Academia.eduAcademia.edu

Figure 6 – uploaded by Daniel Pardo

See full PDFdownloadDownload figure
Fig. 6. (a) Minimum noise figure (b) associated gain, (c) noise resistance, (d) magnitude and (e) phase of T,,, at 19 and 94 GHz as a function of the drain current for an extrinsic bias Vz, = 0.8 V. The values are obtained experimentally (solid lines) and with the MC simulation (dotted line, black circles).

Figure 6 (a) Minimum noise figure (b) associated gain, (c) noise resistance, (d) magnitude and (e) phase of T,,, at 19 and 94 GHz as a function of the drain current for an extrinsic bias Vz, = 0.8 V. The values are obtained experimentally (solid lines) and with the MC simulation (dotted line, black circles).

Related Figures (9)

Fig. 1. Equivalent circuit of the devices. The shaded area represents the intrinsic elements that are obtained by the MC simulation, while the dotted box encloses the “intrinsic” equivalent circuit from the point of view of experimental measurements (just excluding the contact parasitics).
Fig. 1. Equivalent circuit of the devices. The shaded area represents the intrinsic elements that are obtained by the MC simulation, while the dotted box encloses the “intrinsic” equivalent circuit from the point of view of experimental measurements (just excluding the contact parasitics).
Fig. 2. (a) Scheme of the noise tuning configuration and (b) its equivalent as seen from the output.
Fig. 2. (a) Scheme of the noise tuning configuration and (b) its equivalent as seen from the output.
Fig. 3. Comparison of the 7z-V.;, characteristics (also the extrinsic g,,, and gq are shown in the inset for V;, — 0.8 V) measured in a real HEMT (solid line) with those obtained from the MC simulations (dotted lines, black circles).
Fig. 3. Comparison of the 7z-V.;, characteristics (also the extrinsic g,,, and gq are shown in the inset for V;, — 0.8 V) measured in a real HEMT (solid line) with those obtained from the MC simulations (dotted lines, black circles).
Fig. 4. Comparison of the small-signal parameters measured experimentally (solid line) with those calculated with the MC simulation (dotted lines, black circles) for Vz, = 0.8 V: (a) gate-source, drain-source and gate-drain capacitances, (b) transconductance and drain conductance, and (c) intrinsic current gain cutoff frequency.
Fig. 4. Comparison of the small-signal parameters measured experimentally (solid line) with those calculated with the MC simulation (dotted lines, black circles) for Vz, = 0.8 V: (a) gate-source, drain-source and gate-drain capacitances, (b) transconductance and drain conductance, and (c) intrinsic current gain cutoff frequency.
Fig. 5. P,R, and C parameters as a function of the drain current for Vz, = 0.8 V. The inset represents their frequency dependence for J, = 60 A/m.
Fig. 5. P,R, and C parameters as a function of the drain current for Vz, = 0.8 V. The inset represents their frequency dependence for J, = 60 A/m.
Fig. 7. Frequency dependence of (a) Fynin and (b) G,;, obtained with the MC simulation for V,, = 0.8 V and Jz = 60 A/m (the point where the absolute minimum noise figure is found) for devices with different widths. Also shown are the intrinsic results (dash-dot line) and those for a device with W = 100 jm and without T-gate, modeled by a higher gate resistance, R, = 25 (dash- dot- dot line).
Fig. 7. Frequency dependence of (a) Fynin and (b) G,;, obtained with the MC simulation for V,, = 0.8 V and Jz = 60 A/m (the point where the absolute minimum noise figure is found) for devices with different widths. Also shown are the intrinsic results (dash-dot line) and those for a device with W = 100 jm and without T-gate, modeled by a higher gate resistance, R, = 25 (dash- dot- dot line).
Fig. 8. (a) Funin and (b) G,,, as a function of W obtained with the MC simulation for the same bias point of Fig. 7 and for different frequencies: 50 GHz (circles), 100 GHz (triangles) and 200 GHz (squares). The shaded area in (a) represents the range of W for which there is no problem of adaptation at the different frequencies.
Fig. 8. (a) Funin and (b) G,,, as a function of W obtained with the MC simulation for the same bias point of Fig. 7 and for different frequencies: 50 GHz (circles), 100 GHz (triangles) and 200 GHz (squares). The shaded area in (a) represents the range of W for which there is no problem of adaptation at the different frequencies.
Fig.9. (a) Finin and (b) G',,, as a function of the number of fingers of the gate obtained with the MC simulation for the same bias point of Figs. 7 and 8 and for different frequencies and total device widths: 50 GHz and W = 100 pm (circles), 100 GHz and W = 40 jrm (triangles) and 200 GHz and W = 20 pm (squares).
Fig.9. (a) Finin and (b) G',,, as a function of the number of fingers of the gate obtained with the MC simulation for the same bias point of Figs. 7 and 8 and for different frequencies and total device widths: 50 GHz and W = 100 pm (circles), 100 GHz and W = 40 jrm (triangles) and 200 GHz and W = 20 pm (squares).
electronic noise. physics from the University of Salamanca in 1993 and 1997, respectively. Since 1993, he has been with the Electronics Group, Department of Applied Physics, University of Salamanca, as a Grant Holder In 1996, he became Assistant Professor. He worked one year at the Institut d’Electronique et de Microélectronique du Nord, Villeneuve d’Ascq, France. In 1999, he became Temporary Associate Professor. His present research interest is the modeling of submicron heterojunction devices, focusing on the study of electronic noise. Daniel Pardo was born in Valladolid, Spain, in 1946. He graduated in physics in 1971 from the University of Valladolid, where he also received the Ph.D. de- gree in 1975, He graduated in physics in 1971 from the University of Valladolid, where he also received the Ph.D. de- gree in 1975. From 1971 to 1981, he was with the Electronics Department of the University of Valladolid working on the characterization of semiconductor materials and modeling of semiconductor devices, where he was promoted to Associate Professor in 1978. In 1981, he joined the Applied Physics Department, University of Salamanca, where he has been Full Professor since 1983 and Head of the Electronics Group. His current research interest is the Monte Carlo simulation of semiconductor devices with special application to noise characterization.
electronic noise. physics from the University of Salamanca in 1993 and 1997, respectively. Since 1993, he has been with the Electronics Group, Department of Applied Physics, University of Salamanca, as a Grant Holder In 1996, he became Assistant Professor. He worked one year at the Institut d’Electronique et de Microélectronique du Nord, Villeneuve d’Ascq, France. In 1999, he became Temporary Associate Professor. His present research interest is the modeling of submicron heterojunction devices, focusing on the study of electronic noise. Daniel Pardo was born in Valladolid, Spain, in 1946. He graduated in physics in 1971 from the University of Valladolid, where he also received the Ph.D. de- gree in 1975, He graduated in physics in 1971 from the University of Valladolid, where he also received the Ph.D. de- gree in 1975. From 1971 to 1981, he was with the Electronics Department of the University of Valladolid working on the characterization of semiconductor materials and modeling of semiconductor devices, where he was promoted to Associate Professor in 1978. In 1981, he joined the Applied Physics Department, University of Salamanca, where he has been Full Professor since 1983 and Head of the Electronics Group. His current research interest is the Monte Carlo simulation of semiconductor devices with special application to noise characterization.
Related topics:
Contact ResistanceNoise FigureExperimental MeasurementNoise MeasurementEquivalent CircuitsElectrical and Electronic Engineering
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved