Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
04/01/08 17:08
Read: times


 
#152848 - not a good ides
Responding to: ???'s previous message
Each iteration through the main function block forces a read from the ADC data buffer to see if there has been any changes and is compared to the current eeprom value and a decision is made to take action or not.

This of course is after the adc is sampled 32 times


That way a single 'noise read' will affect your reading

You have to check the individual reads and 'say' "till n reads y away from old read have been seen, do not use them".

Of course, heavy analog filtering coud do that too; however, a pulse will, with analog filtering, affect the result, with digital filtering it will not.

Erik

List of 12 messages in thread
TopicAuthorDate
Adding hysteresis on top of my sampling            01/01/70 00:00      
   my method            01/01/70 00:00      
   For this to work...            01/01/70 00:00      
      adaptive delta ... ?            01/01/70 00:00      
      All data is keep in eeprom            01/01/70 00:00      
         What does the oscilloscope tell you?            01/01/70 00:00      
         not a good ides            01/01/70 00:00      
   Median filter ?            01/01/70 00:00      
   are you using an RC low pass before the ADC?            01/01/70 00:00      
      check here            01/01/70 00:00      
      Found a minor issue in hardware...not dealbreaker            01/01/70 00:00      
   Low Pass Filter in Software            01/01/70 00:00      

Back to Subject List