Announcement

Collapse
No announcement yet.

Mobo in coma

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

    Mobo in coma

    Hi,
    I would like to ask all of you, experienced eli doctors, for little advice with my faint mobo. Ima just regular home user with DMM and courage.

    so in galaxy far aw.......
    Ive cleaned and re-pasted my notebook. Was stupid enough to use 10+ years old silver-based paste and possibly spilled something on gfx/cpu parts on their dye. Ive cleaned it as hard as I could.
    (acer aspire 5552G) s1s4 socket, phenom cpu, on Rs880m north bridge , dedicated gfx hd5650.
    mobo shema

    Now mobo switches on for like 3s , diodes shining, fan goin , then it dies. If its cold first-time start it takes like 5 seconds.
    It does not get into POST unfortunately (got POST card), therefore bad power is my suspision.

    *So Ive tested every rail, for bad voltage levels, found nothing.
    *All PowerGood signals are there.
    *Ive tryed to push hard BGA chips towards mobo for cold contact, no change.
    *None of the caps under cpu/gfx seems shorted.
    *Ive found litle trick in manual, when you press some combo on keyboard while pluging power, it enters different BIOS execute branch , where BIOS reads flash drive and re-flashes itself from file. While this "mode" is executed (without flash drive offcourse, dont wanna mess BIOS) , motherboard does not shutoff itself after 3secs , and stays ON untill I pull the plug. Even in this "mode" I didnt found faulty power rail or anything, however NorhtBridge gets hot quite fast , waayyy faster then gfx or cpu (Ive mounted cooling for this).
    *Ive soldered out "power jumper" off NB_CORE power rail (heard that Rs880m is shiz), to see if anything changes, dies in 3secs again.


    Im courious about some signals, there is logic AND that reads SB_PWRGD (is 1) and strange "LDT_RST" (reads 0V, but inverted output from SB). Dont understand the whole "LDT" / "HDT" abreviation, anyone know?


    Unfortunately dont have any logic analyzer or scope , so there I am =S.
    Hopefully with little trick , that plugs small diode and capacitor together I would be capable of spoting quick positive signals off rails that could escape DMM. Dunno how would that helped me, but thats it , exhausted and looking for heads up =)

    #2
    Re: Mobo in coma

    Originally posted by g0rak View Post
    . Dont understand the whole "LDT" / "HDT" abreviation, anyone know?
    AMD processors use HyperTransport technology for processor-memory communication. Under HT technology, the communication device (bus) is effectively within the CPU itself; this is known as the HT bus. There are two types of buses in this system, known as LDT (Lightning Data Transport) bus and HT bus; LDT connects with the motherboard chipset, and HT bus communicates with the system memory.

    Comment


      #3
      Re: Mobo in coma

      This sounds sort of like a CPU problem. Do you know if you got any compound on the little components on the top-side of the CPU? Also try booting it without the CPU, does the same thing happen?
      Dell E7450 | i5-5300U | 16GB DDR3 | 256GB SSD

      Comment


        #4
        Re: Mobo in coma

        Thx for answers.

        Yea both my gfx and cpu got some paste "under" those tiny fagot parts on them. Does not matter in case of CPU, because mobo should start to POST without it right ?
        Spilled GPU might seem bad, but Ive actualy tested those tiny parts , and they seem fine , I mean they usualy form some groups with same resistance (like 75 / 30)... and none off them are shorted. Actualy Ive tryed to measure resistance of that silver compound and found that Im unable to measure anything, even with 2Mohm range , and like half milimeter distance for probes.

        Comment


          #5
          Re: Mobo in coma

          Out of curiosity, can bad northbridge cause mobo not-POST ?
          I mean If it (NB) would be unsoldered from mobo , would the "working" mobo POST ? Or it depends... (on bios etc..)?

          Comment


            #6
            Re: Mobo in coma

            No. If the CPU is not working it will not POST. The BIOS contains code, it is program in memory, not a processor, the CPU has to be there to execute the code.

            Comment


              #7
              Re: Mobo in coma

              That is the word on the table ! Thank you sir ! I did not realized that
              For another thing , how usualy mobos act when turned on without CPU ? Is possible that Im simply hunting ghost, with "just" dead cpu ?

              Comment


                #8
                Re: Mobo in coma

                You're welcome. I don't think it would do anything at all without the processor. I suppose it could depend on the MB though. I haven't fired up a MB without the CPU that I recall.

                Comment


                  #9
                  Re: Mobo in coma

                  If you power it on without cpu usually the power on leds light for nearly 5 secs and then shuts down immediately.

                  Comment


                    #10
                    Re: Mobo in coma

                    Originally posted by lefos View Post
                    If you power it on without cpu usually the power on leds light for nearly 5 secs and then shuts down immediately.
                    Hmm that is some heavy weight info too =). Didnt find that anywhere and its actualy quite crucial behavior imho. Thank you sir !


                    That kinda vouches for my conclusion ... dead cpu, since the silver paste spilling incident

                    Comment

                    Working...
                    X