1-Tone Nonlinear Simulations > Harmonic Gamma Opt. - PAE, Output Power, Gain
Description
This setup determines the optimal source and load impedances to present to a device. It is very similar to the Harmonic Impedance Opt. setup previously described, except that allowed source and load reflection coefficients are defined as circular regions of the Smith chart, instead of defining ranges of impedances. It optimizes the source and load fundamental and harmonic reflection coefficients (up to the 5th) simultaneously, to maximize power-added efficiency, and deliver a specified power to the load. It differs from the load- and source-pull simulations in that it varies both source and load reflection coefficients simultaneously, at both fundamental and harmonic frequencies. A sample device is provided. You must replace this device with your own device, and modify the biases, as needed.
Needed to Use Schematic
A device using a nonlinear model
Main Schematic Settings
Input frequency and range of allowed values for the available source power, desired power delivered to the load and minimum power-added efficiency. Also, the range of allowed source and load reflection coefficients must be specified, as circular regions of the Smith chart, at the fundamental and harmonic frequencies.
Data Display Outputs
HarmGammaOpt1tone.dds, "Power, Gain, Spectrum" page:
For the best impedance values found during the optimization:
- Power-added efficiency
- Power delivered to the load in dBm and Watts
- Power available from the source and power (at the fundamental frequency) delivered to the device
- Operating power gain (power delivered to the load / power delivered to the device)
- Transducer power gain (power delivered to the load / power available from the source)
- Thermal dissipation in the device
- DC power consumption
- Total input power (DC power consumption + power delivered to the device at fundamental and all harmonic frequencies)
- Total output power (power delivered to the load at fundamental and all harmonic frequencies)
- Output spectrum (dBm) and harmonic distortion in dBc.
HarmGammaOpt1tone.dds "Opt Source and Load Z's" page:
- Smith chart showing the optimal source impedances at fundamental and harmonic frequencies
- Smith chart showing the optimal load impedances at fundamental and harmonic frequencies
- Smith charts showing the source and load impedances renormalized to an arbitrary impedance
- Listings of optimal source and load impedances and reflection coefficients
IHarmGammaOpt1tone.dds "Waveforms" page:
- Input and output voltages versus time
- Input and output currents versus time
- Input current versus input voltage and output current versus output voltage
Schematic Name
HarmGammaOpt1tone
Data Display Name
HarmGammaOpt1tone.dds
Notes
- You can delete one of the two supplies and/or replace the voltage sources with current sources, and the PAE calculation will still be valid. You can modify the components in the bias network, realizing that the DC power consumption is computed as (the DC voltage at the Vs_high node) * (the DC current in the Is_high current probe) + (the DC voltage at the Vs_low node) * (the DC current in the Is_low current probe).
- For some load and source impedances, the device might be unstable. For this reason, you might want to simulate the stability circles of the device at a particular bias point, to check for instabilities. To do this, copy the biased device into the schematic generated from the menu selection DesignGuide > Amplifier > S-Parameter Simulations > S-Params., Noise Fig., Gain, Stability, Circles, and Group Delay. The stability circles are on one of the data display pages that will be updated after you run a simulation using this schematic. Avoid using source and load impedances within the unstable regions if the source and load stability circles are inside the Smith chart. You might also want to use some of the other DesignGuide schematics to test for stability with a large input signal.
- If you don't think that reflection coefficients at the fourth and fifth harmonics (for example) are going to have much effect on the performance of the device, you can fix these values by changing the word opt in their equation definitions to noopt. These equations can be seen by editing the VAR block, Load_Gamma_Parameters , and modifying the equation for angle_L_4th, for example. If you set this equal to 0 rather than 0 opt(-pi to pi) , then the angle of the reflection coefficient will be fixed at 0 radians. The variable sample_radius_L_4th as well as the variables for the 5th harmonic and for the load can be modified in the same way. This will speed up the optimization.
- The speed and success of the optimization will depend on the parameters that you set on the Nominal Optimization controller. Refer to the ADS Optimization and Statistical Design manual for more details.
Privacy
Statement
|
Terms of Use
|
Legal |
Contact Us
|
© Agilent 2000-2008 ![]()