Power Transformer Designer

A full-stack web application for designing power transformers, inductors, and pulse transformers using industry-standard methodologies including McLyman's Area Product (Ap) and Core Geometry (Kg) methods.

🎯 Features

Design Methods

McLyman's Ap Method: Classic area product approach for initial transformer sizing
McLyman's Kg Method: Core geometry method optimized for high-power designs
Erickson's Kgfe Method: Loss-optimized design for high-frequency applications
Pulse Transformer Design: Specialized design for capacitor-discharge pulse applications
Inductor Design: Energy storage method with air gap calculations

Analysis Capabilities

Loss Analysis: Core losses (Steinmetz equation) and copper losses with AC effects
Thermal Analysis: Temperature rise prediction and thermal management
Cross-Validation: Multi-method confidence scoring for design verification
Winding Design: Optimal conductor sizing, turns calculation, and layer configuration

Database Integration

Local Core Database: 100+ ferrite and silicon steel cores with detailed specifications
OpenMagnetics Integration: Access to 10,000+ cores from major manufacturers (TDK, Ferroxcube, Ferrіte, etc.)
Material Database: Comprehensive material properties and Steinmetz coefficients
Loss-Based Core Selection: Intelligent core selection optimized for minimum losses

Export Capabilities

MAS JSON: OpenMagnetics-compatible format for FEA tools
FEMM Lua Scripts: Direct export for Finite Element Method Magnetics simulation
Design Archives: Complete design data in JSON format for documentation

🚀 Quick Start

Prerequisites

Python 3.11+ (managed via uv)
Node.js 18+ and npm
Git

Installation

Clone the repository

git clone <repository-url>
cd transformer_designer

Docker Deployment (Recommended) Build and run the entire stack using Docker Compose:
```
docker-compose up --build
```
- Frontend: http://localhost:3000
- Backend API: http://localhost:8000
- API Docs: http://localhost:8000/docs

Manual Backend Setup

cd backend
uv sync
uv run uvicorn main:app --host 127.0.0.1 --port 8000 --reload

Manual Frontend Setup
```
cd frontend
npm install
npm run dev
```

📁 Project Structure

transformer_designer/
├── backend/                    # FastAPI backend
│   ├── main.py                # Application entry point
│   ├── calculations/          # Core design algorithms
│   │   ├── ap_method.py      # McLyman's Ap method
│   │   ├── kg_method.py      # McLyman's Kg method
│   │   ├── erickson_method.py # Erickson's Kgfe method
│   │   ├── pulse_transformer.py # Pulse transformer design
│   │   ├── winding.py        # Winding calculations
│   │   ├── losses.py         # Loss analysis
│   │   ├── thermal.py        # Thermal analysis
│   │   └── cross_validation.py # Design validation
│   ├── models/               # Pydantic data models
│   ├── routers/              # API endpoints
│   ├── integrations/         # External integrations
│   │   ├── openmagnetics.py # OpenMagnetics API
│   │   ├── mas_exporter.py  # MAS format export
│   │   └── silicon_steel.py # Silicon steel cores
│   ├── data/                # Core and material databases
│   └── tests/               # Comprehensive test suite
├── frontend/                # Nuxt 3 frontend
│   ├── pages/              # Application pages
│   │   ├── index.vue      # Home page
│   │   └── design/        # Design interfaces
│   ├── components/        # Vue components
│   ├── composables/       # Composable functions
│   └── assets/           # Styles and static assets
├── pdfs/                  # Reference documentation
└── README.md             # This file

🛠️ Usage

Transformer Design

Navigate to Design → Transformer (/design/transformer)
Enter your requirements:
- Power: Output power (W)
- Frequency: Switching frequency (Hz)
- Voltages: Primary and secondary voltages
- Topology: Forward, flyback, push-pull, etc.
- Duty Cycle: Operating duty cycle
- Waveform: Square, sine, trapezoidal
View results:
- Selected core with specifications
- Winding configuration (turns, wire gauge, layers)
- Loss breakdown (core + copper losses)
- Temperature rise estimate
- Alternative core suggestions with OpenMagnetics integration

Inductor Design

Navigate to Design → Inductor (/design/inductor)
Specify parameters:
- Inductance: Required inductance value (H)
- Current: DC and peak current (A)
- Ripple: Current ripple percentage
- Frequency: Operating frequency (Hz)
Results include:
- Core selection with air gap
- Number of turns
- Wire specifications
- Saturation margin

Pulse Transformer Design

Navigate to Design → Pulse (/design/pulse)
Enter pulse specifications:
- Source Capacitor: Capacitance and voltage
- Pulse Width: Duration of pulse
- Repetition Rate: Pulse frequency
- Isolation: Required voltage isolation
- Load Impedance: Output load characteristics
Get specialized pulse design with:
- Core sized for volt-second product
- Winding configuration for low-leakage
- Isolation considerations
- Energy transfer efficiency

🔌 API Endpoints

Design Endpoints

POST /api/design/transformer - Design a power transformer
POST /api/design/inductor - Design an inductor
POST /api/design/pulse - Design a pulse transformer

Database Endpoints

GET /api/cores - List all available cores
GET /api/materials - List material properties
GET /api/openmagnetics/cores - Search OpenMagnetics database
GET /api/openmagnetics/cores/{shape} - Get cores by shape

Export Endpoints

POST /api/export/mas - Export to MAS JSON format
POST /api/export/femm - Export to FEMM Lua script
POST /api/export/json/download - Download design as JSON

Utility Endpoints

GET /health - Health check
GET / - API information and endpoints

🧪 Testing

Run the comprehensive test suite:

cd backend
uv run pytest

Test coverage includes:

Design method algorithms
Loss calculations
Thermal analysis
Winding calculations
API endpoint validation
Integration tests

🎓 Design Methods Explained

McLyman's Ap Method

The Area Product method is ideal for initial transformer sizing:

Ap = Wa × Ac (window area × core area)
Fast, reliable approximation
Best for: Initial estimates, simple designs

McLyman's Kg Method

Core Geometry method for optimized designs:

Kg = Ap² / MLT (area product squared / mean length per turn)
Better core utilization for high power
Best for: Power transformers, optimized designs

Erickson's Kgfe Method

Loss-optimized design for high efficiency:

Minimizes total losses (core + copper)
Accounts for frequency-dependent effects
Best for: High-frequency converters, efficiency-critical applications

📊 Export Formats

MAS JSON (OpenMagnetics)

Standard format compatible with OpenMagnetics tools and FEA software:

Complete magnetic model
Winding specifications
Material properties
Geometry definitions

FEMM Lua Scripts

Direct export for electromagnetic simulation:

Automated geometry creation
Material assignments
Excitation setup
Boundary conditions

🔧 Development

Backend Development

cd backend
uv run uvicorn main:app --reload  # Auto-reload on changes

Frontend Development

cd frontend
npm run dev  # Hot module replacement enabled

Adding New Cores

Edit backend/data/cores.json following the existing format:

{
  "name": "ETD39",
  "Ae": 125e-6,
  "Aw": 124e-6,
  "Ve": 16000e-9,
  "MLT": 81.5e-3,
  ...
}

Adding New Materials

Edit backend/data/materials.json with Steinmetz coefficients:

{
  "name": "3F3",
  "Kc": 48.4,
  "alpha": 1.32,
  "beta": 2.6,
  ...
}

📚 References

This tool implements methodologies from:

Colonel Wm. T. McLyman - Transformer and Inductor Design Handbook (4th Edition, 2011)
- Area Product (Ap) method
- Core Geometry (Kg) method
- Winding design principles
Robert W. Erickson - Fundamentals of Power Electronics
- Kgfe loss-optimized design
- High-frequency effects
- Converter topologies
OpenMagnetics Standard - MAS format specification
- Industry-standard magnetic model format
- FEA tool compatibility

🤝 Contributing

Contributions are welcome! Key areas for improvement:

Additional core geometries and manufacturers
Enhanced thermal modeling
More converter topologies
UI/UX enhancements
Test coverage expansion

📝 Documentation

User Guide: user_guide.md - Detailed usage instructions
Backend README: backend/README.md - Backend-specific details
API Documentation: http://localhost:8000/docs (when running)
Project Structure: project_structure_report.md

🔍 Troubleshooting

Backend won't start

Ensure Python 3.11+ is installed
Run uv sync to install dependencies
Check port 8000 is available

Frontend connection issues

Verify backend is running on port 8000
Check CORS settings in backend/main.py
Clear browser cache and reload

No cores found

Verify backend/data/cores.json exists
Check file permissions
Review backend logs for database loading errors

📄 License

This project is licensed under the MIT License. See LICENSE file for details.

🙏 Acknowledgments

McLyman's seminal work on magnetics design
OpenMagnetics project for database integration
FastAPI and Nuxt communities
Open-source power electronics community

Version: 0.3.0
Status: Active Development
Last Updated: December 2025

Name		Name	Last commit message	Last commit date
Latest commit History 40 Commits
.agent		.agent
.github/workflows		.github/workflows
backend		backend
frontend		frontend
pdfs		pdfs
.gitattributes		.gitattributes
.gitignore		.gitignore
README.md		README.md
docker-compose.yml		docker-compose.yml
package-lock.json		package-lock.json
power_transformer_framework_plan.md		power_transformer_framework_plan.md
power_transformer_framework_plan_2.md		power_transformer_framework_plan_2.md
project_structure_report.md		project_structure_report.md
test_output.txt		test_output.txt
user_guide.md		user_guide.md

Denys/transformer_designer

Folders and files

Latest commit

History

Repository files navigation