Reliability of Multisim Simulations: Can We Trust the Results?

[Crazymechanic]

Hi folks I would like to know maybe someone has some insight if you make a circuit in multisim (I have the 12 edition) and u go through everything then you run the simulation and it shows that all is ok and working like you thought it should what are the chances that multisim could be lying? I know it sound stupid but are there any errors or bugs in that program so that when I would build that circuit on real board it would not work?

Thanks.

 

[phyzguy]

Usually if you have a problem where the circuit works in the simulation but not in real life, the problem is not with the accuracy of the simulator, but with something you left out of the simulation. For example, in your simulation you usually assume all circuit elements are connected directly to the power supply, but in real life the elements are connected to the power supply through wires which have resistance and inductance. Or perhaps you assume that all elements are at the same temperature when in real life power dissipation in the elements makes them have slightly different temperatures. A very common problem is in statistical variation of the elements. I don't know if you are talking a circuit made with discrete elements or an integrated circuit, but in both cases the elements vary. How robust is your circuit if the elements don't have the values (width, length, resistance, capacitance, etc.) that you assumed, but instead vary by say 20%? What if one is 20% too large and another is 20% too small? Try thinking along these lines and see if you can make your circuit robust enough to tolerate variations.

 

[Crazymechanic]

I actually played pretty much with these things. especially with resistor values i made tolerances like some 30% and it still works atleast the main thing what it has to do it does.
also I tried different voltages in and it worked .
I'm talking about a smps few mosfets driven by astable multivibrator.
The only thing I can't understand is that ok i feed the circuit from a dc power source in multisim everything's ok all measurments real.
Then I put an ac source through a bridge rectifier and some inductors before and nothings happening no matter what diodes i use i get only some 99 volts on the dc side even with a normal size electrolytic capacitor.
I put ground too and all is fine.I can't seem to understand why is multisim making this error?

 

[the_emi_guy]

...Then I put an ac source ...

"AC source" (VAC) in Spice is for small signal AC analysis (small signal transfer functions). You want to use a sinusoidal source (VSIN) in a transient analysis.

 

[Carl Pugh]

AC into rectifiers and a capacitor will give the peak value of the AC.
AC into rectifiers and a inductor will give the average value of the AC.

 

[Crazymechanic]

One more thing I can't understand is that when I have made a circuit that's working great now I save and close that multisim file and make another one where I experiment with some parts.then when done experimenting I make everything back as it was in my other tested version and in this one the circuit doesn't work.
How come if in the previous simulation the same circuit the same parts everything worked great and in a new one nothing works?