1
\$\begingroup\$

I'm trying to think of a way to decrease the temperature that my PCB is running at, but I don't want to add a heatsink if I don't have too since I want this board to have as low of a profile as possible. My board is a DC-DC converter with the Driver IC and my MOSFET's (NTTFS2D1N04HLTWG) reaching a max temp of around 70 degrees Celcius, and my Inductor (IHLP2525CZER1R0M01) reaching a max temp of 80 degrees Celsius.

This question is meant to be more conceptual, but would either increasing the copper pour of my PCB or increasing the copper weight be a good alternative to using a heatsink? My current PCB is using 1oz copper on the exterior, and 0.5oz on the interior. Here's my PCB for a reference.

enter image description here

enter image description here

\$\endgroup\$
7
  • \$\begingroup\$ You did not tell us what MOSFET package you are using. It is highly dependent on that. Why is your inductor so hot? Did you size it properly? \$\endgroup\$
    – DKNguyen
    Sep 15, 2021 at 1:28
  • \$\begingroup\$ @DKNguyen The package FET size I'm using are 3mmx3mm. I don't know why in my worst case application, the inductor get's that hot. The average current flowing is 3A \$\endgroup\$
    – Jay
    Sep 15, 2021 at 1:54
  • \$\begingroup\$ What MOSFET package specifically? What inductor? \$\endgroup\$
    – DKNguyen
    Sep 15, 2021 at 2:02
  • \$\begingroup\$ @DKNguyen I edited my post to include the FET and inductor PN \$\endgroup\$
    – Jay
    Sep 15, 2021 at 2:28
  • \$\begingroup\$ With that MOSFET you could benefit from multiple heavy layers of copper and thermal vias under and around the IC to dissipate the heat. Not too far though since there are diminishing returns. Maybe 1 square inch. If you are using the 10uH inductor in that datsheet you are operating it near its I^2R current so it would be no surprise it is as hot as it is. \$\endgroup\$
    – DKNguyen
    Sep 15, 2021 at 2:37

1 Answer 1

1
\$\begingroup\$

Generally yes, using 1oz or 2oz copper does help add thermal mass and reduces ground system sheet resistance.

One thing to be aware of when using thicker copper foil is that it affects fine-pitch traces - normally we assume the vertical shape of the etched traces is ideal, since the traces are more wide than thick. But very thin traces etched in relatively thick 2oz foil, start to have noticeable trapezoidal walls due to non-ideality of the etch process. PCB vendors generally know how to compensate to some extent, but if there are any 0402 or smaller components on the board, I'd stick with 0.5oz copper for the outer layer. In your example I don't see anything that looks like it would be a problem.

Side note about the inductor heating: check the datasheet for the inductor, there are two ways that the max current is rated. There is a magnetic saturation limit where the effective inductance is reduced by 10%, and there is a temperature limit where the component temperature reaches some threshold. Some vendors (like TDK or Sumida) give both limits separately, other vendors might list only one of these. Either way, if the inductor is getting that hot, it's likely undersized for the peak current. There is a practical limit to how much energy an inductor can store, ½LI². For your Vishay inductors, the Heat Rating test condition is based on a temperature rise of 40°C above ambient, so 70°C sounds like it is at or near heat saturation.

\$\endgroup\$
3
  • \$\begingroup\$ Just so I understand, is my inductor reaching a high temperature because it's too small in size to try to store the amount of energy it's given? And because of that heat is generated? \$\endgroup\$
    – Jay
    Sep 15, 2021 at 2:55
  • \$\begingroup\$ On each charge/discharge cycle, the SMPS controller ramps up the inductor current and then ramps it down again, so what matters here isn't the average current but rather the peak inductor current. Another way to look at it is that the SMPS uses the inductor like a bucket to scoop up some energy on each cycle, then dumps the energy into the output capacitor, transforming the voltage. Smaller scoop and larger average load current requires higher switching frequency, which increases inductor switching losses. \$\endgroup\$
    – MarkU
    Sep 15, 2021 at 3:07
  • \$\begingroup\$ If you search amazon/etc for Raspberry Pi accessories, you may find some 10mm x 10mm heat sinks with adhesive pads; sticking some small heat sink on the top of the inductor itself may help dissipate the excess heat. (Your inductor is a shielded type.) \$\endgroup\$
    – MarkU
    Sep 15, 2021 at 3:10

Your Answer

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

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