0
\$\begingroup\$

I'm fairly new to electrical engineering, tried my hand at a power supply circuit based on the Texas Instruments buck converter TPS65261RHBR. Goal is to step down from a 12V lawnmower battery to produce 5V, 3.3V, and 1.8V to power a number of other circuits. Unfortunately my design doesn't seem to work, the output tracks don't have the right voltages, trying to get to a workable solution for my next prototype. Here's the schematic for my design, it's very similar to the reference design from TI's datasheet: http://www.ti.com/lit/ds/symlink/tps65261.pdf

Just got printed on a PCB, the results are bad: none of the outputs are producing the correct voltages, 3.3V/5V are 0V, the 1.8V is at 0.6V. I've tested all the input points are about 12.6V. Upon reviewing the datasheet it looks like inductor/cap selection isn't great (especially for the 5V output), but I'm worried this isn't the sole problem because the voltages are so far off, and the splitter resistors seem to be correct.

enter image description here enter image description here

\$\endgroup\$
9
  • 2
    \$\begingroup\$ Show the layout too, it's critical for an SMPS. \$\endgroup\$
    – user133493
    Jan 22, 2019 at 2:52
  • 1
    \$\begingroup\$ gasoline or electric lawn mover? \$\endgroup\$
    – Jasen Слава Україні
    Jan 22, 2019 at 3:31
  • \$\begingroup\$ did you measure input current, are the capacitors short-circuited? \$\endgroup\$
    – Jasen Слава Україні
    Jan 22, 2019 at 3:33
  • \$\begingroup\$ @replete Adding layout. It's a bit different than the reference layout, mostly in that I used vias to ground rather than top ground planes connected to the inner ground plane \$\endgroup\$
    – James Suffolk
    Jan 22, 2019 at 16:50
  • \$\begingroup\$ @Jasen gas, I didn't check input current \$\endgroup\$
    – James Suffolk
    Jan 22, 2019 at 16:51

2 Answers 2

Reset to default
0
\$\begingroup\$

Judging by the footprint of the 3 L's your DCR is too high.[mΩ]

Just like in real estate the 3 L's are critical for L5,6,7 DCR specs.;)

Rules of Thumb

L/R >> 1/fs switching rate and DCR <<= RdsOn of FET

Suggested

Manufacturer Wurth Electronics Inc.
Manufacturer Part Number 744311470
Description FIXED IND 4.7UH 6A 19.5 mOHM SMD
Current - Saturation 7A
6.90mm x 6.90mm

ALways review the design selection manual for the EVAL board before making any design.
Always include all {critical LC} BOM parts, design specs and test point results, before asking an SMPS question

I am guessing from layout you chose an 805 inductor or 2mm part 4.7UH 1.1A 175 mOHM which is about 10x too high and the size needs to be 3x bigger in square.

Also for e-Caps, ESR <<= RdsOn and ESR*C << 1/fs

For higher power, a better design uses 10x the inductance transformer and a triple secondary winding but similar low DCR. This is how all PC PSU's are done with multiple outputs.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks, this is what I was looking for. I'll adjust my BOM and update this post with new design \$\endgroup\$
    – James Suffolk
    Jan 22, 2019 at 19:41
  • 1
    \$\begingroup\$ Make the inductor current loop area tiny. \$\endgroup\$
    – Tony Stewart EE75
    Jan 22, 2019 at 20:13
1
\$\begingroup\$

Better late than never, but here's the design after I corrected it. There were multiple problems. I switched to power inductors and refined the values of some of the components based on feedback I received here. I changed the layout to be closer to the reference design for the SMPS. But the biggest problem I think was that I forgot a pulldown resistor on the mode pin and connected it directly to ground.

Haven't yet remade the PCB with this design, but I think this should correct the issues.

enter image description here enter image description here

\$\endgroup\$

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.