Announcement

Collapse
No announcement yet.

A1418 820-00431-A Looping LEDs

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

    A1418 820-00431-A Looping LEDs

    Hello. Attempting to repair 2 different logic boards with i5 CPU + 8GB iMAC A1418 2k.

    Each shows the first bottom LED on = power supply is good.

    Power button is pressed -> LED 2 and then LED 3 each turn ON and then reset and recycle to only LED 1 (power supply) ON.

    Checked against another logic board and found that inductor linked to FDMF6808N near CPU is @ 12.x Mohms.

    Yet, on the board with the 'power cycling' is @ 2.x Mohms or lower.

    Should FDMF6808N be replaced? Found some @ MobileSentrix. Weird that ONSEMI / Fairchild is not showing this component.

    Is this an EOL component?

    Welcome any suggestions for this repair. Now also have the DS-809SE tool as of yesterday. Should be able to read out the flash firmware for public posting.

    #2
    Re: A1418 820-00431-A Looping LEDs

    Hello,
    have you done anything else with that board ?

    I have (had) the same problem (cycling LEDs 1-3)

    First step was checking for obvious visible problems, didn't find any except for an unusally large amount of loose solder balls all over the board on both sides.

    After reassembling the iMac the problem stayed the same for a few minutes, then it changed to LED 1+2 on steady and fan starting/running at full speed.

    Schematics around U7000/ISL6372 of 820-3588 are similar, so i was able to find a short on PPVCCIO_S0_CPU (--> dead CPU) and/or pin 13/14 of U7000 (CPU_VIDSOUT/VIDALERT_L)

    I think i have found/checked all other parts that may short PPVCCIO_S0_CPU, so this board is done for now. Probably killed by a solder ball rolling to the wrong place)

    Is there a datasheet for ISL6372 available anywhere ?

    PS i have succesfully replaced FDMF6808N/FDMF6708N with FDMF3030 (on 820-3662 etc.)

    Comment


      #3
      Re: A1418 820-00431-A Looping LEDs

      Hi. Thanks for your comments. With regret, no - had to shelve the repair till later.

      If the fan is running on full speed - check your thermal sensor for the hard drive. It has to be present or as a plan B - run a program to dial down the fan speed but only if you are confident the thermal cooling is ok in your box. For example, if using a SSD, you could run a program to override the fan speed.

      Some logic boards have the need for a 2 pin thermal sensor which can be built using a 5 cent transistor (2N3904):

      https://www.youtube.com/watch?reload...&v=inLZ9_SQOHU

      Before the above noted repairs, your hands could almost burn from the backside of the chassis with over heating. After the new paste & thermal pads - cool as a whistle. Be sure that the large heatsink plumbing is tight on the CPU & video card chips.

      The argument we have is that it took around 10 years for the quirks to be visible. With some fresh thermal paste & pads which have advanced a great deal since this box was manufactured, the fix works.

      Hope to study this unit in the near future.
      Last edited by piernov; 08-04-2021, 06:20 AM. Reason: typical reflow bullshit whatever

      Comment

      Working...
      X