Academia.eduAcademia.edu

Figure 1 – uploaded by R Torres

See full PDFdownloadDownload figure
Fig. 2. Layout for a four-finger cell MOSFET showing a distributed model for the source resistance. Fig. 1. Microwave MOSFET layout for one of the considered devices.

Figure 2 Layout for a four-finger cell MOSFET showing a distributed model for the source resistance. Fig. 1. Microwave MOSFET layout for one of the considered devices.

Related Figures (16)

Fig. 3. Equivalent circuit for the two fingers of the right end of Fig. 2 consid- ering the device symmetry.
Fig. 3. Equivalent circuit for the two fingers of the right end of Fig. 2 consid- ering the device symmetry.
Fig. 4. Jq as a function of V,. for different Ny with a fixed total gate area. difference between manufacturing processes. In addition, above a given N¢, high frequency and dc measurements are almost in- dependent of this parameter due to the small change of the total number of vias and the total source/drain area.
Fig. 4. Jq as a function of V,. for different Ny with a fixed total gate area. difference between manufacturing processes. In addition, above a given N¢, high frequency and dc measurements are almost in- dependent of this parameter due to the small change of the total number of vias and the total source/drain area.
for the case of the source terminal, as shown in Fig. 5. The pa- rameters A and B are given by However, these equations are not practical to fit the exper- imental data. In order to simplify them, the assumption that Rynet_s 18 much smaller than R.oy4_, can be considered. Here, Reont_s consists of the resistance of the metal-to-semiconductor interface. Thus, the term R,,+_, can be neglected when its ex- ponent is greater than 2.
for the case of the source terminal, as shown in Fig. 5. The pa- rameters A and B are given by However, these equations are not practical to fit the exper- imental data. In order to simplify them, the assumption that Rynet_s 18 much smaller than R.oy4_, can be considered. Here, Reont_s consists of the resistance of the metal-to-semiconductor interface. Thus, the term R,,+_, can be neglected when its ex- ponent is greater than 2.
Fig. 5. Equivalent circuit for the source external resistance. This can be rewritten for several cells as follows:
Fig. 5. Equivalent circuit for the source external resistance. This can be rewritten for several cells as follows:
A and B can then be rewritten as where N, = (GWeotat)/Ny Ne, @ is the number of vias per unit length, which is a constant, and N,, > 3. In this work, @ is equal to two vias per | ym. In this way, R, and Ry resistances can be determined by
A and B can then be rewritten as where N, = (GWeotat)/Ny Ne, @ is the number of vias per unit length, which is a constant, and N,, > 3. In this work, @ is equal to two vias per | ym. In this way, R, and Ry resistances can be determined by
Fig. 7. Micrograph of one of the considered transistors. Fig. 6. Experimental setup used to perform RF measurements under different bias conditions, illustrating the SDA, VNA, and device-under-test (DUT). (a) Photograph. (b) Schematic.
Fig. 7. Micrograph of one of the considered transistors. Fig. 6. Experimental setup used to perform RF measurements under different bias conditions, illustrating the SDA, VNA, and device-under-test (DUT). (a) Photograph. (b) Schematic.
Fig. 8. DUT, indicating the ground—signal—ground probes.
Fig. 8. DUT, indicating the ground—signal—ground probes.
Fig. 10. Fit of the drain resistance against total number of fingers with different Vas at Vas =OV. Fig. 9. Fit of the source resistance against total number of fingers with different Vas at Vag = OV.
Fig. 10. Fit of the drain resistance against total number of fingers with different Vas at Vas =OV. Fig. 9. Fit of the source resistance against total number of fingers with different Vas at Vag = OV.
Fig. 11. Simulated and experimental $,,; of microwave MOSFETs for different number of fingers at V,, = 0.58 V and Vg, = 0 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained average values of R,, Ra, and Ry.
Fig. 11. Simulated and experimental $,,; of microwave MOSFETs for different number of fingers at V,, = 0.58 V and Vg, = 0 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained average values of R,, Ra, and Ry.
Fig. 14. Simulated and experimental $,; and 1.3.22 of microwave MOSFETs for different number of fingers at V,, = 0.50 V and Va, = 0.7 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained average values of R,, Ra, and R,. Fig. 13. Simulated and experimental phase of microwave MOSFETs for dif- ferent number of fingers at V,, = 0.58 V and V4, = 0 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained av- erage values of R,, Ra, and R,.
Fig. 14. Simulated and experimental $,; and 1.3.22 of microwave MOSFETs for different number of fingers at V,, = 0.50 V and Va, = 0.7 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained average values of R,, Ra, and R,. Fig. 13. Simulated and experimental phase of microwave MOSFETs for dif- ferent number of fingers at V,, = 0.58 V and V4, = 0 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained av- erage values of R,, Ra, and R,.
Fig. 12. Simulated and experimental S22 of microwave MOSFETs for dif- ferent number of fingers at V,, = 0.58 V and V4, = 0 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained av- erage values of R,, Ra, and R,.
Fig. 12. Simulated and experimental S22 of microwave MOSFETs for dif- ferent number of fingers at V,, = 0.58 V and V4, = 0 V. The curve with diamonds corresponds to the simulation of the MOSFET with the obtained av- erage values of R,, Ra, and R,.
Related topics:
Electronic EngineeringMaterials ScienceMOSFET compact modelingElectrical and Electronic Engineering
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved