Noisy output
The ADC Pi is an Analogue to Digital converter for the Raspberry Pi
21/08/2020
Posted by:
Manbehindthemadness
So I'm using the built-in exemplary Kalman filters from the get hub repository. The output is still fairly noisy, is there a way to adjust the number of samples that are being evaluated to get a smoother measurement? My application requires the measurement fluctuation within +-2.5 degrees.
22/08/2020
Posted by:
andrew
Could you post a link to the Kalman filter you are using so I can see how it works?
If you are getting excessive noise it may be better to try and eliminate the noise before it enters the ADC instead of trying to filter it afterwards.
Check the power supplies you are using for the Raspberry Pi and the sensor to see if they are noisy. You could try adding capacitors or LC filters to the power supply to smooth the output and reduce high-frequency noise.
You could try adding an RC low pass filter between your sensor and the ADC to reduce noise levels. Analog has a good article on filter design for ADCs on their website that may help you with your problem.
If you are getting excessive noise it may be better to try and eliminate the noise before it enters the ADC instead of trying to filter it afterwards.
Check the power supplies you are using for the Raspberry Pi and the sensor to see if they are noisy. You could try adding capacitors or LC filters to the power supply to smooth the output and reduce high-frequency noise.
You could try adding an RC low pass filter between your sensor and the ADC to reduce noise levels. Analog has a good article on filter design for ADCs on their website that may help you with your problem.
22/08/2020
Posted by:
Manbehindthemadness
I will have a look at that article, I am using a Kalyan filter based on the one you supplied in the OEM code repo out on github. The power supply is a regulated 2s1p 7.4v 11ah lipo pack with discharge protection down stepped to 5v 3a via UBEC.
22/08/2020
Posted by:
Manbehindthemadness
Note: documents in Portable Document Format (PDF) require Adobe Acrobat Reader 5.0 or higher to view.
Download Adobe Acrobat Reader or other PDF reading software for your computer or mobile device.