AWR eBooks

RF Electronics Chapter 2: Computer Simulation Page 44 2022, C. J. Kikkert, James Cook University, ISBN 978-0-6486803-9-0. Table 2.4. Touchstone file for setting the project frequencies. #Hz S MA R 50 5e+007 0 0 0 0 0 0 0 0 6e+007 0 0 0 0 0 0 0 0 7e+007 0 0 0 0 0 0 0 0 8e+007 0 0 0 0 0 0 0 0 9e+007 0 0 0 0 0 0 0 0 1e+008 0 0 0 0 0 0 0 0 1.1e+008 0 0 0 0 0 0 0 0 1.2e+008 0 0 0 0 0 0 0 0 1.3e+008 0 0 0 0 0 0 0 0 1.4e+008 0 0 0 0 0 0 0 0 1.5e+008 0 0 0 0 0 0 0 0 1.6e+008 0 0 0 0 0 0 0 0 1.7e+008 0 0 0 0 0 0 0 0 1.8e+008 0 0 0 0 0 0 0 0 1.9e+008 0 0 0 0 0 0 0 0 2e+008 0 0 0 0 0 0 0 0 2.1e+008 0 0 0 0 0 0 0 0 2.2e+008 0 0 0 0 0 0 0 0 2.3e+008 0 0 0 0 0 0 0 0 2.4e+008 0 0 0 0 0 0 0 0 2.5e+008 0 0 0 0 0 0 0 0 The line beginning with # is a header, specifying the parameters to be used. # MHz, denotes that the frequencies are in MHz, Hz denotes the frequencies are in Hz. S denotes that this is an S parameter file, MA indicates that the complex data is in polar form (magnitude, angle), the angle of which is always in units of degrees; Selecting Select Scripts Simulation Match_Proj_Freq_to_Datafile_Freq will set the project frequencies to the ones in the Touchstone file of table 2.4. In many cases automatic measurements are made on the network to be determined so that more than one thousand measurements result. The frequency setting script described above is then the only practical way to match the project frequencies to the measurements. Once the project frequencies have been set to correspond to the measured frequencies, the Touchstone file can be deleted from the project if desired. Figures 2.54 and 2.55 compare the impedance measurements of the calculated Bandpass T circuit of figure 2.51 and those of the model using the starting values in figure 2.52. The curves with the markers, i.e. the top two curves in figure 2.54 and the bottom two curves in figure 2.55, are the starting impedances of the model. The curves without the markers are the impedances of the Bandpass T circuit. There are thus significant differences between the circuit and the model at the start of the optimisation. Optimisation goals are set to reduce the phase and magnitude errors to zero. The circuit simulator is then set to optimise the guessed component values to minimise the error equations shown in figure 2.53. In this application, the simplex optimiser results in a fast optimisation, to produce the results, in figures 2.56 and 2.57. These figures compare the impedance measurements of the calculated Bandpass T circuit of figure 2.51with that of the model of figure 2.52 and show that after optimisation, the impedances for the calculated Bandpass T matching network are exactly the same values, as those of the modelled network. RF Electronics: Design and Simulation 4 4 www.cadence.com/go/awr

• Cover