In this episode, we discuss the latest with my solar setup (EcoFlow battery, Eco-Worthy panels) and take a deep dive into local AI — exploring LLMs you can run at home, privacy implications, and practical use cases. We also field live chat questions, talk about Ollama, LM Studio, voice cloning, and more. Thanks for listening!















In this episode of Home Gadget Geeks, I share the latest updates from my ongoing solar power experiment — including new insights from my EcoFlow Delta 3 battery system and a fresh set of Eco-Worthy 100-watt solar panels mounted to the front porch. I break down how the system is performing day to day, what kind of watt-hour output I’m seeing in different weather conditions, and why small-scale solar setups can be more about learning and data than raw power savings.



I also compare EcoFlow’s app and battery management features with my older Bluetti unit, explaining why I’ve come to prefer EcoFlow for its detailed analytics, charging control, and real-world usability. From wiring choices to panel placement angles and future expansion plans, this episode dives deep into the practical side of building out a personal solar lab at home.



In the second half of the show, I turn to another passion project: running local AI models on my Mac Mini M4 and Windows desktop. Using tools like LM Studio and Ollama, I explore how large language models can run offline, how they compare to cloud AI systems like ChatGPT, and where local AI might fit into personal or privacy-sensitive workflows. We discuss practical uses for local AI — from reconciling financial data to experimenting with personal analytics — and consider how this technology might evolve as the current AI bubble continues to expand.



Finally, I close with a few thoughts on AI voice cloning, AI-generated media, and what it means for authenticity in the age of synthetic content. It’s a packed hour that blends renewable energy, practical tech experimentation, and the very human side of staying current with emerging technologies — all from the Home Gadget Geeks studio.



Home Gadget Geeks #660 – Solar and Local AI FAQ



1. What kind of solar setup is being discussed?



Jim has set up a Delta 3 EcoFlow battery and initially two 100-watt 18-volt solar panels, with plans to expand to six panels. The setup is being used to experiment with residential solar energy capabilities.



2. What are the advantages of the EcoFlow battery over the Blue Eddie?



EcoFlow’s app offers better reporting, control (like setting charging thresholds), and overall user experience. Blue Eddie serves as backup but lacks detailed analytics and direct battery control between fixed charge levels.



3. How much electricity does the current solar panel setup generate?



With two panels, typical generation is around 1 kilowatt-hour per day, depending on weather, location, and sun exposure. This is used mainly for experimentation and backup, not to power the whole house.



4. What are common challenges of expanding a residential solar system?



Wiring and hardware costs (especially copper) can be significant. Panel placement, angle, and managing inverter/battery limits are important. There’s a need to balance panel quantity with inverter and battery capacity.



5. Why consider local AI models (like LM Studio) instead of cloud AI services?



Local AI offers enhanced data privacy and control since data doesn’t leave the device. It can also be responsive without API costs. However, it requires more powerful local hardware and some techni...