Here is a better version if you are allowed to use diodes:

 

breadboarding

Hmm, being this is a fairly straight forward circuit, you may want to double check your wiring. What type of op-amp are you using?

 

Hmm, being this is a fairly straight forward circuit, you may want to double check your wiring. What type of op-amp are you using?

I've tried LM741 and LM358P

 

Hi, still does not work.

Try increasing the capacitor to say 10uF. The LM741 (if that is what you are using) has a terrible slew rate. It also tends not to work well for a 'comparator' type circuit like this.

 

Ok.
I got much better performance when I slowed it down by changing the capacitor from 0.1uF to 10uF. It appears the op-amp cannot keep up, so slowing it down helps a lot. Also the second circuit I showed will work better than the first. But even in the second circuit use a 10uF. It does not work well unless the overall frequency is fairly low.

 

Hi I just built the one with the diodes and it seems to work, but the output should for example always be -14V (LOW) to 14V (HIGH). with this the circuit you sent, the average voltage is always zero, when the PWM is 100% for example its now at 0V and should be at 14V. Do you maybe have a solution for that?

Thanks in advance.

 

Hi I just built the one with the diodes and it seems to work, but the output should for example always be -14V (LOW) to 14V (HIGH). with this the circuit you sent, the average voltage is always zero, when the PWM is 100% for example its now at 0V and should be at 14V. Do you maybe have a solution for that?

Thanks in advance.

Do you have R6 and C2 going to the negative supply rail? The first one I sent out did not do that, but it is important. Did you change the capacitor to 10uF?

 

Hi I just built the one with the diodes and it seems to work, but the output should for example always be -14V (LOW) to 14V (HIGH). with this the circuit you sent, the average voltage is always zero, when the PWM is 100% for example its now at 0V and should be at 14V. Do you maybe have a solution for that?

Thanks in advance.

Are you using a 10uF capacitor?

 

Hi I just built the one with the diodes and it seems to work, but the output should for example always be -14V (LOW) to 14V (HIGH). with this the circuit you sent, the average voltage is always zero, when the PWM is 100% for example its now at 0V and should be at 14V. Do you maybe have a solution for that?

Thanks in advance.

Are you sure the negative power input to the op amp is going to the negative 15V supply? A common mistake is have the negative power rail go to ground. I am just asking because I cannot see the breadboard circuit.

 

Okay I found the problem, I'm using a super cheap DSO138 oscilloscope and the button to switch from AC to DC coupling was stuck on AC. Now it shows the correct signal. Thanks a lot!

 

Could you maybe provide a small explanation why this works because I don't get the whole circuit, I also tried using a diode to put the signal on the schmit trigger but that only worked for small changes in duty cycle.

thanks in advance

 

ok, sorry about that, here is one possible solution replace R1 and R2 with a potentiometer wiper going to positive input. Replace R3 with a rheostat. The frequency and duty cycle can be set but it is not a linear relationship.

View attachment 250920

I also want to understand this so could you draw it with potentiometers so I can maybe see the mistake I made.

Thanks in advance

 

R1 and R2 are just one pot

But the lower end of the pot should go to the negative rail, not to ground, if you want the full duty cycle range.

 

I edited my schematic a bit and now I have full range of the duty cycle circuit but the frequency changes with it, what is the cause and solution to that?

 

ok, sorry about that, here is one possible solution replace R1 and R2 with a potentiometer wiper going to positive input. Replace R3 with a rheostat. The frequency and duty cycle can be set but it is not a linear relationship.

View attachment 250920

I also got this circuit to work now, but it has the same problem as mine, at high and low % duty cycle, the frequency changes with the duty cycle.

 

I edited my schematic a bit and now I have full range of the duty cycle circuit but the frequency changes with it, what is the cause and solution to that?View attachment 250953

it seems that the behaviour here is the same with or without the diode.

 

But the lower end of the pot should go to the negative rail, not to ground, if you want the full duty cycle range.

yes, I've made those changes.

 

That will always be a problem because the time period varies with the reciprocal of resistance. When you have a potentiometer to set the duty cycle the time period will be
\(
t\propto \frac {1}{R}+\frac{1}{R_{pot}-R}
\)
where R is the setting of the potentiometer, and t is not constant as R varies
The only way to achieve independent adjustment of mark-space ratio and frequency is by using a triangle-wave oscillator and a comparator.

 

Try your latest circuit (post #34), but without the diode, the 10k resistor and the 30k resistor. If you use the LM358, rather than the poor old 741 op-amp, connect the unused op-amp inverting and non-inverting inputs (not outputs) to ground to avoid noise pick-up.
Hint: If you get into the habit of giving each schematic component a reference designator (e.g. R1, C2 etc), it makes it easier to discuss circuits.

 

I also want to understand this so could you draw it with potentiometers so I can maybe see the mistake I made.

Thanks in advance

Here is the original circuit drawn with potentiometers. I will try to provide a circuit description for both a little later today.