Puppet Master final project page with links to all my Fab Academy work:
http://as220.org/~anna)

I wish that I could relate tips for troubleshooting the Modela milling machine in this post - but I have hit a brick wall. (However, I recommend you try the steps below if you are troubleshooting your own milling machine.) My boards have been milling extremely hairy for weeks - punctuated by a workable board or two out of 50 or so milled.

I have tried:

• changing the bit
• changing / tightening the set screws
• moving the position of the spindle (from high to low - low to high)
• scraping the bed with razorblades to ensure that the boards are level (we accumulate tape and glue on the bed sometimes).
• moving the position of the copper to be milled on the bed (sometimes this helps and but I can not replicate the same results in the same place twice.)
• I have tried using different batches of whatever copper stock we have left in the lab.
• I read the manual - no real help there - mostly Modela specific software tips.
• I tried booting into Windows (we usually run the machine using Ubuntu) to see if there was a “recalibration” or “troubleshooting” program or setting in the Modela software - I couldn’t find anything to run that was machine based.

When I first began using the Modela back in January - I often had these types of problems, but as I became better acquainted with the machine, the boards came out better overall, with an occasional hairy board. However, in the last few weeks I mill almost exclusively mangled boards with the occasional usable board being produced.

As I am completely out of time - I thought I would try to cut my remaining infrared prototype boards for my final project / networking on the vinyl cutter out of flexible copper sheets. The boards are very tiny - I am not sure I can get them off off the sticky background without destroying them. I cut many multiples to use as spare parts for when parts get torn. Hopefully this will work and I can finish my projects.

Puppetmaster Project Overview

My final project for Fab Academy is a universal remote I am calling PuppetMaster, due to its ability to control out of reach objects / devices using the fingers. This post details the PuppetMaster boards v.1 - v.2 that use capacitive sensing.

Capacitive Sensing Board Versions

The first working version of the prototype uses capacitive sensing built on the Hello World Step response example, but converted for use with a attiny 44.

I fabbed the board(s) myself, using the standard fab inventory parts.

PuppetMaster Board Version 1 —> pitfalls / issues

I designed my own board based on the hello step response with 3 additional inputs for each of the four fingers. (The hello step response has only one input). As the hello step response example uses the attiny 45 (and does not have enough input / charging pins for 4 inputs), I modified Shawn’s version of the hello step response to work with the attiny 44. This initial version of the PuppetMaster board has a FTDI header and 4 inputs / charge pins. PuppetMaster v.1 However, there was a issue with this initial board and I could not get the modified code for the attiny 44 to work with the Hello.Step.45.py code. I attempted triage, but the board remained flaky. There seemed to be an issue between the ground pin on the chip and power. In addition, I realized that I routed the Tx to Rx wrong on the FTDI header pins. I was using jumpers to work around it initially, but after the version 1 board’s persistent flakiness, I decided to make a version 2 board. I am not including the schematics & Eagle files for version 1 because of these issues.

Code Modification:Â From attiny 45 to attiny 44

I modified the hello step response assembly language code to work with the attiny 44 microcontroller. Both sets of .asm files are listed below. Attiny 45 FTDI Hello Step Response Code Attiny 44 FTDI Hello Step Response Code

Puppetmaster Board Version 2

Here is the version 2 board with the correctly mapped FTDI header and some slightly shifted components to allow for more space between some problematic components. This board version has been flashed with the attiny 44 code (above) and works with Hello.Step.45.py code.

Eagle Files

Download - PuppetMaster v.2.0 Board Download - PuppetMaster v.2.0 Schematic

CAD Image Files Used to Mill board v.2

Putting the Version 2 Board Together

I ran out of time in the lab and did not get a chance to mill out the board. Instead, I scored the perimeter with a utility knife and broke off the excess edge. Then I used a rasp to file the edges down.

Capacitive Sensing in Action

Using Neil’s term.py program to visualize the initial charging graph and sensor input.

Puppetmaster Overview

My final project for Fab Academy is a universal remote I am calling PuppetMaster, due to its ability to control out of reach objects / devices using the fingers. This is the revised final project proposal - the first proposal was overly ambitious and encompassed too many sensor types in too short of a time period.

First Working Prototype: The first working prototype (illustrated in the PuppetMaster v.1-v.2 [capacitive sensing] post) uses thin copper sheets to create a capacitive sensor between the index finger and the thumb.

For Fab 6: By Fab Academy graduation I propose that PuppetMaster will consist of a fabbed board that reads gestural input from accelerometers.

Overall Project Goals / Design Considerations:

• Create a remote control that takes input from finger movements / gestures
• Remote is small and wearable (will likely look like a bracelet or a watch) with connections to the fingertips.
• Remote is comfortable enough for long-term wear, will help to relieve hand / joint fatigue

Project Plan

Short Term Project Stages (Completed by Fab 6)

All board(s) will be fabbed, using the standard fab inventory parts with a few exceptions, (the sensors used in the later stages).

The First Stage [capacitive sensing]

• Completed June 2nd - consists of using capacitive sensing as input data.

The Second Stage [ accelerometer input]

• Completion before Fab 6 in August, will include the addition of accelerometer on each finger to record control gestures.
• Simple gestures will be used for input controls.

Long Term Project Stages (Work in Progress - NOT for Fab 6)

• Additional stages (to follow in the in the Fall) will include one mode of control (IR) but other modes (radio, bluetooth) will be added as the project progresses. (See stages outlined in the table below.)
• Eventually, a dial or switching mechanism will be added to switch between the modes. Although the other modes of control mentioned above will be explored, the gestural input will be the primary focus.
• In the final version of this prototype, the user of PuppetMaster will be able to input information via gestures and remote control objects by using sensors embedded in the fingertips.
• These additional modes (combined with the necessary code) will allow the user to control almost any device with hand gestures. I also want to develop a complete gestural language (possibly based on sign language)in tandem with the addition of 6-axis sensors (3 axis gyro, 3 axis accelerometer) to the device.

Fab Academy Module 15:

Invention, Intellectual Property and Business Models

The Fab Academy Assignment

Develop a plan for distributing your final project.

Distribution Plan

I plan to distribute my PuppetMaster (see Final Project pages in the main navigation for details on this project.) final project through my website as downloadable schematic and board files. Releases will be made available as individual stages are completed. If the project becomes popular, I may consider distributing it as a kit.

It is unlikely that I will patent this device, as patents are expensive and afford little protection unless you are the owner of a large company with deep pockets who is prepared to litigate.

I instead intend to use the Creative Commons for both licensing and enableling others to find these plans. (a non-commercial, modifications allowed, share alike license seems to make the most sense). This will enable me to keep the copyright to my work while allowing others to us, modify and improve it.

I will create and register the actual license when my final project is closer to completion.

Skills Learned

• Creative Commons licensing structure.
• A broad overview of patents and MIT distribution models

The project page for my PuppetMaster final project (and all of my Fab Academy project content) is located here: http://as220.org/~anna/

View the project page here.

Final Project Overview

My final project for Fab Academy is a universal remote I am calling PuppetMaster, due to its ability to wirelessly control out of reach objects / devices using the fingers. This device enables the user to remotely control infrared receiving devices (such as a television or stereo) from a maximum distance of 100 - 150 feet. The user of PuppetMaster will be able to operate the remote by using switches embedded in the fingertips. The board and power supply will be worn around the wrist like a watch or bracelet. (This makes weight and a small form factor a major design consideration for this project). Additional controls will be added in later stages. (see details below)

Form Factor / General Idea Mockup

If IR model - Infrared LED would be placed on top of hand in wrist strap.

I intend to fab the boards myself, using the standard fab inventory parts with a few exceptions, (especially the sensors used in the later stages). The first stage (to be completed by June 1 for the fab academy final project) consists of one mode of control (IR) but other modes (radio, bluetooth and additional gestural sensors) will be added as the project progresses. (See stages outlined below.) A dial or switching mechanism will be added to switch between the modes. These additional modes (combined with the necessary code) will allow the user to control almost any device with hand gestures. I am also keen to develop a gestural language in tandem with the addition of 6-axis sensors (3 axis gyro, 3 axis accelerometer) to the device.

I intend to use a lithium battery for lighter weight and longer life - that may need to come later, I am not sure I will have time to design a power supply.

Overall Project Goals:

• Create a remote control that takes input from finger movements / gestures
• Remote is small and wearable (will likely look like a bracelet or a watch) with connections to the fingertips.
• Remote is comfortable enough for long-term wear, will help to relieve hand / joint fatigue

Short Term Goals By Stage

Prototype in stages, adding functionality and additional control channels / technologies in each stage

1.0 Stage 1: Infrared

1.1. Use infrared to turn off on devices, navigate devices (most likely television)

1.1.1. Most likely commands:

1.1.1.1.On

1.1.1.2.Off

1.1.1.3.Up channel

1.1.1.4.Down channel

1.1.1.5.Other specialty features?

1.2. Power supply -> LiPo model uses lightweight lithium battery and outputs 5V

1.2.1 This is perfect for the TV-B-Gone style board I want to create / modify / fab

1.3 Parts List for puppetmaster (phase 1 - IR)

1.4. Complete working prototype by 6/01/2010

2.0 Stage 2: Radio Control

2.1. Add radio control to Stage 1 prototype

2.2. Enable mode switching on input device - different modes - different gestures

2.3. Experiment with driving a RC device.

2.4. Complete working prototype by 6/30/2010

3.0 Stage 3: Bluetooth

3.1. Add Bluetooth to Stage 1 prototype

3.2. Enable mode switching on input device - different modes - different gestures

3.3. Experiment with driving a RC device.

3.4. Complete working prototype by 7/30/2010

4.0 Stage 4: 6-axis motion sensors

4.1. Add additional sensors to interface:

4.1.1. 6-axis motion sensors (3 axis gyro, 3 axis accelerometer) to input gestures.

4.2. Complete working prototype by 11/01/2010

5.0 Stage 5: Gestural Language

5.1. Utilize 6-axis motion sensors (3 axis gyro, 3 axis accelerometer) to input gestures.

5.2. Creation of full blown gestural language (possibly based on sign language)

5.3. Complete working prototype by 11/01/2010

6.0 Stage 6: Personal LAN

6.1. Connect to the internet to create personal LAN.

6.1.1. Most likely by fabbing a board that can plug into an iphone,

6.1.2. This board could then provide any other user devices with data access

6.1.3. Could connect to Premonition system

6.2. Complete working prototype by 11/01/2010

Schedule:

I've been lasercutting a lot of textiles for a while now. My diorama will be illuminated by a chandelier of lasercut textile and flameworked glass plants. The LEDs will be attached to a photosensor. As ambient light fades, the LEDs will get brighter. These are LEDs that will be within the textile plants.

I assembled these form studies to get a better sense of scale for larger chandeliers. I flameworked these glass components. These forms will also house LEDs. Above are some dolls that I sculpted. I also lampworked the glass eyes, and made the clothes and wigs.

The diorama will display automated figures controlled by servo motors. The figures and the diorama itself will include some cast parts.

My method of distribution for this project is to document the animation, and to use photographs of the diorama as illustrations. My intention is to make a series of these animated dioramas, and use the documents to delineate longer narratives; to illustrate books and compose longer animations that can be distributed.

I initially wanted to do reverse-osmosis water desalination, but then I discovered forward osmosis, which uses ammonia salt to pull the water through membrane, detailed in this page about Yale's evaporative draw-solution system, but then Shawn emailed me a link to Ion Concentration Polarization, which reduces filter clogging, and decided to switch to making an ICP water filtration device with shrinky-dinks In the mean time I noticed that water desalination isn't very fabby, so decided to tack on a PPM/Ph meter and use feedback to calibrate the device.

I then proceeded to accomplish nothing. I don't know how to attach electrodes to anything small, and DuPont doesn't provide samples of Nafion, so I'll have to order some, and right now I'm just concentrating on getting parts together for the salinity meter, but that's a well-documented project I haven't done any research of my own on.

I'm considering switching to building a long-range capacitative+inductive RFID meter with an SD card to do a generic survey of RFID at this point, I guess I didn't panic early enough.

-\n

Stravinsky's first opera was The Rake's Progress, and for years after he always intended to compose another. In 1953 he met with Dylan Thomas in New York to hash out the details of a libretto for an ambitious opera to be commissioned by Boston University. Several possible scenarios were tossed out, including a postapocalyptic love story and a libretto written by a large number of monkeys on typewriters. As fortune would have it, the poet died before the opera could be realized. This machine tells the story of the unfinished opera using the vernacular designs and mechanisms of pinball.

Design

The game components will tell the story of the opera by combining the following elements:

4 possible scripts
12 characters
12 props
12 locales
26 typewriter keys

Each of the elements is keyed to a target on the playfield and has corresponding LED feedback.

General color guidelines for the design are:

Black and white cross-hatch  
Spot complementary colors: orange/blue, purple/yellow  
Collaged b&w paper images

Components

The various components fall into these categories:

 Manual Input
       Flipper
       Plunger
       Coin op

Automatic mechanisms
    Ball return
    Tilt and bump sensing

Tracked Input
    Targets
    Bumpers
    Holes
    Switches

Output
    Lights
    Sound and music
    Score
    Backglass display

In the triage process of the final week I have focused on getting two output systems working; the LED array drivers and the sound and music generator.

The Electronics

The two working playfield ornaments are based on the Fabian Arduino-compatible board.

The Fluxamafet: sixteen N channel MOSFETs on an Atmega168. Controls banks of LED arrays.

The Fluxamamidi: An Atmega168 that generates MIDI sequences for all the sound effects and music in the game. Can read MIDI patches from an SD card and play back according to script and input.

The Fluxamabox (UNFINISHED): A MIDI to audio sound generator based on the ATSAM2195.

The LED arrays

I started by milling a bunch of breakout boards grouping six LEDs into different configurations:

Coupled with the controller, they formed a dreaded floating octopus of wires:

Although they look nice from the front:

For the next group I decided to use the vinyl cutter, which works well for this kind of application:

Programming interface

A Processing interface for visualizing the array. Because Processing is syntactically similar to Arduino, you can cut and paste the model and upload as working firmware.

Here's the Processing code.

Flippers

For now, the flippers are activated by (120V AC) sewing machine solenoids that I happen to have a lot of. I also have a bunch of 5V electromechanical relays that can switch them. Eventually I will swap these out for two-coil flipper solenoids comparable to those at pinballmedic.net.

The Playfield

Just got my pinball balls in the mail:

Clay model to determine form factor.

Structured light scanning attempt. Figuring out slices for laser cut frame.

PCB designs for buttons and bridge.

Here is a very brief clip of a nodding head.

I attached a doll head that I sculpted (with flameworked glass eyes) to a servo motor. This will be part of a larger diorama of dancing automatons that I'm working on.

I've been in L.A. for about a week now and I'm really eager to get back to the lab. I have so many ideas about how to incorporate every lesson thus far into my big project, especially making molds to cast multiples of tiny body parts for the characters I've been sculpting. I'll also be molding some 3-D rendered forms. Complicated and interactive automata dioramas!

I hand- sculpted the heads above in Cernit. I flameworked the glass eyes, and the "hair" on the right is made of trochus shells.

I'm working on creating an chording keyboard

http://en.wikipedia.org/wiki/Chorded_keyboard

for my final project. Hopefully I'll be able to use the skills I learn in the FAB Academy to make one that actually works.

Pinball machines need a lot of coils. I realized that my first step in creating a pinball machine would be to build a tool to wind accurate coils. The pinball machine hobbyist scene is pretty well organized, so I quickly found this chart at pinballmedic.net detailing the various coil winding schemes that I will need:

Coil chart at pinballmedic.net