Which is The Best Configuration for 120 mm Liquid Cooling Fans?

[nextpage title=”Introduction”]

We tested 12 different configurations for installing fans on radiators of liquid cooling systems. Which one is the best? Let’s see.

Some time ago, most liquid coolers (a.k.a. watercoolers) were kits that the user had to assemble himself, at the risk of making a big mess inside the computer. Sealed “all-in-one” systems were rare; there were few options in the market.

Nowadays, however, sealed liquid (a.k.a. maintenance-free) cooling systems are a good option for the user who wants to keep his or her computer cool with low noise and high flexibility. Several brands sell those coolers, offering from basic to high-end models.

Those liquid cooling systems have a block that is in contact with the CPU and transfers the heat to the flowing liquid; the pump (usually integrated into the block) that makes the liquid flow; hoses to connect everything; and a radiator that cools the liquid, transferring the heat to the air. But the radiator needs to be cooled by fans, and there is not much agreement about how to install the radiator and the fan (or fans) on the case.

In this article, we tested several methods of installing the radiator and the fans. We measured noise and temperatures on each test in order to see how the radiator position (and the position of the fans relative to the radiator) influences the temperature and noise.

The positions we tested and the results achieved are described in the following pages.

[nextpage title=”Mounting Positions”]

We used the Zalman LQ315 liquid cooling system for this test because it is an average sealed watercooler, with a 120 mm radiator. It comes with a single fan but can be mounted with two.

We installed the radiator first on the rear panel of the case, and then on the top panel of the case. For each position, we tested the system with the fan (or fans) blowing air from the inside to the outside of the case (i.e., fans working in exhaust mode) and then from the outside to the inside of the case (i.e., fans working in ventilation or intake mode).

For each of those tests, we measured noise and temperatures with one 120 mm fan and then with two 120 mm fans. When only one fan was used, we performed two tests: first with it  mounted between the case panel and the radiator (which are calling “Position 1”) and then with it mounted on the inner side of the radiator (which are calling “Position 2”).

When we combined all different possibilities, we ended up with a total of 12 tests.

Figures 1 to 5 show some of the different positions we used.

Figure 1: Rear panel installation, one fan, position 1, exhaust

Figure 2: Rear panel installation, two fans, exhaust

Figure 3: Rear panel installation, two fans, intake

Figure 4: Top panel installation, one fan, position 1, intake

Figure 5: Top panel installation, one fan, position 2, exhaust

[nextpage title=”How we Tested”]

We tested the cooler with a Core i5-2500K CPU (quad-core, 3.3 GHz), which is a socket LGA1155 processor with a 95 W TDP (Thermal Design Power). In order to get higher thermal dissipation, we overclocked it to 4.5 GHz (100 MHz base clock and x45 multiplier), with default core voltage (Vcore). Notice this is a higher overclock setting than we use on our cooler reviews.

The Zalman LQ315 liquid cooling system we used on this test comes with only one fan, but we used a second fan from the same model (Zalman ZP1225ALM, 120 mm, 2.4 W, 2,000 rpm) on the tests with two fans.

We measured noise and temperature with the CPU under full load. In order to get 100% CPU usage in all cores, we ran Prime 95 25.11 with the “In-place Large FFTs” option. (In this version, the software uses all available threads.)

Room temperature measurements were taken with a digital thermometer. The core temperature was read with the SpeedFan program (available from the CPU thermal sensors), using an arithmetic average of the core temperature readings.

During the tests, the panels of the computer case were closed. The front case fan was spinning at minimum speed in order to simulate the “normal” cooler used on a well-ventilated case. We assume that is the common setup used by a cooling enthusiast or overclocker.

The sound pressure level (SPL) was measured with a digital noise meter, with its sensor placed near the top opening of the case. This measurement is only for comparison purposes, because a precise SPL measurement needs to be made inside an acoustically insulated room with no other noise sources, which is not the case here.

Hardware Configuration

• Processor: Core i5-2500K
• Cooler: Zalman LQ315
• Motherboard: ASUS Maximus IV Extreme-Z
• Memory: 16 GB G.Skill Sniper (DDR3-1600/PC3-12800), configured at 1,600 MHz
• Hard disk: Mushkin Chronos 120 GB SSD
• Video card: MSI GeForce 210 1 GB
• Video resolution: 1920×1080
• Video monitor: Samsung SyncMaster P2470HN
• Power supply: Seventeam ST-550P-AM
• Case: Cooler Master HAF 922

Operating System Configuration

• Windows 7 Home Premium 64 bit SP1

Software Used

• Prime95
• SpeedFan

Error Margin

We adopted a 2°C error margin, meaning temperature differences below 2°C are considered irrelevant.

[nextpage title=”Our Tests”]

The data acquired in our tests is shown in the table below. Keep in mind that “pos 1” means the fan was located between the radiator and the case panel, and “pos 2” means the fan was attached to the internal side of the radiator.

Configuration	Room Temp.	Noise	Speed	Core Temp.	Temp. Diff.
Rear, 1 fan, exhaust, pos 1	25 °C	50 dBA	2,050 rpm	75 °C	50 °C
Rear, 1 fan, exhaust, pos 2	25 °C	48 dBA	2,050 rpm	72 °C	47 °C
Rear, 1 fan, intake, pos 1	25 °C	57 dBA	2,050 rpm	71 °C	46 °C
Rear, 1 fan, intake, pos 2	26 °C	49 dBA	2,000 rpm	69 °C	43 °C
Rear, 2 fan, exhaust	26 °C	52 dBA	2,000 rpm	69 °C	43 °C
Rear, 2 fan, intake	26 °C	59 dBA	1,950 rpm	68 °C	42 °C
Top, 1 fan, exhaust, pos 1	25 °C	50 dBA	2,000 rpm	70 °C	45 °C
Top, 1 fan, exhaust, pos 2	25 °C	52 dBA	2,050 rpm	70 °C	45 °C
Top, 1 fan, intake, pos 1	25 °C	55 dBA	2,000 rpm	72 °C	47 °C
Top, 1 fan, intake, pos 2	25 °C	52 dBA	2,000 rpm	69 °C	44 °C
Top, 2 fan, exhaust	25 °C	56 dBA	2,000 rpm	67 °C	42 °C
Top, 2 fan, intake	26 °C	60 dBA	2,000 rpm	68 °C	42 °C

The graph below shows the difference of the core temperature minus the room temperature for each test.

On the next graph, you see the noise level for each test.

[nextpage title=”Conclusions”]

Looking at the graphs, our first conclusion is that the position of the radiator does not matter. The temperatures with the radiator at the rear panel of the case were basically the same as what we saw when the radiator was installed on the top panel.

The second conclusion is that, in terms of temperature, two fans are better than only one. With two fans installed in push-pull configuration, it doesn’t matter if you install them blowing air to the inside or to the outside of the case. However, two fans make much more noise than only one.

A curious result was obtained on the test with one fan, with the radiator at the rear panel of the case. With the fan blowing air to the outside of the case in both positions, and blowing air to the inside of the case with the fan installed between the case panel and the radiator, the cooling performance degraded. The noise level was too high when the fan was installed as intake and installed between the case panel and the radiator.

Installing one fan at the case’s rear panel, as intake (i.e., blowing air from the outside to the inside of the case), and with it installed on the inner side of the radiator seems to be the best option, reaching an optimal combination of cooling performance and low noise level.

Keep in mind that the results we obtained are valid only for the hardware configuration we used. Other configurations of CPU, case, cooler, or fan models can behave differently.