The switch points for ones and zeros for a (non Schmidt trigger) input is
<=3D 1/3 VCC =3D 0
 >=3D 2/3 VCC =3D 1
 > 1/3 VCC and < 2/3 VCC =3D unknown
The input could switch anywhere between 1/3 VCC and 2/3 VCC.  For any=20
given input, however, it SHOULD switch at the same point each time.  1=20
to 0 might be different from 0 to 1.

In the circuit shown, when switch 10 (top switch) is pushed, the=20
capacitor will charge up to 1/2 VCC!  Therefor, there is a 50-50 chance=20
that the input will never reach the 0 to 1 switch point.

You could fix this by tying Rcal to VCC instead of ground, at the cost=20
of having to do RC timing each time you check for a switch closure.

Kerry




YES NOPE9 wrote:
> Here is the circuit.......
> https://plus.google.com/u/0/photos/102396176247280453443/albums/590179856=
8029457921/5901798571057911522?pid=3D5901798571057911522&oid=3D102396176247=
280453443
>
> There may be up to 10 pushbutton switches.  Only one may be pressed at a =
time.
> Each of the resistors [ R0 thru Rn ] is 10K.  C1 is 0.1 uf .  Rcal is 100=
K .
> At startup the PIC configures a pin as an output and pulls C1 to a *one*.
> Then the pin becomes an input and the PIC measures how long the capacitor=
 takes to become a *zero*.
> This time is used as a calibration value.   ( should be around .01 second=
s )
>
> Periodically the PIC looks at the pin.  IF the pin has become a *one* the=
 PIC sets the pin to zero.
> Then the pin becomes an input and the time to charge C1 to a *one* is mea=
sured.
> The time measured determines which button has been pressed.  If the time =
is 10% of calibrated time , then button nearest C1 has been pressed.  A tim=
e of 20% indicated the next button up.
>
> Is the zero/one voltages for the pin stable enough to do this ?  Can I av=
oid hysteresis ?   Or is there a stable relationship between one to zero ..=
.. and zero to one ?
>
> Perhaps I will have to measure three times to confirm the correct button =
was determined.
>
> thanks for any comments .....
>
> Gus
>
>
>  =20

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.