Viewing page 4 of 5 pages.
Previous 1 2 3 4 5 Next
TEST SYSTEM ASSEMBLY
Setting up a system in the PC-610 was fairly easy thanks to
the roomy case. I wasn't too thrilled about the folded sheetmetal type of
motherboard standoffs as I'm used to using the threaded brass studs, but they
worked OK so I guess I shouldn't complain. I had
a difficult time routing the wires and cable.
The fixed right side really puts a crimp on my style and there's not a lot
of good places to route or hide the wiring inside the case itself. Of course
this is completely trivial to the average builder, but it's something that
I obsess about.
Both the front and rear case fans were pre-installed by Coolcases, even down to the nice cable routing job. The fans
are plugged into the pre-mounted Mitron rheobus that was included by Coolcases
as an optional extra. The side fan was not screwed into place (probably due
to the potential for shipping damage) but it took just a few seconds to install
it during the build. All fans were decoupled from the case by the O-ring mounting
system that was added by Coolcases.
The drive rail system for the external drives was easy to use
and install and the installation was made even easier because of the very
easy to remove front bezel. All you do is pivot those two cam levers up and
the front bezel pops right off. This is much easier, and more robust than
the typical split plugs or screws that hold most bezels in place. I mounted
a DVD rom drive in the top 5.25" bay and, being the Luddite that I am,
a floppy drive in the top 3.5" bay. The included rheobus was pre-installed
in the lower 3.5" bay.
|
HDD Decoupling Experiments
As an experiment, I removed cage
from the case and experimented with suspension mounting the drive, hanging
it on 3/16" bungee cords beneath the 3.5" drive bays. This did wonders
for the seek noise. It reduced the overall volume of the seeks as well as
making it sound softer and more muffled than when it was solidly mounted.
The drive temp did go up a few degrees with the suspension mount though. However,
it wasn't anything to worry about because the drive was suspended directly
in front of the 120mm intake fan which provided plenty of cooling airflow
to the drive, even though the fan was running at 6V for the majority of my
testing. HDD idle noise was too low to hear with the case closed, whether solid
or suspension mounted.
|
The hard drive was easy to install in the removable drive bay.
All you need to do is remove one thumbscrew from the bay and it slides right
out the side. There are no grommets on the drive bay itself to dampen the
noise from the hard drive. Grommet-mounted HDD's have become standard for me now; I could easily notice the increased seek noise with the "solid mounted" Barracuda IV in this case.
The seeks were a bit louder and "sharper" than I've become used
to lately.
The PSU bolted right into place using the typical ATX hole patter
for the bolts. My board-specific I/O shield snapped right into place after
I removed the generic one that was included with the case.
The heatsink of choice, the big Zalman 7000AlCu,
just barely cleared the PSU by about 1/4". This is not atypical of mid-tower cases, but slightly different positioning of the motherboard mounting holes would have provided a bit more room for big CPU heatsinks.
As mentioned before, I opted to screw mount the AGP and PCI cards, rather than relying on the quick-mount plastic levers.
The front I/O panel included the typical power and reset switches, along with a power LED and
HDD activity LED. All of these hooked up to the board easily and worked
fine. The front I/O also included a pair of
audio in/out jacks and two USB 2.0 ports. My motherboard has no internal audio
headers so I couldn't hook up the audio I/O wiring but I did connect the two
USB 2.0 headers to my board. The installation was a snap and both channels
worked fine at full USB 2.0 speeds.
Finished. Note how roomy the case is,
even with a full-sized ATX board mounted.
TEST SYSTEM HARDWARE
* Intel 875PBZ motherboard, with BIOS P05.
* Intel P4 2.4C CPU, at stock speed and voltage (1.525v, 66.2 Watts max)
* Zalman 7000AlCu running at 6-7V,
* 2 x 256MB of Mushkin PC3200 Level II RAM running at 400MHz.
* ATI 7500 graphics card, passively cooled.
* 80GB Seagate Barracuda IV hard drive.
* Toshiba SDM116 DVD drive.
* 400W Seasonic Super Silencer, Rev A1 (original super quiet version)
* 2-3 low noise Globe #S1202512L-3M thermally controlled case fans (supplied with case).
Other:
* CPU temperatures read from the internal thermal diode with Motherboard Monitor
5
* Ambient temperature was 21°C
* Idle temps were determined by starting the machine from cold and allowing
the temps to come to a stable point.
* Load temperatures were generated by 30
minutes of two instances of CPUBurn.
TESTING
Now that we've got it together, let's fire it up and listen to it. After all,
that's what SPCR is all about.
The build process itself was relatively uneventful.
All the hardware fit just like it should into this well-built case.
Fans and Controls - While not as effective
as the softer elastomer-type of fan mounts, the rubber o-rings used to damp mount the fans are easier
to use and seem to do a pretty good job of decoupling the fan vibrations from
being transferred into the case. Of courese, it helps that the case panels
are sturdy.
According to Globe's
specs, the 120mm low speed Globe fans fans blow 67CFM at 2000 RPM at 12 volts with
an SPL of 34 dBA. Based on my calibrated ears, I'd say that was right in the
ballpark, but obviously 34 dBA is way too loud for SPCR. Judging by my previous experience with 120mm fans, 67CFM is also well beyond what's needed to cool this system.
All three fans were set to the lowest voltage via the rheobus. I measured this at 5.8-6 volts with my multimeter.
All the fans started and ran fine at this voltage, but when I experimented
with bypassing the rheobus and hardwiring the fans to 5 volts I discovered
that none of them would start at 5V. I elected to keep them on the rheobus. At 6 volts I was getting around 640-650 RPM
for each fan.
With All Three Fans at 6V - The fan noise, while lower than I expected, was noticeable from a meter away. I
could definitely hear the front and side fans running, even though they were
pleasantly quiet. The noise consisted of a very low hum, a slight amount of
wind noise and a very quiet clicking sound.
Without the Side Fan - Not needing the side fan, I removed it and screwed
a piece of 1/8" thick rubber on top of the fan filter to close off the
hole. I then tested the case with just the front intake and rear
exhaust fans, both running at 6V. There was a noticeable reduction in the
noise; it was now quiet enough that I could just barely hear it from a meter
away. Idle and load temps actually improved a degree or two. My thinking is that the side fan, while
blowing additional air into the case, was messing up the airflow so that
the cooling effect was less optimal.
Front Fan Off - Next I to shut off the front intake fan to see how much noise
was escaping through the front grill from the front fan. I've generally found
that a front fan is unnecessary for a well-ventilated case with good exhaust airflow,
and usually just adds extra noise. With the front fan stopped, the noise was reduced another step, but at the expense of a
few degrees of HDD cooling. Ideally, I'd fit a very quiet 80mm
or 92mm Panaflo L1A running at 5 volts to blow a little cooling air over the
HDD without adding as much noise as the larger 120mm fan. I was not able to
try that easily because the fan grill had been punched to 120mm and
I had no place to mount a smaller fan.
Positive Pressure - My last fan experiment was to increase
the voltage of the front fan to about 9-10 volts to give the case something
closer to positive pressure, sometimes recommended for keeping dust
out of the case. At that setting my idle and load temps
were the same but I didn't really have positive pressure yet, according to
the calibrated toilet paper anemometer. I cranked the front fan voltage up
to 12 volts and tried again. Now I actually had air blowing out of the small
holes and cracks in the case, signifying positive pressure.
The case and CPU temps actually went up a few degrees. This
result is the same as I've found on other well ventilated cases. The positive
pressure may tend to reduce dust buildup, but at the expense of the best airflow
pattern inside the case.
FAN EXPERIMENTS SUMMARY
|
Temps
.
|
3 fans
|
2 fans, no AGP intake
|
rear exhaust fan only
|
positive pressure
|
|
Idle
|
Load
|
Idle
|
Load
|
Idle
|
Load
|
Idle
|
Load
|
|
Case
|
27°C
|
36°C
|
26°C
|
35°C
|
26°C
|
36°C
|
28°C
|
39°C
|
|
CPU
|
28°C
|
50°C
|
27°C
|
47°C
|
27°C
|
48°C
|
30°C
|
52°C
|
|
HDD
|
33°C
|
33°C
|
33°C
|
33°C
|
37°C
|
39°C
|
34°C
|
35°C
|
Overall, I'm pleased with the temperatures in the PC-610. The supplied
Globe fans running at 6 volts are almost as quiet as my reference "OEM"
120mm Panaflos at 5 volts and seem to move as much air. The temps in the PC-610
using the most optimal fan configuration are basically the same as when this
system is in my reference Antec SLK3700BQE case.
As a final experiment, I replaced the rear Globe fan with my normal OEM Panaflo
to see how this affected the temps and the noise. Running the Panaflo at its
normal 5 volts gave me exactly the same temps and very slightly lower noise than with the Globe fan.
If I also replaced the front 120mm Globe fan with a very quiet
80mm fan I'd still have cool HDD temps and probably
the same noise as in the SLK3700BQE case.
Thermal Fan Speed Control - Sharp eyed readers will recall that the supplied Globe fans
are thermally controlled. So how well does that work?
The fans come with a thermal probe attached at the end of a
14" long lead. I initially set up the rear
fan with the thermal probe suspended about 1" above the CPU fan intake.
The speed at 6 volts was around 640 RPM. After running two instances
of CPUBurn, the speed of the thermally
controlled fan did not go up at all even though the case temp increased
about 10°C to around 35°C.
Figuring that this temperature was too far below its
thermal threshold, I moved the thermal probe to the bottom of the Northbridge heatsink. The speed of the fan then increased to a whopping 660 RPM under full load.
So for
plan "C" I wedged the thermal probe right down against
the base of the CPU heatsink itself, also touching the IHS of the CPU.
In this configuration I got an RPM increase of about 80 RPM, to around 720
RPM under full load. This is a little over 10%.
I tried the same thing with the other two fans and got basically the same
results.
I don't know if the thermal ramping is constrained by the low voltage
because I didn't try the same experiment with the fans at 12 volts. [Editor's Note: Mostly likely, the thermal controller was made ineffective by the low voltage.] These
fans are too loud to consider using them at 12 volts in a quiet system so
I figured it didn't much matter.
| Help support this site, buy from one of our affiliate retailers! |
|
