I recently dug up my Lego RCX, and installed BrickOS on it, which allows me to make much more powerful programs for it (in C). I was fooling around when I remembered that I had told somebody that was looking for a way to measure a fan's rpm that they could use a RCX for this, so I decided to try it myself...
I picked a random "trash" fan, and painted half the hub white with tippex, put the light sensor on a stand as close to the hub as possible, hooked the fan up to my test PSU (==PSU minus fan, shorted out start-up pin), and a fanmate. I made a program that counts the amount of black-to-white transitions in one minute, checking every 10ms.
Now the results are what bother me: this particular fan starts out at about 1680 rpm, and speeds up to about 1780rpm in 3 minutes, but goes on to about 1900rpm 15 min later.
So does anybody here know if it is the fan/fanmate/PSU that is "warming up" or some problem with my measuring equipment?
PS: I have checked that the light sensor reliably and clearly sees the difference between black and white, I have also put a box over the system to avoid light interference...
Do fans/PSU's have a warm-up time?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
It could be the fan bearings warming up and developing less fricton.
However, at 1900 RPM you're transitioning every 15ms, which is a bit close to the frequency you're checking. It could be that the actual variation is quite small but you're getting odd effects because the frequency you're sampling at is so close to the frequency of the transitions. If you can I'd suggest increasing the sample rate, one every 2ms or less would be better.
However, at 1900 RPM you're transitioning every 15ms, which is a bit close to the frequency you're checking. It could be that the actual variation is quite small but you're getting odd effects because the frequency you're sampling at is so close to the frequency of the transitions. If you can I'd suggest increasing the sample rate, one every 2ms or less would be better.