Research and D esign of an X -B and UHF P ower A mplifier

Introduction. A method for designing power amplifiers for use in the transmitting channels of X-band transceiver modules is investigated. The design process was aimed at optimizing the relationship between the basic amplifier characteristics, including the operating frequency band, output power level, output linearity, high harmonics suppression, etc. Aim . Investigation of a method for designing an X-band UHF power amplifier, which is capable of optimizing the relationship between its main characteristics. Materials and methods . Theoretical calculations were combined with experimental studies into the design of a UHF power amplifier. The stages of the design process are described in detail, including major ideas, principal circuits, and strip circuits. Evaluations were conducted using the Keysight ADS high frequency circuit simulation tool. Results. A method for designing X-band UHF power amplifiers on the basis of a close combination of theory, simulation, and experimental adjustment was described in detail. Conclusion. A prototype of an X-band PA was developed; an approach to developing a methodology for manufacturing, measuring, and testing X-band PAs is described.

present an approach to designing an X-band UHF PA that optimizes the ratio between its important parameters. Then we describe the design process and investigate the developed system. Finally, some conclusions are drawn.
The process of designing an X-band UHF PA. Input specifications. The input specifications for PA design are presented in Tab. 1.
The design process includes the following steps: selection of an amplifier element; circuit selection and setting the static operating point; estimation of load/source impedances; impedance matching circuit design; voltage divider circuit design, etc. The basic expressions for theoretical calculations can be found in the design materials published as recommendations. Care should be taken to ensure optimal matching of the main characteristics of the amplifier based on not only theoretical statements, but also experimental adjustments.
Selection of the amplifier element. 1. Based on the input specifications, we preliminarily selected an MMIC transistor, which is a TGF2977-SM transistor based on a high electronic mobility technology (HEMT) manufactured by the Quorvo company as a UHF amplifier element with a broadband. The main parameters of TGF2977-SM are presented in Tab. 2 [17, 18].  Fig. 1 shows the actual representation (a) and the QFN-type functional pin assignment (b) of TGF2977-SM.
2. According to [18], the model for the TGF2977-SM element, referred to as HMT-QOR-TGF2977-SM-01, is a large-signal non-linear model whose parameters are extracted from a large-signal model. This data is mapped against scatter matrix measurements [ ] S and load data of a large Load and Pull signal in the range 6 11 GHz.  Fig. 2, a shows the HMT-QOR-TGF2977-SM-01 reference plane. Fig. 2, b demonstrates the element pins (after packaging) for data output (by Keysight ADS highfrequency circuit simulation tool).
3. Validation of the large signal model is of particular importance in UHF PA design, since this model allows the characteristics of the amplifier to be accurately determined. This process is carried out using load and source pull measurements [19]. The validation process is performed under the condition: 9 GHz; Circuit selection and setting of the static operating point. 1. In order to simplify the design and to ease measurement and adjustment procedures when scaling the production process, we selected the circuit diagram of a single amplifier based on the TGF2977-SM transistor with a common source (CS) to provide the gain ratio and output power.
2. The selection and setting of the operating point for the UHF PA circuit is usually carried out based on the characteristics of components provided by the manufacturer. However, for the majority of new generation components, manufactures determine the optimal mode of their operation, rather than statistical characteristics. Thus, experimental measurements can be carried out based on the analysis of calculation and simulation results. Following the recommendations of the manufacturer, the setup procedure for TGF2977-SM in the AB mode is as follows: set up GS = 4 V; V set up the current limit of DS~3 0 mA; I power supply DS = 32 V; V make adjustments slowly to DS 25 mA; I → set up the pulse current limit of PDS~5 00 mA. I

Estimation of load/source impedances.
An estimation of optimal source and load impedance values is performed separately by the source/load pull technique.
1. The load impedance is implemented by the load-pull technique in the ADS as shown in Fig. 4, a. The PA is designed considering the requirement of high performance. Therefore, it is necessary to estimate impedance at the 0 f frequency and at adjacent frequencies in the operating frequency range of the system. At 2nd and higher harmonics, impedance estimation is aimed at reducing the power loss across them. However, according to [2], when the PA element is set to the AB mod, the power loss at the 2nd harmonic is much larger than that of the other higher harmonics. Therefore, the determination of the load impedance at harmonics higher than the 2nd order can be omitted. However, in order to ensure a uniform quality during mass production and its adjustment, we added a pseudo short circuit for the 3rd harmonic in the output impedance matching circuit. In Fig. 4, b, lines show the level of output power and PAE for the diagram presented in Fig. 4, a. It can be seen from Fig. 4, b that the optimal load impedances at the center and neighborhood frequencies for the The source impedance was determined for the maximum value of PAE. In terms of the model, the source impedance is a complex conjugate of the input impedance of the PA element, estimated such that the reflected power loss at its input is minimized. The source impedance was optimized using the source pull technique as shown in Fig. 5, a. Fig. 5, b presents the contour lines of the output power and PAE. Since the output power level is determined as the target parameter, the optimal source impedance is selected at the maximum power of ( ) 3. The determination of impedance values for high harmonics is carried out both at the source and load sides using the load-source pull technique. Fig. 6 presents the results of estimating the source and load impedance at the 2nd and 3rd harmonics. Since these values fall into the boundary region of the Smith diagram, then, according to [13,14], the phase difference between the instantaneous value of the current and the voltage will vary within the limits 90 .
±° As a result, the power loss at higher harmonics will be suppressed.
To further confirm that optimal source/load impedances match, the input/output reverse loss of the PA requires verification. Fig. 7 presents the results of determining the input (red) and output (green) characteristics of the reverse loss of an ideal PA.

Design of an impedance matching circuit.
Impedance matching circuits are used to convert impedances into optimal source and load impedances in order to ensure minimal losses.
1. Fig. 8 shows the strip circuit (a) and the layout circuit (b) of the input impedance matching circuit.
At the strip circuit level (Fig. 8, a), the ideal paths will be converted to strip paths on the Rogers high frequency material RO4003C, for which the layout of the input impedance matching circuit is shown in Fig.  8, b. According to [2,14,15], to evaluate the quality of the input impedance matching circuit, the insertion loss characteristic of the following form can be used 2. Fig. 9 shows the strip circuit (a) and the layout circuit (b) of the output impedance matching circuit. The output impedance matching circuit is designed similarly to the input impedance matching circuit, which is responsible for transmitting a signal from a source with an impedance of 0 ,~( 9.6 4.1) L f Z j − Ω (which is the optimal load impedance of the input impedance matching circuit of the PA element) to a load with an impedance of 50 Ω at the frequency 0 9.4 GHz, f f = = while reducing the power loss at the 2nd and 3rd harmonics. Fig. 9, a shows that there will be an open-circuit transmission line from the PA element side with the length 4 λ at frequencies  Fig. 10 shows that the insertion loss characteristics of the output impedance matching circuit, depending on the frequency, demonstrate acceptably small values. At the 0 f frequency, the insertion loss IL has a value 0.42 dB, thus basically satisfying the design requirements.

Design of a voltage divider circuit.
1. Similarly to any radio electronic circuit, the voltage divider circuit of an PA performs the function of supplying DC power to establish the operating point of the active element and suppresses the RF signal to avoid loss and bad interference to a DC source.
2. According to [12], the voltage distribution for the circuits working in the X band, which was applied in this study, will use a conical opencircuit transmission and the 4 λ path segments as a short circuit capacitor (instead of pass capacitor) at the 0 f frequency as shown in Fig. 11. The results of measuring and testing the quality of the voltage divider circuit through the parameters Results of PA design and comments. 1. Fig. 12 presents a schematic of layout of the X band PA.
2. Fig. 13, a,  3. In order to evaluate the capability of the designed PA to ensure linearity, we performed a direct test measure of the signal form at its output    A comparison of the obtained PA characteristics with those presented in Table 1 shows that the designed X-band PA unit meets all the input requirements.
Conclusions. 1. The results of a study into the design of UHF power amplifiers working in the X band are presented. The authors' intention was not to propose a new design approach, but rather to clarify a method that combines theoretical calculations, experimental measurements, and experimental adjustments. This approach allowed the authors to harmonize and ensure optimization of some relationships between the basic characteristics of the amplifier, including frequency bandwidth, output power, output power variation, high harmonic suppression, and output signal linearity. The described implementation has practical significance in terms of simplicity and convergence between theory and experiment.
2. The results obtained can be used to develop a technological process and procedure for manufacturing X-band UHF PAs even under unfavorable technological and economic conditions.
3. The developed X-band UHF amplifier may find application in multifunctional electronic radio systems, such as communication, radio navigation, etc.

Author's contribution
Xuan Luong Nguyen, synthesis and analysis of approaches to solving engineering design problems and PA technology; theoretical analysis and selection of PA elements meeting the set technical requirements; drawing structural and functional diagrams; selection of options for a schematic diagram corresponding to the amplification modes.
Thanh Thuy Dang Thi, scientific advisor and scientific support, including simulation models for PA and PA elements by common ultra-high frequency design tools; analysis of the obtained results; supporting implementation using the Keysight ADS high frequency circuit simulation tool.
Van Bac Nguyen, scientific, technical, and technological support, including the implementation of PA design and technology using high frequency material RO4003C; participation in the process of measuring and correcting technical characteristics; supporting measurement and adjustment of technical specifications.
Phung Bao Nguyen, scientific advisor, consulting and controlling the entire process, including simulation, design, and manufacturing; measuring, calibrating and processing the data on the PA technical characteristics.