0
\$\begingroup\$

Looking for the thinnest wire for very short distance power transmission.

Supply is 5V, 10A. Ambient temperature is 35C.

There are 5 consuming elements, each consuming 10W. The elements are linked together with wires, 10mm apart. Looking at wire tables, it seems that 20AWG is the correct wire to use but this is for "Chassis Power Transmission" and I can only assume it is more of a thumb rule.

Using a PCB trace calculator, for transmitting 10A over 10mm the recommended trace is 0.25mm^2 of copper, same is 23AWG wire. As distance is very short, voltage drop isn't a factor.

Is there a more accurate way to determine the minimum wire gauge other than "try and see if it works". enter image description here

\$\endgroup\$
5
  • \$\begingroup\$ Yes, they are rules of thumb - the actual limiting factors are voltage drop and temperature rise. \$\endgroup\$ – user253751 Feb 2 at 16:49
  • \$\begingroup\$ So you need to understand the thermal environment at the conductors themselves, not just the ambient. You also need to worry about devices that are placed close to the conductors and may also get hot. You need to do some thermal modeling. \$\endgroup\$ – Elliot Alderson Feb 2 at 16:52
  • \$\begingroup\$ You also need to consider mechanical factors. A PCB trace is completely supported from point to point. The board also conducts heat away and dissipated it to the surrounding air. For chassis wiring, people usually decide on a size wire that is adequate for most needs and don't try to determine the minimum for every use. \$\endgroup\$ – Charles Cowie Feb 2 at 17:24
  • \$\begingroup\$ Honestly, IMO I'd do the "try it and see if it works" method. That's not an ideal engineering practice, but it's tried-and-true and often is all you need. I think that would dial you into an answer much faster than theoretical modeling and calculations. Find the minimum that works, then bump up a size or two for some breathing room. YYMV If you're doing medical or aerospace, maybe you have to do better. If you're just building a one-off in the lab for experimenting ... that's a wholly different situation. \$\endgroup\$ – Kyle B Feb 2 at 19:34
  • \$\begingroup\$ At these very short wire lengths, consider that the CONTACT RESISTANCE will probably similar magnitude to the wire resistance itself. (i.e. where the wires make their connections is not zero ohms). In other words, calculating and modeling isn't gonna be easy as you have to take that into account. \$\endgroup\$ – Kyle B Feb 2 at 19:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.