High Efficiency Classes of RF Amplifiers

Similar documents
A New Topology of Load Network for Class F RF Power Amplifiers

Design and simulation of Parallel circuit class E Power amplifier

ANALYSIS OF BROADBAND GAN SWITCH MODE CLASS-E POWER AMPLIFIER

High Efficiency Class-F MMIC Power Amplifiers at Ku-Band

1 GHz Current Mode Class-D Power Amplifier in Hybrid Technology Using GaN HEMTs

Push-Pull Class-E Power Amplifier with a Simple Load Network Using an Impedance Matched Transformer

Design and Layout of a X-Band MMIC Power Amplifier in a Phemt Technology

In modern wireless. A High-Efficiency Transmission-Line GaN HEMT Class E Power Amplifier CLASS E AMPLIFIER. design of a Class E wireless

High Power Two- Stage Class-AB/J Power Amplifier with High Gain and

Switching Behavior of Class-E Power Amplifier and Its Operation Above Maximum Frequency

Design of a Broadband HEMT Mixer for UWB Applications

Application Note 5057

High-efficiency class E/F 3 power amplifiers with extended maximum operating frequency

Design of Class F Power Amplifiers Using Cree GaN HEMTs and Microwave Office Software to Optimize Gain, Efficiency, and Stability

California Eastern Laboratories

Expansion of class-j power amplifiers into inverse mode operation

Application Note 5379

A High Efficiency and Wideband Doherty Power Amplifier for 5G. Master s thesis in Wireless, Photonics and Space Engineering HALIL VOLKAN HUNERLI

Application Note 1373

CHAPTER 4 ULTRA WIDE BAND LOW NOISE AMPLIFIER DESIGN

Dual-band LNA Design for Wireless LAN Applications. 2.4 GHz LNA 5 GHz LNA Min Typ Max Min Typ Max

This article has been accepted and published on J-STAGE in advance of copyediting. Content is final as presented.

Design of an Efficient Single-Stage and 2-Stages Class-E Power Amplifier (2.4GHz) for Internet-of-Things

Downloaded from edlib.asdf.res.in

Load Pull Validation of Large Signal Cree GaN Field Effect Transistor (FET) Model

The Design of A 125W L-Band GaN Power Amplifier

6.776 High Speed Communication Circuits and Systems Lecture 14 Voltage Controlled Oscillators

Linearization of Three-Stage Doherty Amplifier

ECEN 5014, Spring 2009 Special Topics: Active Microwave Circuits Zoya Popovic, University of Colorado, Boulder

TUNED AMPLIFIERS 5.1 Introduction: Coil Losses:

Uneven Doherty Amplifier Based on GaN HEMTs Characteristic

A 1-W GaAs Class-E Power Amplifier with an FBAR Filter Embedded in the Output Network

6-18 GHz MMIC Drive and Power Amplifiers

Linearization Method Using Variable Capacitance in Inter-Stage Matching Networks for CMOS Power Amplifier

Methodology for MMIC Layout Design

NI AWR Design Environment Load-Pull Simulation Supports the Design of Wideband High-Efficiency Power Amplifiers

Inverse Class F Power Amplifier for WiMAX Applications with 74% Efficiency at 2.45 GHz

Wideband and High Efficiency Feed-Forward Linear Power Amplifier for Base Stations

Analyzing Device Behavior at the Current Generator Plane of an Envelope Tracking Power Amplifier in a High Efficiency Mode

RF/Microwave Amplifier Design Using Harmonic Balance Simulation With Only S-parameter Data

DESIGN OF HIGH POWER AND EFFICIENT RF LDMOS PA FOR ISM APPLICATIONS

A CMOS GHz UWB LNA Employing Modified Derivative Superposition Method

A Novel Dual-Band Balanced Power Amplifier Using Branch-Line Couplers with Four Arbitrary Terminated Resistances

Class E/F Amplifiers

DESIGN OF AN S-BAND TWO-WAY INVERTED ASYM- METRICAL DOHERTY POWER AMPLIFIER FOR LONG TERM EVOLUTION APPLICATIONS

CHAPTER 3 CMOS LOW NOISE AMPLIFIERS

Application Note 1285

A 2.5-GHz GaN power amplifier design and modeling by circuit-electromagnetic co-simulation

Oscillators. An oscillator may be described as a source of alternating voltage. It is different than amplifier.

ATF-531P8 E-pHEMT GaAs FET Low Noise Amplifier Design for 800 and 900 MHz Applications. Application Note 1371

LABORATORY #3 QUARTZ CRYSTAL OSCILLATOR DESIGN

Design of High Efficiency Class E Amplifiers

0.5GHz - 1.5GHz Bandwidth 10W GaN HEMT RF Power Amplifier Design

Techniques to reduce electromagnetic noise produced by wired electronic devices

RECENT MOBILE handsets for code-division multiple-access

CLASS-C POWER AMPLIFIER DESIGN FOR GSM APPLICATION

10W Ultra-Broadband Power Amplifier

1 of 7 12/20/ :04 PM

15 GHz Voltage Controlled Osc Odeneho Anaman 10 GHz Voltage Controlled Osc Enoch Wong

ATF-531P8 900 MHz High Linearity Amplifier. Application Note 1372

Pulsed IV analysis. Performing and Analyzing Pulsed Current-Voltage Measurements PULSED MEASUREMENTS. methods used for pulsed

Low Noise Amplifier for 3.5 GHz using the Avago ATF Low Noise PHEMT. Application Note 1271

CHAPTER 4 LARGE SIGNAL S-PARAMETERS

Application Note 1299

A 600 GHz Varactor Doubler using CMOS 65nm process

Streamlined Design of SiGe Based Power Amplifiers

A Simulation-Based Flow for Broadband GaN Power Amplifier Design

A COMPACT SWITCHING MODE CLASS-F POWER AMPLFIER DESIGN

Design of High PAE Class-E Power Amplifier For Wireless Power Transmission

Leveraging High-Accuracy Models to Achieve First Pass Success in Power Amplifier Design

A COMPACT DUAL-BAND POWER DIVIDER USING PLANAR ARTIFICIAL TRANSMISSION LINES FOR GSM/DCS APPLICATIONS

High Power Wideband AlGaN/GaN HEMT Feedback. Amplifier Module with Drain and Feedback Loop. Inductances

Highly Linear GaN Class AB Power Amplifier Design

Today s wireless system

Highly linear common-gate mixer employing intrinsic second and third order distortion cancellation

A 2 4 GHz Octave Bandwidth GaN HEMT Power Amplifier with High Efficiency

New LDMOS Model Delivers Powerful Transistor Library Part 1: The CMC Model

A GHz MONOLITHIC GILBERT CELL MIXER. Andrew Dearn and Liam Devlin* Introduction

Wide-Band Two-Stage GaAs LNA for Radio Astronomy

L AND S BAND TUNABLE FILTERS PROVIDE DRAMATIC IMPROVEMENTS IN TELEMETRY SYSTEMS

Linearization of Broadband Microwave Amplifier

A GHz MICROWAVE UP CONVERSION MIXERS USING THE CONCEPTS OF DISTRIBUTED AND DOUBLE BALANCED MIXING FOR OBTAINING LO AND RF (LSB) REJECTION

Design of Low Noise Amplifier Using Feedback and Balanced Technique for WLAN Application

Design of an RF CMOS Power Amplifier for Wireless Sensor Networks

MAHALAKSHMI ENGINEERING COLLEGE TIRUCHIRAPALLI UNIT III TUNED AMPLIFIERS PART A (2 Marks)

i. At the start-up of oscillation there is an excess negative resistance (-R)

A linearized amplifier using self-mixing feedback technique

Analysis and Synthesis of phemt Class-E Amplifiers with Shunt Inductor including ON-State Active-Device Resistance Effects

Simulations of High Linearity and High Efficiency of Class B Power Amplifiers in GaN HEMT Technology

RF POWER amplifier (PA) efficiency is of critical importance

Feedback and Oscillator Circuits

K-BAND HARMONIC DIELECTRIC RESONATOR OS- CILLATOR USING PARALLEL FEEDBACK STRUC- TURE

Including the proper parasitics in a nonlinear

LINEARIZED CMOS HIGH EFFECIENCY CLASS-E RF POWER AMPLIFIER

Cardiff, CF24 3AA, Wales, UK

High Gain Low Noise Amplifier Design Using Active Feedback

Progress In Electromagnetics Research C, Vol. 19, , 2011

High efficiency linear

A highly efficient 3.5 GHz inverse class-f GaN HEMT power amplifier

87x. MGA GHz 3 V Low Current GaAs MMIC LNA. Data Sheet

Transcription:

Rok / Year: Svazek / Volume: Číslo / Number: Jazyk / Language 2018 20 1 EN High Efficiency Classes of RF Amplifiers - Erik Herceg, Tomáš Urbanec urbanec@feec.vutbr.cz, herceg@feec.vutbr.cz Faculty of Electrical Engineering and Communication Brno University of Technology DOI: - Abstract: This article is dealing with high efficiency RF amplifiers in modern classes F, E and J. The first part is focused on basic function, main parameters and the output matching topologies of the mentioned classes. Output voltage and current waveforms were simulated for each class of high efficiency amplifiers. The primary focus of this work is the practical design of class F amplifier for 435 MHz band with E-pHEMT transistor. Power added efficiency (PAE) of amplifier achieved 58% and output power was 27 dbm with 14 dbm of input power. Amplifier was realized exclusively with lumped components in order to adhere to the given dimensions. Class F amplifiers designed at megahertz frequencies and with E-pHEMT transistor are quite rare and this article could help designers with understanding narrowband F-class amplifiers with higher efficiency. This amplifier can be used in long range IoT application, because of its low consumption of energy which is necessary in this modern technology. All results were simulated within ADS Keysight environment. Every simulation was realized with nonlinear models from Modelithics.

High Efficiency Classes of RF Amplifiers Erik Herceg, Tomáš Urbanec Faculty of Electrical Engineering and Communication VUT Brno Email: {urbanec, herceg}@feec.vutbr.cz Abstract This article is focused on high efficiency RF amplifiers in modern classes as F, E and J. The first part of this paper is dealing with basic function, main parameters and the output matching topologies of the mentioned classes. For every class of high efficiency amplifiers output voltage and current waveforms were simulated. More attention is devoted to practical design of class F amplifier for 435 MHz band with E-pHEMT transistor. Power added efficiency (PAE) of amplifier achieved 58% and output power was 27 dbm with 14 dbm of input power. Amplifier was realized purely with lumped components to adhere given dimensions. Class F amplifiers designed at megahertz frequencies and with E-pHEMT transistor are quite rare and this article could help designers with understanding narrowband F-class amplifiers with higher efficiency. This amplifier can be used in long range IoT application, because of its low consumption of energy which is necessary in this modern technology. All results were simulated with ADS Keysight environment. Every simulation was realized with nonlinear models from Modelithics. 1 Introduction High efficiency amplifiers can be used where low power consumption is needed, for example battery-powered devices etc. In the case of high efficiency amplifiers is important to take into consideration the fact that the distortion is higher than in linear classes. In primary classes as A, AB, B the maximum efficiency in ideal conditions is limited up to 78.5%. High efficiency amplifiers can theoretically achieve 100% efficiency in ideal operation so the difference is significant. The rule of increasing efficiency is different in each class which is described in [1]. Using the transistor nonlinear model is the basic condition for designing of functional high efficiency amplifier. Nonlinear model enables to simulate the output current and the voltage hence the right operation of the amplifier can be demonstrated. Furthermore, nonlinear models allow to simulate Source and Load Pull (SP and LP) and at the same time it is possible to obtain a favourable combination of efficiency, available gain and output power in which the transistor operates the best. Class F amplifiers are taking advantage of forming voltage sine wave at fundamental frequency to square voltage in ideal case [2]. Fundamental frequency is affected by odd or even higher harmonics. The amplifier is working in class F in case it uses odd harmonic, otherwise it is working in inverse class F. For F-class amplifier, the odd harmonics termination should be ideally matched to infinity (Z L(odd) = ) and for even harmonics termination should represent short circuit (Z L(even) = 0). These conditions can be achieved more easily at higher frequencies, because microstrip transmission lines can be used and design is much simpler. The bias point is set to class B and the conduction angle is 180 so the current waveform is half sinusoidal, therefore there is the harmonic content in current spectrum but for forming the voltage waveform is not high enough. The solution is to overdrive transistor, what causes that the current waveform is deformed by the cut-off and magnitude of harmonic content in current spectrum will be increased as is shown later, in more detailed description. Class E amplifiers are considered as switching amplifiers. Switching frequency is equal to half of the fundamental frequency, so the duty cycle is 50%. In phase of sinewave, the switch is closed at 0 and opened at 180. The class E amplifier circuit includes shunted switching capacitor, series inductance and series resonant LC circuit. The series resonant circuit acts as a short circuit at fundamental frequency and open for higher harmonic content [3]. The main disadvantage of these types of amplifiers is slow discharging of shunt capacitance which leads to disability of usage at higher frequencies. For achieving the maximum efficiency, there are no such strict conditions with higher harmonics matching circuit as in class F amplifiers. The bias point should be set between 150 and 200, i.e. from strong class B to weak class C. Amplifiers in class D can be included between switching classes of amplifiers where transistor acts also as switch. When the switch is open then drain current reaches its maximum, so transistor is shunted due to low resistance and in opposite case if the transistor is off then voltage reaches its maximum. The last analysed class of amplifier is class J. Class J is used for increasing efficiency by shifting phase of second harmonic waveform. The phase of second harmonic waveform should be the same as the phase of fundamental, so the total voltage corresponds to sum of these waveforms. Nowadays many articles and books have been published with explanation of high efficiency amplifiers function. Operation of class F were described from very low frequencies up to tens of gigahertz [4]. Class E amplifiers are more suitable for lower frequencies, more accurately at UHF band maximum achieved efficiency was 80% [5]. Amplifiers in class J are more suitable for higher frequencies, at UHF band, class J amplifiers were not designed. Above 80% efficiency was achieved with class F at amplifier in C-band with LC ladder output topology [6]. Comparing the efficiency achieved in different classes for frequencies near X-band, the most efficient are amplifiers in class F and inverse class F with effectivity above 80%, second most effective is class E with effectivity below 80% and last is class J, where highest achieved efficiency was near 75% [7]. 1

2 Parameters of individual classes In this chapter, basic equations, waveforms and function will be explained for better understanding of high efficiency amplifiers issues. When ideal transistor is used, parasitic components are omitted. In real design parasitic components as drain-source capacitance and series inductance are producing most of issues when realising high efficiency nonlinear amplifiers. 2.1 Class F amplifiers As mentioned before, the class F amplifier uses higher harmonics to change form of fundamental sine wave to square wave. The example of combination 1 st, 3 rd and 5 th harmonic voltage is shown in Figure 1. Sum of these harmonics is represented by red line. The overdriving the amplifier with higher input power has the most significant influence on the magnitude of the components in current spectrum. With higher input power, peak of current sinusoidal waveform crashes almost to zero as shown at Figure 2. The deformation of the waveform results in the increase of current content in the spectrum and by the matching odd harmonics to high impedance causes forming the voltage waveform. Adequate topology to achieve conditions of open and short circuit can be achieved with resonators. Example circuit is shown in Figure 3. Resonator connected in series is designed to resonate on 3 rd harmonic, shunt resonator at fundamental frequency and its character capacitive of first and inductive of second resonator acts also as resonator but as short circuit for 2 nd harmonic. Figure 3: Topology of class F amplifier. Figure 1: Forming the square voltage waveform with higher harmonics. Current and voltage waveform of ideal class F amplifier is shown in Figure 2. Load for 3 rd and 5 th was set to 500 Ω and even harmonics were shorted. This was achieved with variable load which is available in simulation software. Simulated waveforms were obtained from ideal transistor with ideal IV curves and with knee voltage equalled to 0 Volts. Bias point of transistor has to be set to class B where the conductance angle of current waveform is 180 which means that the current waveform is composed of many higher harmonics. In this stage the magnitude of current components in current spectrum is not high enough for forming the voltage waveform. Input and output match has to be chosen with given requirements due to high gain compression of this class. Gain compression can achieve up to 6 db. The main issue during the design of class F amplifier was given by parasitic components of the transistor which are dependent on V DS and which are usually not specified by manufacturer. 2.2 Class E amplifier with shunt capacitance Unlike class F amplifier, class E is more applicable at lower frequencies due to limitation of the discharging shunt capacitance. Example topology of class F amplifier is shown at Figure 4. The increasing power dissipation caused by higher harmonics is prevented by adding a series LC resonator to the output circuit. Through the shunt capacitor is the opposite voltage half wave returning to the series resonator. Series inductance is used as phase shift to eliminate the DC offset and the capacitor in LC resonant circuit is also used as DC blocking capacitor. Figure 2: Class F current and voltage waveforms. Figure 4: Scheme of class E Amplifier. 2

Voltage and current waveforms with ideal switch are shown at Figure 5. For simpler description, transistor was replaced with ideal switch with output series capacitor. In this case the efficiency achieved only 81%, because of the dissipation of the higher harmonics created in switch, so the DC power is transformed to RF power at higher harmonics and that causes lower efficiency. Principle of achieving high efficiency is to move 1 st and 2 nd harmonic load impedance to opposite region in smith chart. In this case the voltage waveform can be phase shifted in two ways. The first way is move 1 st harmonic load impedance to capacitive region of Smith chart and 2 nd harmonic to inductive region of Smith chart. Second way is when 1 st harmonic is shifted to inductive and 2 nd is shifted to capacitive region. This means that class J amplifier uses capacitive termination which is mostly considered as linear. The voltage is approximately increased by square root of 2 compared to class B. Modes of class J amplifier are shown at Figure 7. Waveforms are highly idealized. Middle curve marked as β = 0 is class B operation, β = 1 is operation of Class J amplifier and last curve represents inverse class J amplifier. The meaning of β attribute can be explained by following equation: v t = v B t (1 + β cos(θ)), (1) in which v(t) is voltage in class J amplifier, v B(t) is voltage of amplifier operating in class B and angle θ is the phase of the voltage waveform.. Figure 5: Ideal waveforms of class E amplifier. The ideal switch was replaced with nonlinear model of FET transistor and the topology mentioned earlier was applied. The final waveforms are shown at Figure 6. Reached efficiency was 87%. As in class F, the amplifier needs to be overdriven to achieve the desired efficiency. Figure 7: Modes of Class J amplifier. 3 Practical design of class F amplifier Figure 6: Waveforms of class E amplifier with shunt capacitor. 2.3 Class J amplifiers As mentioned in introduction, class J amplifier is increasing voltage of fundamental frequency with 2 nd harmonic. The output voltage waveform is combination of 1 st and 2 nd harmonic which evidently needs to have the same phase. The main advantage is design without the resonators as in class E and F which can be difficult to design properly. This article deals also with the practical design of class F amplifier [7]. Amplifier was designed for 435 MHz and desired output power was 27 dbm with 16 db gain. Used type of transistor was E-pHEMT which was applicable from 400 MHz to 6 GHz. Class of operation was set to class B with V DS set to 4.5 V and V GS = 0.4 V with quiescent current equalled approximately to 100 ma. Amplifier had to be stable up to 3 rd harmonic frequency, because amplifier is not used in linear region but almost in saturation. Stability in this wideband region was solved with series RLC circuit between gate and drain of transistor. For better stability, vias were added directly under the transistors source pad as a manufacturer s recommendation. Selected topology was similar as shown in Figure 3, so with two resonators. All resonators were designed with lumped components, because the quarter of wavelength at fundamental frequency is more than 8 cm and the design was limited by required dimensions of PCB. 3

Transistor was matched from Z L= 14 + 12j Ω to output impedance equalled to 50 Ω at fundamental frequency. Even harmonics were shorted and odd were set to high impedance. All capacitors in resonators were replaced with capacitor trimmers for better tuning. To verify right matching of input and output, S-parameters were measured; S 11,S 12 and S 22 parameters are shown at Figure 8. The gain at fundamental frequency achieved 24 db and S 11 and S 22 were less than -10 db, what can be considered as right matched amplifier. As input matching was chosen high pass LC circuit which also acts as DC block at input. Ideal output impedance was selected from load pull simulation with regard to compromise between PAE and output power. High impedance for 3 rd harmonic and short circuit for 2 nd harmonic was set with multiresonance circuit. Figure 11: Photography of realised amplifier. [7] 4 Conclusion Figure 8: Measured S-parameters. Transistor worked at its limit conditions, because it was used for 435 MHz and the minimum usable frequency was 400 MHz. Results of designed amplifier are shown at Figure 9 and Figure 10. Achieved efficiency was 58% and output power 27 dbm. Gain compression was significant due to class B operation and low P1dB point which was caused by border conditions of the transistor. Figure 9: Results of measured gain and output power. The topic of this article was dedicated to RF amplifiers in general and its most important functions. Basic function was explained at simulated waveforms of individual classes of RF amplifiers. As published in other articles, class F amplifiers can achieve highest efficiency however it is more complicated for design. Class E amplifiers are simpler to design and are more applicable at lower frequencies due to maximal switching frequency of shunt capacitor. Class J amplifiers are as class F amplifiers more applicable at higher frequencies and realisation is possible with microstrip lines. Second part of article is dealing with practical realisation of class F amplifier at 435 MHz. Topology used in amplifier was chosen with multiresonance circuit. Whole amplifier was designed with lumped components on FR4 substrate. Output resonators were tuned with tunable capacitors to correct adjustment of given matching conditions for odd and even harmonics. Achieved efficiency was 58% and output power was 27 dbm at 14 dbm input power. If there is requirement of better efficiency, a transistor with higher load impedance or different topology might be used. In this case of amplifier, the multiresonance output matching circuit was difficult to design and was not set to ideal conditions. In general E-pHEMT transistors have lower load impedances, for example GaN HEMT have higher output impedances so better effect of multiresonance circuit could be achieved. Voltage waveforms of this amplifier are influenced only by voltage of third harmonic, no higher harmonics are used. The efficiency could be improved by using higher harmonics. Acknowledgement Figure 10: Results of measured PAE. Research described in this paper was financed by the Czech Ministry of Education in the frame of the National Sustainability Program under grant LO1401 and by the Internal Grant Agency of Brno University of Technology project no. FEKT-S-17-4713. For research, infrastructure of the SIX Center was used. 4

Literature [1] CRIPPS, S. C. RF power amplifiers for wireless communications. 2nd ed. Boston: Artech House, 2006. ISBN 1-59693-018-7. [2] RUDIAKOVA, A., KRIZHANOVSKI, V. Advanced design techniques for RF power amplifiers. Dordrecht, The Netherlands: Springer, 2006. ISBN 1-4020-4638-3. [3] SOKAL, N. O., SOKAL, A. D, Class E-A new class of high efficiency tuned single-ended switching power amplifiers" IEEE J. Solid-State Circuits, vol. SC-IO, pp. 168-176, June 1975. [4] OZALAS, M.T. High efficiency class-f MIMIC power amplifiers at Ku-band. In: The 2005 IEEE Annual Conference Wireless and Microwave Technology, p.137-140 DOI:10.1109/WAMIC.2005.1528402. [5] BELTRAN, R. A. UHF class-e power amplifier based upon multi-harmonic approximation. In: 2013 IEEE MTT-S International Microwave Symposium Digest (MTT). IEEE, 2013, p. 1-4. DOI:10.1109/MWSYM.2013.6697389. [6] KURODA, K., ISHIKAWA R., HONJO K. Parasitic Compensation Design Technique for a C-Band GaN HEMT Class-F Amplifier. In: IEEE Transactions on Microwave Theory and Techniques. 2010, p. 2741-2750. DOI:10.1109/TMTT.2010.2077951. [7] MOON, J., KIM J., KIM B. Investigation of a Class-J Power Amplifier With a Nonlinear C OUT for Optimized Operation. In: IEEE Transactions on Microwave Theory and Techniques. 2010, p. 2800-2811 DOI:10.1109/TMTT.2010.2077970. [8] HERCEG, E. Power Amplifier for 435 MHz with High Efficiency. Brno, 2017. Diploma Thesis. Brno University of Technology. Supervisor: Ing. Tomáš Urbanec, Ph.D. 5