Thank you Olin for reading my postings and replying.

   I agree with you, but with some remarks. *IF*s are not so big as
it may seems.
   Price of SPDT is quite low. To check it, I bought today new
chinese mouse with 3 SPDT, PS2 cable, a lot of components
inside for slightly more then one US$.
   I crashed one SPDT just for curiosity. It can't "perform
break with make" simultaneously by design.
   "Hysterisis dead band of the schmitt trigger input" is as described
in DS30292B-page 156 (for PIC16F873/874/876/877,
PIC16LF873/874/876/877):
Input Low Voltage with Schmitt Trigger buffer 0.2VDD
Input High Voltage with Schmitt Trigger buffer 0.8VDD

   Nevertheless, I agree with you, that even in this case software
debouncing should be implemented as you mentioned.

   Let me keep for a while my private REPLY TO address in my
postings to the list, since as I've noticed, James Newton also
did so. Look at REPLY TO address of his posting:
[SX]: SXLIST EMBEDDED STARTER CONTEST: April Winner(s)

   Let us ask James Newton (Admin 3), what option is preferred for the list.

   Mike Singer.



Olin Lathrop wrote:

>Please get rid of you private REPLY address to that default replies go to
>the list.
>
>>    Schmitt Trigger is a thing to construct with a bounceless input
>> by hardware design. I've attached a .jpg with it. May be I'm wrong?
>>    What is the sense in programmatic debouncing with a Schmitt
>> Trigger. Is not it enough simply to desable interrupt while interrupt
>> handling?
>
>Your circuit will work *IF*:
>
>1 - The switch is SPDT as you showed.
>
>2 - The switch performs break before make.
>
>3 - The voltage resulting from the resistor divider when the switch is
>completely open is guaranteed to be within the hysterisis dead band of the
>schmitt trigger input.
>
>That's a lot of IFs, particularly the one about needing a SPDT switch.  This
>discussion had to do with the more normal low cost SPST switch to ground
>with a passive pullup to power.  If using port B with internal pullups
>enabled, this requires no other parts than the switch itself.  This scheme
>does, however, require software debouncing if you are trying to resolve both
>the press and release states.

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