Announcement

Collapse
No announcement yet.

MS-7061 - Partially starting up

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

    MS-7061 - Partially starting up

    I brought this board because i have another one the same (which i still haven't fixed, due to parts not arriving) and it can be good to have two for the sake of comparison
    It was a non-working board, with no shorted power rails
    I connected it to a PSU, and the CPU fan briefly spun (can't recall if it was before, or after, pushing the power button)
    As i was probing the thermal protection chip ATTP1 voltages, the PSU flashed and made a bang sound, and blew the top off of a TNY276PN chip that's inside the PSU.
    May have just being coincidence

    So i got another PSU, (this time, connected to a dim bulb tester), but no rail voltages attempt to come up (even after connecting the PSU directly to the wall socket)

    ------

    PSON (green wire) is not being pulled low when pushing PWRBTN,
    so the PSU wont start

    RSMRST# comes up as 3.2V when the PSU is plugged in

    This board is starting powered on, so to speak - because voltages of around 3.1V are already present on SLP_S3# and SLP_S5# without pushing the power button
    So, i hold the power button in for a few seconds, to make the voltages drop off
    Briefly pushing then power button brings them back up again

    Looking at the attached schematic (thumbnail):

    PSON (-PSON) goes to pin 6, PSON_OUT#, of U1 (ATTP1, Thermal protection chip)
    When power button is pushed, pin 3 (3.5V) drops to 76mV and pin 6 (5V) stays high
    Holding the power button in brings the voltage back up on pin 3
    If i remove the CPU, the same happens (except pin 3 goes from 4.63V to 76mV when button is pressed)
    So i guess that that means that power is getting to the base of the transistor Q4 in on order to pull pin 3 low.

    From my understanding of the description of ATTP1, it seems that if pin 3 (PSON_IN#) is pulled low (as a result of the attached transistor being turned on) then pin 6 (PSON_OUT#) should also be pulled low (except in the event of a thermal event, where pin 6 is pulled high), and then the PSU should start up.

    I checked the resistors and capacitors that are attached to pins 4 and 5, and they are good
    Using another CPU (with lower frequency) increases the pin 3 voltage to 4.64V, but still same story as above.

    The base of the transistor Q4, SLP 3# looks like comes from the southbridge, and it's also on the ACPI chip.
    I checked the southbridge chip for visible damage, but it looks OK.

    Any tips/suggestions would be appreciated
    Attached Files
    Last edited by socketa; 05-19-2022, 05:02 PM.

    #2
    Re: MS-7061 - Partially starting up

    Ok, i had a feeling that i may have made a mistake
    So i went back and rechecked what was happening with SLP_S3# and SLP_S5#
    When i plug in the PSU, and turn on the PSU switch, the voltage to those two rails remains at about zero.
    Then when i push the motherboard power button, the voltages on both of those rails come up to about 3.1V
    So that's good.
    Since SLP_S3# provides voltage for the pull down transistor that connects to PSON_IN#, would it be prudent to bridge pins 3 and 6 of the thermal protection chip U1, and see what happens?
    Last edited by socketa; 05-25-2022, 01:56 AM.

    Comment


      #3
      Re: MS-7061 - Partially starting up

      Scrub that last post
      I put the replacement parts onto the other MS-7061 board that i have, and it works, and i noticed that it starts as soon as i turn on the switch that's on the PSU, it starts up
      So re-checked with this board, and it behaves similarly, as mentioned in the first post

      I compared to voltages of the pins of U1 with my other working board:

      Pin___________This board____________________Good Board
      ______________On/Off_______________________On/Off
      1_Siren#_______0.66/1.53____________________0.08/0.11
      2_GND___________0/0_________________________0/0
      3_PSON_IN#____0.07/3.56____________________0.04/3.94
      4_TXD_________0.69/0.90____________________0.75/0.76
      5_TXR_________1.66/4.44____________________3.69/3.68
      6_PSON_OUT#__5.08/5.06____________________0.12/5.07
      7_VREF________1.74/4.86____________________3.68/3.67
      8_5VSB________5.08/5.07____________________5.07/5.07

      Resistances to ground (with or without CPU installed) are
      1 -
      2 0K
      3 6.7K
      4 57.5K
      5 63K
      6 3.3K
      7 53K
      8 2.3

      Pins 4 (57.5Kohm), 5 (63Kohm), 7 (53Kohm) (which are all connected via resistors, and go to CPUTD), are 15Kohm higher, with and without the CPU installed on this board, compared to the working board
      Seems to me, that for pin 7 to be lower resistance than the other two pins, that the extra 15Kohm resistance would have to be from the direction back into the chip.
      From looking at the voltages, it looks like an over-temp siren is being generated
      The sensor is not installed underneath the CPU, so i guess that it uses a sensor in the CPU.
      Attached Files
      Last edited by socketa; 05-31-2022, 05:07 AM.

      Comment


        #4
        Re: MS-7061 - Partially starting up

        changing tact
        and looking at what's happening at the Q4 transistor:
        B_______C______E
        0.79V___1.02V___0V___(when power switch signals on)
        0V______1.92____0V___(when power switch signals off)
        and on the good board:
        0.71V____0V____0V___(when power switch signals on)
        0V_____3.95V___0V___(when power switch signals off)
        Last edited by socketa; 06-20-2022, 01:45 AM.

        Comment


          #5
          Re: MS-7061 - Partially starting up

          When i unplug the spinning fan on the good motherboard, i can see the temp of the CPU rising in the BIOS hardware monitor, and it goes back down when i plug it back in.
          Looking at the attached schematic, as well as the actual board, there is no CPU sensor (RT1) installed (it's location under the CPU is unoccupied),
          R2 and R3 resistors, that go to the CPU, are not installed,
          and the components labelled with an "X" are not installed.
          So how the heck is it possible for the I/O chip (and the AMD Thermal protection, U1) to monitor the CPU temperature?
          Attached Files
          Last edited by socketa; 06-22-2022, 02:48 AM.

          Comment


            #6
            Re: MS-7061 - Partially starting up

            The sensor in the CPU is being used - even though the components are not installed, they are connected by a PCB trace.
            OK - that's not the fault then, since i tried another CPU

            Comment


              #7
              Re: MS-7061 - Partially starting up

              OK
              i removed the attp1 chip and sat another one on top of the pads
              the resistance readings are now the same as the good board
              so it looks like the original chip is faulty
              should i solder it on and try to start it up?
              any safeguarding measures that i can take?
              i don't want to blow up another PSU like what happened before (even though it could just have being coincidence)

              Comment


                #8
                Re: MS-7061 - Partially starting up

                fixed!
                the chip was faulty
                maybe it was damaged by a power surge:
                soldered a new one on, and it didn't work, and it tested bad (Mohms VREF resistance to ground pin) - took it off, and it tested good
                soldered it back on, and the board booted up
                even though the soldering looked good, maybe one of the legs was insulated with flux.

                Comment

                Working...
                X