Couldn't read your attachment ..
Anyhow, try this: connect 100kΩ (or more) pull-down resistor between each input and 0V (GND) ..
It is possible that through internal capacitancies one input charges the other inputs ..
Regards,
IanP
I allready use 10k pulldown res but no change was apear.and another problem is that chanel 1-7 was not work good .when i connect them to 5v it shows 4.357
if you have got real life circuit up and runnimng try this file i have set it up the same as your drawing the only thing i'm not sure of is all you transistor and buttons are ment to do
arash_micro
the hex file you can copy it into the proteus folder and select and sim it or burn the hex file into the pic your using and run it in real time (in your circuit) just like yoou would of had to you hex file you would of created
If you can try it and lets us know how you get on wuth it.I'm just Curious to see if it has cured your problem. because normally if it works in proteus it work the same in real life. Now i have tested that on my development board and it works ok to me. see you need to eliminate your software or hardware as i said if it work in proteus then it should work the same real time very rare that i find you get different results
please can u send ur circuit drawing as an image
i think it will be easy to help u if i can see ur HW, anyway ur problem will be
--crosstalk between chanels
--the input reistance of each channel
--ground loops
if u send ur circuit i can send u modifications to try it