Overheating issues

[user7092]

Hello all. I'm a bit new on posting about computer issues on a forum, so please feel free to ask for any additional information, if needed.

Here's the story. Last May, I built my own computer and it has worked absolutely great up until about a month and a half ago. The weather is getting warmer and I can understand that it may be overheating since I don't have the best air conditioning in this place, but it's by no means THAT hot in the computer room.

In any case, what happens is, whenever I play a game for a bit, or sometimes even a while, my computer will just shut down. Sometimes it only takes about 10 minutes of gameplay. Other times it takes about an hour. I've spent the last month and a half looking up solutions to this issue and I can't find anything that works.

I thought maybe it wasn't overheating, so I've tried other strategies, but each one failed. I've made sure it's overheating now, because when I put a giant box fan up against my case, it will run for hours without shutting down. I've tried some tips to help prevent overheating, but they aren't working, and I'm at a brick wall now. I don't know what else I can do, because I don't know WHAT is overheating. Here are some of the things that I've tried:

Re-applying thermal paste to CPU heatsink
Re-applying thermal paste to GPU heatsink
Adding a 120mm fan on the side of the fan to blow air into the case, mostly in the GPU/southbridge heatsink area.
Underclocking my graphics card
Upgrading graphics card driver
Blowing out all the dust from the PSU, case, case fans, GPU fan, CPU fan, etc.
Increasing GPU fan speed with RivaTuner

And a few other things I can't think of at the moment.

I am using Everest Ultimate to monitor my temps and they all seem fine for what the averages are. Heck, I've actually managed to get them much cooler in the last month or so. My graphics card used to run at 74C when playing games, but now it runs about 55-65C (just depends on what I'm doing in-game).

I will post some of my hardware and the temps I am getting. I don't know if there's any more info I should provide, but please let me know if there is. Here is what I have:

GPU: EVGA nVidia GeForce 260 GTX
CPU: QuadCore Intel Core 2 Quad Q9400, 2.67 GHz
Mobo: EVGA nVidia nForce 790i SLI FTW
PSU: OCZ GameXStream 700W
Case: NZXT Zero 2 (in case air flow is to be discussed)

My temps, under load, according to Everest, are:
Motherboard 40-42C
CPU 54-57C
CPU Core 1: 52-57C
CPU Core 2: 48-55C
CPU Core 3: 48-55C
CPU Core 4: 51-57C
SPP: 50-52C
MCP:60-66C
GPU: 52-62C (rarely will it get 65, but I've seen it)
GPU Memory: 48-54C
GPU Ambient: 46-50C
Hard drive: 32-35C

I've looked up these values for my specific hardware models and people reported that their systems were stable even with temps higher than those. I know that the MCP seems pretty hot, but I triple-checked that one and it seems it's normal that it runs so hot on this type of board. Same with the graphics card; there were people running these at 80C and doing just fine.

I can't figure out what's overheating and it's driving me absolutely insane for the month and a half I've tried to fix it. Every time my computer shuts off, I drop down to the floor and quickly touch the components inside to see what feels hot to the touch. The MCP heatsink is pretty hot but it's not like I burn myself instantly when I touch it (I'd have to hold my finger on it for a while if I wanted to). The GPU back plate seems a bit hot, but it's always ran hot even when I first built the system a year ago, and it never overheated then. Other than that, everything else seems about normal temperature when I touch it.

Since the system only overheats when I game, it seems like it would be the GPU or CPU that was overheating. The CPU temps seem fine and I've put new thermal paste on it a few weeks ago, which lowered the temps by a few degrees. The GPU is the most likely candidate, but the temps are never consistent when it shuts down. Sometimes the GPU is at 62C and it will shut down, other times it is only 58C, and it still shuts down. So that makes me think it's not the GPU. But what else could it be? I've even thought that maybe, something is overheating that isn't showing a temperature read. Something tucked away on the motherboard, or something?

As you can see, I'm very stressed out with this machine; help would be VERY much appreciated. I want to be able to play my video games for more than 15-30 minutes again, without fear of them shutting down and making me go back and re-do the whole level/mission/whatever. If anyone has any ideas of what is happening, or can give me some tips on what to do, please speak up. I really don't want to go off and buy a new graphics card, install it, then find out the system STILL shuts down on me...

Side note: I listed my PSU because some people have reported similar symptoms to mine being related to a weak power supply, but 700W should be more than enough for everything I need. After all, the system has ran fine for a year, so why stop now? Maybe the PSU is going bad?

 

[pmv]

Aren't spontaneuous shutdowns a sign of a struggling PSU? Do you get sensible voltages reported by monitoring programs? Those temps look fine to me. PSU's can lose power with age, no?

Someone who knows what they are talking about will probably be along soon, but in the meantime, that's my contribution.

 

[user7092]

The thing is, I don't get why it would be the PSU, either. In some senses, I can see why, but when I put a box fan in place of the side panel of my case, my computer has no trouble staying on for long periods of time. I'm pretty inexperienced when it comes to power supplies, so I'm not too sure.

The computer only shuts off when gaming. If I turn it back on a few moments afterwards, it will shut down again as it's going through POST or booting Windows. If I wait a few minutes, I can get back in, but if I start gaming again too quickly without waiting, it will shut down in 3-5 minutes of gameplay. If I leave the computer off or idle for 30 mins-1 hour and try to game, I can usually go about 30-45 minutes before it shuts down. That's why I'm not sure the PSU is involved, but like I said, I'm no power expert.

As for voltage readings, they look pretty sensible to me. Copied over from Everest, I've got:

CPU Core 1.10 V
+3.3 V 3.25 V
+5 V 5.04 V
+12 V 12.36 V
+5 V Standby 4.92 V
VBAT Battery 3.10 V
North Bridge Core 1.34 V
DIMM 1.47 V
GPU Vcc 3.27 V

Those are just the ones under the sensor tab. I believe there are other readings in other tabs, but I think they link into those I just posted. There may be more, not sure, if someone could tell me what ones to post, I'd gladly post them.

Thanks for such a quick reply, Pmv. I appreciate the contribution! To be honest, though, I didn't think that someone would reply so fast, haha. I was just on my way to bed. 5:01 AM here, gotta love staying up on the weekends!

I'll check this as soon as I get up. Thanks again.

 

[superccs]

The chipsets on your mainboard are notorious for getting whicked hot, why not drop $8 and get a cooling fan for your chipset.
http://www.newegg.com/Product/Produc...82E16835185039

Though with the voltage you are running to your quad should be closer to 1.25v, it may be linked to this... http://www.tomshardware.com/forum/252088-30-solution-790i-freezes but that doesn't explain why you can't resume after a crash right away. Suggesting temp issue.

Best of luck.

 

[Intexity]

have you physically checked the fans while running the computer? As in is one failing under direct load?? If possible (just cause of the gaming connection) if you can put the gpu into another computer and see what happens. Could be the card itself especially if it is failing under load. Your temps seem pretty close to ideal...sounds like either your psu or your gpu is having issues under load...

 

[nboy22]

I don't think this is due to overheating.

I think it's the PSU. Your computer shutting off randomly while only playing games is a perfect indication that your PSU is going bad or just cannot supply enough power.

For instance, you fire up the game, the GPU and CPU start both drawing a lot of power from the PSU. The power the GPU and CPU are trying to draw is probably too much for the PSU to handle.

Open your case and find out what kind of wattage is printed on the side of the PSU, and the brand name, then report back here.

 

[WoodButcher]

Is the PSU fan ramping up when you are under load? is it running at all? have you dusted the PSU? sounds like the PSU is overheating.

Open your case and find out what kind of wattage is printed on the side of the PSU, and the brand name, then report back here.

it's in the OP
PSU: OCZ GameXStream 700W

 

[user7092]

@Superccs: I don't think that fan would fit on my mobo because of where my graphics card is. It's in the first PCI-e slot and the fan would have to be placed right underneath it, and there's not a big enough gap to put it there. Unless it goes somewhere else? I'm not too familiar with chipset fans, sorry. I also don't think it's the chipsets that are overheating, but I could be wrong. I will check the voltage on the CPU and see if I can raise that up, to see if it helps. Thanks!

@Intexity: Yes, I check the fans when I'm idle and under load, they are all running just fine. I wish I had another computer to put the graphics card in, but I'm afraid not. I do have a friend that could probably do it, but we don't see each other often... I'll try to line something up with them. Thanks for the suggestion!

@nboy22: In the first post are the wattage and brand name of the PSU. I agree with you in the fact that it seems like the GPU and CPU may be drawing a lot of power, but I'd think that 700W would be fine. It also doesn't explain why I can run fine if I have a box fan pushed up to the side of the case. Unless somehow, the PSU is drawing so much power that IT'S the one overheating, but I don't think that's it, either. The PSU's built-in fan is pretty strong, and it doesn't blow out hot air from the back, either. It's a little warm, but nothing terrible. But, who knows... it's such a mystery to me. Thanks for your suggestion!

 

[user7092]

Is the PSU fan ramping up when you are under load? is it running at all? have you dusted the PSU? sounds like the PSU is overheating.

Haha, you posted when I was posting the other comment. I haven't physically OPENED the PSU but I did take some compressed air and shot it through all the open holes. Some dust came out, but not a whole lot. The fan is running as strong as ever, and it doesn't blow out hot air, just mildly warm.

 

[user7092]

I was curious so I did some more investigating. Everest reported that my CPU had kind of a low voltage, but I looked around a bit more and found a range that it was, the 1.10 was just the lowest. I wasn't sure if I should trust it so I went into my BIOS to look at the voltage and it said 1.20V. Not sure if I should increase that or not.

Also, something interesting: Core Temp, Speed Fan, and Everest all throw out similar CPU temps... around 35-37C when idle. But in BIOS, I went into the temp screen and CPU was at 49-51C! What could be the issue? I know that sometimes the temp reader software you download can be inaccurate, and BIOS is probably the best thing to trust, but this just doesn't seem accurate. If it's running at 50C idle, who knows how hot that sucker gets under load. Everest/Core Temp/Speed Fan all say about 58-62C, but BIOS could say something much more outrageous. I wonder what's up.

Maybe I should try gaming until it shuts down on me, then quickly boot it up, go into BIOS, and see what the CPU temp is. Yeah, I think I'll do that... I will post back the results soon. I've got some Arctic Silver 5 on standby if I need it, even though I've done this a few times already in the past month.

Edit: I've been playing for 25 minutes now and no signs of overheating. I've noticed that my CPU temps aren't getting up to 57C on Everest now, though. They're about 47-50C or so. Maybe it really is the CPU that's overheating? Since it's not shutting down on me at the moment, I'll play until it gets really hot, shut it down, then check BIOS.

 

[WoodButcher]

Stock intel cooler? Pushpins all secured properly?

 

[Intexity]

well as far as your voltage...i have learned (from reading was waiting to finish liquid cooling before oc'ing) that upping your voltage one step at a time and running it after each step wont do any harm.abviously if it crashes its too high. the voltage range is 0.85V – 1.3625V so maybe go up and down to see if anything works...hopefully someone with more knowledge will comment on this post. just stating things that i have researched in the past...

 

[user7092]

Stock intel cooler? Pushpins all secured properly?

Stock, and all pushed in completely. I was looking at getting a new heatsink fan a few weeks ago, but I didn't think my temps were too high.

Also, as an update to the last post I made, I got it to shut down again after playing a game for about 30 minutes (after I posted it). I quickly turned it back on after it shut off and it shut down again a few times on me as I was trying to get into BIOS, but I finally got in, and the CPU temp was only reading 53C, even though it was at 51C idle. So, maybe it's not the CPU...

Thanks everyone, for trying to help me.

 

[user7092]

well as far as your voltage...i have learned (from reading was waiting to finish liquid cooling before oc'ing) that upping your voltage one step at a time and running it after each step wont do any harm.abviously if it crashes its too high. the voltage range is 0.85V – 1.3625V so maybe go up and down to see if anything works...hopefully someone with more knowledge will comment on this post. just stating things that i have researched in the past...

I went into BIOS again to look at some settings. Here are what I have to report for voltages:

CPU Core 1.237V
CPU FSB 1.1V
CPU PLL 1.5V

Now, for settings:
Limit CPUID MaxVal - Disabled
Intel SpeedStep - Disabled
CPU Thermal Control - Disabled
C1E Enhanced Halt State - Enabled
Execute Disable Bit - Enabled
EVGA Vdroop Control - With Vdroop (Meaning: Enabled)

I'm going to look up these values and settings and see if they should be turned off/on and see if I need to tweak the voltage a bit. I might try what you said, just lower the voltage gradually by little bits at a time, and we'll see if that helps. Thanks for the suggestion!

If anyone else has any experience with the settings I listed above, I'd be happy to hear some opinions. Thanks.

 

[WoodButcher]

For bios settings, generally speaking "optimized defaults" is a safe place to start. Your temps in the bios seem high. The heatsink, is it a copper core or just aluminum?
I looked at your case, http://www.newegg.com/Product/ImageG...omputer%20Case and I'm thinking the airflow sucks.
I believe you posted that it ran ok with a fan blowing at the open side of the pc? The side panel has room for fans, if they are not installed the holes should be covered. If not the rear fans are pulling air in the side and right out the back without cooling anything. As it is sold the front fan is the only fan that will do any work except for the fact that it simply blows at the side of the HD cage killing any possibility of air movement. Tape the inside of the case panel with masking tape to see if I'm right and if so think of a way to make it pretty.

 

[user7092]

For bios settings, generally speaking "optimized defaults" is a safe place to start. Your temps in the bios seem high. The heatsink, is it a copper core or just aluminum?
I looked at your case, http://www.newegg.com/Product/ImageG...omputer%20Case and I'm thinking the airflow sucks.
I believe you posted that it ran ok with a fan blowing at the open side of the pc? The side panel has room for fans, if they are not installed the holes should be covered. If not the rear fans are pulling air in the side and right out the back without cooling anything. As it is sold the front fan is the only fan that will do any work except for the fact that it simply blows at the side of the HD cage killing any possibility of air movement. Tape the inside of the case panel with masking tape to see if I'm right and if so think of a way to make it pretty.

Here's the kicker: When I load fail-safe defaults in BIOS, the computer won't boot into Windows at all. No error messages, just stays on a black screen the whole time. I've loaded the "optimized defaults" a few days ago and it didn't change much of anything for me. The heat sink is copper.

I have a fan on the side panel that blows air into the case. As you can see, there are 4 slots to put them, but I put one 3-speed fan in the bottom left (so that it's right over top the GPU/southbridge heatsink). The fan is pretty nice, blows a good amount of air into the case, then the rear fans and PSU fan blow it out the back (PSU sucks it up from the bottom and blows it out the back). Do you think I need to add another fan to the side?

I will try gaming again with the box fan set to the side panel, to see if it shuts down. I want to make sure it works every time 100% with the box fan; if it does, I will try your masking tape technique and seal up the empty fan slots on the side, while keeping the 1 slot with the fan open. Thanks so much for the suggestion, it might really work (if it shuts down even with the box fan, then we've got issues... PSU maybe, but we'll see).

 

[pmv]

Does the windows event viewer say anything of any significance about events immediately before a shutdown?

I suppose you could try adding a fan to every one of the side spaces! Which would presumably make the problem _worse_ if its a PSU thing, but better if its a heat thing.

 

[RadiclDreamer]

Even though your PSU SHOULD be delivering 700, it doesnt seem like it IS.

 

[user7092]

@Pmv: I checked that before, too. All it says is, "System shutdown at (time) was unexpected." I'm really starting to wonder if it's a PSU issue, too. I keep leaning on the idea that it is, then backing away, then leaning again. The issue is that the shutdowns are never consistent. They always vary, making an accurate diagnosis extremely difficult.

@RadiclDreamer: I'm wondering that, myself. I have a friend who has the exact same PSU and it works fine in her rig. I might have to get her to lend it to me for a few hours, just wish she didn't live so far away. Haha.