Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Modifying computer power supplies

  1. Feb 25, 2006 #1
    These are some typical values for a computer's power supply:
    +3.3V@28A, +5V@30A, -5V@0.3A, +12V@15A, -12V@0.8A, +5VSB@2A

    How difficult would it be to change the voltages and current? For example to +30V@20A and -30V@20A.
     
  2. jcsd
  3. Feb 25, 2006 #2

    Pengwuino

    User Avatar
    Gold Member

    Probably not worth it. From what I know, the better power supplies are well regulated and various components might disallow the power supply to work if something was out of place or outside of the perameters set by the components to allow.
     
  4. Feb 26, 2006 #3

    russ_watters

    User Avatar

    Staff: Mentor

    Well, 600w is more than most computer psus are capable of, so it probably isn't even possible. With resistors, you can probably make one generate a lower voltage, but a higher voltage - no.
     
  5. Feb 26, 2006 #4

    Pengwuino

    User Avatar
    Gold Member

    Don't the better PSU's have circuits that regulate the voltage though?
     
  6. Feb 26, 2006 #5

    Cliff_J

    User Avatar
    Science Advisor

    Sure there's going to be circuits to regulate the voltages, but only within a certain range.

    So even though it may be possible to swing the outputs slightly by adjusting the PWM circuit, a new transformer setup to handle the increased VA would be far more likely and also the recitfier diodes as well, plus the heatsink....
     
  7. Feb 26, 2006 #6

    FredGarvin

    User Avatar
    Science Advisor

    I've watched some ee's sitting at their desk winding home made transformers and trying to keep count of how many turns they had. It looked pretty painful to me. Why not just get a good test lab variable power supply and then worry about the rest of the regulating components that will give you headaches?
     
  8. Feb 26, 2006 #7
    Do you know how much one of those power supplies cost in relation to a computer's power supply? rhetorically....
     
  9. Feb 26, 2006 #8

    Pengwuino

    User Avatar
    Gold Member

    A good 600 watt powersupply is probably $80.

    I purchased a 6-15V variable voltage 500W benchtop powersupply for about $120. I wonder how much a 30V power supply would be... hopefully not toooo much more?
     
  10. Feb 27, 2006 #9

    FredGarvin

    User Avatar
    Science Advisor

    True. They aint cheap. You definitely won't get one for a hundred bucks.
     
  11. Feb 27, 2006 #10

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Here is one result of a quick google of "30V Power supply"

    Any specific reason you need 30V? If you could make do with 24V you will find a lot more and probably cheaper Power supplies. 24V is an industry standard so they are quiet common.
     
  12. Feb 27, 2006 #11

    russ_watters

    User Avatar

    Staff: Mentor

    ...And a lot of 24v transformers put out a nominal voltage of more like 27, and a lot of equipment has a high tolerance for variations in voltage...
     
  13. Feb 27, 2006 #12
    I need it for an amplifier system I'm building, and it is using one of those T class (or tripath-class D) chips that requires 30V.
     
  14. Feb 27, 2006 #13

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I assume that we are talking about a DC Power supply. If you were to build a linear 30V Power supply you would need to start with a transformer that puts out nearly 50VAC in order to get a rectified 30VDC output.
     
  15. Feb 27, 2006 #14

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You may want to verify that the high frequency content of a switching power supply will not effect your system.

    It may be that a 24V supply will simply reduce your output...Just a hypothesis, not firm knowledge on my part. I pretty sure that you will not be able to get what you need by modifying a computer power supply.
     
  16. Feb 27, 2006 #15
    Correct, by using 24V the output would be reduced. I was only trying to get the recommended supply for max output power.

    This website:
    http://www.coilws.com/Toroidal Power Tx/Toroid_Pwr_PN.php?orderby=data3&orderdir=desc
    has some good toroids, but they're a bit pricy. I starting to think about why home audio amplifiers are so expensive.
     
  17. Feb 28, 2006 #16
    What about rectifying a 110V AC input then stepping it down to 30V, would that be easier and/or cheaper?
     
  18. Feb 28, 2006 #17

    russ_watters

    User Avatar

    Staff: Mentor

    D'oh - not paying enough attention.
     
  19. Feb 28, 2006 #18
    Since when? A full wave bridge rectifier being fed with a 50 VAC source will output (when filtered with a cap) about 70 volts DC.
     
  20. Feb 28, 2006 #19
    You cannot use a transformer to step up or down DC. Since this has clearly escaped your grasp I would recommend you stay away from modifying anything electronic, let alone building something from the ground up until you pick up a bit more knowledge.
     
  21. Feb 28, 2006 #20
    I cannot remember using the word transformer, I know that DC cannot be transformed. One of the main advantages of AC over DC is the fact that AC can be transformed.

    I was referring to a DC to DC converter (regulator).
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Modifying computer power supplies
  1. Computer Power Supply (Replies: 15)

  2. Power supply (Replies: 5)

Loading...