Hao Jie Chan's Kingdom: January 2016
SD Wing Zero Custom is done! Click here to check it out!

Ad 468 X 60

built of paper.

Sunday, January 17, 2016

A Buck Converter Power Supply

Power Supply

This project has been in my mind for a long time. I have always wanted an easy to use power supply, without resorting to Arduino for a 5V supply. With the addition of a cheap DIY oscilloscope which requires a 9V supply, this project became more and more important.
With the start of my sem break, this long overdue project finally got the chance to be completed.

Power Supply2
Another shot at the completed project exterior.

Power Supply Interior
The interior.

Power Supply Interior2
Okay, time for some explaination. This “power supply” is basically just a buck converter, I am using a cheap one from China, using MP1584EN as the switching IC.
My plan was to change the SMD potentiometer, to a bigger one which can be turned by hand. (I was frustrated by the need of a screwdriver just to change the voltage level.

The input is taken from a laptop adapter, which I have one laying around.  So, it is a 19-21V to maybe 0.8V-18V supply with a maximum of 3A load ( that’s what the datasheet and seller claims, I would not use it for more than 2A personally. )

Power Supply Interior3
The potentiometer solder pads were quite small and were ruined when I desoldered the pot. Heck I even spoil the pot! Then I put a 10K potentiometer initailly, found out that my input would not rise over 2V. I figured my resistance was too low for the feedback path. I then change the pot to a 100K pot. Better, but the voltage stopped at 10V. Can be used, but I wasn’t satisfied. I found that the resistor value is different for 2 batches of boards. Refer to the following photos.

MP1584EN board
This is from the old batch, green colour board. Now If we zoom in to the resistor,

MP1584EN board closeup
It’s 322, as in 3.2K ohm.

MP1584EN board new
The new one, (as in the one I used), blue board, has a 822 SMD resistor, 8.2K ohm. The potentiometer forms a potential divider with the resistor to produce a feedback to the IC. The IC requires 0.8V at the feedback. 100K is simply too small for me to generate up to 18V.

Thus, I hacked the board, again. I have some 10K SMD resistors around. I simply soldered them on top of each other, to reduce the resistance. SO in the end, it’s 8.2K//10K//10K = <5K ohm!
And that worked! Circuit theory rocks! Open-mouthed smile Sorry I just have to bask in my glory for a little while.
Now, I can move on the next project, which is still undecided.

 See you next time!


  • Facebook
  • Twitter
  • Myspace
  • Google Buzz
  • Reddit
  • Stumnleupon
  • Delicious
  • Digg
  • Technorati
Author: Chan HJ
An enthusiast in papercraft. Paper model designing sounds fun too! More about me at my Blog Bio or