Kai Chappell
b826337b36
Replace module-level singleton with @st.cache_resource decorator. This properly survives Streamlit reruns without losing the server reference, preventing "port already in use" errors when refreshing the browser in Docker. The cache is tied to the Streamlit process lifecycle, so when the process restarts, both the cache and daemon threads are cleared together. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
py_dvt_ate
Coupled Physics DVT Simulation Platform
A software simulation environment for offline development of ATE (Automated Test Equipment) characterisation algorithms. Accurately models thermal-electrical coupling, enabling DVT engineers to develop and validate test sequences without physical laboratory access.
Overview
py_dvt_ate simulates a complete DVT test bench:
- Thermal Chamber - Temperature control with realistic ramp and settling behaviour
- Programmable Power Supply - Voltage/current control and measurement
- Digital Multimeter - DC voltage measurement with configurable integration time
- DUT Models - Device Under Test simulation with thermal-electrical coupling (e.g., LDO voltage regulators)
Key Features
- Coupled Physics Simulation - DUT self-heating affects electrical parameters realistically
- SCPI Protocol - Industry-standard commands for instrument control
- Hardware Abstraction - Same test code works with simulated or real instruments
- Multiple Interfaces - CLI, programmatic API, and Streamlit dashboard
- Data Persistence - SQLite for metadata, Parquet for time-series measurements
Documentation
| Document | Purpose |
|---|---|
| Requirements | Defines what the system must do |
| Technical Specification | Specifies how to implement the system |
| Architecture Decisions | Explains why decisions were made |
Installation
# Install with development dependencies
pip install -e ".[dev]"
Quick Start
Interactive Dashboard
Launch the Streamlit dashboard to visualise the physics simulation and run tests:
py-dvt-ate dashboard
This opens a browser window with:
- Live Simulation - Real-time temperature/voltage charts with physics coupling
- Test Execution - Run TempCo characterisation tests
- Results Viewer - Browse and analyse historical test results
CLI Commands
# Start the simulation server (TCP ports for SCPI instruments)
py-dvt-ate serve
# List available tests
py-dvt-ate tests list
# Run a TempCo test
py-dvt-ate tests run tempco --config config/tempco_test.yaml
Programmatic API
from py_dvt_ate.instruments import InstrumentFactory
from py_dvt_ate.simulation import SimulationServer
# Start simulation server
server = SimulationServer()
server.start()
# Create instruments via HAL
factory = InstrumentFactory()
instruments = factory.create_from_config("config/default.yaml")
# Control instruments using standard interfaces
instruments.chamber.set_temperature(85.0)
instruments.psu.set_voltage(1, 5.0)
instruments.psu.enable_output(1, True)
voltage = instruments.dmm.measure_dc_voltage()
print(f"Output voltage: {voltage:.4f} V")
Project Status
Status: MVP Complete (v0.1.0)
The core vertical slice is functional:
- Physics engine with thermal-electrical coupling
- Virtual instruments (chamber, PSU, DMM)
- Hardware Abstraction Layer
- SCPI-over-TCP server
- Test framework with TempCo test
- Streamlit dashboard
- SQLite/Parquet data persistence
See the requirements document for the full scope and future phases.
Technology Stack
- Language: Python 3.11+
- Physics: NumPy, SciPy
- Configuration: Pydantic, YAML
- CLI: Typer
- Dashboard: Streamlit
- Data: SQLite, PyArrow (Parquet)
Author
Kai Chappell
Licence
Proprietary - All rights reserved. See LICENSE for details.