Pspice question....

Hi everyone...

I am trying to model and simulate an astable multivibrator using Pspice but i am having some difficulty....

Both transistors are coming on at the same time so I never get the oscillations I expect. What do I need to do to make the model act more like a "real life" astable mulivibrator?

(I have read it is possible for both transistors to come on at the same time in real life as well if the voltage is turned up slowly enough, but this is obviously not what I want if I'm trying to make a multivibrator.)

Is there some PSPICE trick I am missing?

 

The trick is you need to set an initial condition (.ic) on one node. You can pick a node like one of the bases, and set the initial voltage to something like -1V. I'm not sure how you do that in Pspice, but I know it can be done.
The other way is to make the circuit slightly asymmetrical, such as making one of the caps 1% larger than the other.

 

Interesting idea....

I tried changing one of the caps by 1% but it didnt work....I'll try a few more options...

 

I think I got it....I changed the initial charge on one of the capacitors to .1volts and the circuit appears to be oscillating.

Now I just have to work on getting a shorter charge up time which means changing the cap or resistor values....

But at least it works.

 

Indeed....oscillators of any kind...either "linear" or multivibrators are very difficult to model in spice...because of differen startup and run conditions.

It's usually better to model the AC, the DC and the TRANSIENT models separately.

eric

 

http://physics.msuiit.edu.ph/spvm/papers/2002/emoricha.pdf

I think MATLAB is a little more suited to modeling feedback systems.

eric

 

Accurately modelling in Matlab from the raw-equations might be a little more tricky than first seems, but it is certainly doable, for example with a 555: http://www.ansijournals.com/jas/2005/1012-1015.pdf

Modelling it systematically using Simulink might be an easier (and more productive) approach.

Dave