Battery monitoring - ever tried AN1072?

[flotulopex]

Yes George,

For any reason, either ADCON1.ADCS setting will affect the results.

For those interested in the maths (where "2n" is 1024 because I use 10bits in ADC):


Regarding my results and comparing them to the table A-1 (Appendix A: AC Result Table in AN1072), something must be wrong.

A 135 ADC result corresponds to 4,52V and 147 to 4,16V (?!).

Grrrrrr! I didn't notice I hadn't removed the voltage regulator on my breadboard. My PICkit2 programmer delivers 4,78VDC (even if it shows 5V in the "VDD PICkit2" field) so the "new" ADC values I get now seem to be much closer to what they are supposed to be "on the paper".

For 4,82V displayed on my FLUKE87, the ADC result is now 127 (exact as table A-1) and @4,31V I get 142 (exact as table A-1).

((1024 - 1) / 4,82) * 0,6 = 127 and ((1024 - 1) / 4,31) * 0,6 = 142

[Ioannis]

Glad you did it Roger! Bravo!

Ioannis

[towlerg]

For any reason, either ADCON1.ADCS setting will affect the results.

I don't know the internals of ADCIN, but unless it sets the ADCS bits to something else, you're ADC clock source will be Fosc/2 which at 8Mhz will give a Tad of 250ns. Fosc/16 or Fosc/32 are recommended.

George

[Ioannis]

George, If you do not want any headaches, just use the Frc. The internal 500Khz clock is good enough for most applications.

Plus, you do not worry for any timing violations.

Ioannis