Announcement

Collapse
No announcement yet.

ASRock B75M-GL dead /water damage/

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

    ASRock B75M-GL dead /water damage/

    I valued ventilated case over a solid one (little holes all over, including top), and my son naturally spilled water over case, destroying a MB.
    Very little amount of water managed to MB, but it turned off immediately, of course he tried to power it on few times after that before calling me.

    Board is tested dead, no quarter sppin or anything, with a working PSU from another machine. Now, I found a MOSFET (73T03GH) in CPUVcore which is shorted all-over (D to G to S), and I wonder

    Other MOSFETS look good on board. What are the chances that this is the only problem, and could it kill the CPU?
    Any way to verify it w/o another MOSFET?
    I found them in China only, involving month or so shipping.

    It's hard for me to find any way to test, because even if I pull it out, any measurements I make alredy give so small readings, CPUVcore being low-resistance load as it is.
    Attached Files
    stay classy

    #2
    Re: ASRock B75M-GL dead /water damage/

    If this is a high-side MOSFET it can kill the CPU. But it may be fine too, the only way to find out is to test it on another board (or try to fix this one).
    The easiest way to find out if this is a high-side MOSFET is to measure resistance between yellow and black wire on the P4 power connector. With CPU connected you will measure a few ohms, without CPU you will measure a few kiloohms. If you get more than a few kiloomhs with the CPU connected, no high-side MOSFET is shorted.

    It can be a low-side MOSFET. In this case you will usually measure a full short between either the output inductors (black gray cube next to the MOSFET) and ground, or across the output capacitors (the blue/gray capacitors inbetween the inductors and the CPU). A full short is lower than 1 ohm. CPU resistance is usually at least a few ohms.

    The board should turn on when you disconnect the P4 power connector but obviously the CPU will not be powered on.
    OpenBoardView — https://github.com/OpenBoardView/OpenBoardView

    Comment


      #3
      Re: ASRock B75M-GL dead /water damage/

      Thank you very much, really informative!

      P4 connector - 3.7 ohm with CPU, 8.8 kohm w/o.

      output caps-same.
      No short between coils and gnd, so I think it's high side.


      Originally posted by piernov View Post
      If you get more than a few kiloomhs with the CPU connected, no high-side MOSFET is shorted
      So, when I remove shorted one, I should measure few kiloohms on P4 with CPU in?

      Furthermore, I tried powering board on w/o P4. No signs of life (I believe PSU should start, right?)
      stay classy

      Comment


        #4
        Re: ASRock B75M-GL dead /water damage/

        Originally posted by Joe Black View Post
        Thank you very much, really informative!

        P4 connector - 3.7 ohm with CPU, 8.8 kohm w/o.

        output caps-same.
        No short between coils and gnd, so I think it's high side.
        Indeed it seems so.

        Originally posted by Joe Black View Post
        So, when I remove shorted one, I should measure few kiloohms on P4 with CPU in?
        Yes, that's the first thing you have to do.

        Originally posted by Joe Black View Post
        Furthermore, I tried powering board on w/o P4. No signs of life (I believe PSU should start, right?)
        It should start but try to remove the MOSFET anyway, maybe it's still preventing the board from powering on.
        OpenBoardView — https://github.com/OpenBoardView/OpenBoardView

        Comment


          #5
          Re: ASRock B75M-GL dead /water damage/

          Great, Ill remove the mosfet and try to power on without P4.

          So, if I got it right, MOSFETs are naturally closed, and when all are good, that's why there's few kohms on P4 (example)
          When they open, then P4 "sees" CPU, and (as on mine) P4=3.7 or so ohms.
          Asuming that coils and in and out caps are OK.

          Makes perfect sense now, thank you so much!
          stay classy

          Comment


            #6
            Re: ASRock B75M-GL dead /water damage/

            I gave it a thought and it seems obvious to me now that CPU must be dead.

            12V on P4/no chopping from MOSFETs/directly to CPU/fried CPU.

            No short, so no protection or anything. Correct?
            stay classy

            Comment


              #7
              Re: ASRock B75M-GL dead /water damage/

              However, if caps have survived higher V, board should be fine after replacing MOSFET, and fully working with another CPU.
              stay classy

              Comment


                #8
                Re: ASRock B75M-GL dead /water damage/

                CPU may or may not be fine. The PSU will see a short on the 12V line and shuts down immediately. Don't know how it works but some way the 12V is being sent to ground rather than the CPU logic itself. But it may still die.
                The caps usually won't die from a high-side MOSFET short because the PSU shuts down too quickly.
                Don't try to power on the board without the MOSFET and with the P4 power connector connected. It will probably fry something else.
                OpenBoardView — https://github.com/OpenBoardView/OpenBoardView

                Comment


                  #9
                  Re: ASRock B75M-GL dead /water damage/

                  Originally posted by piernov View Post
                  Don't try to power on the board without the MOSFET and with the P4 power connector connected. It will probably fry something else.
                  Thanks for the heads-up.

                  OK, I'll keep your advice in mind. Am going to order MOSFETs, and when they arrive try to work on the board. Will make sure to post results here.
                  stay classy

                  Comment


                    #10
                    Re: ASRock B75M-GL dead /water damage/

                    Got a new bundle, also 1155 socket.
                    Tested CPU in working board, no POST. Fans spin, but no POST, so I guess CPU is dead.

                    Just in case someone has similar issues. Will try to repair the board when I get another MOSFET and test it with working processor. Will post results, also.
                    stay classy

                    Comment

                    Working...
                    X