Announcement

Collapse
No announcement yet.

Zotac RTX 3060Ti Graphics card PCI-E data lanes

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

    Zotac RTX 3060Ti Graphics card PCI-E data lanes

    Hi all,
    Have a Zotac RTX 3060Ti that I recently picked up, gives POST error in PC, wont start.

    Following the usual advice of things to check I found it was failing at PEX, looks like the enable isnt coming from upstream.
    But then thought I'd check the data lines on Diode test, found the SMBus and all the transmitter pairs on front showing as O/C, I thought its likely the card had been used as a miner and they possibly plugged the 1x riser in wrong way - but that shouldnt have affected the rest of the lanes - just the ones in the 1x positions - No?
    Also noted that if I reverse bias the lines, I do get a reading of appx 0.325V on the pairs and 0.5v on the SMBus - Where's this reading coming from if the inputs are damaged - is it some leakage through the core?

    Ultimately do you think the GPU is completely done?

    #2
    Re: Zotac RTX 3060Ti Graphics card PCI-E data lanes

    BUMP - Anyone with any opinion on this?

    Comment


      #3
      Re: Zotac RTX 3060Ti Graphics card PCI-E data lanes

      Diode mode is measured with red probe on ground and black probe on the signal.
      OpenBoardView — https://github.com/OpenBoardView/OpenBoardView

      Comment


        #4
        Re: Zotac RTX 3060Ti Graphics card PCI-E data lanes

        Originally posted by piernov View Post
        Diode mode is measured with red probe on ground and black probe on the signal.
        I wondered this, even checked if my meter(s) wanted the leads in different sockets for diode test mode.
        Since, if thats the case - I DO in fact get valid readings, but when I double checked some repair vids online, they have the black lead on chassis and red on contacts...

        Also that way round will give me valid readings for B side(front) connections on transmit lines (suppose thats receive on card) whereas conventional (Blk-Gnd) testing on A side(rear) for the Receive lines (card transmit) give readings.

        I would like this to be valid as it means I've still got hope for the card...

        But just took out a working card from one of my other computers, that gave volt drop readings on front terminals-both ways round. Rear lines, measured after the caps, only gave reading conventional way round

        Still not definitive?

        Comment

        Working...
        X