Compact Liquid Cooling Systems Roundup. Part I: Water on a Budget

The mass production closed-loop liquid-cooling systems once again would like to attempt dethroning air super-coolers. This time, however, we decided to gather them together in a series of large roundups. Today we will discuss five products from Corsair, Thermaltake and Zalman.

by Sergey Lepilov
03/06/2013 | 01:54 AM

There were a lot of compact serially manufactured liquid cooling systems for CPUs at the end of 2012. Nearly every manufacturer either announced or put on sale an inexpensive close-loop solution based on the Asetek platform.

 

We test liquid cooling systems on a regular basis and have already written a dozen reviews about them but in the last months the manufacturers have for some reason focused in that direction, flooding newsfeeds with their promotions of “new” liquid cooling products. Corsair and Thermaltake have updated their product line-ups whereas Zalman has released as many as three new coolers and showcased one more original product with some liquid inside. Cooler Master has announced four coolers of this class and NZXT, two coolers. Even Scythe has unveiled its very first serially made liquid cooling system.

It’s hard to explain the manufacturers’ increased interest towards such solutions. Our tests have repeatedly proved that compact liquid cooling systems have no advantages over today's top-end air coolers. Anyway, we cannot neglect this trend, especially as we've received over a dozen new products. You can see most of them in the photo below:

Considering the abundance of products to test, we will publish two reviews about them. The one you're reading now covers the less ambitious and less expensive products while the top-end ones will be discussed in our next review. So, let's get started.

Technical Specifications and Recommended Pricing


* - These are the radiator dimensions without the decorative panels.

All liquid-cooling systems reviewed today are closed-loop products. They come pre-filled with environmentally safe coolant and do not require servicing or adding any coolant later on. We are going to discuss them in alphabetical order.

Testing Participants

Corsair Hydro H55 Quiet

The junior model among the four Corsair products we’re going to test comes under the name of Hydro H55 Quiet. It is packed into a small cardboard box which is mostly white in color:

The cooler’s photo and model number can be seen on the front of the box. Its specifications and accessories are listed on the sides.

 

You can also find a list of supported platforms and diagrams comparing the cooler’s efficiency and noise level with those of an Intel boxed cooler.

There’s a soft cardboard box inside the external one. It has compartments for the cooler, fan, and accessories.

The accessories include retention and back plates, screws, plastic locks and metallic bushings – everything you need to install the Corsair Hydro H55 Quiet on any of the supported platforms. You also get some documentation:

The product is manufactured in China and costs $69.95. The warranty period is as long as 5 years.

Like the majority of the products covered in this review, the Corsair Hydro H55 Quiet is based on the popular Asetek 550LC platform which is inexpensive and easy to install and use. We first met this platform about 2 years ago in our review of the Antec KÜHLER H2O 620 and it hasn’t changed since. It still consists is a radiator with fan, two flexible pipes and a pump/waterblock combo:

The pipes are 300 mm long and have an outer diameter of 11 mm. They are squeezed tight by the input and output fittings, giving us no cause to worry about the reliability of these connections.

The aluminum radiator consists of eleven 1.5mm flat pipes with a corrugated ribbon in between.

The body of the radiator is only 17mm thick while its total thickness is 27 mm. There are threaded holes on both sides of the radiator for M4 screws to fasten fans. The overall dimensions of the radiator are 152x120x27 mm.

A sticker on one side shows a barcode and a power rating which is 3.9 watts.

Two nozzles for pipes stick out of the radiator on the other side:

The pump/waterblock unit on the other end of the pipes measures Ø72x28 mm.

Its performance is not specified (just like the performance of the other pumps in this review). It is connected to a mainboard’s 3-pin fan connector so we could easily measure its power consumption. We were surprised to find that this parameter varied between the coolers although 4 out of the 5 pumps were in fact identical. The pump of the Corsair Hydro H55 Quiet consumes 2.14 watts.

The pump/waterblock unit weighs about 220 grams. Its ceramic bearing is expected to serve for at least 50,000 hours and its noise level is within 26 to 37 dBA. That’s all the information we could get about it, though. “High-efficiency thermal interface” of obscure origin is pre-applied on the copper base of the waterblock:

The waterblock itself is hardly impressive in design. It features a microchannel structure with slim copper fins that are 6-7 mm in height.

This simple solution is actually quite efficient as we already know. The fittings can be turned around to make it easier to install the waterblock on the CPU and lay out the pipes inside the computer case.

The waterblock’s sole is finished well and has no irregularities like some air coolers. The thermal grease imprint on the CPU's heat-spreader is quite normal, even though not ideal:

 

The Corsair Hydro H55 Quiet is equipped with one 120mm fan that has a black frame and a glossy black impeller with seven sharp blades:

 

The fan has no speed regulation and rotates at a constant 1700 RPM. Its air flow is specified at 57 CFM and its noise, at 30.3 dBA. The service life of its sleeve bearing is 50,000 hours.

Like the other coolers in this test session, the Corsair Hydro H55 Quiet is compatible with all modern platforms from both Intel and AMD. You use the same retention frame to install the waterblock irrespective of your CPU:

First you insert plastic tips into its ends. The position of the tips depends on your CPU socket type.

Then you can either screw the frame to the CPU socket without fastening it tight:

Or you can use a fixing ring to secure the frame on the waterblock (you’ll see the photo in the descriptions of the other liquid cooling systems below).

After that you attach the radiator with fan to your computer case’s back or top panel.

With the mainboard installed, you mount the waterblock on your CPU, fastening the screws.

That’s all. You only need to connect the pump and fan to your mainboard’s connectors and start your computer up. The whole procedure of installing the Corsair Hydro H55 Quiet or similar coolers takes no more than 15 minutes.

Corsair Hydro H60 High Performance

The packaging of the Hydro H60 High Performance is the same in design and information content but silvery in color:

Specifications and a list of accessories in several languages can be found on the box. The cooler's performance and noise level are again compared against a standard boxed cooler from Intel.

 

 

The accessories differ, though. We can see a different type of fasteners and lots of screws, bushings and spacers:

Like its junior cousin, the Corsair Hydro H60 High Performance is manufactured in China and shipped with a 5-year warranty. It is somewhat more expensive at $79.95.

This cooler looks exactly like its predecessor: a pump with a waterblock, two pipes and a radiator.

 

We can see that the pipes are larger here, though. They are 14 rather than 11 mm in diameter to reduce resistance to the coolant and minimize its evaporation. The length of the pipes has remained the same at 310 mm.

 

The aluminum radiator is almost the same size (158x120x27) as the previous cooler’s, except that it’s larger in the pipes area. Moreover, the number of flat heat pipes has increased from 11 to 12 in the radiator while the spacing between them is diminished.

 

The radiator’s body is the same thickness as the Hydro H55 Quiet – 17 mm.

The sticker on the radiator doesn’t tell us any useful information.

 

The Corsair Hydro H60 High Performance differs from the other coolers in this test session in its pump/waterblock unit. The latter is not round but square and measures 63x63x32 mm:

 

Unfortunately, that’s the only difference we can point out. The scanty technical info suggests that neither the pump nor the waterblock differ in their interior design from those of the Corsair Hydro H55 Quiet. The power consumption of the pump is 1.96 watts according to our measurements. That’s the most economical pump in this test session.

There’s the same thermal interface on the waterblock’s base but it has a square rather than a round shape:

Thanks to that, the waterblock has contact with the full surface of the CPU’s heat-spreader, so the thermal grease imprint is almost perfect:

 

The Hydro H60 High Performance differs in its fan, too. It comes with a high-quality PWM-regulated Corsair SP120.

 

The slim radiator doesn’t need a high-pressure fan with a large dead zone beneath the motor, though. That's why we guess Corsair's AF120 would be a better choice with its stronger air flow and smaller motor (meaning a smaller dead zone underneath). And two such fans would just be perfect. Alas, the manufacturer has preferred another fan which is specified to have a peak speed of 2000 RPM, a noise level of 30.9 dBA and an air flow of 54 CFM.

The Hydro H60 High Performance is installed in the same way as the Corsair Hydro H55 Quiet. If you’ve got a mainboard for Intel CPUs, you insert threaded pins into the socket’s mounting holes.

  

Then, magnetic retention frames are installed on the waterblock and used to fasten the latter to the mainboard and CPU.

In fact, installing the Hydro H60 High Performance is even simpler and easier than installing the Hydro H55 Quiet.

Thermaltake Water 2.0 Performer (CLW0215)

Thermaltake’s products can be easily identified by their characteristic black-and-red packaging. The new Water 2.0 Performer is no exception:

There’s a photo of the product on the front of the box. Its name and key features are also indicated there. The sides of the box are informative, too:

  

The accessories are the same as those of the junior Corsair except for the fans and installation guide:

Manufactured in China, the cooler comes at a recommended price of $69.99. Its warranty period is 3 years.

Looking at this variation of the Asetek 550LC platform, we can see nothing particularly new, save for the label on the waterblock’s plastic cap:

We have the same radiator measuring 151x120x27 mm, two 300mm pipes (11 mm in diameter), and a Ø72?28mm pump/waterblock unit.

The aluminum radiator consists of 11 heat pipes and a corrugated ribbon in between.

The power rating we can read from the radiator’s sticker is 3.1 rather than 3.9 watts, though.

We don’t know the reason for this difference but according to our measurements the pump of the Thermaltake Water 2.0 Performer turned out to be the most voracious at 3.34 watts, although it doesn’t necessarily has the highest performance.

 

Everything is standard for the Asetek 550LC platform here, including the thermal grease imprints:

 

The single difference between the Water 2.0 Performance and other Asetek 550LC variation is about fans. The Thermaltake version has as many as two of them.

These are Power Logic PLA12025S12HH-LV fans that can also be seen on the Bigwater A80 and Frio Advanced. The speed of the fans is PWM-regulated in a range of 1200 to 2000 RPM. The peak air flow is specified at 83.4 CFM for each fan while the noise level is supposed to be no higher than 27.4 dBA.

The peak power consumption of the fans is up to 6 watts at 0.5 amperes which is twice the power consumption of the pump and only 1/40th the consumption of a Radeon HD 7970 GHz Edition, for example.

Here is the fixing ring we mentioned in our description of the Corsair Hydro H55 Quiet:

It is inserted from below the waterblock and helps fix the retention plate. This makes it even easier to install the water-block on the CPU.

The installation procedure is the same as with the Hydro H55 Quiet, for example, except for the second fan:

Now let’s move on to the next product.

Zalman LQ310

As opposed to Corsair and Thermaltake, Zalman is a new brand on the market of compact liquid cooling systems. You may remember the famous Zalman Reserator series but it didn't include small and affordable products. Now Zalman thinks it necessary to offer such solutions. There are three of them and they differ in price as well as efficiency. Let's first take a look at the junior model which is called Zalman LQ310.

The small and glossy box is perhaps the prettiest among the products included into this review. There’s an enlarged photo of the waterblock and fan and a description of the cooler's key features on the front of the box:

You can find all manner of information on the packaging, including detailed specs, a list of supported platforms and photos of key system components.

  

It’s the third time in this review that we see these accessories, actually:

Manufactured in China, the Zalman LQ310 costs $67.99 ($20 can be had back on a rebate card). The warranty period is 5 years.

What’s the difference between the Zalman LQ310 and other Asetek 550LC variations? Yes, in the design of the cap of its waterblock.

You will hardly find any other differences in these photos:

Here’s the radiator:

And the pump-waterblock unit:

 

And the thermal grease imprints:

 

The fan is original, though. It is a simple black thing with only seven blades and a diameter of 120 mm but it carries the Zalman logo:

 

The ZP1225ALM is PWM-regulated within a range of 900 to 2000 RPM. Its noise and air flow are not specified. The service life of the improved sleeve bearing is 50,000 hours or over 5.7 years of continuous operation.

We don’t think we need to describe the installation procedure once again. We’ve done it already above as well as in some previous reviews.

 

Zalman LQ315

The packaging of the Zalman LQ315 is designed in the same way as the previous model’s except for some variation in color.

We can find the same information on the box but some specifications are different, of course.

  

There’s a multilingual Zalman manual among the accessories:

The Zalman LQ315 costs $99.99 ($20 can be returned on a rebate card).

This cooler doesn’t differ from its junior cousin much. A close-loop system, it consists of two components connected with pipes:

On the other hand, this system has a thicker radiator and a taller waterblock.

The aluminum radiator is different indeed. It measures 153x120x38 mm, being 10 mm thicker than the other radiators in this review. It has only eight channels and a larger corrugated ribbon with perforation:

So, the distance between the edges of the band is larger by half compared to ordinary Asetek 550LC coolers, which should make the Zalman LQ315 more efficient at medium and low speeds of the fan despite the thicker radiator. We'll check this out shortly.

The height of the pump is increased from 28 to 32 mm:

 

This doesn’t mean any improvements in terms of coolant flow or something. The pump is taller just because it has to accommodate the integrated highlighting of the letter Z on its cap:

This marketing trick should appeal to fans of Zalman, Zorro, Godzilla, Jay-Z as well as to many other people who just like the 26th letter of the Latin alphabet.

Here’s a couple of photos showing the thermal grease imprints of this cooler:

 

The Zalman LQ315 has the same fan as the Zalman LQ310:

 

Its impeller is 113 mm in diameter whereas its motor is 42.5 mm in diameter. The 4-wire cable is 345 mm long. According to our measurements, the fan consumes up to 2.47 watts and has a startup voltage of 3.3 volts. That’s all we can tell you about it.

Besides the blue highlighting of the waterblock, the installed Zalman LQ315 can be identified by the captions on the sides of the radiator:

Now that we’ve checked out the key features of five compact liquid cooling systems, we can proceed to test them in action.

Testbed and Methods

The tests were performed inside a system case with the removed side panel. Our testbed was configured as follows:

For the primary tests and summary diagrams we overclocked our six-core processor with the clock generator frequency set at 125 MHz, the multiplier at 35x and “Load-Line Calibration” enabled to 4.375 GHz. The nominal processor Vcore was increased to 1.385 V in the mainboard BIOS. After that we tested the new cooling systems at even higher frequency and voltage settings. Turbo Boost was disabled during this test session, and Hyper-Threading technology was enabled to increase the heat dissipation. The memory voltage was at 1.65 V and its frequency was 2000 MHz with 9-11-10-28 timings. All other parameters available in the mainboard BIOS and related to CPU or memory overclocking remained unchanged.

All tests were performed under Windows 7 Ultimate x64 SP1 operating system. We used the following software during our test session:

So, the complete screenshot during the test session looks as follows:

The CPU was loaded with two consecutive LinX AVX test runs with the settings as indicated above. The stabilization period for the CPU temperature between the two test cycles was about 8-10 minutes. We took the maximum temperature of the hottest CPU core for the results charts. Moreover, we will also provide a table with the temperature readings for all cores including their average values.

The ambient temperature was checked next to the system case with an electronic thermometer with 0.1 °C precision that allows hourly monitoring of the temperature changes over the past 6 hours. The room temperature during our test session varied between 23.6-24.0°C.

The noise level of each cooler was measured between 1:00 and 3:00 AM in a closed room about 20 m2 big using CENTER-321 electronic noise meter. The noise level for each cooler was tested outside the system case when the only noise sources in the lab were the cooler and its fan. The noise meter was installed on a tripod and was always at a 150 mm distance from the cooler fan rotor. The tested cooling systems were placed at the edge of the desk on a sheet of polyurethane foam. The lowest noise reading our noise meter device can register is 29.8 dBA and the subjectively comfortable noise level in these testing conditions was around 36 dBA (do not mistake it for the low noise level). The fan(s) rotation speed was adjusted in the entire supported range using our in-house controller by changing the voltage with 0.5 V increment.

We are going to compare the cooling efficiency and noise levels of the participating systems against those of Phanteks PH-TC14PE with two default fans:

  

The rotation speed of all fans was controlled using our special controller with ±10 RPM precision.

Besides default fans that come with the cooling systems, we also tested them with two alternative 120 mm Corsair AF120 Performance Edition fans installed on the radiators for air intake/exhaust:

Unlike the photo above, during our cooling efficiency tests we installed all the fans in such a way that they could take the air from the outside of the system case and push it inside. According to our preliminary results, this airflow organization provided 2-3°C lower CPU temperatures than when the air was pushed out of the system case through the radiator. Moreover, this also provided better cooling to the heatsinks on the around-the-socket components. Nevertheless, I have to add that this isn’t a universal fan positioning for liquid-cooling systems and requires additional case fans exhausting the air out of the case (we used 120 mm and 200 mm fans).

Performance

Cooling Efficiency

As usual, we wanted to test each cooler throughout the entire speed range of its fan(s) with a step of 200 RPM but not all of them could cope with the load at low speeds (800, 1000 or even 1200 RPM). So, if you don’t see a result for a certain cooler, it means that cooler didn't pass the test at that speed. You can see the results in the table as well as in the diagram below:

Let’s first discuss the performance of the coolers in their default configurations. The advanced air cooler is superior to them, which is no surprise. Affordable liquid cooling systems cannot compete with it, although are comparable in price. Even in the quiet mode (at 800 RPM) the Phanteks PH-TC14PE is a mere 1°C worse at peak load than the two best liquid cooling systems at the maximum speed of their fans.

The mentioned systems are the Thermaltake Water 2.0 Performer and Zalman LQ315. The former is ahead thanks to its two 120mm fans while the latter has a thicker and larger radiator than the others. It must be noted that the Zalman LQ310 is only 1°C worse at the maximum speed and 2°C worse at 1600 RPM. When the speed of the fan is even lower, the junior LQ310 is 4°C worse than the LQ315. It doesn't cope with our overclocked CPU at all at 1000 RPM. The Thermaltake Water 2.0 Performer remains in the lead at the lower speeds of its two fans.

The two products from Corsair differ greatly in their performance. The Hydro H60 High Performance is 2°C inferior to the leaders at the maximum speed of its fan whereas the Hydro H55 Quiet is worse by 8 to 9°C and cannot keep the CPU stable at speeds below 1400 RPM. On the other hand, the Corsair Hydro H60 High Performance also looks worse than the products from Zalman and Thermaltake at 1600 and 1200 RPM. Thus, the different pump and waterblock design do not help Corsair against the competition in this test.

The picture changes when we install two 120mm Corsair AF120 Performance Edition fans on our liquid cooling systems. Even though 4 out of the 5 radiators are identical, like the pumps and waterblocks, the systems perform differently. The Zalman LQ315 with thicker radiator looks best at any fan speed. It is the only cooler to keep the CPU stable at 800 RPM. The Zalman LQ310 is only 1°C worse at 1600 and 1200 RPM but falls behind by 4°C at 1000 RPM.

The two Corsair products have improved their position. The additional fan makes them much more efficient. The Hydro H60 High Performance is a mere 1 or 2°C worse than the leaders at the maximum 1600 RPM as well as lower speeds. Even the Hydro H55 Quiet looks better with two fans.

The Thermaltake Water 2.0 Performer, the leader in its default configuration, has lost its ground. With two Corsair AF120 Performance Edition fans its result is worse by 3°C at 1600 RPM and by 4°C at 1200 RPM. The original fans from Thermaltake seem to be more optimal for this cooler than Corsair’s.

We also tried to overclock our CPU further with the most efficient liquid cooling systems in their default configurations but even the Thermaltake Water 2.0 Performer with its two fans working at max speed couldn’t cope with the 4.5GHz CPU at 1.43-1.44 volts, aborting the test with an error:

The Phanteks PH-TC14P? didn’t find this overclocking to be a problem even at higher voltage. It kept the CPU stable, never letting it get hotter than 77°C.

Now let’s see which of these coolers is the quietest.

Noise Level

It is the fan that is the noisiest component of each liquid cooling system in this review. Anyway, we checked out the pumps, too. We found them to be similar subjectively but the noise depends on the orientation of the pump and its operation. Right after you turn it on, the pump has air bubbles inside and produces a characteristic noise but quiets down in 1 or 2 minutes.

The products from Zalman and Thermaltake and the junior Corsair are equipped with identical pumps but the best one (in terms of noisiness) is in the Corsair H55 – 33.4 dBA. Next goes the Thermaltake with 34.2 dBA. The pumps of the Zalman systems are louder: 36.2 dBA with the LQ315 and 39.8 dBA with the LQ310. The Corsair H60 has the quietest pump whose noise is only 32.8 dBA according to our measurements.

We measured the noise level of each cooler’s fan throughout the entire speed range. The results are shown below:

As you can see, the five liquid cooling systems are similar to each other in terms of noise. It’s good that all of them are preferable to the Phanteks PH-TC14PE in this parameter, even though the latter is not a quiet air cooler itself. The Corsair Hydro H60 High Performance is the quietest, the high-quality Corsair SP120 fan proving its worth again. It is comfortable at speeds up to 1200 RPM. The Corsair Hydro H55 Quiet is similar in this respect. The two fans of the Zalman systems are also close to them in noisiness. Take note that, although the LQ310 and LQ315 have identical fans, their noise level is different due to variation between different samples of the same fan. The Thermaltake Water 2.0 Performer with two 120mm fans is the noisiest, remaining comfortable at speeds up to 1000 RPM.

Conclusion

Based on the same platform, the compact liquid cooling systems discussed in this review have a lot of things in common. As we've found out, they mostly differ in packaging, fans and price (the latter parameter doesn't vary much, though). Even their accessories differ in user manuals only. Could we expect them to differ much in performance? Of course, not. And that's exactly what we've got in our tests.

The Thermaltake Water 2.0 Performer with its two 120mm fans is the most efficient cooler we’ve tested today. However, it is the noisiest, too. This can only be justified by its low price. The Zalman LQ315 is second in performance. It is the only cooler in this review to have a thick radiator. It would certainly be the winner if it had a second fan. The Zalman LQ310 and the Corsair Hydro H60 High Performance share third place but the Hydro H60 has a remarkably quiet pump, so it is preferable over the Zalman system. And finally, the Corsair Hydro H55 Quiet is the least efficient of all, being perhaps the most simplistic of these systems in design.

Summing up this review, we have to remind you once again than none of these compact liquid cooling systems could surpass the best of air coolers. That’s why we cannot recommend purchasing, installing and using any of these systems over an efficient super air-cooler. Perhaps more advanced solutions will change our opinion. We’ll see in the next roundup.