Thermalright Copper Heatsinks – Pre-production Test

SUMMARY: Another new copper heatsink joins the “performance elite” – Thermalright.

The good folks at Thermalright sent me a pre-production sample of their heatsinks to test. I am impressed! Assuming Thermalright delivers the same outstanding performance of these samples, I think purchasers will be pleased. I do have one caveat: The outstanding performance was with a Delta 38 cfm fan; this is one I would not recommend due to its highly irritating noise level, as many have found out. Even so, the Thermalright delivers equal or better performance with a quieter 24 cfm fan compared to others using the Delta 38 cfm unit.

Thermalright x Fan

70 fins for a fins surface area of about one square foot (the larger model).

If you look closely at the Thermalright, you will see 70 fins about as thick as an index card; this yields a total estimated surface area of one square foot. LOTS of cooling potential. More important, the fins are soldered onto the copper base which results in very efficient heat transfer between the base and fins, compared to some others that use a press fit.

Thermalright

The SK6 weighs in at about 415 grams – the smaller model (SK6ab) could not be tested – the fins were too close for the clip to ride in its channel.

The package is really quite small – it measures 2 1/4″ square on the base by 2 1/4″ high with the fan. The fan the larger heatsink (SK6) uses is a SUNON that has slightly less airflow than the YS Tech. I tested the SK6 with both it and a Delta 38 using motherboards and a static test block.

My pal Andy Lemont at Millenium is very kindly using his metal working skills to build me a test block that will enable me to do static tests in conjunction with motherboard tests. As shown below,

Testbed

the Static Test Bed comprises a solid aluminum block machined to simulate the footprint of an AMD T-Bird. On the base is a heater that is powered by a lab power supply which can be set to a precise watt load. A second power supply is used to power the fan. The heatsinks are prepared as usual, with a thermocouple bored into the center. Temps are measured by a digital thermometer, in my case an Omega HH23.

This is a close-up of the aluminum block (a great find at my local metals scrap yard) with the Thermalright on it:

Testbed Block

The aluminum block will be sheathed in insulation to avoid heat loss.

Machined into the block are lugs which simulate the socket’s clip loading.

The procedure is turn the power supply on, “dial in” a wattage setting (in this run I used 50 watts – we are adding heaters to get to 70 to 100 watt loads) and let the block warm up with the heatsink powered up. After about 15 minutes, temps stabilize and temp readings are then taken. Results are then tabulated along with results from motherboard tests. It is my intention to include static testing along with AMD and Intel CPUs on motherboards, so that readers will have confirming results from multiple platforms.

THE TEST

The following test results were on an Iwill KK266, T-Bird 1133 @ 1397, 1.9 volts; Duron 800 @ 1000 on an ABIT KT7, 1.9 volts, and Celeron 533 @ 800, 1.6 volts, on an Iwill W02-R – all running Prime 95 to stress the CPU – and a static test at 49.7 watts. A 1/16″ hole was drilled into the heatsink’s base and a thermocouple was epoxied into place over the center of the CPU’s core. Ambient temps are measured 1″ from the heatsink fan’s intake. Noise was measured using a Radio Shack sound meter placed 8″ from the fan intake.

For this test, two fans were used: Thermalright’s stock SUNON fan rated @ 23.5 cfm, 4500 rpm, 33.5 dBA, and the Delta @ 38, 7000 rpm, 46.5 dBA. Noise levels measured at 8″ from each fan’s intake: 53 and 69 dBA respectively.

TEST RESULTS
TEST

CPU Temp

Ambient Temp

Delta

C/W

MBM

Static – Sunon 23.5

33.0 C

19.8 C

13.2 C

0.27

NA

Duron – Sunon 23.5

39.3 C

23.4 C

15.9 C

0.25

38

T-Bird – Sunon 23.5

42.9 C

23.5 C

19.4 C

0.21

44

Celeron – Sunon 23.5

27.1 C

22.7 C

4.4 C

0.22

32

Static – Delta 38

28.5 C

19.5 C

9.0 C

0.18

NA

Duron – Delta 38

35.0 C

23.0 C

12.0 C

0.19

34

T-Bird – Delta 38

37.6 C

23.0 C

14.6 C

0.16

36

Celeron – Delta 38

26.3 C

22.6 C

3.7 C

0.19

32

Delta = CPU temp – Ambient Temp
C/W = Delta / CPU Watts

Interpreting C/W: For every watt the CPU radiates, the heatsink will cool the core by the (C/W x watts) plus ambient temp. For example, at an ambient temp of 25 C, a C/W of 0.25 with a CPU radiating 50 watts means that the CPU temp will be 50 x 0.25 = 12.5 C over ambient temp, or 37.5 C. The lower the C/W, the better.

Outstanding performance with the Delta and very competitive with the Sunon. I noticed that there is good airflow through the fins and onto the motherboard, which aids in cooling power regulators and other nearby chips. The heatsink was almost cool to the touch, with the exception of the T-Bird at 1397 – warm but not hot.

The Sunon fan is much easier on the ears and not too shabby on performance either – it’s about the same level as many heatsinks using Delta 38s to attain C/Ws in the 0.25 range. Even for hot T-Birds, the relatively quiet Sunon is not a bad choice; for Intel’s power misers, the Thermalright with the Sunon looks like a GREAT cooling solution – super performance at low noise.

For performance at the expense of your hearing, the Delta 38 clearly is outstanding; unfortunately, noise at this level is not something I would use, but that’s your choice.

I tested a pre-production sample – basically a hand-made unit. Availability is expected around May 10, so as soon as a production sample is available, we’ll run another test series. MSRP for the SK6 is $45 with fan and $39 without. You can purchase from Thermalright directly when available.

Email Joe


Be the first to comment

Leave a Reply