Stable ADC readings

[mazoul72]

i am so disapointed because it doesn't work on test real card and works on simu.
strange...
i have microcode studio with 15 days trial .(compiler 3.0.0.1)

TRISA = %00001111 ' RA0 et RA1 en entrée

TRISB = 255 ' Disconnect all resistor from the circuit (Haute impédance)

PORTB = 0 '

'

' ADC

' ---

DEFINE ADC_BITS 10 ' Set number of bits in result

DEFINE ADC_CLOCK 3 ' Set clock source (rc = 3)

DEFINE ADC_SAMPLEUS 50 ' Set sampling time in

ADCON1 = %10000010 ' RA0 à RA3 and RIGHT JUSTIFIED



'

' Software Variable

' =================

ADCRes VAR WORD ' ADC Reading

btn var byte ' Data to be scan

i VAR BYTE '

j var bit

ADCChannel VAR byte '

ADCButton VAR word [3]'

'

' Software constant

' =================

j = 0

ADCSafety CON 10 ' min/max range for ADC Reading

'

ADCNone CON 814 - ADCSafety '

ADCButton[0] = 90 ' button 1,7 or 4,10

ADCButton[1] = 164 ' button 2,8 or 5,11

ADCButton[2] = 270 ' button 3,9 or 6,12



Start:

TRISB = 255

ADCIN ADCChannel,ADCRes ' Read ADC

'

if (ADCRes < ADCNone) Then ' Any button down?!?

' - Yes!

FOR i = 2 TO 0 STEP -1 ' Loop through all 3 calculated posibilities

' Test against calculated ADCresult

' and allow a range of +/- ADCSafety

IF (ADCRes > (ADCButton[i]-ADCSafety)) AND (ADCRes < (ADCButton[i]+ADCSafety)) THEN

' Valid ADC result, clear the coresponding I/O

btn = ((DCD i) << (ADCChannel*3)) ^255

if (ADcchannel = 3 and btn = 255) then btn = ((btn - (3*adcchannel)) << i)

gosub choix

pause 200

i=0 ' and getOut of here

ENDIF '

NEXT '

else ' - NO!

ADCChannel = ADCChannel + 1 'Switch to the other ADC channel

if ADCchannel = 4 then adcchannel = 0

ENDIF

GOTO Start

choix:

select case btn

' premier bloc de 6 boutons
case %11111011 ' AN0-270 --- 47k and 5k = 4.6k

TRISB.0 = 0 ' 5k and 5k = 2.5k

TRISB.1 = 0

case %11111101 ' AN0-164 --- 47k and 10k = 8.2k

TRISB.1 = 0 ' 5k and 5k = 2.5k

TRISB.2 = 0 ' 47k and 47k = 22k

case %11111110 ' AN0-90 --- 47k and 22k = 15k

TRISB.2 = 0

TRISB.3 = 0 ' 10k and 10k = 5k

case %11011111 ' AN1-270 --- 47k and 2.2k = 2.1k

TRISB.3 = 0 ' 47k and 15k = 11.3k

case %11101111 ' AN1-164 --- 100k and 22k = 18k

TRISB.4 = 0 '

TRISB.0 = 0 '

case %11110111 ' AN1-90

TRISB.5 = 0

TRISB.1 = 0

TRISB.3 = 0

' deuxième bloc de boutons

case %11111111 ' AN2-270 --- 47k and 5k = 4.6k

TRISB.7 = 0 ' 5k and 5k = 2.5k

TRISB.6 = 0

case %01111111 ' AN2-164 --- 47k and 10k = 8.2k

TRISB.5 = 0 ' 5k and 5k = 2.5k

TRISB.4 = 0 ' 47k and 47k = 22k

case %10111111 ' AN2-90 --- 47k and 22k = 15k

TRISB.2 = 0

TRISB.3 = 0 ' 10k and 10k = 5k

case %11011000 ' AN3-270 --- 47k and 2.2k = 2.1k

TRISB.7 = 0 ' 47k and 15k = 11.3k

case %11101100 ' AN3-164 --- 100k and 22k = 18k

TRISB.5 = 0 '

TRISB.0 = 0 '

case %11110110 ' AN3-90

TRISB.5 = 0

TRISB.7 = 0

TRISB.3 = 0

end select

return

[mister_e]

Dunno buddy, replace your resistor with LED to Vdd and see what happen, Code seems to be OK.

On your PIC16F877, make sure you have all VDD and VSS pin connected, MCLR to VDD and the right capacitor around the crystal. If you're using an >4MHz one, set the HS_OSC fuse. Always make sure you also disable LVP in the configuration fuse too.

[mazoul72]

Dunno buddy , your program Code works very well with some resistors and one led on my PIC Hardware platform test .
With it , the intensity varies according to different buttons .
Perhaps microcode studio with 15 days trial .(compiler 3.0.0.1) ?
my programmer is willem with winpic800 .

[mazoul72]

i 've just try compiler 2.60 plus patch 2.60c under microcode studio 3.0.0.5
and verify fuse under willem (hs) , always the same .
arg , no understand why if the code is good .........................../

[mazoul72]

STUPID I AM , no pull-up resistors for RA2 and RA3....

[Ioannis]

Post #52

Ioannis

[mister_e]

----======----



----======----