SN74LVC1G123 retrigeebale monostable timing

[yassin.kraouch]

Hi,
what is the maximum outputpulse width that i can put on this restriggable monostable ?
https://www.ti.com/lit/ds/symlink/sn74lvc1g123.pdf
and what is the maximun current needed in pin A and B ?
if i want to have an output pulse duration of 3min, i need a big capacitor and resistor, right ?

 

[FvM]

It's about factor 1000 beyond datasheet specified time intervals, so the you can't rely on the datasheet to determine if it's possible and how.

It might work, possibly needing some protection means to isolate the large capacitor. But it's easier to refer to a timer that supports longer delays by design, e.g. a CMOS 555.

 

[yassin.kraouch]

yes but the CMOS 555 has not the feature of retriggable mode, it can't be retriggebale,

 

[yassin.kraouch]

any help please ?

 

[Analog Ground]

A delay of 3 minutes should be done digitally if you want accuracy and repeatability. Have you ruled out the 74HC5555 suggested in your other post?

 

[yassin.kraouch]

why you say it should be done digitally ? why it cant be done by a capacitor + resistor?

 

[FvM]

why it cant be done by a capacitor + resistor?

I can be done purely analog.

If you have designed an analog 3 min timer, you'll be interested to use a digital time multiplier along with the analog timer. Can you guess why?

 

[Analog Ground]

It can be done with a resistor and capacitor. However, the best accuracy and repeatability is achieved with a digital (counter) solution. The resistors and capacitors tend to have low tolerance and high temperature drift. Precision resistors and capacitors cost more, etc. My comment is simply a digital solution is more stable, accurate and requires less tweaking. In the case of a 74HC/HCT5555, it also may have a lower parts count and take up less room on a board.

 

[yassin.kraouch]

why ? you see this single retiggable monostable is small, and you need just to add res+cap in order t get the time, for example 10megohm and 0.16uf will give you around 180 sec, but the digital timer is 16 pins and is much bigger,

- - - Updated - - -

sorry the link http://www.nxp.com/products/logic/multivibrators/series/74LVC1G123.html

 

[FvM]

10megohm and 0.16uf will give you around 180 sec

Something wrong with your pocket calculator? Time constant is 1.6 sec, may result in about 1 sec monostable interval.
But does the chip work with 10 Mohm? The highest timing resistor menitoned in the lvc1g123 datasheet is e.g. 200k.

 

[yassin.kraouch]

sorry i mean 16uF, but they did not give requirement on the capacitor or the resistor? why not it cannot work ? they have no maximum value right ? and also they have no requirement on timing,
do you think that with microcontroller it can be done ?

 

[Analog Ground]

555 timer period is 1.1*R*C. 180 seconds is going to need something like 1 megaohm and 180 uF.

 

[FvM]

but they did not give requirement on the capacitor or the resistor? why not it cannot work ? they have no maximum value right ? and also they have no requirement on timing.

That's not the right way to read a datasheet. They do specify RC ranges with respective timing in a diagram. Unless otherwise said, you should use the chip within this range, or need to conclude from other specifications that a wider range is possible. There are two restrictions to consider:
- peak current/dissipated energy when discharging the capacitor. You should consider that capacitors above a certain value might damage the chip.
- Rt/Ct pin leage currents. There's a specification that raises doubts if 10 Mohm will work.

do you think that with microcontroller it can be done ?

Perfectly, e.g. a tiny 6-pin PIC.

 

[yassin.kraouch]

i found this application note from TI, but i did not understand if the output lengh can reach 180s **broken link removed** ,
normally is it related to timer right ?

 

[yassin.kraouch]

That's not the right way to read a datasheet. They do specify RC ranges with respective timing in a diagram. Unless otherwise said, you should use the chip within this range, or need to conclude from other specifications that a wider range is possible. There are two restrictions to consider:
- peak current/dissipated energy when discharging the capacitor. You should consider that capacitors above a certain value might damage the chip.
- Rt/Ct pin leage currents. There's a specification that raises doubts if 10 Mohm will work.


Perfectly, e.g. a tiny 6-pin PIC.

Hi Fvm, i found a solution to avoid discharge capacitor problem, see page 3 of this chip, https://www.fairchildsemi.com/ds/74/74VHC123A.pdf simply i need to add a protection diode.

 

[FvM]

The good thing is that the Fairchild VHC123 datasheet allows for "infinite" capacitor values. That can be read so, that the circuit is protected against high currents through the internal transistors charging or discharging the capacitor. I'm not sure if this also the case with LVC1G123, hopefully it is.

The protection diode that avoids uncontrolled capacitor discharge with supply voltage shorts is important however. You surely noticed the suggestion of maximum timing resistor of 1 Mohm, which is already quite high related to maximum Rt/Ct pin leakage of 2.5 uA at elevated temperatures. You'll end up with large tantal or aluminium electrolytic capacitors. That's the point where digital timers are clearly smarter.

 

[yassin.kraouch]

i did not understand well how this diode protect the internal circuit ? it allow current to flow through it to Vdd ?

 

[FvM]

The problem arises if the capacitor is charged and you short the power supply. Then the capacitor would be discharged through ESD diodes which have 20 mA maximum rating.