Electronics > Projects, Designs, and Technical Stuff
12V Signal On/Off Delay
Ian.M:
That's neat but wont work on a slowly falling input signal as the rate of charge transfer through C1 must develop enough voltage across R1 to turn Q1 on hard enough. I calculate it as needing min 0.022 V/us slew rate which comes out to slightly over 0.45 ms fall time. OTOH if you use a 556, with the other half's Threshold and Trigger strapped together as a Schmidt inverting buffer with 33% Vcc hysteresis, to drive it it will work on even very slowly ramping signals, if you can tolerate the lag caused by the wide hysteresis ban .
eblc1388:
Yes, its nice to know there is a fallback solution should the current simple implementation doesn't work out satisfactorily. It will still be a single chip solution, albeit a 556 instead.
jfiresto:
Look at TI's CD14538 dual one-shot. It has a 3-18V supply range and negative and positive edge, schmitt trigger, trigger inputs. With two RC pairs as edge detectors, you might be looking at the IC and perhaps seven added discretes, six of which are cheap. Or you could use both halves and fire one on positive edges and the other on negative, with independent leading and trailing edge pulse times.
Vakito:
Thanks everyone for the responses so far, I think my best bet would be to use a 4077 XNOR as the rising and falling edge detector followed by a 555 monostable for the timing.
What size capacitor would you recommend between the Vdd and Vss of the chips if any?
Thanks
Ian.M:
Depends. Traditional bipolar 555s (e.g. NE555) tend to 'crowbar' the supply with shoot-through as they switch. The only hope of taming them so they are good neighbors is 10uF to 100uF of low ESR electrolytic decoupling as close to the supply pins as you can get it. CMOS '555's are much better behaved.
CMOS chips however generally only need 10nF to 100nF ceramics for adequate decoupling. If you are using 4000 series at low voltage, their slow transitions mean you can even get away without decoupling every chip if your layout is fairly tight.
Navigation
[0] Message Index
[*] Previous page
Go to full version