3
\$\begingroup\$

I'm working on a boost converter board to power an OLED display, specifically a Midas Display MCOT128064SV-WM, off of 5V USB bus power (the VBUS pin on a Raspberry Pi Pico powered via USB).

enter image description here

I referenced this TI WEBENCH design:

enter image description here

I had this design printed and assembled, but it outputs around 12.2V (originally stated as 12.5V) when I applied power, instead of the 13V I'm expecting. Is there anything wrong with this circuit? I'm new to electronics design, so I'm pretty sure I'm missing something.

Edit to add: \$V_{out}\$ reaches 12.18V, and \$V_{FB}\$ only reaches 0.824V, instead of 0.795V.

Edit: After posting this, I found out R2 was ordered as a 1Ω resistor, not 100kΩ. Effectively, the upper half of the resistance divider around FB was 10% too low, so the output voltage was also about 10% too low.

\$\endgroup\$
3
  • \$\begingroup\$ U1 will increase the output until it gets 0.795 V on \$ V_{FB} \$. Can you measure yours and post the measurement into your question. Record both \$ V_{OUT} \$ and \$ V_{FB} \$. \$\endgroup\$
    – Transistor
    Jun 19, 2022 at 13:49
  • 1
    \$\begingroup\$ Does the inductor you used to build your circuit have equal or better specs as the one in the WebBench design - specifically its ESR value and not just the inductance? \$\endgroup\$
    – brhans
    Jun 19, 2022 at 14:20
  • 3
    \$\begingroup\$ I’m voting to close this question because it is the electronic equivalent of a typo: a mistake committed so often that posting every instance of it would overwhelm everyone, and it's super unlikely that anyone else would have the same problem and could find this question. \$\endgroup\$
    – Kuba hasn't forgotten Monica
    Jun 19, 2022 at 18:41

1 Answer 1

Reset to default
5
\$\begingroup\$

I had this design printed and assembled, but it outputs slightly under 12.5V when I applied power, instead of the 13V I'm expecting

So, to put that into context, it's a 3.85% error. Just saying...

The reference voltage for the TLV61046 at 25°C can be between 0.783 volts and 0.807 volts with a nominal at 0.795 volts. So, if the reference voltage in the chip is at 0.783 volts, the output voltage will be less than the ideal predicted one of 13 volts.

According to your resistor values, with 13 volts present on the output, the FB input will be 0.7934 volts but, factor-in some resistor tolerances to this (say 2%) and it might be more like 0.809 volts.

Taking all of this into account means that the actual output voltage could be as low as: -

$$\text{13 volts}\times \dfrac{0.783}{0.809} = \text{12.58 volts}$$

Add in some multimeter calibration errors and temperatures higher or lower than 25°C and you could easily get under 12.5 volts. So what accuracy does your meter brochure state? How old is your meter? When was it last calibrated?

If you are using 2% or 5% resistors it will be a lot more. If you meter's battery is close to depletion it could be more.

Trying to measure \$V_{FB}\$ with a standard 10 MΩ input impedance meter is going to alter \$V_{FB}\$ by nearly 10% and the regulator will adjust it's output by that amount to compensate. Be aware that this may upset things a little.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ I might've been too hasty asking my question here. I've just found out I ordered a 1Ω resistor for R2, instead of 100kΩ... So sorry for wasting your time! Apologies for misquoting my measured value for \$V_{out}\$ as well. \$\endgroup\$
    – kaine119
    Jun 19, 2022 at 14:39
  • \$\begingroup\$ So, in effect (with respect to my answer) it becomes a resistor tolerance issue. Instead of 1.1 Mohms, you had 1 Mohms i.e. it is 10% too low and this means the output voltage is around 10% too low @kaine119 \$\endgroup\$
    – Andy aka
    Jun 19, 2022 at 14:43

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