can a cpu get to cold

[knuckles356]

i just bought a cpu cooler and it came with a fan control. do i need to use this or will the fan run at full blast and just be noisey. which i dont mind. the cooler is a zalman cmps9500

 

[spathotan]

Dont need the fan controler, PWM will control the fan speed.

And dont worry, that cooler isnt going to put your CPU down to -200c.

 

[knuckles356]

lol, cool thanks man

 

[werxen]

this is actually an interesting subject. i was thinking about making a little 'shed' for my computer attached to the house. in winter times it gets very cold here in colorado. i wonder if i could see a significant difference in my overclocking.

 

[cjl]

You could, but I might worry about the hard drives. While CPUs and graphics like being really cold for good overclocks, hard drives actually work best between 30 and 45C. This is about the temperature they typically operate at in a case, but if you had them outside, in subzero celsius temperatures, they might actually get cold enough to cause premature failure.

 

[The_Blood_Raven]

The only way to get a CPU too cold is to get it to absolute zero where the electric currents and chemical reactions would fail. The only other problem is if you CPU temperature is well below room temperature which might create condensation, frying the CPU. You wont have to worry about condensation with ANY air cooling solutions nor almost any watercooling setups.

 

[V3NOM]

^or if you bought an AMD. :lol:

 

[knuckles356]

wow, i learn alot on this forum. dont worry i will keep the questions comeing

 

[werxen]

The only way to get a CPU too cold is to get it to absolute zero where the electric currents and chemical reactions would fail. The only other problem is if you CPU temperature is well below room temperature which might create condensation, frying the CPU. You wont have to worry about condensation with ANY air cooling solutions nor almost any watercooling setups.

thats true although:

1) theoretically nothing has reached absolute zero, but it can get close
2) CPU's CAN get too cold however im not sure what happens. i dont know if its the condensation from the LN2 or if something physically happens to the transistors if they become too cold (possible metal contraction?) maybe you can answer that for me. im too lazy to google and you sound like a guy who actually knows his stuff.

 

[B-Unit]

Heheh, ive considered the same thing werxen. My dream is actually to use some dryer ducting and some type of filtration, and just feed cold air into the case. Probably more effort than its worth, but on days like today I could get some impressive clocks I'm sure.

 

[RetiredChief]

werxen
Positive/negative temp coeffient may have something do do with that. Most (NOT ALL) conductors have a Positive temp coeffient in that their resistivity increases as temp increases ( reason filiment light bulbs/toasters/electric heaters work and do not go "poof". In the case of semiconductors, most have a negative temp coeffient. This explains why heat is detrimental to semiconductors. Above a given temp thermo runaway occures. As the temp goes up, resistance goes down causing current to increase which inturn causes temp to go up more until it goes poof. As you go to "Very cold" you could reach a point where conductors have almost no resistance while the semiconductors resistance becomes very high and would require a higher voltage to "open" gates. You could also have an effect simular to water which has its highest density at 4 C. Corralation - at absolute zero, 0 resistance; Slightly higher, Very high resistance.

Yes I'm sure temp coeffients regarding expansion/contraction could play a part as the contraction of metal contacts to substrate could cause the contact to fail.

 

[knuckles356]

what i was thinking is that there is a operating temp at which the cpu would run best at. not to cold and not to hot. am i wrong in thinking that?

 

[spathotan]

Colder it is the better it runs. Some CPUs have a "cold bug", which I think usually happens around -215c?

 

[werxen]

werxen
Positive/negative temp coeffient may have something do do with that. Most (NOT ALL) conductors have a Positive temp coeffient in that their resistivity increases as temp increases ( reason filiment light bulbs/toasters/electric heaters work and do not go "poof". In the case of semiconductors, most have a negative temp coeffient. This explains why heat is detrimental to semiconductors. Above a given temp thermo runaway occures. As the temp goes up, resistance goes down causing current to increase which inturn causes temp to go up more until it goes poof. As you go to "Very cold" you could reach a point where conductors have almost no resistance while the semiconductors resistance becomes very high and would require a higher voltage to "open" gates. You could also have an effect simular to water which has its highest density at 4 C. Corralation - at absolute zero, 0 resistance; Slightly higher, Very high resistance.

Yes I'm sure temp coeffients regarding expansion/contraction could play a part as the contraction of metal contacts to substrate could cause the contact to fail.

interesting... i will read into this more because i have a few questions but i want to make sure they are not stupid questions before i ask :lol: this thread actually interests me

 

[knuckles356]

is this the same with video cards, sound cards, psu, and hard drive? colder the better?

 

[werxen]

is this the same with video cards, sound cards, psu, and hard drive? colder the better?

yes. general concensus with PC equipment is the cooler the better. i wonder how guys in Australia and Hawaii with no A/C can manage to overclock...

 

[spathotan]

Not harddrives, probably not PSU's either.

 

[werxen]

Not harddrives, probably not PSU's either.

ugh... why do you put fans on HDDS and why do PSU's come with fans? please refrain from giving advice.

 

[knuckles356]

i think he meant that they can get to cold unlike a cpu, i have read that hdds can get to cold and malfunction

 

[spathotan]

ugh... why do you put fans on HDDS and why do PSU's come with fans? please refrain from giving advice.

Please, stop trolling. This thread is about cold, you would know this if you were actually reading instead of trying to be a wise-ass. 32c or 50c ambient is not cold, below 0 or freezing is cold, which is what we are talking about. Cold cold.

Cjl clearly explains this in his above post, which...once again you would know if you had been reading.

 

[B-Unit]

The issue with hard drives is that the read/write heads float extremely close to the magnetic platters. So close infact, that even the tinyest dust and water particles are in effect boulders. When you get a hard drive good and cold, then fire it up, the sudden increase in temperature can cause condensation on the platters, which means goodbye hard drive. For that reason, its not advisable to use a hard drive in extremely cold situations (IE anything below 15-20C)

 

[werxen]

Please, stop trolling. This thread is about cold, you would know this if you were actually reading instead of trying to be a wise-ass. 32c or 50c ambient is not cold, below 0 or freezing is cold, which is what we are talking about. Cold cold.

Cjl clearly explains this in his above post, which...once again you would know if you had been reading.

its not trolling dude. use common sense. would you put liquid nitrogen on a harddrive or PSU to keep it cool?

 

[spathotan]

its not trolling dude. use common sense. would you put liquid nitrogen on a harddrive or PSU to keep it cool?

Nobody is talking about LN2. The conversation switched over to putting hardware outside when its extremely cold, as a matter of fact a conversation that YOU brought up. Youre not even reading your own post anymore.

 

[cjl]

its not trolling dude. use common sense. would you put liquid nitrogen on a harddrive or PSU to keep it cool?

No. However, if you put it outside in the winter, like you said you might, I can almost guarantee you that the hard drives would run below 30C, unless you have the worst case ventilation I have ever seen. Therefore, they would be running outside of their optimal range, and if they got below 10 or 15C, a definite possibility when many days are well below 0 here (right now for example - it's 3F out right now, around -16C), it could significantly increase their chance of failure.

 

[knuckles356]

out of curiousity what it a high temp for a cpu, mine seems to run at 73%C. i havent put on the zalman cooler yet