Viewing page 6 of 7 pages. Previous 1 2 3 4 5 6 7 Next
THERMAL TESTING AND SONICS
Thermals and noise are what it all comes down to with a case that's being reviewed
by SPCR, so let's see how things turned out with the P150.
After everything was assembled, I ran Prime95 for 24hrs to make sure things
were stable, and then I began my thermal testing. The results show no big surprises,
with the 100-ish watt output of the 3.0GHz P4 dumping a lot of heat into the
Load Temps, w/ CPUBurn
°C rise refers to the rise in temperature over the
21°C ambient at load.
Over the course of the past few months I've had nearly this exact
same set of hardware running in an Antec SLK3700BQE and the new Antec Sonata
II (except for the PSU and the front intake fan):
- A Seasonic
Super Silencer400 with one mod: A Panaflo M1A was swapped for the stock fan. This is the quietest PSU I've ever used.
- 120mm Nexus front intake fan at 5V
- DIY suspension for HDD.
- Supplied Antec SmartPower II 450 PSU
- HDD mounted in stock drive sled on 3/8" thick Sorbothane strips
- No front intake
All three systems were using a conventional
single 80mm rear exhaust fan PSU configuration, along with a Tri-Cool exhaust
fan running at its lowest setting. Because of the different PSU's and intake fan setups being used, this isn't
exactly an "apples to apples" comparison, but it should be fairly
close. Here's a chart showing the results:
Comparative System Temps w/ CPUBurn
The board temp in the BQE and
the Sonata II were the same, while in the P150 it was 7°C
cooler. The CPU temp in the BQE and the Sonata II were
very nearly the same, while in the P150, it ran 6-7°C cooler.
Since the PSU fan CFM was nearly the same for all three different
samples, the board and CPU temps are most probably affected by the impedance in the airflow path through the case. Judging by these results, the P150 has better airflow than either of
the other cases.
The HDD temps are interesting. The BQE was configured with
a 5V 120mm Nexus intake fan mounted directly in front of the fully suspended
Samsung SP80 HDD, with the normal BQE drive case almost entirely
cut away for improved airflow. The Sonata II's HDD was mounted on a set of 3/8"
x 3/8" x 3" Sorbothane strips and had no front intake fan installed.
The P150's drive was mounted in the stock suspension with a 5V
92mm Nexus Real Silent fan mounted in front of the drive.
mounting vs. fan location was virtually identical between the BQE and the P150. The 5V 120mm Nexus should put out somewhat more airflow than the 5V 92mm
Nexus in the P150 rig. Yet the BQE's drive temp was 2°C warmer. Why? My best guess is because
the P150 has lower airflow impedance airflow than the BQE, particularly through the bezel.
The HDD temps in the Sonata were higher, most likely because there was not fan for the HDD in the Sonata. This is due to the oddball positioning of the
front intake fan on the back side of the Sonata's HDD cage, nearly at the center of the
case. It also doesn't help that the Sonata II's drive cage is configured in
the "rotated 90°" layout. It allows easy access to the HDD with the side panel off, but it
blocks more airflow through the 3.5" drive cage. I tried a 5V 120mm Nexus fan on the Sonata II drive cage to see
if it would help things. It did lower the HDD temp a bit,
but at the expense of slightly warmer board and CPU temps. I elected to run
without a front fan for the remainder of my testing.
The Antec HDD suspension worked as well as any of my various "home-brewed"
setups, with absolutely none of the hassles. It eliminated any HDD vibration conduction to the case, which is
always a problem when any HDD is bolted directly into the
case. It sharply reduced the already quiet seek noise of the Samsung,
and nearly eliminated rotational or bearing noises. This suspension
is wonderfully effective and really easy to set up. Antec gets a
big "A+" for this one.
Neo HE 430 PSU
The PSU was a big unknown going into
this testing. While some of the newer actively cooled Antec PSUs have been noticeably
quieter than their earlier offerings, none have been "world class"
when it comes to low noise. The specs and design of the Neo HE seemed to indicate
that this may be a winner, but the proof is in the testing.
From idle to any typical sort of system load, the Neo HE is a very
quiet PSU. For comparison, its a bit quieter than the Tri-Cool fan running
on the lowest of its three settings. The 80mm Neo HE fan has a slightly
higher pitch noise signature than the 120mm Tri-cool at low. It's also a very smooth sounding
fan with barely any perceptible mechanical noise. The vast majority of
its very quiet sonic output is the gentle woosh of air turbulence.
After loading the system at 2xCPUBurn load for about 20-30 minutes (at
about 165W of power usage according to the Power Angel) the Neo HE's fan
started ramping up in speed. The noise signature remains smooth, but the
air turbulence noise began to be audible over the sound of the Tri-Cool case fan. At an
ambient temperature of 21°C, the board temp is only in the mid to high 30°C
range when the PSU fan starts to ramp up. This ramping occurs sooner, at a lower temperature than with
the fan-modified Seasonic Super Silencer 400 PSUs that I'm used to. At full load, the exhaust temperature
of the Neo HE is noticeably warmer than that of the Seasonic under the same
These observations are based on my rather high expectations
for PSU noise, based on the extremely quiet versions I'm used
to working with. The Neo HE is actually very quiet, and what noise you can
hear is smooth and inoffensive. with the slightly earlier than normal
fan ramping activity, its one of the few (if not the only) things I can find
to even sort of gripe about on the P150.
A HIGHER POWER SYSTEM IN THE P150 (by Mike Chin in the SPCR Lab in Vancouver)
Antec supplied another P150 sample for acoustic measurements. The components used by R.H. were not available in the lab, so I decided to assemble a system a bit more thermally challenging from components in the lab:
The distance between the back case fan and the Scythe Ninja HS
is 1.5~2" here.
The system was left running Windows XP Pro for a couple of hours before testing was begun. When 2xCPUBurn processor loading was run for >20 minutes, the AC power draw of the system reached 221W, nearly 60W higher than the test system R.H. built. The hard drive temperature stabilized at a perfectly cool ~33°C throughout testing; no disk-intensive activity was done.
The first test was done with no fan mounted on the Sythe Ninja heatsink. This was similar to one of the configurations tried in the P180 Part 2 review, with an Intel 660 processor rated for 115W TDP, considerably higher than the D 820's 95W TDP. In the P180, with the top vent left open and only the back 120mm fan running, the Ninja HS without a fan was enough to keep the Intel 660 CPU from throttling, at 65°C.
The second test was done with a Nexus 120mm fan mounted on the Ninja HS, blowing towards the back case fan.
The overall noise at idle for both tests was dominated by the back-mounted exhaust case fan. The Tri-Cool 120 measures 20~22 [email protected] in the low setting in free air, but screw-mounted in almost any case, the measured and subjective noise goes up. A touch of low level, low frequency noise could be heard. My guess is that this is most related to the resonant frequencies of the air in the case being excited by the vibrations of the hard mounted fan. The PSU could be heard only from very close to the back of the case, and the WD Raptor HDD could be heard only as a noice source during seek.
The ramping up of the Neo HE 430 PSU fan noted by RH in his system test was clearly seen here as well. The fan voltage went up almost 1.5V, with a corresponding increase in measured noise. The PSU fan was the only variable noise source in the system, so all the SPL increase can be attributed to the PSU. The increased noise was audible, but fairly subtle.
During the first test with the heatsink run fanless, the CPU temperature reached the reported 76°C
within a scant 10 minutes. CPU throttling was not indicated by the Throttlewatch utility, and the continuing CPU temperature rise seemed to indicate no throttling. If the CPU speed and voltage was throttled, the temperature would have stopped rising. The CPU load test was stopped for fear of CPU damage. This was a bit disappointing, but not completely unexpected. Unlike the P180 in which this fanless configuration proved thermally successful, the PSU in the P150 is directly over the CPU, and there no intake vent over the heatsink.
Putting the Nexus 120 fan on the HS and running it at ~7V using a Zalman Fanmate fan controller solved the thermal problem. The increase in noise caused by the extra fan was just barely audible, but only with some effort at close distance.
I experimented with the extra holes on the back panel over the PCI slots. Leaving them uncovered, I could feel air being pulled into the case through those holes, by the back case fan and the PSU fan, presumably. Closing them up should improve CPU temperature perhaps, I thought, but this was not the case; there was no thermal impact on the CPU. It's quite possible that things could be different with a VGA card in the PCIe or AGP slot. Especially a hot video card.
* * *
The thermal results indicate that the P150 does not quite match the P180 for cooling efficiency, at least not with a hot processor. The P180's main advantage is the separate cooling / acoustic isolation chamber for the PSU that allows the CPU to be unaffected by the PSU heat (and vice versa) and makes a top 120mm fresh air intake or exhaust possible for the CPU as well.
Noisewise, the HDD suspension is the P150's great feature. But a notebook HDD in the suspension is a bit of overkill unless all the other components in the system can be kep at around or below the 20 [email protected] level. You'd have to start by soft mounting and perhaps reducing the speed of the stock TriCool back panel fan. Then the CPU and VGA heatsink fans would also have to kept that quiet, and finally, you'd want to keep the total system power draw to under ~150W AC.
The P150 has reportedly been under development for the past 1.5 years. It was put on the back-burner during the development of the P180.
Now that the P180 has hit the market and seems to be a roaring success, Antec
has finally decided to release the P150.
Some of the P150's design is said to be based on lessons learned during
the design of the P180. Whether by coincidence
or design, in the P150, Antec seems to have tackled some of the complaints about
the P180 that I recited at the very start of this article.
Never mind the P180; this review is about the P150... and I really like it. Aesthetics aside, the
P150 does a good job of doing what we want it to do: Provide a foundation for a quiet PC. The damped side panels, the quiet
PSU, the suspended HDD mounting system, the cable management hooks and the well-designed front bezel/filter/fan setup ? all of these together make an excellent
platform for a quiet system.
|Help support this site, buy the Antec P150 Mid Tower Case from one of our affiliate retailers!|