PC restarting randomly (video card?)

K

Kiri

Limp Gawd
Joined
Apr 10, 2008
Messages
176
Hi, I'm posting this here because I believe the problem could be graphics card related.

For about a week now, my desktop PC has been randomly restarting (no BSOD or anything).
I have turned off the automatic restart setting in system settings, but it is still just restarting.
There also does not seem to be a minidump directory, so I can't check the logs.

I have made a couple of changes to my PC recently which could be the cause.
First, my PSU died about two weeks ago. I actually came home to the PC constantly turning itself off and attempting to restart every few seconds (it wasnt even booting up), and making a 'crackling' kind of sound.
Anyway, I got a new 630W PSU. (My old one was 650W).

Everything seemed to run fine after the new PSU went in.

Second, I recently bought a second monitor for a multi-monitor setup. However, since I also have my TV hooked up to my PC, I needed to buy a 2nd graphics card.
I think my problems started a couple of days after installing the new monitor and graphics card.
I can't seem to notice any pattern to when it restarts. It is not when the CPU or GPU is under load. I played games for a few hours with no problem.
But it has seemed to restart more frequently when viewing something on youtube (but not always).

Anyway, here are the system specs:
Windows 7
- Kuroutoshikou 630W PSU
- Asus P6T MB
- ATI HD 4850 (this is the card I have had for a couple of years and has been fine) *Both my monitors are plugged into this card.
- ATI HD 5450 by Gigabyte (this is the new card which I suspect may be causing the problem) *Only my TV is plugged into this card.


The HD5450 is installed in the 2nd SLI slot on the MB (should I try putting it in a regular PCI slot instead?)


I have been running HWMonitor also, but the temperatures don't seem to get particularly high.


I have actually just removed the HD5450 card to see if the problems persists, so I will post back with that too.


In the meantime, does anyone have any ideas about how I can go about diagnosing this problem?
Thanks!
 
Usually random restarts with no blue screens is cpu overheating. You might want to replace your cpu heatsink or at least make sure it is clean and reapply the thermal paste.
 
Your motherboard could also have been damaged by the repeated reboots with the old power supply. If it was that bad off, there's no telling what kind of voltage spikes it was spitting out.
 
sounds like an overheating issue to me. what are your temps exactly?
 
i think you may have a dud psu there... first of all why did you buy some no name brand psu? thats the last thing you want to skimp on
 
dinlee23 said:
i think you may have a dud psu there... first of all why did you buy some no name brand psu? thats the last thing you want to skimp on
Click to expand...

This, a bad PSU will cause this issue and since it was recently replaced that would be my first guess(I've also never heard of that brand).
 
Try a quality PSU, just had a very similar machine in my office this week, threw in a Thermaltake and off it went.
 
I threw together a system with a crap Logisys power supply, and it randomly rebooted. I monitored everything, it all seemed good. Checked the windows log and the only thing mentioned was power interruppted unexpectedly. Swapped PSU's and the problem was gone. Bad part was that some damage was done to the motherboard by this point and it only lasted a couple of weeks, before it stopped POSTing. Point is get a new psu before it does more damage.
 
I had restarts and thought my new PSU just crapped out on me, turns out the 24pin was loose... Did you check to make sure everything is plugged in nice and tight?
 
Man, that could be anything. Did you check the alternator?
 
caniba said:
Usually random restarts with no blue screens is cpu overheating. You might want to replace your cpu heatsink or at least make sure it is clean and reapply the thermal paste.
Click to expand...

THIS.

I actually had this problem when I installed my three new GTX670 today. I took out a couple of fans that were mainly to cool the heatpipes on my old GTX480s and messed up daisy-chaining all the power connectors back together. I was marveling at how quiet my computer was with the new cards but dealing with restarts and very slow performance. I thought it was some kind of incompatibility between my mobo and the new cards and spent two hours troubleshooting it. When I brushed up against the CPU heatsink swapping the cards in and out to test them individually and it was scorching hot, I finally made the connection and felt like a complete retard.
 
I used to get random shutdowns just because the CMOS battery was dead. But in that case you'd see your BIOS reset...
 
Thanks for the suggestions everyone. I have a few reasons why I went with this PSU.
1) I live in Japan, so it is cheaper to buy a local product than one that is imported from the US or wherever
2) This brand seems to be very well rated here, and has won awards, so I'm guessing it's not too bad. You can see the spec page here: (in Japanese) http://www.kuroutoshikou.com/modules/display/?iid=1547
3) My previous PSU which died and started crackling and sparking was a Corsair TX650W, which is supposedly a high quality unit and is not cheap (especially here).


By the way, here is a screenshot of my HWMonitor readings. As far as I can tell everything looks ok.
http://i.imgur.com/w2eYR.png
 
I will re-check all the cables and make sure they are firmly inserted.
Then I'll check the CPU heatsink situation as suggested.

If its neither of those, then I'll consider that PSU may be the problem
My corsair is actually still under warranty... if I could just find the receipt.. (><) lol


But anyway, you don't think it could be something to do with the new graphics card?
Since I took it out last night, I haven't had the problem again. (still too early to draw any conclusion from that though).
 
Oh, I just noticed that HWMonitor has reported my CPUTIN at a max of 92C, and AUXTIN max 510C!!
Should I be worried about that? lol
Surely 500ºC is not possible...
 
Kiri said:
Thanks for the suggestions everyone. I have a few reasons why I went with this PSU.
1) I live in Japan, so it is cheaper to buy a local product than one that is imported from the US or wherever
2) This brand seems to be very well rated here, and has won awards, so I'm guessing it's not too bad. You can see the spec page here: (in Japanese) http://www.kuroutoshikou.com/modules/display/?iid=1547
3) My previous PSU which died and started crackling and sparking was a Corsair TX650W, which is supposedly a high quality unit and is not cheap (especially here).


By the way, here is a screenshot of my HWMonitor readings. As far as I can tell everything looks ok.
http://i.imgur.com/w2eYR.png
Click to expand...

i saw your cpuz screen shot and are those temps on idle? if so 50c is quite high... and yeah 92c is approaching TJmax, thats when the cpu throttles down or restarts the whole system to avoid failure, which you dont want happening...

also saw your psu specs and it seems okay if what it said about the unit is true. from past experience my pc restarts randomly from a bad CPU OC so you might want to check that if you overclock, and even then the restarts are definitely directly related to your temps
 
Yeah, I can see quite a bit of dust caked up in the heatsink, so I'm going to remove it and give it a good clean. I guess I will have to buy some thermal paste to reapply it too.

I still haven't had a restart since I took out the second video card though. Which is making me wonder whether the 2nd card is either making my CPU run hotter somehow, or there is some other kind of incompatibility? Anyway, I guess I'll clean the CPU heatsink first and then put the card back in and see how it goes.
 
Restarts could also mean loose monitor or TV connection to the video card. This happened to me with my new 680. First I thought the card or the PCI-E slot was defective but turned out that the DP cable I used to connect my video card to the monitor was gone so the system restarted randomly during a slight jerk of the case or a heavy gaming session when the fans on the video card spins vigorously.... Could also be loose card on the slot although that should not be a problem for your 5450 as I presume the card is not heavy but its better to double check if there is any slack. Check for loose contacts of any type from the power supply to other components in your system and do a general tightness check to see whether all connectors are well inside their pin holes.
 
dremic said:
sounds like an overheating issue to me. what are your temps exactly?
Click to expand...

Yes it does.

dinlee23 said:
i think you may have a dud psu there... first of all why did you buy some no name brand psu? thats the last thing you want to skimp on
Click to expand...

He obviously didn't know better. But that might not be the actual problem.

Ebernanut said:
This, a bad PSU will cause this issue and since it was recently replaced that would be my first guess(I've also never heard of that brand).
Click to expand...

It certainly could cause this issue, but we need more infromation. No one asked if this only happened while gaming or what. If it only happens when gaming then the PSU could easily be the issue. If it happens at the Windows desktop, it's less likely.

Sodapopjones said:
I had restarts and thought my new PSU just crapped out on me, turns out the 24pin was loose... Did you check to make sure everything is plugged in nice and tight?
Click to expand...

Again possible.

Koween said:
I used to get random shutdowns just because the CMOS battery was dead. But in that case you'd see your BIOS reset...
Click to expand...

I seriously doubt that. I've seen CMOS batteries completely die and the machines ran fine until you power cycled them. Then you'd have to setup all CMOS information again.
 
Dan_D said:
...
I seriously doubt that. I've seen CMOS batteries completely die and the machines ran fine until you power cycled them. Then you'd have to setup all CMOS information again.
Click to expand...

Might be a certain configuration or something else, but that was the issue. It would shutdown randomly, once a week or sometimes once in an hour. After that it wouldn't start - the fans would spin only while I kept the power button. After a few tries it'd start and POST screen would say that the BIOS has been reset and I had to reapply overclocks again. I think the clock also reset each time...
Anyway, I wrestled with that issue for about 6 months. It didn't look that the battery could be a problem, at least when the power was on. I checked the voltage with a voltmeter and it was below 2V. Replaced it with a new battery and had no problems ever since.
The system was an overclocked Q6600 @ 3.4GHz. / EVGA 780i.
It could be that this was only a part of the issue, but I checked everything else I could come up with - downclocked, checked for loose connections, checked for dead hardware, disconnected everything but the essential hw etc.
 
Koween said:
Might be a certain configuration or something else, but that was the issue. It would shutdown randomly, once a week or sometimes once in an hour. After that it wouldn't start - the fans would spin only while I kept the power button. After a few tries it'd start and POST screen would say that the BIOS has been reset and I had to reapply overclocks again. I think the clock also reset each time...
Anyway, I wrestled with that issue for about 6 months. It didn't look that the battery could be a problem, at least when the power was on. I checked the voltage with a voltmeter and it was below 2V. Replaced it with a new battery and had no problems ever since.
The system was an overclocked Q6600 @ 3.4GHz. / EVGA 780i.
It could be that this was only a part of the issue, but I checked everything else I could come up with - downclocked, checked for loose connections, checked for dead hardware, disconnected everything but the essential hw etc.
Click to expand...

I found the problem. Now it makes sense. If any boards could defy logic, physics and do something almost technically impossible, it's one of those.
 
OK, I'm going to attempt to clean the heatsink and reseat it tonight. Any tips on that btw? I've tried blowing air through, but it hasn't had much effect. I'll have to get a brush or something I guess...

But besides that, its now been 3 days since I took out the second video card, and haven't had a reboot since. So I'm really thinking that that is the problem. But what should I do about it? Do I have to get a different video card? I don't really understand what can be the problem here. The 2nd card is not getting very hot, and it is only being used to drive my TV. I'm not attempting to play games on it or anything..
 
Kiri said:
OK, I'm going to attempt to clean the heatsink and reseat it tonight. Any tips on that btw? I've tried blowing air through, but it hasn't had much effect. I'll have to get a brush or something I guess...

But besides that, its now been 3 days since I took out the second video card, and haven't had a reboot since. So I'm really thinking that that is the problem. But what should I do about it? Do I have to get a different video card? I don't really understand what can be the problem here. The 2nd card is not getting very hot, and it is only being used to drive my TV. I'm not attempting to play games on it or anything..
Click to expand...

the simplest thing you can do without actually take your computer apart, is run memtest. hell, even windows 7 has a built-in memory diagnostic.
 
Well I wish I hadn't decided to take off the heatsink and clean it now.
I got it nice and clean, but I busted one of the fastener things when I tried to remount it.
Now the computer shuts down as soon as it loads the OS.
I guess I'll have to go and buy a new heatsink/fan tomorrow morning :(
 
You must log in or register to reply here.
Back
Top