Exact match. Not showing close matches.
'[OT] Button Toggle Followup'
Just a report on my fix for that button toggle problem, for anyone
This circuit seems to be nice and stable under ESD attacks:
| |\ | |\ 74HC14 (2 gates)
| / | \ | | \
+----+--/ --+--R1K----+--| >O--+--| >O--+---- OUT
| Btn(N.O.)| | | / | / |
| | 0.01uF --- |/ |/ |
| | --- |
_+_ | | |
___ 0.1uF | GND |
| | |
If anybody sees any difficulties with it, please let me know, it's going
in a real product real soon, and I'm a relative newbie at this :-)
I first tried Paul's suggested 100K and .0047 uF values for the RC into
the first gate, but that seemed to completely disable the button, it
wouldn't toggle anymore (and I admit I never tried to figure out exactly
why: my first guess is that the 100K absorbs too much of the charge that
gets transferred on a button press).
I also tried a few other things, and Dwayne's suggestion to just add a
cap of 1/10 the value of the "memory" capacitor worked great in one
direction: if the output was high, I couldn't toggle it with a spark, but
if it was low, it would toggle. I'm sure that's trying to tell me
something...and BTW it was only negative polarity sparks at the lab that
would toggle it.
People also suggested zeners to ground, and that's a great idea.
Actually, the larger circuit is full of them, sprinkled about in
strategic locations, so I think I'm OK. Perhaps another in parallel with
the 0.1uF cap would be a good idea.
Anyway, adding the 1K resistor calmed the circut down completely, I can't
toggle it with a spark in either direction now, at least with my piezo
barbecue lighter, "home grown" approach. I'm going back to the test lab
Monday for some more calibrated zapping :-)
Thanks to all who responded, I appreciate the help.
At 07:31 01/15/99 -0800, Dave Johnson wrote:
>Just a report on my fix for that button toggle problem, for anyone
>[nice circuit snipped]
>If anybody sees any difficulties with it, please let me know, it's going
>in a real product real soon, and I'm a relative newbie at this :-)
i don't see any difficulties, but i'm interested why you would do it that
way, if you're going to a PIC input anyway from the circuit (which could
easily do the toggling) -- or do you feed some other stuff, too?
I have a quick question; What happens if you hold the push button down
- does this circuit auto-toggle? (I'm guessing not, because of the 10k
resistor, just making sure - Checking MY sanity, not yours <G>)
Dave Johnson wrote:
Mark Willis wrote:
>I have a quick question; What happens if you hold the push button down
>- does this circuit auto-toggle? (I'm guessing not, because of the 10k
>resistor, just making sure - Checking MY sanity, not yours <G>)
No, with the button held down the input comes from the center of a
divider made up of 470K and 10K, either from Vdd to GND or vice versa,
depending on the current state. Because the 10K is so much smaller than
the 470K, the input stays below (or above) the gate threshold
indefinitely: no cycling.
Gerhard Fiedler wrote:
>i don't see any difficulties, but i'm interested why you would do it that
>way, if you're going to a PIC input anyway from the circuit (which could
>easily do the toggling) -- or do you feed some other stuff, too?
No, I'm only feeding the PIC, but I basically wanted "sticky" button
presses to make sure I don't miss them, for a couple reasons: first, in
one "mode" a button press needs to wake the PIC from sleep, and I thought
that precluded a software solution (I think I may be wrong about that,
though). Secondly, often the PIC is in a state where timing (cycle
counted) is critical, and I didn't want to add code that only got
executed on a button press: that would mess up my isochronicity (is that
a word? :-) So at the time this seemed a good solution. And it still
seems like a good one.
That said, however, I see now that with some more work I probably *could*
have done a software-only solution, and it would have saved some space on
my very crowded little board, not to mention a little money, at the
expense of some more code (which I have room for, it turns out). Sigh.
But since we're going into production in less than a month, I'm sticking
to my guns :-)
More... (looser matching)
- Last day of these posts
- In 1999
, 2000 only
- New search...