Monostable with Configurable Multifunction Gate

"D from BC" schreef in bericht news: snipped-for-privacy@4ax.com...

Suppose next circuit will do:

|| +------||----+ | || | | | | .-----. | | | | | +---|C | | | | | | +---|B Y|--++-- | | | | ---)--o|A | | | | | | | '-----' | .-. | | | - | | ^ '-' | | | ---+-------------+- created by Andy´s ASCII-Circuit v1.24.140803 Beta

petrus bitbyter

Is that a diode on the output? If so, I'm not clear on it's purpose. The output is push pull. Dunno if inputs B and C are ok with the negative voltage from the capacitor. I suppose internal clamps might protect the ic.

Other than that, looks good. Alternatively, I think it still works if input B is tied high.

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

The pulse goes into InA. The Y output feeds a capacitor that feeds back to InB A resistor pulls InB low. InC is tied InB

Petrus posted the same..

Thanks..

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

By any chance do you know why schmitt gate logic like the NL7SZ57 have an almost 2:1 max/min range (for Vcc~5V) for the thresholds?

Even the

quad nand schmitt imo has a huge Vt+ range and Vt- range for just

25C. Is there that much process variation? I don't think this is good for designs that rely on predictable thresholds.

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

It's probably a lot better in practice. The spec is such that it doesn't need to be tested at manufacture. Only simple-minded tests are done to ensure connectivity. Nothing else is tested.

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
 I love to cook with wine     Sometimes I even put it in the food

DF

The thresholds and gm have to match quite well to land it in the middle of the span. The gate has an NMOS and PMOS in series fighting over which way the line is supposed to be. The one driven by the input is stronger and always wins at some point.

It doesn't take much process variation to move the number around a lot and testing it isn't so easy. A wide spec means that they can be sure of it "by design" and only test for a workin gate.

Yup.. I'm thinking that too.. I wouldn't be surprised if after testing

10 IC's I'll discover the trigger points only have a small deviation.

btw.. I have to make a correction to what I wrote earlier. The schmitt nand from above almost has a 2:1 ratio for the threshold min to max. The NL7SZ57 is not that bad. Ex: Vt+ at Vcc=5.5V,25C is min=2.61V and max 3.33V. That's tighter than the hc132 schmitt nand.

I'm conserned with the trigger level range cause I'm making an overvoltage shutdown circuit and I want to trigger at 100V. An attenuator knocks 100V down to 2.61V. (2.61V being the lowest Vt+ trigger level on the NL7SZ57.)

Calculating the overvoltage trip range.

2.61V/100V = R2/R1+R2 = 3.33/Vout

Vout=127.6V would be the max overvoltage voltage if in the unlikely event the input has the max trigger level of 3.33V.. Uhhh.. it's ok for my app..

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

"D from BC" schreef in bericht news: snipped-for-privacy@4ax.com...

Yes, it is a diode and it should be connected to the B and C inputs instead of to the output just to protect them.

|| +------||----+ | || | | | | .-----. | | | | | +---|C | | | | | | +---|B Y|--+---- | | | -----)--o|A | | | | +---+ '-----' | | | .-. - | | ^ | | | '-' | | --+---+--------------- created by Andy´s ASCII-Circuit v1.24.140803 Beta

petrus bitbyter

.PDF

=A0 =A0...Jim Thompson

I have used a HCxx schmitt as a low voltage lockout. Within a batch the thresholds have a lot less spread than from batch to batch.

You can do better with trusting the E-B drop of a transistor to hold still. I'm thinking of a common emitter PNP stage.

You may consider making a discrete Trigger-Smith with better control of the thresholds, rather than relying on the not-so-well specifyed thresholds of the NL7SZ57. A rail-to-rail (or clamped) comparator might help.

Pere

Yup.. I've also considered using a 555. D from BC myrealaddress(at)comic(dot)com BC, Canada Posted to usenet sci.electronics.design