Hey,
This is the hardware:
1 Intel 4 wire pc-fan
1 pull-up resistor (to 5v), from a HALL(open-colector output)
The fan has 2-pulses per revolution, from a HALL, and while viewing on the osciloscope i saw that instead of nice spike-s, or something square-wave like, there was a spike and is oscilated down to 0.
Does anybody know the problem?
I don't know if it makes a difference but the Intel fan spec calls for a pull-up to 12v.
Shouldn't matter for open collector - any voltage above Vsat ought to work... Does sound odd.
Is the motor being powered? If not the 12v power is also used to power the hall sensor.
The pull-up is to 5v, but at 12v is the same, only diference is a bigger spike, the motor is powered from 12V. What I want is a nice clean spike, ore something like a 5% on time square wave.
Can you post a photograph of the oscilloscope trace?
kurama:
The pull-up is to 5v, but at 12v is the same, only diference is a bigger spike, the motor is powered from 12V. What I want is a nice clean spike, ore something like a 5% on time square wave.
Perhaps a low-pass filter (signal through resistor and capacitor from input pin to ground) would make for a neater pulse. You could further condition the pulse with an analog comparator (high-gain op-amp with the negative input set to a level below the ringing).
I was thinking to use a Schmitt triger, because for a low-pass filter i would need to make a precise cut-off frequency.
because for a low-pass filter i would need to make a precise cut-off frequency.
No ... why should it matter if your cut off frequency is say 20Hz or 20.1Hz.
However these things are designed to be counted simply so I am assuming you are doing something wrong. A picture of what the trace is might give us a clue as to what it is.