▶️ Usage Guide — SemiDevKit

This document explains how to run the main tools included in SemiDevKit, covering:

• 🧪 TCAD Playgrounds
• 📐 BSIM4 Analyzers
• 🧮 Paramus Physical Edition
• 🏗 OpenLane-Lite (RTL → GDSII)

Make sure that installation and setup are completed before proceeding.

🧪 1. Running TCAD Playgrounds

1.1 MOSFET / MOSCAP Playground

(tcad_playground)

This playground solves 1D Poisson and drift–diffusion models for MOS devices.

cd tcad/tcad_playground
python mosfet_vgid_1d.py
python mosfet_vdid_1d.py
python moscap_cv_1d.py

📂 Output figures are saved in:

tcad/tcad_playground/fig/

1.2 PZT P–E Loop Playground

(tcad_playground_pzt)

This playground simulates ferroelectric polarization–electric field (P–E) hysteresis.

cd tcad/tcad_playground_pzt
python pzt_pe_hysteresis_1d.py

📂 Figures are saved in:

tcad/tcad_playground_pzt/fig/

📐 2. Running BSIM4 Analyzers

2.1 DC Analyzer (VG–ID / VD–ID)

This analyzer evaluates DC characteristics using BSIM4 models and ngspice.

cd bsim/analyzer_dc
python run/run_vgid.py
python run/run_vdid.py
python plot/plot_vgid.py
python plot/plot_vdid.py

📂 Results are generated under:

bsim/analyzer_dc/results/

2.2 CV Analyzer (Cgg–Vg)

This analyzer extracts capacitance–voltage characteristics.

cd bsim/analyzer_cv
python run_cv.py
python plot_cv.py

📂 Results are stored in:

bsim/analyzer_cv/results/<node>/

2.3 DIM Analyzer (L / W Sweep)

This analyzer evaluates device scaling effects by sweeping gate length and width.

cd bsim/analyzer_dim
python run/run_vg_dim.py
python run/run_vd_dim.py
python plot/plot_vg_dim.py
python plot/plot_vd_dim.py

📂 Results are stored in:

bsim/analyzer_dim/results/<node>/

2.4 Reliability Analyzer (HCI / NBTI)

This analyzer simulates reliability degradation mechanisms.

cd bsim/analyzer_reliability
python run/run_hci_nmos.py
python run/run_nbti_pmos.py

📂 Plots and extracted degradation data are saved in:

bsim/analyzer_reliability/results/

🧮 3. Paramus Physical Edition

Paramus Physical Edition generates BSIM4 model cards directly from simplified physical parameters.

Example: Modelcard generation

cd bsim/paramus_physical
python paramus.py --node 130nm --type nmos --out nmos130.sp
python paramus.py --node 130nm --type pmos --out pmos130.sp

📂 Preset parameter files are located in:

bsim/paramus_physical/presets/

📄 Generated .sp modelcards appear in the current working directory.

🏗 4. Running OpenLane-Lite

⚠ Requirements:

• Linux or WSL2
• Docker (running)

Execute the minimal RTL → GDSII flow:

cd openlane/openlane-lite
./docker/run_in_docker.sh ./scripts/run_flow.sh

📂 Results (GDS / DEF / logs) are generated in:

openlane/openlane-lite/runs/<design-name>/

🛠 5. General Troubleshooting

❌ Python cannot find module

→ Ensure that your virtual environment is activated.

❌ ngspice not found

→ Install ngspice and verify that it is available in PATH.

❌ Docker cannot run

• 🐧 Linux: ensure your user belongs to the docker group
• 🪟 Windows: ensure Docker Desktop is running and WSL2 backend is enabled

🧭 6. Next Step

You can now:

• Proceed to 4_license.md
• Explore each module in more detail
• Modify scripts and parameters to experiment with device behavior

🎯 You are now ready to use SemiDevKit