Maxwell’s Disciples: The Living Inverter Today, most power inverters are "black boxes"—when they break or when you add more solar panels, you usually have to throw the whole unit away. We want to change that. Our project, Maxwell’s Disciples, is a vision for a "Living Inverter" that is self-aware, upgradeable, and designed to last.
The Vision Instead of a static piece of hardware, we are designing a modular ecosystem. Our goal is to create an inverter that tracks its own "health" by measuring vitals like voltage, current, and temperature. By analyzing this data, the system can calculate a real-time health score, warning you before a failure happens instead of leaving you in the dark.
Scalability & Efficiency We aren't just looking to monitor; we want to expand. Our idea is to use a modular architecture that allows users to add more power capacity whenever they need it. By adding parallel MOSFETs, we can physically lower the resistance of the system, making it more efficient as it grows. This is based on the principle:
$$R_{total} = \frac{R_{DS(on)}}{n}$$
In this formula, ( n ) is the number of modules added. As ( n ) increases, resistance drops, and efficiency goes up. We plan to use firmware to calculate this efficiency in real-time:
$$\eta = \frac{V_{out} \cdot I_{out}}{V_{in} \cdot I_{in}}$$
The Challenge Ahead Since we are currently in the ideation phase, we know we have big hurdles to clear. We need to solve "current hogging"—ensuring that parallel components share the load equally so they don't burn out. We also need to design a PCB layout that stays clean and accurate even in a high-power environment.
Why It Matters We believe the future of green energy isn't just about better batteries; it’s about smarter, transparent hardware. We want to move away from "disposable" tech and toward "snap-on" upgrades. Our goal is to build a system that tells you its own "expiry date" and grows with your home, making renewable energy more sustainable and affordable for everyone.
Built With
- altium
- android-studio
- nodered
- platformio
- python
Log in or sign up for Devpost to join the conversation.