Overclocking with the CopperPins Heatsink

---

With the making of the CopperPins done it's now time to test out the cooling power it contains. For our tests we are going to be using a PPGA Celeron 300 and a Pentium III 500E FC-PGA. The 300Mhz Celeron is an old processor true, but this particular chip is very overclockable and I know a little more about what it can do, what it has never been able to do, and what it won't do, better then any other processor I have at the moment. The FC-PGA is really just in there for fun. I don't have a mother board with a FSB setting above 155Mhz to really test it out :(

Outline of tests:

Using an MSI-6163 mainboard, some PC-100 ram and an ISA SVGA video card we set about overclocking the respective chips with the Frosty CuPin Heatsink attached.

A small amount of silicon thermal compound was used as an interface between the copper base and the dies/heatspreaders. To attach the heatsink to the socket, wire was looped around the clips of the socket 370, and then through the pins of the heatsink (much like the Alpha pal35 does). Given the weight of the copper heatsink this method mostly just held the heatsink in place, rather then apply any form of significant downward force on the die.

Celeron 300 Tests

First on the testing block is the PGA Celeron 300. Ambient temperature was 24.2 C. The thermistor was located on the surface of the copper base directly above the CPU die, and insulated from the surrounding air with a small bit of anti-static foam.

Remembering that this is just a 300 Mhz processor we can see a nice jump up to a stable 504Mhz. I was able to get the BIOS to post up to 558Mhz at the default voltage. Only a small 1 watt 40mm fan was used in these tests. The reason being to test the heatsink with the minimum of cooling attachments to really judge what the heatsink can do itself. So far it looks pretty good. This particular 300 has always been able to do 504Mhz, but this is the first time it has even posted at 558Mhz. With additional fans and changes to the Vcore it may be possible to inch it up to stable operation there.

Anyhow it performed fairly well, and from the range of temperature values it took all the increases in temp up until around 40 degrees, where changes then became largely muted.