HF Outphasing Transmitter Using Class-E Power ... - IEEE Xplore
HF outphasing transmitter using class-E power amplifiers Ramon Beltran*, Frederick H. Raab#, and Arturo Velazquez* *
CICESE Research Center, Ensenada, B. C., Mexico Green Mountain Radio Research, Colchester, VT, USA
#
Abstract— Chireix outphasing systems have been used successfully with power amplifiers operating in class-B and F. However, when class-E PAs are used in these systems, either the efficiency or the dynamic range is poor. The asymmetric combining technique described here uses transmission lines or equivalent networks of different electrical lengths to position the impedance loci to provide good dynamic range while maintaining high efficiency. This in turn allows linear power amplification of amplitude-modulated signals with high average efficiency. The concept is verified by a 28-W PEP transmitter that operates at 1.82 MHz and achieves an efficiency of 85% or better over an amplitude range of 10 dB. Index Terms— Average efficiency. Class-E, power amplifier (PA), Chireix, asymmetric combiner, outphasing.
I. INTRODUCTION
H
IGH-EFFICIENCY linear amplification is of interest in modern communication systems as it increases talk time, decreases power consumption, decreases heat dissipation, and improves reliability. It is achieved by combining highefficiency power amplifiers (PAs) and a transmitter architecture such as envelope elimination, Doherty, or outphasing [1]. An outphasing transmitter (Fig. 1) produces a variableamplitude output by varying the phases of the driving signals to its RF-power amplifiers. The phase modulation causes the instantaneous vector sum of outputs of the two PAs to follow the desired signal amplitude. In a microwave implementation, power combiners based upon transmission lines are used. The outphasing transmitter, also known as linear amplification using non-linear components (LINC) was originally developed to provide linear amplification with active devices that have poor linearity [1], [2]. Chireix added complementary shunt reactances at the inputs of the combiner to improve the efficiency near the amplitude of the unmodulated AM carrier. Outphasing is attractive because signal phase can easily be modulated over a wide bandwidth. Research to date on using the Chireix technique at microwave frequencies has focused on using class-B, -D, and -F PAs, [3], [4]. The class-E amplifier offers excellent efficiency (ideally 100 percent) at RF and microwave frequencies [1]. Hybrid combining presents the PAs with constant, resistive loads. This allows good dynamic range but produces poor efficiency for all but peak output (both PAs in phase). The
978-1-4244-2804-5/09/$25.00 © 2009 IEEE
Fig.1. Class-E PAs outphasing with asymmetric transmission line combiner.
impedances presented to class-E PAs by a Chireix combiner are generally not suitable for class-E operation and produce poor efficiency, poor dynamic range, or both. This paper presents a new concept that uses class-E PAs and asymmetric combining to produce high efficiency over a wide range of amplitude, Fig. 1. This is confirmed by simulation and experiment with a prototype that uses two class-E PAs with 96% efficiency and 14-W output power each (resistive load), operating at 1.82 MHz, delivering 28-W at full output (0°) in asymmetric outphasing configuration. This frequency allows implementation of the PAs using low-cost MOSFETs, true transient class-E operation, and direct observation of the drain waveforms. The amplifiers output networks and asymmetric combiner (transmission lines W1 and W2 in Fig. 4) were implemented using lumped elements. II. CONVENTIONAL OUTPHASING In an outphasing transmitter, the output signal is proportional to the sine of the difference in phase between the two inputs to the combiner; i.e.: V0 m = V DD sin ϕ m , and the phase modulation must be therefore: ϕ m (t ) = arcsin E(t ) . Saturated PAs operating in classes A, B, C, D, and F act approximately as the voltage sources needed for proper operation. The impedance locus produced by simple outphasing causes the PAs to be subjected to reactive loads that degrade their efficiencies. (square markers in Fig. 2). The addition of the Chireix shunt impedances moves the impedance loci to regions of higher efficiency (diamond markers in Fig. 2). Conventional Outphasing with Class-E PAs The ideal load-pull contours of an ideal class-E PA are shown in Fig. 2. Comparison of the impedance loci of the conventional simple outphasing (square markers), to those of asymmetric class-E (circular markers), shows immediately
757
IMS 2009
that there are problems in using class-E PAs in the conventional outphasing architecture (square markers). The outputs of the two PAs vary, but are not changed by the same amount. In fact, they can maintain a nearly constant output voltage over a wide range of phase differences because the output of PA2 increases while that of PA1 decreases. Equally important is that efficiency of PA2 decreases rapidly as its load impedance moves toward the right side of the Smith chart. This is verified by direct observation of the drain waveforms (Fig. 3) where the asymmetric loading of the PAs is observed for the conventional simple outphasing using class-E PAs (square markers), while both PAs present very similar drain waveforms at PEP (0°- Fig. 3 (a)), the reactive loads cause the waveforms to vary asymmetrically (amplitude and shape) for drive phases others than 0° (Fig. 3, (b)-(d)), both output and efficiency then drop as the phase difference is increased. While the PEP efficiency is compared with that of a single class-E PA that is better than a linear amplifier, the averageefficiency characteristics are about the same [5]. When Chireix reactances (diamond markers) are added, the system presents very good efficiency over all drive phases. However, it delivers nearly constant output power regardless the drive phase (Fig. 4). As a result, it does not provide a useful dynamic range required for communication applications. One important observation made during the experiments is that for certain Chireix reactances and certain phase differences, one PA is presented with negative load resistance. As a result, it may rectify the signal from the other PA, causing it to have a negative dc-input power, which in turn results in erratic measurements of efficiency. III. ASYMMETRIC OUPHASING FOR CLASS E
Fig. 2. Impedance loci for outphasing with ideal class-E.
(a)
(b)
(c) (d) Fig. 3. Drain waveforms (PA1 right, PA2 left) of the conventionalsimple outphasing class-E PAs; (a) 0° (20 V/div), (b) 90°, (c) 135°, and (d) 170° (40 V/div).
The new architecture for outphasing of class-E PAs is shown in Fig 1. Transmission lines W1 and W2 have electrical lengths of θ+δ and θ-δ, respectively. The essence of the new combining technique is as follows: (1) Transmission lines W1 and W2 rotate the impedance loci to center them on the line at 65° corresponding to the maximum efficiency of the class-E PAs. This requires θ ≈147.5°. (2) As a result of centering the impedance loci, the amplitudes of the two PA outputs ideally vary identically with phase difference. However, their variation with ϕ m is not in general linear nor a simple function, and in practice they may not be exactly equal. Predistortion is used to set the phases to produce the desired amplitude. (3) The phase shifts within the two PAs also vary, and are not necessarily the same in both PAs. Predistortion compensates for the phase shifts. (4) Either differential line length δ or shunt susceptance Bs can be used to move the impedance loci closer to the η=1 line.
Fig. 4. Conventional Chireix-outphasing class-E efficiency and output power.
The value of δ, Bs, or both can be chosen to optimize the average efficiency for a given signal or set of signals. (5) In practice, it is also necessary to give due attention to obtaining a reasonable dynamic range of output voltage.
758
Experimental Verification of the Asymmetric Outphasing The prototype system operating at 1.82 MHz is used to verify the new technique. These results are also confirmed by simulation using Pspice. Prototype Two identical class-E power amplifiers were implemented using L-C series-tuned output networks with a Q of 10. The two amplifiers are assembled on a single board using IRF510 MOSFETs. The properly tuned PAs achieve a drain efficiency of 96 % with an output power of 14 W each, (Fig. 5). The prototype asymmetric combiner is implemented using lumped elements T networks instead of transmission lines W1 and W2, (Fig. 5). The values of the LC elements are determined by equating the ABCD matrix of a transmission line with that of the desired T network. Results of Simulation The results of simulation are shown in Fig 6. The new technique not only has a good dynamic range of amplitude variation, but also maintains an efficiency of 85% or better for amplitudes from 0.8 W to full output (27.5 W). In contrast, the efficiency for outphasing with ideal class-B PAs is no better than 78%. Indeed the asymmetric class-E outphasing ideally presents better efficiency when δ=0° (asymmetric but not optimized) over most output voltages compared to that of class-B, Fig. 7.
Fig. 5. Prototype class-E outphasing transmitter with combiner based upon asymmetric lumped-element T networks.
Measured Performance The measured characteristics of the prototype are shown in Fig. 8. The output power varies from 25.7 W at PEP (0°) to nearly zero (0.54 W) for a phase difference of 180°. The efficiency is better than that of the ideal class B over the entire range of amplitudes (signal processing is not included). The value of the difference δ in line length does not appear to be critical, and high efficiency is obtained for 10°, 18.4°, 22° and 48°, as shown in Fig. 9.
(a)
IV. APPLICATIONS Maintaining good instantaneous efficiency over a wide range of amplitudes will result in significant improvements in the average efficiency when producing amplitude-modulated signals. The predicted average efficiency of the asymmetric class-E outphasing is 98.7% for a signal with a square-root raised-cosing (SRRC) envelope (3.8-dB peak to average ratio) and 88.7% for a Rayleigh-signal with a 10-dB peak to average ratio ( ξ ). In contrast, the average efficiencies for an ideal linear class-B PA are only 51.8% and 28%, respectively. Fig. 10 shows average efficiencies comparison among optimized ideal Chireix-outphasing with class-B and F PAs and the asymmetric class-E outphasing for SSB ( ξ = 10 dB). Multitone (Rayleigh envelope with ξ = 5 , ξ = 10 and ξ = 20 dB) and SRRC ( ξ = 3.8 dB). The highly reactive loads make a Chireix transmitter with class B and F PAs very inefficient at low outputs, Fig 8, this impact on the average efficiencies, Fig 10.
(b) Fig. 6. Simulated class-E outphasing performance, output power and efficiency vs. drive phase (a), and efficiency vs. output voltage (b), for θ =147.5° and δ=17.4°.
Fig. 7. Simulated class-E outphasing efficiency vs. normalized output voltage for various values of δ with θ =147.5°.
759
(a)
Fig. 10. Average efficiency for various signals.
V. UHF IMPLEMENTATION (b) Fig. 8. Measured class-E outphasing performance, output power and efficiency vs. drive phase (a) and efficiency vs. normalized output voltage (b), for θ =147.5° and δ=17.4°.
Preliminary experiments suggest that this technique can also be used at microwave frequencies. Class-E operation at UHF and microwave frequencies is approximated by using transmission-line networks to control a finite number of harmonics [6]. The asymmetric combiner is implemented with true transmission lines. REFERENCES
Fig. 9. Measured comparison of efficiency vs. output voltage for different δ in its respective T networks.
760
[1] F. H. Raab, et al,” Power Amplifiers and Transmitters for RF and Microwave,” IEEE Trans. on Microwave Theory and Techniques, vol. 50, no. 8, March 2002. [2] H. Chireix, "High power outphasing modulation," Proc. IRE, vol. 23, no.11, pp. 1370 - 1392, Nov. 1935. [3] T. Hung, et al, “CMOS Outphasing Class-D Amplifier With Chireix Combiner” IEEE Microwaves and wireless components letters, vol. 17, no 8, August 2007. [4] J. Grundlingh, K. Parker, G. Rabjohn, “A high efficiency Chireix Outphasing power amplifier for 5 GHz WLAN applications”, IEEE MTT-S International Microwave SymposiumDigest, 2004, Fort Worth, TX, USA, Jun. 6-11, 2004. [5] D. Kawamoto, et al, "Design of a generalized phase controlled class E inverter," IEEE Int. Symp. on Circuits and Systems (ISCAS 2001), Sydney, Australia, May 6 - 9, 2001. [6] F. H. Raab, “Class-E, Class-C, and Class-F Power amplifiers based upon a finite number of harmonics”, IEEE Transactions on Microwave Theory and Techniques, vol. 49, no. 8, August 2001.
CICESE Research Center, Ensenada, B. C., Mexico Green Mountain Radio Research, Colchester, VT, USA
#
Abstract— Chireix outphasing systems have been used successfully with power amplifiers operating in class-B and F. However, when class-E PAs are used in these systems, either the efficiency or the dynamic range is poor. The asymmetric combining technique described here uses transmission lines or equivalent networks of different electrical lengths to position the impedance loci to provide good dynamic range while maintaining high efficiency. This in turn allows linear power amplification of amplitude-modulated signals with high average efficiency. The concept is verified by a 28-W PEP transmitter that operates at 1.82 MHz and achieves an efficiency of 85% or better over an amplitude range of 10 dB. Index Terms— Average efficiency. Class-E, power amplifier (PA), Chireix, asymmetric combiner, outphasing.
I. INTRODUCTION
H
IGH-EFFICIENCY linear amplification is of interest in modern communication systems as it increases talk time, decreases power consumption, decreases heat dissipation, and improves reliability. It is achieved by combining highefficiency power amplifiers (PAs) and a transmitter architecture such as envelope elimination, Doherty, or outphasing [1]. An outphasing transmitter (Fig. 1) produces a variableamplitude output by varying the phases of the driving signals to its RF-power amplifiers. The phase modulation causes the instantaneous vector sum of outputs of the two PAs to follow the desired signal amplitude. In a microwave implementation, power combiners based upon transmission lines are used. The outphasing transmitter, also known as linear amplification using non-linear components (LINC) was originally developed to provide linear amplification with active devices that have poor linearity [1], [2]. Chireix added complementary shunt reactances at the inputs of the combiner to improve the efficiency near the amplitude of the unmodulated AM carrier. Outphasing is attractive because signal phase can easily be modulated over a wide bandwidth. Research to date on using the Chireix technique at microwave frequencies has focused on using class-B, -D, and -F PAs, [3], [4]. The class-E amplifier offers excellent efficiency (ideally 100 percent) at RF and microwave frequencies [1]. Hybrid combining presents the PAs with constant, resistive loads. This allows good dynamic range but produces poor efficiency for all but peak output (both PAs in phase). The
978-1-4244-2804-5/09/$25.00 © 2009 IEEE
Fig.1. Class-E PAs outphasing with asymmetric transmission line combiner.
impedances presented to class-E PAs by a Chireix combiner are generally not suitable for class-E operation and produce poor efficiency, poor dynamic range, or both. This paper presents a new concept that uses class-E PAs and asymmetric combining to produce high efficiency over a wide range of amplitude, Fig. 1. This is confirmed by simulation and experiment with a prototype that uses two class-E PAs with 96% efficiency and 14-W output power each (resistive load), operating at 1.82 MHz, delivering 28-W at full output (0°) in asymmetric outphasing configuration. This frequency allows implementation of the PAs using low-cost MOSFETs, true transient class-E operation, and direct observation of the drain waveforms. The amplifiers output networks and asymmetric combiner (transmission lines W1 and W2 in Fig. 4) were implemented using lumped elements. II. CONVENTIONAL OUTPHASING In an outphasing transmitter, the output signal is proportional to the sine of the difference in phase between the two inputs to the combiner; i.e.: V0 m = V DD sin ϕ m , and the phase modulation must be therefore: ϕ m (t ) = arcsin E(t ) . Saturated PAs operating in classes A, B, C, D, and F act approximately as the voltage sources needed for proper operation. The impedance locus produced by simple outphasing causes the PAs to be subjected to reactive loads that degrade their efficiencies. (square markers in Fig. 2). The addition of the Chireix shunt impedances moves the impedance loci to regions of higher efficiency (diamond markers in Fig. 2). Conventional Outphasing with Class-E PAs The ideal load-pull contours of an ideal class-E PA are shown in Fig. 2. Comparison of the impedance loci of the conventional simple outphasing (square markers), to those of asymmetric class-E (circular markers), shows immediately
757
IMS 2009
that there are problems in using class-E PAs in the conventional outphasing architecture (square markers). The outputs of the two PAs vary, but are not changed by the same amount. In fact, they can maintain a nearly constant output voltage over a wide range of phase differences because the output of PA2 increases while that of PA1 decreases. Equally important is that efficiency of PA2 decreases rapidly as its load impedance moves toward the right side of the Smith chart. This is verified by direct observation of the drain waveforms (Fig. 3) where the asymmetric loading of the PAs is observed for the conventional simple outphasing using class-E PAs (square markers), while both PAs present very similar drain waveforms at PEP (0°- Fig. 3 (a)), the reactive loads cause the waveforms to vary asymmetrically (amplitude and shape) for drive phases others than 0° (Fig. 3, (b)-(d)), both output and efficiency then drop as the phase difference is increased. While the PEP efficiency is compared with that of a single class-E PA that is better than a linear amplifier, the averageefficiency characteristics are about the same [5]. When Chireix reactances (diamond markers) are added, the system presents very good efficiency over all drive phases. However, it delivers nearly constant output power regardless the drive phase (Fig. 4). As a result, it does not provide a useful dynamic range required for communication applications. One important observation made during the experiments is that for certain Chireix reactances and certain phase differences, one PA is presented with negative load resistance. As a result, it may rectify the signal from the other PA, causing it to have a negative dc-input power, which in turn results in erratic measurements of efficiency. III. ASYMMETRIC OUPHASING FOR CLASS E
Fig. 2. Impedance loci for outphasing with ideal class-E.
(a)
(b)
(c) (d) Fig. 3. Drain waveforms (PA1 right, PA2 left) of the conventionalsimple outphasing class-E PAs; (a) 0° (20 V/div), (b) 90°, (c) 135°, and (d) 170° (40 V/div).
The new architecture for outphasing of class-E PAs is shown in Fig 1. Transmission lines W1 and W2 have electrical lengths of θ+δ and θ-δ, respectively. The essence of the new combining technique is as follows: (1) Transmission lines W1 and W2 rotate the impedance loci to center them on the line at 65° corresponding to the maximum efficiency of the class-E PAs. This requires θ ≈147.5°. (2) As a result of centering the impedance loci, the amplitudes of the two PA outputs ideally vary identically with phase difference. However, their variation with ϕ m is not in general linear nor a simple function, and in practice they may not be exactly equal. Predistortion is used to set the phases to produce the desired amplitude. (3) The phase shifts within the two PAs also vary, and are not necessarily the same in both PAs. Predistortion compensates for the phase shifts. (4) Either differential line length δ or shunt susceptance Bs can be used to move the impedance loci closer to the η=1 line.
Fig. 4. Conventional Chireix-outphasing class-E efficiency and output power.
The value of δ, Bs, or both can be chosen to optimize the average efficiency for a given signal or set of signals. (5) In practice, it is also necessary to give due attention to obtaining a reasonable dynamic range of output voltage.
758
Experimental Verification of the Asymmetric Outphasing The prototype system operating at 1.82 MHz is used to verify the new technique. These results are also confirmed by simulation using Pspice. Prototype Two identical class-E power amplifiers were implemented using L-C series-tuned output networks with a Q of 10. The two amplifiers are assembled on a single board using IRF510 MOSFETs. The properly tuned PAs achieve a drain efficiency of 96 % with an output power of 14 W each, (Fig. 5). The prototype asymmetric combiner is implemented using lumped elements T networks instead of transmission lines W1 and W2, (Fig. 5). The values of the LC elements are determined by equating the ABCD matrix of a transmission line with that of the desired T network. Results of Simulation The results of simulation are shown in Fig 6. The new technique not only has a good dynamic range of amplitude variation, but also maintains an efficiency of 85% or better for amplitudes from 0.8 W to full output (27.5 W). In contrast, the efficiency for outphasing with ideal class-B PAs is no better than 78%. Indeed the asymmetric class-E outphasing ideally presents better efficiency when δ=0° (asymmetric but not optimized) over most output voltages compared to that of class-B, Fig. 7.
Fig. 5. Prototype class-E outphasing transmitter with combiner based upon asymmetric lumped-element T networks.
Measured Performance The measured characteristics of the prototype are shown in Fig. 8. The output power varies from 25.7 W at PEP (0°) to nearly zero (0.54 W) for a phase difference of 180°. The efficiency is better than that of the ideal class B over the entire range of amplitudes (signal processing is not included). The value of the difference δ in line length does not appear to be critical, and high efficiency is obtained for 10°, 18.4°, 22° and 48°, as shown in Fig. 9.
(a)
IV. APPLICATIONS Maintaining good instantaneous efficiency over a wide range of amplitudes will result in significant improvements in the average efficiency when producing amplitude-modulated signals. The predicted average efficiency of the asymmetric class-E outphasing is 98.7% for a signal with a square-root raised-cosing (SRRC) envelope (3.8-dB peak to average ratio) and 88.7% for a Rayleigh-signal with a 10-dB peak to average ratio ( ξ ). In contrast, the average efficiencies for an ideal linear class-B PA are only 51.8% and 28%, respectively. Fig. 10 shows average efficiencies comparison among optimized ideal Chireix-outphasing with class-B and F PAs and the asymmetric class-E outphasing for SSB ( ξ = 10 dB). Multitone (Rayleigh envelope with ξ = 5 , ξ = 10 and ξ = 20 dB) and SRRC ( ξ = 3.8 dB). The highly reactive loads make a Chireix transmitter with class B and F PAs very inefficient at low outputs, Fig 8, this impact on the average efficiencies, Fig 10.
(b) Fig. 6. Simulated class-E outphasing performance, output power and efficiency vs. drive phase (a), and efficiency vs. output voltage (b), for θ =147.5° and δ=17.4°.
Fig. 7. Simulated class-E outphasing efficiency vs. normalized output voltage for various values of δ with θ =147.5°.
759
(a)
Fig. 10. Average efficiency for various signals.
V. UHF IMPLEMENTATION (b) Fig. 8. Measured class-E outphasing performance, output power and efficiency vs. drive phase (a) and efficiency vs. normalized output voltage (b), for θ =147.5° and δ=17.4°.
Preliminary experiments suggest that this technique can also be used at microwave frequencies. Class-E operation at UHF and microwave frequencies is approximated by using transmission-line networks to control a finite number of harmonics [6]. The asymmetric combiner is implemented with true transmission lines. REFERENCES
Fig. 9. Measured comparison of efficiency vs. output voltage for different δ in its respective T networks.
760
[1] F. H. Raab, et al,” Power Amplifiers and Transmitters for RF and Microwave,” IEEE Trans. on Microwave Theory and Techniques, vol. 50, no. 8, March 2002. [2] H. Chireix, "High power outphasing modulation," Proc. IRE, vol. 23, no.11, pp. 1370 - 1392, Nov. 1935. [3] T. Hung, et al, “CMOS Outphasing Class-D Amplifier With Chireix Combiner” IEEE Microwaves and wireless components letters, vol. 17, no 8, August 2007. [4] J. Grundlingh, K. Parker, G. Rabjohn, “A high efficiency Chireix Outphasing power amplifier for 5 GHz WLAN applications”, IEEE MTT-S International Microwave SymposiumDigest, 2004, Fort Worth, TX, USA, Jun. 6-11, 2004. [5] D. Kawamoto, et al, "Design of a generalized phase controlled class E inverter," IEEE Int. Symp. on Circuits and Systems (ISCAS 2001), Sydney, Australia, May 6 - 9, 2001. [6] F. H. Raab, “Class-E, Class-C, and Class-F Power amplifiers based upon a finite number of harmonics”, IEEE Transactions on Microwave Theory and Techniques, vol. 49, no. 8, August 2001.
