Cascade NF with MWO using s-parameter files

[enrico]

cascade nf

Hi All,

Everytime when I try to plot the noise figure from a cascade using standard s-parameters files involving amplifiers and filters in between I get a warning message in the microwave office (or similar like Genesys) saying that for example the filters does not contains noise parameters, so therefore the device will be treated as noiseless.

Since the filters are passive devices, so the noise figure it is their own insertion loss I cannot understand how to contour this problem.

Suppose I have a cascade of two amplfiers with their full and respective s-parameters files containing noise data and a filter in between. The filter also is modeled using its respective s-parameter file but does not contains data.
Since the the noise figure from the filter is commanded by its own insertion loss, How this problem could be solved ? (in case I want to plot the noise figure of the whole chain including the filter).

It must be a way how to tell it to the simulator, in e.g. microwave office since the S21 (or S12) plot represents its own noise figure.

Any ideas are very welcome !

 

[DDavid]

s-parameter files

Hi,

If I understand you correctly you have Sparameters of the each amp.
But the question is the the same s parameters include noise parameters,
what i mean is that u need not only S11, S22, S21, S12 also
K, gamma min etc.

I attached sparameters example with noise, take a look!!!

David

 

[vfone]

cascade microwave amplifier site:edaboard.com

A System Simulator needs only the insertion loss to calculate the NF, but a linear Circuit Simulator needs parameters as: Rn (equivalent noise resistance) and Fmin (minimum noise figure, which in your case is the insertion loss)

 

[enrico]

cascade s-parameters

Hy Guys,

Many thanks for the replies.

Yes that is correct. The s-parameters from the amplifiers also contains noise parameters like you mentioned, but the filter does not.

In this case, the data that you get with any filter vendor does only contains the full s-parameters (S11, S12, S21 and S22) but does not contains noise data.

If not, so how can it possible to plot the total noise figure of those linear devices together ? So from what I have read it is not possible because we need additional data like noise for the filter(s) most likely due to mismatches between the elements.

Still, as I can see from your previous comments the Fmin could be derived from the filter insertion loss, that is ok and I can add this into the s-par matrix. What about the Rn (Equivalent Noise Resistance) ?


Any inputs are welcome from your side. Anyway many thanks for your previous comments.

 

[zeshan102]

insertion loss s21 noise figure

Noise figure of filter is its insertion loss so total noise figure can be calculated by adding the insertion loss of filter and amplifier noise figure.

 

[DDavid]

s parameter cascade

Lets' say u have link like he following:

AMP1 - Filter - AMP2

Put all each of then separate schematic and check NF (for the AMP you need noise parameters like we spoke before, regarding to the filter you do not need nothing except the IL because like you remember IL=G=NF in passive elements)

Then after you will see NF of each AMP cascade all three of them and ...

David