hey all i've built the cpemma transistor based fan controller, and it works absolutely fantastic, the problem im having is that some of the fans cut out when the voltage goes below a certain value, i can figure out what that cut out voltage is on the fans, but its tricky figuring out whether or not i've turned the pot too far and the fans cut out, without having to actually dive into the case and have a look at the fan. are there any circuits out there that could perhaps incorporate some form of a motion sensor, that can be mounted close/aimed at the fan so that it outputs a signal when the fan stops spinning or there is no motion, it could work the other way around, i.e, stop outputting a signal when there isnt motion. is there any such a circuit that can be constructed fairly cheaply, and could someone link to a schematic. thanx.
yup, most of the fans do have rpm sensing wires on them, but when plugged into the mobo, or the aerogate controller, and the fan is slowed down, past a certain point, it just seems to stop outputting a signal, but the fan still spins, which was why i thought itd be best if i perhaps tried some completely independant way. thanx.
I'm hoping that's a problem with the acceptable speed range for BIOS/Aerogate reporting, my cpu fan shows 0 RPM in BIOS but 750RPM in MBM5 after tweaking the fan type settings so it's still sending a signal. If not, my next Eye Candy idea fails. Idea is to take the sensor signal to the clock input of a binary ripple counter chip, with a 4.7k pull-up resistor to the supply. A led on the Q4 output would then blink nice but dim. The original full-speed signal also switches the open-collector repeater Q1 which can be connected to the motherboard fan connector for a digital speed result. I like simple. I've still to test, so play at your own risk. A few counters will do, 4024 is the smallest and cheapest.
