thread top ad



No announcement yet.

820-3437B quarter fan in loop

  • Filter
  • Time
  • Show
Clear All
new posts

  • 820-3437B quarter fan in loop

    Hi team,

    PPBUS_G3H: 8.59V
    PP3V42: 3.401V
    PPRTC_G3H: 3.310V
    PP5V_S5: 4.994V
    PP3V_S5: 3.310V
    S4 State pulsing

    SMC_ADAPTER_EN : 3.366V
    PM_BATLOW_L : 3.305V
    PM_DSW_PWRGD: 3.398V
    SMC_PM_G2_EN: 3.396V

    Checked all SPI Bus Series Termination traces and resistors
    Put flux and heated SMC and SPI

    Re programmed U6100 with medusa: NOK
    Replaced U6100 and re programmed (Cleaned Pram, ME, SN Crc ): Not Ok
    Still quarter Fan Spin
    Sometimes When the medusa is connected fan starts running normally.
    Re programmed with 00165 Dump bios just for try : Board Starts but no Chime, No backlight and No display.
    Back to 3437B bios and pulsing state
    Maybe not SPI finally

  • #2
    All big coils pulse too?
    Also check ALL_SYS_PWRGD and pins 3/7 of U1950.

    Did you try SMC bypass?


    • #3
      “sometimes runs normal with Medusa connected” any signs of corrosion on J6100?


      • #4

        All big coils pulse too?
        Also check ALL_SYS_PWRGD and pins 3/7 of U1950.
        Pulsing Too , everything pulses to Vcore

        Did you try SMC bypass?
        YES: Pusles too : forces the fan to enter in quarter Spin loop


        Yes checked and resoldered all pins : No sign of corrosion directely on the J6100 but Tests Points around have the gold color turned to brown

        Board Starts only when the Medusa is writing which is very ennoying , I have to disconnect reconnect the charger many times during the process for the medusa to finish writing.
        Last edited by Bouba148; 07-19-2019, 11:54 AM.


        • #5
          Maybe Crap under CPU preventing PM_SLP_S4_L to be high?


          • #6
            If coils pulse, means PM_SLP_S4_L pulses too.

            "Board Starts only when the Medusa is writing"
            Yesh, stays on, but doesn't work.
            To avoid future problems with Medusa, dosn't use MacBook charger.
            Inject external 3.3V on 3V3_SUS rail and never get problems again.

            In this case, when the board stays on because of Medusa "help", check all voltages from pages 57/58.


            • #7
              What's your idea reagrding: all voltages from pages 57/58?
              do you think that one of those is shorted to gnd and drops down the PP3V3_S5?


              • #8
                Page 57:

                1.8V S3 REGULATOR:
                1.8V Pulsing because P1V8S3_EN pulsing too.

                1.5V S0 LDO:

                PP1V5_S0 Pulsing as it's directely depending of 1.8V S3

                1.05V SUS LDO:
                PP1V05_SUS: 1.061V OK

                Page 58:

                PP1V5_S0SW_AUDIO_HDA Pulsing

                3.3V SUS Switch OK

                3.3V S4 Switch: Pulsing as depending S4_PWR_EN (Pulsing)

                PP3V3_S3_FET_R Pulsing as P3V3S3_EN Pulsing too

                3.3V S0 and PP3V3_S0SW_SSD pulsing as P3V3S0_EN pulsing

                PP3V3_S4SW_SNS pulsing as SMC_SENSOR_PWR_EN pulsing

                PP5V_S0 Pulsing as P5VS0_EN pulsing too

                PP1V05_S0SW_PCH_HSIO pulsing as depending of PP5V_S0

                PP1V05_S0 Pulsing


                • #9
                  If power sequence is OK, then should be communication problem between some blocks, maybe bad CPU.
                  Check if PLT_RESET_L and BUF_PLT_RST_L pulse both high.


                  thread bottom ad