PCB Trace IR Drop and Temperature Rise Calculator
Enter the current and the trace geometry. The calculator returns resistance, voltage drop against your rail’s tolerance budget, and temperature rise — solved iteratively, because hot copper is more resistive and drops more.
Runs entirely in your browser. Nothing is uploaded, stored, or sent anywhere.
Voltage drop, not current rating
Most trace calculators answer “how much current can this carry”, which is a thermal question with a well-known answer. This one answers the question that decides more designs: how much voltage did I lose, and can the load live with it. On a modern low-voltage rail that constraint bites long before the copper gets warm.
A 1.0 V core rail with ±3% tolerance has 30 mV to spend on everything — regulator setpoint error, load regulation, ripple, transient droop and the trace. Spending half of it on copper is a design decision, usually an accidental one.
R = ρL/A with ρCu = 1.72×10−8 Ω·m
· 1 oz copper ≈ 34.8 µm thickR(T) = R20(1 + 0.00393(T − 20)) ·
V = IR, P = I²RThe thermal feedback loop is real
Copper has a temperature coefficient of about 3900 ppm/°C — its resistance rises 0.39% per degree. A trace carrying current warms, its resistance rises, it drops more voltage, it dissipates more power, and it warms further. At modest currents this converges quickly and the correction is a few percent. At high current density it is the mechanism behind traces that pass on the bench and fail in a sealed enclosure at 60 °C.
The figures above are solved iteratively rather than evaluated at 20 °C, so the resistance shown is the hot resistance, not the datasheet one. A 30 °C rise means about 12% more resistance and 12% more drop than a cold calculation predicts.
The return path is half the circuit, and it is missing here
This calculates one conductor. The current has to come back, and on a board it returns through the ground plane, which has resistance too. If the plane is solid and wide the return drop is small; if it is slotted, crowded with vias, or the return is a trace rather than a plane, the return drop can equal or exceed the outbound one.
That is also where ground-bounce problems come from: the return drop appears as a voltage difference between the load’s ground and the regulator’s ground, which no amount of widening the power trace fixes. If accuracy matters, remote sense at the load closes the loop around both conductors at once and is nearly always the right answer.
Levers, in the order they usually pay
- Shorten it. Resistance is linear in length; moving the regulator is often easier than any copper change.
- Widen it. Also linear, and cheap on an outer layer where space allows.
- Use a plane, not a trace. A pour is dramatically lower resistance than any practical trace width.
- Heavier copper. Effective and costly — 2 oz halves the resistance but raises board cost and constrains minimum trace and space everywhere.
- Remote sense. Does not reduce the drop, but removes it from the error budget, which is usually what you actually needed.
More free tools
Each of these runs entirely in your browser. Nothing is uploaded, stored or sent anywhere, and none of them needs an email address.