On Thursday 19 March 2009, Thomas Ibbotson wrote: > The voltage controlled switch represents the electrical > breakdown above a voltage of 5KV. This circuit works quite > well in reproducing what we see experimentally. However, > there are artifacts caused by the fact that the switch > changes immediately from high to low resistance.
What version of gnucap are you using? If you are using the "stable" version, it is probably a good idea to get the latest development snapshot. There are some significant improvements in time step control. When you get these artifacts, and they are not solvable any other way, often you can get rid of them by changing the differentiation method. ".option method=euler" usually cleans them up good. If you have the latest snapshot, ".option method=gear" also works. It might help to look at the tolerances, particularly vntol, abstol, chgtol, and trtol. Also, look at parameters related to limiting, particularly vmax and vmin. The default values are chosen with low voltage semiconductor circuits in mind. It is likely that you will need different values for high voltage circuits like this. > So, I would like to know if it's possible to have a voltage > controlled switch as above but which has a time-dependent > model for its switching? There isn't one already written, but you could make one. _______________________________________________ Help-gnucap mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-gnucap
