Ads error while importing s2p and simulating

[aashishsharma]

HEY I was importing S2P file into ADS and running s parameter simulation but it gives error -error detected by hpeesofsim during netlist flattening.while evaluating expression 'SNP1.CMP1.all':
the independent variable values are not monotonic for freq at value 2.2e+09
heres link to S2p file given by manufacturer https://www.maximintegrated.com/design/tools/applications/product-software/rf/MAX2644.S2P
Anyone with any ideas whats wrong
Thanks in advance

 

[RealAEL]

That file actually has 3 set of s-parameter/noise data at three different RBIAS settings each with the same frequency range. To use these within ADS you would need to edit the data into 3 different file, Tables 1 & 4 to file 1, Tables 2 & 5 to file 2 and Tables 3 & 6 to file 3. Then use the single s2p file that you require.

 

[aashishsharma]

That file actually has 3 set of s-parameter/noise data at three different RBIAS settings each with the same frequency range. To use these within ADS you would need to edit the data into 3 different file, Tables 1 & 4 to file 1, Tables 2 & 5 to file 2 and Tables 3 & 6 to file 3. Then use the single s2p file that you require.

thanks that solved the problem

 

[aashishsharma]

hey although i was able to plot s(2,1) vs frequency but i actually want to simulate that LNA(max2644) to see how the output would look like if I provide input with 2.4ghz+ some khz freq. That 2 port generic model doesnt provide me with terminals for bias voltage etc given in datasheet.
so how can I simulate that particular IC. also What kind of simulation should i run. I just want to see how my output voltage and current will vary.(circuit given in datasheet above)
Also note that manufacturer havent provided any model for that IC otherwise I would have done it in LTSPICE
any ideas????
thnkns in advance

 

[RealAEL]

If you want see voltage/currents or to do any bias dependant simulation then you would need to be doing something like a Harmonic Balance or Transient simulation. S-Parameter simulation will not enable that sort of analysis.

The next issue is that the model of the active device would need to be a nonlinear/Spice like model. Again S-Parameter models are insuffient for this circumstance. This then leads to the circumstance that in general there are no such models for complete amplifiers. The only thing that would approach this would be an X-Parameter model but these are not supported by many manufacturers yet and I don't think it would really address this issue.

The only practical way to do the simulation you want is to use a complete transistor level model for the amplifier schematic. This would only be available from the manufacturer but I doubt that they would publish this data for end users.

 

[aashishsharma]

If you want see voltage/currents or to do any bias dependant simulation then you would need to be doing something like a Harmonic Balance or Transient simulation. S-Parameter simulation will not enable that sort of analysis.

The next issue is that the model of the active device would need to be a nonlinear/Spice like model. Again S-Parameter models are insuffient for this circumstance. This then leads to the circumstance that in general there are no such models for complete amplifiers. The only thing that would approach this would be an X-Parameter model but these are not supported by many manufacturers yet and I don't think it would really address this issue.

The only practical way to do the simulation you want is to use a complete transistor level model for the amplifier schematic. This would only be available from the manufacturer but I doubt that they would publish this data for end users.

1.what if i use s2p file and use harmonic balance 1 tone simulation and sweep the input frequency???
will it provide me the output voltage???
what about the current.???
2. Also bias information is provided in the datasheet itself so can i skip that part???. I was interested in current drawn by the device so that i can calcultate the the power of my power source for overall design

 

[RealAEL]

You can use s-parameter data to model passive devices, like filters, in a circuit but for active devices they are of little use.

As you said there is no power pin to bias the device so you cannot feed it any DC voltage/current.

In operation the net current at any node during Harmonic Balance simulation is the sum of the DC operating point and then the signal at the fundamental and each of the harmonic tones. The S-Parameter datafile only contains data for 2.2GHz through 2.7GHz in 50MHz steps. There is no 0Hz data to be able to evaluate the DC operating point correctly. The simulator would need to extrapolate and guess the DC performance. Simulation at any other frequency would also need to be extrapolated/interpolated from the available values. It would be a waste of time for so many reasons. You would see the voltage variations from the fundamental but the s-parameter data would not generate any harminic tones and the DC operating point would be totally incorrect.

 

[Ow@i$]

only a x-parameter file or device model from the vendor can do the job that you require!!!