Good choice for aggressive P4/Socket A cooling
SUMMARY: Top Rank CPU cooling for AMD or Intel, depending on noise preference.
A while ago we took a look at the Thermalright SLK800. Not one to stand still, Thermalright has superseded it with a new version, the SLK-800U. The major difference between them is that the SLK-800U is larger and the mounting system improved over the previous version:
P4 Mounting includes a back plate, mounting “wings”, retention clips and spring loaded bolts while AMD Socket A mounting uses just the back plate and spring loaded bolts; note that for P4 use, the standard P4 retention bracket can be used without the spring loaded bolts. Use of the back plate requires motherboard removal.
Note that the SLK-800U’s base does not fully contact the P4 IHS.
The fan is mounted using wire clips (for 70mm and 80mm fans). Also included are foam pads for AMD CPUs, mounting guides for P4 use and a tube of generic white thermal grease. Detailed mounting instructions are HERE. Note also that some motherboards may require “custom standoffs” – some are listed HERE. In addition, some motherboards may have motherboard components close to the mounting hole – some care must be exercised in this event.
The base finish shows signs of polishing marks:
I could both hear and feel ridges on the base; lapping will improve performance.
One notable factor in the SLK-800U’s favor is the amount of air flowing down through the fins from the 80mm fan overhanging the heatsink’s sides – the top measures 87 x 56mm. There is a significant secondary cooling benefit as a result, and components around the heatsink should run cooler. In addition, since the motherboard acts as a secondary CPU heatsink, there will be a positive impact on CPU cooling.
I tested the SLK800 with Vantec’s Tornado, a Sunon 80 x 38mm fan, model #PMD1208PMB1-A rated at 84.1 cfm @ 5700 RPM, 55.2 dBA, 9.1w; this fan features vanes in the base to straighten airflow off the blades. I varied voltage to get a range of fan speeds and measured rpms using an Omega digital tachometer.
I measured noise using a Radio Shack sound meter placed 8″ from the fan’s intake and recorded noise at a number of rpms – these should not be far off the mark far other 80mm fans at similar rpms (Note: Manufacturers meter sound at 3 feet from the fan, so the dBAs I measured do not correspond to the manufacturers’ ratings – I have normalized my dBA readings for reference).
The SLK800 was tested on the CPU Die Simulators which give results that are unaffected by motherboard influences. I then tested it on an Acorp 4S845A motherboard with a modified P4 1500 to read CPU case temps as an example of what users might see on a P4 system².
| Heatsink | Die Temp | Ambient Temp | Delta | C/W |
| SLK-800U, 5490 rpm, 74/54¹ dBA | 37.9 C | 20.4 C | 17.5 C | 0.23 |
| SLK-800U, 4496 rpm, 68/49¹ dBA | 39.3 C | 20.7 C | 18.6 C | 0.25 |
| SLK-800U, 2993 rpm, 56/41¹ dBA | 42.4 C | 21.0 C | 21.4 C | 0.29 |
| Heatsink | Die Temp | Ambient Temp | Delta | C/W |
| SLK-800U, 5480 rpm, 74/54¹ dBA | 32.6 C | 21.2 C | 11.4 C | 0.15 |
| SLK-800U, 4495 rpm, 68/49¹ dBA | 32.3 C | 20.1 C | 12.2 C | 0.16 |
| SLK-800U, 2992 rpm, 56/41¹ dBA | 34.9 C | 20.4 C | 14.5 C | 0.20 |
¹ The first number is the actual reading, the second normalized to the fan’s dBA rating.
| CPU/Motherboard | CPU Case | Ambient Temp | Delta | C/W | MBM Temp |
| Acorp 4S845A, 5612 rpm | 30.2 C | 21.8 C | 8.4 | 0.15 | 26 C |
C/W = Delta / CPU Watts
Interpreting C/W: For every watt (CPUw) that the CPU
consumes, the HSF will limit the CPU’s temperature rise to (C/W x CPUw)
plus the temperature at the HSF’s fan inlet. For example, at an ambient temp of 25 C, a C/W of 0.25 with a CPU radiating 50 watts means that CPU temp will increase 50 x 0.25 = 12.5 C over ambient temp, or 37.5 C. The lower the C/W, the better.
Die Simulator results place the SLK-800U, running at about 5500 rpm, in the top rank of heatsinks (Heatsink Ranking).
The difference between aggressive cooling/high noise and “reasonable” cooling/low noise is about 5-6 C over 100 watts. For a CPU radiating half that, the difference starts to fade very quickly (note: 38 dBA is the generally accepted “office” noise standard for PCs).
The C/Ws shown here are at high load for at least 30 minutes – the number of times most users even come close to this loading is minimal.
The upshot is that for the AVERAGE user who is not pushing the CPU and spends most of the time for surfing and email, a quiet fan, such as the Vantec Stealth, might do the job OK. Each user must make a choice, balancing cooling needs vs noise tolerance.
Thermalright’s SLK-800U is an excellent choice for aggressive CPU cooling. However, the more aggressive the fan used, the higher the noise level – no surprise. At a minimum, a rheostat might be required if your PC is “24-7”.
The SLK-800U is available from The Heatsink Factory.
² Wiring to read internal Socket A temps precluded mounting the backplate on my Shuttle AK3.1.
Be the first to comment