Why does electrical equipment get hot?
-
What did you upgrade from? What else did you change at the same time as your CPU? Even though CPU power consumption has apparently peaked and turned downward a bit on the low-midrange end, graphics hardware in particular is consuming ever more power. Top end nVidia cards are drawing almost 200W, ATIs topend card is at 225W. Consumer level cards haven't grown nearly that fast but there's some growth there as well. Unless you have major airflow problems with an apparently low to mid range system 9 exhaust fans is major overkill.
-- If you view money as inherently evil, I view it as my duty to assist in making you more virtuous.
dan neely wrote:
What did you upgrade from?
Two hard drives and RAM seemed to have tipped the scale. On warm days, the system would shut down at inopportune moments. So, replaced everything with new CPU, GPU, 2 more HD, another DVD, pretty lights. Big tower - lot's of space. Eliminated ribbon cables. No airflow problems now. PS fan, CPU fan, Graphics fan, 4 LED output fans, 2 input fans. Hums right along. Know what I mean?
Gary
-
It's because of all the smoke moving around inside the cpu getting the work done. It's a little-known fact that all electronic equipment works using smoke. It is for this reason that when you let the smoke out, it stops working. ;)
-
If it didn't need power you wouldn't have to plug it in. Most of the power is turned into heat, a bit into other radiation - light, radio, x-rays if a CRT, and some sound. A tiny tiny little bit of energy might be stored in a flash memory device. I'm not sure if anyone knows the limits of how much energy is actually needed to process / store information (interesting philosophical question - it is Sunday), but I suspect that it is many orders of magnitude less than we use.
Peter "Until the invention of the computer, the machine gun was the device that enabled humans to make the most mistakes in the smallest amount of time."
In principle, a computational device does not need to create any heat if no information is lost in the computation. This usually means returning the question along with the answer at the end of a computation. So now we have the design, we just need those engineer types to work out the details.
-
dan neely wrote:
What did you upgrade from?
Two hard drives and RAM seemed to have tipped the scale. On warm days, the system would shut down at inopportune moments. So, replaced everything with new CPU, GPU, 2 more HD, another DVD, pretty lights. Big tower - lot's of space. Eliminated ribbon cables. No airflow problems now. PS fan, CPU fan, Graphics fan, 4 LED output fans, 2 input fans. Hums right along. Know what I mean?
Gary
ghle wrote:
Hums right along. Know what I mean?
yeah, I've largely moved my cooling over to 120mm low speed fans instead of faster 80mm ones. Not silent, but much quieter.
-- If you view money as inherently evil, I view it as my duty to assist in making you more virtuous.
-
ghle wrote:
Hums right along. Know what I mean?
yeah, I've largely moved my cooling over to 120mm low speed fans instead of faster 80mm ones. Not silent, but much quieter.
-- If you view money as inherently evil, I view it as my duty to assist in making you more virtuous.
-
In my PC I have one whopper of a CPU heat sink and more fans than you can shake a stick at, this is because I know these components get hot. But what causes this heat? Is it some type of friction down on the atomic layer, or what?
Me: Can you see the "up" arrow? User:Errr...ummm....no. Me: Can you see an arrow that points upwards? User: Oh yes, I see it now! -Excerpt from a support call taken by me, 08/31/2007
What is heat? Vibrational energy (and in gases, translational and rotational, as well). Do not confuse this with the 'warmth' coming off of your system, which is energy in the form of radiation (to keep it simple, we'll say it's infrared). The IR radiation is one of (beloved) entropies methods of dispersing concentrations of energy - it's way of transferring it through a vacuum. (Yes, we've convection, as well, for those fans to assist). First of all, whilst electricity moves through your equipment (and everything else) at the speed of light, ELECTRONS DO NOT! They actually move relatively slowly. In order to travel through a medium, they must overcome resistance. This has quite a few sources: A lot of it is electromagnetic in nature, wherein the induced fields ultimately are inopposition to the fields that induced them. This is working on the wave-nature of the electricity (i.e., an alternating current of rather high frequency these days). One way you can help yourself get a feel for one of these is that the structure (of a conductor) at the atomic level is trying to reorient itself with the field. There is always some degree of lag; the two fields are thus in some degree of oppostion. Back to those moving electrons: as they seep through your system, they are not moving smoothly, but rather from one region of lower potential energy to another. This structure is due to the atomic structure of the conductive medium. The regions are damn small. However, to move up the hill (so to speak) from one region to the next, requires energy (i.e. a push we call voltage). This extra push we may call a work function (certain license taken with the term). When it goes back down to the lower energy state, the energy is released - well MOST OF IT. That bit left over, which was part of the electron/atom system was distributed about the entire system to some extent. That bit left over is dispersed into atomic (better said, Molecular) motion. Your getting hotter. This disperses from electronic through vibrational, &etc. modes (degrees of freedom) - another pleasure from our friend entropy perefers this. You always 'lose' a bit, even though you're going up & down the same hills of energy. No free-wheeling coasting. It has to do with the irreverability of real systems. Basically a universal tax on everything. This should be enough (for now).
"The difference between genius and stupidity is that genius has its limits." - Albert Einstein
-
In about 100 years, physicists will discover that there's something smaller than quarks even. Hamsters spinning in little cages. So, give those hamsters a rest and turn your computer off. :) Marc
-
In my PC I have one whopper of a CPU heat sink and more fans than you can shake a stick at, this is because I know these components get hot. But what causes this heat? Is it some type of friction down on the atomic layer, or what?
Me: Can you see the "up" arrow? User:Errr...ummm....no. Me: Can you see an arrow that points upwards? User: Oh yes, I see it now! -Excerpt from a support call taken by me, 08/31/2007
What generates the heat is the power dissapated by the component. Energy is used and must be converted to some other form of energy, in this case heat. My understanding is that as you increase the clock frequency of a transistor, you get more leakage current, and in line with Power = Voltage x Current, you have power that needs to be dissapated, hence the heat.
-
In my PC I have one whopper of a CPU heat sink and more fans than you can shake a stick at, this is because I know these components get hot. But what causes this heat? Is it some type of friction down on the atomic layer, or what?
Me: Can you see the "up" arrow? User:Errr...ummm....no. Me: Can you see an arrow that points upwards? User: Oh yes, I see it now! -Excerpt from a support call taken by me, 08/31/2007
Thanks to all who took the time to reply - even those with comedy answers (hint: don't give up the day job guys :) ) - I find the whole subject absolutely fascinating, and am even now inspired to do an evening course in electronics! I got frustrated with CP not letting me vote y'all a 5 at the same time, so even if you didn't get one consider yourself on my Christmas card list!
Me: Can you see the "up" arrow? User:Errr...ummm....no. Me: Can you see an arrow that points upwards? User: Oh yes, I see it now! -Excerpt from a support call taken by me, 08/31/2007
-
dan neely wrote:
faster 80mm ones. Not silent, but much quieter.
Sorry, can't hear you... :((
Gary
If you want a decent quiet case, I can thoroughly recommend the following: Antec Solo Case 2 x 92mm Nexus Real Silent fans in the front 1 x 120mm Nexus Real Silent fan in the Rear 1 x Noctua CPU heat sink and fan This thing is very, very quiet and cool, which is no bad job given I have: Q6600 8GB Corsair "go faster" stuff 2 x 8800GTS's 2 x 150GB Raptors 2 x 500GB Samsung Spinpoints :)
Me: Can you see the "up" arrow? User:Errr...ummm....no. Me: Can you see an arrow that points upwards? User: Oh yes, I see it now! -Excerpt from a support call taken by me, 08/31/2007
-
If it didn't need power you wouldn't have to plug it in. Most of the power is turned into heat, a bit into other radiation - light, radio, x-rays if a CRT, and some sound. A tiny tiny little bit of energy might be stored in a flash memory device. I'm not sure if anyone knows the limits of how much energy is actually needed to process / store information (interesting philosophical question - it is Sunday), but I suspect that it is many orders of magnitude less than we use.
Peter "Until the invention of the computer, the machine gun was the device that enabled humans to make the most mistakes in the smallest amount of time."
Actually not that philosophical. There is a discrete amount of energy that electronic engineers and solid state physicists have agreed upon as the minimum amount to accurately represent digital quantities. We are a few orders of magnitude from this number still but the gap is closing fast. Additionally, when designing a power supply circuit, to deliver optimal energy from your voltage source, you have to dissipate(emit as heat) as many watts of power in the supply as you do in the target circuit. And why do computers use energy in the first place? Computers used to use a lot more energy before they started using CMOS (complimentary metal oxide silicon) technology. Thanks to CMOS technology energy is technically only dissipated when a logic gate switches state from 1 to 0, CMOS is larger, more difficult to fabricate, and slower than other logic technologies but overcomes the heat and energy barriers and therefore in practical applications can be fabricated to be smaller and faster.