This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TiNA-TI 9.1

Hi

 

I´m trying Tina, for the first time (I currently use other P-spice versions), and it looks fine and very friendly.

But, checking out some of my old circuits I saw an error.  I found that when I try to simulate the transient response of a simple LC parallel circuit with initial conditions, and no disipative elements (zero series R in inductor and infinite parallel R for the cap ) the oscillations dump to zero 

regards

 

n

  • Nolberto,

            The reason you are not seing an oscillation at w=1/sqrt(LC) is because of the numeric integration method you are using for your transient simulations. The default numeric integrator for PSPICe and TINA is the "GEAR" method which is efficient in most cases but produces erroneous results in theoretically lossless or underdamped oscillatons like the one you provided in your example. To change the numeric integration method just go to the Analysis tab from the pull down menu and select "options". In the Options window go to the "transient" group and change the integration method to trapezoidal and rerun the simulation. Let me know if this helps.

    Best regards,

    -Marcos 

  • Why is Gear the default then?  Just because it's faster?

  • Jonathan,

    Actually, TRAP is faster. Gear is considered less accurate by some because it tends to dampen oscillations (whether they are real or not) which is why lossless circuits and oscillators are not handled well with it. Most experienced SPICE users will find out which kind of circuits work well on one versus the other. Gear works better on SMPS devices and signal chain applications tend to use Trap. PSpice defaults to Gear (I believe they may have removed Trap at one time or another as well) and that tends to be the leader that most follow. Many simulators provide a modified Trap version as well that can be more robust and more accurate, however, convergence is still a chore for SMPS devices.

    There are many different opinions on this topic and I have used both types as needed. Please remember that a simulator is just a linear equation solver trying to solve some very non-linear relationships. It is best to understand how the simulator you are using works and why so you can decide how to best use the tool to gain the understanding you need from the simulation.