Kettle Companion Redesign - Design Review Summary
Purpose
This document consolidates the design decisions, requirements, constraints, and rationale developed during the review process for the Kettle Companion redesign. It is intended for internal review and sign-off, including review by Andy Stanford-Clark.
Project Goal
Replace the existing WeMos D1 mini plus single RGB LED shield implementation with a simpler, more robust, lower-cost integrated PCB and enclosure approach suitable for prototype builds now and scalable manufacturing later.
Legacy Hardware Being Replaced
- LOLIN / WeMos D1 mini controller board
- Single RGB LED shield using one addressable RGB LED
- Existing enclosure originally designed around the stacked board arrangement
Chosen Electronic Architecture
Controller
- Selected module: ESP32-C6-MINI-1-N4
- Rationale:
- Supports Wi-Fi and Zigbee as required
- Also supports Thread / Matter-related ecosystem paths through IEEE 802.15.4 capability
- Better long-term platform than ESP8266 for future product evolution
- Suitable for native USB-based programming/debug strategy
LED
- One central SK6812-compatible addressable RGB LED
- LED to remain centered in the same functional position as the original shield LED
- LED to point upward
USB
- USB-C connector for both power and programming
- Intended production USB role: power plus firmware/setup
- Use simple 5 V USB-C sink operation, not USB-PD
- Use a mechanically reinforced USB-C receptacle with shell stakes / anchor tabs / through-board retention features
- Pads-only SMT USB-C retention is not acceptable
- Hidden test pads shall be provided for power, ground, EN/reset, boot access, and UART
- No visible external reset button is required in the baseline design
Power Strategy
- 5 V from USB-C input
- 3.3 V local regulation for the controller
- LED powered from appropriate rail per final electrical implementation review
Locked Mechanical Constraints
- PCB must fit within the original D1 mini footprint envelope or be smaller
- PCB must not exceed the envelope defined by the D1 mini mechanical reference
- USB-C connector must remain in the same effective enclosure position as the original board connector location
- RGB LED must remain central on the board in the same effective optical position as the original shield LED
- Board design should move from stacked boards to a single integrated PCB
Enclosure Direction
The enclosure is being redesigned as part of a transition from 3D printing to injection moulding.
Enclosure Goals
- Lower build cost
- Faster assembly
- Higher reliability
- Secure USB insertion without stressing PCB solder joints excessively
- Simple drop-in placement with low risk of assembly error
Chosen Retention Strategy
Primary retention:
- Snap-fit enclosure approach
Fallback retention:
- Single screw option designed in as a backup if snap-fit prototype testing is not robust enough
Mechanical Strategy
- Enclosure should self-locate the PCB
- USB-C side should act as a primary locating datum
- Board should be supported so plug insertion forces are reacted by enclosure geometry as much as possible
- Avoid dependence on header pins or stacked-board solder joints for structural strength
- Avoid adhesive as the primary retention method
Optical Strategy
Final decision
- No light pipe
- The enclosure itself should diffuse the light
- The LED should point upward in the center
- The desired visual result is a soft diffused glow, not a focused indicator point
- Prototype validation should tune enclosure material, thickness, finish, and spacing rather than adding a separate optical part first
Rationale
- Lower part count
- Simpler assembly
- Lower tolerance stack-up risk
- Better fit with low-cost manufacturing goals
- Better fit with the desired whole-product glow behavior
- Existing design discussion indicated that a light pipe would focus light too much in one area
- Product height and distance from LED to enclosure top can be used to control brightness and hotspot behavior
RF / Antenna Direction
- Use the PCB antenna variant of ESP32-C6-MINI-1-N4 for the first revision
- Do not add antenna complexity prematurely
- Prototype and validate RF behavior before considering alternatives
- One antenna system is sufficient for the target 2.4 GHz wireless use cases; separate Wi-Fi and Zigbee antennas are not required in the baseline design
Manufacturing Direction
Current phase
Future intent
- Evolve toward production in thousands
Recommended philosophy
- Design now with simple assembly and scalable production intent
- Minimize loose parts and manual assembly steps
- Prefer enclosure geometry over extra hardware where possible
- Use the enclosure and connector mechanics together to improve durability
Reference Inputs Reviewed
Uploaded files and references reviewed during the process
- D1 mini mechanical reference PDF
- D1 mini schematic PDF
- SK6812 datasheet PDF
- Enclosure photos with existing board installed
- Enclosure STL reference
- Multiple chat screenshots containing Andy's design feedback and dimensional notes
Key extracted technical references
From the uploaded D1 mini schematic reference:
- USB-C sink-style implementation with CC resistors was identified in the reference design
- Local 3.3 V regulation and USB-related circuitry were identified as useful architectural references
From the uploaded SK6812 datasheet:
- 5 V-class LED supply behavior
- 800 kHz signaling
- Reset low timing requirement around 80 us
- Threshold behavior indicating that MCU-to-LED signaling must be reviewed carefully in the final schematic implementation
Review Process Followed
- Identified the legacy hardware stack being replaced
- Confirmed the legacy LED arrangement is a single central LED, not a 7-LED shield concept
- Compared controller options and selected ESP32-C6-MINI-1-N4 because Zigbee is a real requirement
- Chose USB-C for both power and programming
- Locked the mechanical constraints around D1 mini footprint, central LED position, and connector location
- Reviewed enclosure photos and discussed robustness for plug insertion
- Shifted enclosure strategy from stacked-board support to a single integrated PCB
- Evaluated retention options and selected snap-fit as primary with single-screw fallback
- Evaluated optical options and removed the light pipe in favor of enclosure-based diffusion
- Aligned the design direction with low build cost, low assembly effort, and reliability goals
Key Outcomes for Sign-off
Electronics
- ESP32-C6-MINI-1-N4 selected
- One central SK6812 selected
- USB-C power plus programming selected
- Hidden test pads selected for factory/service access: power, ground, EN/reset, boot access, UART
- Production USB behavior narrowed to power plus firmware/setup
Mechanics
- Single integrated PCB selected
- Keep within original D1 mini board envelope
- Keep USB-C and LED in original functional locations
- Reinforced USB-C connector required
Enclosure
- Injection-moulded redesign assumed
- Snap-fit primary retention
- Single-screw fallback retained in design reserve
- No light pipe
- Enclosure provides diffusion
Open Items Remaining for Implementation
These items do not block the design direction, but should be validated during implementation:
- Final connector part selection based on mechanical robustness and enclosure capture details
- Final PCB support / snap geometry in the injection-moulded enclosure
- Prototype validation of LED diffusion through final enclosure material and thickness
- Prototype validation of PCB antenna performance in the real enclosure
- Final pin mapping and USB implementation details in the schematic
Summary
The redesign direction is now intentionally simpler than the legacy arrangement:
- one board instead of stacked boards
- one central RGB LED
- USB-C with reinforced mounting
- enclosure-diffused glow instead of a light pipe
- snap-fit assembly with single-screw fallback
This direction best matches the stated goals of lower cost, simpler manufacturing, improved robustness, and a scalable path from prototype to volume production.