High-Efficiency Doherty Amplifiers Using Class-F and Inverse Class-F Load Networks with a Quarterwave Transmission Line

Parent Category: 2015 HFE

By Andrei Grebennikov

In modern wireless communication, broadcasting, and industrial systems, it is required that the power amplifier operate with high efficiency and low harmonic output level simultaneously. To increase efficiency of the power amplifier, it is possible to apply a switchmode Class-E, Class-F, inverse Class-F, or mixed Class-E/F technique [1, 2]. 

Highly efficient operation of the power amplifier can generally be obtained by applying bi-harmonic or polyharmonic modes when an additional single- or multi-resonant circuit tuned to the odd or even harmonics of the fundamental frequency is added to the load network. An infinite number of odd-harmonic resonators results in an idealized Class-F mode with a half-sinusoidal current waveform and a square voltage waveform at the device output. Similarly, in inverse Class-F power amplifiers optimized by odd-harmonic short-circuit termination and even-harmonic open-circuit peaking, the fundamental and harmonic load-network impedances result in a half-sinusoidal voltage waveform and a square current waveform at the device output to obtain maximum efficiency.

The average efficiency of the power amplifier can further be increased by using a Doherty configuration with a Class-B or Class-AB bias in the carrier amplifier and a Class-C bias in the peaking amplifier [2]. In some cases, the Class-F design strategy can be applied to the carrier amplifier, whereas the peaking amplifier operates in a conventional Class-C mode [3]. For example, the load network of a Class-F/Class-C Doherty amplifier can be connected to the drain through the short-circuited quarterwave transmission lines providing inherent even-harmonic suppression [4]. In this case, the drain current of the Class-B biased carrier device contains the DC, fundamental-frequency, and even-harmonic components, whereas the drain current of the Class-C biased peaking device is purely sinusoidal. Implementation of both the carrier and peaking amplifiers with a Class-F load network containing a series quarterwave line can provide an increase of the overall efficiency and improvement of the harmonic suppression level [5]. In an unbalanced Doherty configuration when the carrier amplifier operates in a Class-F mode, while the peaking amplifier operates in an inverse Class-F mode, the backoff peak efficiency point can be shifted from the conventional 6 dB to 8.2 dB [6].

1509 dohertyAmps fig01

Figure 1 • Schematic of Class-F power amplifier with parallel transmission line.

Class F with Quarterwave Transmission Line

Ideally, a control of an infinite number of the harmonics maintaining a square voltage waveform and a half-sinusoidal current waveform at the drain can be provided using a parallel quarterwave transmission line and a series resonant circuit tuned to the fundamental frequency [2]. Figure 1 shows the circuit schematic of a Class-F power amplifier with a parallel quarterwave line using a 28-V 10-W Cree GaN HEMT CGH40010F device. For even harmonics, the short circuit on the top side of the quarterwave line is repeated, thus producing a short circuit at the drain. However, the short circuit at the top side of the quarterwave line produces an open circuit at the drain for odd harmonics. Here, the loaded quality factor of the series resonant circuit tuned to the fundamental frequency is sufficient to provide the sinusoidal output current flowing into the 50-Ω load.

To better illustrate the drain voltage and current waveforms with minimum effect of the device parasitic output parameters, a sufficiently low operating frequency of 200 MHz was chosen. In this case, a simple lossy RL input shunt network is used to match the device input impedance to a 50-Ω source and to compensate for the device input gate-source capacitance of about 5 pF at the fundamental frequency that results in a return loss greater than 20 dB. The length of the parallel quarterwave line was optimized to maximize efficiency taking into account the device parasitic parameters such as the drain-source capacitance Cds and series drain inductance provided by the combined bondwire and package lead inductance. The series 40-Ω resistor connected to the device gate is added to provide unconditional operation stability. The simulated drain voltage appears close to a square waveform and drain current similar to a half-sinusoidal waveform are shown in Fig. 2(a), where small waveform ripples (minimized by load-network parameter optimization) can be explained due to effects of the device output drain-source capacitance of about 1.3 pF and package parasitics at higher-order harmonic components. As a result, a maximum drain efficiency of 84.9% with a power gain of 17.2 dB at an input power of 23 dBm is obtained with a sine-wave driving signal, as shown in Fig. 2(b). Harmonic components are suppressed by greater than 42 dB.

1509 dohertyAmps fig02

Figure 2 • Simulated waveforms, gain, and efficiency of Class-F GaN HEMT power amplifier with parallel transmission line.

Figure 3 shows the circuit schematic of a Class-F power amplifier with a series quarterwave line using a 28-V 10-W Cree GaN HEMT CGH40010F device. This type of a Class-F power amplifier was initially proposed to be used at higher frequencies where implementation of the load networks with only lumped elements is difficult and the parasitic device output (bondwire or pack-age lead) inductance is sufficiently small [1]. For even harmonics, the short circuit on the load side of the transmission line is repeated, thus producing a short circuit at the drain. However, the short circuit at the load produces an open circuit at the drain for odd harmonics with resistive load at the fundamental. Unlike the case with a parallel quarterwave transmission line, such a Class-F load network configuration with a series quarterwave line can provide an impedance transformation. The loaded quality factor of the shunt resonant circuit is high enough to provide the sinusoidal current flowing into the 50-Ω load.

1509 dohertyAmps fig03

Figure 3 • Schematic of Class-F power amplifier with series transmission line.

In Class F, the optimum load resistance RL can be obtained for the DC supply voltage Vcc and fundamental-frequency output power Pout delivered to the load as [2]

1509 dohertyAmps eq01     (1)

Inverse Class F with Quarterwave Transmission Line

An idealized inverse Class-F operation mode can be represented by using a sequence of the series resonant circuits tuned to the fundamental and odd harmonics. An infinite set of the series resonant circuits tuned to the odd harmonics can be effectively replaced by a series quarterwave line with the same operating capability [7]. Such a circuit representation of an inverse Class-F power amplifier with a series quarterwave line loaded by the series resonant circuit tuned to the fundamental frequency and based on a 28-V 10-W Cree GaN HEMT CGH40010F device is shown in Fig. 4.

1509 dohertyAmps fig04

Figure 4 • Schematic of inverse Class-F power amplifier with transmission line.

The high-Q series-tuned output circuit presents to the transmission line a load resistance at the frequency of operation. For even harmonics, the open circuit on the load side of the transmission line is repeated, thus producing an open circuit at the drain. However, the quarterwave line converts the open circuit at the load to a short circuit at the drain for odd harmonics with resistive load at the fundamental frequency.

In inverse Class F, the optimum load resistance RL can be obtained for the DC supply voltage Vcc and fundamental-frequency output power Pout delivered to the load as [2]

1509 dohertyAmps eq02      (2)

Similarly to the conventional Class-F GaN HEMT power amplifier, a simple lossy RL input shunt network is also used to match the device input impedance to a 50-Ω source and to compensate for the device input gate-source capacitance at the fundamental frequency. The length of the series quarterwave line was optimized to maximize efficiency taking into account the device parasitic parameters. The simulated drain voltage (close to half-sinusoidal) and current (close to square) waveforms are shown in Fig. 5(a), where small deviations from the ideal waveforms (with optimized load-network parameters) can be explained due to effect of the device output drain-source capacitance Cds and package parasitics. In this case, a maximum drain efficiency of 85.2% with a power gain of 14.6 dB at an input power of 27 dBm is obtained with a sine-wave driving signal at 200 MHz, as shown in Fig. 5(b).

1509 dohertyAmps fig05

Figure 5 • Simulated waveforms, gain, and efficiency of inverse Class-F GaN HEMT power amplifier with transmission line.

Doherty Amplifier Using Class-F Mode

Figure 6 shows the circuit schematic of a Doherty configuration based on a Class-F mode with a parallel quarterwave line for the peaking amplifier and on a Class-F mode with a series quarter-wave line for the carrier amplifier using two 28-V 10-W Cree GaN HEMT CGH40010F devices. Here, a series quarterwave line also provides an impedance transformation required at backoff output power levels. According to Eq. (1), both carrier and peaking devices see the same load impedance close to 50 Ω at Vcc = 28 V and Pout = 10 W, assuming the saturation drain voltage of about 2-3 V. The output quarterwave transmission line with a characteristic impedance of 35.3 Ω is necessary to match with a standard 50-Ω load. The quadrature 90° hybrid coupler is used at the input to split signals between the carrier and peaking amplifying paths.

1509 dohertyAmps fig06

Figure 6 • Circuit schematic of GaN HEMT Doherty amplifier using Class-F mode.

In this case, the conventional balanced Doherty structure, where both the carrier and peaking amplifiers are operated in a Class-F mode, provide a maximum output power of 43.2 dBm with a drain efficiency of 85.6% and a power gain of 13.2 dB and a backoff peak efficiency point at an output power of 37.0 dBm (6.2-dB backoff) with a drain efficiency of 68% and a power gain of 17 dB, as shown in Fig. 7.

1509 dohertyAmps fig07

Figure 7. Power gain and efficiency of GaN HEMT Doherty amplifier using Class-F mode.

Doherty Amplifier Using Class-F and Inverse Class-F Modes

Figure 8 shows the circuit schematic of a Doherty configuration using a Class-F mode with a parallel quarterwave line for the peaking amplifier and an inverse Class-F mode with a series quarterwave line for the carrier amplifier. Here, a series quarterwave line also provides an impedance transformation required at backoff output power levels. The quadrature 90° hybrid coupler is used at the input to split signals between the carrier and peaking amplifying paths. From Eqs. (1) and (2), it follows that the carrier device operating in an inverse Class-F mode sees the load impedance by 1.5 times greater than the peaking device operating in a Class-F mode, thus resulting in an unbalanced Doherty structure in nominal operation.

1509 dohertyAmps fig08

Figure 8 • Circuit schematic of GaN HEMT Doherty amplifier using Class-F and inverse Class-F modes.

As a result, when using an impedance inverter with the characteristic impedance of Z1 = 50 Ω and an output transformer with the characteristic impedance of Z2 = 35.3 Ω, a maximum output power of 43.3 dBm with a drain efficiency of 87.0% and a power gain of 14.2 dB and a backoff peak efficiency point at an output power of 38.3 dBm (5-dB backoff) with a drain efficiency of 80.0% and a power gain of 16.3 dB are achieved, as shown in Fig. 9(a). In this case, the DC current flowing through the carrier device is 0.53 A, whereas the DC current flowing through the peaking device is 0.34 A, demonstrating an unbalanced operation condition. To provide the balanced operation mode with the same DC current of 0.46 A in both the carrier and peaking ampli-fiers corresponding to a 6-dB backoff efficiency point, the characteristic impedances Z1 and Z2 should be set to 55 Ω and 32.3 Ω, respectively.

1509 dohertyAmps fig09

Figure 9 • Power gain and efficiency of GaN HEMT Doherty amplifier using Class-F and inverse Class-F modes.

Further increase in the characteristic impedance of the impedance inverter to Z1 = 60 Ω and the corresponding decrease in the characteristic impedance of the output transformer to Z2 = 30.3 Ω result in a maximum output power of 43.8 dBm with a drain efficiency of 85.2% and a power gain of 15.5 dB and a backoff peak efficiency point at an output power of 35.8 dBm (8-dB backoff) with a drain efficiency of 82.9% and a power gain of 14.0 dB, as shown in Fig. 9(b). Here, the DC current of the peaking device increases to 0.54 A, while the DC current for the carrier device reduces to 0.38 A. To achieve a 9-dB backoff peak efficiency point, it needs to increase the characteristic impedance of the impedance inverter to Z1 = 65 Ω and to decrease the characteristic impedance of the output transformer to Z2 = 28.3 Ω. In this case, the DC current through the carrier device becomes equal to 0.35 A, whereas the DC current through the peaking device grows to 0.72 A (maximum allowable drain current for CGH40010F is 1.5 A). Thus, a 4-dB improvement in backoff peak efficiency point can be achieved by simply increasing the characteristic impedance of the impedance inverter from 50 to 65 Ω (1.3 times) and decreasing of the characteristic impedance of the output transformer from 35.3 to 28.5 Ω (by less than 1.25 times).

About the Author

Dr. Andrei Grebennikov (Senior Member of IEEE) of Microsemi Corp., Aliso Viejo, Calif., received his Dipl. Eng. degree in radio electronics from the Moscow Institute of Physics and Technology and Ph.D. degree in radio engineering from the Moscow Technical University of Communications and Informatics in 1980 and 1991, respectively. He obtained long-term academic and industrial experience working with Moscow Technical University of Communications and Informatics (Russia), Institute of Microelectronics (Singapore), M/A-COM (Ireland), Infineon Technologies (Germany/Austria), Bell Labs, Alcatel-Lucent (Ireland), and Microsemi (USA) as an engineer, researcher, lecturer, and educator. He lectures as a Guest Professor at University of Linz (Austria) and presented short courses and tutorials as an Invited Speaker at International Microwave Symposia, European and Asia-Pacific Microwave Conferences, Institute of Microelectronics, Singapore, Motorola Design Centre, Malaysia, Tomsk State University of Control Systems and Radioelectronics, Russia, and Aachen Technical University, Germany. He is an author and co-author of more than 100 papers, 25 European and US patents and patent applications, and eight books dedicated to RF and microwave circuit design.

References

1. H. L. Krauss, C. W. Bostian, and F. H. Raab, Solid State Radio Engineering, New York: John Wiley & Sons, 1980.

2. A. Grebennikov, N. O. Sokal, and M. J. Franco, Switchmode RF and Microwave Power Amplifiers, New York: Academic Press, 2012.

3. P. Colantonio, F. Giannini, R. Giofre, and L. Piazzon, “Theory and experimental results of a Class F AB-C Doherty power amplifier,” IEEE Trans. Microwave Theory Tech., vol. MTT-57, pp. 1936-1947, Aug. 2009.

4. K. W. Eccleston, K. J. I. Smith, and P. T. Gough, “A compact Class-F/Class-C Doherty amplifier,” Microwave and Optical Technology Lett., vol. 53, pp. 1606-1610, July 2011.

5. Y. S. Lee, M. Woo, and Y. H. Jeong, “High-efficiency Doherty amplifier using GaN HEMT Class-F cells for WCDMA applications,” Proc. 2008 IEEE Int. Microwave Millimeter-Wave Technol. Conf., vol. 1, pp. 270-273.

6.S. Goto, T. Kunii, A. Inoue, K. Izawa, T. Ishikawa, and Y. Matsuda, “Efficiency enhancement of Doherty amplifier with combination of Class-F and Inverse Class-F schemes for S-Band base station application,” 2004 IEEE MTT-S Int. Microwave Symp. Dig., pp. 839-842.

7.M. K. Kazimierczuk, “A new concept of Class F tuned power amplifier,” Proc. 27th Midwest Circuits and Systems Symp., pp. 425-428, 1984.