Originally posted by _John_Lennon_+8 January 2004 - 02:36--></div><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td>QUOTE (_John_Lennon_ @ 8 January 2004 - 02:36)</td></tr><tr><td id='QUOTE'> <!--QuoteBegin-lynx@7 January 2004 - 22:27
84375 rpm?

That&#39;s the sort of reading you get with a crap fan (it&#39;s not Speedfan&#39;s fault).

What? Intel standard? Nuff&#39; said&#33;
Look, speed sensors ARNT THAT COMPLICATED.

If speedfan cant read anything under a 20 dollar fan, than I dont think its the fan company&#39;s fault. [/b][/quote]
Exactly my point, speed sensors are not complicated at all.

Speedfan can only read the info presented by the sensor - if the sensor gives a false reading then Speedfan will report it.

One of the drawbacks of pwm speed control can be the inability to give accurate tachometer readings if the HARDWARE is not designed properly.

The tachometer works by switching the signal on and off to create pulses, and the sensor counts the pulses. Pwm speed control works by switching the supply voltage on and off, but at a much greater rate. If the pwm switching is allowed to pass through the tachometer to the sensor, the sensor will be measuring the pwm switching, not the tachometer.

This can be prevented by using a couple of very cheap components in the fan design, but since pwm speed control is only just becoming commonly used (for computer fans) it has not been important in the past. Fan makers who have not updated their designs will suffer from this sort of problem.

As you see, it IS the fan company&#39;s fault, but price doesn&#39;t come into it. I suspect some expensive fans may well suffer from the same problem simply because they weren&#39;t designed to cope with pwm speed control.