Random system shutdown

veri745 · 04-04-2007, 05:38 PM

I know there are several other posts with similar titles. I've read most of them, and was either not convinced my problem was the same, or have already tried the suggestions.

I've tried other forums for this issue already, please review these links:
Abit Forum
OCZ Forum

* Module size and kit size:-2 x 1gb Corsair XMS
* CPU model and rated speed:-Athlon X2 4200+
* Overclocked CPU speed:-2200 Mhz - stock
* Divider and FSB used:-
* Motherboard and revision:-Abit Fatal1ty AN8
* Motherboard Bios rev:-
* Video card:-BFG GeForce 6800GTOC
* GPU Temps: 58 C idle; 70 C load
* Power supply:- OCZ Modstream 520W
* Voltages (tested using DMM):
12V rail: 12.01 idle; 11.87 load
5V rail: 5.11
3.3V rail: 3.37
* CPU and System Cooling:-Zalman 9500; 4x 80mm case fans
* CPU and System temps:-cpu: 30 idle, 45 max load
* HDDs: 80GB IDE (Windows); 160GB SATA (games, programs); 200GB IDE (storage)
* Country:-US

Problem: System power off
Frequency: Random
Repeatability: No, but occurs MUCH more frequently in 3D applications

System will power back up immediately. No message from Windows saying computer was not shut down properly. No errors in system log.
Problem occurs most often in games (Counter-Strike: Source, Europa Universalis III).
Occasionally it has shut down during normal Windows use (internet, messaging).

I have basically ruled out an overheating issue (please read my other forum links)
I have tested my memory extensively using memtest86 for several hours.
I have tested my processor for several hours using Prime95.
According to the power supply calculator linked in some of the other threads, my system need 334W + 30% = 435 W, which my power supply should handle easily.
I use ABIT uGuru to monitor system temperatures and voltages, and have also checked my PSU voltages using a DMM.
Video card has been replaced with no change in symptoms (I tried an x850xt which I was RMAing my 6800).

Let me know if you need any more information. I am not a novice, but this issue has me stumped.

Tumbleweed36 · 04-04-2007, 06:00 PM

Hi,

Just for comparisons of voltages, temps, and fan speeds, why not try both Speedfan and SensorviewPro for other views of what is going on in there:

Speedfan will tell you the temperatures and a whole lot more.

http://www.sofotex.com/SpeedFan-download_L4655.html

SensorView Pro

http://www.stvsoft.com/download.php?id=122

veri745 · 04-04-2007, 06:54 PM

k, i guess i'll humor you. These readings are almost EXACTLY what I get out of uGuru (infact, Speedfan uses uGuru to get it's readings).
The HDD temps are nice though.

Speedfan reports:
Local: 44C
Remote: 45C
HD0: 34C
HD1: 31C
CPU: 24C
SYS: 36C
PWM: 35C
Core: 28:
Voltages:
HTV: 1.18
VCORE: 1.37
DDRVDD: 2.77
3.3v: 3.37
5V: 5.17
12V: 12.16
SB: 1.52
nForce4: 1.6
5VSB: 5.17
Fan Speeds:
CPU: 1740rpm
NB: 4500rpm
AUX1: 2400rpm
OTES1&2 (idle): 0 rpm (OTES fans are controlled by temperature, and don't kick on until CPU temp hits 35C or so)

Tumbleweed36 · 04-04-2007, 07:07 PM

Hi,

Darn, the symptoms are classical. Either Memory (you have tested), the temperatures (reported good), or the voltages (reported good), or video card (you have changed it), so this is a tough one. I wonder if one of your sensors on the board could be faulty and not reporting actual temps or voltages? O.K., I know, just thinking out loud.

You do have a very good power supply in there, so that has me puzzled. I wonder a couple of things:

> Are we seeing a voltage drop while putting this baby under stress? Even the best power supplies sometimes will have drops from time to time. Some are very sporatic, but the thing that bothers me is you have a good one. To bad you don't have another supply to at least try.

> Could the temps be hotter than reported? (you tried the fan trick)

> If a sensor could be reporting a temp wrong, could your thermal solution not be working properly?

> Is your heatsink fan blowing down on the heatsink?

> What case cooling do you have?

> Have you ever tried to build it out of the case to eliminate a small and intermittent short of some kind. Using minimal components?

Tell you what, I will keep thinking on this one and get back with you in when I have had some time to think about it. This one is a toughie.

1 g0t 0wn3d · 04-04-2007, 07:10 PM

i dont really see a problem could your gpu be overheating or is it onboard?

**Doby** · 04-04-2007, 07:21 PM

My guess is a flaky psu seeing as you tried the house fan trick and have a pretty good understanding of temps and there affect so I am sure you got a handle on them.

The other reason for my "guess" is because the problem seems to have gotten a bit better after blowing out the psu with compressed air.
I do like that psu but any psu at any time can be flaky and randomly drop a rails voltage leading to a reboot or shutdown. Also seeing as this is a shutdown more that a reboot (and all other temps besides psu) are good suggests to me the psu is overheating and just cutting power. Why could be any number of things internal to the psu.

My suggestion is to try another name brand quality unit that is adaquate for the system.

**Doby** · 04-04-2007, 07:29 PM

Quote:

> Are we seeing a voltage drop while putting this baby under stress? Even the best power supplies sometimes will have drops from time to time. Some are very sporatic, but the thing that bothers me is you have a good one. To bad you don't have another supply to at least try

That is exactly my thinking its just that Tumble types so much faster than me

veri745 · 04-05-2007, 10:13 AM

Alright, for Tumbleweed:
I attached the Speedfan log file, which ends at my computer shutting down (didn't take very long that time, as you can see). There doesn't really seem to be anything overly suspicious there that I see. The only thing that changes at all is the 12V and my cpu temp, which appears to be running very cool right now (my case is open).

As far as my HSF condition, my Zalman 9500 is new in the last two months. I installed it with AS5, and the heatpipes are currently very cool to the touch. This shutdown issue started approximately a month after I installed the HSF.

I'm currently running this rig in a Lian Li aluminum case (the silver version though)
I have not yet tried to run it outside the case. I have tried unplugging my 200GB hard drive and taking out a stick of RAM.

For 'I got owned':
My 6800GT is not onboard. Under load, the maximum temp I've seen in around 70 C. The 'slowdown' threshhold for this video card is 127 C. The card overheating would not cause a complete shutdown, even if it were.
FYI, once upon a time I did have a shoddy power supply in this rig. When that one was going out (voltages fluctuating a lot), my video card would often ward me that it was not getting enough power and was 'slowing down'. This has not happened recently.

For Doby:

Sadly, I do not have another power supply handy, nor do I have a spare $100 to buy a new one randomly. I may see if one of my friends would trade me power supplies for day, but I'm really not convinced that it is the issue.

Thanks for all the responses guys; I appreciate it. Hopefully we can get this one worked out.

P.S. although I have reinstalled all of my drivers as a result of this problem, I might try to do an Windows repair today if I have time. Is this advisable?

**Doby** · 04-05-2007, 03:10 PM

Assuming you are from the US you could go to a store like bestbuy or staples,one that has a liberal return policy and purchase a antec trio 650w, try it and see if it solves the problem, if not return it for a refund. Just make sure you ask about returns.

Looking over your log the 12V does seem to fluctuate although not below the plus or minus 5% tolerance for a cpu rail if this was in a short period of time I don't like to see that. Lets get Tumbles take on this.

I don't think a windows repair at this point will solve anything because if windows or a driver was causing this the computer would normally reboot or bsod, just shutting down suggests hardware is overheating and with your nice temps the only thing I can think of would be the psu overheating

Tumbleweed36 · 04-05-2007, 05:36 PM

Hi,

I think that Doby has a good plan and that is to try one from someone that you can return it if the problem is not solved. The fluxuation in the +12 volt rail is quite significant although as mentioned, not out of the +/- 5% allowance. For all practical purposes, the facts you have given us are fine or within range. One other question, is the power you are putting in to this (from the wall plug) pretty steady in your area or is there quite a bit of voltage fluxuation?

veri745 · 04-05-2007, 06:08 PM

Hey guys. I acquired another power supply from a friend. It's an Antec SmartPower 450W, and it runs my friend's rig fine, which is more power hungry than mine (Athlon 4000+, GeForce 7600, 4 HDDs, 2 optical drives)

Same power off issue after running Europa Universalis III for about an hour.

This would lead be to believe it is NOT the power supply?

What kind of options does that leave for possible sources of this kind of problem?

Things I might try:
memtest my memory sticks individually, and for longer period.
build my rig out of case (ugghhhh)
repair my Windows installation
install Windows and a game on a separate HDD

Those last two are not my idea of a good time, but I'm totally stumped now...again...

*edit* I attached the Speedfan log again; It ends right before my system shut down. There appears to be about the same amount of fluxuation on the 12V lead, but the voltages are slightly higher (better or worse?)

**Doby** · 04-05-2007, 06:51 PM

Although I believe the 450w antec is on the lite side for your system and therefore not a totally conclusive test I did notice that the log shows more stable +12V readings, at least there not dipping below 12V. So I think there may be another problem.

Tumbleweed said this was going to be a tuffy

Definitely run memtest on the ram sticks individually, the program will not be effective if not. But still if it was ram related the computer would restart or bsod not shut down.

In your bios look around for something like system event and see if there is something turned on that would make the computer shutdown after some sort of event or system failure. I am grasping at straws I know.

Also make sure your cpu shut down temp threshold is not set to low in bios, I can't see how this would be possible because of your good cpu temps but I would just look at the cpu temp and threshold the bios is giving just in case.

I'll keep thinking but I dunno

**Doby** · 04-05-2007, 07:09 PM

Some other thoughts,,,,,,,,,,,

I would remove all unnecessary cards, usb devices,routers ect. Disconnect the front usb and speaker connections from the motherboard and use the rear ones for now and see if it makes any difference

Tumbleweed36 · 04-05-2007, 07:17 PM

Hi,

I also think that Doby has you on a good plan. Why not try a basic build without taking the Motherboard out of the case. If you try that and still get the shutdowns, then you do need to do an out of the case build. I think in your case, not removing it from the case at this time might be helpful to you.

veri745 · 04-06-2007, 12:17 AM

Possible good news?

I removed all peripherals, unplugged my front and rear accessory usb ports, switched back to my original power supply, reseated my video card and audio daughter card, and reseated my power/reset switches, power/HDD LED connections, and PC speaker.

I was able to run Europa Universalis III for a good 2.5-3 hours without any power-off's.

I'll keep my fingers crossed and continue testing it tomorrow.

**Doby** · 04-06-2007, 01:40 PM

Yea that is good news I think you have somewhat isolated the problem. It still has to be a short and my first guess would be the front usb ports and seeing you are a gamer you were using a joystick or something in them.

I would connect up the front usb ports only and use them while playing the game to see if you can trigger the shutdown.

If it does not do it I would add one thing at a time till you find the problem device, if not usb then my second guess would be some kind of faulty card

veri745 · 04-18-2007, 09:42 PM

Alright, I had a couple weeks there of no problems. Now I have a different problem but that's besides the point.
I have experienced one or two times when my comp. shut off unexpectedly since my last post, but they've been extremely few and far between, and never when playing a game, which is mainly when the problem occurred before.

I'm still hesitant to call this one 'resolved' because I have suspicions of some underlying problem somewhere that I eventually hope to fix during the upgrade process. I have way too many hardware problems it seems.

Cheers!

04-04-2007, 05:38 PM	#1 (permalink)
veri745 Registered User Join Date: Apr 2007 Posts: 8 OS: XP Pro My System	Random system shutdown I know there are several other posts with similar titles. I've read most of them, and was either not convinced my problem was the same, or have already tried the suggestions. I've tried other forums for this issue already, please review these links: Abit Forum OCZ Forum * Module size and kit size:-2 x 1gb Corsair XMS * CPU model and rated speed:-Athlon X2 4200+ * Overclocked CPU speed:-2200 Mhz - stock * Divider and FSB used:- * Motherboard and revision:-Abit Fatal1ty AN8 * Motherboard Bios rev:- * Video card:-BFG GeForce 6800GTOC * GPU Temps: 58 C idle; 70 C load * Power supply:- OCZ Modstream 520W * Voltages (tested using DMM): 12V rail: 12.01 idle; 11.87 load 5V rail: 5.11 3.3V rail: 3.37 * CPU and System Cooling:-Zalman 9500; 4x 80mm case fans * CPU and System temps:-cpu: 30 idle, 45 max load * HDDs: 80GB IDE (Windows); 160GB SATA (games, programs); 200GB IDE (storage) * Country:-US Problem: System power off Frequency: Random Repeatability: No, but occurs MUCH more frequently in 3D applications System will power back up immediately. No message from Windows saying computer was not shut down properly. No errors in system log. Problem occurs most often in games (Counter-Strike: Source, Europa Universalis III). Occasionally it has shut down during normal Windows use (internet, messaging). I have basically ruled out an overheating issue (please read my other forum links) I have tested my memory extensively using memtest86 for several hours. I have tested my processor for several hours using Prime95. According to the power supply calculator linked in some of the other threads, my system need 334W + 30% = 435 W, which my power supply should handle easily. I use ABIT uGuru to monitor system temperatures and voltages, and have also checked my PSU voltages using a DMM. Video card has been replaced with no change in symptoms (I tried an x850xt which I was RMAing my 6800). Let me know if you need any more information. I am not a novice, but this issue has me stumped.

04-04-2007, 06:00 PM	#2 (permalink)
Tumbleweed36 Asst Manager Hardware Join Date: May 2005 Location: USA Posts: 14,184 OS: XP Professional	Re: Random system shutdown Hi, Just for comparisons of voltages, temps, and fan speeds, why not try both Speedfan and SensorviewPro for other views of what is going on in there: Speedfan will tell you the temperatures and a whole lot more. http://www.sofotex.com/SpeedFan-download_L4655.html SensorView Pro http://www.stvsoft.com/download.php?id=122 __________________ ----------

04-04-2007, 06:54 PM	#3 (permalink)
veri745 Registered User Join Date: Apr 2007 Posts: 8 OS: XP Pro My System	Re: Random system shutdown k, i guess i'll humor you. These readings are almost EXACTLY what I get out of uGuru (infact, Speedfan uses uGuru to get it's readings). The HDD temps are nice though. Speedfan reports: Local: 44C Remote: 45C HD0: 34C HD1: 31C CPU: 24C SYS: 36C PWM: 35C Core: 28: Voltages: HTV: 1.18 VCORE: 1.37 DDRVDD: 2.77 3.3v: 3.37 5V: 5.17 12V: 12.16 SB: 1.52 nForce4: 1.6 5VSB: 5.17 Fan Speeds: CPU: 1740rpm NB: 4500rpm AUX1: 2400rpm OTES1&2 (idle): 0 rpm (OTES fans are controlled by temperature, and don't kick on until CPU temp hits 35C or so) Last edited by veri745 : 04-04-2007 at 07:00 PM.

04-04-2007, 07:07 PM	#4 (permalink)
Tumbleweed36 Asst Manager Hardware Join Date: May 2005 Location: USA Posts: 14,184 OS: XP Professional	Re: Random system shutdown Hi, Darn, the symptoms are classical. Either Memory (you have tested), the temperatures (reported good), or the voltages (reported good), or video card (you have changed it), so this is a tough one. I wonder if one of your sensors on the board could be faulty and not reporting actual temps or voltages? O.K., I know, just thinking out loud. You do have a very good power supply in there, so that has me puzzled. I wonder a couple of things: > Are we seeing a voltage drop while putting this baby under stress? Even the best power supplies sometimes will have drops from time to time. Some are very sporatic, but the thing that bothers me is you have a good one. To bad you don't have another supply to at least try. > Could the temps be hotter than reported? (you tried the fan trick) > If a sensor could be reporting a temp wrong, could your thermal solution not be working properly? > Is your heatsink fan blowing down on the heatsink? > What case cooling do you have? > Have you ever tried to build it out of the case to eliminate a small and intermittent short of some kind. Using minimal components? Tell you what, I will keep thinking on this one and get back with you in when I have had some time to think about it. This one is a toughie. __________________ ----------

04-04-2007, 07:10 PM	#5 (permalink)
1 g0t 0wn3d Hardware Team Join Date: Jan 2007 Location: Sedona, az Posts: 6,276 OS: Vista basic 32bit My System	Re: Random system shutdown i dont really see a problem could your gpu be overheating or is it onboard? __________________ Bench Test \| PC Wizard \| System specs \| PSU info \| PSU calculator \| Memtest86+ \| Speedfan

04-04-2007, 07:21 PM	#6 (permalink)
Doby Moderator Hardware Team Join Date: Jan 2006 Location: Central PA (USA) Posts: 7,150 OS: XP sp2/ Vista	Re: Random system shutdown My guess is a flaky psu seeing as you tried the house fan trick and have a pretty good understanding of temps and there affect so I am sure you got a handle on them. The other reason for my "guess" is because the problem seems to have gotten a bit better after blowing out the psu with compressed air. I do like that psu but any psu at any time can be flaky and randomly drop a rails voltage leading to a reboot or shutdown. Also seeing as this is a shutdown more that a reboot (and all other temps besides psu) are good suggests to me the psu is overheating and just cutting power. Why could be any number of things internal to the psu. My suggestion is to try another name brand quality unit that is adaquate for the system. __________________ Would you like to help TSF fight cancer and other diseases??? JOIN TEAM TSF

04-05-2007, 03:10 PM	#9 (permalink)
Doby Moderator Hardware Team Join Date: Jan 2006 Location: Central PA (USA) Posts: 7,150 OS: XP sp2/ Vista	Re: Random system shutdown Assuming you are from the US you could go to a store like bestbuy or staples,one that has a liberal return policy and purchase a antec trio 650w, try it and see if it solves the problem, if not return it for a refund. Just make sure you ask about returns. Looking over your log the 12V does seem to fluctuate although not below the plus or minus 5% tolerance for a cpu rail if this was in a short period of time I don't like to see that. Lets get Tumbles take on this. I don't think a windows repair at this point will solve anything because if windows or a driver was causing this the computer would normally reboot or bsod, just shutting down suggests hardware is overheating and with your nice temps the only thing I can think of would be the psu overheating __________________ Would you like to help TSF fight cancer and other diseases??? JOIN TEAM TSF

04-05-2007, 05:36 PM	#10 (permalink)
Tumbleweed36 Asst Manager Hardware Join Date: May 2005 Location: USA Posts: 14,184 OS: XP Professional	Re: Random system shutdown Hi, I think that Doby has a good plan and that is to try one from someone that you can return it if the problem is not solved. The fluxuation in the +12 volt rail is quite significant although as mentioned, not out of the +/- 5% allowance. For all practical purposes, the facts you have given us are fine or within range. One other question, is the power you are putting in to this (from the wall plug) pretty steady in your area or is there quite a bit of voltage fluxuation? __________________ ----------

04-05-2007, 06:51 PM	#12 (permalink)
Doby Moderator Hardware Team Join Date: Jan 2006 Location: Central PA (USA) Posts: 7,150 OS: XP sp2/ Vista	Re: Random system shutdown Although I believe the 450w antec is on the lite side for your system and therefore not a totally conclusive test I did notice that the log shows more stable +12V readings, at least there not dipping below 12V. So I think there may be another problem. Tumbleweed said this was going to be a tuffy Definitely run memtest on the ram sticks individually, the program will not be effective if not. But still if it was ram related the computer would restart or bsod not shut down. In your bios look around for something like system event and see if there is something turned on that would make the computer shutdown after some sort of event or system failure. I am grasping at straws I know. Also make sure your cpu shut down temp threshold is not set to low in bios, I can't see how this would be possible because of your good cpu temps but I would just look at the cpu temp and threshold the bios is giving just in case. I'll keep thinking but I dunno __________________ Would you like to help TSF fight cancer and other diseases??? JOIN TEAM TSF

04-05-2007, 07:09 PM	#13 (permalink)
Doby Moderator Hardware Team Join Date: Jan 2006 Location: Central PA (USA) Posts: 7,150 OS: XP sp2/ Vista	Re: Random system shutdown Some other thoughts,,,,,,,,,,, I would remove all unnecessary cards, usb devices,routers ect. Disconnect the front usb and speaker connections from the motherboard and use the rear ones for now and see if it makes any difference __________________ Would you like to help TSF fight cancer and other diseases??? JOIN TEAM TSF

04-05-2007, 07:17 PM	#14 (permalink)
Tumbleweed36 Asst Manager Hardware Join Date: May 2005 Location: USA Posts: 14,184 OS: XP Professional	Re: Random system shutdown Hi, I also think that Doby has you on a good plan. Why not try a basic build without taking the Motherboard out of the case. If you try that and still get the shutdowns, then you do need to do an out of the case build. I think in your case, not removing it from the case at this time might be helpful to you. __________________ ----------

04-06-2007, 12:17 AM	#15 (permalink)
veri745 Registered User Join Date: Apr 2007 Posts: 8 OS: XP Pro My System	Re: Random system shutdown Possible good news? I removed all peripherals, unplugged my front and rear accessory usb ports, switched back to my original power supply, reseated my video card and audio daughter card, and reseated my power/reset switches, power/HDD LED connections, and PC speaker. I was able to run Europa Universalis III for a good 2.5-3 hours without any power-off's. I'll keep my fingers crossed and continue testing it tomorrow.

04-06-2007, 01:40 PM	#16 (permalink)
Doby Moderator Hardware Team Join Date: Jan 2006 Location: Central PA (USA) Posts: 7,150 OS: XP sp2/ Vista	Re: Random system shutdown Yea that is good news I think you have somewhat isolated the problem. It still has to be a short and my first guess would be the front usb ports and seeing you are a gamer you were using a joystick or something in them. I would connect up the front usb ports only and use them while playing the game to see if you can trigger the shutdown. If it does not do it I would add one thing at a time till you find the problem device, if not usb then my second guess would be some kind of faulty card __________________ Would you like to help TSF fight cancer and other diseases??? JOIN TEAM TSF

04-18-2007, 09:42 PM	#17 (permalink)
veri745 Registered User Join Date: Apr 2007 Posts: 8 OS: XP Pro My System	Re: Random system shutdown Alright, I had a couple weeks there of no problems. Now I have a different problem but that's besides the point. I have experienced one or two times when my comp. shut off unexpectedly since my last post, but they've been extremely few and far between, and never when playing a game, which is mainly when the problem occurred before. I'm still hesitant to call this one 'resolved' because I have suspicions of some underlying problem somewhere that I eventually hope to fix during the upgrade process. I have way too many hardware problems it seems. Cheers!