Installed 275 GTX, now PC reboots after about 30 min.

[klizzey]

Hello, i'm a bit worried about my computer randomly shutting off on me after just installing a 275 GTX. I posted here yesterday with questions about my PSU and after some positive responses decided to go with buying the card.

I currently have The Apevia 750 Watt PSU and a BFG Tech 275 GTX OC
( http://www.apevia.com/ProductsInfo.asp?KEY=ATX-LCD750W )

The computer turns on fine, the card runs fine, it's running games fine, but so fine it has shut down on me twice mid gameplay. It doesn't reboot, it just shuts off. Is this a PSU problem? Can it not handle powering the card under stress or is it the card itself the issue here? I thought it may be an overheating issue, but the GPU never passed 70 C under load, the CPU runs 50-70, and HD runs fairly low. In order to fit the GPU in, I had to move my HD from one slot to another, not sure if this had anything to do with it. It's also OC'd, could this perhaps be causing it, should I tune it down some? Fairly concerned with it shutting off at random and would like some insight as to what you guys think it is. Thanks.

 

[Crashman]

Hello, i'm a bit worried about my computer randomly shutting off on me after just installing a 275 GTX. I posted here yesterday with questions about my PSU and after some positive responses decided to go with buying the card.

I currently have The Apevia 750 Watt PSU and a BFG Tech 275 GTX OC
( http://www.apevia.com/ProductsInfo.asp?KEY=ATX-LCD750W )

The computer turns on fine, the card runs fine, it's running games fine, but so fine it has shut down on me twice mid gameplay. It doesn't reboot, it just shuts off. Is this a PSU problem? Can it not handle powering the card under stress or is it the card itself the issue here? I thought it may be an overheating issue, but the GPU never passed 70 C under load, the CPU runs 50-70, and HD runs fairly low. In order to fit the GPU in, I had to move my HD from one slot to another, not sure if this had anything to do with it. It's also OC'd, could this perhaps be causing it, should I tune it down some? Fairly concerned with it shutting off at random and would like some insight as to what you guys think it is. Thanks.

It could be a bad power supply, I've never used Apevia so I can't tell you much about them.

 

[Twoboxer]

Rapid shutdowns are most often caused by heat or power, or an unstable OC exposed by load.

You've checked heat and its apparently not the issue.

Drop your OC back to stock - that's easy to check. If it runs stable you can take it back up some *later*.

If it still shuts down, buy a Corsair psu.

 

[klizzey]

I decided to do some testing with it and here is what I got.

It ran without a problem for over an hour at the desktop, browsing, listening to music, watching videos, no problem. I even did a quick stress test at which it was processing over 2000 FPS. It ran for 15 minutes without a problem.

After all checked out, I proceeded to log back onto Champions Online and flew around the city a bit and then hovered looking toward the most graphic intensive part of the main city. After 30 minutes of doing so, I decided to get into some combat; after not even 5 minutes of fighting stuff, the computer just shut off. One thing I did realize was that my Quad cores were running very high. Core 0 was running in the 90's along with core 1, while core 2 and 3 hovered in the 80's. This to my knowledge is very high. Could it be that during the combat, the CPU output increases, it gets even hotter, overheats and turns the system off to prevent further damage? The GPU on the other hand never went over 65C.

As far as the PSU goes, it's working fine from what I can tell. I checked the LCD display and the system was never using more then 300W. Only thing out of the ordinary was that the -12V was showing 16. Where the 12v1 and 12v2 were running at 12. Not sure what the -12V is about, but could that be an issue?

 

[rand_79]

get a better cpu cooler., try for now running without being overclocked to see if the problem happens.

also make sure all your accessories arent all plugged into one of the 12 volt rails..

you have 2 30amp 12volt rails maybe one has a high load and one has a low load?

 

[Twoboxer]

Your cpu temps are too high. This probably is the cause, but it needs to be fixed regardless.

Could be case air flow, poor installation of or a weak cpu heat sink/cooler, or the result of OCing with a stock cooler. Take your OC's off and run stock until you get things under control.

A properly installed, appropriate heat sink is step one. But remember, they only get heat off the cpu to the extent the case gets hot air out.

Running the PC on its side with the cover off can help cool things and *may* give you a chance to test whether heat is the cause of your shutdowns. But your temps are too high, regardless.

 

[klizzey]

Cleaned the heatsinks/fans, tighened up the loose cables, unhooked heatsink and put it back on tight, but now my computer shuts off right as the screen to enter bios passes by. I hear the card turn on, I see the CPU's heatsink/fan running, the PSU is running, but then it just shuts off like it doesn't even have enough juice to power everything up.

 

[klizzey]

Fixed the above issue with it turning off during boot up. Back to original issues.

 

[ohiou_grad_06]

What is airlow like inside the case? What I personally have is a pci slot cooling fan directly under my video card to pull heat outside the case. The cpu gets very hot after a certain period but it only started after the new card was inserted, however you seem to have plenty of juice.

It sounds like maybe there is not great cooling on the video card and that it may be exhausting it's heat back into the case thereby making the cpu get hot which shuts off your pc t save the cpu from frying. A way to test is take the side off the case, get a box fan or desk fan, turn it on high and run your system with the fan running directly into the case. If the shutdowns cease, you have your culprit. I've also seen that if a power supply puts off too much heat it can cause the original symptom, and with that new card that power supply is probably working much harder than before. These are the places I would start.

 

[klizzey]

Next plan of action at bottom for those not caring to read my shenanigans.

The airflow is pretty good. my PSU vents out the hot air rather quickly out the top back, and the GPU vents it out the bottom back of the case. The side panel has a nice large fan that also vents out hot air. Ontop of that, I have a good table fan blowing through the mid-back vents ( I have a rather large case so the air being vented out the top and bottom stays clear of the fan blowing cool air in ).

Well, some of the things i've done so far are :
- Refixed the CPU heatsink and tightened it.
- Cleaned the case/fans, etc.
- Did a complete display driver wipe and installed the latest Nvidia ones.
- Stress tested CPU, GPU, RAM, HD, etc. using the tools provided by Nvidia.
To touch up on the above, the tools they provide arn't exactly very taxing, as it hardly pushed anything even close to max, but with GPU running at 60C, Cores at 70-80C, there was no problem and the computer didn't even flinch.
- I ran Champions Online in it's own safe mode, which lowers all settings to default and resolution to 600x400, with this I was able to run SpeedFan next to the window and I was actually able to engage in combat and fight stuff. Now although this wasn't taxing on the GPU, so I can't completely rule out GPU not getting enough juice under heavy loads, it pushed my CPU the same way the max settings do. Speedfan clocked me at 90C (Speedfan users claim that it's CPU readings are normally about 15C too hot) on core 0 and mid to upper 80's on the other 3 cores. I have an Intel 6600 Quad. I think the culprit is indeed the CPU. The reason I was able to stay on the game without an issue the first time while flying up high was because the CPU wasn't being heavily taxed. Soon as I drop down and the flashy effects, mobs, other people, etc. start coming up, it begins to heat up and continues to do so untill it gets so hot that the CPU has no choice but to shut the computer off to prevent damage. Another thing pointing me this way is that I was running FarCry2 without anissue, but 20 minutes or so in after engaging some baddies, the computer turned off. The GPU was heavily taxed right off the bat, from start to end, but the CPU slowly heated up and continued to do so untill it got so hot it had to turn off.

My next plan of action is to go out tomorrow and purchase some Arctic 5 and apply a new coat. It's been about 2 years now since I affixed the heatsink to the CPU and added their thermal paste so I think that paste on there is pretty much gone, especially now after fiddling with it. This should help tremendously with dissapating the heat from the CPU to the heatsink and should hopefully solve my problem.

If anyone else has an idea, please toss it in the hat, i'll post back tomorrow after applying new thermal paste with results.

 

[Twoboxer]

Well, we dropped a lot of ideas for you and you went your own way, which is fine.

Last thing I'd say is side fans blowing out are not generally a good idea.

Good luck!

 

[ohiou_grad_06]

Yeah, I think your fan setup is backward actually man. I would personally turn all fans to blow inward, and then have only the rear exhaust fans blowing out of the case. Like maybe one under your power supply blowing out, and a pci slot cooler fan blowing out. On my rig, I've got a 9600gt and an Athlon x2 5200+.

My cpu and gpu are now both idling at about 40 degrees celcius. Keep in mind I've been lazy and haven't cleaned them out in about 6 months, so they are due. But if I keep them cleaned out, the gpu usually runs about 38, then cpu in the mid 30s. All my cooling set is is one 250 mm side fan pulling air in, an 80mm fan under the power supply pulling air out, and a pci slot cooling fan under the video card pulling air out.

What you want is to create a vacuum of cooler air by forcing cool air in and have 1 or 2 exhuast fans pull air out the back side of the case. Flip your fans around and see if that doesn't solve the problem.

 

[redwine01]

i had this problem far ago when i bought my 2nd pc. i was wondering what's happening whenever i m inside game it was restarting after a while. So i decided to check what happening. what i found out is that when i started my pc the cpu fans were not spinning. i thought wat the hell then i saw that those ignorant pc people forgot to plug my cpu fan to power it on. i pluged it and then my pc runs just fine. from the explanation u gave i think it is because of the cpu running too hot and for safety reasons ur pc is restarting because of too much heat. what u need to do is:

check if the heatsink/fan is installed properly
check if there is too much paste. the paste should be thin
set the bios in default
set the gfx clocks to normal (if u overclocked it manualy if it was preoverclocked card then u dont have to do this)
install the chipset driver and gfx driver properly. incorrectly installed driver will cause random shutdowns.

i hope this will solve the problem. if not then try to update ur bios to the latest version.

 

[klizzey]

Well, so far i've done everything suggested in the thread short of buying a new PSU because I doubt it's a PSU problem, but that may be my next option.

I can get into games and have push the computer to it's peak and it not reboot. One example being Champions Online, I was able to run around, switch zones, do missions, etc. without it shutting off, but 30 minutes in, it out of nowhere shut off while just flying.

I've replaced the arctic 5 on my CPU and under Crysis loads it never went above 65C on Speedfan, which more accurately would place me at 50-55C under max loads.

I was able to run around the final level of Crysis:Warhead, with all the bugs flying around for about 5 minutes (Probably the most GPU/CPU intensive part of the game imo), but soon as a cutscene rolled in and it got about 20 seconds into it, the computer just shut off. Same spot in the same cutscene on both attempts. If it was my PSU, it would shut off soon as I began running around and shooting on Crysis, as that would really force the CPU/GPU to work, but it doesn't, instead it does it at the same exact spot in the same cuttscene.

I've underclocked the GPU, tried it factory set, same with CPU, same deal, turns off out of the blue. The reading on my PSU's screen never shows the wattage going over 300, and I have a 750W PSU. I don't have many things operating inside my computer either, so it's not like it's being overloaded by appliances. The GPU temp never climbs over 65 under loads either.

I honestly have no idea what the issue is now that my CPU is running cool, GPU is running cool, and everything is working as it should, untill it shuts off that is.

 

[alikum]

You could have already damaged your CPU. This is just a wild guess but my friend had a similar situation where one of his cores died (well maybe half dead). This could explain why.

 

[klizzey]

All 4 of my cores are operating, 1-2 handling most stress, 3-4 kicking in to about half.

I may try to remove some older memory in a second here and see if it's bad memory causing it to die randomly. Doesn't appear like the CPU was ever in danger and it's running even cooler now.

Also, does anyone happen to know the GTX 275's shut off point? Could it be 70C or somewhere around there? I don't think it would be so low, but i'm beginning to wonder now. Tried to dig up the info, but couldn't find it.

 

[Bluescreendeath]

GTX200s have a max temperature of 90'C-100'C before failure. It's designed to reach around 80'C at load.

70'C should be not much of a problem for it.

 

[redwine01]

yes 70'c shouldnt be any problem with GTX275 my friend's OC 275 goes up to 88'C so its nothing comparing to urs. 1 more suggesion i can give u is in bios there is a option called "Shutdown Temperature" it is under power management or PC health something like this. Check it out make it 75'C and u r good to go. another thing u may want to do is reseating the ram. Even if these doesnt help check for viruses in ur system. And at last test ur pc with a low end graphics card that doesnt require much power. if it doesnt reboot then definately ur psu cant provide enough to ur GTX275 but it should be providing more than enough but mayb it is faulty. if these doesnt help then i m not sure whats wrong with the system.

 

[klizzey]

Thanks for the suggestions. I'm about to reseat the ram, and perhaps even switch the GPU into a different slot.

This problem is really starting to baffle me as I just ran 3dmark 06 and it passed through all the tests without as much as a studder. It finished all of the tests and the computer was still on. Right afterwards, I ran Crysis, about 5 minutes in, the computer turned off. The 3dmark 06 stressed the system more then Crysis did, so I really don't get why it's turning off.

If the ram doesn't fix it, I'm going to go ahead and do a complete wipe of drivers and reinstall from there, because that's the only thing I can think of that would cause issues during games only.

 

[Twoboxer]

Hello, i'm a bit worried about my computer randomly shutting off on me after just installing a 275 GTX. . . . It doesn't reboot, it just shuts off. Is this a PSU problem? . . .

Just a reminder.

 

[overshocks]

You never mentioned your hardware set up, what heatsink, how much overclock..case..etc..
You guys didn't point this out already? Shame, the blind helping out another blind wont work, will it? lol..
you can clearly see it's not a PSU problem..because it will shutdown at load, not when hes at 30 mins or whatever time.
it's his cpu being so hot, 100C core temp is max for his cpu, what you use to monitor temps?
dude you need to provide more info. isnt there a thread that teach you how to ask a question properly?

 

[overshocks]

GTX200s have a max temperature of 90'C-100'C before failure. It's designed to reach around 80'C at load.

70'C should be not much of a problem for it.

Actually the failure point is 110C , which is thermal shutdown point. 90-100 is close, but still runable.

 

[redwine01]

he already said that his cpu doesnt go more than 70'c. and even if u assume he is using stock cooling system it shouldnt atleast restart the system unless the cpu is really much overheated. And i had experience running my cpu at 65'c at load once when i had a dusty heatsink i remove the dust and it goes back to 36'c with the stock cooling system and a little overclock. my system didnt restart or shutdown at 65 i just noticed it in speedfan settings. a stock cooler may be not as gr8 as aftermarket one but atleast not that bad to have a system restart or burn down. so i highly suspect that the heatsink is not properly installed or dusty too much or the fan is not running. as i have said in my earlier post to check these issues. and OP replied he did checked everything but nothing happened. i still think its a heat problem becoz i had the experience of shutting down pc when it got overheated from fan not running. Lets see what OP replies after reseating ram and cleaning all the driver and installing newer one.

 

[klizzey]

Thanks for the help guys, with the suggestion I got, I was able to rule stuff out one by one and find the culprit.

In the end it came back to what people in the beginning were suggesting. PSU problem. The PS can easily operate everything in the system, but once it needed everything to run at peak output, it just didn't have enough juice to keep everything powered. Although I had set the GPU back to stock settings, the stock settings were only 15-20 Mhz lower and I hadn't even thought of lowering it further.

Anyway, long story short, taking the GPU down from 645Mhz Core bus and 1180Mhz Memory bus to 615Mhz Core and 1100Mhz Memory did the trick. It keeps the GPU from needed more power then I have, but still lets me play my games at max settings. This is a temporary fix of course untill my new PSU gets in.

In the end, thanks to this problem, I ended up fixing up my CPU temps to stay below 60C and I got my card working in the end. Thanks a lot for the help!

 

[klizzey]

And Overshocks, we had already ruled out CPU overheating because I had cleaned it and fixed it up to where it was staying below 60. With that out of the way, there was no reason to post what heatsink/fan I have or any further information about it.

And isn't there a thread out there that teaches you how not to come off as so dull?