Close
0%
0%

VederPSU

A tiny and efficient uC controlled lab power supply

Similar projects worth following
The VederPSU is a project to design a small, efficient lab power supply that is comparable to a commercial lab PSU in terms of power output, regulation and noise.
This project started a few years ago but has recently gone trough a big redesign, which is the reason I am posting the project here now.

The VederPSU is a small and efficient lab power supply. Some of the requirements are:

  • Small, less then 10*10*5cm in size
  • Because of the size limitations, the VederPSU needs to be efficient as there is no space for a big heatsink
  • 0 to 30V output.
  • 0-2A output
  • Low noise (less then 30mV)
  • Usable standalone and PC controlled

Current status:

Hardware works apart from the following bugs:

  • LT3081's don't regulate fully to 0V, a negative supply is needed to fix this issue
  • 2 grounds not connected in schematic, this requires 2 bugwires to fix.
  • The current 5V regulator has a maximum input of 20V, which is a problem with a 24 to 36V input it can get..

untitled.sch

Schematic rev 0.3, eagle file

x-kicad-schematic - 841.95 kB - 02/12/2017 at 21:18

Download

untitled.brd

PCB rev 0.3, eagle file

x-kicad-pcbnew - 364.54 kB - 02/12/2017 at 21:18

Download

Schematic_vederpsu_new.pdf

Adobe Portable Document Format - 38.19 kB - 07/20/2016 at 21:43

Preview

  • Revision 0.3

    riktw02/12/2017 at 21:15 0 comments

    The revision 0.2 board worked mostly, but has a few issues. One was that idiot me forgot to connect 2 pins to the ground. The other issue is that the LT3081's need a bit of load to regulate to 0V and that the 5V regulator used has a maximum input of 20V.

    Revision 0.3 fixes these issues.

    The 5V regulator has been replaced by a 78L05, the SOIC version of the well known 7805, which is capable of 30V input. Not perfect but acceptable for now. A SMPS would be a nicer solution in the future.

    The problem with 2 gound pins not connected has been fixed. A small power LED has been added as well as an indication the device is powered.

    At last a LM2687 has been added to generate -5V at a few mA's This is used to create some load for the LT3081's so the output will regulate to 0V. It's afforable and works perfect for this application.

    The new files have been added for the schematic and PCB layout.

    On a more interesting note, a PCB that will be used as a front panel containing an LCD, 2 rotary encoders and a uC is being designed, this together with the rev 0.3 VederPSU design will make a more complete and functional lab power supply. Hopefully an update will be here in about a month on that.

  • The new board.

    riktw07/20/2016 at 21:34 0 comments

    The new board:

    A few months ago I stumbled across this app note from Linear tech, describing how to build a lab power supply unit with their new LT3081 IC. This IC is very interesting for this application as it already has a current limiting function build in. It also has a temperature sensor build in and a way to monitor the current. This makes it near perfect for a lab power supply. The downsides are the price, the maximum output current and that a slight negative voltage is needed to really regulate to 0V.

    Based on the older version of the VederPSU a new schematic was made, using the same DAC and ADC from the old VederPSU but with a different power stage based on the LT3081 app note. The reason for using the same ADC and DAC (the MAX11613 and MAX5802) is that these are one of the very few ADC's and DACs with an I2C interface and a separate voltage reference input. I want a single voltage reference for the whole system to make calibration easier and to be able to pick a voltage reference of choice.

    The DAC has 2 outputs, one for the voltage and one for the current. With the current voltage reference they can go from 0 to 4.096V. The voltage output is multiplied by 10 using an opamp to create a 0-40V signal to control the output voltage with. The 0-4.096V for the current is converted to 0-200uA using an opamp. This combined with a current mirror controls the current limit of the LT3081's.

    The ADC has 4 inputs, from which one is used as a voltage reference input. The 3 left are used to measure the output voltage, the output current and the temperature of the LT3081's.

    The schematics can be found on the project page for anyone who wants to take a look.

  • History

    riktw07/19/2016 at 19:35 0 comments

    i.e. how it started.

    A few years ago a classmate and I started slowly working on a portable lab power supply. The initial reason was because at school we where not allowed to drag lab power supplies with us to the project rooms. We had to test in the lab, which was crowded and sometimes full, and work on all other aspects of projects in the project classrooms.

    A few revisions later a mostly working prototype was realised but by then I was doing my internship and the project ended up collecting dust, both figuratively and literally. Some info on how it worked can still be found in an old old blog post on my blog: http://justanotherelectronicsblog.com/?cat=3

    Quickly summarised the last "old" prototype uses an NXP LPC11U37 microcontroller, a maxim ADC and DAC (MAX11613 and MAX5802 because they both have a voltage reference input) and a voltage reference (LM4040) on the microcontroller side. The actual power supply side uses am LT3083 as linear regulator, am LM2670 as switching pre-regulator and a TC74 temperature sensor. To measure the current an LT6105 was used, combined with some opamp magic for current limiting and source/sense measurement.

    The last "old" prototype

    The last old style prototype of the VederPSU.

    As said, this worked reasonably well, apart from noise in current limit mode. I will upload all the files and relevant stuff here or on a github in case it is useful for anyone. The next project log will feature a more extensive write-up of the new hardware and software.

View all 3 project logs

Enjoy this project?

Share

Discussions

nicolas Jonasson wrote 06/16/2020 at 17:16 point

Love it! Have you considered using the LT3091 as well, to create a +/- power supply? : )

Would you mind sharing the code as well?

  Are you sure? yes | no

George wrote 02/08/2017 at 04:37 point

Do you have a parts list of your project, either I'm blind or ... of course this is my first time on hackaday so maybe I'm missing something.

  Are you sure? yes | no

nicolas Jonasson wrote 06/16/2020 at 17:14 point

Use the .brd Eagle CAD file, you should see the BOM there.

  Are you sure? yes | no

Benjamin E. wrote 01/13/2017 at 16:29 point

Nicely done! I need to build something like this. Would be quite handy for when I need to prototype some microcontroller stuff, since my computer is not at my electronics bench/workbench... and my preferred power supply is a heavy linear bench-top type. I do have a switching one which goes up to 60V (the linear one goes up to 30V ... still plenty for microcontroller stuff!), but it's output is very noisy even though it has a linear output regulator in it.

(Oh and don't forget that when I got that switching bench PSU, it was brand-new but already broken. The regulator for the fan had completely exploded in two. I mounted the replacement on the main heatsink (which is grounded), and it works fine ... I'm guessing it blew up because it had overheated since it had no heatsink, but the strange thing is, the power supply appear to be brand new, unopened! It could have also been a chinese knockoff, which are known to spontaneously explode while well within the "supposed" specified operation conditions... Actually, speaking of such stuff, the two 100V 1000uF filter caps in it appear to be knock-off Nichicon capacitors! They are not listen in the Nichicon datasheet, and have sloppy labels, unlike real Nichicon capacitors.)

P.S. I'm thinking that if I do build a portable PSU, it will probably either have it's own battery pack, or run off of a laptop-style battery pack. The pre-regulator will probably be of the SEPIC kind, though I'm not 100% certain yet. (It could be a buck if I use a higher voltage, home-built battery pack.)

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates