Geometry, Economics, and Failure Modes for Real-World Designs
AI-Augmented Learnings Series – Volume 1
Available on amazon. https://a.co/d/7z06COk
Design solar and storage systems that don’t catch fire, go broke, or leave you in the dark.
Practical Solar Power Systems is a concise handbook for technically minded readers who want PV and battery systems that work in the real world—weak grids, hot roofs, messy user behavior and all. Instead of drowning you in code clauses or sales talk, it focuses on a few clear mental models for geometry, wiring, batteries, and economics, and then shows you how to use an AI assistant as a tireless junior engineer and tutor.
This book is written for engineers, serious DIYers, and practitioners who like numbers more than slogans. If you can follow a spreadsheet and basic electrical safety, it will walk you from “I want solar” to a defensible design: modules, inverters, batteries, wire sizes, protection, and economic justification.
In this book you’ll learn how to:
- Think of solar as geometry plus efficiency—why cross-sectional area, tilt, and seasonal loads matter more than panel marketing numbers.
- Understand IV curves, MPPT, and overpaneling so you design for the year, not one perfect hour of sun.
- Spot and prevent arcing and rooftop fire risks with sane wiring, connectors, and layout choices.
- Treat batteries like tires, not magic boxes—chemistry trade-offs, SOC windows, temperature limits, and why lead-acid often dies young in hot climates.
- Size storage using a marginal payback vs. endurance method: the economic lower bound from the payback of the next 5 kWh chunk, and the upper bound from how many cloudy days you want to ride through.
- Work through full examples of grid-tied, hybrid, and off-grid systems with realistic constraints, including weak grids and generators
Built for AI-augmented learning
Every chapter is structured to be used with an AI assistant from day one. You get checklists, design procedures, and end-of-chapter problems that double as conversation starters with an AI tutor. You’ll learn to:
- Explain your own reasoning step-by-step and ask an AI to critique it, like a senior engineer would.
- Remix the examples using your own site: climate, tariffs, hardware, and risk tolerance.
- Use “rabbit hole” questions to explore deeper topics—panel efficiency limits, the newsvendor problem, optimization, AC machines, and more—at your own pace.
The goal is not to let AI design systems unsupervised, but to train you to use it as a fast, honest junior: something you can argue with, correct, and lean on for structure while you keep final responsibility for safety and performance.
About the author
Dave Keil is a professional engineer who has designed and built solar and storage systems for harsh, isolated environments, including the performance-focused world of international solar car racing. With a background in chemical and mechanical engineering, finance, and project management, he cares less about perfect lab curves and more about whether systems stay safe, boring, and economical ten years after commissioning.