by Sergey Lepilov
02/17/2010 | 01:54 PM
A person who is not familiar with the CPU air-cooling field will hardly understand anything from the title of this article, and medical doctors will even doubt the sanity of the patient who may have written this phrase. Only overclockers with at least four years of experience will understand what we are going to talk about today. Before we start, we would like to explain (just in case) that among the existing CPU coolers there are really legendary solutions such as Scythe Ninja, Thermalright Ultra-120 eXtreme and Zalman CNPS7xxx. Of course, the famous Thermaltake Big Typhoon that was launched back in early 2005 is also one of them. Since then it has become a beloved favorite of many overclocking fans, some of whom have been using this cooler ever since. Why do we go back to the Big Typhoon in the introduction to our review of Propeller 120 cooler from Tuniq Co. Ltd.? Well, you will understand that shortly.
Tuniq Propeller 120 comes in a large box with a plastic carry handle and a colorful external slip-cover made of thin cardboard. There is a huge aircraft carrier on the front side of the box with the cooler photo and list of key features displayed as icons right below it:
It is hard to tell what the connection between Tuniq Propeller 120 and aircraft carrier really is. Maybe the heatsink arrays that seem to be floating on air resemble the front of the aircraft carrier in some way, or maybe there is some other reason for that, but nevertheless, this military ship is the one put on the front of the new cooler box.
The remaining sides of the cooler box are dedicated to more traditional computer related topics:
Tuniq Propeller 120 comes bundled with a universal backplate, two retention kits for Intel and AMD platforms, variable fan rotation speed controller in the form of a rear panel bracket, Tuniq TX-3 thermal paste optimized for coolers using core-contact technology and a booklet with installation instructions:
The cooler is made in China and is priced at $59, which is quite a lot for a top cooler using core-contact technology.
Tuniq Propeller 120 measures 128x127.5x125 mm and weighs 590 g without the fan. The new solution looks very attractive:
The cooler is built on four copper nickel-plated heatpipes with a diameter of 8 mm. They go through the cooler base, forming the bigger part of it. Core-contact technology in Tuniq’s interpretation implies that the heatpipes contact the core directly, although in reality they contact the heat-spreader, because there no processors with exposed cores out there. However, Tuniq didn’t dare call this technology “Heatspreader-contact” and that was definitely the right thing to do.
Heatpipes hold two heatsink arrays of aluminum plates, each featuring 51 plate. Each plate is 0.45 mm thick and they are spaced out at 2.0 mm from one another.
The plates are shaped as rectangular trapezoids 5.95 cm tall, with 2.0 cm top and 5.2 cm bottom sides. The calculated effective surface of Tuniq Propeller 120 cooler heatsink is pretty modest and makes only 4,980 cm2. For example, the best air cooler out there, Noctua NH-D14, has a heatsink with over 12,000 cm2 effective surface area, while the best top-cooler from Scyth, Zipang 2, has a heatsink measuring 5,890 cm2.
The surface of the heatsink plates is exactly the same as that of Tuniq Tower 120 Extreme: it is all covered in small bumps that create extra turbulence for the airflow and increase the heat transfer efficiency:
The design of Tuniq Propeller 120, namely its two-array heatsink, reminds us of a formerly popular Thermaltake Big Typhoon that was extremely efficient compared with its competitors back in the days. The only constructive difference is that the heatsink arrays of Tuniq Propeller 120 are pierced with four 8 mm heatpipes that go through the cooler base, while Thermaltake Big Typhoon has a total of six 6 mm heatpipes coming out of its base on both sides. However, it is impossible to introduce this type of base in a core-contact cooler that is why the Tuniq solution has only four heatpipes that go through the base rather than come out of it.
At the base, the heatpipes sit in an aluminum plate that flows into a small heatsink taking the heat off the top of the heatpipes in the base. The gaps between the heatpipes here are 1.8 mm. their surface is very even and very well finished:
If you read our review of Alpenföhn Nordwand cooler based on five 8 mm heatpipes, you should remember that the main issue with that cooler was the improper contact between the heatpipes and the LGA1366 CPU heat-spreader: only three central heatpipes out of five were in fact in full contact with the processor. Since Tuniq Propeller 120 only has four 8 mm heatpipes, it doesn’t suffer from the same issue:
As you see, all four heatpipes contact with the processor heat-spreader, which means that they will work more effectively than five heatpipes like that, with only three and a half being in direct contact with the processor. Ideally, we would like to find and test a cooler using heat-pipe direct touch or core-contact technology that would have no aluminum inserts between the heatpipes at all, but since there hasn’t been a single cooler like that yet, there must be an issue of ensuring that heatpipes stay together in the base. Hopefully, the manufacturers will soon come out with a solution to this matter.
Tuniq Propeller 120 is equipped with a single 120 mm fan with clear plastic frame and blades:
The fan specifications are identical to those of the Tuniq Tower 120 Extreme fan: 1000-2000 RPM rotation speed, 16-20 dBA noise, 90.65 CFM maximum airflow. The cooler comes with a variable controller that will allow you to adjust the fan rotation speed. It can be installed onto the case rear panel. the fan uses a magnetic fluid dynamic bearing (MFDB) that should last 50,000 hours or over 5.7 years.
The fan is attached to the heatsink with four soft silicone mounts absorbing vibrations and thus lowering the level of generated noise:
Besides, these mounts raise the fan by about 3-4 mm above the heatsink plates, which reduces the airflow resistance.
There are four blue LEDs in the corners of the fan frame:
They create a nice-looking blue glow inside the system case:
Tuniq Propeller 120 is compatible with all contemporary AMD and Intel processors. It uses the same type of retention as Tuniq Tower 120 Extreme and installs in exactly the same way. To install the new cooler onto the processor you must use the enclosed universal backplate with spindles that screw into it first:
After that you install the retention plate onto these spindles and tighten it with four spring thumb-screws:
There are slits for a slotted screw-driver at the top of these screws, but you won’t be able to use it, because it can’t reach the screw tops beneath the heatsink. However, you will be able to tighten them up just with your fingers, without any additional tools. If your fingers are not strong enough, then you could also use pliers, just make sure not to overdo it.
The new cooler can be installed in any of the two possible positions without interfering with the heatsinks over voltage regulator components or memory DIMMs.there is no mention of the preferable installation position in the Tuniq Propeller 120 manual that is why we first tested the cooler with the heatpipes going along the processor heat-spreader parallel to the memory DIMMs:
As we have expected, this position turned out less effective than when the cooler heatpipes go across the LGA1366 processor heat-spreader perpendicular to the memory modules. However, we were really amazed with the temperature difference in this case that reached 9°C under peak load in favor of the second position. It must be the overall heatpipes orientation rather than their position in relation to the processor heat-spreader that mattered most in this case, because in the first case the heatpipes ends in the right heatsink array point down, which negatively affects their efficiency. However, when we turn the cooler by 90°, all heatpipes are horizontal and work with equal efficiency. By the way, during the test session, Tuniq Propeller 120 heatsink warms up so much that you can only touch it for 2-3 seconds, no more than that.
The default fan can be easily removed from the silicone mounts and replaced with a 120 mm Noctua NF-P14 fan, for instance:
We are also going to test the new cooler in this configuration during our performance session.
We are going to test the cooling efficiency of our today’s testing participants and their competitors in the following testbed:
During this test session we managed to overclock our 45nm quad-core processor (with polished heat-spreader) with the multiplier set at 21x and “Load-Line Calibration” enabled to 3.8 GHz using the weakest cooler of the testing participants. The nominal processor Vcore was increased to 1.3 V in the mainboard BIOS.
The memory voltage was at 1.6 V and its frequency was around 1.43 GHz (7-7-7-14_1T timings). All other parameters available in the mainboard BIOS and connected with CPU or memory overclocking remained unchanged (set to Auto).
All tests were performed under Windows 7 RTM x64 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 Linpack 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 processor core of the four for the results charts. The ambient temperature was checked next to the system case with an electronic thermometer with 0.1 °C precision that allows monitoring the temperature changes over the past 6 hours. During our test session room temperature was at 24.8-25.2 °C.
For comparison purposes we also included the best top-cooler - XILENCE Black Hawk Copper and the best tower-cooler – Noctua NH-D14:
“Black Hawk” was tested with its default fan, and Noctua NH-D14 was tested with one 140 mm Noctua NF-P14 fan installed between the heatsink arrays. Both coolers were tested in quiet mode and at maximum fan rotation speed. The fan rotation speeds were set with the help of our controller with ±10 RPM accuracy.
Before we compare Tuniq Propeller 120 against the two best cooling solutions out there, let’s check out the let’s check out the dependence of its cooling efficiency on the rotation speed of its default cooling fan and the use of a larger 140 mm Noctua NF-P14 fan:
.png)
We can see that the dependence of Tuniq Propeller 120 efficiency on the fan rotation speed is extremely high. The higher is the fan speed, the better the cooler works. At 600 RPM the default fan is barely blowing that is why the cooler can’t cope even with the lowest CPU overclocking letting it past the overheating protection boundaries. In the interval between 800 and 1400 RPM the peak processor temperature lowers by 10°C. Here I have to say that the Tuniq Propeller 120 heatsink arrays are not so dense (the gaps between the plates are 2 mm), so this noticeable temperature drop should result from the fact that the heatsink is relatively small for 8 mm heatpipes, rather than from the ability to overcome high heatsink resistance. The same conclusion can be drawn from the results obtained with a 140 mm Noctua NF-P14 fan: the cooler works 5°C better with this fan than with the default one at 810 RPM and 3°C better at 1200 RPM.
Now let’s check out the comparative results chart:
.png)
Unfortunately, Tuniq Propeller 120 gets defeated by the other two testing participants. New Tuniq cooler with its default fan at 1940 RPM cools the overclocked processor 4°C worse than XILENCE Black Hawk Copper in quiet mode at 1000 RPM. With the same 140 mm fan Tuniq Propeller 120 works a little better, but the lag is still quite significant. Noctua NH-D14 with the same exact fan at the same exact speed is 20°C more efficient in quiet mode at 810 RPM and 17°C more efficient at 1200 RPM. Of course, we understand that both, Noctua and XILENCE coolers, are in fact more expensive than Tuniq Propeller 120, but in one of our recent reviews we have also tested cooling solutions prices similarly to Tuniq Propeller 120 that still demonstrated much more impressive results. Take, for instance, Scythe Grand Kama Cross: it is even cheaper than the new Tuniq solution, but cools the CPU just a little less efficiently than the Black Hawk made of solid copper.
During maximum overclocking tested, the copper top-cooler doesn’t leave the Tuniq newcomer a single chance:
Tuniq Propeller 120 | XILENCE Black Hawk Copper |
XILENCE Black Hawk Copper ensures 5°C lower CPU temperature than Tuniq Propeller 120 even though the processor is overclocked higher and works at higher core voltage. And as for the potential of Noctua NF-D14, you are already well familiar with it from our previous reviews, so we won’t repeat it here.
So why did the new Tuniq Propeller 120 cooler perform so low? Let’s check out the detailed results table for each processor core.
Click to enlarge
Note that the temperature difference between the cores cooler with the new Tuniq solution using core-contact technology is considerably bigger than by other two testing participants. In our opinion, it indicates that the heat transfer is not even and maybe there is no secure contact between the processor heat-spreader and the heatpipes in the cooler base, as well. Despite an almost ideal thermal compound imprint on the cooler base, one of the heatpipes could be pushed away from the processor heat-spreader when the cooler was pressing against it, or the 1.8 mm gaps between the heatpipes did have their negative say in the final result. However, it would be good to recall that Tuniq Tower 120 Extreme and Alpenföhn Nordwand have the same exact retention mechanism (although the heatpipes are configured differently in their base), but they work much more effectively than the new Propeller.
If we exclude the possibility of poor contact as well as any significant effect from the gaps between the heatpipes in the cooler base (the new TX-3 thermal paste didn’t help), then we could blame the too small heatsink that is simply unable to dissipate all the heat from the powerful processor. I believe that very high heatsink temperature during the tests as well as very high dependence on the fan rotation speed and airflow intensity could be considered indirect proof of this point. Besides, the heatpipes inside the heatsink arrays are not arranged in the most efficient way, either: the bottom of the heatsink plates is empty, while all four heatpipes go through the very top part of the heatsink arrays.
Despite its relatively low efficiency, Tuniq Propeller 120 does have a few advantages. The new cooler is universal, which means it supports all contemporary platforms. It uses simple and reliable retention that doesn’t require any special tools and guarantees secure hold and high pressure. Besides, Tuniq Propeller 120 comes bundled with a quality fan with blue LEDs that remains very quiet up until 1100 RPM and generates very moderate noise until 1400 RPM speed. Moreover, Tuniq Propeller 120 boasts unique looks and is very lightweight. And even though it won’t become the new “Big Typhoon”, we sincerely hope that Tuniq engineers will take into account our comments and will significantly improve their upcoming new versions of Propeller 120.
