having the cake and eating it too - passive notch filter design

[phasetrans]

It should be mentioned that LTspice is not a control plane for the original Berkeley SPICE, it is a heavily optimized bespoke tool. And a tool under active development and internal use. The latter is probably why it has so many Easter eggs.

Source 1: the primary LTspice author himself (https://theamphour.com/196-an-interview-with-mike-engelhardt-spice-simulator-synteresis/)

Source 2: a friend who works for a company that uses LTspice internally, controlled by software, as part of their full board-level simulation pipeline process.

[Rob Strand]

It should be mentioned that LTspice is not a control plane for the original Berkeley SPICE, it is a heavily optimized bespoke tool. And a tool under active development and internal use. The latter is probably why it has so many Easter eggs.

With commercial code a lot of time and people goes into the user interface, especially in the light of customer feedback.   When your a one man show you can only spend so much time doing this.  The first priority is for the simulation to be accurate and reliable.  I haven't hand many problems with LTspice sims, other than normal numerical issues.    When I run a simulation I'm fairly confident of the results.

Source 1: the primary LTspice author himself (https://theamphour.com/196-an-interview-with-mike-engelhardt-spice-simulator-synteresis/)

Berkeley SPICE (the original) had lots of problems and discontinuities for the models contained within.

To me these things matter.   Quite a few times I simulated transistor circuits on the older simulators and got non-physical negative collector voltages.    These were exactly the cases where hand calculations are all too hard.   I've spend time myself preventing discontinuities and non-physical results on tube models.   You can throw a lot of time into it.   Also you have to decide what the model will do in cases which never appear in the datasheets or literature.

[phasetrans]

...I've spend time myself preventing discontinuities and non-physical results on tube models.   You can throw a lot of time into it.   Also you have to decide what the model will do in cases which never appear in the datasheets or literature.

Do you have a triode and pentode that you are willing to share? I gave up on LTspice for tube circuits because of problems with the models I could find.

[Rob Strand]

Do you have a triode and pentode that you are willing to share? I gave up on LTspice for tube circuits because of problems with the models I could find.

I don't have a problem giving it out but it's on one of my old (broken) PCs which I've shelved for a rainy day.

The models I did were on microsim spice.  I haven't converted them to LTspice but that's only the symbols.     Off-hand I took the excem tubemodels and tweaked them.   Most of the time was spent bullet-proofing the Pentodes.    On the whole the Pentodes cause the most problems.   I did get a power amp simulation working which matched the oscilloscope waveforms under clipping.

The excem models were available on-line at one point (maybe called lampes, in French) but i can find the link ATM.

www.excem.fr/download/usergui5.pdf

Another set of models was the Koren models.   IIIRC these are hard to tweak by hand,

http://www.normankoren.com/Audio/Tubemodspice_article.html