Recent shutdowns, reboots, other strange and various problems 
Bookmark this page:
 Yahoo!
 Google
 Windows Live
 del.icio.us
 digg
 Netscape
|
|
|||||||||||||
|
Posted by John Falstaff on February 20, 2006, 1:45 pm
Please log in for more thread options with no problems at all, until just a few days ago. Now it's driving me nuts. The system is as follows: Athlon XP 2200+ MSI KT2 Combo 512 MB of PC2700 GeForce4 Ti200 Samsung 710N monitor Maxtor 80GB IDE Sony optical drives (2) Sparkle 300W PS WinXP Home Ed. SP2 A few days ago the LCD monitor went crazy with odd-colored stripes, the system rebooted several times. Shut down and restarted with the same result. Later on, restarted and it was fine. That monitor problem has not returned. Yesterday the system repeatedly died on me. Sometimes it would reboot, fail to load Windows and reboot again, etc. Sometimes it would run long enough for me to get into BIOS, where all voltages looked correct, CPU and mainboard temp very moderate. Of course I suppose the voltages might have been correct only when it was running properly, but they did not seem erratic at all. Sometimes it would not run even enough for me to get into BIOS, and sometimes starting it up would not put *anything* on the monitor at all . Completely blank. It seemed more likely to start up after it had cooled for a while, so I assumed some sort of hardware problem. I've checked the connections, both outer and the cables inside, and that doesn't seem to be the problem. Because it had often failed to finish loading Windows I restarted and used F8 to get to the last good configuration, and that seemed okay at first. I also went back to an earlier restore point since I thought all these screwups might have garbled something. Today, more of the same. I have the side of the case off, and all the fans are running fine including when it dies. Right now I'm running in Safe Mode with network so I can use the cable modem. It's this machine I'm on now. It's been running just fine on Safe Mode, within the limitations of that mode of course. I have been able to get it to load Windows normally only a couple of times today, and it soon dies when I do so. I suspect the power supply, partly because the machine sometimes would not only fail to boot but even fail to put anything on the screen at all, and also I've noticed the PS fan noise seems to change at the time the system dies--though I'm not absolutely sure which comes first. But just in case I've missed something obvious (that happens), anything else I should check before I start looking for a new power supply? | |||||||||||||
|
Posted by kony on February 20, 2006, 2:51 pm
Please log in for more thread options A system reboot may point to a windows problem (or windows driver, application, etc- software). Check Event Viewer and disable Windows' "Automatically Reboot" setting in Startup & Recovery. If a Bluescreen error is produced, write down the main description and the stop error code. >
>Yesterday the system repeatedly died on me. Sometimes it would reboot, fail >to load Windows and reboot again, etc. Sometimes it would run long enough >for me to get into BIOS, where all voltages looked correct, CPU and >mainboard temp very moderate. Of course I suppose the voltages might have >been correct only when it was running properly, but they did not seem >erratic at all. Sometimes it would not run even enough for me to get into >BIOS, and sometimes starting it up would not put *anything* on the monitor >at all . Completely blank. If you had been in and out of the system, possibly this slightly changed electrical contacts of some part- confirm all cards, cables, memory, etc are well-sealed and aligned- not just aligned with motherboard brackets for example but aligned with the motherboard, since some cases are off tolerance a bit. Examine the motherboard and video card for failed capacitors- swollen, vented tops or residue on tops or bottoms. "Some" Ti41200 video cards were prone to this problem, especially a few Gainward cards but others too. In some cases the cards would even respond better to a warmer system, because capacitor function is better at higher temp (even though the temp decreases their useful lifespan). >
>It seemed more likely to start up after it had cooled for a while, Does it always shut off in the same way? Does it only try to reboot, or turn completely off, and if completely off do you ever have to unplug it from the wall to reset the power supply? Did you use decent synthetic based thermal compound (or the original AMD heatsink thermal pad), OR a lower quality silicone grease? The latter may degrade over time and while CPU temps may not look alarmingly high, portions of the CPU can be overheating due to uneven contact with the heatsink. Difficult condition to pinpoint except to just pull the 'sink off, examine the top and bottom of the CPU for dark areas (which occur from overheating but don't necessarily mean it's permanently damaged yet) and reapply the heatsink with a good quality synthetic grease such as Arctic Silver, Alumina, or any other specificially synthetic type. >so I
>assumed some sort of hardware problem. I've checked the connections, both >outer and the cables inside, and that doesn't seem to be the problem. >Because it had often failed to finish loading Windows I restarted and used >F8 to get to the last good configuration, and that seemed okay at first. I >also went back to an earlier restore point since I thought all these >screwups might have garbled something. > So is this problem progressively getting worse or about the same frequency? At some point you may want to run memtest86+ and Prime95's Torture Test (large in-place FFTs setting) to check stability of memory and CPU. If either has errors, even a formerly "good" backup or restore point could become corrupted even as it's being restored. >Today, more of the same. I have the side of the case off, and all the fans
>are running fine including when it dies. Right now I'm running in Safe Mode >with network so I can use the cable modem. It's this machine I'm on now. >It's been running just fine on Safe Mode, within the limitations of that >mode of course. I have been able to get it to load Windows normally only a >couple of times today, and it soon dies when I do so. You might also scan the system for viri, spyware, etc, though at this point there are a lot of potential issues and some feedback on the aforementioned things might be useful. >
>I suspect the power supply, partly because the machine sometimes would not >only fail to boot but even fail to put anything on the screen at all, and >also I've noticed the PS fan noise seems to change at the time the system >dies--though I'm not absolutely sure which comes first. So what exactly does "system dies" mean here? Does it freeze or reboot or shut off or? Fan noise change can be a failing fan but in this case more likely it is as you suggested, related to PSU voltage changes. Does the fan seem faster or slower? Your motherboard appears to use 12V for CPU power which may be a fairly balanced load on the PSU, yet the 300W Sparkle of that era may not have a lot of 12V amps and could be stressed. >
>But just in case I've missed something obvious (that happens), anything else >I should check before I start looking for a new power supply? Just the above things come to mind, though when selecting a new PSU you might choose one with more 12V amps, perhaps closer to 18A. Newegg has a pretty good deal on a Thermaltake 430W @ $30 AR right now, "IF" you end up needing a PSU. http://www.newegg.com/Product/Product.asp?Item=N82E16817153023 | |||||||||||||
|
Posted by John Falstaff on February 20, 2006, 7:31 pm
Please log in for more thread options
> On Mon, 20 Feb 2006 13:45:06 -0500, "John Falstaff"
> >>I built this system in July 2004 and it's been run about 10 hours every
>>day, >>with no problems at all, until just a few days ago. Now it's driving me >>nuts. >> >>The system is as follows: >> >>Athlon XP 2200+ >>MSI KT2 Combo >>512 MB of PC2700 >>GeForce4 Ti200 >>Samsung 710N monitor >>Maxtor 80GB IDE >>Sony optical drives (2) >>Sparkle 300W PS >>WinXP Home Ed. SP2 >> >>A few days ago the LCD monitor went crazy with odd-colored stripes, the >>system rebooted several times. Shut down and restarted with the same >>result. >>Later on, restarted and it was fine. That monitor problem has not >>returned. >
> A system reboot may point to a windows problem (or windows > driver, application, etc- software). Check Event Viewer and > disable Windows' "Automatically Reboot" setting in Startup & > Recovery. If a Bluescreen error is produced, write down the > main description and the stop error code. OK, I will do that. >
> >>
>>Yesterday the system repeatedly died on me. Sometimes it would reboot, >>fail >>to load Windows and reboot again, etc. Sometimes it would run long enough >>for me to get into BIOS, where all voltages looked correct, CPU and >>mainboard temp very moderate. Of course I suppose the voltages might have >>been correct only when it was running properly, but they did not seem >>erratic at all. Sometimes it would not run even enough for me to get into >>BIOS, and sometimes starting it up would not put *anything* on the monitor >>at all . Completely blank. >
> > If you had been in and out of the system, possibly this > slightly changed electrical contacts of some part- confirm > all cards, cables, memory, etc are well-sealed and aligned- > not just aligned with motherboard brackets for example but > aligned with the motherboard, since some cases are off > tolerance a bit. I haven't been into the system for some time before this happened. The last time was probably when I changed video cards, and that was months ago. But I'll check that anyway. >
> Examine the motherboard and video card for failed > capacitors- swollen, vented tops or residue on tops or > bottoms. "Some" Ti41200 video cards were prone to this > problem, especially a few Gainward cards but others too. In > some cases the cards would even respond better to a warmer > system, because capacitor function is better at higher temp > (even though the temp decreases their useful lifespan). OK, I'll check those things too. >
> >>
>>It seemed more likely to start up after it had cooled for a while, >
> Does it always shut off in the same way? Does it only try > to reboot, or turn completely off, and if completely off do > you ever have to unplug it from the wall to reset the power > supply? Sometimes it tries to reboot, sometimes not. I've never unplugged it from the wall, didn't know that resets the power supply. I'll try that too. >
> Did you use decent synthetic based thermal compound (or the > original AMD heatsink thermal pad), OR a lower quality > silicone grease? The latter may degrade over time and while No, it was a boxed CPU and I used the stock HSF without changing anything. > CPU temps may not look alarmingly high, portions of the CPU
> can be overheating due to uneven contact with the heatsink. > Difficult condition to pinpoint except to just pull the > 'sink off, examine the top and bottom of the CPU for dark > areas (which occur from overheating but don't necessarily > mean it's permanently damaged yet) and reapply the heatsink > with a good quality synthetic grease such as Arctic Silver, > Alumina, or any other specificially synthetic type. I doubt that's the problem, as I put this together about a year and a half ago and I'm always careful about seating the HSF. >
> >>so I
>>assumed some sort of hardware problem. I've checked the connections, both >>outer and the cables inside, and that doesn't seem to be the problem. >>Because it had often failed to finish loading Windows I restarted and used >>F8 to get to the last good configuration, and that seemed okay at first. I >>also went back to an earlier restore point since I thought all these >>screwups might have garbled something. >> >
> > So is this problem progressively getting worse or about the > same frequency? This dying and/or failure to start just started yesterday. It seems to have gotten somewhat worse, but when it started it was pretty bad. The earlier rebooting problem when the display went nuts apparently cleared up later the same day, and of course I don't know whether it's related to this or not. I suspect it probably is because it's the only other time this machine has ever tried to reboot by itself. > At some point you may want to run
> memtest86+ and Prime95's Torture Test (large in-place FFTs > setting) to check stability of memory and CPU. If either > has errors, even a formerly "good" backup or restore point > could become corrupted even as it's being restored. I did run Memtest86+ for about an hour, a few days ago when I had that monitor problem. No errors. I've never run Prime95 and don't have it, but I'll look for it--if I ever get this working properly. I also have a Ghost backup of the whole system, done several days ago. Of course that doesn't help me either at the present time. >
> >>Today, more of the same. I have the side of the case off, and all the fans
>>are running fine including when it dies. Right now I'm running in Safe >>Mode >>with network so I can use the cable modem. It's this machine I'm on now. >>It's been running just fine on Safe Mode, within the limitations of that >>mode of course. I have been able to get it to load Windows normally only a >>couple of times today, and it soon dies when I do so. >
> You might also scan the system for viri, spyware, etc, > though at this point there are a lot of potential issues and > some feedback on the aforementioned things might be useful. McAfee antivirus did find a worm on its own, and I think that was just yesterday too. I promptly did a full system scan then and it turned up nothing else. I don't recall the name of the worm. That was before this problem started, so maybe that's related. Never thought of that until you mentioned it. >
> >>
>>I suspect the power supply, partly because the machine sometimes would not >>only fail to boot but even fail to put anything on the screen at all, and >>also I've noticed the PS fan noise seems to change at the time the system >>dies--though I'm not absolutely sure which comes first. >
> > So what exactly does "system dies" mean here? Just stops. Sometimes I'm away from the machine and when I come back, the monitor is blank, its power light is blinking, the system appears to be running (i.e., power light is on, fans are running) but there's no response to any keyboard or mouse action. Last time it just stopped dead on me (i.e., monitor went blank) while I was doing something on the Internet. Hmmm. Maybe I should try a different monitor. I did check the monitor connections at both ends and they're good, pressure on them doesn't produce anything on the monitor. Still, it doesn't seem like it could be a monitor problem because I have it running right now (again in Safe Mode) and it's fine--in this mode. > Does it
> freeze or reboot or shut off or? Monitor goes blank, sometimes it tries to reboot, sometimes not. Turning the machine off manually and attempting to restart after a short pause sometimes doesn't do anything (monitor stays completely blank, doesn't even show any of the video card or BIOS info), but sometimes turns on enough for me to get intothe BIOS or Safe Mode. > Fan noise change can be a
> failing fan but in this case more likely it is as you > suggested, related to PSU voltage changes. Does the fan > seem faster or slower? Slower (slightly reduced sound), just before the computer dies. That had me wondering if perhaps the PSU has a heat problem or something, and is reducing power to protect itself which in turn makes my system collapse. But I am totally ignorant about PSUs. > Your motherboard appears to use 12V
> for CPU power Yes, it has that extra 12V 4-pin connector. > which may be a fairly balanced load on the
> PSU, yet the 300W Sparkle of that era may not have a lot of > 12V amps and could be stressed. I wish I could see the label with that info, but it's on the other side. This PSU goes into the case upside down as far the labels are concerned. Dunno why they do it that way. >
> >>
>>But just in case I've missed something obvious (that happens), anything >>else >>I should check before I start looking for a new power supply? >
> Just the above things come to mind, though when selecting a > new PSU you might choose one with more 12V amps, perhaps > closer to 18A. Newegg has a pretty good deal on a > Thermaltake 430W @ $30 AR right now, "IF" you end up needing > a PSU. > http://www.newegg.com/Product/Product.asp?Item=N82E16817153023 Yes! As a matter of fact I've already bought one, intending to use it in my next system. :-) Kony, thanks so much for all the suggestions. I do recall reading about swollen or leaking capacitors being common problems on motherboards, now that you mention it. I will check out everything you've mentioned and come back here with the results, either later tonight or tomorrow. Maybe I'll also change video cards just in case that's the problem--I have one or two GeForce2s I'm not using. Thanks again, Jack | |||||||||||||
|
Posted by John Falstaff on February 20, 2006, 9:02 pm
Please log in for more thread options
[ . . . ] Kony, you're a genius. And a much appreciated one, too. I couldn't see any swollen or leaking capacitors on the motherboard. But after what you wrote, about capacitors often being a problem with GeForce4 Ti4200 cards and especially Gainward (which is *exactly* what I have), and considering that these rebooting etc. problems started a few days ago with the video problems mentioned, I decided that was the most likely culprit--and, of course, the easiest one to check out first. So I pulled the Gainward card and sure enough, a few of the capacitors seem to have popped their vents (I presume that's what the cross-shaped indentations are on top) and leaked a tiny amount of brown stuff. I put in an old GeForce2 Ti and started 'er up, Windows sorted out the drivers and asked for a restart, and now everything is working perfectly again. Without your help I'd be preparing to put in a new PSU, to no purpose. Thanks a million! Jack | |||||||||||||
| Similar Threads | Posted |
| Recent shutdowns, reboots, other strange and various problems | February 20, 2006, 1:45 pm |
| Strange DVD-RW problem.... | April 1, 2007, 9:07 am |
| STRANGE PROBLEM WITH MY PC | January 24, 2006, 7:59 am |
| A recent epidemic of hangs? | February 23, 2005, 8:16 am |
| Strange problem on bootup | June 23, 2005, 4:18 am |
| Strange usb port problem | October 17, 2006, 3:51 am |
| Comp Freezing, and strange slowdowns | October 3, 2005, 8:17 pm |
| Re: Strange graphic card problem | August 16, 2007, 5:57 pm |
| Has the recent price drop in CPUs made it to the end user? Forgive me if this is a cross post. | July 30, 2006, 6:20 am |
| Why my PC reboots rather than shutsdown | July 25, 2005, 7:05 am |
| Random reboots | October 5, 2005, 2:07 am |
| PC Starts then reboots | October 5, 2006, 10:04 am |
| Computer intermittently reboots | August 30, 2005, 5:58 pm |
| Random reboots under Windows x64, HELP. | January 27, 2006, 6:52 pm |
| Frequent Reboots - How to Correct? | July 22, 2006, 9:31 am |
>with no problems at all, until just a few days ago. Now it's driving me
>nuts.
>
>The system is as follows:
>
>Athlon XP 2200+
>MSI KT2 Combo
>512 MB of PC2700
>GeForce4 Ti200
>Samsung 710N monitor
>Maxtor 80GB IDE
>Sony optical drives (2)
>Sparkle 300W PS
>WinXP Home Ed. SP2
>
>A few days ago the LCD monitor went crazy with odd-colored stripes, the
>system rebooted several times. Shut down and restarted with the same result.
>Later on, restarted and it was fine. That monitor problem has not returned.