I have an HTPC that I use for primarily recording shows (playback is handled by other client devices). It had been working fine till late Saturday night. Here's the configuration:
PC Chips M811LU Socket A Motherboard
AMD AthlonXP 2200+,
Crucial 2048MB RAM - brand new purchased in December and ran through overnight memtest in January.
Windows XP MCE 2005 with update rollup 2 and SP3
Hauppauge HVR 1600 TV tuner card
When I turned it on Sunday morning, I noticed it would randomly keep rebooting. I spent about 3 hours trying out different things. Things that I consistently noticed in this time were:
a) When the computer is first powered ON after it has been turned OFF for atleast few minutes, it seems to take few seconds (almost 5-10 seconds) before the POST screen even shows up on the monitor.
b) With each attempt at booting, the PC gets a bit further than the previous attempt at booting up. So the first couple times, it does not even make it past the RAM check. After that, it takes it a couple more attempts to get past the peripheral detection stage. Finally, after it gets past the peripheral detection stage, it takes a few attempts before Windows completely loads. Even after Windows finishes loading, it never seems to stay ON for more than 2-3 minutes. So I am unable to get too far with running any diagnosis tools for any meaningful amount of time. I was hoping to atleast go into Windows Event Viewer to see if it reported any issues while booting that may give a clue but was not able to do so before it rebooted.
c) Another odd thing I had noticed when I had first plugged in the computer to diagnose the problem was the computer power light stayed off, the HDD light stayed on. Also, the DVD drive kept constantly blinking and I was unable to eject the DVD drive.
Things I tried unsuccessfully in the three hour frame to help figure out what had gone wrong and/or to resolve the issue:
1. Disconnected DVD burner for the entire time I was trying to diagnose/repair the issue.
2. Swapped Video cards - Kept swapping between two different video cards and didnt notice a difference in the behavior
3. Removed all peripheral cards - only other peripheral card I had in the system besides the video card was the TV tuner card.
4. Swapped power supply (Ultra V-series 500W) with Antec EA-380.
5. Disconnected all HDDs except the one with the OS. After that swapped bootable HDD with a different bootable HDD.
6. Swapped RAM with a pair of older RAM chips
Up until this point, it would not even get past the initial POST screen.
7. Removed and replaced the CMOS battery - After doing this, it would atleast make it through the CMOS RAM check but would then give CMOS checksum and stop booting. So I went into the BIOS and loaded the failsafe boot option and even turned off most (but not all) unused on-board perihpherals (e.g. parallel port, etc. I should also note that I had removed the CMOS battery for only maybe 5 seconds or so before putting it back in, so not sure if it was long enough to reset the memory. But I am hoping the reset with the jumper (see below) might have
8. Inspected motherboard to see if any capacitors might have blown (bulging or leaking capacitors) but didnt notice any.
9. Disconnected and turned off all case fans and turned them back on one by one.
10. Disconnected fans from motherboard and plugged into power supply using adapter and vice versa.
11. Reset the CMOS memory by shorting the jumper as described in the motherboard manual (see attached). The jumper was left in the reset position for several minutes, so I am assuming this had reset the CMOS memory even if the battery method may not have been done correctly? After resetting this way though, I did not do much additional testing as the symptoms were no different than before resetting the jumper.
Other things that have crossed my mind but I have not tried yet:
1. Removing CPU fan and heatsink to check for overheating issues. This is because I had applied a fresh coat of Arctic Silver 5 last summer because the CPU kept overheating/causing BSOD issues (but even at that time, it would atleast run for 10-15 min before shutting down). Since I applied Arctic Silver 5 less than a year ago, I wonder if CPU overheating could still be a possible cause?
Would appreciate any opinions on what could be causing this behavior and/or any suggestions for things to try help determine the cause of the problem.
Thanks much for reading this entire post and any input you may provide,