74HC14 Circuit Not Working

[RyanG]

The circuit in the attached picture is not working correctly. I get how it is supposed to work, but I don't get a square wave output. When I adjust the pot I just change the voltage. I checked it with a scope.

[rx8pilot]

The input is just a variable voltage? Why are you expecting an adjustable PWM?


EDIT: I think I get it - the cap is going above and below the schmidt threshold. Did you put this on a breadboard? Screen shots?

[RyanG]

I was looking for an easy adjustable PWM and I found this on the Internet, so I was trying to use it. According to the picture it should put out a PWM signal.

[rx8pilot]

What did you probe with your scope? Pin 4? The 74AC14 will not output a variable voltage.

[RyanG]

Yes, pin 4. I have no idea what is going on here to be honest. If it is working, then it's a very small range somewhere around 45%-55%.

[Brumby]

Those outputs should be either GND, Vcc or switching between the two.

Is there anything else connected to pin 4 other than the scope?

[edpalmer42]

What do you see if you probe the top of the capacitor?  What changes if you adjust the pot?

Ed

[RyanG]

The only thing connected to pin 4 is the scope, and I know that the pins are either "on" or "off." But, I don't get why I get such a small range of values. Without the scope connected, I used a LED and the brightness barely changes when I adjust the pot.

[Brumby]

I'm thinking it might be time for some photos.

Can you take a photo of your setup?  It should be close-up, well lit and in sharp focus for us to see exactly what's where.  Show it with the scope probe attached, so we know where everything is.

Also, images of what is on the scope's screen - one for the pot at minimum, middle and maximum positions, if possible.


Two other questions -
 - What are you using for a power source?
and
 - Where is the scope's ground clip placed?

[grumpydoc]

Check your wiring.

Check that all unused inputs to the 74HC14 are grounded or tied to Vcc

Post photos

[RyanG]

I have checked the wiring several times. All of my unused inputs are tied to GND. Power source is 5v 7A. The ground clip of the scope is on the - rail of my breadboard. The difference between the maximum and minimum of the voltage reading on the scope from pot minimum to maximum is ~0.05v, and it is the only thing that I can see change. I can do pictures if absolutely necessary, but I don't have the good breadboard wires at the moment. I just have longer jumper wires, so it is messy looking.

[rx8pilot]

Post photos - messy or not. It is not a beauty contest.

What happens when you probe Vcc? Does your scope show 5v?

[RyanG]

Yes, the scope shows 5v. Picture of breadboard and schematic attached. When I attach the scope it replaces the series resistor and LED. I removed the wires for the other inputs so you can see the rest of the circuit on both the breadboard and the schematic. The diodes are not touching, it is just the angle of the picture. I also tried another 74HC14 just in case the one I was using was bad. Every other component was also tested.

[Brumby]

Well, for starters, I don't see pin 7 connected to 0V.

[grumpydoc]

I have checked the wiring several times. All of my unused inputs are tied to GND. Power source is 5v 7A. The ground clip of the scope is on the - rail of my breadboard. The difference between the maximum and minimum of the voltage reading on the scope from pot minimum to maximum is ~0.05v, and it is the only thing that I can see change. I can do pictures if absolutely necessary, but I don't have the good breadboard wires at the moment. I just have longer jumper wires, so it is messy looking.

OK, make sure your Schmitt trigger is working

Ditch the caps and diodes, connect one end of the 10k pot to Vcc, the other to ground and connect the wiper to pins 1 & 3.

Start with the input voltage at 0V, verify that the output of both gates is close to 5V. Gradually raise the input voltage and check that, at some point, the output flips to 0V (that point should be about 3.5V with a 5V supply). It will be worth noting the input voltage at the trigger point - it won't be the same for each gate but will be close. If you want you can verify that as you reduce the input voltage from the first trigger point you need to reach a lower threshold (about 1V) before the output flips high again.

[grumpydoc]

Well, for starters, I don't see pin 7 connected to 0V.

+1

[rx8pilot]

Put a bypass cap on Vcc, 1uf or so. It should be right at the Vcc pin.

What signal do you see on the inputs?

Sent from my horrible mobile....

[rx8pilot]

Yeah.... pin 7 is rather critical.

Sent from my horrible mobile....

[stj]

the schematic shows a 74AC series, not a HC series - the outputs are different.
besides, using a cap/resister tap circuit often used to generate a reset pulse to create a self-oscillating stage is very poor design to be extremely polite.

use a 555 like everybody else.
if you need pulse-shaping, use a 555 to drive your HC schmit-triggered inverter.

[RyanG]

The pin 7 wire came out when I pulled the other ground wires for the picture by accident. I'm not that good at this but also not that dumb. 

[Zero999]

the schematic shows a 74AC series, not a HC series - the outputs are different.
besides, using a cap/resister tap circuit often used to generate a reset pulse to create a self-oscillating stage is very poor design to be extremely polite.

use a 555 like everybody else.
if you need pulse-shaping, use a 555 to drive your HC schmit-triggered inverter.

It should work for the 74HC series too.

What's wrong with this design? The only problem I can see is poor frequency stability and accuracy, which is a non-issue for a simple PWM circuit.

I have checked the wiring several times. All of my unused inputs are tied to GND. Power source is 5v 7A. The ground clip of the scope is on the - rail of my breadboard. The difference between the maximum and minimum of the voltage reading on the scope from pot minimum to maximum is ~0.05v, and it is the only thing that I can see change. I can do pictures if absolutely necessary, but I don't have the good breadboard wires at the moment. I just have longer jumper wires, so it is messy looking.

OK, make sure your Schmitt trigger is working

Ditch the caps and diodes, connect one end of the 10k pot to Vcc, the other to ground and connect the wiper to pins 1 & 3.

Start with the input voltage at 0V, verify that the output of both gates is close to 5V. Gradually raise the input voltage and check that, at some point, the output flips to 0V (that point should be about 3.5V with a 5V supply). It will be worth noting the input voltage at the trigger point - it won't be the same for each gate but will be close. If you want you can verify that as you reduce the input voltage from the first trigger point you need to reach a lower threshold (about 1V) before the output flips high again.

Yes, this is good advice. Check the Schmitt trigger circuit is working before doing anything else.

[RyanG]

I'm getting about 3.6v and 1.1v for the Schmitt trigger points

[Zero999]

The next steps are to test the diodes and the potentiometer.

[grumpydoc]

I'm getting about 3.6v and 1.1v for the Schmitt trigger points

That looks about right

Just to confirm that the circuit should work here is my breadboarded version



Slight differences from your original circuit, I could only quickly put my hands on a 5k pot so I increased the cap to 220nF. The resistor on the LED is 200ohms rather than 220, finally there's a 220nF decoupling cap on the supply (badly placed, BTW as it should be close to the chip but I just stuck it where it was easy). You can also see the logic analyser leads. None of those changes should make much difference to be honest.

The 74HC14 is a pull I had lying around. Provenance unknown but that looks like a 1993 date code!

And changing the duty cycle works as well although one limitation of this circuit does show up which is that the frequency is not independent of duty cycle varying from approx 1.2 to 1.5kHz

[rdl]

I also put one together and it worked pretty well other than the frequency stability. I used a 10pf capacitor and a CD74HC14E. The square wave looked fine for a breadboard circuit. At midrange on the adjust pot, frequency was about 6 MHz and varied from around 5 to almost 9 MHz as the duty cycle was adjusted. I didn't see the frequency stability get much better when the cap was changed to get down in the single digit KHz range. This kind square wave generator is not something I ever really looked at. Maybe there are better circuits.