逆F类放大器漏极电压波形研究

为了更清晰地了解逆F类功率放大器通过规整漏极电压和电流波形实现高效率的本质,对其漏极电压波形进行了理论分析和仿真验证。结果表明,在仅考虑二、三次谐波,且当谐波比例k=1/4时,v(θ)波形最平坦,最近似于半正弦波。利用电磁软件仿真设计并制作了一款k=0.271的逆F类功率放大器,经测试在输出功率为36.33 dBm时,功率附加效率达到了71%。     

一种高效率E-1 / F 类GaN HEMT 射频功率放大器

使用GaN HEMT 功率器件,设计了一款5G 低频段的高效率E-1 / F 类射频功率放大器。为降低晶体管寄生参数及高次谐波对逆E 类(E-1 )功放开关特性和输出性能的不良影响,将具有寄生参数补偿的逆F 类(F-1 )谐波控制网络引入逆E 类功放输出匹配电路中,实现了对二次谐波和三次谐波分别进行开路和短路处理,从而获得逆E 类功放要求的良好开关特性。同时,得益于逆F 类功放优良的谐波控制效果,改善了功放漏极电压和电流波形,大大降低其漏极峰值电压和电流,进而提升了功放的输出性能。实测结果表明,该功放在3. 3 ~ 3. 6 GHz 的300MHz 有效工作带宽内的功率附加效率为59. 1% ~ 71. 4%,最大漏极效率高达75. 6%,输出功率在40. 2 ~ 41. 5dBm之间,增益平坦度在依1dB 以内。最后利用20 MHz 带宽的单载波LTE 信号作为测试信号,基于广义记忆多项式数字预失真器对该功放进行线性化后,功放输出的邻信道功率比改善了近15 dB。     

宽带F 类功率放大器的设计

提出了一款4G 频段全覆盖高输出功率高效率功率放大器。设计采用的是Cree 公司提供的GaN HEMT 晶体管CGH40025F。基于F 类功率放大器的设计理论,通过对晶体管的输入输出端均采用谐波控制网络,并将渐变式阻抗匹配这种宽带匹配方法应用到输入输出端的基波匹配当中。在实现二次谐波阻抗匹配至低阻抗区,三次谐波阻抗匹配至高阻抗区的同时基波阻抗被匹配至50Ω附近,从而有效提高了功率放大器的输出功率、效率和带宽。最终的测试结果表明在1. 7 ~ 2. 7 GHz 频率范围内,漏极效率维持在62. 55% ~ 76%,输出功率在20 ~ 41W,增益在10 dB 以上。仿真与实测结果基本一致。     

0.4~2.2 GHz连续F类功率放大器设计

连续F类功放(PA)通过引入修正因子,减轻功放对基波和二次谐波阻抗的要求,从而扩展F类功放的带宽。本文在连续F类功放基础上,提出一种新型修正型连续F类工作模式。通过引入电阻性二次和三次谐波阻抗进一步扩展连续F类功放的设计空间,使其可实现多个倍频程带宽。基于此理论,采用负载牵引方法对基波和谐波进行最佳阻抗提取,结合实频技术设计功放匹配网络,实现了一款超宽带高效率功率放大器。该功放在0.4~2.2 GHz频段内,饱和输出功率为39.8~41.4 dBm,漏极效率在59%~79%之间,增益大于10 dB。仿真与测试结果良好,验证了该方法的正确性。     

LC并联谐振型高PAE混合型功率放大器设计与实现

针对功率放大器的输出功率和功率附加效率PAE(Power Added Efficiency)问题，提出了一种基于凹陷栅极0.25μｍ GaAs pHEMT技术的高效混合型功率放大器。首先在E类负载网络中加入LC并联谐振器，并将其调谐到开关频率的二次和三次谐波，然后通过结合E类和F类功率放大器来增加直流电源电压，从而有效增加功率放大器的输出功率和效率。此外， 较高的电源电压降低了功率放大器的工作电流，从而有效降低了功耗，对提出的混合型功率放大器进行了具体实现。测试结果显示，当频率为1.8ＧＨｚ时，该混合型功率放大器在32ｄＢｍ的输出功率下的PAE可达53％。     

基于谐波调谐的双频高效功率放大器设计

为了满足无线通信系统对低功耗双频功放的需求,分析高效功放的阻抗条件,提出了一种新型双频输出匹配电路,包括谐波控制电路和基波匹配电路两部分.首先,通过调谐晶体管的谐波阻抗减小漏极电压和漏极电流波形的重叠,从而提高功放的效率;其次,通过公式推导得出双频阻抗匹配电路参数,将晶体管在两个频率下的最佳基波阻抗匹配至50Ω.为验证...     

采用寄生补偿的高效率逆F类GaN HEMT功率放大器

为了提高无线通信系统的工作效率,提出了一种基于逆F类结构的新型高效率功率放大器。结合功率晶体管的寄生参数,设计了一种添加了寄生补偿电路的输出端谐波控制网络,对2到5次谐波阻抗进行了处理。针对电路中寄生反馈元件的存在,在输入端对2次和3次谐波阻抗也分别进行了开路和短路处理。该功率放大器选用GaN工艺的HEMT器件作为功率晶体管,当工作在940MHz频率时,经测试所获得的最大漏极效率为87.4%,最大功率附加效率为78.6%,饱和输出功率为39.8dBm。     

C波段高效率逆F类GaN MMIC功率放大器

为了解决晶体管寄生参数对逆F(F-1)类功率放大器效率的影响,采用了一种新型的输出谐波控制结构。首先,设计二次和三次谐波控制电路,同时将直流偏置电路加入二次谐波控制电路,降低了电路设计的复杂度。其次,为了解决寄生参数对F-1类功放本征漏极端阻抗的影响,采用一段串行微带线进行寄生补偿。最后,通过微带线和电容进行基波和负载之间的匹配。为验证方法的有效性,采用0.25 μm氮化镓高电子迁移率晶体管(GaN HEMT)工艺,设计了一款工作在5.7 GHz~6.3 GHz的F-1类微波集成电路功放。版图后仿真结果显示,F-1类功放的漏极效率DE为57.2%~62.3%,功率附加效率PAE为51.8%~57.4%,饱和输出功率为39.0 dBm~40.4 dBm,增益为9.0 dBm~10.4 dBm。版图面积为3.2×1.7 mm²。     

应用于4G-LTE无线通信系统的F类高PAE射频功率放大器

为了有效实现高谐波抑制并提高功率附加效率,提出了一种适用于4G-LTE无线通信系统的高效F类功率放大器。该功率放大器使用了低电压p-HEMT晶体管和小型微带抑制单元,能够在低射频输入功率下产生n次谐波抑制和较高的功率附加效率（power added efficiency,PAE）。采用谐波平衡法对提出的功率放大器进行了仿真分析,并对其进行了实际制造。通过实际测量对仿真结果进行了验证。测量结果显示,提出功率放大器的工作频率为1.8 GHz,带宽为100 MHz,平均PAE为76.9％,且具有2V的极低漏极电压。射频输入功率范围分别为0-12 dBm时,最大输出功率和增益分别为23.4和17.5 dBm。     

一种连续F类/逆F类模式转换功率放大器

传统的F类和逆F类功率放大器的带宽不宽,且对于功放输出信号的谐波控制比较严格。在连续类功放理论的基础上,设计了一款在工作带宽内连续F类和连续逆F类模式转换的功率放大器。设计的功放采用了Cree 公司的CGH40010F GaN HEMT 晶体管。通过调整功放管输出端的谐波控制网络,控制谐波阻抗在Smith 圆图中位置分布,从而在带宽内同时实现连续F类和连续逆F类的工作模式。制作了测试板,结果表明在2.4～4.2GHz的带宽内,增益在11dB 以上,漏极效率为55%～82%,输出功率在39.5～41.9dBm。采用了10MHz 的LTE 单载波信号进行功放的数字预失真测试,功放的输出ACPR改善了6dB以上。     

Class-F power amplifiers with maximally flat waveforms

Class-F power amplifiers (PA's) employ harmonic-frequency resonators to shape their drain or collector waveforms to improve efficiency. Generally, the output network must present the drain with either an open or short circuit at the harmonic frequencies. At VHF and higher frequencies, the drain capacitance, lead inductance, lead length, and dispersion make implementation of reasonably ideal tuned circuits difficult. However it is possible to control the impedances at a finite number of harmonics. This note first derives the basic relationships among the Fourier coefficients of the waveforms and the performance of the amplifier. Fourier coefficients for maximally flat waveforms are then derived for inclusion of up to the fifth harmonic. Amplifier performance is then tabulated as a function of which harmonics are included in the voltage and current waveforms. Efficiency increases from 50% of class A toward 100% as harmonics are added. Power-output capability increases by up to 27%     

TRAPATT oscillations in a p-i-n avalanche diode

The characteristics of TRAPATT oscillations in a p-in diode are discussed and an approximate semi-analytical solution for the diode voltage waveform is derived when the diode current is a square wave. It is shown that a traveling avalanche zone is not necessary to generate a dense "trapped" plasma and that the boundary conditions prevent the trapped plasma from completely filling the depletion layer. Typical voltage waveforms and corresponding diode power, efficiency, and impedance at the fundamental and higher harmonics are presented. When the diode current is a square wave the diode does not necessarily exhibit a negative resistance at all higher harmonics. A computer program for TRAPATT oscillations in a p-i-n diode is described. Its running time is two or three orders of magnitude less than more exact time domain computer analyses. Typical results of diode power, dc to RF conversion efficiency, and required circuit impedances are presented for several different current waveforms which are composed of up to the seventh harmonic of a square wave and the first two harmonics of a half-wave sine wave. It is shown that high-efficiency oscillations are possible with diode currents composed of only the fundamental and one harmonic.     

Analysis of a Fully Matched Saturated Doherty Amplifier With Excellent Efficiency

A saturated Doherty amplifier based on class-F amplifiers is analyzed in terms of its load modulation behavior, efficiency, and linearity. Simulations included the amplitude ratio and phase difference between the fundamental and third harmonic voltages, the current/voltage waveforms, load lines, and the third-order intermodulation amplitudes/phases of the carrier and peaking amplifiers. The saturated doherty power amplifier was implemented using two Eudyna EGN010MK GaN HEMTs with a 10-W peak envelope power. For a 2.14-GHz forward-fink wideband code-division multiple-access signal, the doherty amplifier delivers an excellent efficiency of 52.4% with an acceptable linearity of -28.3 dBc at an average output power of 36 dBm. Moreover, the amplifier can provide the high linearity performance of -50 dBc using the digital feedback predistortion technique.     

Exact analysis of class E tuned power amplifier with only one inductor and one capacitor in load network

Presents an exact analysis of that amplifier, an accurate and simple design procedure, and experimental results which confirm the theoretical analysis. The following performance parameters are determined for optimum operation with any switch duty ratio: the current and voltage waveforms, the peak values of collector current and collector-to-emitter voltage, the output power, the power-output capability, and the component values of the load network. The output harmonic spectrum is given for several values of switch duty ratio. The transistor power losses and collector efficiency are also estimated. The measured collector efficiency was over 95 percent with 3 W output power at 1 MHz, using a BSX60 TO-39 transistor.     

Class-F power amplifiers with reduced conduction angles

This paper determines the characteristics of class-F power amplifiers (PAs) with “class-C” conduction angles (current flow for less than half of the RF cycle). Analyses are conducted for both clipped-sinusoidal (classical class-C) and rectangular current waveforms. It is not possible to define the voltage and current waveforms arbitrarily if the production or sinking of power at the harmonic frequencies is to be avoided. As a result, only a few cases are of practical interest. In these cases, the conduction angle is set to null a specific harmonic in the current. The nulled harmonic is then used to flatten the bottom of the voltage waveform, resulting in increased efficiency and increased power output     

C-band GaAs FET power amplifiers with 70-W output power and 50% PAE for satellite communication use

This paper describes a successfully developed high-power and high-efficiency C-band GaAs FET amplifier for satellite communication systems. To realize high efficiency in a high-power amplifier, an HFET well-designed for high-power applications is developed, and precise design for an amplifier is carried out. The HFET employed achieves reduction in a gate leakage current while maintaining a high maximum drain current. For precise design of an amplifier, large-signal FET model parameters are extracted using pulsed I-V and S-parameter measurements. Based on this model, second harmonic impedances as well as fundamental impedances are determined for obtaining high efficiency and input- and output-matching circuits which are assembled in a compact package are designed to achieve a high-efficiency internally matched amplifier. As a result, the amplifier delivers a high saturated output power of 70 W and a high power-added efficiency of 51%. These characteristics are the record power performance in C-band in terms of simultaneous achievement of high power and high efficiency. A low third order intermodulation distortion of -35 dBc is also obtained at a drain voltage of 10 V.     

High-Efficiency Power Amplifier Using Novel Dynamic Bias Switching

A novel bias-switching scheme for a high-efficiency power amplifier is proposed. Two voltage levels for the drain bias of the RF power amplifier are generated using a combination of a class E dc/ac inverter and a class E rectifier with offset voltage. When signal peaks occur, the output of the class E dc/ac inverter is rectified and the rectified dc is added to the offset voltage by the class E rectifier, which boosts the drain bias of the RF power amplifier. Except during peaks, the drain bias of the RF power amplifier is connected to the offset voltage directly. Since the efficiency when there are no peaks is very high due to the direct connection between the offset voltage and drain bias, the overall efficiency of the RF power amplifier can be improved dramatically in high peak-to-average power ratio (PAPR) systems. The measured results show that the drain bias of the RF power amplifier is boosted up to approximately 1.8 times the offset voltage when the RF peaks generate. The overall efficiency of the proposed bias-switching amplifier is improved by 62% compared to that of the fixed bias amplifier in high PAPR systems     

Electrical and Temperature Stress Effects on Class-AB Power Amplifier Performances

Normalized degradations of drain efficiency and output power as a function of conduction angle, maximum drain current, and maximum output voltage are modeled. Good agreement between the model predictions and Cadence radio-frequency simulation results is obtained. The output power and efficiency of class-AB power amplifiers (PAs) degrade with channel hot electron stress due to reduced conduction angle, drain current, and output voltage. The degradation is enhanced at high temperature. In addition, the PA third-order input intercept point and adjacent channel power ratio all decrease with stress     

Class-E, Class-C, and Class-F power amplifiers based upon a finitenumber of harmonics

Class-E operation at UHF and microwave frequencies is achieved by using transmission-line networks to provide the drain harmonic impedances of an ideal class-E power amplifier (PA). This paper develops a technique for analysis of such amplifiers that are based upon a finite number of harmonics. The technique is generally applicable to classes E, C, and F as well as PAs with harmonic reactances not corresponding to those of established classes. The analysis shows that the maximum achievable efficiency depends not upon the class of operation, but upon the number of harmonics employed. For any set of harmonic reactances, the same maximum efficiency can be achieved by proper adjustment of the waveforms and the fundamental-frequency load reactance. The power-output capability depends upon the harmonic reactances and is maximum for class F