Some Sort of Multivibrator

out

__________ _| |_____

The circuit does pulse extension and works with DC too.

If input is steady H, the output is steady H.

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

Thee 555 is inverting, meaning you'll need to add parts to it for H ->

H.

Surf for retriggerable one-shot, and buy what you want off-shelf.

...Jim Thompson

As in Fairchild MM74HC4538. (Most all semi houses make it.)

...Jim Thompson

Lets see if I can specifiy your problem.

You seem to want a circuit whose output will go high shortly after the input goes high, and stay high for as long as the input is high, and remain high for 8msec after it has gone low.

If the input goes high a second time during the 8msec of pulse stretch, the output should remain high, and rmain high for 8msec after the falling edge of the second pulse.

This implies that you want to OR the input with the positive going output of a retriggerable monostable - like the 74xx123 - which is triggered by the falling edge of the input pulse. The 74123 has a finite, if small, propagation delay from input to output - for the Philips 74HC123 typically 30nsec at 4.5V, worst case 51nsec - which would put a narrow, negative-going glitch in your output.

For many applications this won't matter; for those where it does a minimally delayed version of the input pulse could be OR'd with the other two signals to fill the gap.

The advantage of the 74123 over your RC network plus diode is that the monostable period is much more precisely defined; the tolerance on your improved monostable has to reflect the - large - tolerance on the switching threshold of whatever NOR gate you use to realise the circuit.

Hope this helps.

-- Bill Sloman, Nijmegen

Yup..that's it.. Ahh..Those are the waves I should have posted at the beginning so it would be clear. Hopefully my posted circuit problem presentations will improve..

If I construct this with a quad schmitt NOR, I have a spare NOR to use as an oscillator (My app also needs a crude triangle generator.)

Here's a crappy alternative:

Schmitt trig. input OR gate used. Note: C discharges through diode leakage. R put in parallel with C for shorter times.

This circuit may not function for very short pulses due to the C charge up time. iows.. No pulse extension for very short pulses. The source is from a logic IC with a 35mA max Voh spec. (I don't know yet how short the source pulses will be.) Fixable with more input current.

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

Sure it retriggers but the output clears after a time when the input stays high. The ltspice circuit I posted has an output that stays high for as long as the input is high.

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

I'll think about that for a bit..

Here's the logic I'm looking for:

The output goes H and stays H when the input goes H and stays H.

Rising edges reset the timer.

Falling edges start the timer.

The ouput falls low only after a time period after the input falls low.

I guess the circuit could be called a positive logic pulse stretcher.

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

If you read thru the ap notes you'll find "pulse stretching"... RTFM ;-)

...Jim Thompson

I looked at the waveforms on page 3. It'll stretch short pulses but not stretch long pulses.

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

Repeating myself, surf on "pulse stretching".

...Jim Thompson

A simple way to do this with a 74xx123 + timing R/C + a say 2N7002 is to use the 74xx123 backdoor. The input will be the gate of the 2n7002 MOSFET; the drain of which will connect to the R/C common point, the source to GND. The monoflop will stay with Q=3D1 while there is a 1 on the gate of the MOSFET (perhaps something with a lower threshold voltage than the

Dimiter

------------------------------------------------------ Dimiter Popoff Transgalactic Instruments

------------------------------------------------------

Original message:

This is just a test - I posted a message a few mionutes ago on this thread via google and it appeared on their index, but not as a message (this happens frequently if one checks immediately after posting). But this time the message did not appear somewhat later, rather, it disappeared from the index (unprecedented to me, hence the test message).

Dimiter

Thanks.. I'll consider this too..

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

Yup..that'll work too but I hoping to get all the logic from 1 IC (as originally posted) except for the R and C..

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design

Then it's a delay off that is resetable. a simple task for a 555.

Just invert the input Pulse to the trigger via a NPN (open Collector) or (Emitter follower) trans..

Look at the bottom for the "Missing Pulse Detector" Figure 10A

that will generate the results you're looking for.

I want to make an amendment to the circuit I pointed to.

The discharge pin needs not be in use, but tie the trigger and threshold together along with the way it is now, this way the input pulse discharges the cap each time before it reaches the threshold. This will give a consistent time since it'll hold the cap from charging while the pulse is on. Of course, common emitter NPN should be used for this, to be driven from the input pulse.

If the + input can provide the needed load current, a retrigerrable 555:

In ---+----->|--------+-----Load | | | ----- | | | 555 | | +-----|trig | | | out|---+ ----- | [Cap] | Gnd ------------------+

Yeah that works too. If I get too much supply rail noise, I'll why. :) Rising and falling edges might be on the slow side but that might be ok.

D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design