Re: Spice simulation pictures

To: tesla-at-pupman-dot-com
Subject: Re: Spice simulation pictures
From: huffman <huffman-at-fnal.gov>
Date: Mon, 14 Oct 1996 08:48:11 -0500
    [The following text is in the "ISO-8859-1" character set]
    [Your display is set for the "US-ASCII" character set]
    [Some characters may be displayed incorrectly]

Hi Phil,
It would be interesting to model a coil using a geometry based software. I
have access to ANSYS which will do all kinds of fancy field analysis from a
physical input, but it is very hard to input the geometry.
I use AutoCAD and can't imagine making a 'real' copper spiral of wire, i.e.
1000+ turns. It would be sure to bring my system to a halt.
Dave
----------
> From: Tesla List <tesla-at-poodle.pupman-dot-com>
> To: Tesla-list-subscribers-at-poodle.pupman-dot-com
> Subject: Re: Spice simulation pictures
> Date: Saturday, October 12, 1996 3:25 PM
> 
> >From pgantt-at-ix-dot-netcom-dot-comSat Oct 12 13:06:04 1996
> Date: Fri, 11 Oct 1996 09:15:40 -0700
> From: pgantt-at-ix-dot-netcom-dot-com
> To: tesla-at-poodle.pupman-dot-com
> Subject: Re: Spice simulation pictures
> 
> On 10/10/96 23:24:33 you wrote:
> >
> >Hi GL,
> >I have never tried to closely model a TC. I suspect the distributed
> >capacitance (self capacitance)
> >is probably overshadowed by the addition of a lumped capacitor at the
end.
> >I wanted to try a 1/4
> >wavelength xmission line, but couldn't get it to simulate. (step size to
> >small)
> >A good question is; how much C becomes to much? I have heard of some
very
> >large toroids on some fairly
> >small coils. I doubt if spice can answer this question, I will however
play
> >with it to see what happens.
> >
> >Dave (flaming stones as necessary)
> >
> 
> Having used various forms of Spice over the years, I tend to think that 
> HSpice is the only tool possibly capable of performing accurate
simulations 
> on a coil.  Unlike other varieties of Spice, you can input actual
geometries 
> into HSpice, and providing you have appropriate models, you may be able
to 
> get some reasonable results.  I also tend to think that this task would
be 
> formidable to say the least.
> 
> A simulator that performs analysis on fields may be better suited to the 
> task (like Quad Design's TLC).  However, to do an accurate simulation
with 
> TLC, you need to do a layout of the geometry using CAD tools, and 
> backannotate the geometrical design to the simulator.
> 
> Phil