I have not worked on my SPICE modelling program for a few weeks now.
I have been spending time on my Big Radio Project.

Now that is completed I can return to my SPICE modelling program.
When last I posted I mentioned that I have run across the models of Adrian Immler and that I wanted to evaluate them and get them into my program. They look like very good models, assuming you can extact parameters for them that will fit your data.

I have not been able to do that very well programmatically. His models do not reduce into simple explicit equations of independant variables. Which is fine for SPICE, a SPICE engine is meant to handle that.

But it is not fine for the non linear regression I want to do to extract model parameters. I have tried every trick and simplification I know, and I can get things to work not too badly for small amounts of data, but I am really not satisfied with it. I have put in the ability to plot the results of the model against the measured data so the user can interactively change the parameter values and see the results. This may be the only way to get a good fit to the data for the Immler models. But it doesnt allow the automated parameter extraction and model comparison that the program is intended to do.

The other thing to do of course is use the models provided by Adrian Immler.

So why not just release the program as it is now, with the understanding that the Immler models are not automated ? Because there are other things I need to clean up. As it is now it is just something I use, I need to clean up the menus, and remove things that dont work, or features I started to add but didnt finish etc.

I have set a target of Sept 1 to finish this program.
Oh, I forgot to say what yeare. OK, Sept 1 2020.