[Carsten]

Indeed. Didn’t see that 🙂

[Carsten]

There are textbook formulas for that. you could add “parameter sweeps” for W and h and use normal formula blocks to calculate Z0. Then you should be able to plot with cartesian etc. Maybe you have to add a little rf simulation to satisfy Qucs

[Carsten]

Hi DrZrD,

What do you mean by “SCH assignments”?

Could you explain the problem in other words, please?

[Carsten]

A simplified example:

The inductor on the left is interpreted as being parallel to R1

The netlist generated for the left circuit alone seems ok:

R:R1 gnd _net0 “50 Ω” “26.85” “european” “SMD0603”
L:L1 _net0 _net0 “39 nH” “0” “” “inductor_1mH”
.SP:SP1 “lin” “0.01 GHz” “1.2 GHz” “100” “no” “1” “2” “none”
Pac:P4 _net0 gnd “1” “50 Ω” “0 dBm” “1 GHz” “26.85” “con_2”

I don’t like that.

[Carsten]

Please post the .sch file.

[Carsten]

Hello Dieter,

I can image being this person. I designed filters for MRI coils so far.

You can check my linkedIn: https://www.linkedin.com/in/carsten-k%C3%B6gler-584a2387/

You can also contact me there.

Best wishes

Carsten

[Carsten]

usimmics does not support complex source impedances yet. Does the effect disappear if the source impedances are real?

[Carsten]

> I need to assign the power port for wideband impedance matching to a S parameter file.

You want to export the resulting s-parameters to a touchstone file?

[Carsten]

QucsStudio/uSimmics has not been developed for Linux. There are glitches when using it via Wine. Michael usually does not provide support for these issues.

[Carsten]

Hi Mike,

Voltage/current probes only provide a measurement if an AC simulation block is present. To achieve this, you need to deactivate the S parameter simulation block. The original Qucs could have both simulation blocks activated. Perhaps that’s why it’s in the documentation.

Best

Carsten

[Carsten]

Components -> Paintings -> Text

[Carsten]

Can confirm it.

BTW: Models “Two mutual inductors” and “Three mutual inductors” works fine.

[Carsten]

Hi,

In the “Equation defined component in frequency domain”, the variable ‘freq’ is defined. So you can do things like this (equals a 10nH inductor)

But ‘freq’ is not defined in equation blocks.

[Carsten]

According to F1 help:

dBm(x,Z)
convert voltage across impedance Z into (active) power in dBm (parameter Z is optional, default 50)

dBm() should be applied to voltages. S-parameters are dimensionless voltage ratios.

S21 is 1 for such a circuit. An impedance of 50 Ohm and 1 V_peak would dissipate 1/100 W of power. Equals +10dBm

[Carsten]

S-Parameters are linear, so the power level has no influence. I suspect, a power level is not even defined / needed by the solver.

How did you measure the power? The voltage and current probes do not work with a S-Paramter-Block active

[Author]

Posts