Table of Contents
Introduction
But first, a brief explanation of what a exactly what a peltier is: A peltier uses a TEC (thermo-electric cooler) to act like a “heatpump” to draw heat away from the object it is cooling. They work by applying a voltage across a thermocouple, which causes a temperature difference. More information on peltier coolers can be found here. By using a powerful enough peltier cooler, you can achieve below-ambient temperatures on the object you are cooling.
However, there are a few drawbacks. First, peltier coolers are fairly expensive. CoolIT’s cheapest cooler to offer peltiers retails for about $160. Second, they require quite a bit of power. A single peltier cooler used by CoolIT can draw up to 15watts. With their high-end cooler using 6 peltiers, the cooler needs more than 80watts of power. Thats almost as much power draw as the CPU they are cooling. Third, peltiers actually add heat to your system. That 15watts of power has to go somewhere, and it is converted into heat. The exciting thing about the CoolIT coolers is that by putting the peltiers right by the exhaust fan for the system, that additional heat does not affect your computer hardware, because it is immediately vented out the back of the computer.
Lastly, I need to caution that using peltier coolers can actually be dangerous for your computer hardware. Any time you are working with below-ambient temperatures, you have to worry about condensation — just like a glass of cold water on a hot day. When we left the two coolers featuring peltiers running on a bench, the coolers accumulated condensation after about 20 minutes. While the heat from a CPU would prevent this from happening (the heat would not allow these coolers to operate at such a low temperature), a system that does not POST properly can potentially allow condensation to happen. When a system does not POST, the CPU is usually not being given very much voltage, which means it will not be providing any heat. This would allow the cooler to eventually get below the ambient temperature. After a few minutes, condensation can form on the cooler and drip onto your components, causing sparks to fly!
Test Setup
|
|
The ambient temperature was always within one degree of 24 degrees Celsius. After installing the cooler, the system was allowed to warm up for 45 minutes before our temperature logging was started. To put the system under as much load as possible, we ran an instance of Prime95 on each core of the CPU with a single loop of 3DMark06 running at the same time. After the stress test was complete, the system was given two minutes to cool down before the temperature logging was stopped.
The Coolers
The Arctic Cooler Freezer 7 is easily our biggest seller. It is light, easy to mount, and does a decent job cooling. The fan mounts using four rubber connectors to reduce fan vibrations. On CPU’s faster than the 3Ghz, it does start to have trouble cooling the CPUs. On our test system, the CPU got very close to 80c, which with Intel CPUs is the point where the CPU starts to throttle back to protect itself from overheating. This is currently the quietest CPU cooler we carry. | |
The Puget Hydro CL1 is a self-contained CPU liquid cooling solution. It has a lot of the benefits of liquid-cooling: better temperatures (which allows for overclocking) and fairly quiet operation, but also has the benefits of air-cooling: the ease of installation and maintenance. The radiator screws in place over the rear 120mm fan in your case. By default, the rear fan is usually left as an exhaust. Through our testing, we have found that in cases like the Antec P182 (which has a top exhaust fan), the cooler performs much better by reversing the rear fan and turning it into an intake. Since the case has a top exhaust fan, the impact on system temperatures is acceptable. In cases that do not have a top exhaust fan, this cooler will still out-perform the Freezer 7, but only by 5-6 degrees Celsius. | |
This is another self-contained CPU liquid cooling solution, which is very similar to the Puget Hydro CL1. The radiator size is nearly identical. The only real difference between the two coolers is that on the Hydro CL1, the pump is located on the CPU block, whereas on the CoolIT Pure, the pump is separate from the CPU block. Also just like the Hydro CL1, this cooler does not out-perform the Freezer 7 by very much until the fan is turned into an intake. Once the fan is changed to an intake, it performs almost exactly the same as the Hydro CL1. However, this cooler is actually too large to fit into an Antec P182, due to the added plate needed to mount the pump. It would fit fine in a case like a Lian-Li PC7H, but since that case does not have a 120mm top exhaust fan, the cooling potential is reduced. The cooler can be forced into a P182, but the side panel shows a noticeable bow at the rear. The fan in this cooler by default is plugged into a circuit board that provides power to the pump, blue LED, and the fan. In that configuration, this cooler offers no fan ramping. However, it is easy enough to bypass the circuit board and plug the fan into the motherboard CPU fan header. | |
This cooler features three peltier coolers attached to a closed-loop liquid cooling solution. With the addition of peltiers, this cooler can only be configured as an exhaust. The fan is a smaller 92mm fan, even though the unit mounts to a 120mm fan mount. There is a switch for Low, Medium, and High which is supposed to control the fan speed, but our sample appeared to be defective, as the switch did nothing. The unit was stuck on the high setting. The peltiers also do not ramp, allowing for the possibility of condensation forming on the heatsink. Also like the CoolIT Pure, this cooler does not fit well into a P180. If we rotated the radiator 90degrees, putting the LED panel towards the top of the case, we can just barely get it to fit. However, one of the peltiers still hits the side panel, causing a noticeable bow on the case side panel. This cooler is very loud. Just from listening, we estimated that it is about 2-2.5 times louder than the Hydro CL1 cooler, or about 3 times louder than the Freezer 7. It is definitely the loudest cooler we have ever installed into a computer. Allowing the fan to ramp would help, at the cost of performance. Performance-wise, this cooler did not out-perform the Hydro CL1 by more than a few degrees. Since it is more expensive, louder, and nearly the same performance as a Hydro CL1, I cannot recommend this cooler. | |
The big brother of the CoolIT Eliminator, this cooler has 6 Peltier coolers attached to a closed-loop liquid cooling solution. This unit uses a 120mm high-flow fan. One large advantage of this cooler is that it features ramping for both the fan and the peltiers according to the fluid temperature. This allows for the cooler to ensure that the fluid is always above ambient temperature, thus preventing the cooler from collecting condensation. Unlike the other CoolIT coolers, this cooler will fit into an Antec P180. Like the CoolIT Eliminator, this cooler is very loud. It is roughly the same noise level as the Eliminator. According to the specs, it is a few dBa louder, but it was hard for us to tell any difference. This cooler features a USB interface to the computer, allowing for extra configuration of the cooler. From the software, you can monitor CPU and fluid temperatures, as well as the power draw (performance) of the peltiers. You can set min/max temperature ranges as well as min/max peltier power draw. You can also select to have the system shut down if the pump fails. This cooler did out-perform all the other coolers, but at $320, it is easily the most expensive as well. From our records, this cooler performs about the same as a full liquid cooled setup. It is cheaper than a liquid cooled system, but is also quite a bit louder. |
Temperature Graphs
| Freezer 7 | Hydro CL1 |
| CoolIT Pure (default) | CoolIT Pure (Intake) |
| CoolIT Eliminator | CoolIT Freezone |
A few of the graphs show a large spike in System temperature at the beginning of the graph. That was artificially added so that the scale of each graph would be the same.
Temperature Comparison
CPU Temperature Comparison
This graph is a close-up of only the CPU temperatures. The first thing to note is that the CoolIT Pure actually performed worse than a Freezer 7 in its default configuration. Turning the fan into an intake like we do with the Hydro CL1 helps immensely. In that configuration, the Pure does out-perform the Hydro CL1, but only by a few degrees. The Eliminator cooler performs almost identically to the Pure when configured as an intake. The CoolIT Freezone is the only cooler to really outperform either the Hydro CL1 or the Freezer 7. Under load, the Freezone lowered the CPU temperatures by about 17 degrees compared to the Freezer 7.
Conclusion
Of all the CoolIT coolers we tested, the Freezone is the only we would consider. This cooler could work well for those who want the performance of liquid cooling at a lower cost (and with less maintenance), as long as noise is not an issue.