So, now that I have a general idea of how the Beyond 2000 Plus works, I have to make a decision on how I’m going to replicate its functionality.  What I’ve done in the past with radio controlled cars is just tap into the outputs of the microcontroller and control the stock h-bridge directly with the Arduino.  The problem here though is that space is a concern.  I have a feeling that no one will want to use a sex-toy with a huge controller.  I suspect from looking at the design that some of the problems I had with the original was noise and stray voltage coming from the motors resetting the micro.  If I build my own controller and h-bridge from scratch, or at least used a better designed one, I’ll hopefully be able to avoid these problems.

I have a suspicion that vibrator use is somewhat similar from use to use (ie: vibration starts off low and increases in intensity over time).  If I was able to log (to an EEPROM perhaps) changes made to the levels and modes of the vibe over usage time, I may be able to more or less accurately reproduce those changes on successive uses.  Of course, these changes would be merely suggestions that could be changed at the will of the user, and those “overrides” would be logged and incorporated into the system’s “knowledge”.  I suspect that an averaged reproduction of common usage would have a very subtle but pleasing effect.  Imagine a vibrator that predicted your whims based on previous usage.  Seems the next logical step in adult themed technology!

So, 5 buttons, some sort of L.E.D. display to indicate motor speed/mode, a dual h-bridge, and an Arduino to control it all should be a good start.  I’ll add in other components later, when I’m sure the basics are working.

I could see clearly, even before I opened it up, what looked like 4 transistors on the outer two sides of the control section of the Beyond 2000+ and instantly suspected a dual H-bridge design.  I figured with the complexity of the ‘modes’ the whole thing would be powered by some kind of microcontroller.  Exploring how the microcontroller in the Beyond 2000 Plus controlled the h-bridge to drive the motors and also how it drove the LED display seemed a worthwhile endevour.  A group of people, smarter than I am, most likely spent a great deal of time designing how this thing was going to work and if I’m going to try to replicate it, I will probably gain some useful info from their finished product.

True enough, the circuit design powering the motors was a dual H-Bridge with the 2 sets of 4 transistors lining the edge of the board.   Transistors are PNP and NPN silicon power transistors: B772 and D882 respectively.  Datasheets: NEC B772 , NEC B882

Looks like they have 4 catch-diodes to deal with voltage spikes from the motors.  These are very common small signal diodes, possibly 1N4148.  These can handle current up to 200mA, which seems kinda small for the motors we’re using and the spikes that could be expected.   This may be why it would shut off periodically.

Notice the 16V 470μF electrolytic capacitor?  Assuming this is to help smooth out voltage drops when the motors run.

The microcontroller is labeled “Pilot Boat” and “PB35 0301″.  I couldn’t find any information on it online, I’m assuming it’s not custom, but I haven’t sniffed out what it is exactly.  It’s a 28pin SMD design.

The motors run at different speeds, so there is obviously some kind of PWM at play here.

The LED array seems to be arranged in a matrix display requiring what I think is only 8 pins to control the 16 LEDs.  Some kind of ‘plexing going on here.

Motors are the simple brushed DC type and have been measured to run peak at 1A for the vibration motor and a little less for the shaft motor.  Neither motor has a filter capacitor attached.

Batteries are 4 AA cells in series producing 5-6v of power depending on type and charge.

There seems to be no voltage regulation whatsoever, so everything runs straight off the batteries.

Buttons are that rubber type material found on many remote controls and have conductive circle backings that bridge exposed PCB traces.

The next installment will take what I’ve learned here to help design an home-brew solution.  Stay tuned…

Please see the previous post for the motivation behind this one…

First step was to disassemble the intended target. I was excited to get into the thing so I forgot to take some “before” shots. This internet image will have to suffice. The Beyond 2000+ is made almost entirely out of plastic with the shaft and stimulator covered in a sheath of some kind of jelly or silicone. The shaft consists of a plastic tube, which contains the motor used for rotating the beads and head, and supports the bead and head structure. Within this white bead/head structure, I suspect there is a gear mechanism for increasing torque, but I haven’t taken it apart to confirm. The tube and sheath were simply glued onto the controller base and it all came apart fairly easily. The glue was weak, and a little leverage pealed away the jelly cover and the plastic tube with little damage. With a little careful stretching, the jelly simply slid off and the two mechanical structures slid right out.

It’s my hope to keep the shaft/bead section as well as the jelly cover for use in the home-brew version.  The next entry will take a closer look at the Beyond 2000+’s control unit.

The bottom line of all this information is this:  For a product that was very expensive, it didn’t work very well.  It malfunctioned constantly!  It would cut out at inopportune times, turn on at even more inopportune times (ie: inlaws are visiting), and don’t even think about using it in a colder than average environment.   I don’t know if you’ve ever tried to return a sex toy; for some reason retailers frown upon it…  Suffice to say, a replacement didn’t fare too much better either.  I’m not sure there is anything sadder than what would be an excellent grown up toy that remains unused in a bedside drawer…

So, when something isn’t working right, and you can’t get it fixed or get a refund, then there is absolutely no loss in opening it up to check out what gets its motors running.  It is my hope to delve into the innards of this device, reverse engineer as much as I can, and determine if I can emulate it’s functioning with my own hardware running Arduino code…  I’ll post my work on this blog and I guess we’ll see how it goes!

LEDFrameReprogram.pde

//  Automatic LED License Plate Frame Reprogrammer
//  IRremote library by Ken Sherriff

#include <IRremote.h>
IRsend irsend; //setup irlibrary

const int buttonpin = 12;
const int codearraylen = 4;  //how many characters in message

const unsigned long messagecode[codearraylen] = {0xFF40BF,0xFF22DD,0xFF18E7,0xFF7887}; //Message - Codes for P W N D

const unsigned long startcode[4] = {0xFF926D,0xFF12ED,0xFFB04F,0xFF629D}; //Codes for Edit, Enter, Clear, Y
const unsigned long endcode[1] = {0xFF12ED}; // Code for Enter to finish message
const int betweendelay = 100;

void setup()
{
 Serial.begin(9600);
}

void loop() {
 while (digitalRead(buttonpin) == HIGH){
 Serial.println("Running");
 // send initial setup codes - ie: Edit, Enter, Clear, Y
 for (int i = 0; i < 4; i++){
 irsend.sendNEC(startcode[i], 32);
 delay(betweendelay); // delay between individual codes
 }
 // send message codes
 for (int i = 0; i < codearraylen; i++) {
 irsend.sendNEC(messagecode[i], 32);
 delay(betweendelay);
 }
 // send finishing codes - ie: just Enter - might want more here later
 for (int i =0; i < 1; i++){
 irsend.sendNEC(endcode[i], 32);
 delay(betweendelay);
 }
 }
}

I sampled the remote’s buttons using the library’s receive function so I could use the values in the above code. The sampled codes are as follows:

IRLibrary_LED_Frame_Codes.txt

IRLibrary E-Plate Codes

Edit FF926D
Enter FF12ED
Shift FF50AF
CAP FF906F
Clear FFB04F
ON/OFF FF38C7
RUN FFF807

Insert FFD22D
<- FFD02F
-> FF10EF
INV FF28D7
SPD FFE817

A FFBA45
B FFFA05
C FF3AC5
D FF7887
E FF7A85
F FFB847
G FF1AE5
H FF58A7
I FFDA25
J FF9867
K FFD827
L FF9A65
M FF5AA5
N FF18E7
O FFC03F
P FF40BF
Q FFC23D
R FF807F
S FF02FD
T FF827D
U FF42BD
V FF00FF
W FF22DD
X FF609F
Y FF629D
Z FF20DF
! FFE01F
? FFA25D
# FFE21D
$ FFA05F

1 FF0AF5
2 FF8A75
3 FFCA35
4 FF8877
5 FFC837
6 FF48B7
7 FF4AB5
8 FF08F7
9 FFAA55
0 FFEA15

Hardware is consistent with my previous post. All that’s left is figure out how to amp up the IR LED output with a transistor (don’t expect more than a foot of range the way it is).

When I get that working I’ll etch it all onto a PCB.

Stay tuned.
]]> http://themessyworkbench.wordpress.com/2010/08/13/led-license-plate-frame-reprogramming-%e2%80%93-software-success/feed/ 0 tehnick http://themessyworkbench.wordpress.com/2010/08/08/led-license-plate-frame-reprogramming-first-failure/ http://themessyworkbench.wordpress.com/2010/08/08/led-license-plate-frame-reprogramming-first-failure/#comments Sun, 08 Aug 2010 23:59:52 +0000 teknick http://themessyworkbench.wordpress.com/?p=175 Continue reading →]]> After checking out Ken Shirriff’s IR Library for the Arduino, it seemed like an appropriate place to get started building a self-contained LED License Plate Frame reprogrammer. I started with his code found here, just to see if I could get an Arduino to play back a learned signal.  Wired up an Arduino-bootloaded Atmega168 with the necessary components and hooked up a IR sensor and LED as per Ken’s diagram.   And then I met with failure.  The code seemed to look like it was doing what it was supposed to do; it would read and report back with a code over serial, and the LED lit up.  No matter what I did though, it wouldn’t trigger the plate frame.

Out came the IR Toy again to get to the bottom of this!   First I sampled the output of the remote, then sampled the output of the Arduino IR library.  Here is the result.  The signal for the remote is overlayed under the signal from the Arduino.

The Arduino successfully copied and re-transmitted the code, but a tad too quickly.  I don’t unfortunately know enough about the tolerances of IR receivers, but clearly the difference is significant enough to make a difference to the frame.  I also don’t know enough to determine if this is a code problem or perhaps my 16Mhz crystal isn’t oscillating at a very accurate rate.  Will have to explore more to find out!

Edit:  Problem Solved!

According to Ken’s page,

IR sensors typically cause the mark to be measured as longer than expected and the space to be shorter than expected. The code extends marks by 100us to account for this (the value MARK_EXCESS). You may need to tweak the expected values or tolerances in this case.

MARK_EXCESS was increased to 200 in irremote.h and, when tested, the timing from the Arduino matched the timing from the original remote.  And the LED frame responded appropriately. Success!

More to come…

Turn down the resolution of the video if it’s too choppy for your machine!

Another demonstration is here.

I thought I’d make a list of components included so far in the Camera Trigger project.  Keep in mind that things might change at any time and so this list could be updated…  If you have any questions about why I chose something, or what something is, don’t hesitate to leave a question in the comments…

And without further ado, the components list:

• Atmel ATmega328 for brains
• HD44780-compatible LCD – Yellow text on blue background – backlit.
• Quadrature Encoder with switch from Seeedstudio.com
• PS2501 4-channel Opto-isolator (I think from Sparkfun)
• 24LC256 – 256 kbit I2C EEPROM
• 16Mhz Crystal and two 22pF Ceramic Capacitors for a clock input.
• Push button for microcontroller resets
• 0.1μF Ceramic Capacitor for the microcontroller reset pin
• 10K Potentiometer – control LCD contrast
• IR and Yellow LEDs – for infrared transmission and monitoring
• NPN Transistor for LED switching
• Serial Port for connecting to camera cable – Cable is sourced from DealExtreme.
• 2 3.5mm Audio Jacks – for connecting to external flashes
• On/Off switch
• Resistors of various values for LEDs, LCD backlight, the optoisolator inputs, and two 10K for the pull-ups for the reset switch and rotary encoder button.
• LM7805 5v Voltage Regulator
• Electrolytic decoupling capacitors 100uF/25V, and 10uF/25V
• 1N4001 1A/50v rectifier diode
• PTC resettable fuse because I sometimes accidentally try to blow things up
• FTDI 5v Breakout board from Sparkfun (only used for programming the microcontroller – will not be included in final board)
• Lots of wire to be replaced by traces on a PCB when designed and etched…

Once I have the software where I want it, and have worked out any bugs in the hardware I’ll put a schematic up.  The board design will go up as well when it’s completed…

If you’re new to powering electronics, it may be worthwhile to check out Sparkfun’s Power Supply Tutorial.  That’s all for now!

More to come…