Viewing page 3 of 3 pages. Previous 1 2 3
DON'T FORGET THE THERMAL FAN!
Just as I was about to write up and post this review, I recalled that the fan speed on previous Intel HSF controlled by a thermistor in the hub of the fan. Hmmm. Given that the temperature in the lab was a cool 20°C throughout the day, I needed to find a way to heat up the fan without going through the rigmarole of installing the test bench system into a case.
The first attempt involved a Canadian copper penny and a 30W soldering iron. The penny was used as a heat spreader so the plastic of the hub would not be burned.
After tiring of holding the soldering iron for a couple minutes with only a marginal increase in fan speed, this method of heating up the fan was abandoned. The big gun was brought out: A 1200W hair dryer.
A thermal sensor mounted on the edge of the fan frame recorded the temp, which jumped from ~26°C to nearly 40°C in a matter of 10 seconds of pointing the hot hairdryer at the HS. The fan speed, which had been 2500 RPM at 12V, jumped to ~5800 RPM almost instantly. Taking the dryer away slowed it back down to ~2500 RPM in less than 30 seconds.
A bit more experimentation showed that even at 30°C, the fan speed climbed up to ~3000 RPM. So with the 30~40°C temperatures that prevail around the CPU in a case, the fan would ramp up and probably keep the Prescott P4-2.8 from climbing not much above 60°C. It does become much louder, with a pronounced screaming whine when the RPM goes beyond 3,000.
There is little point comparing this cooler to any of the top ranked models on SPCR's Recommended HS list. It loses on both cooling and acoustic fronts.
The new "Prescott" stock Intel HSF run at the standard 12V is an adequate product from a cooling point of view. It is somewhat quieter than the previous "high end" copper base Intel HSF we tested, although it also does not appear to cool as well at that noise level. The thermistor control in the fan hub works to ramp up the fan speed very quickly and dramatically at higher temperature, so with poor case cooling or a very hot processor, you'd probably have acceptable CPU cooling but at the cost of very high noise.
As the test results indicate, cooling performance at reduced fan voltage is really not very good, and the noise reduction is not worth the trouble. No one who has a concern about noise would want to use this HSF unless ambient noise is very high.
* * *
Discuss this this article in the SPCR Forums.
|Help support this site, buy from one of our affiliate retailers!|