Current setup for board includes a hall effect sensor, that toggles when magnets are connected. It may be simpler to just keep querying the RFID board, and if the RFID chip has been removed, then stop playing.

Should be functionally the same, but may be worse for power consumption, which so far hasn't been modeled or thought about.

I would like to replace all of the print statements with log statements, and I would like the momobox to keep track of some basic on-board logging. Some thoughts on the logger:

• I need a switch to make the Pico read-only. As described here, the Pico can self-write to its own filesystem, or it can allow the PC its connected to to write, which is used for easily updating the code. The linked article suggests making a switch, and adding it to the boot.py file. I propose to add the following to boot.py:

import board
import digitalio
import storage

switch = digitalio.DigitalInOut(board.GP22)
storage.remount(mount_path="/", readonly=not switch.value)

Currently GP22 is the pin that detects whether an SD card is inserted. What the code above would mean is that the file-system would be read-only (able to update the code) if the SD card was not present. Since the SD card is required to use the momobox, part of updating the onboard code would be to boot the momobox with the SD card removed.

• I would want to minimize and limit the size of the logs, since the on-board memory is limited to 1 MB, and some of that space is already spoken for. I know that the python logging module allows for this to some extent, so I'll have to implement it.
• On the subject of the above, it would be useful to minimize the size of the log file by minimizing log messages; it's okay to be terse

Currently the only option for volume control is to turn the screw on the audio board. End user's key stakeholder requests easier adjustment.

Not sure how to implement this easily, may take some creative thought.

Current system only has 1 MB of memory. Need to add the SD board and SD card.

Speaker mount points are not easy to work with; no good tool access, not even worth trying to get the square nuts into the bottom mount points. Needs to be rethought.

Best idea so far: Passive slot at the bottom of the box that accepts the speaker flange. Maybe keep improved screw bosses near opening of top of box, but there is also an option to make a flexured post or latch to simply snap the speaker into place.

Heat-set inserts are great, and also expensive. They cost about $17 on McMaster, and I'd like to offer users a cheaper solution.

Will probably go with some sort of press-in or slot-in hex or square nuts. Currently the heat-set inserts are used in the lid to box connection and the RFID board to lid connection.

The latter could be tapped M3 fasteners since the RFID board nominally only goes on once; the lid to box connection benefits greatly from being threaded.

Need to incorporate a battery so that momobox runs wirelessly.

Start a basic bill of materials, either in markdown table or csv format. Should include the following columns:

• Component Name
• Quantity Required
• Description or Notes
• Source (link to purchase)
• Price

Okay for this to be wrong / incomplete. Also okay for components to be identified as needed and not fully specified. Multiple options for components also okay.

The firmware runs continuously when the batteries are connected. Goal is to have the pico go into sleep mode after some amount of idle time, and then awake upon a signal from the hall sensor.

A good starting point with resources is here: https://ghubcoder.github.io/posts/deep-sleeping-the-pico-micropython/

May be better to also read this post about dormant and sleep modes: https://ghubcoder.github.io/posts/awaking-the-pico/

This is micropython based (momobox currently uses circuit python), so may not be fully applicable.

A good thing to start with is to measure the current draw from the batteries during the following operations:

• idle time
• RFID request
• playing a song
• song paused

Current box design is monolithic. Looks nice, but I think there are advantages to a separate bottom:

• iterating on the box design likely doesn't require a new bottom. In the latest print the bottom took 3 hr.
• a totally flat bottom separates from the bed (warps) prematurely
• a ribbed bottom requires support material which is hard to remove.

Input: momie left on top of box. Song completes. Same momie is removed and reattached.

Expected result: song starts from beginning

Observed behavior: song does not restart. Placing a different momie on top of box works as expected

Need to learn to program some fundamentals on the Pi. Issue is closed when I can do the following:

• control a program by opening or closing a physical switch (or connect a wire on the breadboard)
• write and retrieve a value that is persistently stored on the Pi
• copy a file from my computer to the Pi
• read from a file on the Pi
• delete a file from the Pi

Good explanation and possible solution here: https://forums.adafruit.com/viewtopic.php?t=188012

And here:
https://forums.adafruit.com/viewtopic.php?f=8&t=187616#p909859

Need to order some 10nF and 100nF capacitors and check my resistor supply, then check on a breadboard

• Magnet needs to be positively secured to top of lid (maybe just use superglue)
• Bosses for inserts need to be larger diameter so they don't deform when insert installed
• Verify positive separation between magnet and RC522 board
• Need clocking feature between lid & base
• Make material between magnet and top of lid as thin as possible
• Not enough wire routing space for RC522 harness

Currently the red RFID board LED is on whenever the unit is powered. Nice to see that it's on, but the light is off-center and not there intentionally. One or more dedicated LEDs to communicate status would be better

• Board mount fasteners to lid are not accessible. Removing the board mount requires partial removal of the pico and the SD card board. Thru-holes have been added to the board mount, and corresponding bosses need to be added to the lid.
• Battery clip blocks board mount fasteners on back of board
• Harness routing features, particularly for RFID harness, should be added to ensure that nothing gets pinched when the lid is closed.

We're tired of having to swap the batteries every day. While it would be nice to reduce power consumption (see #16), it's also necessary that it can charge simply by plugging it in.

Some very preliminary research: https://www.circuitbasics.com/how-to-power-your-raspberry-pi-with-a-lithium-battery/

As I understand it I'll need three components, which I now have on order:

• A re-chargeable lithium-ion battery (3.7 V)
• A voltage regulator board - This keeps the power to the Pico at 5 V regardless of the battery state of charge
• A battery charge controller

My hope is to use the above together so that the user can plug in directly to the charge controller, and still be able to use the box when it's plugged in. At the very least, not need to reconfigure anything while charging. If this fails, I bought a more expensive all-in-one board that I can use instead: https://www.digikey.com/en/products/detail/adafruit-industries-llc/2465/5356834

Also purchased another Pico with pre-soldered headers to try this with, as well as a second breadboard.

While I'm waiting for parts to arrive, I have some CAD work to do to figure out how I'm going to mount two additional boards and a new battery within the existing housing. I'll probably want to move the Pico back and put the charge port in its place, and 3D print a new battery clip.

The "official" method of talking to the Raspberry Pi Pico is through the Thonny editor, which isn't my preferred development environment.

Check out this article and see if I can't get Raspberry Pi Pico to talk directly to VSCode, which is my editor of choice.