Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.803682
Title: Contributing to second harmonic manipulated continuum mode power amplifiers and on-chip flux concentrators
Author: Poluri, Nagaditya
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The current cellular network consumes a staggering 100 TWh of energy every year. In the coming years, millions of devices will be added to the existing network to realize the Internet of Things (IoT), further increasing its power consumption. An RF power amplifier typically consumes a large proportion of the DC power in a wireless transceiver, improving its efficiency has the largest impact on the overall system. Additionally, amplifiers need to demonstrate high linearity and bandwidth to adhere to constraints imposed by wireless standards and to reduce the number of amplifiers required as an amplifier with a broader bandwidth can potentially replace several narrowband amplifiers. A typical approach to improve efficiency is to present an appropriate load at the harmonics generated by the transistor. Recently proposed continuous modes based on harmonic manipulation, such as class B/J continuum, continuous class F (CCF) and continuous class F-1 (CCF-1), have shown the capability of achieving counteracting requirements viz., high efficiency, high linearity, and broad bandwidth (with a fractional bandwidth greater than 30%). In these classes of amplifiers, the second harmonic is manipulated by placing a reactive second harmonic load and the reactive component of the fundamental load is adjusted while keeping a fixed resistive component of the fundamental load. The first contribution of this work is to investigate the reason for amplifiers designed in classes B/J continuum and CCF to achieve high efficiency at back-off and 1dB compression. In this thesis, we demonstrate that the variation of the phase of the current through the non-linear intrinsic capacitances due to the variation of the phase in the continuum of drain voltage waveforms in Class B/J/J* continuum leads to either a reduction or enhancement of intrinsic drain current. Consequently, a subset of voltage waveforms of the class B/J/J* continuum can be used to design amplifiers with higher P1dB, and efficiency at P1dB than in Class B. A simple choice of this subset is demonstrated with a 2.6GHz Class B/J/J* amplifier, achieving a P1dB of 38.1dBm and PAE at P1dB of 54.7%, the highest output power and efficiency at P1dB amongst narrowband linear amplifiers using the CGH40010 reported to date, at a comparable peak PAE of 72%. Secondly, we propose a new formulation for high-efficiency modes of power amplifiers in which both the in-phase and out-of-phase components of the second harmonic of the current are varied, in addition to the second harmonic component of the voltage. A reduction of the in-phase component of the second harmonic of current allows reduction of the phase difference between the voltage and current waveforms, thereby increasing the power factor and efficiency. Our proposed waveforms offer a continuous design space between class B/J continuum and continuous F-1 achieving an efficiency of up to 91% in theory, but over a wider set of load impedances than continuous class F-1. These waveforms require a short at third and higher harmonic impedances, which are easier to achieve at a higher frequency. The load impedances at the second harmonic are reactive and can be of any value between -j∞ and j∞, easing the amplifier design. A trade-off between linearity and efficiency exists in the newly proposed broadband design space, but we demonstrate inherent broadband capability. The fabricated narrowband amplifier using a GaN HEMT CGH40010F demonstrates 75.9% PAE and 42.2 dBm output power at 2.6 GHz, demonstrating a comparable frequency weighted efficiency for this device to that reported in the literature. IoT devices may be deployed in critical applications such as radar or 5G transceivers of an autonomous vehicle and hence need to operate free of failure. Monitoring the drain current of the RF GaN MMIC would allow to optimize the device performance and protect it from surges in its supply current. Galvanic current sensors rely on the magnetic field generated by the current as a non-invasive method of current sensing. In this thesis, our third major contribution is a planar on-chip magnetic flux concentrator, is enhance the magnetic field at the current sensor, thereby improving the current detection capability of a current sensor. Our layout utilizes a discontinuity in a magnetic via, resulting in penetration of the magnetic field into the substrate. The proposed concentrator has a magnetic gain x1.8 in comparison to air. The permeability of the magnetic core required is 500, much lower than that reported in off-chip concentrators, resulting in a significant easing of the specifications of the material properties of the core. Additionally, we explore a novel three-dimensional spiral-shaped magnetic flux concentrator. It is predicted via simulations that this geometry becomes a necessity to enhance the magnetic field for increased form factor as the magnetic field from a single planar concentrator deteriorates as its size increases.
Supervisor: DeSouza, Maria Merlyne Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.803682  DOI: Not available
Share: