The Question: If you use the 74LS123 with an external resistance of 10 k but do not use an external capacitor, what do you get for the pulse width?

The Solution Attempt: So first I took a look at the data sheet (http://www.futurlec.com/74LS/74LS123.shtml) which presents some confusing information. Operating Rule 1 says an external resistor and external capacitor are required for proper operation. But then the same rule goes on to state that the external capacitor's capacitance may range from 0 to any necessary value. That sounds to me then like an external capacitor isn't required. So I look for a formula or chart for the pulse width but all that's provided is a formula for when (external capacitance) >> 1000 pF and a graph for when 10 pF < (external capacitance) < 1000 pF, so I'm not sure how to find the pulse width for (external capacitance) = 0. How might I proceed?

 

Did you see the page 2?
And the important thing is this:
when 10 pF < (external capacitance) < 1000 pF

Sometimes when the capacitance of capacitor is too small, it won't generate the oscillation or astable, so choosing the values of capacitor from 10pf that just make sure the IC can be working fine.

 

In any practical circuit there will be some parasitic capacitance from wires, traces and the internal geometry of the chips. 10pf is such an incredibly tiny value, it is hard to imagine that it can be constructed with any degree of precision. What exactly is the tolerance and temperature coefficient of a typical 10 pf capacitor? So any given construction may or may not work and may or may not be repeatable.

 

Thank you for the responses. It sounds like you guys are suggesting what I suspected which is that it won't function properly as a monostable multivibrator without a capacitor. Is this correct? And if so, what WILL happen? Is its behavior just undefined? Will the output just be stuck HIGH or stuck LOW?

 

You might well think about the response as having a distribution with a mean and a variance. A small variance from the mean implies a more predictable response, while a large variance implies a broader, less predictable, response. What you want is enough capacitance so that the strays and parasitics are small in relation to the chosen value.

 

Simple solution. Why don't you just try it out?
With 10kΩ resistor, a pulse width of 100ns is typical.