BJT astable multivibrator generate triangle wave

Question

In this circuit there is a bjt astable multivibrator. It generates a square wave. And according to the description, at the output of the opamp it generates a triangle wave from this square wave.

I've tried to implement this circuit in LTSpice.

When I simulate output of the first circuit (astable multivibrator) looks fine.

But if you look at the output of the opamp there are some glitches rather than triangle wave.

What might be wrong with this circuit here.

Edit :

I changed the value of R6 resistor to 1000000 ohms but still output of the opamp doesn't look like a triangle wave.

Edit2 : After playing around with potentiometer values,I should be able to see a waveform which looks like triangle at the output of opamp.

But if I increase the simulation time from 100 milli seconds to 700 milli seconds I see DC 15 volts at the output of op amp. I don't know why this happens.

Answer 1

So the output of vibrator is a pulse which should be the input of the integrator as to gain the triangular waveform. In your simulation R6 =1 milliohm instead of 1 Mega.

You are basically shorting the integrating capacitor.

The correct symbol for megaohm = Meg in LT spice.

EDIT: As Bimpelrekkie mentioned, since you have a open-loop system your op- amp will drift.

The simulation below shows how you can charge and discharge the capacitor based on the output.

How it works ?

So at t=0 , assumme output is zero and the output of the flip flop is zero. The input at summing amplifier has only the negative dc voltage, this causes a linear rise due to the integrating capacitor.

Once the (arbitrary) threshold of 7V is reached the Comparator U2 goes high and sets the Flip flop which has an output voltage of 2V. As a result, the voltage at the output of the summing amplifier jumps from -1V to (-1V + 2V) 1V, this causes the voltage to fall linearly. Once the voltage goes below 0.05 V, U3 goes high which resets the flip flop and the summing point jumps back to -1 and the cycle repeats.

Do not mind the simulation is not very neat , but gives a idea how you could approach the problem.

Answer 2

I see DC 15 volts at the output of op amp. I don't know why this happens

I do ;-)

This circuit has an essential flaw.

The integrator relies on the fact that the capacitor is charged and discharged with exactly the same amount of charge every time. It is an integrator so keeps adding / subtracting charge into C3.

That means that the square wave generator needs to generate a perfect 50% duty cycle square wave, which it cannot. That's why the potmeter U1 is there, to dial out any offsets.

However, that will not prevent any charge/discharge imbalance from adding up so the sawtooth voltage will eventually "drift" up or down. In your simulation the output drifts down as the charge/discharge imbalance accumulates.

This behavior is a consequence of the simplicity of this circuit. It does not monitor its output voltage to compensate for the drift.

To solve this use a circuit which decides when to start charging/discharging based on the value of the output voltage. The 555 timer chip works like that, it has a window comparator to keep the voltage across the timing capacitor between 2/3 Vdd and 1/3 Vdd. Of course the 555 isn't a proper sawtooth generator, it is just the principle I refer to. Needless to say, a "better" sawtooth generator is going to be a bit more complex.

Another alternative is to generate the sawtooth digitally using a DDS like the AD9833 on Ebay you can buy modules with this chip.