New algorithm for hall sensor

I am always facing this issue, I am reading the hall sensor values directly using Adc, but even when the motor is not moving there is small difference of adc count between hall sensors and fluctuating and i am assuming some frequency and performing wrong calculations. To avoid this i want to calculate the slope of the adc counts and if the slope crosses some minimum threshold i will assume it is valid signal else it is some noise. Is it correct method of doing or alternate methods available? Please help.

I am reading the hall sensor values directly using Adc, but even when the motor is not moving there is small difference of adc count between hall sensors and fluctuating

Click to expand...

When dealing with these chip sensors it is always useful to have a calibration procedure in order to set up the same baseline for all of them, and course, store it in any non volatile memory of your system. In addition, the bandwidth of these devices are too high so that it is expected any fluctuation due to noise or even physical small movements not perceptible for us, so I would also think about taking the average of a number of samples and/or adding a small value capacitor at the microcontroller's analog entry.

Hi,

noise and fluctuations my have various sources:
(even three lines of text can´t give useful informations)

* noisy power supply
* wrong schematic
* wrong device selection (OPAMP, capacitors...)
* wrong filters
* bad PCB layout
* noisy sensor
* thermal drift of sensor

Now it´s on you to find the root of error or to cure it´s result.

Good informations are:
* schematic
* layout
* scope picture of error (magnitude, frequency /-spectrum...)
* timing chart
* desription of magnitude in volts or LSB, plus information like Vp, Vpp, Vavg, VRMS..

Klaus

Thank you for the reply. I came to know from online that reading analog hall sensor values using adc is not the best method and need to convert to digital before reading them using some comaparator etc. I want to know if it is correct? This digital method also has the advantage of giving it to input capture module. Right now i don't have the required hardware to perform and meet the above requirement. So i have already compromised on performance?
Next finding the root cause at hardware level. I am more of software than hardware and not fully under my control and don't have very in depth knowledge of hardware so it is very difficult for me to figure out the hardware issues. So very confused and trying to sort out using some software methods.