Thermaltake CL-P0025 Silent Tower CPU heatsink/fan

After two and a half years of heatsink reviews, we finally tackle one from the most prolific of aftermarket CPU cooler brands. Thermaltake’s Silent Tower is one of the new breed of tall heatpipe-equipped designs with fan airflow that goes across the board rather than down or up. They’ve incorporated some nice design features and a couple of real muffs. Are the snafus fatal?

November 27, 2004 by Ralf
Hutter with Mike Chin and Sean Boyd in the sound lab.

Thermaltake is one of the most prolific
aftermarket suppliers of PC accessories. The brand is well known for the “Xaser”
line of aluminum cases as well as a wide variety of heatsinks and
newer line of power supplies. The latest trend in high performance heatsinks, driven by the prodigious heat output of the latest processors,
are very large designs using heatpipes. Never one to miss out
on a trend, Thermaltake have recently released several colossal new heatsinks
and a few of them could be of interest to Silent PC Review readers.

The Silent Tower is a large, actively cooled, heatsink that uses a
copper base and heatpipes to transfer heat to 59 aluminum fins. It’s
covered by an aluminum shroud that acts as the mounting point for a 92mm fan
molded from virginal white plastic. While it looks like weight would be a problem,
it’s actually not too bad at
640 grams, fan included.

The base of the Silent Tower.

Thermaltake claims six copper heatpipes are soldered to the base and also
clamped from above by a copper plate, but it really appears to be three pairs of pipes in a U-shape. As best as I can tell, the 59 aluminum
fins are only press fitted over the heatpipes and are not directly soldered to them.

Heatpipe attachment to base. Clamp also serves to position the mounting
bracket.

The overall footprint of the Silent Tower is small, despite its lofty height.
The aluminum fin section starts 1.75″ above the “keep-out” area, which
provide plenty of clearance for any normal memory DIMMs, power supply components
and all but the most massive northbridge heatsinks. The prospective buyer might
want to check out case dimensions though, as the 138mm height of this heatsink
may interfere with smaller cases.

The blue anodized fan shroud is made up of three stamped sheets
of .040″ thick aluminum that have been pop-riveted together. The shroud
is attached to the heatsink by four machine screws that bolt it onto the top
aluminum fins. The fan shroud also serves as the attachment point for the included
92mm fan. The opposite side of the shroud is drilled and tapped to enable the
mounting of an auxiliary 92mm or 80mm fan for increased air pressure through
the heatsink fins.

The fan is a Thermaltake-badged Panaflo FBL, which uses the patented
LAC Augmented Fan
design under license. It looks virtually identical to the 80mm version found
on Thermaltake’s successful Silent Boost heatsink. This type of fan design
supposedly increases airflow without increasing noise by making use of the blade-edge
turbulence normally lost in conventional frame fan designs. The fan is of fixed
speed and terminated with the standard 3-pin motherboard connector and includes
rpm monitoring capability. Cross-referencing the Panaflo
FBL spec sheet (PDF file link), it looks like the TT fan is somewhere between
the stock Panaflo FBL Medium and High speed models, in terms of
amperage and CFM. Considering that Panasonic lists the M at 30 dBA/1m,
and the H at 35 dBA/1m, TT 21dBA spec looks absurdly low.

Silent Tower with shroud removed. Cooling fins are slid over heatpipes
and stacked on each other.

Thermaltake also includes a complete set of mounting hardware
that includes all the brackets, bolts, pads and other assorted parts that you’d
need to mount the Silent Tower on your favorite K7 (holes in the board around the CPU socket required),
K8, Socket 478 or LGA 775 motherboard. All the mounting methods
clamp the heatsink directly to the motherboard. This is probably a
good idea due to the weight and height of this heatsink, but unfortunately means
that the motherboard will have to be removed from the case in order to mount
the heatsink.

Also included is a 2 gram tube of thermal compound labeled “Powerful
Thermaltake Compound“. How powerful it actually is will have to remain a
mystery because, as with all my heatsink reviews, I used Arctic
Silver Ceramique. Thermaltake also provides a lavishly illustrated instruction
manual along with the heatsink. It’s quite complete but isn’t designed to be
used by anyone with 40+ year old eyesight. I had to break out the old magnifying
glass to make out of many of the pictures and diagrams.

INSTALLATION ISSUES

The Silent Tower was mounted on my standard P4 heatsink test motherboard. Assembly was
fiddly and not straight forward.

• A steel “H”
  bracket, a rubber pad and an antistatic plastic sheet all mount on the
  underside of the board with four long studs. These are all fastened from the top
  side of the board using four nuts and a set of fiber washers.

• Next the HSF is
  placed on top of the CPU and the top “H” bracket is slipped into position
  over the four protruding studs and into the slot on top of the base of the heatsink.

• Then the whole thing is clamped together by tightening four little nuts down
  onto the protruding studs.

This sounds fairly easy but in practice it’s not.

The biggest problem is that there’s no way to tell when you’ve tightened the
top clamping bracket enough. The mounting studs are fully threaded so there’s no way to tell how much tension
is “too much”. It doesn’t help that the top “H” bracket
flexes as you tighten up the nuts. This adds a mushy feel to the whole operation
making it quite difficult to tell how tight things really are.

Another issue
is that the entire heatsink itself can slide from side to side while tightening
up the nuts due to the design of the clamping bracket. Even with the board out in the open, it’s difficult to
access to the tiny nuts on the side nearest the northbridge heatsink. Thermaltake did not supply a wrench to make this easier. I used a set
of needle-nosed pliers, which worked, but not easily.

When I had it tightened up as much as I thought was “enough”,
I noticed that I could still slide the heatsink from side to side. After I got
the system running, I experimented with tension. Tightening the
nuts down so much that the heatsink wouldn’t move didn’t make any difference in CPU temperature. With this much tension, the top bracket became quite
noticeably bent, which didn’t make me feel comfortable.

Mounting hardware. Note the clamp underneath the board to distribute
the force.

All the above applies to mounting a P4 CPU with a large integrated heatspreader
on top of it. If I was mounting this on an AMD CPU with a bare core, I’d be
extremely concerned about chipping the core. It’s too hard to tell if you’re applying
pressure evenly across the CPU and when to stop applying pressure.

The weight of the very
tall heatsink will apply a lot of cantilever force to the CPU core when the motherboard
is mounted in the typical upright position. I didn’t see a problem with the test
setup, but the P4 (and the Athlon 64) has the sturdy heatspreader to distribute the clamping
force, something that a bare core CPU like the AMD XP doesn’t have.

My heatsink reviews are conducted with the reference motherboard sitting
flat on a piece of antistatic foam. For this review I also tested the heatsink
with the board mounted in a vertical position to see if there was any difference
in cooling ability: There wasn’t.

Mounted. Note the clearance in the “keep out” area and the overall
height.

ON THE TEST BENCH

For this test I compared the 92mm Silent Tower to my sort-of-reference 92mm fan equipped
Zalman
7000AlCu heatsink. I used the same processor that I’ve been using in
all my heatsink tests, a Pentium 4, 2.4C. It may not put out as much heat as
a new Prescott, but using this processor will allow me to directly compare the
performance of the Silent Tower against all the other P4 heatsinks that I’ve
tested during the past year.

I was not able to test the Silent Tower in full accordance with the standard
SPCR testing method, which calls for the HS to be used with SPCR Reference
80mm Panaflo L1A. Nothing but 92mm fans will fit on the intake side.
I decided to test several different types of quiet 92mm fans on
the Silent Tower, to get a good feel for its performance. I used the included
Thermaltake augmented fan, as well as a 92mm Panaflo L1A and a 92mm Nexus. I also set up the heatsink using two Nexus 92 fans in
a “push-pull” configuration, which Thermaltake claims gives
better cooling.

Ye Olde Test Setup.

HS TEST PLATFORM & PROCEDURE

The P4 HS test platform is an open system not enclosed in a case.

Intel P4-2.4C Northwood core – Maximum power is 66.2W.
Intel 875PBZ motherboard – Intel 875P Chipset; on-die CPU
thermal diode monitoring
ATI Radeon 7500 passively cooled video card (AGP)
Mushkin PC3200 Level II – 2 x 512MB DDRAM
Seagate 80GB Barracuda IV hard drive
Seasonic SuperSilencer 300W (rev A1) PSU modded with 5V Panaflo
M1A
Arctic Silver Ceramique Thermal Compound
Two-level metal platform with rubber damping feet. Motherboard on top; other
components below.
CPUBurn processor stress software
Motherboard Monitor 5.3.4.0 software to track CPU temperature
and fan speed

Each heatsink was cleaned and installed as per the
manufacturer’s and Arctic Silver’s instructions. Prime95 was then run for
8 hours to verify system stability and cure the Ceramique. The system was
then shut down and not restarted until the next morning when the actual
testing was done. The system was allowed to cool between tests for 30 minutes.
Each test was run for 30 minutes even though all temperatures generally
stabilized within 15 to 20 minutes.

Each heatsink was tested three times on consecutive mornings to check
to the consistency of the results. All results were within 1-2°C of
each other and the average readings are included in the charts.

Ambient temperature was measured at 71°F (21°C) over the entire series
of tests. No tests were run unless the ambient temperature was at that reference
level.

* All temperatures in degrees Celsius.
* Diode: Reading from P4-2.4C CPU diode via Motherboard Monitor.
* Temp Rise refers to the difference between ambient temperature and the
diode reading. .
* °C/W refers to the °C of temperature rise per watt of heat dissipated
by the CPU.

COOLING AND ACOUSTICS FINDINGS

I tested the heatsink with each of the fans at 12V, 7V and 5V, mounted in the
default Thermaltake “blowing” orientation. I also did one series of
tests using two of the Nexus fans mounted in a “push-pull” configuration.

The first column on the left contain links to sound recordings in MP3 format. The second column shows the sound pressure level of each fan, measured at a distance of one meter in a room of ~16 dBA ambient noise. This information comes from the SPCR Acoustics Lab, with the data and recordings provided by Mike Chin and Sean Boyd. The stock Silent Tower fan was not measured or recorded because it was not present in the lab.

It may be enlightening to download the MP3 files and listen for yourself. You don’t have to take my word about what they sound like.

AT 12 Volts, the HS cools very well with every fan. Unfortunately, all but the
92mm Nexus are too noisy for serious consideration by SPCR readers. The stock
Thermaltake is the loudest by a large margin, followed by the L1A Panaflo,
then the Zalman with the Nexus coming in as the quietest by far. The Thermaltake
fan is particularly obnoxious, with a loud whine, a raspy buzz, noticeable
clicking and a lot of air noise. By no stretch of the imagination could this
fan be considered “silent”, or even quiet.

The
Panaflo cools just as well as the stock fan, but had no raspiness, less whine
and much less air turbulence noise. So much for the “augmented” fan design giving
more airflow with less noise. The 92mm Nexus is the standout, giving excellent
cooling performance at very low noise levels. The Nexus at 12V is very quiet,
perhaps as loud as the Zalman 7000 fan at 6-7V.

AT 7 Volts, the Thermaltake and the Panaflo cool quite well, but the
Thermaltake fan is not much quieter than at 12V. It still
has the clicks, whine, buzz and turbulence noise, albeit at a bit lower level. This excessive air turbulence noise seems to be a common trait on every
augmented fan I’ve ever heard. The L1A isn’t too bad at 7V, but is certainly
louder than the Zalman fan even though both provide the same cooling performance.
The Nexus is nearly silent at 7V, but obviously running out of steam; the cooling performance is barely acceptable.

AT 5 Volts, the stock fan was finally becoming somewhat
quiet but it still had a distinct whine, a low metallic hum and audible clicking,
with little air noise. The Panaflo at 5V is, well, a 92mm Panaflo:
Not one of Panaflo’s quietest with an audible clicking and a bit of whine,
but is probably sufficiently quiet for many less-than-hardcore silent PC builders.
The Zalman is subjectively quieter than the Panaflo with slightly better cooling ability.
The Nexus is super quiet but at the expense of any useful airflow.

WITH HOTTER CPUs

Thermaltake rates the Silent Tower to “P4 478, up to 3.2Ghz”,
but I only tested it with the (relatively) moderate output of the 2.4C processor.
We can however, use the °C/W numbers to extrapolate the theoretical
performance of the Silent Tower HSF to any other P4 processor, including the
100+ watt P4 3.2GHz Prescott. According to Intel’s “TDP” specs, the 3.2
Prescott puts out 103W. The higher “MP” power rating for the 3.2Ghz Prescott
is 115 watts so we’ll use those numbers to calculate the estimated maximum CPU
temperature for a 3.2E running 2x CPUBurn:

These extrapolated temperatures show that the Silent Tower with the stock fan would effectively cool even a toasty 3.2E Prescott processor, at >7V fan settings. It’s a shame that the stock fan is so loud. Swapping
the quieter Panaflo 92L gives similar cooling but
whether it’s quiet enough at 7V is up to the individual user.

DUAL PUSH-PULL FANS BETTER?

I mentioned earlier that I tested the Silent Tower in a two fan “push-pull”
configuration. Thermaltake
promotes this as a feature and includes two sets of pre-tapped mounting holes
on the “downwind” side of the shroud, one in a 92mm pattern and the
other in an 80mm pattern. Since the cooling ability with the stock fan and the
Panaflo is quite good but too loud, I decided
to test the push-pull configuration using 92mm Nexus fans.

The Nexus 92
is very quiet, but lacking in airflow at lower voltages. A pair in a push-pull configuration would increase the air pressure through the heatsink fins,
possibly improving cooling with less loss of airflow induced by back pressure.
I tested this configuration at all three reference fan voltage levels and at
each level got about 1-3°C better cooling compared to running just one fan.
That’s slightly better performance, but at the expense of a tiny bit more
fan noise. Is it worth it? You be the judge.

FINAL CONCLUSIONS

Thermaltake offers a mixed bag with the Silent Tower. Once installed properly, it certainly
performs well even with fairly low airflow, and its design lends itself to being
used in a ducted system for efficient system cooling. However, the awkward mounting system
is a near-fatal flaw. I’d never use this on a CPU without a heatspreader. It’s hard to recommend even for heatspreader-protected CPUs because there are many other HSF options at similar or better prices that provide as good or better cooling and acoustics without the mounting hassles. It’s not as if well-proven heatsink mounting mechanisms don’t exist: It’s yet another example of engineers who have forgotten about using springs to control clamping force.

SPCR has always insisted that after-market heatsinks must be appraised not only for the quality of the main heatsink but also the ease and efficacy of the mechanical mounting system Without providing the typical user a positive way to achieve safe, correct pressure between the heatsink base and the CPU, even the most awesome heatsink metallurgy cannot ensure adequate cooling performance. The Silent Tower and its mounting system is akin to a high powered sports car with skinny tires that belong on a soapbox racer.

Despite the “Silent” nomenclature,
as with so many similarly labeled mainstream products, the stock fan misses the mark by a huge margin. If the mounting system was better
engineered, and if it was sold without a fan, I’d be a lot happier
with it. As it stands, the Silent Tower is difficult to recommend except for hardcore modders if it can be obtained at a fire sale price.

Our thanks to Thermaltake
for this opportunity
to review the Silent Tower.

* * *

Discuss this this article in the SPCR Forums.