Announcement

Collapse
No announcement yet.

Computer troubles...

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

    Computer troubles...

    My home PC is giving me some trouble, it's not booting up properly.

    The power cycles on and off continuously. It doesn't show any signs of activity and only turns on for about 3 seconds before turning off for one, then repeats this cycle continuously. I have bench tested the power supply and that seems OK. It stays on when PS ON is held low. I've also scoped the rails on the outputs and they seem reasonable when under normal load, 12V at 12.2V, 5V at 5.09V, 3.3V at 3.37V. I'm thinking it's the motherboard, which is a Gigabyte Z97X-UD3H.

    Does anyone have any experience of this issue?

    Have done the obvious, like disconnecting all peripherals and drives, all fans, and running with only one stick of RAM. Will try another PSU just in case, also need to reset the CMOS or maybe use the backup BIOS flash on the board (apparently it has two).

    Took power supply apart (EVGA NEX650G), no bad capacitors visible.
    Last edited by tom66; 02-08-2018, 02:03 AM.
    Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
    For service manual, schematic, boardview (board view), datasheet, cad - use our search.

    #2
    Re: Computer troubles...

    i have seen a bad cpu do that.
    i think it had a short in a section that is not active at boot.

    Comment


      #3
      Re: Computer troubles...

      Originally posted by stj View Post
      i have seen a bad cpu do that.
      i think it had a short in a section that is not active at boot.
      I'm hoping it's not that. I'd expect CPUs to be quite reliable but I'll need to rule it out...
      Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
      For service manual, schematic, boardview (board view), datasheet, cad - use our search.

      Comment


        #4
        Re: Computer troubles...

        simple test,
        if you go into the bios settings - does it stay on?

        my one did - it was only when the o.s. started ramping up the system that it doed.

        Comment


          #5
          Re: Computer troubles...

          No, it only powers up for about 3 seconds and during that time there's no video. There's barely enough time for the CPU fan to start spinning.

          It's like one of the very first checks is failing.
          Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
          For service manual, schematic, boardview (board view), datasheet, cad - use our search.

          Comment


            #6
            Re: Computer troubles...

            Bad BIOS would be my first guess, short on CPU VRM would be my second, dead chipset would be the third.
            Originally posted by PeteS in CA
            Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
            A working TV? How boring!

            Comment


              #7
              Re: Computer troubles...

              try to clean ram slot thoroughly with isopropyle and reseat ram

              Comment


                #8
                Re: Computer troubles...

                Check VGA cable (if have there). I had similar problem on mine. Turned out it was bad video cable (why it did power cycle I can't explain - simply I have no idea, I know it is weird issue).
                Laptops and computers repair.

                Comment


                  #9
                  Re: Computer troubles...

                  Originally posted by tom66 View Post
                  Have done the obvious, like running with only one stick of RAM.
                  try running without any ram at all. the board should beep and complain of no ram inserted. could be the ram gone bad in this case. what ram is it anyway? if its corsair ram, quite likely it may have gone bad. i recently had one of my p4 pcs go bad due to the ram. cause was found to be a bad psu killing the crucial/micron ram.

                  Comment


                    #10
                    Re: Computer troubles...

                    Originally posted by tom66 View Post
                    No, it only powers up for about 3 seconds and during that time there's no video. There's barely enough time for the CPU fan to start spinning.
                    Also, check the socket and the traces to the socket!
                    ASRock B550 PG Velocita

                    Ryzen 9 "Vermeer" 5900X

                    16 GB AData XPG Spectrix D41

                    Sapphire Nitro+ Radeon RX 6750 XT

                    eVGA Supernova G3 750W

                    Western Digital Black SN850 1TB NVMe SSD

                    Alienware AW3423DWF OLED




                    "¡Me encanta "Me Encanta o Enlistarlo con Hilary Farr!" -Mí mismo

                    "There's nothing more unattractive than a chick smoking a cigarette" -Topcat

                    "Today's lesson in pissivity comes in the form of a ziplock baggie full of GPU extension brackets & hardware that for the last ~3 years have been on my bench, always in my way, getting moved around constantly....and yesterday I found myself in need of them....and the bastards are now nowhere to be found! Motherfracker!!" -Topcat

                    "did I see a chair fly? I think I did! Time for popcorn!" -ratdude747

                    Comment


                      #11
                      Re: Computer troubles...

                      Fixed it. Was very simple.

                      I simply jumpered the CLR_CMOS jumper on the motherboard while the system was powered. This erases the configuration of the system. The system immediately rebooted and worked; I just had to reconfigure my SATA drives to get Windows to boot.

                      I tried the backup flash method (which involves holding the power button while power cycling using the switch on the PSU) but that did not help so it was not a corrupted BIOS flash, just corrupted configuration.

                      Really odd though that the configuration would go bad after some time, but happy in the end that it was an easy fix. Also seems silly that the system had no way of recognising it was in a boot loop (e.g. five failed boot attempts in a row, load default config).
                      Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
                      For service manual, schematic, boardview (board view), datasheet, cad - use our search.

                      Comment


                        #12
                        Re: Computer troubles...

                        that's shitty coding, the bios should have checksum'd the cmos before trusting it!

                        Comment


                          #13
                          Re: Computer troubles...

                          Originally posted by stj View Post
                          that's shitty coding, the bios should have checksum'd the cmos before trusting it!
                          A checksum isn't the cure. All that (theoretically) does is ensure the data hasn't been "corrupted" outside of the control of the BIOS.

                          If, OTOH, the BIOS crams a bogus value somewhere and then factors this into its checksum calculation, there is nothing "wrong" with the resulting checksum... it will continue to vouch for the "bogus value". Every time the BIOS checks that checksum!

                          You need, instead, sanity checks on individual values (and groups of values) that can affect the operation of the machine in ways that could lead to a crash, failure to boot/POST, etc.

                          And, ideally, to choose data representations that make "bogus" values harder to create (or easier to recognize). For example, an "elapsed time" datum need never represent a negative value. So, choosing a representational form that supports negative values just means you've now got a whole range of bogus values that you have to explicitly check for ("Is the value less than zero? If so, it's bogus...")

                          Comment


                            #14
                            Re: Computer troubles...

                            Originally posted by stj View Post
                            that's shitty coding, the bios should have checksum'd the cmos before trusting it!
                            It wasn't the BIOS, it was the CMOS...
                            ASRock B550 PG Velocita

                            Ryzen 9 "Vermeer" 5900X

                            16 GB AData XPG Spectrix D41

                            Sapphire Nitro+ Radeon RX 6750 XT

                            eVGA Supernova G3 750W

                            Western Digital Black SN850 1TB NVMe SSD

                            Alienware AW3423DWF OLED




                            "¡Me encanta "Me Encanta o Enlistarlo con Hilary Farr!" -Mí mismo

                            "There's nothing more unattractive than a chick smoking a cigarette" -Topcat

                            "Today's lesson in pissivity comes in the form of a ziplock baggie full of GPU extension brackets & hardware that for the last ~3 years have been on my bench, always in my way, getting moved around constantly....and yesterday I found myself in need of them....and the bastards are now nowhere to be found! Motherfracker!!" -Topcat

                            "did I see a chair fly? I think I did! Time for popcorn!" -ratdude747

                            Comment


                              #15
                              Re: Computer troubles...

                              You may want to double check for a weak CMOS battery; sometimes being low in voltage may change some setting on the fly and produce unexpected or strange problems.
                              The most notable that remember was one that "decided" to boot without sending video to any output on any graphics card, but without errors (in fact, via teamviewer was possible to use the pc). The other got a sudden lost of the mouse and refused to get detected in at least two windows versions and a couple of Linux distros; I think even in DOS wasn't detected.
                              Both returned to normal just unplugging the power and disconnecting the battery, so as a rule of thumb with computers with random problems one test that I do is take out the CMOS battery to see if changes something.

                              Hope this helps.

                              Comment


                                #16
                                Re: Computer troubles...

                                Originally posted by hasefroch View Post
                                You may want to double check for a weak CMOS battery; sometimes being low in voltage may change some setting on the fly and produce unexpected or strange problems.
                                The most notable that remember was one that "decided" to boot without sending video to any output on any graphics card, but without errors (in fact, via teamviewer was possible to use the pc). The other got a sudden lost of the mouse and refused to get detected in at least two windows versions and a couple of Linux distros; I think even in DOS wasn't detected.
                                Both returned to normal just unplugging the power and disconnecting the battery, so as a rule of thumb with computers with random problems one test that I do is take out the CMOS battery to see if changes something.

                                Hope this helps.
                                Oddly, the system kept the time post-reset (so I don't think the battery was low), but I'll keep an eye on it, thanks.
                                Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
                                For service manual, schematic, boardview (board view), datasheet, cad - use our search.

                                Comment


                                  #17
                                  Re: Computer troubles...

                                  The CMOS checksum fail also depends on how such a checksum is calculated. Simple checksums like sum-of-all-values can fail to detect many types of error, for instance substituting 0x00 with 0xff will fool such a checksum.
                                  Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
                                  For service manual, schematic, boardview (board view), datasheet, cad - use our search.

                                  Comment


                                    #18
                                    Re: Computer troubles...

                                    Originally posted by tom66 View Post
                                    The CMOS checksum fail also depends on how such a checksum is calculated. Simple checksums like sum-of-all-values can fail to detect many types of error, for instance substituting 0x00 with 0xff will fool such a checksum.
                                    You choose/design the polynomial to detect the types of errors you expect to encounter.

                                    All checksums/hashes reduce information -- you're trying to "summarize" some (large?) number of bytes of data with just a few bytes (e.g., two, for a 16 bit checksum). That opens the door for any number of "collisions" in the dataspace -- different combinations of data bytes that yield the same checksum/hash.

                                    E.g., a simple "check digit calculation" will yield the same result for the sets of data (1,2,3,4,5}, {2,3,4,5,1}, {3,4,5,1,2}, etc.

                                    So, if you expect your valid "1,2,3,4,5" to be corrupted to "4,3,5,2,1", then that sort of approach doesn't buy you much!

                                    Note that naively storing multiple copies of the data also buys you very little (esp given the overhead involved).

                                    And, a single checksum/hash for data having varying degrees of importance is also foolish. You probably care differently about the "asset identifier" stored in the BIOS NVRAM than you do about the "boot order" or IP address of the LoM system! So, why fold "less important" parameters into the calculation of a hash for MORE important parameters, thereby increasing the odds that the hash will fail (on POST) rendering those more important parameters as "suspect"?

                                    Comment


                                      #19
                                      Re: Computer troubles...

                                      Glad it was this easy and you are back up and running. Be aware that on some motherboards, setting the clear CMOS jumper to clear with power on will corrupt the BIOS, requiring a reflash.

                                      Another trick to keep in mind for the dual-BIOS boards is that if you short any two data lines of the main BIOS chip while powering on the system, the motherboard should automatically read from the backup BIOS.
                                      Originally posted by PeteS in CA
                                      Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
                                      A working TV? How boring!

                                      Comment


                                        #20
                                        Re: Computer troubles...

                                        Originally posted by Curious.George View Post
                                        You choose/design the polynomial to detect the types of errors you expect to encounter.
                                        Indeed, if you are a competent engineer!

                                        Sometimes, the lazy approach is taken, which can be to sum values together, which is vulnerable to many types of error and really only detects single bit errors reliably. A correctly designed CRC will pick up most types of bit substitution and flip errors. An even smarter system would use something like Reed-Solomon, allowing errors to be identified and corrected.

                                        Who knows what the manufacturer of the BIOS or motherboard did in this instance?
                                        Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
                                        For service manual, schematic, boardview (board view), datasheet, cad - use our search.

                                        Comment

                                        Working...
                                        X