Trace Width for Pulsed Current

[MeerkatAirforce]

I am trying to figure out how to size a trace for a burst of current. All I can find online is resources which tell me how to size traces based on constant current based on the IPC-2221 standard.

I have a circuit which produces 27A for 6 milliseconds before a long break, and one which produces 15A for 4 milliseconds before a long break. Does anyone know how to appropriately size the width of a trace based on a pulse of current like this?

[mikeselectricstuff]

Thermal issues are likely to be fairly minimal at those pulse lengths, so probably best to just calculate based on your maximum allowable voltage drop.

[SiliconWizard]

Yes. For such short times and low duty cycle, what will really matter here is the fusing current. (I recommend this tool: https://saturnpcb.com/saturn-pcb-toolkit/ )
The fusing current essentially depends on trace width and copper thickness.

To give you an idea, a 5 mm-wide trace @18 µm copper weight can take almost 1000 A for 6 ms.

For the required trace width, I'd take the maximum width to stay well below the fusing current and the trace width required for passing the average current (according to your duty cycle) for the given trace length and the max temperature rise you have defined. Hope this is clear.