Gianni Bosi

Behavioral Modeling of GaN FETs: A Load-Line Approach

In this paper, a new model formulation is presented that correctly accounts for low-frequency dispersion (i.e., trapping and thermal phenomena) affecting field-effect transistors (FETs). In particular, for the first time a behavioral description is applied only to the intrinsic current generator, enabling the correct measurement-based evaluation of the intrinsic device operation. The model, which is by construction technology independent, has been extensively validated considering a GaN FET. This choice is justified by the large interest around this technology and by the presence of dispersion effects that must be accurately accounted for.

A Load–Pull Characterization Technique Accounting for Harmonic Tuning

A novel methodology for the characterization of the nonlinear dynamic behavior of electron devices (EDs) is presented. It is based on a complete and accurate ED characterization that is provided by large-signal low-frequency I/V measurements, performed by means of a low-cost setup, in conjunction with any model-based description of the nonlinear reactive effects related to ED capacitances. The unique feature of the proposed technique is that a fully harmonic control of waveforms at the current generator plane is achieved, and as a consequence, high-efficiency operation can be simply investigated. Different experimental data are presented on GaAs and GaN transistors, and to definitely verify the capability of the new approach, the design of a class-F GaN power amplifier is deeply investigated as a case study.

Non-linear look-up table modeling of GaAs HEMTs for mixer application

A non-linear model suitable for mixer applications is developed for advanced 0.15 μm GaAs HEMTs. The proposed model is based on a look-up table approach. The validity of the described modeling technique is confirmed by the good agreement between model simulations and measurements.

Linear versus nonlinear de-embedding: Experimental investigation

The manuscript presents a comparison between nonlinear and linear de-embedding procedures for the identification of the I/V dynamic characteristics at the transistor current-generator plane. These approaches, without the need for modeling device trapping and thermal effects, allow to retrieving the waveforms of the electrical quantities at the current generator that governs device performance in terms of output power and efficiency. It will be demonstrated that the accuracy of the selected procedure determines the accuracy of the obtained results. Simulations and measurements carried out on a 0.25 × 600 μm2 GaN HEMT are reported as case study.

A new description of fast charge-trapping effects in GaN FETs

A nonlinear multi-bias model oriented to accurately predict the effects of charge-trapping in Gallium Nitride (GaN) HEMTs is proposed. As a case study, we considered a 0.25-μm 8×75-μm GaN HEMT. The model is identified by using CW low-frequency time-domain data and validated through high-frequency vector nonlinear measurements.

Extremely low-frequency measurements using an active bias tee

An active bias tee suitable for small- and large-signal low-frequency (5 Hz to 400 kHz) characterization of electron devices has been designed and manufactured. Different experimental results, carried out on 0.25 μm GaAs and GaN HEMTs, confirm the validity of the proposed bias circuit.

Transistor vector load-pull characterization for millimeter-wave power amplifier design

The manuscript presents a load-pull characterization technique for the design of power amplifiers in the millimeter-wave frequency band. The proposed approach is based on a recently proposed characterization technique which, by exploiting direct low-frequency nonlinear electron device measurements in conjunction with a model-based description of the device strictly dynamic nonlinearities, achieves a similar level of accuracy provided by expensive nonlinear measurement setups operating at microwave frequencies. The proposed characterization technique, validated by means of measurements carried out at 20 GHz on a 0.15-μm GaAs pHEMT device, has been exploited for the first time in order to draw load-pull contours up to 60 GHz.

A GaN power amplifier for 100 VDC bus in GPS L-band

An innovative approach in RF GaN technology suitable for Global Positioning System (GPS) L-band transmitters is presented. The solution operates directly from the 100 VDC voltage bus without the need for a DC-DC step-down converter to typical 28 V or 50 V. It is based on a mature GaN HEMT process which is radiation tolerant and reliable for space applications. The prototype single stage amplifier reported in this paper operates at 100 V bias and achieves 100 W CW output power with 19 dB gain and high efficiency > 60% at the L1 frequency band of 1575 ± 50 MHz. Preliminary results at L2 and L5 GPS bands are also reported. The demo amplifier consists of a single die with 12 mm gate periphery in a hermetic package. Higher power can easily be attained by utilizing multiple chips in one package. The proposed technology reduces SWaP figure of merit and simplifies the design of transmitters for GPS satellites.

Theoretical consideration on harmonic manipulated amplifiers based on experimental data

This contribution aims at the experimental confirmation of the advantages of the harmonic manipulation theory, using a low-frequency low-cost characterization setup. A 0.5-μm 10×100-μm (1-mm gate periphery) GaN HEMT has been characterized synthesizing at the current-generator plane different load conditions, realizing tuned-load and Class-F operation. The measurements clearly demonstrate the importance of by synthesizing the required loads at the correct reference plane, giving an experimental proof of the performance predicted by theoretical analysis.

Characterization of charge-trapping effects in GaN FETs through low-frequency measurements

In this paper, an empirical characterization technique of charge-trapping effects in GaN FETs based on low-frequency measurements is proposed. It is applied on a 0.25 × 600 μm2GaN HEMT. Experimental results confirm theoretical assumptions reported in literature.