Throw away your network analyser and use Spice!
 

On Wed, 11 Aug 2004 18:59:22 -0700, John Larkin
wrote:

RFSIM99 does nice frequency-domain analysies, in all sorts of formats
- Smith, rectangular plots, polars, tabular s-param listings, stuff
like that. It does lumped parts, transmission lines, and some active
stuff. Free, too!

It's in a box here, near the bottom:

http://home.sandiego.edu/~ekim/e194rfs01/

That's certainly one highly useful piece of artillery for the RF
designer!
I don't think it actually allows one to *extract* S- parameters
through modling a BJT, for example.
I haven't had the chance to check it over in detail yet, but it seems
you have to insert already predetermined S-params from datasheets or
wherever and it will generate matching networks from there. So it
doesn't actually amount to a 'virtual VNA' - or does it??

p.
--

"What is now proved was once only imagin'd." - William Blake, 1793.

On Thu, 12 Aug 2004 13:44:11 +0100, Paul Burridge
wrote:

On Wed, 11 Aug 2004 18:59:22 -0700, John Larkin
wrote:

RFSIM99 does nice frequency-domain analysies, in all sorts of formats
- Smith, rectangular plots, polars, tabular s-param listings, stuff
like that. It does lumped parts, transmission lines, and some active
stuff. Free, too!

It's in a box here, near the bottom:

http://home.sandiego.edu/~ekim/e194rfs01/

That's certainly one highly useful piece of artillery for the RF
designer!
I don't think it actually allows one to *extract* S- parameters
through modling a BJT, for example.


If you draw a part as a lumped equivalent schematic (ie, a Spice
model) it will give you the s-params. If you are very lucky, the part
manufacturer will give you that Spice model; few do.

I haven't had the chance to check it over in detail yet, but it seems
you have to insert already predetermined S-params from datasheets or
wherever and it will generate matching networks from there.

I think it will do that, too.

So it
doesn't actually amount to a 'virtual VNA' - or does it??


I don't entirely know what you mean by that. It will give you the
s-params of a given simulated network.

The really hard thing is to have the s-params and go backwards to a
lumped Spice model. There is some software that does this, but I don't
know of any that I can afford.


John

On Thu, 12 Aug 2004 13:44:11 +0100, Paul Burridge wrote:

On Wed, 11 Aug 2004 18:59:22 -0700, John Larkin
wrote:

RFSIM99 does nice frequency-domain analysies, in all sorts of formats
- Smith, rectangular plots, polars, tabular s-param listings, stuff
like that. It does lumped parts, transmission lines, and some active
stuff. Free, too!

It's in a box here, near the bottom:

http://home.sandiego.edu/~ekim/e194rfs01/

That's certainly one highly useful piece of artillery for the RF
designer!
I don't think it actually allows one to *extract* S- parameters
through modling a BJT, for example.
I haven't had the chance to check it over in detail yet, but it seems
you have to insert already predetermined S-params from datasheets or
wherever and it will generate matching networks from there. So it
doesn't actually amount to a 'virtual VNA' - or does it??

p.

Heh, heh, heh. You *can* measure s-params from a circuit in
(P)Spice, but Burridge has me killfiled and will never know it if no
one replies to me - not that he'd listen :)
--
Best Regards,
Mike

On Thu, 12 Aug 2004 17:34:15 -0400, Active8 wrote:

On Thu, 12 Aug 2004 13:44:11 +0100, Paul Burridge wrote:

On Wed, 11 Aug 2004 18:59:22 -0700, John Larkin
wrote:

RFSIM99 does nice frequency-domain analysies, in all sorts of formats
- Smith, rectangular plots, polars, tabular s-param listings, stuff
like that. It does lumped parts, transmission lines, and some active
stuff. Free, too!

It's in a box here, near the bottom:

http://home.sandiego.edu/~ekim/e194rfs01/

That's certainly one highly useful piece of artillery for the RF
designer!
I don't think it actually allows one to *extract* S- parameters
through modling a BJT, for example.
I haven't had the chance to check it over in detail yet, but it seems
you have to insert already predetermined S-params from datasheets or
wherever and it will generate matching networks from there. So it
doesn't actually amount to a 'virtual VNA' - or does it??

p.

snip

The vendor article I read has a better way to do it. Probably doable
in RFSim, too. I just looked at the moron's abse post.
--
Best Regards,
Mike