Laplace expression possible with LTSpice lossy transmission line?

[BobW]

I'm trying to model a lossy transmission line in LTSpice. I want the
simulation to work up to about 15GHz and include skin effect and
frequency-dependant dielectric constant of pcb traces and material.

I know that in LTSpice it's possible to include Laplace expressions in
signal generators. I was hoping that it would work in their lossy
transmission, too, but there's nothing in their help files and nothing
online.

When I use the simple model statement it works well (along with the
appropriate .param statement for R, L, and C:

.model BobLossyTLine3inch LTRA(len=3 R={RperInch} L={LperInch} C={CperInch})

However, when I try to include some type of Laplace expression for
resistance or capacitance, I can't find anything that will work. I suspect
there's no way to do it.

Has anyone had any success with this?

Thanks.

Bob
--
== All google group posts are automatically deleted due to spam ==

[cassiope]

While we don't know much about LTSpice's internals, I suspect that the
fractional exponents necessary
to model lossy transmission lines are beyond its capabilities. i.e.
there are no simple lumped-component
models, which I suspect would be necessary for any spice to use.

I've done comparable things modeling fractional-exponent equivalents
with "enough" RLC sections to
make the approximation work. You might try that.

HTH...

[oopere]

Hi, Bob

If I recall correctly, a transmission line may be modeled as (*):

*---[Z]------ ------------------[Z]----*
+ + |+ +| + +
V1 E1 O F(V2,E1) O F(V1,E2) E2 V2
- - | | - -
*------------ -------------------------*

For an RLCG transmission line, you have s-dependent impedances Z while F
has other s-dependent terms than exp(-s*T).

You may simulate the impedances and the controlled source with the
Laplace functions on almost any spice.

(*) Google up: "method of characteristics" transmission lines

Pere