-1
\$\begingroup\$

I designed a PCB with two LM61495 buck converters. They should produce a 3V3 and 5V output. However, for both circuits, the output is 12V. The picture below shows the output of one of the channels measured. All voltages measured are 1/10 of the actual voltages.

enter image description here

Given that I have two circuits with the exact same error, I rule out an manufacturing error. The schematics of the two circuits can be seen below.

enter image description here

enter image description here

I am not able to detect an error in those schematics that could explain the output. As far as I can see, the circuit is exactly like the example in the datasheet.

Do you have an idea what the problem could be? The waveform looks a bit suspicious to me (not a normal ripple) but I can not deduce an error from this.

\$\endgroup\$
6
  • 2
    \$\begingroup\$ Is it switching? \$\endgroup\$
    – bobflux
    Commented Jan 4 at 22:08
  • 2
    \$\begingroup\$ Why did you rule out a manufacturing error? Did you verify there is none? For example, I have been bitten by manufacturing error. The feedback resistor that was mounted had ohms instead of kilo-ohms it read in the BOM, so the prototype burned up with overvoltage, because I did not verify the feedback resistors before applying power. Lesson learnt never to trust anything unless verified by measurements before applying power. \$\endgroup\$
    – Justme
    Commented Jan 5 at 0:00
  • \$\begingroup\$ @bobflux how can I measure whether it's switching? I assumed it is switching given the ripple but I now realize that this conclusion is wrong and the ripple might come from the input voltage. \$\endgroup\$
    – mlx11
    Commented Jan 5 at 7:58
  • \$\begingroup\$ @Justme I ruled out an manufacturing error due to the problem appearing twice. But you are right, this reasoning is wrong. I'll try to measure/verify all the values. \$\endgroup\$
    – mlx11
    Commented Jan 5 at 7:59
  • \$\begingroup\$ @bobflux Looking at the output waveform again, the rippfle frequency does not match the expected 400kHz of the buck converter. So it might not be switching \$\endgroup\$
    – mlx11
    Commented Jan 5 at 8:04

1 Answer 1

Reset to default
0
\$\begingroup\$

I'm not sure how this can be but, the device in your schematic has 20 pins whereas the device in the data sheet you linked has 16 pins like this: -

enter image description here

I'll leave it to you to figure out reasons why. Your schematic: -

enter image description here

There are only 16 pins: -

enter image description here

\$\endgroup\$
4
  • 1
    \$\begingroup\$ The OP;s schematic symbol (and presumably PCB footprint) has an extra pin for each of the corner pads on the package which appear larger than all the others. \$\endgroup\$
    – brhans
    Commented Jan 4 at 21:00
  • 1
    \$\begingroup\$ exactly @brhans. I verified the footprint and it is correct. The additional pins are in the corner. \$\endgroup\$
    – mlx11
    Commented Jan 4 at 21:06
  • \$\begingroup\$ @mlx11 count the pins on the 1st image and there are 16. There are only 16 pins and that image is from the data sheet link in the question. I've added another image. \$\endgroup\$
    – Andy aka
    Commented Jan 4 at 21:32
  • 2
    \$\begingroup\$ @Andy the footprint is correct. It models each corner pad as 2 pads (for whatever reason). So the VCC pad is VCC_1 and VCC_2 which model the same pad. \$\endgroup\$
    – mlx11
    Commented Jan 4 at 23:06

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