Width of pcb traces?

Is there a basic rule/formula to choose the width of a power-routing trace based on it’s max power? So, for example, what width to choose for a main 5v trace with 2A max - 10W (from the regulator) vs. 12v trace w. 3A max - 36 W? I’d like a general rule, but specific answers are welcome as well.

I’ve used this many times:


Just make sure you get all the parameters set correctly and you’re reading from the correct set of results, otherwise it can be confusing. The first few times I played with it, I didn’t change the default of 2 oz copper and I was looking at the internal trace results instead of external.

I think that’s exactly what I was looking for. Thanks ;D

You first have to answer the question “how hot am I willing to allow this trace to get”? You can make the trace as thin as you want and it will carry the current, but it will get hot and if it gets hot enough, it will melt and open (just like a fuse).

Obviously, you will also lose a lot of power in the trace if it does get hot, even if it’s not melting (it basically acts like a resistor).

So really your question is: “What width to choose for a main 5V trace with 2A max. AND maximum allowable temperature rise of XX degrees Celsius above ambient?” (the 5V doesn’t really matter…it’s all about the current)

You fill in the XX.

Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

(the 5V doesn’t really matter…it’s all about the current)

So I should be considering current, not power then? Out of the interest of my understanding of electronics theory, why is that?

The question is why you’re considering it at all. What are you worried about:

  • not losing “too much” power in the traces?
  • not allowing the traces to heat up?
  • not melting the traces?

Traces have resistance. Current through resistance causes power loss, and heat. That’s it. The several trace width calculators available on-line just help you compute how hot the traces will get (and how much power they dissipate) as a function of width, length, current, etc.

How you use that information is up to you.

Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

Yes well the reason I asked was because I’d rather not have my traces heat up and destroy themselves, so I was wondering what the general consensus was for width as a function of (what I thought was to be power) current. Of course the thickness of the trace would also matter, but I assumed this was kept to a standard in hobby-grade pcb fabrication.

What really controls the width on hobby (DIY Boards) I make myself is my method of fabrication.

On a typical board I make, I use 0.8mm for power traces, 0.5mm for data traces and 0.35 for traces that slip between standard DIP pin spacing.

With these sizes I seldom have issues with resist application (Toner Transfer) and don’t lose traces while etching (I have found that traces smaller than .3mm tend to disintegrate during the etching process I use.

So, Unless you start working with High Voltage or High Current, you should be able to use the sizes I use with relative success.

Thanks for the input, but 2 things:

I make use of fab-houses (i.e. BatchPCB), so I don’t etch at home.

And what would you consider high voltage/current in this case?

I am nowhere near a PCB expert, but on our amplifier systems at work were drawing ~3 amps on a 35v AC system and the main busses are reinforced with 1/16 inch thick nearly quarter inch wide copper strips (which makes rework a bitch)

Is it overkill? does that apply to a digital system? I dont know, but on a ac analog system that can sustain some heavy hits that is what is needed or else the traces get hot enough to bubble from the board and eventually fail due to time and air exposure