Making Coffee with Flutter
summary
A 1995 Illy Francis X1 gets gutted and fitted with a temperature probe, solid‑state relay and ESP8266 running a PID controller that achieves ~0.1° C accuracy; extraction target stays 88–96° C and P gains are adjustable live. A Raspberry Pi runs an MQTT broker and a Flutter UI (uses Riverpod, fl_chart, mqtt_client) to tune PID and view temperature charts; performance on Raspberry Pi OS is solid, and for production ARM64 devices the recommendation is Flutter Elinux with Wayland and MQTT broker integration.
AI-written summary · about
chapters (21)
- 0:00Introduction to talk
- 0:49My journey to coffee machines
- 4:03The coffee machine itself
- 5:05Basics of coffee science
- 7:33How the machine works
- 9:20Hardware modifications
- 10:04Software and PID control
- 11:50Why Flutter for control
- 12:54Future plans for the system
- 15:46Why embedded Flutter
- 17:12GUI frameworks comparison
- 20:24Why Flutter fits embedded
- 21:05Running Flutter on embedded
- 22:08Flutter embedder options
- 23:14Display servers explained
- 25:43Choosing the right embedder
- 27:11Flutter ELinux by Sony
- 29:20Embedded system considerations
- 30:02Step-by-step installation guide
- 31:49Development and deployment
- 32:10Summary and resources
