Announcement

Collapse
No announcement yet.

Macbook Pro A1990 - 820-01041 Stuck at 5V / 0.02A

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

    Macbook Pro A1990 - 820-01041 Stuck at 5V / 0.02A

    Hello,
    I am currently working on a Macbook Pro A1990 (motherboard 820-01041) which is not charging.
    The Macbook consumes 5V and 0.019A at USB-C. There seems to be a partial short-circuit on the PPBUS_G3H rail as I have 0V on this rail.

    I've physically separated (by removing F7000 and F7001) PPBUS_G3H and the ISL9240 charging chip to see which side the short is coming from. Removing these fuses gives me 12.6V at the output of the ISL chip (PPVBAT_G3H_CHGR_REG).
    However, I can't work out exactly why I don't have the PPBUS_G3H rail.

    Here are the values I measured:

    PPBUS_G3H: Diode = 0.006 / Resistance to ground = 6.7 ohm
    PP20V_USBC_XB_VBUS = 5V
    PP3V3_UPC_XB_LDO = 3.3V
    PP1V8_UPC_XB/XA/TB/TA_LDOA = 1.8V
    PP1V8_UPC_XB/XA/TB/TA_LDOD = 1.8V
    PP1V1_UPC_XB_LDO_BMC = 0V
    PPDCIN_G3H = 5V
    SMBUS_SMC_4_G3H_SDA = 0V
    PP3V3_G3H_RTC_X = 0V /Diode = 0.4 / Resistance to ground = 800kOhm

    I noticed that the RB703 and RB731 resistors have lower values than expected: RB703 = 7.3kOhm and RB731 = 16kOhm (instead of 27.4kOhm).

    I've tried injecting 1.0V on the PPBUS_G3H rail but nothing heats up and it draws 0.330 Amps maximum.

    Don't hesitate to ask me if you need any further information. Thanks in advance!

    #2
    Remove all power. Meter in resistance mode.

    Measure the resistance from the PPBUS_G3H rail where you are measuring 6R7 ohms (one meter probe) to the any inductor on the board. You are hunting for a low resistance path.

    Continue to test from the same PPBUS_G3H point to each and every inductor. Start with the CPU region. If you find a low resistance then most likely you have a defective high side mosfet which usually means the CPU and/or parts on that power rail are dead. Also usually a no-fix situation.

    Comment


      #3
      Thank you very much for your prompt reply.

      If I've understood correctly, I need to measure the resistance between the point where I measured 6.7ohm and all the inductors near the CPU? For example, I put my multimeter in resistance mode, one probe on fuse F7000 where PPBUS_G3H goes and the other probe on inductor L7420 (where there is CPUGT_SW2 on one side and PPVCCGT_CPU_PH2 on the other)? I'm getting 0.01ohm on this inductor.

      Comment


        #4
        Yes. This is of concern. The PPBUS_G3H = PPBUS_HS_CPU = Vin on U7420 buck regulator. Internal to U7420 is a high and low side mosfet pair. If there is a leaky mosfet internal to this regulator, then you will show this low resistance. Respectively, the high voltage of PPBUS_G3H would have powered this very low voltage (~1 volt) CPU core voltage rail. Usually his means the CPU is dead and this will be a no-fix case.

        Can you confirm if F7000 is removed off the board? You are measuring from F7000 PCB pad # 2 (aka downstream / consumer side of the regulator circuit) to the inductor L7420 (either side of this inductor is ok)?

        Also, when you inject 1 volt (nothing higher) onto PPBUS_G3H pin # 2 downstream side, do you see this same voltage @ L7420? You will need extra hands to perform this measurement but important to know. If yes, then it appears the internal high side mosfet has leaked for the Vcore rail.

        Next, without power, measure the resistance of L7420 to ground. Most likely the CPU is dead but let us check a bit more.

        Comment


          #5
          Yes F7000 is removed off the board and I'm measuring from F7000 PCB pad#2 to the inductor L7420 and I'm getting 0.01 ohm.

          When i inject 1.0V on the F7000 pad#2 (F7000 removed) i'm getting the same voltage on L7420.

          Now, without power, the resistance to ground of L7420 is 6.7ohm (same as PPBUS_G3H rail).

          So the first problem came from the U7420, which then fried the CPU? Is there any way of replacing the CPU without losing the data?

          Comment

          Working...
          X