University of Messina Uses NI AWR Design Tools to Model Microwave Amps


Reading time ( words)

The University of Messina, in Messina, Italy, has always been characterized by its high quality of research and teaching. It is committed to European higher education so as to foster mobility, increase employment, and attract students and academics from within as well as outside Europe. The University of Messina participates in partnerships with foreign universities, promotes international research projects, and supports European and international exchange programs.

The Design Challenge

The University of Messina Microwave Electronics Lab (EleMIC) uses NI AWR Design Environment software for research and all microwave electronics education levels (B.S., M.S., Ph.D.). The challenge for Professor Alina Caddemi is to not only introduce her students to the secrets of microwave circuit analysis and design, but also to make it a truly exciting experience.

Of particular interest to Prof. Caddemi is device modeling. In this domain, she and her group have 25 years of experience and more than 200 international papers published. A tough benchmark has been the design of single-stage, high-power microwave amplifier using complex device models for nonlinear analysis of the device behavior as well as optimization. The group recently used Microwave Office circuit design software to tune the behavior of a gallium arsenide (GaAs) high-electron mobility transistor (HEMT) wafer previously characterized in the microwave range (2-18 GHz) under dark and laser-light illuminated conditions. A key challenge was the noise analysis and modeling aspect that was markedly changed by the light exposure. 

The Solution

NI AWR Design Environment fulfills the EleMIC’s requirements for having very agile software tools to accomplish the fine tuning required for model refinement, complete and versatile nonlinear analysis tools, and easy management of layout generation for EM simulation. For the work presented in the IEEE conference paper, the software was used to analyze in detail the performance changes of a GaAs HEMT-based low-noise amplifier under visible (650 nm) laser exposure. The device employed was first characterized in dark and illuminated conditions over the 2-18 GHz frequency range in terms of DC, scattering and noise parameters. Subsequently, it was modeled by means of a linear circuit model with associated noise sources in both conditions. By employing the experimental results, a low-noise amplifier was then designed and optimized for dark-mode operation in the 7.5-8.5 GHz frequency range. Finally, the performance of the LNA under light exposure was evaluated. 

On the basis of device performance variations under VIS light exposure, either a severe degradation of the noise figure, a slight increase of the amplifier gain, or a moderate variation of the input/output matching was expected. Instead, the obtained results exhibited a moderate degradation of the overall LNA performance due to device illumination, except for the input matching level which showed to be improved. NI AWR Design Environment was then used to analyze the correlation between the device model variations and the amplifier behavior to get a deeper insight into the observed trends.

On the basis of measured results, a low-noise amplifier was tailored on the device performance in dark conditions. The design was carried out Microwave Office, with performance goals centered in the 7.5 to 8.5 GHz frequency band. Alumina (εr = 9.9, tg δ = 0.0002, h = 250 μm) was adopted for the substrate with 5 µm-thick gold metallization. The amplifier gain is above 10 dB, while the noise figure is 1.4 ± 0.15 dB over the entire band. Input and output matching levels stay higher than 10 dB within the chosen frequency band. The stability was monitored inside and all over outside the operating band by the value of coefficient. Modeling of the bonding wires has been accomplished by means of the resident multilevel bond wire model BWIRES2 that define bonding scheme parameters in compliance with EIA/JEDEC Standard No. 59. 

Due to the required stabilization of the amplifier, we noticed that the reflection coefficient “seen” on the left side of the device input was not really close to Γopt within the operating band, i.e., broad-band stabilization was achieved at the expense of the noise mismatch.

Why NI AWR Design Environment

Prof. Caddemi chose NI AWR Design Environment software because it has a very user friendly interface, is easy to use, has a very good element help function that is highly beneficial for educational purposes, has powerful optimization capabilities and achieves very fast and accurate EM analysis. Productivity, ease of use, simulation speed, availability of models, and technical support were cited as the most helpful features. For Prof. Caddemi, the most valuable aspect of the software is that it is a powerful tool suite for expert engineers and novice students. 

For more information, visit www.ni.com.

 

Share


Suggested Items

Simon Fried: Additive Manufacturing Through Printed Electronics

10/17/2018 | Barry Matties, Publisher, I-Connect007
Simon Fried, president of Nano Dimension, discusses how the company has taken the additive manufacturing process to the next level through printed electronics. He also shares his thoughts on the growing demand for 3D circuits, as well as how this could potentially be a game-changer for PCB designers.

Cadence: Bullish on AI

10/15/2018 | Andy Shaughnessy, Design007 Magazine
David White has been involved with artificial intelligence research for almost 30 years. Now, David is the senior group director of R&D for Cadence Design Systems, and I knew we’d have to speak with him for this issue on AI. In a recent interview, we discussed his decades of work in AI, Cadence’s research into AI and machine learning, and what he believes AI could mean for the EDA tools of the future.

Artificial Intelligence: The Future of EDA?

10/10/2018 | Andy Shaughnessy, Design007 Magazine
Artificial intelligence (AI) has been making inroads into a variety of industries in the past decade or so, from automobiles to medical devices. Naturally, EDA tool companies are taking a look at AI. Does AI offer a way forward for PCB design tool developers? I recently interviewed Paul Musto, director of marketing for the Board Systems Division of Mentor. We discussed Mentor’s plans for integrating AI into EDA tools, and why we may be at the very beginning of understanding the pros and cons of this new technology.



Copyright © 2018 I-Connect007. All rights reserved.