The ability of an "S" gate to source current depends on the voltage. The 2 mA spec is at a voltage of maybe 0.5V. As the voltage at the gate input rises the source current is reduced. At a voltage of 1.6 volts the source current from the "S" input is reduced to a level where 1.6 V across the 33KΩ resistor is no longer sufficient to allow the capacitor to continue to discharge below 1.6V. This would not depend on the ability of the signal generator to sink current.

 

No, I'm trying to say the the RC is not an integrator.

I understand the desired effect is to produce a delay, but I still call it an integrator.

I agree, that something does not make sense, I just cant put my finger on it. I do agree that the signal generator should be able to sink some amount of current, but I've never actually looked at the specs of one to know how much that should be.

Similar questions here. I'm looking for a schematic for one of my signal generators...

 

This would not depend on the ability of the signal generator to sink current.

But the signal generator should be able to discharge the cap in a few RC time constants.

 

The RC is integrating current -- not voltage. You're correct.

 

Since I rarely use XOR gates, I tend to implement a doubler using AND and INVERT:
RCA ICAN-6267 (1975 version)

 

The ability of an "S" gate to source current depends on the voltage. The 2 mA spec is at a voltage of maybe 0.5V. As the voltage at the gate input rises the source current is reduced. At a voltage of 1.6 volts the source current from the "S" input is reduced to a level where 1.6 V across the 33KΩ resistor is no longer sufficient to allow the capacitor to continue to discharge below 1.6V. This would not depend on the ability of the signal generator to sink current.

I think exactly the same as you.
And all of your confusions comes from the fact that I'm not the person who build and test this circuit. My friend ask my this question and He did not give me all the details. And this is why I do not know all the facts. And because I don't now the answer and nothing reasonable came to my mind I decided to ask this question here. Additional I've never used TTL gates in my hobby carrier. As a hobbyist I was only using HC and 4000 series gates, and this is why I totally forget about this "huge" input current. And I'm sure that the problem lies there.

 

No need for apologies, we all stumble around sometimes trying to understand things we cannot see. Some of what I know came from the school of hard knocks, puzzling about things I could not explain. In the early days of firmware development when our debug tools consisted of bone knives and stone axes I was presented with incomprehensible mysteries on a daily basis.

 

I've never used TTL gates in my hobby carrier. As a hobbyist I was only using HC and 4000 series gates, and this is why I totally forget about this "huge" input current. And I'm sure that the problem lies there.

I started out working on HP3000 computers that were a refrigerator sized box of TTL circuits. When I started electronics as a hobby, that's what I used because I was familiar with it. Back in those days, your 5V supply needed to provide amps. In those HP computers, the main 5V supply could provide 100A. When shorted, it provided enough current to weld. That's why we were told not to wear jewelry on the line. One guy learned the hard way to follow safety rules. He burned his watch band into his wrist.

 

I do agree that the signal generator should be able to sink some amount of current, but I've never actually looked at the specs of one to know how much that should be.

The output of my hobbyist Global Specialties 4001 signal generator is 4 paralleled 7400 NAND gates; so sink 64mA, source 3.2mA. My other signal generators are spec'ed to drive 50 ohm loads, so sink/source at least 100mA.

 

The output of my hobbyist Global Specialties 4001 signal generator is 4 paralleled 7400 NAND gates; so sink 64mA, source 3.2mA. My other signal generators are spec'ed to drive 50 ohm loads, so sink/source at least 100mA.

That's useful information -- thanks.