Every day, computer hobbyists strive to squeeze every possible drop of potential out of their systems. These enthusiasts will eventually turn to overclocking, the mad art of running their components beyond their rated speed. In this eternal quest, the overclocker’s enemy is, as you all well know, heat. Every computer component uses electricity and producing no other work (kinetic movement, light, etc) all the energy consumed must be radiated as heat.
For years, systems engineers sought ways to remove this heat from the system in nearly every possible way, from the simplest to the most extreme. This is a brief article touching on the various methods of cooling your components. This may even be viewed as a history of cooling, seeing as each method builds on the benefits of the last. Every lesson learned here was first learned by the professionals, and then taken advantage of or learned anew by us, the hardcore hobbyists. The power-users and system abusers.
Convection and Radiation:
These are far and away the simplest methods of component cooling. No fans, no active components aiding the cooling whatsoever. Just vents. (And not even necessarily that.) Most enthusiasts and hobbyists these days would look at a system cooled by this method alone and say that it has no cooling whatsoever. Not technically true.
For instance, Nortel, maker of PBXs uses convective cooling in its low end 11C PBX. This system must dissipate about 500 watts of heat per cabinet, fully loaded, and the system must meet Nortel’s standard for stability and reliability. The boards within the system are mounted vertically; the cabinets are ventilated top and bottom. The heating of the air around the components causes the air to rise like in a chimney, drawing the heat out of the system. The natural convective currents of air are very effective here.
Heatsinks naturally improve the convective or radiative heat transfer of an object by increasing the surface area, and possibly by managing the airflow to improve convection. Another important consideration may be the material of the enclosing case. If the case is made of plastic, or any thermally insulative material, it will contain the heat as the components heat the air within the case. A case of aluminum or any thermally conductive material will shed heat to the surrounding air as the interior heats.
You should all be very familiar with fans for cooling. Cooling fans have been a part of PCs since the first. Rather than explain the history or operation of fans, or the thermodynamics of conduction, I’ll instead point out a few facts on improving your cooling.
Turbulent air cools better. Say, for sake of argument, you have a simple tube with a fan in the middle. The fan pulls air from one side of the tube, and blows into the other. If you have a hot component on the exhaust side of the fan, it will be more efficiently cooled than on the intake side.
This is because the air on the exhaust side of the fan is more turbulent. For lack of a better explanation, the loops and whorls of turbulent air moving across the surface pick up more heat. The effective surface area of the object is increased. (Actually, it was explained to me by saying the effective surface area of the air is increased.) The total volume of airflow remains the same, but turbulent air just cools better.
The corollary of this: Use directed flow from intake fans to cool specific areas or your hottest components, use exhaust fans to extract the heated air from the case. Most people have already learned this lesson.
Go with the flow. Obviously, your cooling system will work better if your airflow doesn’t have to fight the convection of the heated air in your system. Intake near the bottom, exhaust near the top. Try to maintain a uniformity of flow, a consistency of direction. Front to back, or back to front.
Don’t let your fans fight. Sibling rivalry it ain’t, but you still don’t want your fans duking it out. Fans pulling against each other or blowing into each other leave dead zones where there may be a reasonable amount of turbulence, but there’s very little actual airflow. This creates hotspots.
This includes cooling loops. I’ve seen people adding blowholes in pairs on their case side covers; one intake, one exhaust, each immediately next to the other. While this is assuredly very effective at cooling the area between the blowholes, it may create boundary conditions that actually impede the airflow through the case around that area.
Just like when supermarkets and department stores use those big vertical flow “air doors” to keep the cold side cold and the hot side hot. Try running both those blowholes as intakes, maybe even at half voltage. You may see your cards stay just as cool, and your overall case temperature drop.
I saved the heatsinks for after fans since most of us use them together, either as heatsink-fan combos, or with a fan directing flow over, around or through a heatsink. You all know that heatsinks do their jobs in two ways: they provide additional mass for the heat to propagate into, and they provide additional surface area by which to radiate or conduct heat.
Which is more important? It depends on your needs, but I don’t think any of us are too concerned about moderating temperature change rates. We just want to keep our friggin silicon cool, right? So we want surface area, and the more the better.
Turbulence can help, but only to a point. In our heatsinks, turbulence still improves the effectiveness of the heat transfer. We want turbulence in the airflow to help the air carry away more heat. But we don’t want so much turbulence that we reduce the total airflow through the ‘sink.
HP for instance, had a cast and machined heatsink, which featured tall, thin vanes, in the usual long straight rows. However, the vanes weren’t completely straight. Each one was twisted slightly out of line, into a little 15-degree zigzag formation. This increased the amount of turbulence in the airflow (as well as increasing the ‘sink’s surface area a small amount), without reducing the airflow.
Copyright Scott Morrison 2000
What are commonly referred to as peltier coolers are probably a little more accurately called thermoelectric coolers, seeing as the current technology in TE cooling has stepped away from Peltier’s original concept of just using the electrical junction of two dissimilar metals.
Modern TE coolers use semiconductors to pull the heat from one side to the other. The physics behind TE coolers are beyond what I want to get into here. Suffice to say that they’re a solid-state device that uses externally sourced current to move heat energy from one place to another.
Everybody reading this should already know the basics of peltier operation. A peltier is a thin sandwich of semiconductors between two slices of thermally conductive substrate, possibly aluminum or ceramic. A peltier has a hot side and a cold side. They don’t simply create some freaky “cold field” or something like that… the heat they remove from the cold side must be shed on the hot side, along with all the heat from the electrical energy the Peltier dissipates.
The heat from the current isn’t inconsiderable, either. It may actually be several times the energy removed from the cold side. So the peltier cools our silicon… but that means we’ve got to cool our peltier, right? Damn straight.
What does that mean? That means any system cooled by a peltier has got to have some damn fine cooling. Some people make do with a large enough heatsink-fan on their pelts, just extracting the heat energy to the air inside the case. Then they have to have enough airflow to change the air in the case often enough to keep the pelt from breathing hot air. Others go all out for water cooling (see below.)
On the plus size, there’s been news recently that some of those phy-sci brainy types have found out how to improve the efficiency of TE coolers pretty radically, between materials and design improvements. With some hope and luck, we should see facets of the improved technology on the street by this time next year. The reports promised a 20x improvement in efficiency, but I don’t think we’ll see all that for quite some time.
For now, though, properly sizing your pelt and your heat sink is critical. However, this isn’t the article to read if you want to learn how to do that. Plenty of other very good articles have already been written on the subject.
Ah, here we start getting into the meat of the matter. This is what most overclockers aspire to. A peltier and a watercooler… an overclocker’s dream. There are so many ways of implementing a water cooled system; there’s no way I can touch on it all. Instead, I’ll cover the basics of a typical system. (Besides, this article isn’t about the details…)
What’s the story behind watercooling? What’s so hot (or cool) about it?
Water is capable of carrying a lot more heat away from the surface of your silicon for a couple of reasons. Water has a higher coefficient of transfer, and a higher specific heat. Meaning it can pick up heat faster from a hot object, and it can hold more heat energy per specific volume. There are obvious disadvantages, too. Watercooling a PC is a complex, and possibly very risky maneuver.
What is needed for watercooling? The two heat-moving components are the waterblock, and the radiator. We’ll start where the heat does.
It’s your CPU you’re looking to cool (some have gone so far as to watercool their video chipset and even their system chipset.) Attached to the CPU is the waterblock. Typically, this is a metal block, machined to include internal channels allowing for the flow of water. Once again, surface area is important. You want as much metal in contact with the water as possible, to give the water as much opportunity as you can to extract heat. A few passes, zigzags, maybe even fins in the water flow will all improve the ability of your coolant to extract heat.
The radiator is just as crucial as the waterblock. Once you’ve picked that heat up from the silicon, you must do something with it, or you’re just going to be circulating hot water back to the silicon. There have been folks who’ve made do without a radiator at all, just a large enough reservoir of water. Often it’ll be ice water. But in a continuous running system, you’ll eventually heat the water to the point where it’s not cooling the system anymore.
Eventually you’ve got to get the heat back out of the water. (Okay, I’ve even heard of one guy just running water out of the tap, through the system and down the drain… not very environmentally friendly.) Most people extract the heat from the coolant with a radiator and a fan.
Between the waterblock and the radiator, of course, you’ve got tubing, you’ve got a pump, and you’ve got a reservoir. The most reliable pumps are magnetic impeller pumps. They’ve got no shaft seals to leak, and leakage is a watercooler’s nemesis. Then of course, you’ve got tubing and all the required joints, each one a potential new leak.
Currently, there isn’t much available in the way of professional quality kits or any such thing available for enthusiasts. While you can certainly find kits available, you’ll often find Rubbermaid containers more appropriate for storing leftovers in your fridge being used as reservoirs, and tranny oil coolers being used as radiators.
The waterblocks are often the best thing going about these kits, and they’re usually not the most efficient design. Not to worry though, as a technology, hobbyist watercooling (We need to watercool hobbyists? Damn!) is still in it’s infancy. Things can only get better, as we’ve seen with the introduction of a compact, large-surface area radiator marketed specifically for us (thanks, Senfru.) In addition, many people are coming up with better and better waterblock designs every day.
Now we get into the realm of the more exotic cooling systems. I’ll be touching on the next two systems briefly here; however, I’m already working on a pair of more in-depth articles to explain these… Keep reading…
Copyright Scott Morrison 2000
Direct Immersion Cooling:
I’ve heard rumors of hobbyists doing this. I’ve yet to see anyone actually put together a page on their project, and I don’t have any names of any individuals who’ve done this. I’ve got a good workup coming, and a pretty complete bill-of-materials worked out, so maybe I’ll just do it myself… but I digress.
In direct immersion cooling, the system components (at least the ones without moving parts) are submerged in a non-conductive liquid. The liquid is generally circulated within the enclosure to enhance cooling, and also pumped out of the enclosure for cooling. In the case of the Cray, these were large spill towers and a heat exchanger. I don’t think we need to go that far… a radiator or a peltier cooler with a massive enough heatsink should do the trick.
Within the enclosure, cooling of specific components could be enhanced by directing freshly cooled fluid to the hotspots with directed flow from jets or nozzles. Obviously you won’t be submerging your hard drives, CD-ROM drives, etc in this stuff. You probably won’t even want to use an internal modem with this, because most modems still use reed-switch relays, and the cooling fluid penetrates the relays and interferes with their operation.
The selection of a cooling fluid is the single most critical item. Obviously, you can’t just dunk your box in a tank full of water, or even glycol. However, there are families of chemicals called perfluorocarbons (PFCs) and their cousins, perfluorinated hydrocarbons (PFHCs), which are long chains of carbon and fluorine, which are ideal for this.
Their conductivity and dielectric strength are several orders of magnitude higher than that of air. While they conduct heat only about a quarter as well as water does, they’re still better than air by a factor of five or six times.
The advantages of this method of cooling should be pretty obvious. The high cooling potential of the cooling fluids remove the need for complex, dangerous to your system watercooling equipment, and you don’t have to have huge complexes of fans. One single large fan through a radiator could cool your entire system. Plus, my god man… think of the geek prestige you’d have with an immersion cooled system… your gawd-awful overclocked Athlon and Voodoo5 6000 occupying a big acrylic tank full of crystal clear fluid… I’m experiencing a bizarre tightening of the jeans just thinking about it.
There are disadvantages of course, and some technological hurdles to overcome as well… The properties of PFCS, for instance, provide some unique challenges.
It’s not exactly environmentally friendly… It’s not going to eat the ozone layer for breakfast or anything; in fact it’s non-reactive with ozone. But the chemicals we’re talking about evaporate relatively easily, and they hang out in the atmosphere for a long time. The biologically non-reactive… but they’re also big long chains of carbon and fluorine. They’re greenhouse gases… that is, they promote global warming, and they do it almost as much as the hated CFCs (which eat ozone AND help hold in the heat.) PFHCs don’t have this problem at all, decaying quickly in the atmosphere.
Another disadvantage is the fact that both PFCs and PFHCs have virtually no surface tension, and have an incredibly low viscosity. A pipe joint or pump seal that could hold water at 10 atmospheres could leak these fluids like a cheap showerhead. These are all issues that can be overcome, though, through careful sealing of the enclosure, the use of magnetic impeller pumps and low pressures.
An issue unique to using PFHCs is that it’s a solublizing hydrocarbon. (Solublizing… is that even a word? I read it in a FAQ covering PFHCs.) That means other hydrocarbons are soluble in PFHC. If you were to use some hydrocarbon based plastics, or plastics with hydrocarbon plasticizers, it slowly leaches away the plasticizer, or decays the plastic. Once the fluid is saturated with the plastics, it may redeposit it on surfaces, or worse, builds up phenomenal static charges until it discharges through a conductive path.
However, all these disadvantages can be overcome. 3M, a manufacturer of these chemicals has gone to great lengths to research the issues and find workarounds, and they have great information on their website about the applications of these fluids.
For instance, 3M lists case studies of applications where just the component to be cooled was encased in a liquid cooling jacket, allowing the cooling fluid to be in direct contact with the device to be cooled. This allows very effective cooling of the device without the worries of outgassing of the cooling fluid’s vapor. The drawback here is the most hobbyists don’t have access to the machinery and materials to build the hermetic case around their PIII or Athlon.
The final disadvantage? Probably price. I don’t have pricing yet on the chemicals, and building your PC into a welded acrylic aquarium probably isn’t a cheap proposition either.
Ah… after spending a couple weeks researching direct immersion cooling, I was put off by the possible expense and effort involved. Originating, apparently, in efforts to passively move heat around in spacecraft and reactors, etc., the elegant simplicity of heat pipes struck me as a more elegant solution to my cooling needs.
Commonly used in laptops, occasionally in workstations and other high-end boxes, heat pipes are simplicity itself. A heat pipe is pretty much exactly what it sounds like.
A heat pipe is a hermetically sealed pipe, usually cylindrical, containing a heat conducting fluid. The interior wall is lined with a wicking substance, and the remainder of the heat pipe is simply hollow, a channel for the cooling fluid. In the case of heat pipes commonly used in PCs, the fluid is simply purified water, and the wicking medium is just copper braid, similar to that found in coax insulation, or, more closely, desoldering braid.
Although I’ve found some descriptions of single phase heat pipes, where the fluid does not boil, but simply expands under heating, it seems unlikely you’ll find these in PC cooling applications. In two phase, commonly called phase change or vapor phase operation, the medium boils. Either way, operation is essentially the same. One end of the heat pipe is the hot end or evaporator section; the rest of the heat pipe is the cool, or condenser section. (Yup, just like the evaporator and condenser portions of an air conditioner.)
When heat is injected into the hot section, it creates a pressure differential in the cooling medium, and the fluid flows away from the hot section. I’m assured that the cooling medium doesn’t lose much heat on the way because there’s very little change in pressure. When it reaches the far end of the pipe, where the pipe is typically embedded in a heat sink, the fluids shed heat as the fluid is condenses into the wicking medium, and the fluid travels back to the hot end through the wick, and the process begins again.
The advantages of heat pipe cooling systems? They can actively move heat away from a hot area, spreading it over a larger area for dissipation, and unlike a peltier, they don’t require additional energy to do so. Heat pipes can even be bent to the desired shape; in fact this is a necessary part of some applications, where the pipe is bent to allow more of the pipe to be in contact with the heat source or sink.
A typical heat pipe cooling system would consist of a cold plate attached to the device to be cooled. Embedded in the cold plate would be one end of one or more heat pipes. The heat pipes would be arranged to carry heat away to one or more oversized heat sinks, in which the pipes would also be embedded. Properly designed, a heat pipe system could effectively cool even a peltier without the need for water cooling.
Currently, overclocking enthusiasts are indulging in some rather “brute force” cooling methods. In the Darwinian progression of cooling technology, it seems that, at least for those that need a great deal of cooling, we’ll have a divergence of paths… One path continuing down the road of brute force into the realm of immersion cooling (somebody’s likely to do this just for the hell of it.); the other path leading down the road of increasingly elegant solutions, possibly combinations of the next generation of thermoelectrics and heat pipes, etc.
Sources for additional information
3M Thermal Management Fluids
Indek – manufacturer of heat pipes and compact cooling products:
Noren Products – manufacturer of heat pipes and a wide range of cooling products. These guys are the shiznit when it comes to heat pipes.
Copyright Scott Morrison 2000