Thermaltake offers a dual-orb, dual-fan cooling system for hot VGA cards in the CL-G0102 DuOrb. Are two really better than one?
July 17, 2008 by Lawrence Lee
|Thermaltake Duorb CL-G0102|
As graphics cards become more and more powerful, cooling them becomes a
daunting challenge. In a typical ATX tower system, the graphics porocessing
unit (GPU) is in the middle of the case, upside down, with its own printed
circuit board (PCB) obstructing a lot of the heat being put off. To complicate
matters, GPUs can now easily pull more than 100W and with PCB lengths increasing,
sometimes the system airflow can be effectively cut off.
The Duorb by Thermaltake is one of many GPU coolers available to PC enthusiasts
who want better than stock cooling. It is a dual fan cooler like the Xigmatek
Battle-Axe, but is only equipped with two heatpipes. Its basic design
is similar to the Zalman
VF900-CU, which was a star in its day. The VF900 is simply too small
to effectively cool most current GPUs. A larger radial cooler with a massive
fan would seem like an obvious evolutionary step, but you can only make it
so big before it touches the expansion slots on the motherboard. Thus, Thermaltake
has decided to go horizontal, splitting what would have a been a small, radial
copper cooler into two small semi-radial coolers. The result is the Duorb,
or two orbs as the name suggests.
Thermaltake Duorb: Features & Brief (from
product web page)
Features & Brief
|Universal Design for|
nVIDIA & ATi VGA card.
|Supports all the|
|Two orbs must be|
better than one…
|All copper structure:copper|
base & copper fin provide maximum cooling performance.
| The use of copper|
increases the weight and cost of the product, but its higher thermal
conductivity compared to aluminum may make it a worthy trade-off.
|ScurveTM fan blade design|
and 6-LED lights.
|The fan blades are|
twisted slightly in an unusual fashion.
|Maximum cooling efficiency|
|Thermaltake is last|
manufacturer we would associate with "quietness."
Thermaltake Duorb: Specifications (from
product web page)
nVIDIA 8800 GT/GTS/GTX/Ultra
nVIDIA 7900 GTX
nVIDIA 7800 GT/GTX
nVIDIA 6800 GS/GT
ATi X1950 PRO/XT/XTX
ATi X1900 GT/XT/XTX
ATi X1800 GTO/XL/XT
ATI HD 3870
|Dimension||(L)180.5 x (W)98.5 x (H)30 mm|
|Heatsink Material||Copper base & Copper Fin|
|Heatpipe||6 mm x 2pcs|
|Fan Dimension||80 x 15 mm x 2pcs|
|Started Voltage||7 V|
|Max.Air Flow||23.8 CFM|
|Max. Air Pressure||2.9 mmH2O|
|Fan Speed||2500 RPM|
|Life Expectation||50,000 hrs|
One thing Thermaltake always manages to get right is aesthetics. With all-copper
construction and translucent fans and bright blue LEDs, the Duoorb is an attention-grabber,
at least before it’s installed and impossible to view.
The fans are attached to two metallic rings joined together with a bridge
screwed to the mounting plate. The fans are made of hard, translucent plastic,
and the blades are warped/twisted, possibly in an effort to reduce turbulence.
The heatpipes and fins are firmly bond together creating a form resembling
an UFO. In between the moderately spaced fins, solder is plainly visible.
Each heatpipe emanates from the base and curves into a "C" shape
on each side. One half of the cooler is stretched further away than the
other to create more distance between them. The fans are wired to a single
4-pin molex power connector. Unlike many other Thermaltake cooling products,
no fan controller is provided.
The base is very flat and polished to an almost mirror finish. There are
very faint, nearly invisible machine marks.
The fans, manufactered by Yen Sun Technology measure approximately 72mm
across. The hub has three short spokes which screw onto a circular metal
ring. As the heatsink is radial, the fins closest the fan are more tightly
The Duorb ships with large set of additional heatsinks — 12 for memory
and 4 for VRMs. It should be enough to cover all the extra components that
require cooling on modern video cards.
The installation of the main cooler is fairly simple and similar to the
and Zalman VF1000.
First, a pair of mounting clips are screwed onto the base. Since it’s mandatory
for all cards, we would have liked to have seen them built directly onto
the heatsink or attached straight from the factory.
Four double-sided hex screws are installed in the appropriate mounting
holes. The longer side goes through the mounting holes on the graphics card
and out the other side where barrel-shaped nuts secure them.
Unfortunately, one of the hex screws snapped inside the nut as it was being
tightened. Imprecise machining of the tiny thread caused binding, and just a
smidgen of extra pressure was all it took for the rod to snap. The broken part
was stuck in the net and impossible to remove. This mean we lost not just one
screw but also the matching nut.
The exact same problem occured during our Xigmatek Battle-Axe review. We really
wish aftermarket cooler companies would use shorter, stronger screws.
A search through our spare parts bins turned up nothing that could substitute
for the missing parts. Faced with the choice of securing the heatsink on three
sides or two, we went with two, to balance the pressure on the GPU. A check
of the thermal paste footprint confirmed that good contact was being made, and
the cooler could only be twisted slightly when torque was applied. Extreme care
was used in tightening the two screws/nuts.
We found there was no overhang, meaning the Duorb will fit in any case that
can accomodate a full-sized graphics card. This may vary slightly depending
on the card and GPU location.
The ramsinks were left off our test card because they did not appear to be
re-usable, and the test card seems to function fine without them. We did install
one of them just to check the thermal adhesive, and it was fairly secure. To
say they are low-profile is an understatement — they are only 2mm tall.
Being so short and having only five thick fins it’s doubful they provide much
The clearance under the cooler was ample and uniform at 9mm. There was no reason
the ramsinks could not have been taller — they could be quadrupled in height
and still fit underneath the Duorb.
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
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.
Measurement and Analysis Tools
Our main test consists of ATITool’s artifact scanner running in conjunction
with CPUBurn to stress both the graphics card and processor simultaneously.
It is a realistic test that mimics the stress on the CPU and GPU produced by
a modern video game, only more consistantly. The software is left running until
the GPU temperature stabilizes for at least 10 minutes at which point, both
the CPU and GPU temperatures are recorded. In addition we also take measurements
of the system’s overall noise level and power consumption using a B&K Sound
Meter and a Seasonic Power Angel respectively. If the heatsink has a fan, the
procedure is repeated at various fan speeds while the system fan is left at
the lowest setting of 7V. If it is a passive cooler, the system fan instead
is varied to study the effect of system airflow on the heatsink’s performance.
If artifacts are detected in ATITool or other instability is noted, the heatsink
is deemed inadequate to cool the video card in our test system.
Preliminary testing is also done at idle, and with only CPUBurn running for
comparison. For idle results, the system is left stagnant for 10 minutes before
ATITool is loaded and the first temperature it reports is used. We do this because
on our test platform, after ATITool is loaded, it puts some kind of stress on
the GPU, causing the temperature to climb immediately (even if it is left idle
for hours beforehand) and the power consumption to increase by approximately
10W. We theorize that initially the card is in 2D mode, either underclocked
or undervolted (or possibly both) and that ATITool automatically puts it in
3D mode, which would account for the rise in temperature and power draw. ATITool
is left running in the background for the remainder of testing which is why
the GPU temperature during CPUBurn will appear higher compared to idle. Consider
this the difference between 2D idle and 3D idle.
The ambient conditions were 25°C and 14 dBA at the time of testing.
With the Duorb being fairly light and only occupying one extra slot, it was
very easy to install the test card.
Test Results: ATI Radeon X1950XTX + Thermaltake Duorb
CPUBurn + ATITool
VGA fan speed approximated via Stroboscope.
Fans @ 12V: The fans were very loud with much high-pitched turbulence.
Performance was fairly good with a GPU temperature of 75°C.
Fans @ 9V: The noise level was much improved but it was still whiney
and turbulent. The GPU temperature barely changed from 12V, only 1°C
Fans @ 7V: While they can pass for ‘quiet’ at this speed, the fans
were still quite buzzy and the reduction in noise level allowed an odd, intermittent
electrical squeal to become audible. Performance was only reduced by a single
Fans @ 5V: Very quiet, with a low-pitched mechanical hum evident close-up.
Inside the case, it was inaudible. The GPU temperature did not budge compared
The interaction between the fans generated a lot of noise — each fan on
its own was not that bad, but when their output streams collided it created
a lot of extra turbulence. If they were spaced further apart, the acoustics
would have been better. Alas, the results show that all that extra airflow and
noise amounted to almost nothing — there was barely any change in cooling
performance when the fan speed was lowered. We found the fans’ starting voltage
was about 3.7V (not 7V as listed in the specifications) so they are prime candidates
for undervolting — a built-in fan controller (even a manual one) would
have been a godsend.
The results may have been skewed by our mounting problems, but they would have
been slight. Good, tight contact was being made between the GPU core and heatsink
Thermaltake Duorb vs. Zalman VF1000 vs. Xigmatek
Noise levels were measured in open air. 3°C
was subtracted from the results of the Duorb as the ambient temperature
was 3°C lower at the time the other coolers were tested.
The Duorb, with two heatpipes and two fans, more or less matched the performance
delivered by the Zalman VF1000, with its four heatpipes and one fan. The noise
levels between the two were comparable as well. The Xigmatek Battle-Axe edges
them by a small margin, keeping the GPU at about 70°C no matter how fast
the fans were spinning. To date the only cooler that has broken 70°C at
load on our test VGA card is the Arctic
Cooling Accelero S1 paired with a Nexus 120mm fan.
MP3 NOISE RECORDINGS
The recordings begin with the ambient noise of the test room. Please set your
playback volume so that the ambient noise is almost inaudible, then don’t adjust
the volume control again. For best results, save the sound file to your own
PC, then listen.
Duorb in test system (at 5V, 7V, 9V, 12V @ 1m)
The MP3 is broken into five-second sections beginning with the ambient
in the room (14 dBA), followed by the VGA test system without a video card
installed (17 dBA).
Duorb in open air (at 5V, 7V, 9V, 12V @ 1m)
The MP3 is broken into five-second sections beginning with the ambient
in the room (14 dBA).
Thermaltake products usually rely on fast spinning fans to keep them competitive
with more efficient products, but with the Duorb, the fast, noisy fans were
not necessary. Even at 5V with the fans barely audible, it achieved almost the
same level of performance as it did with the fans churning vigorously at 12V.
The Duorb has few of the qualities befitting a silent PC: Not only do the fans
spin too fast, they’re made of a brittle resonant material, hard-mounted, and
spaced too closely together. The lack of fan speed control compounds the acoustical
The provided RAM heatsinks are small and probably ineffective. Next to Zalman’s
low-profile heatsinks, they seem rather pathetic. Cutting corners like this
when memory chips are getting hotter with each new GPU generation seems short-sighted.
In addition, the mounting system really needs work. Zalman VGA coolers use a
similar method but with thumbscrews and sturdier screws which keeps fidgetting
to a minimum — it’s very simple to use. Another better mounting system
is exemplified by the Arctic Cooling Accelero S1/S2, which uses four screws
and mounting arms that are attached permanently, making installation a one step
Judged purely on cooling performance, the Duorb is excellent, delivering a
high level of performance and at a lower price-point than the Zalman VF1000.
However, its cooling ability is limited by its two heatpipe design and even
ridiculous amounts of airflow cannot improve it further. The extra airflow might
be useful for a super-hot, poorly designed system operated in an unairconditioned
room during a heat wave in the tropics… but that’s a long stretch. The basic
heatsink design is good, but for quiet cooling, a fan controller is mandatory,
along with soft hands to finesse the fragile mounting screws.
Thermaltake Duorb GL-C0102
Pros * Good performance with any amount of airflow
* Occupies only one extra slot
Cons * Fans much louder than needed
* No fan controller included
* Questionable mounting hardware
* Ramsinks are puny
Thanks to Thermaltake
for the Duorb sample.
SPCR Articles of Related Interest:
First Direct-Touch Heatpipe VGA Cooler
Arctic Cooling Accelero
S2 VGA Cooler + Turbo Module
Arctic Cooling Accelero S1 VGA Cooler
Updated VGA Card/Cooler
Zalman VF1000 LED Graphics Card
* * *