[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Tesla simulation software project



Tesla List wrote:
 
> Original Poster: paul-at-abelian.demon.co.uk

> Certainly would. I'm trying to reproduce the base input impedance
> using a non-uniform transmission line modelled as a cascade of around
> 10^3 discrete sections from which the AC equations lead to a set
> of simultaneous equations in the currents (just as your teslasim
> does for the coupled LCR model). The various distributed reactance
> profiles are provided by justifiable physical models.

Isn't 1000 sections a bit too much for a lumped model? A practical
coil has about 10 resonances easily observable. The others are of
too low Q to be significant. This can be modeled with no more than
20 reactive elements.
Something that works very well in discretized models of continuous
devices is to allow the user to vary the number of discrete sections
in the model. After a certain number of sections nothing changes
significantly, and there is no need for more sections.

> The result is not very complex, and not too slow. Right now it gives
> the wrong answers but does show some promise.
> 
> (If there are any coiler/physicist/programmer types out there who
> want to collaborate on a professional standard, peer reviewed
> project, the aim of which is to understand in detail the resonant
> mechanisms of tesla secondaries, please do get in touch!)

Maybe I can help.

Antonio Carlos M. de Queiroz