Typical brushless DC fan RPM myth? or am I a dork?

[bbiandov]

Hello,

For years I have assumed that what the MB sensors and monitor application report is actualy true - that is in terms of CPU and chasis fan RPM. The numbers that I have seen range from 3000 to 6500 RPM. Today however I was hooking up one of those DC brushless fans to AVR project that I am working on and my software was counting 150 RPM. I though I was wrong but then I hooked up the fan to my HP freq counter and what do you know - 300Hz come out of the RPM sensor wire (knowing a pusle is produced every time the field is switched, devide by 2 and you get RPM). So then I though the fan is bad, hooked up another one and got close numbers? What is going on? Am I seeing DA TRUTH or do those little fans actually go to 4000+ RPM and if yes, how can I explain my measurements? THanks

 

[Roff]

You seem to be comparing revolutions per minute to cycles per second (Hz).

 

[bbiandov]

Yep, that is correct. two cycles = 1 revolution. I must thank you for point the M in RPM out. I always though somehow it is revolutions per second. Now that I multiply the value by 60 in order to get the cycles per minute it comes to 3900 RPM. Thanks again!.

FYI here is the math (130Hz/2)X60=3900 so 130Hz coming out of the sensor wire is in fact correct. The fan is switching the coils 130 times per second to achieve rotation (using its internal hall sensor IC and H-bridge) and it takes two switches to achieve full 360 degrees so the fan is doing 65 rotations per second.