Overheating?

[Hectic Glenn]

Hey guys, just need some educated opinions. In the last three days i've been getting shutdowns with bluescreens saying driver failure, which associate with overheating problems. The climate temperature at the moment is very hot, and i'm wondering if that could have had an effect.

The computer has been fine with adjustments/changes to it in the past. I've recently added an external hard drive (with own power supply - mains) which is USB, and a new razer mouse, which i doubt could have any effect on this problem right?

I've opened up the tower, all the fans are moving, i touched my gfx card, and it feels pretty hot (catalyst control panel says 50 degrees at idle, is this acceptable?)

P4 2.66ghz
512mb DDR ram
ati 9600XT 128mb
Windows XP

 

[giant384]

whats your cpu temperature if it has one on your motherboard and give the exact temperature

 

[Asus]

Have you gfx card properties window up showing the tempature. Launch a game and after you have played for a bit, alt+tab out of the game and check what it says for temp.
How warm is it at your place? Room temp.

 

[Hectic Glenn]

Its about 28 degrees outside, because of the heatwave, so the room is about that i'd say. (the computer heat seems to heighten it)

My GPU is 47 degrees and GPU Ambient (whatever that is) is 42 degrees.

I'll try the game thing shortly Asus cheers. The problem of restarting usually occurs after the computer is left standing for some time and when is used again it's not very happy.

I have the Tower case off for now and the card is now 47 degrees, 3 better! Wahey, but i'm not sure if thats good or not.

 

[_Z_Ryuken]

i touched my gfx card, and it feels pretty hot (catalyst control panel says 50 degrees at idle, is this acceptable?)

It is very important to specify whether Celsius or Fahrenheit. I'm guesing Celsius, but just the same.

Anything over 40*C ambient (not doing anything) is very hot.

 

[giant384]

it could be something that using up all your ram dosent seem like a heat issue to me

 

[_Z_Ryuken]

it could be something that using up all your ram dosent seem like a heat issue to me

I disagree. If he is getting bluescreens, it is likely a purely hardware error, unless he is getting mass corrpution of core files, in which case it might still be a hardware problem.

 

[Asus]

Most boards today will automatically shut your PC down if the CPU goes much over 70C. Video cards can confortably go up to 70-80C while playing games. As long as the temperatures are below that I think it would be fine.

I'm not exactly sure how a bluescreen from a driver could be caused by heat to be honest. The only things I can think of that would is the Graphics card (which does not lock up the system but crashes the driver) or the Northbridge chipset (does not have a temp sensor).

If your CPU uses HT, maybe disable that.

 

[_Z_Ryuken]

Most boards today will automatically shut your PC down if the CPU goes much over 70C.
Video cards can confortably go up to 70-80C while playing games.

If the video card hangs or memory gets corrupt generally your PC just restarts. I'm not exactly sure how a bluescreen from a driver could be caused by heat to be honest. The only things I can think of that would give you a driver related error from heat is the Graphics card or the Northbridge chipset.

A friend of mine was experiencing a very silimar problem as this. I determined the power supply was faulty. Upon removal of the supply to MB power cord we found it to be singed. There was [possibly] a slight arc fault causing MB and VC temps to rise and frequent crashing.

My advice to you Glenn, is disassemble your PC and check for burns or other signs of voltage problems.
Your BIOS menu will tell you if you are getting correct voltage.

 

[Asus]

A friend of mine was experiencing a very silimar problem as this. I determined the power supply was faulty. Upon removal of the supply to MB power cord we found it to be singed. There was [possibly] a slight arc fault causing MB and VC temps to rise and frequent crashing.

That's a good idea, to check the Powersupply.
Just wondering but how much room do you have from the back of the case or powersupply to the wall?

 

[bbson john]

It is very important to specify whether Celsius or Fahrenheit. I'm guesing Celsius, but just the same.

Anything over 40*C ambient (not doing anything) is very hot.

No. For once, I miss-installed the fan for the CPU and it was not in operation. When I checked the CPU's temperature in BIOS, its temperature went up to 100*C, which was hotter than boiling water. I opened my computer and checked the components inside. The heat sink nearly burnt my finger. But after I repaired the fan, the CPU worked completely fine, no sequela.

 

[Hectic Glenn]

So just to confirm you think its unlikely my recent additions are to blame or the climate, its just something driver related going wrong. I did the the problem occuring once every 2/3 months, which was fine. (i turn my PC off every night too). But i have had 4 in 2 days, which is why i'm concerned.

I bought the new HD to copy my data over for a reform, which seems like a good idea (2 years of clogging). So with regard to temperatures, i'm not too bad then? And Asus, that could be a problem i have it under my desk pushed up against the left support (solid), quite far back (not much room at the back) and a sub-woofer to the right side.

 

[giant384]

No. For once, I miss-installed the fan for the CPU and it was not in operation. When I checked the CPU's temperature in BIOS, its temperature went up to 100*C, which was hotter than boiling water. I opened my computer and checked the components inside. The heat sink nearly burnt my finger. But after I repaired the fan, the CPU worked completely fine, no sequela.

my god lol 100 C im surprised it didnt burn out

 

[_Z_Ryuken]

No. For once, I miss-installed the fan for the CPU and it was not in operation. When I checked the CPU's temperature in BIOS, its temperature went up to 100*C, which was hotter than boiling water. I opened my computer and checked the components inside. The heat sink nearly burnt my finger. But after I repaired the fan, the CPU worked completely fine, no sequela.

You misunderstand. The components can handle in excess of 100*C, but if it is sitting unused, the temperature should not exceed 40*. If it does, it is not being properly cooled, as you have observed.
In a full-processing state, there very well may be damage incurred by uncontrolled heat. Plastic connectors may melt for example, as I have observed.

So just to confirm you think its unlikely my recent additions are to blame or the climate, its just something driver related going wrong. I did the the problem occuring once every 2/3 months, which was fine. (i turn my PC off every night too). But i have had 4 in 2 days, which is why i'm concerned.

I bought the new HD to copy my data over for a reform, which seems like a good idea (2 years of clogging). So with regard to temperatures, i'm not too bad then? And Asus, that could be a problem i have it under my desk pushed up against the left support (solid), quite far back (not much room at the back) and a sub-woofer to the right side.

The likelyhood of your room temps having any bearing on internal PC temps is very small.
Check for singed/burned connections, and go into BIOS and check your temperatures, voltages, and fan speeds.

A burnt component can cause an increase in impedence, lowering fan speeds, causing overheating, which will crash your PC, and your PC will not be able to tell you what is wrong.

 

[bbson john]

You misunderstand. The components can handle in excess of 100*C, but if it is sitting unused, the temperature should not exceed 40*. If it does, it is not being properly cooled, as you have observed.
In a full-processing state, there very well may be damage incurred by uncontrolled heat. Plastic connectors may melt for example, as I have observed.

I am just surprised that the fan is that important. And the CPU can live over 100*C.

 

[_Z_Ryuken]

I am just surprised that the fan is that important. And the CPU can live over 100*C.

It is only silicon and copper, after all. Increase in heat merely makes the flow of data unstable at a point.

Sudden shocks like from weak static feilds do far more damage.

 

[Hectic Glenn]

Ok i will check the BIOS tomorrow, however checking the connections for burned connections might be tricky. Shall i just pull the plugs out (from the PSU obviously) and check it looks clean?

Also i must mention my card is directly above my sound card, which i think could be restricting cooling? Its always been there without problem but i'm not sure? If i move the soundcard down a PCI slot or two, will my computer say 'you changed slots!? i can't find it so it's not going to work!?'.

 

[_Z_Ryuken]

Ok i will check the BIOS tomorrow, however checking the connections for burned connections might be tricky. Shall i just pull the plugs out (from the PSU obviously) and check it looks clean?

Also i must mention my card is directly above my sound card, which i think could be restricting cooling? Its always been there without problem but i'm not sure? If i move the soundcard down a PCI slot or two, will my computer say 'you changed slots!? i can't find it so it's not going to work!?'.

Just look at all your contacts. Any burning will be obvious. Transparent/white connectors will have a brown spot, black connectors will be less obvious

It should not matter if you move your card. It will probably just reassign an IRQ and figure it out as if nothing has happened. If you have extra space I would make use of it.

Also just for giggles put your hand by your fan exhaust and try to feel the temp of the air coming out. Mine is almost cool enough to direct on myself on a hot day.

 

[Hectic Glenn]

I assume thats the fan right at the back which my processor + heatsink are in front of? Mine isn't face cooling niceness, but it is cool and doesn't feel hot, just a tinge of warmth. Is there any immediate cause for concern in my case or could it just be something software related instead. When i format of course my woes will be gone i hope. (I have a feeling i'm going to need help with that perhaps hehe). Cheers for all your help so far guys, it's been priceless.

 

[_Z_Ryuken]

When i format of course my woes will be gone i hope. (I have a feeling i'm going to need help with that perhaps hehe). Cheers for all your help so far guys, it's been priceless.

Formatting should always be a last resort.
If you have bad hardware, it would be a wasted effort.

Can you tell me your power supply specs? Wattage output, age, make, model, etc.

 

[Asus]

if it is sitting unused, the temperature should not exceed 40*. If it does, it is not being properly cooled, as you have observed.

For an enthusiast, it may be far from what you want but many PC CPUs sit above 40C idle. No problem with that as long as it does not get to extremes and cause stability issues. I agree if you are talking about case temperature though.

The likelyhood of your room temps having any bearing on internal PC temps is very small.

I disagree. Room temperature impacts PC temp quite a bit. Though things like dust and airflow can make a bigger impact.

 

[Hectic Glenn]

I can only tell you my PSU is from a dell dimension 4600, and i believe its 280 watts and about 2 years old.

Asus, do you think in my case then the temperature being 50ish generally is generally ok? With regard to the temperature right now. Also i'd like to clarify i've bought this new external HD to backup my data to format, it was already on the cards. So don't worry about that

 

[_Z_Ryuken]

All I know is my cpu (P3) sits at 30-32*C at idle with one fan in a 72*F room.
My friends P4 sits at 34-36*C with 4 or 5 fans in a 68*F room.

I suppose if your P4 has a few less fans and is in a room around 80*F you might see temps at or above 40*C.
But if that is the case, I still do not think it is being properly cooled. While running a game it will run hotter than 4-5* over what I run as heat is compounded. It may be 12-14* higher, and with a faulty PS, 20-30*C hotter, resulting in crashing and possibly damage.

I can only tell you my PSU is from a dell dimension 4600, and i believe its 280 watts and about 2 years old.

280 watts is sorta small. How many peripherals do you run?

 

[Asus]

Asus, do you think in my case then the temperature being 50ish generally is generally ok?

Is that your CPU?
Well you only know about how hot your components are with sensors. So that would be CPU, Graphics card and case temp. I wouldn't assume that the rest of your component's temps are OK if you are having stablity issues.

50C for a CPU temperature is fine. I would not let it go over 60C for a running temperature. It might make an error when crunching numbers.

 

[Hectic Glenn]

280 watts is sorta small. How many peripherals do you run?

What categories as a peripheral sorry? USB devices i assume don't? I have the gfx card, sound card, PCI network card, DVD-RW drive, and the 4 USB devices currently plugged in.

My CPU said 43, and my card 47/50 now.

 

[_Z_Ryuken]

What categories as a peripheral sorry? USB devices i assume don't? I have the gfx card, sound card, PCI network card, DVD-RW drive, and the 4 USB devices currently plugged in.

My CPU said 43, and my card 47/50 now.

Everything that draws power from your PC including USB devices and hard drives. Everything you add puts strain on the PSU.

Those temps are probably fine. This one thing I'm not sure of, but maybe video cards generally run hotter than everything else. Mine does not have a sensor.

 

[Asus]

Do you know if your CPU is a Northwood or Prescott core?

If you are worried about your Power Supply then check out this power supply calculator.
This one is pretty accurate.

This one thing I'm not sure of, but maybe video cards generally run hotter than everything else.

Yeah they run pretty high compared to CPUs. My GFX card can go up past 70C when playing games. Max temp on AMD CPUs is 85C and my GFX card max temp is somewhere around 115C.

 

[Hectic Glenn]

Prescott it would appear, it also advises 230 watts. People say my PSU isn't good enough, but it always seems to be fine for what i need it for. I don't use too many demanding add-ons.

Do you know the driver fail/ blue screen message on Windows XP i meant?

 

[_Z_Ryuken]

I don't. I can't remember the last time I saw a BoD on an XP system.

230W is fine for most people but the thing is if you are a gamer with a video card and a sound card and USB devices you are probably maxing out the PSU. Constantly maxing it out will shorten it's lifespan.

With a higher output PSU, it's no so much the power you are paying for, but the longevity. Mine is 400w and going on 5 years, still strong. I don't have nearly enough devices to make use of that kind of output.

 

[Asus]

What I would do first is make sure your PC is off the ground and the back of your PC is at least 15cm away from the wall. I assume you have a case fan in the front and back. If you have appropriate cooling then I would not leave the side of the case off. Touch your heatsinks quick. Are they just warm or pretty hot?

Prescott it would appear, it also advises 230 watts.

Did you notice the Capacitor Aging option. Check the notes below. Maybe try it with the USB devices unplugged. (Not including mouse/keyboard)

 

[Qonfused]

Well, it is summer so it will heat up more. For the summer, I just open up the side of my case and point of circular fan at the side on lowest.

 

[Hectic Glenn]

So it in summer your computer can be affected, because i wanted to know that, it seems that might be my case.

I touched my heatsinks, they don't even feel warm today, just off-cold, which i suppose is good.

Would you suggest that when i'm not using a USB device is it better to unplug it completely to reduce strain/ life of a PSU? Also, the largest increased strain on my system is the new 250gb seagate USB 2.0 external hard drive i added, i'm wondering (even though it has its own mains power supply, to avoid strain on the PSU) if it would still be sapping power from the computer? If so i'm very disgruntled about that.

This morning i checked most of the white PSU plugs and theres no sign of burnt parts. I also moved my sound card to the lowest PCI slot, and removed dust from gfx card, sound card, grill at the front (which was caked!), and a little at the back. The exhaust, i think you called it, is clean and the fans in the PSU, to the back of my processor and in my graphics card are all spinning like before. My card temperature is 42 degrees today, with its extra space (at idle).

Also if i plug my mouse into the mouse port rather than USB will it be better on the PSU? I have a USB to mouse port plug, but my mouse doesn't like it. It's got a flashy gold usb plug and doesn't get recognised. If i tried really hard it might work, i'm just wondering if this kind of thing would save power?

 

[Asus]

The PS2 port uses a little less power than USB for the mouse. If you are not using the harddrive you might as well turn it off or unplug it. That is probably more important.

As long as you have proper cooling, I'd keep the side of the case on because that will keep the airflow streamline and hit up all the components it needs too front to back. That is more important than your CPU or graphics card reporting a lower temperature. The reason is because other components are not getting air like they would if the case was closed and they have no sensor to warn you. Not to mention a lot more dust gets in with the side open.

(Heat + Dust = Computer's worst enemy duo)

 

[Hectic Glenn]

Right o, i'll clip stick the side back on shortly, directing the airflow inside makes good sense. I just had a go playing a game and checking the temperature of the card, which showed 53 degrees, so 10 increase from idle. That doesn't seem too bad.

 

[giant384]

dont worry about usb ports they use very little power

 

[_Z_Ryuken]

dont worry about usb ports they use very little power

This is true.

Unless you are still having stability issues I'd say you are good to go.

What are your voltage readings? There should be at least 3.
1.5v
3.3v
12v

They should be within 0.2v. Any more than that and there is a probably a problem somewhere.