Slashdot Mirror
On-CPU Peltiers From AMD?
Hack Jandy writes "Remember those people who lived on the edge and put peltiers between their CPU and heatsink (or your favorite beverage)? A peltier is a devices that gets cold on one side and warm on the other when an electrical current passes through it. It looks like there is talk that AMD will actually incorporate some of these devices on the CPU according to Xbitlabs. AMD already incorporates some degree of the peltier effect with it's Silicon on Insulator."
it's == contraction for "it is": "It's a peltier device."
its == possessive for it: "AMD has a peltier-like effect with its Silicon on Insulator technology."
IT'S (pun intended) NOT THAT HARD, PEOPLE!
Actually, the funny thing is, this hasn't been true for years. Intel CPUS put out quite a lot more wattage (~90w) compared to AMD's top end silicon, around 70 watts. But it made for a mildly funny joke back when the t-bird was spanking the socket 423 chips.
01101001 01100001 01101101 01101110 01101111 01110100 01100001 01101100 01100001 01110111 01111001 01100101 01110010
Just imagine the amount of power required for something like that. Each core of a dual core 65nm CPU will need at least 70 watts of power, giving 140 watts total. In order for a peltier to be useful, the wattage has to be greater than that of the CPUs. So with only the CPU and Pelt, that is a minimum of 300 watts, with something between 400 and 500 being more likely. That is an absurd amount of power and heat.
eclecti.cc
I don't know where you get this information, but the new AMDs are somewhat cooler running that the top end Pentium 4s.
As for heat coming from the other side, that's one of the issues that an on-chip Peltier would alleviate, but presenting a cooler surface to the CPU on the side with the heatsink.
i don't think the realiability is the real issue.
it seems xbitlabs is just.. well.. doing what journalist wannabees like - take some facts and twist them.
for one: YOU DON'T GET MORE -EFFECTIVE- cooling with peltiers. you end up using more power than you would with normal cooling. the total heat output gets _increased_.
unless they(chips) can't work in normal room temps there's not really much point in using peltier cooling in cooling them, except if you somehow manage to get the effect 'for free' or something.
world was created 5 seconds before this post as it is.
The 1990's called, they want your facts back. If anything, lately its Intels P4 generating more heat overall. Anandtech has a useful article on the issue http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2026
As mentioned already, Peltier junctions act as heat pumps. In other words, you put work in to extract heat from the processor, this extra work turning into (you guessed it) even more heat.
While it might contribute to the cooling of the processor, you'll need an even bigger heat-sink and fan stuck on top to dissipate all the extra heat...
(Rant: why can't all processors be like the one in my iBook, designed for power efficiency as well as performance?)
Tedious Bloggy Stuff - hooray?
AMD is patenting this as a way of *getting around* of SOI disadvantages. SOI means silicon on insulator , which is in this case SiO2, which is also excellent thermal (not only electrical) insulator. AMD says that SiO2 conducts heat at least hundred times less than silicon.
What they are saying is that transistors on SOI might behave better, but they are certainly running hotter than their classic countepairs, since layer of SiO2 stands between them and the cooling system.
So AMD is proposing several schemes of embedding TEC device into the insulating layer in the silicon. This layer would:
1. Decrease overall thermal resisstance of the cooling path
2. When powered on, offer bigger thermal diferential, since it could cool embedded side of the TEC significantly below the cooler temperature.
It is unclear if they intend to use this on the whole chip, or just the especially hot areas...
Another flamebait. Processors are getting hotter and hotter. Intel used to generally do better in power consumption than AMD. But that period has ended years ago when Willamette core P4 was out. Have you compared the heatsink size of Prescott core P4 and that of AMD64?
I'd be more worried about the heat coming from the other side. My case is hot enough as is, no need to add to it
RTFA. Which "other side" are you talking about?
According to the wikipedia article, peltier effect is "the creation of heat difference from an eletric voltage." Namely, a peltier moves heat from one point to the other with the help of a electric field.
In this case, the peltier moves heat away from one side, the processor, to the other side, the heat sink, and the later then extracts the heat away. It helps conducting heat when the temperature difference between the CPU surface and the heatsink is higher. Although as a semiconductor itself, the peltier also generate heat. That is far less than the heat it takes away from the core and that heat is in effect taken away as well. So I don't know what you mean by "adding heat to it".
People who dislike China tend to mention Tiananmen Square a lot, but they always forget the Tank Man is also a Chinese.
I remember that SGI once used peltiers and they had to recall them because of failures due to corrosion due to condendsation because the device temperatures fell below the dew point.
Peltier is the guy who discovered the effect. A peltier junction (sometimes called a peltier cooler, which is a stupid name because it's also a heater) is the solid-state heat pump. A peltier junction is not called a peltier.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
So it should be just as reliable as the underlying processor, or any other piece of solid state electronics. This isn't really a cooling solution for a processor, you will still need those in the form of heatsinks, fans, heatpipes, etc. This is just a way to make sure heat can actually get to the processor surface and not get trapped in the middle. Comparing this use of the Peltier effect and traditional overclocking Peltier devices makes no sense - the only thing they share in common is their use of the same underlying physical phenomenon for heat transfer.
The real question in my mind is whether the fundamental efficiency limitations of Peltier junctions will make these processors even worse power-eating, heat dispersing beasts. But it sounds like there must be real heat buildup issues with Silicon On Insulator technology preventing them from ramping speeds up to the point where it would be competitive and ready for the mass market. And this sounds like a clever solution.
Consider a typical EER for a peltier air conditioner of 0.33 compared to 9.5 to 13 for a freon one. It's an interesting physical phenomenon, but a huge waste of energy.
While you are right that Peltier devices move heat from one side (the cool side) to the other (the hot side), they also generate a HECK of alot of waste heat themselves. So, great for cooling CPUs, but ussually used only in conjunction with other "extreme" coolers such as a watercooler or a bong. This is because, the more you cool the hot side, the colder the cold side will be. Peltiers are unique amoung extreme cooling in that they can actually cool lower (ussually far lower, when used in conjunction with other extreme technologies) than room temperature. Extreme OC'ers have sometimes (often!) destroyed their systems by letting water wapour condense onto the back of the sub-zero CPUs!
They're really quite amazing devices. I saw one when they first came onto the consumer PC market, about 4 years ago. Bloke had hooked it up to a car battery, and poured some drinking water onto it. Muttered something about the battery being flat, and a few seconds later scraped *ice* of the surface of the Peltier. They're also used to cool scientific equipment, such as CCD imaging devices in medium-to-high power telescopes.
DanielThe heat dissipated by a heat sink (with a fan or not) is a linear function of the temperature of the surface of the heat sink.
So if I have a CPU which puts out, say, 100W of heat, and I have a particular size of heatsink, the temperature of the heat sink will rise until the dissipation of the sink is 100W (or the chip melts). Say my heatsink dissipates 100W at 100C. (All numbers are made up.) Let it conduct perfectly, too, to simplify the discussion. If I put out more than 100W, the temperature of the heat sink will rise a little.
If my CPU is directly connected to the heat sink, its surface will be at 100C. If I have a Peltier that is configured to be 20 cooler on the cold side, then the CPU surface would be only 80C (assuming the Peltier consumes no power). So if my CPU is designed to run at 80C, the Peltier is fine; alternatively, I could get a bigger heat sink or a CPU fan.
So you could use the Peltier for overclocking, to reduce the needed size of the heatsink, or to remove the need for a fan. A given heatsink will be able to dissipate more heat for a given CPU temperature (since the CPU is cooler than the heatsink).
The problem right now is lack of surface area. My poor little switch is a prime example of this - the heatsink it shipped with had so much glue underneath it and so few fins that it was a nice toasty 70C! Heatpipes are no good also, because you need a large heat difference between the heat source and the outside, so you'll usually end up with about 50C on the cpu and 60C outside. Also, while peltiers are great, as mentioned many times before, they only MOVE heat, so guess what that means? That's right - you better have a heatsink the size of a radiator to dump all that heat out.