S-parameters model for surface mount resistor in microwave

[paradiseseeker8]

How to create an S-parameters model for a surface mount resistor in microwave? let's say for 0805 or 0603 with value 120 ohms and frequency range 2GHz to 12GHz.

Thanks.

 

[volker@muehlhaus]

You can find S-parameter files on the manufacturer download pages.
You can also purchase the Modelithics CLR library.
Or you can measure if you understand the de-embedding process.
Or you can EM simulate as a resistive sheet (assuming that there are no meanders or cuts from laser trimming).

 

[D.A.(Tony)Stewart]

Ask supplier for s- parameters then add same for traces.

Use lumped circuit analysis
Choose values for trace Ls, Cp to each end point then lookup parameters for vendor model and size for L,C,R.

It will be pretty bad at 12GHz for 805 and many 603, unless specifically designed for RF. (flip chip or wraparound.) Orientation also affects values and some designs use different methods to improve performance near SRF.

Consider Vishay Dale Thin Film
Model :FC in 603, 402 ..even 201 and compare.
https://www.vishay.com/docs/60093/fcseries.pdf

 

[paradiseseeker8]

thanks for replying, volker
but the number of s-parameter files provided are very few
and they are modeled for the regular resistor values such as 50 ohms, 100 ohms and 1 kohms but specific value such as 120 not available.
Modelithics CLR library as I noticed it is for the regular values .

I'm still newby this topic, so excuse me how to EM simulate it?

 

[FvM]

To determine S parameters of a chip resistor, it must be placed in a micro strip with known parameters (substrate height, trace width, Er), reference planes have to defined. And it can be shunt or series connected.

 

[paradiseseeker8]

thanks, sunnyskyguy.
Vishay provides S-parameter models for FC series and CH series but both for regular values how can use these files to fit for a specific resistor value.
Choosing Ls and Cp values based on what?

 

[volker@muehlhaus]

but the number of s-parameter files provided are very few and they are modeled for the regular resistor values such as 50 ohms, 100 ohms and 1 kohms but specific value such as 120 not available.

It depends on the library, other vendors have more complete libraries. What I do if I need a value that is not in the library: pick the nearest value from the library, and then add the difference in resistance as an ideal resistor connected in series.

Modelithics CLR library as I noticed it is for the regular values .

I don't understand what you mean? Modelithics has scalable models, so they are very useful and complete. The only disadvantage is price, but the accurate models are worth it, in my opinion.

I'm still newby this topic, so excuse me how to EM simulate it?

If you are not familiar with EM, it doesn't make sense to try that. Getting accurate data from EM needs some time to learn and practice. So in your case use a library.

 

[criterion456]

How to create an S-parameters model for a surface mount resistor in microwave? let's say for 0805 or 0603 with value 120 ohms and frequency range 2GHz to 12GHz.

Thanks.

Now that you have no ready-made s-parameter model ,
just create one




Ls, Cp : Refer to your data sheet
R : just its value


Or as volker@muehlhaus said,
by means of measurement if you understand the de-embedding process.......

 

[volker@muehlhaus]

View attachment 121480

Ls, Cp : Refer to your data sheet

The problem with your model is that all shunt capacitance is missing, so this is not accurate enough. The model topology in the Vishay data sheet is more accurate ... but unfortunately they have not provided any values.

 

[criterion456]

The problem with your model is that all shunt capacitance is missing, so this is not accurate enough. The model topology in the Vishay data sheet is more accurate ... but unfortunately they have not provided any values.

Hmm…….but……..my model……is just based on Vishay.

\(￣口￣)/

Cg is external capacitance to GND,

Thus, would you please explain more ?

all shunt capacitance is missing ?

I am confused………..

TKS~!

 

[D.A.(Tony)Stewart]

The problem with your model is that all shunt capacitance is missing, so this is not accurate enough. The model topology in the Vishay data sheet is more accurate ... but unfortunately they have not provided any values.

With the series I suggest LC values are given for each case size. (smaller is better)
R values are shown in graphs. This does not include external parasitic elements.

 

[criterion456]

With the series I suggest LC values are given for each case size. (smaller is better)
R values are shown in graphs. This does not include external parasitic elements.

for size, the smaller the better

Is the aim to posses higher SRF ?

TKS~!

 

[volker@muehlhaus]

Hmm…….but……..my model……is just based on Vishay.
all shunt capacitance is missing ?

I was referring to criterion456. His model is too simple to be useful. The Vishay model is ok.

- - - Updated - - -

With the series I suggest LC values are given for each case size.

You are right, I missed the L,C specs in the plots.
For the shunt elements, I would assume parallel plate capacitor math, with Al2O3 dielectric for the resistor body and then PCB substrate below with the actual values for permittivity and thickness.

 

[paradiseseeker8]

I contacted Vishay asking about S-parameter model for 120 ohms and 180 ohms in CH series,
they suggested using the electrical model provided in datasheet



and they send me values for capacitors and inductances to use with electrical model