Two design flows are provided in the ADW:
You can do all the active work including defining the specifications for the matching problems to be solved in Microwave Office (TM) or the Advanced Design System (TM) and only design the matching networks (with or without harmonic control) with the Ampsa Wizards. A complete flow is provided for this in the ADW from synthesis, to fine tuning and introducing modifications in the matching network for biasing purposes, reducing discontinuity effects or replacing basic components with alternatives (inductor to spiral inductor or solenoidal coil, etc.) to re-optimizing the matching network to restore its performance after the modifications made. The final matching network can be exported in various formats, including DXF, Sonnet Software (R) format and also as a Microwave Office (TM) script.
The parameters of the discontinuity models used in the ADW can be optimized to allow for easy adjustment of the network to provide the EM performance required. Only two EM simulations are usually required. The first simulation is used to adjust the parameters of the discontinuity models, after which it is re-optimized in the ADW and the second is used to verify that the performance is as required. Note that it may be necessary to reduce large steps in line width by introducing extra intermediate steps (tapering).
When linear cascade-type amplifiers are designed most of the work can be done in the ADW with linear models and boundary-line constraints on the I/V-plane, after which the designed amplifier can be exported for fine-tuning with non-linear models and further processing. The ADW design process usually starts with fitting linear ADW models to the class A or B S-parameters of the transistors to be used at the required dc operating points (the operating point corresponding to the rated output power) and specifying the associated I/V-plane information. A systematic approach is then followed to design each stage of the amplifier, after which the completed design is optimized. The amplifier is then exported for fine-tuning and further processing.