• Home
  • blog
  • Intel’s HSF for high-end Core 2 Extreme CPUs

Intel’s HSF for high-end Core 2 Extreme CPUs

Intel has a new, cool-looking radial heatsink/fan for its high end retail processors. Do the the aftermarket HSF makers have anything to fear?


May 21, 2008 by Lawrence
Lee

Product
Intel FCLGA4-S
LGA775 CPU Cooler
Manufacturer
Intel
(Delta OEM)
Street Price
~N/A

Stock heatsink/fans that come boxed with retail processors are usually just
good enough to cool the CPU. It’s understandable why. Users generally don’t
make their decision on a CPU based on the included stock cooler, which is seen
as a giveaway, a freebee that often gets replaced with a much higher performance
aftermarket cooler. CPU makers don’t see themselves as being in the business
of making coolers. There is very little incentive to include a high performance
or quiet heatsink/fan.

The Intel Core 2 Extreme series of quad core processors are rated for thermal
design power (TDP) of 130W or more, similar to the P4 Prescott or Pentium D
line of CPUs based on the old netburst architecture — a line plagued with
overheating problems. To avoid such issues this time around, the standard LGA775
stock cooler has been replaced in favor of something with more oomph. Generally
we find reference CPU coolers to be adequate, but much louder than we like.
Will this improved heatsink be good enough? Or will we toss it aside like we
usually do? Only one way to find out.


An unusually long sleek black box is required to contain the retail box
QX9650 processor and accompanying cooler.

 


The plastic clamshell holding the heatsink and processor is larger than
necessary, but not the worst offender we’ve seen by far.

 

Intel FCLGA4-S: Specifications
Measured Dimensions (overall)
133(L) x 133(D) x 78(H) mm
(minus mounting frame)
Measured Dimensions (lower
section)
90 (L) x 90 (D) x 37 (H)
mm
Measured Weight 420g
Fan
Model DTC-AAS10 (Delta)
Measured Diameter 105 mm
Rated Current 0.80A
Power Input 12V
Connector 4-pin PWM

PHYSICAL DETAILS

At first glance we like what we see — the FCLGA4-S stands out immediately
as the largest reference heatsink we’ve ever come across. It is a radial cooler
like previous Intel coolers, but now the fins extend past the fan’s diameter,
surrounding it like the classic Zalman flower series.


Staring clockwise from the left: Intel Core 2 Extreme cooler, standard
Intel LGA775 cooler, Zalman 7700 modified with Nexus 120mm fan.

When compared side-to-side, it seems almost unfathomable that the original
LGA775 heatsink was ever tasked to cool a 130W processor.


The Intel cooler on the left, MaxOrb on the right.

The FCLGA4-S is similar in size to the Thermaltake MaxOrb, though much lighter,
and more symmetrical. They both look like giant silver mushrooms.


Underside.

The heatsink is comprised of three main pieces: A copper core and two sections
of nickel-plated aluminum fins circling it. The upper fins have four large gaps
to allow a screwdriver access to the push-pins. The lower section forms a full
circle, and is quite tightly spaced in comparison.


A closer look at the base.

The base is very flat, with miniscule circular machine marks visible (not
on camera though). As usual, thermal compound is conveniently pre-applied.

MORE PHYSICAL DETAILS & INSTALLATION

Unlike previous Intel coolers, the fan is easily removed. It is secured via
three tension clips.


Fan removed.

Three arms extend from the fan hub, pushing against a notch in the fins to
keep it in place. It’s a simple but crude way to mount a fan. Ideally, you
would not want the fan to touch the heatsink in order to avoid vibration transfer
via conduction. The fan is made of rigid, translucent plastic to scatter the
light from three blue LEDs. It’s amusing to see superfluous LEDs on a piece
of hardware with an Intel sticker.


Fan controller switch.

Attached to the fan assembly is a rudimentary fan controller — a simple
switch that sets the fan to either high or low speed.The circuitry is exposed,
without any window dressing.


The core.

The fins are soldered to a copper core, which had an odd-shaped wick sticking
out at the top. Frostytech
apparently cut it open, or obtained exclusive photos from Intel revealing a
vapor chamber inside:

“The 41mm diameter copper ‘thermal chamber’ works
along a similar principle as a heatpipe, using a working fluid under a vacuum.
The low pressure causes the fluid to change states when heat is applied,
allowing the vapour to rapidly conduct heat between hot and cold surfaces.
A solid metal block by comparison would rely upon the metals conductive
properties to move heat from the base and distribute it along the cylindrical
walls to which the fins are soldered.”


A closer look.

“The Intel FCLGA4-S reference heatsink by comparison
has relatively thin copper walls around the hollow chamber… The walls
are no more than 0.5-1mm thick, and a sintered metal wick is evidently used
to return condensed working fluid back to the hot side. The copper base
is surprisingly thin too, about 4mm. The net result is a really lightweight
heatsink for its thermal performance capabilities.”

This is quite a departure for Intel. Up until recently their idea of improving
heatsink efficiency was to put a thin piece of copper in the middle of the base,
and now they’re using a vapor chamber. Aftermarket heatsinks usually use heatpipes
as the main method of heat transfer — the Asetek
VapoChill Micro
is the only product we’ve come across that uses anything
like a vapor chamber. It featured an enclosed reservoir of gas/liquid connected
to three different columns or pipes. It was a mediocre performer.


Mounted.

The heatsink mounted easily on our test platform. It was a fairly snug fit, with
very little rotational leeway. The cooler overhung the edge of our motherboard
by only 16mm.

TESTING

Testing was done according to our
unique heatsink testing methodology
, and the reference fan was profiled
using our standard fan testing
methodology
. A quick summary of the components, tools, and procedures
follows below.

Key Components in Heatsink Test Platform:

  • Intel
    Pentium D 950
    Presler core. TDP of 130W; under our test load, it measures
    78W including efficiency losses in the VRMs.
  • ASUS
    P5LD2-VM
    motherboard. A basic microATX board with integrated graphics
    and plenty of room around the CPU socket.
  • Samsung
    MP0402H
    40GB 2.5″ notebook drive
  • 1
    GB stick of Corsair XMS2
    DDR2 memory.
  • FSP
    Zen
    300W fanless power supply.
  • Arctic
    Silver
    Lumière: Special fast-curing thermal interface
    material, designed specifically for test labs.

Test Tools

  • Seasonic
    Power Angel
    for measuring AC power at the wall to ensure that the
    heat output remains consistent.
  • Custom-built, four-channel variable DC power supply,
    used to regulate the fan speed during the test.
  • Bruel & Kjaer (B&K) model 2203 Sound Level
    Meter
    . Used to accurately measure noise down to 20 dBA and below.
  • Various other tools for testing fans, as documented
    in our standard fan testing
    methodology
    .

Software Tools

  • SpeedFan
    4.32
    , used to monitor the on-chip thermal sensor. This sensor is not
    calibrated, so results are not absolute.
  • CPUBurn
    P6
    , used to stress the CPU heavily, generating more heat than most
    real applications. Two instances are used to ensure that both cores are stressed.
  • Throttlewatch
    2.01
    , used to monitor the throttling feature of the CPU to determine
    when overheating occurs.

Noise measurements were made with the fan powered from the lab’s variable DC
power supply while the rest of the system was off to ensure that system noise
did not skew the measurements. Keep in mind that the fan in the heatsink is
a PWM fan, which typically have a narrower speed range when controlled by direct
changes in voltages. It may also exhibit slightly different acoustics under
PWM speed attenuation.

Load testing was accomplished using CPUBurn to stress the processor, and the
graph function in SpeedFan was used to make sure that the load temperature was
stable for at least ten minutes. The stock fan was tested at various voltages
to represent a good cross-section of its airflow and noise performance.

The ambient conditions during testing were 18 dBA and 20°C.

TEST RESULTS

The material used for the fan made it difficult to get a proper fan speed reading
using a tachometer. After finding the noise level at 12V, 9V, 7V and 5V using
a custom DC fan controller and our B&K SLM, we hooked the fan up to our
test motherboard and used SpeedFan to manipulate and read the fan speeds.

Intel FCLGA4-S Fan (Low) Measurements
SpeedFan Setting
Fan Speed
Noise Level
100%
2050 RPM
38 dBA@1m
90% (~9/12V)
1930 RPM
36 dBA@1m
80% (~7V)
1820 RPM
34 dBA@1m
70%
1690 RPM
32 dBA@1m
60%
1520 RPM
30 dBA@1m
50%
1380 RPM
28 dBA@1m
40% (~5V)
~1300 RPM*
25 dBA@1m
* Estimated (at 40% and below, the fan speed failed to
register)

From noise data alone, we extrapolated that the fan spins at approximately
1300 RPM at 5V, 1800 RPM at 7V, and between 1900 and 2000 RPM at 9/12V. These
results were obtained with the fan set to low — we had no desire to try
it any faster.

Cooling Results

Intel FCLGA4-S
Fan Voltage
Noise @1m
Temp
°C Rise
°C/W
12V
36 dBA
42°C
22
0.28
9V
36 dBA
42°C
22
0.28
7V
34 dBA
43°C
23
0.29
5V
25 dBA
45°C
25
0.32
Load Temp: CPUBurn for ~10 mins.
°C Rise: Temperature rise above ambient (20°C) at load.
°C/W: based on the amount of heat dissipated by the CPU (measured
78W); lower is better.

Fan @ 12V: The fan was very aggressive with a noticeable buzzing.
The fan also transferred vibration to the heatsink’s fins, creating the sound
of rattling metal. With a 22°C rise in CPU temperature, it was fairly
inefficient considering it emitted an unbearable 36 dBA.

Fan @ 9V: The fan sounded pretty much identical as it did at 12V.
Thermals did not change.

Fan @ 7V: The sound level dipped slightly. The rattling of the fins
was still evident, but as fan was spinning slower, the frequency of the clanging
lowered, creating a harsher acoustic profile. The CPU temperature increased
by only a single degree.

Fan @ 5V: The noise level dropped significantly. The rattle was barely
noticeable at a distance of 1m. Within a foot of the heatsink, there was a
tiny bit of clanging like a very light breeze blowing through a set of wind
chimes. The fan whined as well. While this was a great improvement over higher
speeds, it still was far from being quiet. On the bright side, there was only
a three degree difference compared to the fan at 9 or 12V.

While we were impressed with how little the CPU temperature changed between
1900 and 1300 RPM, the noise it produced was irksome. At 5V, it was still
generating 25 dBA. We did not feel it was appropriate to test it any lower
as the fan’s starting voltage was fairly high (8.4V).


Loose fins rattle and buzz at high frequency.

The source of the rattling and clanging was found to be the loose fin sections
in the upper section of the heatsink. During testing we found that compressing
them improved the acoustics significantly. The fan itself was not awful, but
the way it interacted with the rest of the heatsink was.

Comparisons

Comparison: Intel FCLGA4-S vs. Other Radial Coolers
Zalman CNPS8700
Thermaltake MaxOrb
Intel FCLGA4-S
SPL @1m
°C Rise
SPL @1m
°C Rise
SPL @1m
°C Rise
33 dBA
17
35 dBA
17
36
22
28 dBA
19
29 dBA
19
34
23
24 dBA
23
24 dBA
21
25
25

Compared to radial aftermarket coolers, the FCLGA4-S doesn’t do terribly, trailing
the Thermaltake MaxOrb and the Zalman 8700 by 4-5°C. Normally we would rebuke
a heatsink for such a poor result, but considering it comes bundled essentially
for free with a CPU (admittedly a $1,000+ CPU) it’s hard to criticize it for
not matching a $50 product. We do however take issue with the chattering fins.

FINAL THOUGHTS

Intel’s implementation of a vapor chamber cooler left us unimpressed, but
it’s hard to say whether the technology itself is to blame. It is just a stock
heatsink/fan, after all, so the amount of time and money put into its design
and manufacture is limited — it could be a case of poor execution. Without
third party vapor chamber heatsinks to compare it to, we cannot draw any meaningful
conclusions regarding its core design.

This Core 2 Extreme heatsink is more than capable of cooling a 130W TDP processor,
but not quietly. While we were not expecting it to be silent, we had hopes.
The jarring sound of vibrating metal would be annoying even for an average user.
We like how the fan is easily removed but this does not make the fan any easier
to replace. We would have gladly traded this feature for a more secure mounting
that would transfer less vibration. There also should be some kind of mechanism
to keep the fins from vibrating.

From a performance-to-noise standpoint, the FCLGA4-S is a mediore performer
that cannot compete aginst most of the high performance coolers that SPCR reviews
routinely. Althought it offers higher cooling performance, the Intel FCLGA4-S
has the failings of a typical stock reference heatsink. The Thermalrights, Scythes,
Xigmateks and Zalmans of this world need not worry about an imminent invasion
of their niche by the Intel juggernaut.

Intel FCLGA4-S
PROS

* Good performance
* Light

CONS

* Fan too loud
* Loose fins and fan mounting method create metallic rattling

Our many thanks to Intel
for the QX9650 processor and accompanying FCLGA4-S heatsink sample.

* * *

Articles of Related Interest
Thermaltake V1: “Peacock
Tail” Cooler

Thermaltake MaxOrb Heatpipe
Cooler: Maximum Orbness

Two Big Top-downers: Big Typhoon
VX & Xigmatek HDT-D1284

Xigmatek HDT-S1283 & SD964
“heatpipe direct-touch” CPU coolers
Akasa AK-965 socket 775
tower cooler

Zalman CNPS8700 LED CPU Cooler: Update of a
Classic

* * *

Discuss this article in the
SPCR forums.

Leave a Comment

Your email address will not be published. Required fields are marked *