Announcement

Collapse
No announcement yet.

Help with Bench PSU or Conversion

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Help with Bench PSU or Conversion

    Hi.

    I would like to find out if someone here do or did any ATX PSU conversions for variable voltage & current control. It would be really nice to have a fairly large variable option. 1V - 25V+ but since I work on car stuff and laptops mainly anything from 12V to 20V would be nice.

    Current @ 19V atleast 5A | 9A would be really nice for those big gaming laptops. But the main thing would be to have a variable Bench PSU that I could use mainly for testing so that means short bursts 1min - 5mins max esp on heavy load... Other things would include injecting voltages into boards for testing... These would be low current mainly.

    I have a Aero Cool Zerodba 500W ATX Modular PSU just laying here. I want to either convert it or strip its parts and build my own PSU from the ground up using the parts gathered from the many parts I have here. I have been busy on and off designing my PSU in Kicad but its a slow process. Just too busy... I have various components some are really high rated ones esp for the switches. Also have various High Freq Transformers.... Whatever I need I can buy.

    Chip Details:
    - TL3842P
    - LM339N
    - NCP1653

    Attached are photos regarding the PSU in question.

    Building it from the ground up is also possible I have all the tools/parts and things needed its really just time... I can get PCB's made professionally from a local manufacturer. So it won't be ghetto shit. I figure once I have the schematic/ Design done I'd post here sometime to get some feedback on it but until then converting/modding the ATX would be preferred.
    Attached Files
    while alive { live(toFullest=true) }

    #2
    Re: Help with Bench PSU or Conversion

    there's a few online.
    if you cant find what you need here - your in trouble!
    http://www.danyk.cz/

    Comment


      #3
      Re: Help with Bench PSU or Conversion

      Chances are you will not be able to change the voltage all the way down to 1V and up to 25V without severe loss in stability or damage to components. You're better off starting fresh unless you don't care about regulation or want to stay within a few percent of the originally designed voltage.

      Comment


        #4
        Re: Help with Bench PSU or Conversion

        Originally posted by eccerr0r View Post
        Chances are you will not be able to change the voltage all the way down to 1V and up to 25V without severe loss in stability or damage to components. You're better off starting fresh unless you don't care about regulation or want to stay within a few percent of the originally designed voltage.
        Concur. Trying to get that range of adjustability would be very difficult and would probably screw up the control loop stability.

        Further, it would set up problems with the OVP circuits for the outputs monitored by the supervisor circuit.

        If the 3.3V output is mag-amp post-regulated that circuit would also face an input volt-seconds range the circuit is not designed to handle.

        Finally, the output electrolytics are probably a mix of 10V and 16V parts. The output that could be adjusted up to 25V should have 35V rated caps. Changing those parts would also probably screw up the control loop stability.

        That's a lot of 55 gallon drums of worms MJ-meo-dmt is looking to open.

        Also, if the 12V output is rated for (arbitrarily) 20A, adjusting it up to 24V and expecting to load it to 20A could cause power limit problems.
        Last edited by PeteS in CA; 11-15-2018, 12:48 PM.
        PeteS in CA

        Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
        ****************************
        To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
        ****************************

        Comment


          #5
          Re: Help with Bench PSU or Conversion

          @stj Thanks, that's a pretty cool share.

          Hi, nice to meet you both.

          Chances are you will not be able to change the voltage all the way down to 1V and up to 25V without severe loss in stability or damage to components. You're better off starting fresh unless you don't care about regulation or want to stay within a few percent of the originally designed voltage.
          @eccerr0r
          Yeah 1V is a bit too far, I don't really need that, nothing below 12V. And for max id say 20V.
          I also figured I should just start "fresh"... What do you mean regards to "fresh"? Strip it down and replace all the need components like the output caps and so on... or fresh fresh new PCB scratch.

          Well for regulation meaning, to be able to adjust the Voltage/Current fine/coarse? or Regulation in regards to keeping the PSU rated "Safe"
          I guess having it switchable between a few preset voltages would be just fine. Like 12V | 15V | 18.5V | 19.5V | 20V || Max 8A - 10A (thinking about laptops, most req about 3A-5A, "high" performance ones 7A-9A and that's usually between 18V-19V)
          _

          @PeteS in CA Yes I know about the can of worms, I agree! but I don't have a choice it would make my work easier... So I have to pick the best way to go without redesigning/rebuilding to much of the "wheel". Even being able to maybe, make use of the original PCB; would be great cost wise....
          Last edited by MJ-meo-dmt; 11-15-2018, 02:32 PM.
          while alive { live(toFullest=true) }

          Comment


            #6
            Re: Help with Bench PSU or Conversion

            Fresh as in new circuit, little of the "critical stuff" in the old PSU can be reused. The key things that need to be recalculated and reworked: all of the coils including the main transformer, as well as capacitors that match up with them. The voltage feedback is easy peasy compared to the rest of the control loop.

            Regulation, meaning the stability of the control loop. Meaning you can still may get an average of 1 volt...

            ... with ripple of 6V p-p...

            Likely you can safely get 10-15% deviation from the original design voltages. Anything past that, it's hard to say. As PeteS says, going outside will cause problems: going down in voltage will cause stability issues by going out of the intended operational mode, going up in voltage will damage circuits as well as possibly overloading the primary duty cycle, at least in flyback designs; or simply impossible without coil rewinding in others.

            Comment


              #7
              Re: Help with Bench PSU or Conversion

              Not trying to pile on, but you may not be able to get enough duty cycle to get the desired voltages. Oversimplifying some, your output voltage is the input voltage times the duty cycle times the transformer turns ratio. Both the circuit topology (it's maximum duty cycle) and the transformer turns ratio set a maximum output voltage.

              If you stayed 20V max then 25V capacitors could be used to replace 16V caps on the 12V line.

              I think you should look at single output power supplies from whatever distributor(s) sell into SA.
              PeteS in CA

              Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
              ****************************
              To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
              ****************************

              Comment

              Working...
              X