On Sunday 08 June 2003 15:52, Dave Dilatush scribbled:
> Why the transistors?  They just complicate things.  Seems to me
> you can do this with resistors and switches only:

Yes, you're right.  However, 2 of the "switches" are already transistor=20
switches triggered from external signals (w/non-5V levels).  But yes, I c=
an=20
eliminate the transistor on the third.


>
> SW_1 >-----R1-----o------> To PIC comparator input
>
>
> SW_2 >-----R2-----o
>
>
> SW_3 >-----R3-----o
>
>
>                  R4
>
>
>                  GND
>
> In the above, the other sides of switches SW_1, SW_2, and SW_3
> are connected to +5V and the Voltage Reference Module of the
> PIC16F628 is operated in the high range (VRR bit =3D 0), where
> VREF =3D 1/4 * VDD + (VR<3:0>/ 32) * VDD.
>
> With R4 =3D 10K for example, you could make R1 =3D 4.7K which would
> effectively disable the other switches as you described, by
> lifting the PIC comparator pin to about 3.40 volts.
>
> Making R2 =3D 22K would give around 1.56 volts when SW_2 is
> pressed; and if R3 =3D 15K you'd get 2.00 volts for SW_3 and 2.64
> volts when both SW_2 and SW_3 are pressed.
>
> With the comparator threshold levels roughly 156 mV apart, it
> would seem there's enough separation between the input levels to
> tell them apart reliably, no?

Seems to.  I was analyzing slightly differently by thinking of the # of l=
evels=20
possible (f/comparator ref using the internal Vref) and the range in whic=
h=20
these fall, rather than the actual voltage.  The resolution of the compar=
ator=20
seems pretty good but only in a certain range (ie: not all the way from 0=
V to=20
5V).  I guess the non-switched resistor will play a good part in adjust t=
he=20
range of signals I send to the comparator.  I'll have to experiment a bit=
 to=20
see if practise follows theory.

Thanks,
-Neil.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads