Hi Guys,
This is a frequency counter front end (not mine) where the NAND in the chip is not being used,
just the Schmidt trigger part used to clean a signal for input into some logic.

I am also interested in using one this way to square a triangle wave.
My question is about the use of the 0.1uF decoupling. Is it used so the transistor in this circuit can still operate?
If you were connecting logic to this it would not be used right? Maybe just the sensitivity pot which I do understand is a divider.

 

I would say the capacitor is there to protect the logic gate against switching noise present in the rest of the circuit. While it is drawn the way it is, physically it would be connected close to the NAND gate's power and ground. Read this post to review the use of a decoupling capacitor.

 

It’s not the decoupling capacitor, that would be extra connected between the chip’s supply and gnd.
It’s an AC coupling in this circuit.... not decoupling actually.

 

My mistake in reading the schematic. I thought you were talking about C3; I didn't notice C2 was also 0.1μF.

 

P1,C4 is a differential circuit, normally we don't use it that way, it must be disconnect from ground.

 

I didn’t see C3
I don’t need the pot adjustment , the signal is already hitting upper & lower threshold.
I suppose I can try both ways with direct input or not.

 

C3 is there from Vcc to ground that it is used for filtering the Vcc noise, the C1 is missing, I think it just a babel problem.

The C4 and P1(10K) just pull up to Vcc, it is provide a short trigger signal, if you don't want that short then you may get a continuing signal related to the input signal.

 

Ok thanks, that’s fine. I understand why you might want to condition the frequency counter input that way,
but don’t think it’s needed for me right now.

 

If you want a square wave become to a triangle wave, then you can't use 74HC132, becasue the output of 74HC132 still a square wave, it just change the duty cycle.

 

I have a triangle wave I want to make a square wave. But I don’t care about duty cycle.

 

You just need a schmitt trigger ic as mc14584, 74HC14, 74HC132 , no need other components.

 

The purpose of C4 is to AC-couple the signal from the transistor to the Schmitt input. The pot P1 enables the DC level of that input to be biased near the trigger threshold, so that only a small AC-coupled voltage is needed to trigger the Schmitt. The pot is thus a sensitivity control.

 

Transistor T1 is a simple common emitter BJT amplifier. The input signal applied to R1 is clamped to between 0.6V and -0.6V by diodes D1 and D2. R1 prevents destroying the diodes if the input voltage is too high.

The signal is AC coupled to the base of T1 via C2. R2 provides a small amount of current to bias the base-emitter junction of T1.

The output of the BJT amplifier is also AC coupled via C4 to the inputs of the 74HC132 used as a Schmitt trigger to provide hysteresis and a square wave output. Since the signal is AC coupled, a DC bias point must be established using the SENSITIVITY ADJ voltage divider. In most circumstances, this would be adjusted to the midpoint of the pot to provide a bias point that is midway between Vcc and GND.

If your input is a triangular wave this would not be an ideal circuit to use. Since the signal is AC coupled, both C2 and C4 are differentiator circuits which respond to the rate of change of voltage. A square wave has sharper edges as compared to a triangular wave and hence has a higher dV/dt signal.

A better choice for triangular wave would be a comparator circuit with an adjustable reference level.
The 74HC132 alone would act as a comparator with hysteresis with a switching threshold at the midpoint between Vcc and GND.
If you are attempting to modify this circuit for a triangular wave input, simply disconnect C4 and apply the input to pins 1 and 2 of U1A.

Depending on the source impedance of your input signal you may want to remove the SENSITIVITY ADJ connection since this would present a 5kΩ load on your signal. If your signal is already biased between Vcc and GND then the SENSITIVITY ADJ is not needed.

 

Thanks for the info, yes I’m thinking the same initially that I don’t need the AC coupling.
The signal is fed either by a flip flip or a hex buffer and both 74HC.
It’s when I’ve used the flip flop to divide a frequency that the wave becomes a bit of a triangle,
but both thresholds are reached, so I figure the Schmidt trigger will keep all output uniform.
It won’t have unknown input applied like a frequency counter front end might,
and the output is always to clock micro controller pins or other 74HC logic.

Why the wave is triangle in the first place I’m not sure.. possibly a problem,
or maybe it’s input is fast enough to see rise & fall slopes for the device,
but the signal is consistent, so I don’t really mind.

 

If you attach the circuit and draw the waveform on it, maybe we can figure out where the problem is?

 

Thanks for the info, yes I’m thinking the same initially that I don’t need the AC coupling.
The signal is fed either by a flip flip or a hex buffer and both 74HC.
It’s when I’ve used the flip flop to divide a frequency that the wave becomes a bit of a triangle,
but both thresholds are reached, so I figure the Schmidt trigger will keep all output uniform.
It won’t have unknown input applied like a frequency counter front end might,
and the output is always to clock micro controller pins or other 74HC logic.

Why the wave is triangle in the first place I’m not sure.. possibly a problem,
or maybe it’s input is fast enough to see rise & fall slopes for the device,
but the signal is consistent, so I don’t really mind.

What is the frequency of the output signal from the flip-flop?
What is the bandwidth of your oscilloscope?
Are you using a x10 attenuation probe?

 

Either 8 or 12 MHz derived from 16 or 24 MHz, 100 MHz scope and using 10x probe.

I can certainly see the Schmidt trigger. All frequency inputs are uniform,
but this picture is rubbish, I forgot to set in the scope I was using a 10x probe,
and the probe might not be well tuned for this scope.
I can’t otherwise explain the drop below zero where the circuit doesn’t have any supply below zero.

So I think I can call it a success, but will look at the signal properly tomorrow... 2:45am here in Aus right now!

 

The undershoot is caused by ringing in an improperly terminated transmission line. It may have been the result of improper grounding of the scope probe. You can ignore this.

Your leading and trailing edges look good.

 

Thanks, yeah I figured that... I don’t think it could possibly come from the circuit.
I think I have what I want and just haven’t seen it properly yet