Announcement

Collapse
No announcement yet.

Converting APFC PSU to nor.al PSU without PFC? - Hiper 525W

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

    Converting APFC PSU to nor.al PSU without PFC? - Hiper 525W

    As the title says - I have here a Hiper PSU model HPU-4S525 that came with a blown to smithereens APFC board, blown fuse and main caps.


    So far this is what I did:
    -remove the APFC board
    -jumper the PFC spot on board*
    -replace all capacitors - it had bulging caps on both primary and secondary
    -test main FETs - all check fine

    With no success so far. I temporarily replaced the fuse with a 60W lightbulb (the old fuse was blown anyways) so I could prevent any part go nuclear.

    Symptoms:
    -without shorting the PS_ON wire, 5vSB comes up okay at 5.1V and lightbulb flashes one time accordingly then stays dark as it should
    -shorting PS_ON however will result in the lightbulb remaining bright, no voltages present on the secondary, until I cut the AC power.


    Is there anything I might have missed out?

    ICs present on this PSU:

    -TL3842 - near main input wires
    -2x LM339 - one near the secondary caps, the other on a mounted PCB near 12v wires.

    Finally, I'm leaving a few pictures of it in hopes somebody can tell me what I should check and possibly remove.
    Attached Files
    Last edited by Dan81; 03-23-2019, 03:09 AM.
    Main rig:
    Gigabyte B75M-D3H
    Core i5-3470 3.60GHz
    Gigabyte Geforce GTX650 1GB GDDR5
    16GB DDR3-1600
    Samsung SH-224AB DVD-RW
    FSP Bluestorm II 500W (recapped)
    120GB ADATA + 2x Seagate Barracuda ES.2 ST31000340NS 1TB
    Delux MG760 case

    #2
    Re: Converting APFC PSU to nor.al PSU without PFC? - Hiper 525W

    check for shorted output diodes.

    Comment


      #3
      Re: Converting APFC PSU to nor.al PSU without PFC? - Hiper 525W

      Originally posted by stj View Post
      check for shorted output diodes.
      You mean the output rectifiers?
      Main rig:
      Gigabyte B75M-D3H
      Core i5-3470 3.60GHz
      Gigabyte Geforce GTX650 1GB GDDR5
      16GB DDR3-1600
      Samsung SH-224AB DVD-RW
      FSP Bluestorm II 500W (recapped)
      120GB ADATA + 2x Seagate Barracuda ES.2 ST31000340NS 1TB
      Delux MG760 case

      Comment


        #4
        Re: Converting APFC PSU to nor.al PSU without PFC? - Hiper 525W

        just check every diode in it.
        if they are o.k. you next check for shorted ceramic caps.

        Comment


          #5
          Re: Converting APFC PSU to normal PSU without PFC? - Hiper 525W

          Originally posted by Dan81 View Post
          Symptoms:
          -without shorting the PS_ON wire, 5vSB comes up okay at 5.1V and lightbulb flashes one time accordingly then stays dark as it should
          -shorting PS_ON however will result in the lightbulb remaining bright, no voltages present on the secondary, until I cut the AC power.

          Is there anything I might have missed out?

          ICs present on this PSU:

          -TL3842 - near main input wires
          -2x LM339 - one near the secondary caps, the other on a mounted PCB near 12v wires.
          I suspect the TL3842 might be bad too or whatever is driving the main PS MOSFET (there might be a "buffer" driver circuit after the TL3842, implemented by one or two transistors and diodes - if yes, remove and check these for sure). Either that, or it is configured improperly by design. A well-designed PSU with 3842 PWM controller will shut down when it detects an over-power / PWM duty cycle too high condition.

          Another thing to check: voltage across primary caps before and after you attempt to turn on the PSU. Is the voltage across the two primary caps even?

          Also, are you sure the secondary only has two LM339 controllers? I would expect there to be a supervisor chip somewhere - perhaps on that daughterboard. It may be just a tiny 8-pin SMD chip, so have a look. Two LM339 comparators for main PS feedback sounds like an awfully sketchy design.

          Originally posted by stj View Post
          check for shorted output diodes.
          That shouldn't cause the incandescent bulb to stay lit. 3842 PWM IC is a current-mode PWM and if designed properly, should detect when duty cycle goes too high. Thus a shorted output should make the main PS crowbar - that is, turn off and remain off until power is cycled.

          But you do make a good point that it is worth checking the output rectifiers anyways.
          Last edited by momaka; 03-27-2019, 01:59 AM.

          Comment


            #6
            Re: Converting APFC PSU to normal PSU without PFC? - Hiper 525W

            Originally posted by momaka View Post
            I suspect the TL3842 might be bad too or whatever is driving the main PS MOSFET (there might be a "buffer" driver circuit after the TL3842, implemented by one or two transistors and diodes - if yes, remove and check these for sure). Either that, or it is configured improperly by design. A well-designed PSU with 3842 PWM controller will shut down when it detects an over-power / PWM duty cycle too high condition.

            Another thing to check: voltage across primary caps before and after you attempt to turn on the PSU. Is the voltage across the two primary caps even?
            As far as I can see, there is a transistor (couldn't read it right now), followed by a Fuhjyyu TNR 2.2uF 50v cap and a ceramic disk capacitor.

            Originally posted by momaka View Post
            Also, are you sure the secondary only has two LM339 controllers? I would expect there to be a supervisor chip somewhere - perhaps on that daughterboard. It may be just a tiny 8-pin SMD chip, so have a look. Two LM339 comparators for main PS feedback sounds like an awfully sketchy design.
            This is probably the 5th or 6th time I checked, and yes, it's just 2 LM339 controllers and of course the TL3842 IC.

            Originally posted by momaka View Post
            That shouldn't cause the incandescent bulb to stay lit. 3842 PWM IC is a current-mode PWM and if designed properly, should detect when duty cycle goes too high. Thus a shorted output should make the main PS crowbar - that is, turn off and remain off until power is cycled.

            But you do make a good point that it is worth checking the output rectifiers anyways.
            Will check the rectifiers. Though if it was the rectifiers wouldn't the PSU just simply flash the bulb once and then do nothing? (same as if it has only 5vSB)
            Main rig:
            Gigabyte B75M-D3H
            Core i5-3470 3.60GHz
            Gigabyte Geforce GTX650 1GB GDDR5
            16GB DDR3-1600
            Samsung SH-224AB DVD-RW
            FSP Bluestorm II 500W (recapped)
            120GB ADATA + 2x Seagate Barracuda ES.2 ST31000340NS 1TB
            Delux MG760 case

            Comment


              #7
              Re: Converting APFC PSU to nor.al PSU without PFC? - Hiper 525W

              diodes can be leaky without being totally shorted.

              Comment


                #8
                Re: Converting APFC PSU to normal PSU without PFC? - Hiper 525W

                Originally posted by Dan81 View Post
                As far as I can see, there is a transistor (couldn't read it right now), followed by a Fuhjyyu TNR 2.2uF 50v cap and a ceramic disk capacitor.
                But are any of those connected to the Gate of the MOSFET for the main PS? Because that's what we are interested in. Follow the main PS MOSFET's Gate back to the 3842 PWM chip. If it's directly connected to it (or through a resistor only with no transistors in between), then the 3842 PWM chip is very likely bad. And if there is a transistor that drives the MOSFET's Gate, then most likely that transistor is bad. Remove from circuit and check.

                Originally posted by Dan81 View Post
                This is probably the 5th or 6th time I checked, and yes, it's just 2 LM339 controllers and of course the TL3842 IC.
                Ok, wont' but you about this anymore, I promise. I just haven't seen a forward converter PSU with such outdated / simple feedback circuit. Normally they use at least one of those 8-pin protection ICs (like Weltrend WT571002 chips). Only seen flyback topologies use 3842 IC with no supervisor chip and 431 shunt regulator only. But that's a completely different topology and PS design.

                Originally posted by Dan81 View Post
                Will check the rectifiers. Though if it was the rectifiers wouldn't the PSU just simply flash the bulb once and then do nothing?
                Exactly.
                Regardless of whether the rectifiers might be leaky or shorted, the supervisory circuits on the secondary should detect that and turn off. If it doesn't, then that's a really crap PSU design.

                Comment

                Working...
                X