Thursday, December 29, 2011

New Body, New Issues

I wanted to take some decent shots of the new Frankenkindle body, so you can see what it looks like when fully assembled.

Front view.  Agatha Christie approves.

 Another front view, 
slightly different angle for dramatic effect (try to contain yourself).


Rear view.

Note that the electronics haven't changed just yet.  This is simply a new package for the same device we've demonstrated earlier.  

This button arrangement worked out beautifully for my sister, but it turns out it doesn't fit on her tray!  She uses a joystick to drive the chair, which stands about 6" high and sits at her right hand.  Her left hand is strapped into a permanent support on the left side of the tray, giving me about 10.5" of useable space on the tray, right in the middle.  This body is just over 18" wide.
 
I had spent so much time designing the body itself that I didn't consider the available room on her tray. You can't see it, but I'm banging my head on my desk.  Rookie mistake. 

This is what R&D is all about.  Trying to anticipate everything, and then trying to fix what you didn't see coming.  In this case we'll need another body design, but I now have some experience with that so the next one should come much more easily.  The electronics still need to be repackaged as well, so I will try to incorporate that into the next design.  

Stay tuned for more updates as they become available.

Friday, November 25, 2011

New funding opportunity!

Go have a look at our shiny new IndyGoGo funding campaign!

If you can give, that's great.  If not, that's also fine.  I know times are tough, and every dollar counts for everyone.  If you want to help but can't spare a buck or two, I'd appreciate it if you could just tell a friend.  Facebook, Reddit, Google+, I'm not picky. 

Help us get the word out!


Wednesday, November 23, 2011

New body for the FK1

The FK1, as it's being called now, received a new body this week.  Parts were obtained from TAP Plastics in San Rafael, CA, and assembled.  Overall I think it looks very good so far.

Apologies for the image quality - cell phone cameras and all...

FK1 Front Panel Assembled.  Napoleon Dynamite Approves.

 FKI Rear View

 FK1 Front Panel w/ Kindle & Cover

Everything is a learning process.  I'm keeping a list of things to alter/upgrade/change when the FK2 is finally built, and that list keeps growing.  So far it's a secret though, so keep checking back for progress updates!  :)

The next step on the FK1 will be the actual brain transplant.  The control board will be mounted on the large flat area in the rear of the unit and wired to all six control buttons.  The ribbon cable will be routed around the side, and that should do it! 




Tuesday, November 15, 2011

Get your very own Frankenkindle!

Long time no see!  Turns out having an infant at home kinda takes up a lot of one's time.

Last week I received my first request for pricing on a Frankenkindle.  (insert happiness and merriment here).  The eventual goal has always been to offer this up for sale, but the design needs a lot of work.  I've got some great ideas for additional functionality that I'd like to add, but as of now they're just that: ideas.

Stay tuned for information on the next-gen Frankenkindle.  For now though, let's concentrate on what we have right in front of us.

Original Frankenkindle (FK1)

Parts have been ordered from TAP Plastics to create a housing that is easier on the eyes.  I figured it was better to have a platform that could be streamlined and trimmed back later.  It will be constructed of 1/2" plastic and will utilize six arcade-style buttons for control, but will use the same control system we've seen before.  Here's what the design should look like.  Expect pictures after Thanksgiving.

It was this version that I quoted to the reader who requested more information.  Then I realized that there's really no reason to make you guys wait if there's anyone else out there who might be interested.  After all, Christmas is coming up... 

Before we get to a quote, it's important that we all understand a few key points about the FK1:
  • The design isn't finished yet.  
There are a lot of changes and improvements coming soon... (cue cheesy, ominous music)  While the critical page turning functions work beautifully, the user has only the following keystrokes available: Up, Down, Left, Right, Select, and Home.  This is sufficient for selecting and reading books, but it will still require assistance for things like connecting to the Amazon network to download additional books.
  • The design isn't finished yet
The FK1 will be constructed of 1/2-inch plastic.  It will be strong.  It'll probably survive the next big California earthquake.  This was intentional.  Rather than testing structures that may break under unanticipated stresses, I decided to overdesign it.  The housing will be a key area of the next design, which should look better and (with any luck) cost less to produce.
  • The design isn't finished yet  (see a pattern yet?)
The control board that powers the Frankenkindle was hacked together piece by piece as each function was figured out.  While "pretty" isn't usually a requirement for DIY electronics projects, efficiency should be.  This control board will receive the bulk of the design attention in the next revision.  Many new features will be added, and a custom circuit board will be created.  It should be far simpler to assemble, and (again, with any luck) cost less.  The FK1 however, will ship with something similar to the control board you've seen in previous posts.

Finally, understand that the use of any Frankenkindle unit will shred any warranty that Amazon may offer.  for this reason, it is not possible to modify an existing Kindle unit.  Each Frankenkindle (certainly the FK1 and possibly later versions as well) will ship with a new 3rd-generation Kindle.  You will receive a completely assembled unit, and detailed instructions will be included for any operation that may be required for use (for example, replacing the original Kindle cables to add books.) 

So, all that said, if you would like to purchase your very own FK1 unit it can be yours for the bargain basement price of...

$445 (w/ Kindle WiFi)
$495 (w/ Kindle WiFi + 3G)

This won't always be the case, but for now I can only accept payment through PayPal.  Please direct all sales-related inquiries to eewestcoaster@gmail.com.

Tuesday, October 18, 2011

Frankenkindle Field Test - Great Success!

This last weekend my wife and I packed up our new daughter and headed south to Lompoc to spend some time with the family.  The purpose of the visit was twofold: My sister was due to meet and play with both her new niece and the Frankenkindle.  I'll leave it to you to decide which is more important.

The day was a fantastic success on all levels.

When it finally came time to play with the Frankenkindle, my sister took to it immediately.  A very brief introduction was all that was needed, and she was off and running.  Here is some home video we shot of her using the device.  (Apologies for the quality.  My Dad did a great job with the camera but we only got one take because of a dying battery.)


She was a great beta tester, suggesting a number of improvements to the design.  First, the whole system needs to be raised off the table a bit so she can have better access to the buttons (hence the giant textbook).  The next revision will also use different buttons, and she mentioned that she liked these membrane buttons from Sparkfun but would prefer to see them in a single row rather than a 2x2 design.  Excellent feedback - Thanks Sis!

I'm also aware that some of you can't believe I'd gloss over the new baby in favor of electronics.  For you, I present a still shot of my sister holding her new niece, Zoey.

Zoey, Amberly, and myself


Saturday, August 27, 2011

By request: Frankenkindle demo code

Mike at Hackaday suggested that I submit the code used to generate the keystroke sequences that control the Kindle.  He correctly noticed in the prototype video that the pages don't turn very quickly.  I believe that this issue is caused by the custom script running on the Kindle itself.  You can see that the Up and Down buttons used on the menu selection screen operate very quickly, but the page turn operation takes more time.  At this point it's something I think I can live with but I'm always looking for ways to optimize my design.  This project will be completely open source anyway, so there's no harm in publishing it a bit early.

The basic operation is as follows:  The microcontroller waits until a button is pressed, at which point it calls the appropriate subroutine.  These subroutines manipulate control signals going to two multiplexer chips, which effectively act as Single Pole, Quadruple Pole switches.  When activated, they short specific pairs of wires that are connected to the Kindle, mimicking actual button press events.  Note that turning individual pages is not possible from the main Kindle keypad, so the software had to be hacked to allow custom control through specific key press sequences.  Turning a page to the right is accomplished by pressing the Shift key followed by the 5-way Right key.  There is specific timing required to make this work.

TL;DR: Microcontroller waits for you to press a button, then presses a button on the Kindle for you.

Note that the controller itself is a Teensy++ v1.0 (PJRC has since updated the design and currently sells v2.0 boards).  It runs in a mode Paul at PJRC calls Teensyduino, which allows it to use Arduino code.  A full design package (code, schematics, BOM, etc) will be published once I feel the design is complete.  For now, here is the code you saw running in the demo video:


Arduino code

01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
///////////////////////////////////////////////////////////////////////
//
// Frankenkindle v0.1 code
//
// Glenn Johnson
// 8/2011
//
// Monitor simple pushbuttons from V.Reader device and translate
// into Kindle-compatible keypad control signals.  
//
// Note: Chip select signals not used - both multiplexers
//       are always powered.
//
////////////////////////////////////////////////////////////////////// 



void setup() {              

  //Set up serial for debug messages
  Serial.begin(9600);
  
  //set up the digital inputs from the V.Reader
  pinMode(PIN_B0, INPUT_PULLUP);  //Down, marked "Replay"
  pinMode(PIN_B1, INPUT_PULLUP);  //Home, marked as "Exit"
  pinMode(PIN_B2, INPUT_PULLUP);  //Up, marked as "Play"
  pinMode(PIN_B3, INPUT_PULLUP);  //(Not connected)
  pinMode(PIN_B4, INPUT_PULLUP);  //Right
  pinMode(PIN_B5, INPUT_PULLUP);  //Select (Enter), marked as "Bookmark"
  pinMode(PIN_B6, INPUT_PULLUP);  //Left
  
  //set up digital outputs to control the mux
  pinMode(PIN_C0, OUTPUT);    //IN1_1
  pinMode(PIN_C1, OUTPUT);    //IN2_1
  pinMode(PIN_C2, OUTPUT);    //IN1_2
  pinMode(PIN_C3, OUTPUT);    //IN2_2
  //pinMode(PIN_C4, OUTPUT);    //chip select 1
  //pinMode(PIN_C5, OUTPUT);    //chip select 2
  
  pinMode(PIN_D6, OUTPUT);    //LED, for testing
  
  //Set control signals to unused states
  digitalWrite(PIN_C0, HIGH);
  digitalWrite(PIN_C1, HIGH);
  digitalWrite(PIN_C2, HIGH);
  digitalWrite(PIN_C3, HIGH);  
  //digitalWrite(PIN_C4, LOW); 
  //digitalWrite(PIN_C5, LOW);
  
}

void shift()
{
  
  //press SHIFT
  digitalWrite(PIN_C0, LOW);   //IN1_1
  digitalWrite(PIN_C1, HIGH);  //IN2_1
  
  delay(100);

  //release SHIFT
  digitalWrite(PIN_C0, HIGH);
  digitalWrite(PIN_C1, HIGH);
  
}

void left()
{
  
  //press DPAD_LEFT
  digitalWrite(PIN_C0, HIGH);   //IN1_1
  digitalWrite(PIN_C1, LOW);    //IN2_1
  
  delay(100);
  
  //release DPAD_LEFT
  digitalWrite(PIN_C0, HIGH);   //IN1_1
  digitalWrite(PIN_C1, HIGH);   //IN2_1
  
}

void right()
{
  
  //press DPAD_RIGHT
  digitalWrite(PIN_C0, LOW);    //IN1_1
  digitalWrite(PIN_C1, LOW);    //IN2_1
  
  delay(100);
  
  //release DPAD_RIGHT
  digitalWrite(PIN_C0, HIGH);   //IN1_1
  digitalWrite(PIN_C1, HIGH);   //IN2_1
  
}

void up()
{
  
  //press DPAD_UP
  digitalWrite(PIN_C2, LOW);    //IN1_2
  digitalWrite(PIN_C3, LOW);    //IN2_2
  
  delay(100);
  
  //release DPAD_UP
  digitalWrite(PIN_C2, HIGH);   //IN1_2
  digitalWrite(PIN_C3, HIGH);   //IN2_2
  
}

void down()
{
  
  //press DPAD_DOWN
  digitalWrite(PIN_C2, HIGH);    //IN1_2
  digitalWrite(PIN_C3, LOW);    //IN2_2
  
  delay(100);
  
  //release DPAD_DOWN
  digitalWrite(PIN_C2, HIGH);   //IN1_2
  digitalWrite(PIN_C3, HIGH);   //IN2_2
  
}

void enter()
{
  
  //press DPAD_CENTER
  digitalWrite(PIN_C2, LOW);     //IN1_2
  digitalWrite(PIN_C3, HIGH);    //IN2_2
  
  delay(100);
  
  //release DPAD_CENTER
  digitalWrite(PIN_C2, HIGH);   //IN1_2
  digitalWrite(PIN_C3, HIGH);   //IN2_2
  
}


void shiftLeft()
{

  //send SHIFT keystroke
  shift();
    
  delay(200);
   
  //send DPAD_LEFT keystroke
  left();
  
  
  //Two second delay, ensure that the script has time to execute.
  delay(2000);
  
}



void shiftRight()
{
  
  //send SHIFT keystroke
  shift();
  
  delay(200);
  
  //send DPAD_RIGHT keystroke
  right();
  
  //Two second delay, ensure that the script has time to execute.
  delay(2000);
    
}

void shiftUp()
{
  
  //send SHIFT keystroke
  shift();
  
  delay(200);
  
  //send DPAD_UP keystroke
  up();
  
  //Two second delay, ensure that the script has time to execute.
  delay(2000);
    
}



void loop() {
  
  //if a button is pressed, it goes low
 
  if (!digitalRead(PIN_B0))  
  {
    
    //Down
    
    //Kindle command
    down();
    
    delay(1000);
    
  }
  
  if (!digitalRead(PIN_B1))
  {
    
    //Home

    //Kindle command
    shiftUp();
    
    delay(1000);
    
  }
  
  if (!digitalRead(PIN_B2))
  {
    
    //Up

    //Kindle command (Home = Shift + Up)
    up();
    
    delay(1000);
  }
  
  if (!digitalRead(PIN_B3))
  {

    //Not connected
    
  }
  
  if (!digitalRead(PIN_B4))
  {
    //Turn page right

    //Kindle command 
    shiftRight();
    
    delay(1000);
    
  }
  
  if (!digitalRead(PIN_B5))
  {

    //Select

    //Kindle command 
    enter();
    
    delay(1000);
    
  }
  
  if (!digitalRead(PIN_B6))
  {

    //Turn page left

    //Kindle command (page turn left = Shift + left)
    shiftLeft();
       
    delay(1000);
    
  }  
   
}

Thursday, August 25, 2011

Exciting times for the Frankenkindle!


Yesterday was a huge day for the Frankenkindle!


Site traffic jumped, to say the least, as word of the functional prototype demo spread.  Special thanks to Drew from Build Lounge, who saw the post on Reddit and forwarded the link to a few other websites. 

Throughout the day it spread all over the world.  Here are some other sites promoting the project.  I urge you to take a moment to check them out, as many of these sites are dedicated to making devices for disabled individuals just as I am, and could benefit from some additional exposure.  




Tinta-e (from Spain! Translated to English)



The Digital Reader

The Nerd Insurance

MobileRead Forums (these guys are great.  I got the Launchpad script I used to hack the Kindle on a different thread on this site.)

Richard's Notes (personal blog)

I also sat for a couple interviews, so be on the lookout for those stories to be published soon.  I’ll post links when they’re up. 

Stay tuned for an update on the project itself, where I will attempt to answer some of the questions posed by comments to the above linked stories.  There are some fantastic discussions going on in this niche market, and it’s exciting to be a part of it. 

Finally, we also received our first donation!  I’m not sure if it was intended to be anonymous so I’ll err on the side of caution.  You know who you are, and we thank you.  Just to recap, we accept donations through PayPal or Bitcoin (information on the sidebar).  If you want to help but don't have / don't want to send the cash, just have a look at the ads on the side of the page.  Every little bit helps!

Thanks again for taking the time to follow my little project. 

Sunday, August 21, 2011

Frankenkindle prototype demo


That's right, it's alive!

The Frankenkindle is alive, and fully functional.  The 5-way keypad (up, down, left, right and center) and 'Home' have all been implemented with a substitute keyboard.  (Also,if you turn up your speakers you may hear Queen playing Bohemian Rhapsody in the background - don't say I never give you anything.)

Functionally it's complete, but there's still a fair bit to be done to make it more robust.  The Kindle itself needs to be semi-permanently mounted to the front panel rather than just resting on two wood screws.  The circular interface board along with the flat cable connecting it to the Kindle are both quite fragile and will need to be covered.  Finally, the cables will need to be routed in a more organized fashion.

Here are some pictures to better illustrate what you saw in the video.  Keep in mind that this is a prototype if ever there was one.  It's not pretty, but it doesn't have to be.  It's functional.  Once my sister has a chance to play with it I'll be able to take some good notes on what works and what doesn't, and fold them into the next revision.  Right now I'm designing for what I think she'll need.  The only way to find out if it's useful to her is to actually turn her loose on it and watch what happens.

 The disassembled Frankenkindle.  Front panel, Kindle, and control board are all visible.



Side view, showing the circular board used to interface the main control board
to the Kindle's keyboard input.



 Rear view, showing the control board.  


Power is supplied from a 5V "wall wart" style power supply and the 9-pin serial cable is used to provide an  easy quick-disconnect for the new front panel buttons.

Closeup of the main control board.  

5V DC power is fed through the black cable and used to power the Teensy controller directly (the long green board with USB connector).  It is then fed through a series of diodes to provide a sexy red power LED and stable 3.3V for the multiplexers (actually closer to 3.5V, but who's counting?).  The multiplexers are the surface-mount chips on the red breakout boards.  The bank of resistors in the upper right form a series of voltage dividers, useful for dropping the 5V control signals from the Teensy into 3.3V signals suitable for use with the multiplexers.

And there you have it!  More pictures will be posted to document progress on the enclosure itself.  Also, be on the lookout for a full bill of materials (BOM) and more formal documentation.

Thanks for reading! 

Saturday, August 20, 2011

Rookie mistake

We all do it, and we all hate to admit it.  But every now and then a facepalm-worthy event occurs that just can't stay hidden.

A few weeks ago I finished rewiring the main control board on the Frankenkindle, adding quick-disconnect cables and fitting it into the crude wooden stand I made.  Everything looked good, but it stubbornly refused to actually control the kindle. 

I tested everything.  I checked the code, made special modified subroutines for testing, played with the timing, checked continuity on every connection on the board.  Then I did it again.  Then (you guessed it), I did it again.  This worked on the breadboard, so what's different about my new wiring?

Then it hit me.

I came back to this blog post wherein I explained the keyboard connector on the back of the Kindle itself.  Sure enough, there it was in black and white:  "As seen in this picture, pin 1 is at the bottom, and pin 20 on top."  It seems in my infinite wisdom I had failed to document this fact anywhere else in my notes, and wired the Kindle connector board in reverse.  As is usually the case, it's always the simplest things that trip us up. 

I haven't facepalmed that hard in a while.  It was so bad I had to share. 


Thursday, August 4, 2011

It's a girl!

I just wanted to stop by and explain the painful lack of updates recently. 

Short version: it's my wife's fault.  :)

We're expecting a baby, and have been seeing doctors and attending all sorts of classes to prepare for having our lives turned upside down in a little over a month.  Zoey Kristina is set to arrive September 17th, so you can imagine that the past few months have messed with my normal schedule a bit.  As those of you with kids know, this disruption is likely to only get more pronounced in the coming months when she gets here. 

I am still working on the Frankenkindle, albeit sporadically.  I wire-wrapped the whole circuit onto a perfboard, tore it apart and rewired it all over again, then mounted everything onto a hastily constructed stand.  At this point all the pieces are in place, I just need to squash a few small bugs in my code and triple-check the board's wiring.  It shouldn't be long now until I have a working demo to show you.

So, thank you for continuing to follow this project.  I assure you it's not dead, just don't expect updates as often as before.

See you soon!

Wednesday, July 6, 2011

No more Amazon Associates referrals

I live under a rock, so I didn't find out about this until just the other day when Amazon sent out an e-mail blast to all their California associates, informing us that they are severing ties and will no longer offer referral fees.

This is in response to a bill that our esteemed governor just signed that will require the state to collect taxes on internet sales.  Or something to that effect.  I could learn more about it but honestly I don't care.  Politicians will do what they do best - nothing constructive.

So why am I writing about this?  Because it means one potential source of income (and a mighty convenient one, at that) has just dried up and gone away.  As of now, the only financial way to support this site is with a direct donation (we still accept kindly worded comments if you're running short on cash). 

In the sidebar you will see donation links.  As of today we are also accepting donations in the new digital currency known as BitCoin. 

I just wanted to thank you for taking the time to follow this project.  Donations or no, it is the enthusiasm and creativity of people like you that help make the world go around. 

Enough talk - let's go void a warranty!

Tuesday, June 21, 2011

By Request: Kindle 3 Keyboard Map

In a previous post, loyal reader Blaketh requested that I post the keyboard map information I'm using for this project.  I immediately broke a promise to post it in the following few days, so I'll try to make up for it today. 

First, let's look at the connector itself, so you can figure out which end is up.  With the rear cover removed, you can see the keyboard connector in the lower-right corner of the main board.  It's a thin white FFC (Flat Flex Cable) type, and has 20 very small pins. 


As seen in this picture, pin 1 is at the bottom, and pin 20 on top. 

The keyboard itself is nothing more than a bunch of dry-contact switches, which is to say there aren't any sensitive electronics to worry about damaging or cloning.  To simulate a keystroke, all you have to do is short two wires together.  The trick is figuring out which pairs do what...

For your consideration, I present the keyboard map of the Kindle-3. 

(Format: (Short this pin)  (to this pin)  (to generate this keystroke) )


1 7
Menu
1 8
Home
1 9
Back
2 12
Q
2 13
W
2 14
E
3 7
R
3 8
T
3 9
Y
3 10
U
3 11
I
3 12
O
3 13
P
3 14
A
4 7
S
4 8
D
4 9
F
4 10
G
4 11
Home
4 12
J
4 13
K
4 14
L
5 7
Del
5 8
Z
5 9
X
5 10
C
5 11
V
5 12
Back
5 13
N
5 14
M
6 7
.
6 9
Enter
6 10
^ (shift)
6 11
Alt
6 12
Space
6 13
Aa
6 14
Sym
20 15
FW_UP
20 16
FW_DN
20 17
FW_LEFT
20 18
FW_RIGHT
20 19
FW_CENTER

A simple pushbutton can be used to mimic the stock Kindle keys. 

One word of caution - the auto-repeat delay is very short.  You don't notice this on the stock Kindle because those tiny little buttons don't maintain contact when pressed.  I won't pretend to know much about their internal structure, but I can tell you that a standard button will generate a lot of duplicate keystrokes in rapid succession if held down or even pressed slowly.  In my case I'm using a microcontroller to handle all the timing so it's not a big issue.  However, if you're just cloning the keyboard with no additional electronics, you might want to investigate a debounce circuit of some sort or you'll get a lot of duplicate keystrokes.

That's it!  Once again, thanks for reading!

Tuesday, April 26, 2011

The FrankenKindle - It's alive!

Every project needs a good name, and given the nature of this particular endeavor I can't think of anything that would top The FrankenKindle.  Now all I need is a theme song...

This week marked a huge milestone for the project.  The two large page turning buttons on the V.Reader device were used to control the page turn functionality of the Kindle 3.  And, wonder of wonders, it worked!

Here's a quick rundown of the components in the video (skip to the end to see the actual test).

(WARNING: Nerdspeak ahead.  Skip to the bottom if you value your sanity.  You've been warned...)

The V.Reader

We'll be using seven buttons in total from this device, although for this test only the two large arrow buttons on the front panel were used to turn individual pages.  They are simple dry-contact switches, which means that when they're activated it's basically the same as touching two wires together to make a connection.  Two wires were used per switch, and are routed back to the microcontroller on the breadboard.

The controller

For simplicity's sake I chose to use the Teensy++.  It's a USB-powered device based on an 8-bit AVR microcontroller, which allows easy connection to the PC for programming and also provides a convenient 5V supply to the rest of the circuit.  In addition, it is compatible with the Arduino environment which dramatically simplifies the programming phase. This particular unit is overkill for this application but we're hip-deep in the prototyping phase here so the emphasis is more on making it work than making it efficient.  That will come later.

The multiplexer

This is the multiplexer I chose, mainly because it's cheap, intended for digital signals, and has a breakout board already.  If multiplexers confuse you, just think of one of those old-school telephone operators sitting in front of a giant switchboard.  They take a jumper cable and connect two signals together so they can communicate.  That's exactly what happens here.


Let's look at the left-hand circuit for starters.  The short version is that pin 7 (called "1D") may be connected to any one of the four pins above it, pins 3, 4, 5 or 6 (1S4, 1S3, 1S2, or 1S1, respectively).  The signals IN1 and IN2 control which connection is made, and 1EN and 2EN enable it.  The EN signals allow the connection circuits to be handled independently, basically making this into a dual single-pole, quadruple-throw switch.  Each chip, then, should allow up to eight distinct connections to be made.

That's fantastic, except the breakout board supplied by SparkFun doesn't break out the 1EN and 2EN pins, so any connection made on one side of the chip is mirrored on the other.  1D and 2D are still distinct, but they cannot be controlled independently.  If 1D is connected to 1S3, then 2D must also be connected to 2S3.  This limits us to only four independent connections per chip, rather than eight.  (Yes, the two switch systems are electrically isolated, but since it is not possible to independently control them, half of the switch becomes useless for this application.)

Further, since I can't enable or disable the connection I must reserve one combination as a "home" and make sure it's not used.  The reason for this is simple - each connection simulates a key press event on the Kindle.  We must have a state where no keys are pressed, and the only way to do that with this chip as SparkFun sells it is to reserve one of the switch combinations and not connect it to the Kindle at all.  As such, out of eight possible connections we are only able to use three.

Fortunately I decided to buy spares, so I should have enough on hand to complete the first run of the FrankenKindle.  Eventually I'd like to make this project more user-friendly however, and this issue will need to be addressed.  It will likely require creating a custom PCB for the project.

The software

As detailed in a previous post, two hacks were used to improve the functionality of the Kindle for this application, Jailbreak and Launchpad.  When a key is pressed on the V.Reader, the Teensy sees that and uses the multiplexer to mimic pressing the specific key sequence to fire the Launchpad script, which will turn an individual page.  (Whew!)  This same procedure could be used without the software hack, but it would only be able to jump between chapters, not individual pages.

(This concludes the nerdspeak - welcome back!)

The test

Enough talk - here's a short video showing how it works.  Apologies for the horrible sound quality.

Tuesday, April 19, 2011

Successful keyboard test!

They say a picture is worth a thousand words.  I'm not sure who "they" are, and I imagine this word value is up for debate, but the idea is nice.

This video is 2:16, or 136 seconds in length.  At 25 frames per second (I have no idea if that's the case, but it's what Youtube recommends), that's 3400 words.  Don't worry, I won't put you through that.  I just want you to know how valuable this video really is...


Thanks to the great folks over at the Mobileread.com forums, it is now possible to remap hotkey combinations and individual keys to just about any function we could need.  In this case the 5-way Left and Right keys have been paired with the Shift key to allow them to turn individual pages.

One unexpected benefit of this technique is that the keys retain their original functionality if used without the Shift key.  That is, pressing the Right-side 5-way button will jump to the next chapter in the book.  But pressing Shift and then the Right-side 5-way button will advance only one page.

That's great, but the goal of this project was to be able to do this without using the Kindle's onboard keypad at all, instead using a custom keyboard with big cushy buttons.


This video shows a successful test of the system.  There are some very subtle timing issues that will need to be ironed out, and it now becomes clear that an intermediate controller will be required to handle the key sequencing.  This entire process should be transparent to the user; a single keystroke should advance a page. Fortunately there are a host of very simple controllers on the market that will be able to accomplish this task.

We've still got a long way to go, but this was an important milestone.  We've shown that the Kindle can be controlled externally, using whatever buttons we may find convenient.  Once the sequencing and timing issues are resolved, we can move on to actually building the custom keyboard using buttons harvested from the V.Reader.

See you next time!

Monday, April 18, 2011

Success!

I have not yet begun to void your warranties, BestBuy!

I thought my warranty shredding activities would be limited to popping off the rear case and tapping into the keypad's signal connectors.  Nope, it turns out the actual Kindle operating system needs a good dose of warranty-voiding hacks as well.  Fortunately the Great and Powerful Internet (henceforth known as the "GPI") came through once again.  Thanks to the aforementioned hacks (jailbreak, launchpad), I have successfully created hotkey sequences to allow the user to turn individual pages from the main keypad. 

The simple keystroke combination Shift + Right and Shift + Left (that's the left and right keys on the 5-way) will now allow the user to turn individual pages.  Interestingly enough, pressing the Right and Left buttons without first pressing shift retains their original function: jumping between chapters.

This is a huge success.  Proper documentation will follow in a later post, when I'm properly rested and have a chance to produce some decent pictures and maybe a demo video.  For now, it looks like the biggest hurdle so far has been overcome.

Thanks for reading!

Sunday, April 17, 2011

If Engineering were easy, anyone could do it!

No Engineering endeavor goes smoothly every step of the way, and this one seems no different.  After surprising success with the main keypad a couple weeks ago, it quickly became clear that the easy part was over.

It turns out that the main keypad and the individual page turn keys on the device edges use signals which are routed through two different connectors on the motherboard.  The 20-pin FFC (Flat Flex Cable) connector handles the main keypad signals, while the individual page turn keys on the edges are routed through two VERY small connectors, one 6-pin and the other merely 4. 

Pictures will follow in a later post.  For now though, the moral of the story is that tiny connectors are incredibly difficult to deal with.  The creation of a breakout board similar to that of the 20-pin main ribbon cable proved to be a spectacular failure.

So what alternatives do we have?  With the existing physical hack, we're only able to advance between chapters, not individual pages.  It seems clear that the physical keyboard alone won't be sufficient to complete this project.

Fast forward a few hours.  Some quality alone time with Google and more than a few newbie questions posted on www.mobileread.com and a possible solution is found.  Jailbreak the Kindle and install another software hack called Launchpad.  (Here is the page I used for the jailbreak operation, which pulls together information from a few different mobileread.com forum posts and presents it in a nice step-by-step instruction guide)  This should allow the keys to be mapped to just about any function we wish.  The idea will be to map the left and right keys on the 5-way to force them to turn pages, not chapters. 

So far the Kindle has been jailbroken and it's not (yet) a smoking ruin.  The road got a lot longer this week, but we're making good progress.

Stay tuned for pictures and detailed information about the software hacks. 

Sunday, April 3, 2011

Keyboard map complete!

This weekend's mini-project was simple - use a physical pushbutton like this to mimic the behavior of every key on the Kindle 3's main keypad.  The result?  Success!

Of course as we all know, every silver lining has a cloud.  It turns out that the D-Pad (called a "5-Way" in most Kindle documentation), while useful for menu navigation, cannot actually be used to turn a page. 

You read that right.  The Kindle offers pairs of oversized buttons on halfway up either side of the unit for turning pages but does not duplicate this functionality in the 5-Way connector they offer on the main keyboard.  These dedicated keys are routed to the motherboard through their own connectors.  It looks like I'll have to clone this second, much smaller connector as well.

Back to Mouser, and it looks like the smallest FFC (Flat Flex Cable) connector they've got has 6-pins, which matches nicely with one of the small connectors on the side of the Kindle.

The 6-pin page turning keypad connector
 marked by the upper-right arrow

Once the dedicated page turn keys are cloned, this phase of the project will be complete.  

Phase II begins with an organ transplant.  The main keypad from the V.Reader unit referenced in an earlier blog post will be removed and wired into an adaptor that will allow it to be used in place of the Kindle's own keypad.  This adaptor will also need to be created.

Phase III, hopefully the final phase, will be the creation of a simple stand for the assembly.  It will need to position the Kindle in such a way that the screen is easily visible, but will also need to present the custom replacement keyboard so my sister can use it.  Finally, the entire assembly will need to be physically secured to a desk or wheelchair tray to provide stability.

Thanks for reading!

Tuesday, March 29, 2011

Keyboard Test

It's amazing what you can learn when you spend a few hours with a multimeter and a breadboard.  

The Kindle's keyboard does indeed use simple tactile switches as I had initially expected, but the wiring scheme was not very straightforward at all.  For this test I set up a series of tactile buttons from SparkFun to mimic the behavior of the Kindle's directional pad.  Up, Down, Left, Right, Enter (select) and Home were wired.  

The sample keyboard wired and ready for testing.

The test was a success, in that it proved that the keys can be cloned with simple switches.  Nevertheless, it illustrated a few more nuances that will need to be fully explored.

First, the keys auto repeat.  Holding a key will continually execute the given function, and the delay is surprisingly short.  I routinely advanced multiple lines vertically or multiple chapters to either side (yes, chapters - we'll come back to that in a second).  This means that we will probably need to use some sort of microcontroller to handle the timing.  If I'm getting unwanted repeats, how hard will it be for someone with a disability?

The second issue actually involved the left and right keys.  For the purposes of this project it is desirable to use the D-pad instead of the edge-mounted ">" and "<" keys for advancing to the next page, for the simple reason that the edge-mounted keys are routed to the motherboard through an additional, very small 4-conductor cable.  It's the same type of cable as the main keyboard connector, but it would be another part to solder and wire.  It would be far simpler to just use the D-Pad. 

Fortunately the D-pad's left and right buttons can be used in place of the standard keys on the side of the unit.  During testing however, I discovered that my cloned replacements advanced by an entire chapter with each click, not just one page.  A quick Google search for keyboard shortcuts suggests that the Alt key can be used to cause this sort of "fast forward" effect.  Alt+Right will advance to the next chapter (or annotation), and Alt+Left will back up to the previous one.  By pressing my Left and Right keys alone, I jumped by a full chapter, implying that the Alt function was automatically engaged on a signal line that I haven't yet wired.  This should be ironed out in a later test when I'm able to simulate more of the keyboard.  This test was just intended to show that the keyboard can, in fact, be cloned.

Finally there is a very subtle electrical issue that will still need to be addressed, but more research is required to properly document it.  Stay tuned.

You can see the keyboard test here.  Bask in the glow of my amateur film making skills.  Try to contain yourself.



Friday, March 18, 2011

Christmas in March!

It's always a good week when you get packages from both Mouser and SparkFun in the space of two days.  My Hirose connectors and ribbon cables arrived, along with a breakout board from SparkFun.

Remember the goal of this project, to create a custom keyboard for use with a standard Kindle.  The Kindle is a fantastic design, but only if you have small fingers or superior motor control.  For people with certain physical disabilities, this device is completely unusable.

Today we'll begin creating the physical adapter board that will allow us to solder buttons and such together to create a custom keyboard.  We'll worry about the actual layout and physical appearance later.  First we have to prove that this is even possible.  The first task is to gain access to each of the tiny conductors in the Flexible Flat Cable (FFC) on the Kindle motherboard.

The Hirose connector used on the Kindle is a 20-pin type, with 0.5mm pin spacing.  The only breakout I could find readily available is an 80-pin TQFP type, for use with certain microcontrollers.  It's a great design, with duplicate pads on both sides.  There are four possible mounting locations for a connector of this type on each side of the board, which means I can screw up the soldering 7 times and still not be out of luck.

I really only need one connector.  I bought six.  Turns out three was the magic number (surface mount soldering is tricky!).

Thanks to the surprisingly handy magnifying ring lamp my Dad got me back in college, I was finally able to secure the connector to the breakout without shorting any of the leads.  A quick continuity test with my multimeter showed that each pin was securely fastened to the board.

A couple scorch marks never hurt anyone...

Flat Flexible Cable (FFC) test fit

Now that we've got the connector attached to a suitable (albeit oddly shaped) breakout board, let's do one final sanity check.  We need to make sure the other end of the cable actually fits into the Kindle's socket.  This is the acid test, and will prove whether or not we actually bought the right connector.

It fits!

Whew!  That's a relief.  The cable fits securely into both the Kindle connector and my hacked-together breakout board.

That's enough for today.  Now that we have access to the keyboard, we can begin testing to figure out exactly how the buttons work.


Saturday, March 12, 2011

Kindle Teardown

I went to Best Buy and played with the Kindle in preparation for this project.  The first thing I did was to turn it over and look for screws.  To my horror, there weren't any.  I left empty handed and immediately went to my old friend The Internet.   Approximately 14 seconds after I started my search, I'd discovered two excellent videos showing me exactly how to take the Kindle apart.  Well, I should clarify that...  I knew I wouldn't have a problem taking it apart, the trick would be to make sure it still worked when I put it back together again.

Two site of note here, both are recent favorites of mine.  In no particular order, here they are:

www.eevblog.comThis video was the first I found, and was an excellent tour of the Kindle's guts.  Dave demonstrates exactly how to tear the unit down, and even offers some insight into its construction.  This was an excellent find, and I plan to spend a lot of time here going forward.

My only issue with the EEVBlog video above is that he didn't take the front panel apart to look at the keyboard.  That was the whole reason for this project, so I bookmarked his site and continued my search. 

www.tested.com:  This site is chock full of product reviews and teardown videos, and I quickly found the one I was looking for.  A complete teardown and reassembly video of the 3rd generation Kindle.  He didn't spend much time on the keyboard but he did completely disassemble the unit and spent some additional time on the disassembly procedure (especially as it relates to the screwless case!).  With this information I felt confident that I could completely tear down my Kindle without destroying it. It would be up to me to figure out the keyboard.

It should be noted that this site is not in any way affiliated with the above blogs.  They were an excellent resource for this project however, and I would be remiss if I didn't list them here.  Anyone trying a similar project as mine would find their information incredibly useful.


Enough talk!  Let's void a warranty.

The 3rd-Generation Kindle is a great piece of engineering.  It's much smaller than I expected it to be, and internally it's laid out in a way that allows surprisingly easy maintenance.
This Kindle is at the top of the button transplant list.

I've never used another eReader, so I don't have much to compare it to.  That said, I love it.  My sister loves to read but has trouble holding a book and turning the pages.  This will be exactly what she needs, we just need buttons she can work with!  

When I bought the Kindle the clerk at Best Buy very politely listed my options for extended service plans.  I had to stop myself from laughing out loud.  If only he knew that the first thing I'm going to do is take it home and void the warranty.  Let's get to it!

Challenge accepted.

The first thing you might notice is the fact that the back of the Kindle has no screws.  It's like a big sign saying "I dare you to crack this case open."  What do we say to a challenge like that?  "Watch me."

The case simply snaps together.  All you need to get it apart is a flat bladed screwdriver.  Insert the screwdriver into the seam that runs around the perimeter of the unit, pressing outward.  Once you're able to insert the screwdriver under the rear case, rotate it parallel to the table to pry it apart.  With a little luck this will open the case slightly, and you can move down a bit to the next spot.  Repeat the process all the way around and remove the case.

When the back case comes off, take a moment to look around.  Most notable here is the surprisingly large battery.  It is held in place by two screws, which are easily removed.

Battery screws, begging to be removed.

Once the battery is removed, it's time for the motherboard.  There are two sizes of screws used in the assembly of this device.  Those used to attach the battery are the larger of the two, and there are quite a few more on the frame, the 3G modem and the connector for external lights and such.  

Three screws hold the 3G modem (or in this case a plastic dummy) in place.

The rest of the large frame screws.

Once the silver screws are removed, it's time to get out the magnifying glass and tweezers and start in on the tiny little black screws.  There are 11 of them, and they actively try to hide from you. 

It's like playing "Where's Waldo?", but smaller and more expensive.

 With all the screws removed the motherboard is left floating, but it's not quite time to remove it yet.  There are four ribbon cables connecting all the front panel buttons to the motherboard, along with one miniature four wire cable with a JST-style connector for the speakers.  This connector is at the upper right corner of the image below, just to the right of the dummy 3G modem.  The red and black cable can be removed with a flat bladed screwdriver, or with your fingers (provided yours are smaller than mine!).  Three of the ribbon cables have "zero insertion force" (ZIF) sockets, and can be removed simply by flipping the top door of the connector and easing the cable to the side.  The last ribbon cable (bottom left) uses a different socket, and is a simple press fit.  Lift it straight up to disengage.


Four ribbon cables connect the front panel keys to the motherboard.

Once these have been successfully removed, the motherboard is floating free and can be removed.  Slide it toward the top of the Kindle slightly to free the connectors and switches at the bottom of the case, and lift the entire assembly up and away.  

Motherboard removed.  This is the back of the keypad.

When the motherboard is removed, you can infer the purpose of each ribbon cable removed in the previous step.  the two smaller cables are routed to the side buttons for turning pages.  The large ribbon cable on the lower right is routed to the main keyboard, and the final cable on the lower left handles the eInk controller for the display.

The front panel, from behind.  The display controller cable is clearly visible.

Once the keyboard is removed, the front panel is visible from the rear.  There are a few things to note here.  First, the ghost image on the eInk panel.  This type of display retains whatever was last displayed when power is removed.  Second, the keypad is a simple strip of silicone rubber, with small plungers on the rear of each key.  This type of keypad typically uses strips of conductive rubber on the back of each button to close a connection between two exposed traces on the circuit board underneath.  In this case however, the plungers are not conductive.  The assumption was that we would find pairs of exposed PCB traces to which we could solder wires and hijack each button.  Houston, we might have a problem...

The keyboard itself.

Looking at the keyboard itself, it is clear that we'll need a new approach.  There doesn't seem to be any simple way to connect wires to these buttons and have the user close the circuit by pressing something on the keypad.  

Unless...  This keyboard is connected to the motherboard via a 20-connector ribbon cable visible at the lower left of the image above.  Let's have a look at the connector to which it mates.

Flat ribbon cable connector for the keyboard.

This cable is the only connection between the main keyboard and the motherboard.  The signals from each key must be carried in this cable.  It's a 20-pin connector but I count 42 distinct keys on the front (not counting the four side-mounted buttons for turning pages.  This cable is still our best bet for intercepting the keyboard commands and injecting our own, but it's going to take some tinkering to figure out the signals.

First, let's have a look at the connector.  Flipping through the new Mouser catalog that conveniently arrived yesterday, it looks like this is an FFC (Flexible Flat Cable) connector from Hirose.  The arrow in the picture below confirms that the connector is a 20-position variety.

 
Closeup of the keyboard connector.

The closest match in the catalog appears to be the 798-FH28E20S05SH05.  

So, what have we learned?  The initial idea of simply soldering wires to the keyboard is out.  It's not feasible with this particular button style so we need a new approach.  We know that the keyboard signals are wired through this connector and cable back to the motherboard, but that cable is far too small and flimsy to solder to directly.  Fortunately it looks like we got lucky with Mouser and just might be able to find a suitable mating connector of our own.  

The next phase will be the creation of a suitable breakout board for that cable.  Once we're able to wire to all 20 signal lines, we can begin hacking the keyboard.  The goal remains the same: substitute our own buttons for the tiny factory keyboard.  

Time to hit www.mouser.com to order some connectors.