A SUB 1 W LOAD-PULL QUARTER-WAVE PREMATCHING NETWORK BASED ON A TWO-TIER TRL CALIBRATION

Transistors used for cellular and PCS infrastructure applications are required to amplify signals with a peak-to-average ratio that can exceed 13 dB, resulting in a peak envelope power (PEP) approaching 1 kW. This PEP requirement is a consequence of simultaneous amplification of multiple digitally modulated carriers with a time-varying envelope and requires a load resistance in the neighborhood of 0.3 W. Present load-pull technologies based on mechanical tuners is limited to approximately 1 W at cellular and PCS frequencies, which renders these systems incapable of characterizing transistors under these conditions. Quarter-wave prematching networks have been developed to transform the source- and load-pull domains to lower impedance. A variety of techniques have been used to characterize these quarter-wave networks, including standard vector network analyzer (VNA) error correction. This article presents a further refinement of this characterization technique, which is based on a twotier calibration using 7mm and microstrip thru-reflectline (TRL) calibrations.

RF power amplifiers deployed with first-generation cellular base stations were based on cavity combiners and class C-operated silicon bipolar junction transistors for final-stage devices. Up to 10 independent carriers, each constituting one user typically was combined prior to feeding the antenna. This architecture, coupled with the constant envelope property of FM, virtually eliminated the need for linear transistor operation. However, the linearity requirements placed on transistor performance for second- and third-generation wireless base stations are much more demanding. Wireless service providers require that base stations occupy as little volume as possible and, with the adoption of digital modulation; many carrier signals now have a timevarying envelope. The first requirement implies the elimination of the cavity combiner, thereby requiring simultaneous amplification of several carriers. The second requirement implies that quasilinear class AB amplification be used to maintain the integrity of the modulation envelope.

These changes have drastically changed the way in which high power transistors are characterized. Simultaneous amplification of several carriers, each with a time-varying envelope, results in a peak-toaverage ratio that can exceed 13 dB, leading to a PEP demand approaching 1 kW. At the standard 26 V base station supply voltage, a load resistance in the neighborhood of 0.1 W  is required for generating closed load-pull contours of power, gain, poweradded efficiency and adjacent-channel power rejection.

Present high power load-pull technology is based on either active fundamental re-injection or mechanical tuners1-4. Although in principle an active load-pull system can present an arbitrary load impedance, the architecture of these systems is not amenable to generating the extremely high power necessary to emulate a sub 1 W load at 1 kW PEP. The current state of the art in mechanical tuners is limited in resistance to approximately 1 W, although narrowband systems can go lower5. To overcome the limitation posed by mechanical tuners, many researchers have adopted quarter-wave prematching networks to transform the tuner impedance to lower impedance. With this approach, it is possible to present a sub 1 W resistance necessary for high power transistor characterization.

SOLT Calibration with non Insertable thru Connection

This is a method of network calibration that is used in electronics field. It is often termed as on wafer calibration. The standards used in this calibration are termed as short open load and thru which constitutes the SOLT calibration kit. These standards are used to calculate the calibration algorithm which is important to make error free measurements. This method is carried out by connecting a short circuit, and open circuit and a load in successive manner to one port of the and the measurements are taken to complete the reflection calibration within a vector network analyzer plane.

This is typical calibration method that implies three impedance and one transmission standards to specify the reference plane. The three impedances are matched typically with the three standards a Short, Open, Load, and Thru which constitutes the SOLT calibration setup. While as in the other calibration methods like as TRL calibration and others the thru reflection and line standards are used to measure the calibration.

Both of the calibration methods are used and any of the calibration setup can be used to measure the performance of the network as per the availability of calibration standards and functionality of the network analyzer.

The SOLT calibration method is quite useful in deciding the measurement for network performance that is specified as per the mechanical dimensions. While performing SOLT Calibration process you must find the standard values. Then you should attach the calibration setup with the above standards to the Network Analyzer calibration port, the end of a cable, or inside a test fixture where the measurement is to be made. This is considered as the reference plane or measurement plane.

The network connections like as the insert-able connection like as male to female cable connection and others. These connections don’t require other external connections. The external adapters or devices are used to complete the through connection during SOLT calibration. While performing network calibration you must use all the components that are to be used in calibration or else it may result in measurement error.

Calibration can be done without insertable thru this is useful in handling the non-insertable devices. The simplest method of doing this is by using a set of phase-equal adapters with shorts, opens, and loads of male and female adaptors. Now by connecting one adapter to complete the through connection during calibration and then replacing it to swap by another adapter connected with the Device Under Test gives the calibration measurements.

This method of calibration is quite useful in designing the devices operating on the cascading networks and uses high frequencies but can be tested with insertable thru connection. It is quite useful in RF devices. Apart from this there are various other methods of calibration.