0
\$\begingroup\$

I would like to simulate a circuit which uses LM324 operational amplifier.

The complete schematics of the circuit is like that.

It is taken from an article written by George Steber in the 2004 issue of Circuit Cellar magazine.

U1A and U1B generates triangle wave. U2A and U2B form a power buffer for the triangle wave.

Edn magazine

I've understood most of the circuit's working principle. But I couldn't figure out this part.

LM324 step generator

As you can see there is a CD4066 CMOS switch and a voltage controlled switch.

I am not sure whether this switch is between, CD4066's VDD and VSS pins.

This is the same circuit in MicroCap simulator software.

LM324 step generator MicroCap

When I look at the transient analysis result, I see some kind of noise instead of step wave.

LM324 transient analysis

May you provide a working example which doesn't use switches?

Any ideas would be apreciated. Thanks.

Update :

According to a similiar article step generator provides a slow voltage when S1 switch closes.

\$\endgroup\$
2
  • \$\begingroup\$ Let's say you have this book that has 10 chapters and you understand what 9 chapters are about then, you decide to send the chapter you don't understand to me so that I can explain it to you. But I only see that one chapter and can't figure out what is happening at all. Can you see the problem here? You have all ten chapters and I have just (say) chapter 7 and you expect me to explain that to you? How does that work? \$\endgroup\$
    – Andy aka
    Dec 11, 2020 at 11:40
  • \$\begingroup\$ Ok. I am adding overall circuit schematics. \$\endgroup\$
    – user211748
    Dec 11, 2020 at 11:44

2 Answers 2

Reset to default
0
\$\begingroup\$

SW1 is the on-off switch for both the triangle generator and the bottom circuit. It is debounced with a 5ms time constant and then fed into the inputs of the 4066 switches. Unit C of the 4066 switches the op-amp output for either 0V or 6V. It is smoothed with a 24us time constant, a lot faster than your x-axis.

The y-axis of your measurement is not specified. What is the range of the spike, which may occur due to the LM324 being a fairly slow op-amp? It is likely rather small. Using faster op-amps will bring down its duration and height even more.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks for the explanations. In fact I didn't specified y axis. There is an auto setting for x and y axises. Testing above circuit without any correction this time I got a different response. It turns out that y axis is between 687m and 687.25m. \$\endgroup\$
    – user211748
    Dec 11, 2020 at 13:37
  • \$\begingroup\$ Thanks for checking. So the "spike" is roughly .1mV -- rather small. \$\endgroup\$
    – Carsten B.
    Dec 13, 2020 at 11:16
0
\$\begingroup\$

After carefully reading the second article I should be able to get the desired response.

Although there are some minor differences between two circuits their operating logic is similiar.

I've used only one switch and LM358 operational amplifier in this circuit.

LM358 step generator

This is the transient analysis result.

LM358 step transient analysis

\$\endgroup\$
2
  • \$\begingroup\$ The real change you made is in the cap value from 470p to .01u -- 20x larger! This makes the input signal to the op amp much slower in variation, so that the amp now reacts without overshoot. I'd suspect the analog switch was never the problem. \$\endgroup\$
    – Carsten B.
    Dec 13, 2020 at 11:19
  • \$\begingroup\$ Thanks for the feedback. If I have time I may test the original circuit with 470p capacitor. But I suspect the real problem was my understanding of step generator circuit. At first glance, I didn't realize that step generator was providing a step voltage when S1 closes. After reading the article thoroughly I used a time controlled switch to simulate momentary button behavior. Anyway, this article is quite interesting. It mentions about discrete-time digital filter, bilinear transform which are quite new and interesting subjects for me. \$\endgroup\$
    – user211748
    Dec 13, 2020 at 11:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.