the outer film, applied per-section (single mode) or to the whole conductor (bundle/bulk); a thick overmould is just a thick coating, using the bar's own surface area for the outer film (so enter the mould's true external area in bulk mode if it differs). Bulk (CAD) mode synthesises one equivalent conductor from an entered 20°C resistance, volume and surface area — reproducing both the resistance and the thermal mass exactly — and does not re-apply the AC skin factor, since a measured/field-solver resistance already embeds it. Resistance is reported both at the solved operating temperature and at a 20°C reference (the basis field solvers typically report), the two differing by the copper/aluminium resistivity temperature factor. For critical designs, verify against manufacturer test data and, where required, by test.
Validated: the IEC 60865-1 adiabatic short-circuit formula was independently re-derived from the documented equation and matched the calculator's output exactly (a 20 kA/1s fault on a 100 mm² copper section: 322.43°C both ways), and its inverse (minimum area for a given temperature limit) round-trips back to exactly 100 mm². DC resistance matches ρ/A by hand exactly, and the skin-effect factor correctly approaches ks = 1 (no skin effect) at low frequency — the same skin-effect and DC-resistance functions are reused unchanged by the Cable/Wire Sizing and PCB Trace Width calculators, both separately validated.