adc stability for volt meter

late_voyager

New Member
well here we are again. board was printed and built. 2 of my volt readings are very stable, great thanks everyone... but now one reading is wandering from 150 to 700+ on Word variable W0. this value is greatly changing even if nothing connected
when 24v conected i have stable 1.49v at pin c.3?
i am using same 1/15k divider
run the test code and wandering
Code:
#picaxe 20M2

init:
pause 200
FVRSETUP FVR2048      'Set Fixed Voltage Reference = 2.048 Volts 
ADCCONFIG %011         'Set ADC to FVR
let b26 = 200         ' button pause delay
Symbol bbatt = w4        'Word variable for the calibadc10 reading
Symbol sens = w8
symbol crane = w0        ' b0-6
symbol multi = 10        'multiply
symbol div = 32    ' div

Main:
debug
readadc10 C.3, crane ' 24v input
        crane = crane  *39
    bintoascii crane,b2,b3,b4,b5,b6
       serout C.0,N2400,(254,192)
       serout C.0,N2400,(" ",b4,b5,".",b6," volts    ") 
    pause 10
goto main
[code/]

[ATTACH=full]22739[/ATTACH]

nothing connected...
[ATTACH=full]22740[/ATTACH]
 

Attachments

darb1972

Senior Member
Did you CAD your own design? If so, did you run a design error check? The error can pickup on things such as micro breaks in the tracks. Eg: you route the trace to pad/s and although it looks connected to the human eye (particularly when not zoomed in), the track falls just short of the pad. Hard to pick/see without error checking.

Did you etch your own board? Again, as above, depending on many variables it is possible that an issue (that's hard to see) can occur with the remaining tracks and pads.

Unless you have done so, I would suggest running a meter (on ohms) over ALL of your traces, preferrably putting the probe in the centre of each pad.

I would suspect from the description of your fault, that the divider in question might have a poor earth/-Vcc track connection.
 

late_voyager

New Member
Did you CAD your own design? If so, did you run a design error check? The error can pickup on things such as micro breaks in the tracks. Eg: you route the trace to pad/s and although it looks connected to the human eye (particularly when not zoomed in), the track falls just short of the pad. Hard to pick/see without error checking.

Did you etch your own board? Again, as above, depending on many variables it is possible that an issue (that's hard to see) can occur with the remaining tracks and pads.

Unless you have done so, I would suggest running a meter (on ohms) over ALL of your traces, preferrably putting the probe in the centre of each pad.

I would suspect from the description of your fault, that the divider in question might have a poor earth/-Vcc track connection.

I designed board but professional check/made/ they check for broken links etc. it seems strange as getting 1.5 v to pin c3...
seems i get no change on adc reading. but get a change on 1.5 to 0volt on pin... maybe 20m2 is toast?
 
Last edited:
Top