CoolerMaster Hyper 48 HSF: 478, K8, 775

Cooling
Viewing page 3 of 3 pages. Previous 1 2 3

INSTALLATION

The stock P4-478 HS retention frame was used as the motherboard is equipped with a sturdy CPU socket backplate and sits flat horizontally for testing. The clips were very simple to use, and the process only took a minute or two. Except in a very tight case, there is no need to remove a socket 478 motherboard for installation unless you are going to use the retention frame and backplate supplied with the Hyper 48. For assembling a new system, the retention frame and backplate supplied with the Hyper 48 are recommended.


Hyper 48 clipped in place on standard socket 478 HS retention frame.

TESTING

The core of the test system is very similar to that used in the past:

  • Intel P4-2.8A The Thermal Design Power of this P4-2.8 (533 MHz bus) is 68.4 or 69.7W depending on the version. As the CPU is a demo model without normal markings, it's not clear which version it is, so we'll round the number off to ~69W. The Maximum Power, as calculated by CPUHeat & CPUMSR, is 79W.
  • Intel D845PEBT2 motherboard - Intel 845PE Chipset; on-die CPU thermal diode monitoring
  • Panaflo FBA08A12L1A 80mm DC fan
  • nVidia GF400MX VGA card (AGP)
  • OCZ DDRAM PC-3700, 512 MB
  • Seagate Barracuda IV 40G 1-platter drive (in Smart Drive from Silicon Acoustics)
  • Enermax UC-A8FATR4 multifunction monitor/fan controller w/ thermal sensors
  • Seasonic Super Tornado 300 (Rev. A1)
  • Zalman Multi-Connector (ZM-MC1) and Fanmate1 voltage controller
  • Arctic Silver Ceramique Thermal Compound
  • Two-level plywood platform with foam damping feet. Motherboard on top; most other components below. Eases heatsink changes and setup.
  • CPUBurn processor stress software
  • Intel Active Monitor and Motherboard Monitor software to show CPU temperature

The ambient temperature in the test lab was 21°C. Ambient noise in the lab was ~20 dBA. Maximum load temperatures were recorded ~20 minutes into a CPU stress test with CPUBurn.

Two fans were used to test the performance of the Hyper 4: The stock fan and a Nexus 92 . The Nexus is the quietest 92 mm fan we know of, and despite its somewhat low airflow, is becoming a defacto 92mm reference at SPCR in the absence of other really quiet 92mm fans.


The other fan: Nexus 92.

RESULTS: STOCK FAN

CoolerMaster Hyper 48 w/Stock Fan
CFM @ Voltage
Load Temp
°C Rise
°C/W MP
°C/W TDP
SPL & MP3 files
40 @ 12V
42°C
21
0.27
0.30
28 @ 9V
44°C
23
0.29
0.33
17 @ 7V
49°C
28
0.35
0.41
9 @ 5V
63°C
42
0.53
0.61
18 dBA/1m

CFM @ Voltage: Measured in free air with digital anemometer (+/- 10%) with the fan running at the specified voltage
Load: CPUBurn for ~20 mins
°C Rise: Temperature rise above ambient at load
°C/W MP / TDP: Temperature rise per Watt, based on Maximum Power (79W) or Thermal Design Power (69W) rating of CPU
SPL: Sound Pressure Level in dBA/1m measured with high accuracy B & K SLM at 1 meter
MP3 file: High resolution recording of fan on HS at 3" distance. Note that it is preferable to save these sound files to your hard drive, and then play them.

The stock fan is surprisingly quiet. It is by far the quietest CoolerMaster fan we've encountered. At stock 12V, it's nowhere near CoolerMaster's claim of 18.5 dBA, but it is a very respectable 25 dBA. Most of the noise is the whooshing of turbulence, as you can hear with the MP3 file above. There is also a fair amount of hissing or white noise not directly related to the turbulence, which may be caused by the bearings. Still the overall character of the noise is fairly benign; it may be quiet enough even at 12V for many users. Cooling performance is very good.

Dropping the fan voltage down to 9V still gives good cooling results and a significant noise drop to 21 dBA/1m. Again, the character of this noise has more white noise than is usual with a fan at this speed. Check the MP3 file to hear it for yourself.

At 7V we see the cooling capability begin to taper, but it is still usable. A 28°C rise with this CPU is still quite acceptable. The noise is now probably quiet enough for all but the most finicky SPCR readers.

At 5V, the stock fan simply does not provide enough airflow for safe cooling of this CPU in a closed case. It would be unwise to use with anything but lower power systems.

RESULTS: NEXUS FAN

CoolerMaster Hyper 48 w/Nexus 92 mm fan
CFM @ Voltage
Load Temp
°C Rise
°C/W MP 79
°C/W TDP 69
dBA/1m & MP3 files
25 @ 12V
46°C
25
0.32
0.36
15 @ 9V
49°C
28
0.35
0.41
10 @ 7V
56°C
35
0.44
0.51

CFM @ Voltage: Measured in free air with digital anemometer (+/- 10%) with the fan running at the specified voltage
Load: CPUBurn for ~20 mins
°C Rise: Temperature rise above ambient at load
°C/W MP / TDP: Temperature rise per Watt, based on Maximum Power or Thermal Design Power rating of CPU
dBA/1m: SPL measured with high accuracy B & K SLM at 1 meter
MP3 file: High resolution recording of fan on HS at 3" distance. Note that it is preferable to save these sound files to your hard drive, and then play them.

The Nexus fan blows less air than the stock Delta. The 12V airflow of the Nexus is comparable to the Delta at 9V. Not surprisingly, the cooling performance is very similar. So is the noise: Only 1 dBA/1m separates the 12V Nexus and the 9V Delta. But listening to the two MP3 files should tell you what I heard: Despite the close loudness level, the Nexus has a gentler smoother noise signature with much less hissiness.

The Nexus at 9V has a CFM similar to the stock Delta fan has at 7V. Here we see identical cooling performance and the same small SPL difference.

Performance at 7V with the Nexus is borderline; again, we wouldn't recommend this setup for a hot processor. There was no point to run any tests at 5V; the cooling performance would not be adequate.

DIRECT COMPARISONS

Here is a comparison of the Hyper 48 against two well-known competitors:

The Zalman 7000Cu, one of our recommended HSF that a lot of people are familiar with. Its integrated 92mm fan produces ~12 CFM and ~20 dBA/1m at 5V, which is the recommended quiet setting for this HSF.

The very similar (to Hyper 48) Thermalright SP94 with a Panaflo 80L @ 9V, which produces ~20 CFM and ~19 dBA/1m. The 9V drive level for the fan is chosen because it closely matches the loudness level of the Zalman at 5V.

The Hyper 48's stock Delta fan produces ~20 CFM airflow with noise at around ~20 dBA/1m at 8V, the voltage used for this comparison.

The data on the other heatsinks is derived from the same testbed, much of it from this review: Thermalright SP94, SP97 & other heavyweights. Note that in the table below, there are links to MP3 files of the Zalman and the Thermalright HSF with their respective fans at the given voltages.

Comparison of Cooling Performance @ ~20 dBA/1m
Zalman 7000Cu @ 5V
Thermalright SP94 w/ Panaflo 80L @ 9V*
CoolerMaster Hyper 48 w/stock fan @ 8V
°C Rise
°C/W TDP
°C Rise
°C/W TDP
°C Rise
°C/W TDP
26°C
0.37
27°C
0.39
26°C
0.37
* Note that at the same 20 CFM airflow level, the Nexus 92mm fan is very slightly quieter (1 dBA/m) and smoother sounding than the Panaflo 80L. Conversely, at the same 20 dBA/1m SPL, it blows a bit more air than the Panaflo 80L and so a bit of improvement in cooling can be expected. This is not likely to be much greater than one degree Celcius.

What is really interesting is that at the same low noise level, these three heatsinks perform almost identically, at least with the P4-2.8 CPU used here. The only real difference is in the Zalman 7000cu: The integrated fan is louder for a given airflow, but the heatsink itself needs less airflow to achieve the same cooling. In other words, its louder fan is compensated by the higher cooling efficiency of the embedded fan / radial design.

FINAL CONCLUSIONS

The CoolerMaster Hyper 48 is most definitely a very good heatsink. It has admirable cooling properties, comes with an excellent stock fan, and is quiet enough even at 12V to satisfy many quiet PC enthusiasts. It is chunky but not bigger than many other high end heatsinks. Fit should not be a problem in most cases.

The heatsink is not without some downsides. The weight is a concern: At 864 grams, it's nearly double the recommended limit put forth by AMD and Intel. Moving the system around with the HSF mounted would be asking for trouble.

The exclusion of a fan control device (like the FanMate included with the Zalman 7000Cu or the simple speed pot in the CoolerMaster Ultra Vortex) is also a downside. The noise at stock speed isn't bad, but it wouldn't be up to snuff for a silent enthusiast. The HSF performs admirably at 9V or 7V, so one would assume CoolerMaster would want to showcase such performance. We'd like to see this heatsink shipped with some way to slow down the included fan.

Lastly, the current market price is a bit steeper than the competition, but that could change.

All in all, the Hyper 48 is a great step forward for CoolerMaster. We hope to see more heatsinks with such quiet behavior and good performance from them in the future.

Pros

* Excellent cooling performance
* 25 dBA noise floor with stock fan at 12V
* Reasonable size
* "Tried and true" design

Cons

* Too heavy; exceeds HSF weight limits
* No included variable fan rate capability
* A bit pricey for the performance?

Much thanks to CoolerMaster for the Hyper 48 sample.

* * *

Comment on this article in our Forums.



Previous 1 2 3

Cooling - Article Index
Help support this site, buy the CoolerMaster Hyper 48 from one of our affiliate retailers!