It is unfortunately very easy to set up a DPS transmitter design project that is immediately doomed to failure. The reason is that if the implementation technology selection is not carefully matched to the project performance objectives, then, when it comes time to connect the new DPS with the RF PA it will be used with and to do full testing, the chances of it all working together are actually quite small.

The way to succeed is no different from any other complicated project: plan ahead and do the required component and subsystem testing so that a substantial well-understood foundation is established for the system operation. For DPST operation, the testing is much more complicated than for conventional designs because the variable power supply adds a new degree of freedom. Familiar analysis methods based on curves must change to operate on surfaces, for example. Add to this a completely new interface that is fundamentally not well-behaved and the project gets even more interesting. Careful characterization of this new interface is vitally important.

This chapter provides significant detail on the testing required in DPS transmitter development in order to have near certainty of project success. This test pro fi le is the result of 20 years' experience this author has with making DPST designs work. Each of these tests has its important place on the path to project success. Many of the unusual tests in these lists are the result of unpleasant surprises in the test lab.

Characterization planning

The three major blocks of characterization testing required for any successful DPS transmitter design are: (1) the DPS itself, (2) the RF power amplifier on its own, and (3) the transistor used within the RF PA. These blocks are illustrated in Figure 13-1. At this writing, the manufacturers of dynamic power supplies and RF power amplifiers are different companies. This necessitates their interconnection to be considered as a new interface.