DIY Lab Bench power supply build problem!

Discussion in 'Power Electronics' started by utilityhacker, Sep 12, 2016.

  1. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    I had a couple of dc power adapters laying around and so i decided to make a bench power supply from them. I used 2 adapters with 12v,1A and 5v,800ma rating. I connected everything as shown in the attached image. The problem is the 12v supply now gives 13v output and 5v supply now gives 6v output. 3.3v seems to be working fine. I could really use some help.
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,638
    2,343
    Hello,

    Do you have more information or pictures of the used adapters?

    Bertus
     
  3. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    for the 12v adapter
    INPUT: 100-240v, ~50/60Hz, 250ma
    OUTPUT: 12v, 0.5A

    I lost the 5v adapter shell on which the power rating was printed but its pretty much the same as 12v adaptor.
    for the 5v adapter
    INPUT: 100-240v, ~50/60Hz
    OUTPUT: 5v, 0.8A
     
  4. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    Output should be 12v and 5v but its 13v and 6v instead.
    Increase of 1v .
     
  5. bertus

    Administrator

    Apr 5, 2008
    15,638
    2,343
    Hello,

    Are the adapters of the switched kind?
    They probably do not like the grounds tied together.

    Bertus
     
  6. DGElder

    Member

    Apr 3, 2016
    344
    85
    I got that. Did you measure it before you built it into your supply? What did it measure? What is different from how it was connected to the mains then and now and how you are measuring it then and now?
     
  7. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,644
    759
    What is the actual input to the LM317, measured wrt ground? That IC requires a minimum difference between input / output.
     
    bertus likes this.
  8. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    Yes, I did measure it before building it and it measured 12v. Its same like plugging 2 adapters in a single wall outlet and i used a multimeter to measure it. I think it might be because i tied the ground pins to each other. I'll check that abd report back in a moment.
     
  9. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    Output from lm317 seems to be fine 3.1~3.2 (3.3v expected). its just with the 12v and 5v not outputting stated voltage.
     
  10. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    I'll check that and report in a moment.
     
  11. utilityhacker

    Thread Starter Member

    May 1, 2014
    40
    4
    I tested each adapter with their respective grounds, but the voltage still doesn't measure the sames(13v,6v) as it is supposed to(12v,5v). I used the 12v adapter previously to power a 8051 based programmer board and the 5v adapter to power an arduino and both worked fine. I dont understand. If these adapters used to output extra 1v and my boards are still working does this mean that the mcu isn't affecteded by it? Im asking this because im might be using this to power oyher microcontrollers.
     
  12. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,644
    759
    Please just measure and tell the actual input wrt ground.
     
  13. HW-nut

    Member

    May 12, 2016
    45
    11
    The adapters may not be regulated and specifications you have are for a specific load condition.
     
  14. seanstevens

    Member

    Sep 22, 2009
    25
    1
    I agree with HW-nut, generally adapters are not regulated and are for specific load.
     
  15. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,733
    759
    OP is getting unloaded voltage
     
  16. splud

    New Member

    Jun 30, 2013
    6
    0
    I agree with the consensus that this appears to be an issue with the supplies being unloaded - can you try putting a load on them in the range of half of the rated current and then take a voltage reading? Of course, if they're cheap wallwarts, there's a good chance that if you load them to anywhere near their rated current, the voltage is going to drop to below their rated voltage.

    Note that wallwarts are not "bench power supplies". Also, while the linear regulator is best driven from a supply not too much higher than the rated dropout, your 5V supply has a lower total power than your 12V does, so driving a load on the 3.3V is going to have a significant impact on the stability of the 5V supply. If you wanted to stick mostly with what you've got, I'd at least swap out the linear with a switchmode (such as an LM2574-3.3) driven off of the 12V supply.

    If you revisit your power supply, you might consider using just the 12V wallwart (or potentially any source up to circa 17 VDC), and three TPS62160 switchmode regulator ICs (or a TPS62161 for 3.3V, TPS62162 for 5.0V, and a TPS62160 which is voltage adjustable with feedback resistors to give you 12.0V). They're rated for 1A, are quite stable with a 2.25MHz switching frequency, so the inductor is small and they respond quickly to transients, and can operate to Vout==Vin. Yes, I do suggest that your 12V supply still have a regulator on it, rather than just using the 12V as-is.
     
Loading...