# Function generator with an operational amplifier

I am solving a function generator whose frequency of each output is 5kHz. My question is, how can I find the parameters of C3, R5 and R6 to get an output of the same frequency?

In addition to the above I have two more questions:

1. How can I design another op-amp-based circuit to convert the sinusoidal signal to a PWM signal and whose frequency is also 5kHz?

2. How can I vary the amplitude of the waves, in a range of 0V and 10V by using a potentiometer?

Pd: To calculate the desired output of the first op-amp square wave generator I used the astable multivibrator equation. For the second and third OP-Amp circuits, I used the integrator equations. My frequency output is 5000Hz.

• Not fully tested but this does what you want. For unipolar, replace Gnd with V+/2 tinyurl.com/yau2boop May 13, 2022 at 5:51

"My question is, how can I find the parameters of C3, R5 and R6 to get an output of the same frequency?"

Sorry, you cannot. The frequency of oscillation are set by R, R1, R2 and C1.

Or, rather, your output frequency will ALWAYS be 5 kHz, regardless of C3,R5 and R6.

Another note - your "triangle wave" out of U2 will not be a very good one. You need a different approach.

As analogkid answered, you can't use a 5 kHz PWM signal to represent a 5 kHz signal.

"How can I vary the amplitude of the waves, in a range of 0V and 10V by using a potentiometer?"

With a single potentiometer, you can't. Well, you can, but I doubt you'll find it practical.

What you would do is put each output through a multiplier, such as the venerable AD534. You would have to scale each output to 1 volt at the input to the multiplier(s). You would then take a single pot, connect one end to ground and the other end to 10 volts, and the wiper to all three multipliers.

Some things:

a. Without some form of feedback from the triangle output to the square wave generator, the integrator will eventually saturate either nigh or low. this is because the output stage of a 741 is not symmetrical. The high output is not as close to the positive rail as the low output is to the negative rail.

The common way around this is to make the square wave section a simple comparator with hysteresis, and drive it with the output of the integrator. Now the squarewave output levels are determined by the comparator hysteresis, not the device characteristics.

b. Each waveform output is at an opamp output, which is a fairly low impedance. Adding a 10K pot to GND at each output should get you independently adjustable levels for all three waveforms.

c. You can't. You cannot modulate a 5 kHz carrier with a 5 kHz signal.

What PWM frequency do you need. It will have to be many times the 5 kHz sinewave signal frequency, which means that the 741 might not have enough bandwidth. The method is to use another two opamps to produce a stable triangle wave at the PWM freq, and feed this to one input of a comparator. The other comparator input is the sinewave. There are many schematics for this on the innergoogle.