Example of Gain Compression Simulation (ADS only)

Gain-compression simulation example illustrates the setup for a basic gain compression simulation of a power amplifier circuit. The design includes a harmonic balance controller as a comparison.

Note This design, XDB1.dsn , is in the directory examples / Tutorial/SimModels_prj . The simulation results are in XDB1.dds .

Gain-compression simulation example

To perform a gain-compression simulation:

1. From the Sources-Freq Domain palette, select a port-type source such as a _P_1Tone_ or _P_nTone_ . Place the source at the input of the component or circuit under test, then define input power and edit other parameters as required.
2. From the Simulation-S_Param palette, select a Term component and place it at the output of the component or circuit under test. Edit the impedance if necessary.
3. Ensure that the input port source is identified as Num=1, and the output port termination as Num=2. Also ensure that you have labeled the inputs and outputs of nodes at which you want data to be reported.
4. From the Simulation-XDB palette list, select an XDB simulation component, and place it on the schematic, then edit it.
5. Select the Freq tab, and edit frequencies, orders, and fundamentals as required.

   Hint In a one-tone simulation such as this one, MaxOrder is irrelevant.

   • Ensure that values are established for all of the frequencies of interest (for example, RF, LO, and IF frequencies) in the design under test. (Where mixers are employed, set the LO frequency to Freq[1]; it is easier to achieve convergence if the frequency of the signal with the largest amplitude is assigned to Freq[1].) To add or delete frequencies, select them in the Select field, then click Add or Cut . Display them on the schematic to facilitate editing.
6. Select the X dB tab and set the following:
   • In the Gain compression field, either accept the default or enter a new gain-compression value. The default is 1 dB.
   • Under Port numbers , ensure that appropriate values are entered for the input port source (generally Num=1) and the output termination (generally Num=2).
   • Under Port frequencies , set appropriate values for the input and output frequencies. In a circuit employing a mixer, the input frequency would be that of the RF input, the output frequency that of the IF output.
   • Under Power tolerances , set appropriate values for acceptable variations in power at the input and output ports. In the Max. Input Power field, set the maximum input power over which the simulator is allowed to search.
7. To reduce memory requirements significantly where many frequencies and orders are involved, select the Solver tab, then select Krylov Solver .

   Note We do not recommend using the Krylov option on small problems. For details, refer to the topic "Selecting a Solver" in the chapter "Harmonic Balance Basics" in the Harmonic Balance Simulation documentation.

8. From the Simulation-HB palette list, select an HB simulation component, and place it on the schematic. On the Freq tab, set the values for the Fundamental Frequencies . On the Sweep tab, enter Pin for Parameter to sweep . Then, check Use sweep plan and enter Plan1.
9. Simulate . Gain compression data items may be identified with an HB prefix. The following is a Rectangular plot of the results of the simulation. At the top are the gain compression results, with the standard harmonic balance results at the bottom for comparison.
   
   The following graphs and their associated equations illustrate how to extract more meaningful results from the simulation. The equation Gain finds gain by subtracting the input power Pin from the output voltage of the fundamental [Vout[1]). The equation linear adds input power to gain to produce the linear plot in the graph. To find the input power that leads to a compression point of approximately 1 dB, place a marker on each trace and move them so that their difference is as close to 1 dB as the simulation results allow. The simple equation compression subtracts the values of the markers and the compression List plot shows the equation's results to be a gain compression of approximately 1 dB. The next List plot shows the input power and output power at which the approximately 1 dB compression point occurs.