slow time domain simulation

I'm new to CircuitLab but so far it seems really great. I do have a question...

I have an analog circuit with two sine wave voltage sources and a sine wave current source. It's not very complex - just a few zener diodes, resistors and capacitors. When all sources are set with the same phase angle, the simulation runs in a few seconds. When I modify the phase angle of the current source, the same simulation takes 20 minutes. Any ideas what could be going on? I've tried adjusting time steps, which obviously helps, but it is still much slower.

Thanks, Alan

by alehman
September 28, 2013

Can you post the circuit so we can have a look?

:)

by signality
October 01, 2013

Here it is. Thanks for your time.

by alehman
October 09, 2013

I have observed that adding any phase angle to the current source greatly increases the processing time and the results appear to become invalid. It should be a DC current (with small AC ripple) that increases to a value proportional to the cosine of the phase angle - similar to the 0 phase angle case.

by alehman
October 09, 2013

Hi alehman,

The problem is that I1 needs a DC path to ground on both sides. It is grounded one one side but has no DC path to ground on the other.

Connecting a 1G resistor in parallel with I1 fixes the problem and gets you back to sims taking only a few seconds.

If you have further problems, try the same trick with 1G resistors in parallel with each of D3, D4 and D5.

:)

by signality
October 10, 2013

The 1G in parallel with the current source fixed the problem. Thanks! Alan

by alehman
October 12, 2013

Post a Reply

Please sign in or create an account to comment.