2
\$\begingroup\$

I'm routing a PCB. Currently, all power traces are 20 mils. A QFP chip on the board requires multiple voltage rails to operate, I found the easiest solution is passing a power trace through a pin on the chip (the 1.8VI net in the screenshot).

PCB layout

However, it means the trace is as wide as the pin itself. Will it acting as a heatsink and create a solderability / thermal balance problem in hand soldering or reduce the yield in reflow soldering? Should I reduce the trace width when it's passing a pin for thermal relief?

\$\endgroup\$
2
  • 1
    \$\begingroup\$ I thought thermal relief was only for hand soldering and it isn't an issue with reflow where ambient temp is high. Can't heat sink when everything is the same temperature. I've not had problems using traces as wide as the pad when getting boards assembled via oven. \$\endgroup\$
    – DKNguyen
    Commented May 14, 2020 at 14:23
  • 1
    \$\begingroup\$ If you have a connection to a large internal plane you end up with the suface traces hotter than the internal ones. Meaning those pins could not flow as well as the others. \$\endgroup\$
    – Reroute
    Commented May 14, 2020 at 15:25

1 Answer 1

Reset to default
1
\$\begingroup\$

The traces do not connect to a large plane of copper nearby to the connection to the pin so it should not impact reliability. Think of it like if you double the thickness the trace should be twice as long to not be a larger heatsink

In your case the ground trace ends up going through a small trace to plane so the thermal relief is there. And while I cannot tell where the 1.8V trace is going. If its connection to plane is more than about 8mm away from the pin. For that thickness trace it should not effect things.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.