Arctic’s Acclero Hybrid II-120 combines an AIO liquid cooler with a back side heatsink to produce a formidable yet quiet cooling solution for high-end graphics cards.
December 8, 2014 by Lawrence Lee
|
Product
|
Arctic Accelero Hybrid II-120 GPU Cooler |
|
Manufacturer
|
Arctic |
|
Street Price
|
US$115 |
Cooling a modern high-end graphics card quietly is a difficult task due to
the massive amount of heat they put out. Their unenviable position directly
adjacent to the CPU makes it tougher yet. Liquid cooling is a natural fit for
this cause, transferring the heat generated by the GPU core directly to a location
where it can be quickly expelled from the case. This allows all the components
to run cooler, as the heat doesn’t linger inside the case as long.
This advantage was evident when we reviewed the NZXT
G10 GPU bracket, which allows a variety of closed loop CPU coolers to
be adapted for use on a graphics card. Paired with a NZXT
X41 (140 mm radiator) and slapped onto a factory-overclocked GTX
680, the results were tremendous. The GPU core temperature ended up
well under 70°C at an impressive noise level of 17~18 [email protected]. Both the CPU
and PCH ran considerably cooler as well, so the effects were system-wide. Can
the all-in-one unit Arctic Accelero Hybrid II-120 perform similarly?
Formerly known as Arctic Cooling, Arctic is a bit of a legend as one of the
first aftermarket heatsink manufacturers to take on graphics card cooling. A
decade ago they launched the VGA Silencer, a classic that was the GPU-equivalent
of the Zalman CNPS7000. They received similar acclaim for the Accelero S1, a
superb fanless model, and followed that up with the formidable Accelero Twin
Turbo and Accelero Xtreme series.
|
|
The Hybrid II-120 is Arctic’s second foray into liquid GPU cooling. The original Hybrid is a combined air/liquid cooling solution with a small fan blowing over the VRM. This model moves away from that paradigm, shifting VRM cooling to the opposite side of the PCB by utilizing a large fanless aluminum heatsink which can also help cool the memory chips and other components as well. A support bracket, that latches on to the screw points of four expansion slots, helps keep the card from bending from all that weight. Meanwhile, the GPU core itself is cooled by a basic AIO cooler with a 120 mm fan.
|
|
Due to the nature of the heatsink, the accessories are unusual for a GPU cooler.
A large blue thermal pad precut into 50 pieces is used to transfer heat from
the various board components to the back side heatsink. A protective plastic
film ensures the heatsink doesn’t directly contact any circuitry, but you have
to manually cut holes in it so the pads can make contact. Also included is a
syringe of Arctic’s MX-4 thermal compound, all the necessary bolts, screws,
nuts, and washers, and a bit of documentation.
|
Arctic Accelero Hybrid II-120: Specifications
(from the product web page) |
|
| Max. Cooling Capacity | 320 Watt |
| Radiator Dimensions | 150 x 120 x 25 mm |
| Pump Dimensions | 70 x 70 x 28 mm |
| Heatsink Dimensions | 218 x 98 x 23 mm |
| Total Weight | 793 g |
| Fan Speed | 400 – 1,350 RPM (Controlled by PWM) |
| Fan Airflow | 74 CFM / 125.7 m³/h |
| Fan Noise Level | 0.3 Sone @ 1,350 RPM |
| Fan Power | 0.25A/+12V DC |
| Compatibility – Nvidia* | GTX 980, 970, Titan (Black),780 (Ti), 770, 760(Ti), 680, 670, 660 (Ti), 650 (Ti boost), 580, 570, 560 (Ti,SE), 550Ti, 480, 470, 465, 460(SE), GTS 450, 250, 240 (OEM) |
| Compatibility – AMD* | R9 290(X), 280(X), 270(X),R7 265, HD 8870, 7870(XT,GHz), 7850, 6970, 6950, 6870, 6850, 6790, 5870, 5850, 5830, 4890, 4870, 4850, 4830, 3870, 3850, 3690 |
| *Accelero Hybrid II-120 is compatible only with the graphics card using a standard size PCB with a width of 98mm. Graphics cards with different dimensions will not be compatible with this cooler. This includes: -Low profile Graphics card -Graphics cards with extended PCB (such as : ASUS DirectCU II R9 290, GTX780-DC2OC-3GD5) The compatibility list is based on AMD Radeon and NVIDIA’s reference board layout only. ARCTIC holds no responsibility for incompatibility on non reference boards. Please check the height restriction drawing before purchase. In the vast majority of the cases, it is not possible to assembly this cooler when the graphics card is located in the first extension slot of the motherboard. |
|
The official specifications list quite a few video card models as compatible
but there are several caveats.
- The card must have a standard size PCB (98 mm wide).
- It can’t be a low profile model.
- Arctic can’t guarantee it will work on non-reference boards.
- The fan is controllable via PWM but connects directly to the graphics card
with a 4-pin connector. Some video cards utilize 5-pin headers but they are
out of luck as no adapter is included.
PHYSICAL DETAILS
The Arctic Accelero Hybrid II-120’s main cooler has a 7.0 x 7.0 x 2.8 cm base, 41 cm of tubing, and a 14.9 x 12.0 x 2.7 cm radiator, according to my measurements. The aluminum heatsink for the back of the card weighs 340 grams on its own, with dimensions of 21.7 x 9.8 x 2.3 cm.
|
|
|
|
|
|
|
|
|
|
INSTALLATION
The most critical aspect of installation is firm, maximum contact
between the base and the GPU core for efficient heat conduction. Ideally
it should also be a simple procedure with the user having to handle as few pieces
of hardware as possible. The procedure for installing the main cooler itself satisfies these conditions but mounting the additional heatsink is more complex.
 The first step is to apply thermal compound to the center of the copper base, then secure the bolts to the mounting frame using the appropriate holes for the video card in question. Plastic washers go over them to prevent shorts, and depending on the GPU model, a set of washers may also be required to ensure proper spacing. |
|
|
|
|
|
|
|
|
|
|
|
|
INSTALLATION (Con’t)
After installing the graphics card, the last step is to mount the fan to the radiator and the radiator to the case. This is accomplished easily enough but with the nature of this cooler, there are some compatibility issues that need to be pointed out.
|
|
|
|
|
|
|
|
|
|
|
|
TEST METHODOLOGY
Test Platform
- Intel Core i3-2100 processor, Sandy Bridge core, dual
core 3.1 GHz, integrated HD 2000 graphics, TDP of 65W. - Scythe Kotetsu CPU cooler – Scythe Slip Stream 500RPM 120mm
fan. - Gigabyte Z77X-UD3H motherboard, Z77 chipset, ATX.
- Kingston HyperX Genesis memory, 2x4GB, DDR3-1600.
- Kingston SSDNow mS100 mSATA solid state drive – 64GB
- Kingwin Lazer Platinum power supply, ATX v2.2, 80 Plus Platinum,
1000W total output, 83A on +12V rail. - Fractal Design Define R5 case – ATX, stock 140mm fans.
- Microsoft Windows 7 Ultimate operating system, 64-bit
Measurement and Analysis Tools
- Prime95 stability test to stress the CPU.
- FurMark
stability test to stress the GPU. - GPU-Z to
monitor GPU temperatures and fan speeds. - SpeedFan to monitor system temperatures and adjust fan speeds.
- MSI Afterburner to adjust GPU fan speeds.
- Extech
AC Power Analyzer 380803
AC power meter, used to measure the power consumption
of the system. - A custom-built variable fan speed controller to power third party cooler fans.
- PC-based spectrum analyzer
— SpectraPlus with ACO Pacific mic and M-Audio digital
audio interfaces. - Anechoic chamber
with ambient level of 11 dBA or lower.
Test Procedure
Our test procedure is an in-system test, designed to determine whether the cooler is adequate for use in a low-noise system.
By adequately cooled, we mean cooled well enough that no misbehavior
related to thermal overload is exhibited. Thermal misbehavior in a graphics
card can show up in a variety of ways, including:
- Sudden system shutdown, reboot without warning, or loss of display signal
- Jaggies and other visual artifacts on the screen.
- Motion slowing and/or screen freezing.
Any of these misbehaviors are annoying at best and dangerous at worst —
dangerous to the health and lifespan of the graphics card, and sometimes to
the system OS.
Coolers are installed on an ASUS GeForce GTX 680 DirectCU II OC, a factory-overclocked single GPU card that draws 200W~205W by our estimates. The stock VRM heatsink is left on if possible. Noise is measured and recorded with our test system on with the cooler installed. Our mic is positioned at a distance of one meter from the center of the case’s left side panel at a 45 degree angle. The GPU temperature is recorded using GPU-Z.
Our main test consists of FurMark stability test running in conjunction with Prime95 to stress both the graphics card and processor simultaneously. This combination produces more CPU/GPU stress than a typical gaming session. As our test system has limited airflow, our results are not indicative of a real-world situation, but rather a worse-case scenario. If the cooling solution in question can cool the card and its components adequately in this environment it means there will be some degree of thermal headroom when deployed in a more conventional situation.
The cooler’s fan(s) is connected to the motherboard (if possible) and its speed is changed to various levels to represent a good cross-section of its airflow and noise performance. If it’s a passive heatsink, we use different predefined system fan speeds to determine the effect of system airflow on cooling performance.
|
GPU Test System:
Anechoic chamber measurements |
|
|
Case Fan Setting
|
System [email protected]
|
|
High
|
24 dBA
|
|
Med
|
15 dBA
|
|
Low
|
12~13 dBA
|
|
Note: mic is positioned at a distance of one meter from the center of the case’s left side panel at a 45 degree angle.
|
|
|
|
For heatsinks that support but don’t include standard case fans, our reference fans are used. The measured details
of these models are shown below, derived using our standard fan testing methodology.
|
Reference Noctua 140mm fan
Anechoic chamber measurements |
||
|
Voltage
|
Speed
|
|
|
12V
|
28~29 dBA
|
1250 RPM
|
|
9V
|
21 dBA
|
990 RPM
|
|
8V
|
18 dBA
|
880 RPM
|
|
7V
|
15~16 dBA
|
770 RPM
|
|
6V
|
13 dBA
|
660 RPM
|
|
Reference Nexus 120mm fan
Anechoic chamber measurements |
||
|
Voltage
|
Speed
|
|
|
12V
|
16 dBA
|
1080 RPM
|
|
9V
|
13 dBA
|
880 RPM
|
|
7V
|
12 dBA
|
720 RPM
|
|
Reference Nexus 92 mm fan
Anechoic chamber measurements |
||
|
Voltage
|
Speed
|
|
|
12V
|
16 dBA
|
1470 RPM
|
|
9V
|
12 dBA
|
1150 RPM
|
NOISE MEASUREMENTS
|
|
|
Specifications: Arctic Accelero Hybrid II-120 Stock Fan
|
|||
| Manufacturer | Arctic | Power Rating | 3.0 W |
| Model Number | F12 PWM | Airflow Rating | 74 CFM |
| Bearing Type | Fluid Dynamic | Speed Rating | 400 ~ 1350 RPM |
| Frame Size | 120 x 120 x 25 mm | Noise Rating | 0.3 Sone |
| Hub Size | 35 mm | Header Type | 4-pin PWM VGA |
| Blade Diameter | 113 mm | Starting Voltage | < 4.0 V |
| Cable Length | 40 cm | Weight | 100 g |
| Corner Type | Open | Retail Availability | Yes (w/ standard PWM header) |
Additional notes: |
|||
The included 120 mm is nothing remarkable from a design perspective, aside for perhaps the small hub, measuring only 35 mm across. It has nine blades of modest size and curvature with sharp tips. It’s a 4-pin PWM model but the header is the smaller VGA type.
|
Noise Measurements
(GPU Test System w/Accelero Hybrid II-120) |
|
|
Radiator Fan Speed
|
Total [email protected]
|
|
1390 RPM
|
27~28 dBA
|
|
990 RPM
|
20 dBA
|
|
850 RPM
|
17 dBA
|
|
560 RPM
|
15 dBA
|
|
Pump Only
|
14 dBA
|
|
Measuring mic positioned 1m at diagonal angle from
the center of the heatsink. System fans on low speed, CPU fan at 500 RPM. System noise level without GPU: 12~13 dBA. Ambient noise level: 10~11 dBA. |
|
Our GPU test system is very quiet, producing just 12~13 [email protected] when running on integrated graphics. The system fans are set to low speed (5V) via the case’s integrated fan controller and the CPU fan is limited to only 500 RPM.
As with all liquid coolers, the pump produces some noise as well, but surprisingly,
not very much. With the card installed, the pump on, and the radiator fan disabled
momentarily, the overall noise level is only 14 [email protected] — this is with the
pump receiving a full 12V.
27~28 [email protected] is the loudest this machine gets with the radiator fan at full speed and it crosses the quiet threshold (~20 [email protected]) once it gets down to about 1000 RPM. With a low starting voltage, it’s capable of speeds as low as 400 RPM but any level in-between would be more than acceptable from a noise perspective.
|
|
The pump generates a mild but distinct buzzing at all times and occasionally there’s an audible sloshing/gurgling of the liquid inside but overall, it sounds relatively pleasant, at least as far as pumps go. Some tonality is produced at ~160 and ~850 Hz but it’s not noticeable at distance.
|
|
Arctic has a good track record when it comes to stock fans and this one is no different. It has an exquisitely smooth and innocuous sound at speeds under 1000 RPM. Most of the small peaks produced in the spectrum above are from the pump.
TEST RESULTS
|
Load Test Results: Arctic Accelero Hybrid II-120
|
|||||
|
State
|
Idle
|
Prime95 + FurMark
|
|||
|
Radiator Fan Speed
|
560 RPM
|
560 RPM
|
850 RPM
|
990 RPM
|
1390 RPM
|
|
CPU Temp
|
21°C
|
47°C
|
47°C
|
45°C
|
43°C
|
|
PCH Temp
|
41°C
|
51°C
|
50°C
|
47°C
|
46°C
|
|
GPU Temp
|
27°C
|
78°C
|
62°C
|
55°C
|
51°C
|
|
GPU VRM Temp
|
41°C
|
100°C
|
98°C
|
96°C
|
94°C
|
|
System Power (AC)
|
60W
|
302W
|
303W
|
305W
|
305W
|
|
15 dBA
|
15 dBA
|
17 dBA
|
20 dBA
|
27~28 dBA
|
|
|
System noise level without radiator fan (pump only): 14~15 [email protected].
System noise level without GPU: 12~13 [email protected]. Ambient temperature: 22°C |
|||||
The GTX 680 malfunctions when the GPU temperature reaches 88°C but the
Hybrid II stays comfortably away from this mark. Even with the fan spinning
at a paltry 560 RPM, the GPU core stays below 80°C, and a modest increase
in fan speed brings it down by double-digits. The performance is hugely dependant
on the speed of the radiator fan, though the rate of improvement slows beyond
about 1000 RPM.
The back heatsink does a fairly good job as well, keeping the VRMs adequately
cooled at or below 100°C (they can take a lot more than that). The VRM,
CPU, and PCH temperatures all run cooler as the radiator fan speed is ramped
up. It’s clear that pulling the heat away with water cooling is beneficial to
all the components, not just the GPU.
COMPARISON
|
Comparison: CPU + GPU Load
|
|||||
|
Cooler
(Pump/VRM Fan Speed) |
GPU/Rad. Fan Speed
|
GPU Temp
|
GPU VRM Temp
|
CPU Temp
|
|
|
Stock Cooling
|
1100 RPM
|
86°C
|
116°C
|
55°C
|
17 dBA
|
|
Arctic Accelero Hybrid II-120
|
850 RPM
|
62°C
|
98°C
|
47°C
|
17 dBA
|
|
560 RPM
|
78°C
|
100°C
|
47°C
|
15 dBA
|
|
|
NZXT X41 + G10 (1820/1000 RPM)
|
770 RPM
|
63°C
|
94°C
|
45°C
|
18 dBA
|
|
680 RPM
|
68°C
|
94°C
|
46°C
|
17 dBA
|
|
|
System noise level without GPU: 12~13 [email protected].
Ambient temperature: 22°C |
|||||
The Hybrid II absolutely crushes the stock cooler at the 17 [email protected] level. Not
only does the GPU core run much cooler, the VRM temperature improves by almost
20°C, an impressive feat considering it uses a fanless heatsink. The CPU
receives 8°C of relief as most of the heat normally rising from the
GPU is siphoned away to the radiator. The stock solution actually has an impressive
three-slot design with a rather thick heatsink, so it’s closer to an aftermarket
model rather than a typical OEM cooler.
Against the NZXT X41 + G10 combination, the results are very close. Despite
its smaller radiator and fan size, the Hybrid II holds a 6°C advantage when
both coolers are producing an overall system noise level of 17 [email protected]. The CPU
basically gets the same cooling benefit with either the Arctic or the NZXT,
while the latter has a slight edge in VRM cooling.
MP3 SOUND RECORDINGS
These recordings were made with a high
resolution, lab quality, digital recording system inside SPCR’s
own 11 dBA ambient anechoic chamber, then converted to LAME 128kbps
encoded MP3s. We’ve listened long and hard to ensure there is no audible degradation
from the original WAV files to these MP3s. They represent a quick snapshot of
what we heard during the review.
These recordings are intended to give you an idea of how the product sounds
in actual use — one meter is a reasonable typical distance between a computer
or computer component and your ear. The recording contains stretches of ambient
noise that you can use to judge the relative loudness of the subject. Be aware
that very quiet subjects may not be audible — if we couldn’t hear it from
one meter, chances are we couldn’t record it either!
The recording starts with 5 second segments of room ambiance, then the machine idling with the case fans on low speed, then the cooler’s fan
at various levels. For the most realistic results, set the volume so that
the starting ambient level is just barely audible, then don’t change the volume
setting again.
- Arctic Accelero Hybrid II-120
— stock fan off, pump only (14 [email protected])
— stock fan at 560 RPM (15 [email protected])
— stock fan at 850 RPM (17 [email protected])
— stock fan at 990 RPM (20 [email protected])
— stock fan at 1390 RPM (27~28 [email protected]
FINAL THOUGHTS
I expected the Arctic Accelero Hybrid II-120 to perform well but am surprised it manages to compete so strongly with the larger NZXT G41 + G10 combination. It actually delivers slightly better GPU cooling while maintaining an excellent level and quality of noise. The pump is not controllable but even at full speed, it’s one of the least offensive sounding models I’ve encountered, and the included fan has silky smooth acoustics. Arctic’s passive trace-side heatsink approach is not as effective as the G10’s 80 mm fan but it’s more than competent, only trailing the NZXT solution by a few degrees in VRM temperature. It takes some work to assemble, but the design, utilizing thermal pads to contact areas that need additional cooling, allows it to support a wide variety of different board designs.
However, failing to provide proper compatibility for wider PCBs rules out many
variants of non-reference, high-end graphics cards (of the Radeon R9 290X and
the GeForce GTX 780 and 980, for example) — the kinds of cards that can
benefit most from the Hybrid II’s capabilities. It’s technically incompatible
with these models as the clips can’t clamp onto the heatsink, but it can be
jerry-rigged to work for the most part. The heatsink intrudes a good inch into
the space normally reserved for the CPU cooler, and this could cause headaches.
It can also block portions of the motherboard: It covers the expansion slot
above, and on our platform, it interferes with the internal USB 3.0 header.
The fan uses a 4-pin video card connector, which is great as it won’t take up
an extra 3-pin fan header, but if your video card has a 5-pin design, you’re
out of luck as an adapter isn’t included.
If these various compatibility issues don’t apply or are acceptable, the Accelero
Hybrid II-120 is certainly a solid candidate, a formidable yet quiet cooler
worthy of any high-end graphics card. It’s currently selling for around $100,
a fair bit to add to the already high cost of a high-end graphics card, but
it delivers top-notch performance and improves cooling system-wide. It also
offers a bit better value than the NZXT X41 and G10 bracket combination.
Our thanks to Arctic for the Accelero Hybrid II-120 graphics card cooler sample.
* * *
Arctic Accelero Hybrid II-120
is Recommended by SPCR
Articles of Related Interest
NZXT Kraken G10 Graphics Adapter
Prolimatech MK-26 Multi-VGA Cooler
SPCR’s 2012 Graphics Card/Cooler Test System
ASUS GeForce GTX 680 DirectCU II OC
Thermalright Shaman 140mm Fan GPU Cooler
Arctic Cooling Accelero Xtreme Plus GPU Cooler
* * *
