Closed Loop AIO Liquid Coolers: 14-way Mega Roundup Review
by E. Fylladitakis on February 12, 2014 7:00 AM ESTTesting Methodology
Although the testing of a cooler appears to be a simple task, that could not be much further from the truth. Proper thermal testing cannot be performed with a cooler mounted on a single chip, for multiple reasons. Some of these reasons include the instability of the thermal load and the inability to fully control and/or monitor it, as well as the inaccuracy of the chip-integrated sensors. It is also impossible to compare results taken on different chips, let alone entirely different systems, which is a great problem when testing computer coolers, as the hardware changes every several months. Finally, testing a cooler on a typical system prevents the tester from assessing the most vital characteristic of a cooler, its absolute thermal resistance.
The absolute thermal resistance defines the absolute performance of a heatsink by indicating the temperature rise per unit of power, in our case in degrees Celsius per Watt (°C/W). In layman's terms, if the thermal resistance of a heatsink is known, the user can assess the highest possible temperature rise of a chip over ambient by simply multiplying the maximum thermal design power (TDP) rating of the chip with it. Extracting the absolute thermal resistance of a cooler however is no simple task, as the load has to be perfectly even, with the ability to vary the load, as the thermal resistance also varies depending on the magnitude of the thermal load. Therefore, even if it were possible to assess the thermal resistance of a cooler while it is mounted on a working chip, it would not suffice, as a large change of the thermal load can yield very different results.
Appropriate thermal testing requires the creation of a proper testing station and the use of laboratory-grade equipment. Therefore, we created a thermal testing platform with a fully controllable thermal energy source that may be used to test any kind of cooler, regardless of its design or compatibility. The thermal cartridge inside the core of our testing station can have its power adjusted between 60W and 340W, in 2W increments (and it never throttles). Furthermore, monitoring and logging of the testing process via software minimizes the possibility of human errors during testing. A multifunction data acquisition module (DAQ) is responsible for the automatic or the manual control of the testing equipment, the acquisition of the ambient and in-core temperatures via PT100 sensors, the logging of the test results, and the mathematical extraction of performance figures.
Finally, as noise measurements are a bit tricky, we're measuring these manually. Fans can have significant variations in speed from their rated values, thus their actual speed during the thermal testing is acquired via a laser tachometer. The fans (and pumps, when applicable) are powered via an adjustable, fanless desktop DC power supply and noise measurements are being taken 1m away from the cooler, in a straight line ahead from its fan engine. At this point we should also note that the decibel scale is logarithmic, which means that roughly every 3 dB(A) the sound pressure doubles. Therefore, the difference of sound pressure between 30 dB(A) and 60 dB(A) is not "twice as much" but nearly a thousand times greater. The table below should help you cross-reference our test results with real-life situations.
| Noise Level Reference Values | |
| <35dB(A) | Virtually inaudible |
| 35-38dB(A) | Very quiet (whisper) |
| 38-40dB(A) | Quiet (slight humming) |
| 40-44dB(A) | Normal (humming noise, comfortable level) |
| 44-47dB(A) | Loud* (strong aerodynamic noise) |
| 47-50dB(A) | Very loud (strong whining noise) |
| 50-54dB(A) | Extremely loud (level equivalent to a ≈1500W vacuum cleaner) |
| >54dB(A) * | Intolerable for home/office use; special applications only. |
* Noise levels above this are not suggested for daily use
Facebook
Twitter
RSS
139 Comments
View All Comments
faster - Wednesday, February 12, 2014 - link
I want a closed loop system for my video card!My GTX780 is much louder than my CPU fan.
Liquid cooling was supposed to be how one obtained high performance quiet computing. Air cooling solutions should not be able to compete with a liquid cooling solution in the same environment, but it seems they do. How is that?
blanarahul - Wednesday, February 12, 2014 - link
"I want a closed loop system for my video card!"This. All the stupid OEMs want to disgruntle consumers. We already have more than enough great CPU coolers, but very few great GPU coolers. All I want is a card that is NOT pre-overclocked (but can be overclocked at my whims ;) ) and comes with a all-in-one liquid cooling solution. Is that too much to ask??
E.Fyll - Wednesday, February 12, 2014 - link
Oh, that is coming, soon. Stay tuned, should be online within a couple of weeks. :)Dribble - Wednesday, February 12, 2014 - link
True, for cpu's you don't need that big a cooler - all these lower power Intel cpu's don't pull 200W even overclocked. A big air cooler is sufficient. It'll work as well as a single 120mm fan radiator water cooler but is cheaper and more reliable.However graphics cards are another matter altogether - they pump out huge amounts of heat.
Hence either I WC my graphics card in which case I might as well get a system that can WC my cpu too, or I just stick to air cooling.
BuddhaBum44 - Wednesday, February 12, 2014 - link
You can always get the Kraken X40 and the bracket they make for 780s: https://www.nzxt.com/product/detail/138-kraken-g10...bj_murphy - Wednesday, February 12, 2014 - link
Solid review with some good information. I've been waiting for a "compendium" of sorts to link to people, explaining which closed loop coolers are the best. Thanks E. Fylladitakis, looking forward to more great articles!doggghouse - Wednesday, February 12, 2014 - link
What is a realistic load for a CPU? My 4770K has a max TDP of 84W... and I see a 3960X has a max of 130W. Are there actual CPUs that have anything above that, like 200W - 340W? If not, does it make sense to include those loads in the average thermal resistance, since these AIO coolers are going to be applied to a CPU, not to a synthetic load...?E.Fyll - Wednesday, February 12, 2014 - link
It depends on the CPU, of course. An overclocked CPU can easily surpass their max TDP rating. High thermal loads are useful for the extraction of proper thermal resistance ratings, plus they are easily reachable by modern GPUs (and GPUs are relevant, especially with AIO coolers; you'll see why soon enough). Of course, if you know the power requirements of your current CPU (if not overclocked, about 75% of its TDP), you can easily check the graph closest to it.dragosmp - Wednesday, February 12, 2014 - link
Hi,Great review, I like the methodology. It is nice to see all coolers tested with a constant load that is subject to much less randomness than a CPU power output.
I have two questions:
*is it possible to test all coolers at a certain noise level like 40dB +/-0.5dB; 7V testing is not that relevant for a cooler that is silent @10V, why would anyone silence it even further, save electricity?
*could you provide an order of magnitude of what clock speed and voltage a CPU would need to be at to achieve 340W/150W...etc. It would be useful to get our bearings vs the real world. A chart would be nice a bit like this:
.....................150W.................250W
Haswell DC....4.6GHz/1.25V......
Haswell QC....4.2GHz/1.27V......
FX83xx OC....4.1GHz/1.35V......
The reason for the last remark is that buying decisions are made also with cost in mind. One may think: I have max 30°C Tcase, 4.5GHz Haswell, what is the thermal conductivity I would need so the CPU never passes 70°C? Answer ==> review (maybe not the best cooler, maybe not the most expensive...). I have bought windows for my house like that.
E.Fyll - Wednesday, February 12, 2014 - link
40 dB(A) is not really "silent". I would rate <35 dB(A) as silent and still I can notice that in a very quiet room. 40 dB(A) is a slight humming noise, fairly quiet and most people just ignore it, yet it is easily noticeable. It is an interesting idea but that is not really possible when not all products can do at least 40 dB(A) and not practical, as the motherboard does not read the sound level, it just adjusts the voltage. 7 Volts are just high enough to ensure that (almost) every fan will start and about the same voltage as most motherboard will apply in their "quiet" mode. About the CPUs, I cannot do that as that would require me buying and testing every single CPU, which is not possible. Besides, every CPU is unique and the energy consumption also depends on several settings when overclocked, so it could easily bring misleading results. A single different setting can cause a massive change on consumption at the same exact frequency. For instance, a i7-3820 at 4.4 GHz consumes nearly 20% more energy with its voltage upped by 0.1V. It truly is a very interesting idea but a great deal of data and testing is required to create a proper database.