SilverStone SST-LC01 HTPC Case

Cases|Damping
Viewing page 5 of 5 pages. Previous 1 2 3 4 5

THERMAL / NOISE TESTING

A Zalman Fanmate1 fan voltage controller was installed for the Panaflo on the heatsink as well as the Evercool exhaust fan. These controllers were used to set the cooling / noise level. The CPU stress utility CPUBurn was used to maximize the load on the system. Measurements were taken only at least 15 minutes in each test state.

Conditions: Room temperature was 22°C and ambient noise was 17 dBA.

TEST 1: Preliminary testing was done with both fans getting the full output through the Fanmate1, which is approximate 10.4V on both of these samples.

Panaflo 80L on CPU cooler & Case fan at 10.4V
Activity
CPU
System
PSU Fan
Noise, front*
Noise, back*
Idle
47°C
37°C
6.8V
30
33
CPUBurn
59°C
45°C
6.8V
30
33
* In dBA/1m: Measured 1m from the front and back.

The overall noise level at these fan speeds can be characterized as modest, but not really quiet. There is some high pitched component as well as a lower frequency resonance to the character of the sound. The former is coming from the fans. The latter is not, as you might imagine, caused by the case panel resonating.

Because the large cover is supported all around the edges as well as by the center cross bar, it does not appear to rattle or resonate at all. Damping the cover by placing large heavy books on it had no effect whatsoever on the resonant sound, but removing the cover eliminated the resonance. Hence, the lower frequency noise appears to be the natural cavity resonance of the air in the case, exacerbated by the noise of the fans. It might be tamed by careful application of sound absorbing materials to break up standing waves inside the case.

TEST 2: All three of the fan grills were removed at the full speed test run again. There was no effect except for the thermal auto-speed PSU fan at idle: It dropped to 5.8V, suggesting that the improved airflow helped the PSU run a lit cooler at idle. This change in fan speed was not noticeable because of the noise from the other fans.

One interesting thing to note: The hole for the 80mm case fan is at 2~3mm smaller all around than the actual diameter of the fan blades. Making this hole a bit larger would probably help with airflow and turbulence noise.


Grills removed for better beathability.

TEST 3: The fan voltages were dropped to levels where, subjectively, the overall noise became very quiet, very close to my own low noise reference systems. It would be hard to beat this noise performance substantially without going fanless and using a suspended quiet notebook drive.

Panaflo 80L on CPU cooler at 7V & Case fan at 6V
Activity
CPU
System
PSU Fan
Noise, front*
Noise, back*
Idle
51°C
41°C
6.8V
23
25
CPUBurn
70°C
50°C
6.8V
23
25
* In dBA/1m: Measured 1m from the front and back.

The measured and subjective noise level was much lower than before, with even the lower frequency resonance much tamed. CPU temperature climbed too high for comfort, however.

TEST 4: The fan voltages were raised up from test 3 to bring temps down to a safer level.

Panaflo 80L on CPU cooler at 8.5V & Case fan at 8V
Activity
CPU
System
PSU Fan
Noise, front*
Noise, back*
Idle
49°C
39°C
6.8V
26
29
CPUBurn
64°C
48°C
6.8V
26
29
* In dBA/1m: Measured 1m from the front and back.

The measured and subjective noise level is still pretty quiet. It is at a level that would surprise owners of ordinary computers and please many who are seeking a quieter PC. CPU temperature has dropped considerably, and although considered high by some users, systems generally show no signs of instability at this CPU temperature.

Hard Drive Notes: The temperature of the Seagate Barracuda IV hard drive remained steady at 44~45°C (as read by the utility DTemp) throughout the testing, maybe going up by one degree during long sessions of CPUBurn. It appears that the drive position is far enough away that the heat of the CPU does not affect it much in this system. Granted, a hotter VGA card might change things. The noise of the suspended drive was never a factor; it remained below the level of the fans in all the various settings of the above tests.

ACTUAL USE

The system tried above is only one countless combinations that you might assemble in the SST-LC01. This particular combination was not necessarily ideal, it was simply a convenient combination of components already on hand. Some other components would probably have given better results:

1) A CPU cooler designed for upward airflow (rather than downward) like the Alpha or Scythe Kamakaze and Samurai in combination with a right angle deflector or duct to direct the hot air to the exhaust fan.

2) A PSU with a 120mm fan for increased case cooling.

3) A cooler CPU.

On the other hand most VGA cards today would probably be hotter than the MX440 pressed into service here.

Despite the high temperatures seen during the load testing, my preference was for the lowest noise mode of TEST 3. During normal use over several days, CPU temperatures rarely exceeded 60~64°C peaks with ambient room temperatures of 20~24°C. Activities included playback of some DVD movies, Photoshop image editing work, and lots of web browsing.

CONCLUSIONS

Few cases are truly optimized for low noise computing, and even the best mid-towers have some flaws and compromises in this regard. The keys for a silent computer are

  1. unobstructed, generous airflow potential,
  2. solid, non-resonant design and construction.
  3. no direct path between noise source and the operator's ears, and
  4. hard drive decoupling — either built in or ease of employing it.

In most of these regards, the SST-LC01 is very suitable for a quiet HTPC. With a better combination of components than the ones used for the system test, you can achieve lower temperatures with the same low noise performance as my TEST 3 configuration. This aluminum case is more capable in its fundamentals than most of the HTPC oriented cases and a solid starting point for anyone who wants to experiment and mod for

STRENGTHS

* Good intake vents and airflow pathways. Even with the low speed PSU and back panel fans, airflow could be felt at the intake vents.
* Simple straightforward case layout. Except for the slight annoyance of the center bar, it's easy to install a board.
* Sturdy construction and design minimizes panel vibration despite aluminum.
* ATX PSU compatible. What a relief! So many more choices.
* Plenty of room for HDD suspension. With notebook drives, you could really go to town.
* Room for ducting experiments around intake and exhaust.
* Doors make matching of optical drive front panels less of an issue for the stylistically fastidious.
* Many exposed drive bays. A multifan controller in one would be very useful.

CAUTIONS

* Sticky power button. Not a big deal but it's a detail oversight.
* Only one 80mm exhaust fan.
Probably inadequate for many hot component combinations today. Case modders might look to cut out a hole for another 80mm next to the existing one. There seems enough room.
* Little airflow around HDD area. This could be improved with vent holes on the bottom panel beneath the drive bays. The taller than usual feet will help with intake airflow from the bottom.
* The case is a bit too tall and maybe too deep for some audio / video stands or cabinets. This is probably the price for ATX PSU and motherboard compatibility, and a full complement of drive bays.
* Side vents let out noise. But drop the internal noise enough and there's not much noise to come out; this is the basic trick of silencing a PC anyway.

Our thanks to SilverStone Technology for this SST-LC01 sample.

* * *

Discuss this article in the SPCR Forums.



Previous 1 2 3 4 5

Cases|Damping - Article Index
Help support this site, buy from one of our affiliate retailers!
Search: