Thank you.
Now I will make a new symbol lm324.sym. I will add this symbol in my library. After that I will be able to use your circuits that use lm324.

You could also likely just substitute one of your LM324 models for the one I used in the schematic.

 

Are you an active member of the forum, and I prefer to just copy the symbol under a different name (I already did). By the way, my model actually reflects the work of the operational amplifier (unlike others). However, the transistor model doubles time the simulation. The problem with conventional models LM324 (LM2902) discussed in the Yahoo Group.

 

The problem with conventional models

I would suggest a differing approach to modeling: models are useful because they are wrong.

So you are far better off at being able to recognize limitations of models and simulations, in order to be more successful at using simulations.

 

Unfortunately for spice models are not usually known limitations. Here is an example. I developed a 10A current generator for the laser. To increase the efficiency I was doing it on a DC / DC chip. According to calculations, it was good, but the reality of my scheme did not give more than 5A. Reason is the spice model did not account for some of the properties of the chip. As a result, I had to quickly alter the schema and board source current.

 

Don't know your design at all so cannot help in pinging the issue on simulation or others.

I think generally speaking simulation cannot take you all the way to a working circuit, especially for analog designs.

Digital degins are generally more tolerant, due in part for model simplicity.

For analog designs, building the circuit to test simulation validity is a must.

 

I've not parsed the entire circuits in my Steve Spice, but the basic idea is how I'd do it - variable frequency sawtooth generator followed by a comparator with a variable reference.