Av's Physical Computing Blog

FINALthoughts

2005-12-19T08:45:00.000-08:00

I definitely learned a lot in the process of the labs, the ping pong paddles and the flower. I realize now that I need to learn more about mechanics. I know it was possible to use one or two servos for my flower and after a few instances of servos burning in the future this is the way I will go.

I wasn’t thrilled with the way the flower petals were connected to the gears. It was done in a haphazard fashion, which resulted in the angle and length of the petals being inconsistent.

With the massive amount of ports/parts, labeling everything initially would have saved me lots of time and frustration.

I hope to use a perf board on my next project, it would have been easier to mount and I could have used only one perf board instead of two circuit boards. While the flower didn’t look anything like I expected, that was just the aesthetic end of things. In terms of pcomp, I was satisfied. There was way more involved than I thought and I learned a lot of valuable lessons along the way. Next semester: wearable technologies!

DYEINGflower

2005-12-19T06:45:00.000-08:00

The next piece was to find fabric for the petals. I sewed chiffon onto the aluminum frame of one petal. Although I knew how to sew passably, it ended up being significantly harder and more time consuming that I wanted. Terence suggested using pantyhose. I went to Kmart and got a pair of white ladies XL stockings, cut them up and slipped them over the petal frames and it worked liked a charm.

I cut off the tops of the hose slip cover to get a fringy, flowing look. I wanted to add color to the petals so I dyed them…nylon is quite a resistant material so I used my hands to work in the dye. A messy endeavor, but it worked. I used the same method for the leaves/stems. To speed up the drying process, I used a heat gun…which left some odd but cool patterns on the fabric.

REWIREit

2005-12-19T06:14:00.000-08:00

The next day I came in with the mission to map out my port assignments and rewire the boards so that things could be easily identified and rewired. Although it took quite a few hours, it was worth it.

The PIC had just enough ports for me, so I avoided using a shift register. I did run into problems with PORT A. After consulting the datasheet, ADCON %111100 to turn the ports all digital

All the servos were getting unwieldy so it was time to make a proper stand. I bought some masonite, which I realized once again such a great material. Thin, light and cheap.

I cut two circular sheets of masonite and spaced them with dowels. I mounted the servos and held them in place with wire going through drilled holes in the top circular piece.

The construction of the stand took a full day, much longer than I expected.

SERVOtime

2005-12-19T06:01:00.000-08:00

Now that the clock was working I wanted to integrate the servos with the time. I had only tested with four servos, but with six servos Ben Brown suggested a separate power supply. I went to Radio shack and bought a 6V power supply, stripped the ends and plugged it into a second breadboard to power the servos. Milmoe pointed out the servos and the pic needed to share a common ground.

All this was temporary mounted on a stand I procured from the wonderful shop junk pile. After 13 hours in the lab I called it a day.

CLOCKchip

2005-12-16T14:52:00.000-08:00

To keep the time, I used a real time clock module from sparkfun: http://www.sparkfun.com/shop/index.php?shop=1&cat=45

Todd gave me some code/tips for the chip. It was easy to use, note the three pull up 4.7K OHM resistors. (diagram possibly)

I used serout to check the time and debug my code. [see code]

Basically you program the PIC chip with a subroutine to set the time and then reprogram the chip without the subroutine. If you don’t do this, the clock will RESET everytime you turn power on/off.

Todd pointed me to a way to simulate a clock without the IC. I tried this method, but since the chip used in the example was different than the PIC chip I was using, I decided to use the clock chip.

http://www.picbasic.co.uk/forum/showthread.php?t=2129

For the minutes portion of the clock, I wanted to use 12 LEDs (each representing 5 minutes). My plan for the code was to put all the ports into an array and navigate through the array. Unforunately after much discussion, Tom Igoe confirmed this was not possible. Ben Brown suggested I take the long route and code all the conditions. It’s lengthy, but works.


'****************************************************************
'*  Name    : clock.BAS                                      *
'*  Author  : Avani Patel                                    *
'*  Notice  : Copyright (c) 2005 2004                           *
'*          : All Rights Reserved                               *
'*  Date    : 4/21/2005                                         *
'*  Version : 1.0                                               *
'*  Notes   :                                                   *
'*          :                                                   *
'****************************************************************
Include "modedefs.bas"          ' Include serial modes


TX      var    portc.6                      ' Define serial output pin
SDA    var     PORTC.4                ' I2C data pin
SCL    var     PORTC.5                 ' I2C clock pin
SQW     var     portc.3

RTCData VAR BYTE[8] ' Data byte array



input PORTB.6 'hour set
input PORTB.5 'second set
input PORTB.4 'set time

setHour var PORTB.6
setMinute var PORTB.5
setTime var PORTB.4

RTCSec var byte         ' Seconds
RTCMin var byte         ' Minutes
RTCHour var byte        ' Hours
RTCWDay var byte        ' Weekday
RTCDay var byte         ' Day
RTCMonth var byte       ' Months
RTCYear var byte        ' Year
RTCCtrl var byte        ' Control
DecimalHours var byte
DecimalMinutes var byte
DecimalSeconds var byte
address var byte
dataHours var byte
dataMinutes var byte
dataSeconds var byte

'testLight var PORTB.7     'blink this light at the begining_only happens once
i var byte

'use this subroutine only when programming for the first time
gosub setClock

'LED minutes vars
bSpace CON 200 'between blinks
longBSpace CON 1000 'between rounds
seqSpace CON 2000 ' end of main pause

'PORT SETUP'
'output portb.7  'light once, not neccessary

'set remaining ports to output for LEDs
output portc.2
output portc.3
output portd.0
output portd.1
'output portd.2

ADCON1=%10000111 'makes porta digital

'setting alias for LED ports, all on LEFT side of PIC

p1 var porta.0
p2 var porta.1
p3 var porta.2
p4 var porta.3

p5 var porta.4
p6 var porta.5
p7 var porta.6
p8 var porta.7

p9 var portc.2
p10 var portc.3
p11 var portd.0
p12 var portd.1

                 
potVar VAR word 
potMin VAR Byte 
potLED Var BYTE
potMod Var BYTE

'inputVar var byte
'i var BYTE
'end LED vars

'servo hour vars
TRISB = %00000000 

s1 var portb.7
s2 var portb.6
s3 var portb.5
s4 var portb.4
s5 var portb.3
s6 var portb.2

LotOpening VAR BYTE

pulseWidth VAR BYTE
' set up constants with the minimum and maximum pulsewidths
minPulse CON 140
maxPulse CON 220

openPetal CON 140
closePetal CON 220

'write method for opening and closing

' set up a constant with the time between pulses:
refreshPeriod CON 20

' set an initial pulsewidth:
pulseWidth = closePetal 'start with closed petal stata

LotOpening = 1  ' it's opening


'end servo hour vars
main:

I2CRead SDA,SCL,$D0,$00,[RTCSec,RTCMin,RTCHour,RTCWDay,RTCDay,RTCMonth,RTCYear,RTCCtrl]
gosub bcdtodec        
'Serout2 TX,16468,[HEX RTCMonth,"/",HEX RTCDay,"/",HEX RTCYear,"  -  ",HEX RTCHour, ":",HEX RTCMin,":",HEX RTCSec,13,10]
'pause 1000
goto main

bcdtodec:
'Convert BCD to decimal... assume variable RTCSeconds holds the seconds in BCD format...
'Hours
DecimalHours=RTCHour & $70
DecimalHours=DecimalHours>>4
DecimalHours=DecimalHours*10
DecimalHours=DecimalHours+(RTCHour & $0F)
'convert from 24 hr time to 12
if (DecimalHours >= 13)  then
low PORTB.7
    decimalHours = DecimalHours - 12
else
high PORTB.7
    decimalHours = DecimalHours
endif
'address = 2
dataHours = (DecimalHours * 255)/12

'Minutes
DecimalMinutes=RTCMin & $70
DecimalMinutes=DecimalMinutes>>4
DecimalMinutes=DecimalMinutes*10
DecimalMinutes=DecimalMinutes+(RTCMin & $0F)
'address = 1
dataMinutes = (DecimalMinutes * 255)/60

'Seconds
DecimalSeconds=RTCsec & $70
DecimalSeconds=DecimalSeconds>>4
DecimalSeconds=DecimalSeconds*10
DecimalSeconds=DecimalSeconds+(RTCSec & $0F)
'address = 0
dataSeconds = (DecimalSeconds * 255)/60

'GOSUB subdigiout
gosub minLEDs
GOSUB hrServos

Serout2 TX,16468,[dec DecimalHours, ":",dec DecimalMinutes,":",dec DecimalSeconds,13,10]
Serout2 TX,16468,[dec dataHours, ":",dec dataMinutes,":",dec dataSeconds,13,10,13,10]
return

hrServos:
' change the angle for the next time around:
IF (pulseWidth > openPetal) AND (LotOpening=1) THEN 'if the pulse is greater than 50 
    pulseWidth = pulseWidth-1 ' knock it down
    GOSUB move
    HIGH portd.2
ENDIF

IF (pulseWidth = openPetal) THEN 'when pulseWidth reaches totally open state
    LotOpening=0
    
    PAUSE 1000
     
    pulseWidth = pulseWidth+1
ENDIF    

IF (pulseWidth < closePetal) AND (LotOpening=0) THEN 
     pulseWidth = pulseWidth+1
     GOSUB move
     HIGH portd.1
ENDIF

IF (pulseWidth = closePetal) THEN 'when pulseWidth reaches totally open state
    LotOpening=1
    
    PAUSE 1000
    
    pulseWidth = pulseWidth-1
    
ENDIF 

RETURN

move:
'take the output pin low so we can pulse it high
'LOW PORTC.3
' pulse the pin

if DecimalHours==1 THEN
SEROUT2 portc.6, 16468, ["servo 1", 13,10,13,10]
PULSOUT s1, pulseWidth
ENDIF

IF DecimalHours==2 THEN
PULSOUT s1, pulseWidth
PULSOUT s2, pulseWidth
ENDIF

IF DecimalHours==3 THEN
PULSOUT s1, pulseWidth
PULSOUT s2, pulseWidth
PULSOUT s3, pulseWidth
ENDIF

IF DecimalHours==4 THEN
PULSOUT s1, pulseWidth
PULSOUT s2, pulseWidth
PULSOUT s3, pulseWidth
PULSOUT s4, pulseWidth
ENDIF

IF DecimalHours==5 THEN
PULSOUT s1, pulseWidth
PULSOUT s2, pulseWidth
PULSOUT s3, pulseWidth
PULSOUT s4, pulseWidth
PULSOUT s5, pulseWidth
ENDIF

if DecimalHours==6 THEN
PULSOUT s1, pulseWidth
PULSOUT s2, pulseWidth
PULSOUT s3, pulseWidth
PULSOUT s4, pulseWidth
PULSOUT s5, pulseWidth
PULSOUT s6, pulseWidth
ENDIF

' pause for as long as needed:

PAUSE refreshPeriod

RETURN


minLEDS: 'figures out values for lights
   'potMin=23 'temp constant value
  ' potLED=potMin/5  'to give 1-12 for minutes
 'potMod=potMin//5 'modulus value, to give remindar for blinking LED

    potLED=DecimalMinutes/5
    potMod=DecimalMinutes//5
    
    gosub blinkLEDs
RETURN

'DO NOT CHANGE- this sub works perfect!  ONLY mess with pauses!     
blinkLEDs:

If potLED==0 THEN 'less than 5 min
   SEROUT2 portc.6, 16468, ["in pot=0", 13,10,13,10] 
   
   if potMod > 0 THEN
   for i=1 to potMod
       ' high PORTD.2
        high p1
        
        pause bspace
      
       ' LOW PORTD.2 
     low p1 
     
     pause bspace
     
   next i 
   pause longBspace ' long blink seperate out minutes
   ENDIF
      
ENDIF


if potLED >=1 AND potLED < 2 then '5 mins 
   
   SEROUT2 portc.6, 16468, ["in pot>=1", 13,10,13,10]
   
   HIGH p1
   
   If potMod > 0 THEN
      for i=1 to potMod
       
        high p2
        pause bspace
      
     low p2  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace 
   
endif

if potLED >=2 AND potLED < 3 then '10 mins 
   HIGH p1
   HIGH p2
   SEROUT2 portc.6, 16468, ["in pot>=2", 13,10,13,10]
   
      If potMod > 0 THEN
      for i=1 to potMod
       
        high p3
        pause bspace
      
     low p3  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace 
endif

if potLED >=3 AND potLED < 4 then '15 mins 
   HIGH p1
   HIGH p2
   HIGH p3
   
      If potMod > 0 THEN
      for i=1 to potMod
       
        high p4
        pause bspace
      
     low p4  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace
endif


if potLED >= 4 and potLED < 5 then '20 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   
      If potMod > 0 THEN
      for i=1 to potMod
       
        high p5
        pause bspace
      
     low p5  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace
endif


if potLED >= 5 AND potLED < 6 then '25 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   
      If potMod > 0 THEN
      for i=1 to potMod
       
        high p6
        pause bspace
      
     low p6  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
     
   pause seqSpace
endif


if potLED >= 6 AND potLED <7 then '30 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   
      If potMod > 0 THEN
      for i=1 to potMod
       
        high p7
        pause bspace
      
     low p7  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace 
endif

if potLED >= 7 AND potLED < 8 then '35 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   HIGH p7
   
    If potMod > 0 THEN
      for i=1 to potMod
       
        high p8
        pause bspace
      
     low p8  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace 
endif

if potLED >= 8 then '40 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   HIGH p7
   HIGH p8 
   
    If potMod > 0 THEN
      for i=1 to potMod
       
        high p9
        pause bspace
      
     low p9  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
    
   pause seqSpace
endif

if potLED >= 9 then '45 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   HIGH p7
   HIGH p8
   HIGH p9 
   
    If potMod > 0 THEN
      for i=1 to potMod
       
        high p10
        pause bspace
      
     low p10  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace
endif

if potLED >= 10 then '50 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   HIGH p7
   HIGH p8
   HIGH p9 
   HIGH p10
   
    If potMod > 0 THEN
      for i=1 to potMod
       
        high p11
        pause bspace
      
     low p11  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
    
   pause seqSpace
endif

if potLED >= 11 then '55 mins 
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   HIGH p7
   HIGH p8
   HIGH p9 
   HIGH p10
   HIGH p11
   
    If potMod > 0 THEN
      for i=1 to potMod
       
        high p12
        pause bspace
      
     low p12  
     pause bspace
     
    next i 
    pause longBspace 
    ENDIF 
     
   pause seqSpace
endif

if potLED == 12 then '60 mins     'does not exist  AVANI!
   High p1
   high p2
   high p3
   HIGH p4 
   HIGH p5
   HIGH p6
   HIGH p7
   HIGH p8
   HIGH p9 
   HIGH p10
   HIGH p11
   HIGH p12
   pause seqSpace
endif

'Pause 1000 'this needs to compensate for rest of minute
high portd.2
pause 400
low portd.2
SEROUT2 portc.6, 16468, ["end sequence here", 13,10,13,10]
 
goSub turnLow

RETURN


turnLow: 'turns all LEDs off
PAUSE 1000
'LOW p1
LOW p2
LOW p3
LOW p4
LOW p5
LOW p6

LOW p7
LOW p8
LOW p9
LOW p10
LOW p11
LOW p12

RETURN

setClock:
'Write command
RTCSec=$00 ' Seconds
RTCMin=$08 ' Minutes
RTCHour= $01 ' Hours
RTCWDay=$01 ' Weekday
RTCDay=$19 ' Day
RTCMonth=$05 ' Months
RTCYear=$05 ' Year
RTCCtrl=$10 ' Control preset to output 1 second 'Tick' on SQWpin
I2CWrite SDA,SCL,$D0,$00,[RTCSec,RTCMin,RTCHour,RTCWDay,RTCDay,RTCMonth,RTCYear,RTCCtrl]
Pause 1000

return

FlowerINProgress

2005-11-17T15:02:00.000-08:00

Parts List

2 colors of fabric: inside petals, outside petals

12 servos: 6 small, 6 regular size. Need servos with a bit more torque

12 servo arms

servo mounting tape

12 super bright LEDs: 6 blue, 6 white

2 circular wooden bases to mount servos on

This is what the top view of the piece will look like

Code so far...

Based on some servo testing fun, I determined the angel to open and close a petal.

Then a coded functions to control this. Finally I added a switch so I can bloom and close petals as I wish.



DEFINE OSC 4 'ADC DEFINITIONS
DEFINE ADC_BITS 10 'CHANGE ADCON1 IF 10 BIT OR 8
DEFINE ADC_CLOCK 4
DEFINE ADC_SAMPLEuS 50


OUTPUT portd.2    ' LED debug
OUTPUT portd.1

OUTPUT portb.7    ' servo is on this port
OUTPUT portb.6    ' servo is on this port
OUTPUT portb.5    ' servo is on this port
OUTPUT portb.4    ' servo is on this port

INPUT portc.4     ' openFlower switch
INPUT portc.6     ' closeFlower switch

HIGH portd.1
PAUSE 200
LOW portd.1
PAUSE 200

start:
LotOpening VAR BYTE

pulseWidth VAR BYTE
' set up constants with the minimum and maximum pulsewidths
'  minPulse CON 140
'  maxPulse CON 220

openPetal CON 140  ' minPulse
closePetal CON 220 'maxPulse

'write method for opening and closing

' set up a constant with the time between pulses:
refreshPeriod CON 20

' set an initial pulsewidth:
pulseWidth = closePetal 'start with closed petal stata

LotOpening = 1  ' it's opening

main:
' change the angle for the next time around:
if (portc.4=1) then
high portd.1 ' light at end of petal is on this port
PAUSE 500
LOW portd.1
GOSUB bloom  ' petals open
ENDIF

IF (portc.6=1) THEN
LOw portd.1
PAUSE 500
GOSUB encase  'petals close
ENDIF

GOTO main

move:

' pulse the pin
PULSOUT PORTb.7, pulseWidth
PULSOUT PORTb.6, pulseWidth
PULSOUT PORTb.5, pulseWidth
PULSOUT PORTb.4, pulseWidth
' pause for as long as needed:
PAUSE refreshPeriod
RETURN

bloom:
IF (pulseWidth > openPetal) AND (LotOpening=1) THEN 'if the pulse is greater than 50
pulseWidth = pulseWidth-1 ' knock it down
GOSUB move

ENDIF

IF (pulseWidth = openPetal) THEN 'when pulseWidth reaches totally open state
LotOpening=0
pulseWidth = pulseWidth+1

ENDIF
RETURN

encase:
IF (pulseWidth < lotopening="0)" pulsewidth =" pulseWidth+1" pulsewidth =" closePetal)" lotopening="1" pulsewidth =" pulseWidth-1">

PROJECTObservation

2005-11-10T14:48:00.000-08:00

The real life inspiration:

I went to the New York Botanical Garden and apparently lotuses are not around this time of year. Nevertheless I found ample inspiration among the rainforest flora.

Here are some flowers that may work for the hours

Here are some stems that may work for the minutes:

InitialPROTOTYPE

2005-11-10T14:41:00.000-08:00

My first prototype of the timepiece. The small stems in the middle represent the minutes, the inner petals the hours from 1-6, the outer petals the hours from 7-12.

FinalPROJECT: Lotus of Time

2005-11-03T12:55:00.000-08:00

For my final project I had the idea of making a clock [something that I needed in my room anyway]. My goal is to designing a clock of subtle beauty, not harsh and mechanical; a reinterpretation of nature, something that I miss dearly living in the concrete jungle.

The organic design and reverence for nature in India has always been a constant source of inspiration for me.

This past June I went Udaipur in the state of Rajasthan. We toured its famous palace, which in my mind is the Indian counterpart of the Palace of Versailles. The marble carvings found embedded in niches, doorways, floors were exquisite. I was most struck by the simple flower carvings.

The flower pattern inspired an ID project, namely repeating flowers cut out of leather, to be worn as a necklace. I liked the design so much that I realized the petals would be perfect for my clock interpretation.

Ideally, my clock will have the following features:

Tell time, via the petals of the flower and its tendrils

possibly differentiate between AM/PM

Show seasons, via the change in light color

Display date, again via the petals of the flower, altho I have to work out how to diffential the time and the date

Be able to toggle between different time zones. My family is spread all over the world and having this function would be very handy when I need to make phone calls

The remaining questions are:

scale, I like building in large scale, but I may built a small prototype to work out the kinks with

materials. Right now I envision using plexiglass & blue and white wire. The color of the wire is extremely important as I want the wiring to be a part of the design. Plexi is quite expensive, so I may have to substitute another material.

finding the correct LEDs

figuring out how it can be mounted onto a wall, or should it be standalone?

The.USER.Illusion

2005-10-29T16:01:00.000-07:00

The art I have made in the past could be defined as high resolution (3d animation) or low resolution (drawings). In my time at ITP, my purpose is to create art that responds to a user, not something that simply exists to be viewed. It is tempting to throw a lot out to the user at once, but "The User Illusion" illustrates that all these things are filtered differently by all, depending on a number of variables. The variable that interests me the most is increasing conscious bandwidth by tapping into a user's emotions. I saw Daniel Rozin's "Wooden Mirror" at SIGGRAPH years ago. But I only recalled it once I heard of physical computing. The secondary spark in me came years after I interacted with the mirror. When that happened it changed the course of how I see things.

Obs Proj TWO, the Final Stretch [um, Prototype]

2005-10-29T13:32:00.000-07:00

Although the drum trigger was working, it was not giving us reliable results. As suggested in class we decided to work in parallel. We ultimately wanted two working paddles, so we decided to have one hooked up to a midi synth box and the other one using the Sonia sound library in Processing.

I am not taking ICM so I had never used Processing before. Dan Shiffman's book draft was invaluable in getting me up to speed, but soon realized that sound was not Processing's forte. At the end of the day I got Daniel to come by and take a look at my code and even he couldn't figure out why it wasn't working!

So I ditched Sonia for the time being and got the midi synth box from the ER. I got some simple tones to play with the paddle. Going back to my days of playing flute I got the paddles to play Salt 'n Peppa's "Push It".


DEFINE OSC 20
DEFINE HSER_RCSTA 90h ' enable the receive register
DEFINE HSER_TXSTA 20h  ' enable the transmit register
DEFINE HSER_BAUD 31250 ' set the baud rate

TRISD = %00000000   'Set all of PORTD pins to output mode

paddleVar VAR word 'Create variable to store result
pitch var byte(12)

i var byte
i=0

' ranges for reference
pitch(0) = 60' mid c
pitch(1) = 61' c#
pitch(2) = 62' d
pitch(3) = 63' d#
pitch(4) = 64' e
pitch(5) = 65' f
pitch(6) = 66' f#
pitch(7) = 67' g
pitch(8) = 68' g#
pitch(9) = 69' a
pitch(10) = 70' a#
pitch(11) = 71' b


' a,e,d,c,b,g,g,b,c,b,g - salt n peppa push it!
'69,64,62,60,71,67,67,71,60,71,67
song var byte(11)
song(0)=69
song(1)=64
song(2)=62
song(3)=60
song(4)=71
song(5)=67
song(6)=67
song(7)=71
song(8)=60
song(9)=71
song(10)=67

'light debug sequence
 high PORTD.0
 pause 200  
 low PORTD.0

 high PORTD.1 
 PAUSE 200
 low PORTD.1

main:
 ADCIN 3, paddleVar 'coming in at analog port 3

 if paddleVar > 100 THEN 'set to 100 since slight vibrations give off readings
    hserout [$90, pitch(i),$40]
    if(i>11) then 'loop thru song once its at end reset to 0 o
     i=0
     ELSE
     i=i+1
    ENDIF
 ENDIF

 PAUSE 100

goto main

*One thing to remember, when working with the midi box, you need a 20 hz clock and when you program the chip you have to change the oscillator from XT to HS.

ConstructionWOES
On the sensor/paddle side, Steve ran into Danny Tsang [who had done a similiar project for a class] and he told us a better way to rig the paddles. Steve stripped a piezo speaker and took out the piezo film inside, carefully soldered the connections.

We hollowed out the paddle handle, ran wires through it and glued the pieces back on.

Steve used Gorilla Glue, which takes about 3 hours to dry. BUT the glue expands which fixes the sensor into place. [pic here]

We got some code that handled the sensor peak readings and luckily got Sonia to work at the last minute.



' Define ADCIN parameters
DEFINE  ADC_BITS        10             ' Set number of bits in result
DEFINE  ADC_CLOCK       3          ' Set clock source (3=rc)
DEFINE  ADC_SAMPLEUS    20          ' Set sampling time in uS

PeakValue var word
SensorValue var word
LastSensorValue var word
Threshold var word
Noise var word
PingPong VAR WORD

PingPong = 420

' serial pins and data reate:
tx var portc.6
rx var portc.7
n9600 con 16468

Threshold = 50       ' set your own value based on your sensors
PeakValue = 0        ' initialize peakValue
noise = 5           ' set a noise value based on your particular sensor

' Set PORTA to all input
TRISA = %11111111
' Set up ADCON1
ADCON1 = %10000010

Main:
    ' read sensor on pin RA0:
    ADCin 3, sensorValue

    ' check to see that it's above the threshold:
    If sensorValue >= threshold + noise then
        ' if it's greater than the last reading,
        ' then make it our current peak:
        If sensorValue >= lastSensorValue + Noise then
            PeakValue = sensorValue
        endif
    ' if the sensorValue is not above the threshold,
    ' then the last peak value we got would be the actual peak:
    Else
        If peakValue >= threshold then

            ' this is the final peak value; take action
            'serout2 tx, n9600, ["peak reading", DEC peakValue, 13,10]
            'serout2 tx, n9600, [DEC peakValue]
            serout2 tx, n9600, [DEC pingPong]
        endif

        ' reset peakValue, since we've finished with this peak:
        peakValue = 0
    Endif

    ' store the current sensor value for the next loop:
    lastSensorValue = sensorValue
Goto main



//PROCESSING CODE FOR SOUNDS

import pitaru.sonia_v2_9.*;
import processing.serial.*;

/*avani added, put sounds in array, sounds should reside in same folder as .pde file */
String[] wavArray = {"first.wav", "second.wav","third.wav"};// new String[3];

//end avani additions 
int bgcolor;                      // background color
int fgcolor;                      // fill color
Serial port;                              // the serial port
int[] serialInArray = new int[3];    // where we'll put what we receive
int serialPong = 0;
int serialCount = 0;                 // a count of how many bytes we receive

boolean firstContact = false;  // whether we've heard from the microcontroller
int j=0; //wavs index
Sample bounce;

void setup() {
  size(256, 256);  // stage size
  noStroke();            // no border on the next thing drawn

  // print a list of the serial ports, for debugging purposes:
  println(Serial.list());

  port = new Serial(this, Serial.list()[0], 9600);
  //port.write(65);    // send a capital A to start the microcontroller sending
  Sonia.start(this); // Start Sonia engine.
  
}

void draw() {
  background(bgcolor);
  fill(fgcolor);
  // get any new serial data:
  while (port.available() > 0) {
    serialEvent();
    // note that we heard from the microntroller:
    firstContact = true;
  }
}

void serialEvent() {
  processByte((char)port.read());
}

void processByte(char inByte) {
  // add the latest byte from the serial port to array:
  serialInArray[serialCount] = inByte;
  serialCount++;
  
  // if we have 3 bytes:
  if (serialCount > 2 ) {
    bounce = new Sample(wavArray[j]);    
    serialPong = (int)serialInArray[0];
    println(serialPong);
    
    if (serialPong == 52) {
      bounce.play();
      j++;
      if (j>2) { j=0;}
     } serialCount = 0;
  }
}
  
// Safely close the sound engine upon Browser shutdown.
public void stop(){
  Sonia.stop();
  super.stop();
}

[processing code here]

We constructed two Radio Shack project boxes in order to keep the components in place when the user swings the paddle.

So without further ado....see the project in action! That's right! Click here

Lots of fun was had.

Observation Project TWO, it's Analog, it's digital....it ANALOG!

2005-10-22T14:01:00.000-07:00

Steven decided to strip the drum trigger to see what was inside. He found piezo film which basically senses vibration. We rigged the paddle so that the film was placed inside a groove cut for it near the bottom of the paddle. It picked up the vibration and so we figured out YES the drum trigger is analog. Oh and when the vibration was sensed, the LED in Santa's mouth lit up! "Ping Pong, Santa's ON."

Observation Project TWO, Phase II

2005-10-16T18:45:00.000-07:00

We convened and decided to each take a paddle and try out different sensors. We all ended up in the lab at the same time and worked from one board, each testing after another.

Conclusions and findings:

- Drum trigger is digital NOT analog. Had to be hit at a specific pt and not big enough to cover surface area of paddle.

- Flex sensor. Idea was to attach multiple flex sensors to a paddle. Only flex sensor detects when a tennis ball hits it but the range of values is small, very similiar to values when just moving paddle. We tried a various of resistors but this just dropped the values from 900 to 200, but the overall number range stayed the same. After three different flex sensors and no consistent results we decided to try some Piezos.

WeekSIX

2005-10-15T14:12:00.000-07:00

Workshop.
I had put off soldering for quite awhile so I decided to get to it and finally solder my serial cable to the header pins. After that was done, I couldn't get any results in HyperTerminal. Carlin showed me how to use the multimeter to check the connections and see if the signal was going through. In spite of all the signals checking out, once I swapped out the serial header I got results again.

Lesson learned: Solder carefully, very carefully. Save yourself hours of frustration!

Processing.
Since I'm not taking ICM I had no idea what Processing was. I opened it up and after Arly pointed me to Dan Shiffman's site here and I read up on how Processing's Java based syntax flows, I felt more at ease. Dan OSullivan's site has the PIC Basic/Processing code that lets the PIC communicated with Processing. I'm going to strip out all the junk and post a simpler, easier to understand version in a few days.

Brainstorming.
Since college I've always had an interest in furniture design. My ID class spent the last two weeks trash digging in order to acquire enough cardboard to make some chairs. I really dug the whole assignment and had the idea to make a clear chair that shows how many people have sat in it. The idea is still in the early stages so more to come in a week or so.

A life lesson in bad design

2005-10-10T17:38:00.000-07:00

Back in undergrad I did two research seminars on user interface design. It made me shift my personal design paradigm from just making things look pretty to making something functional and usable as well. About a year ago I bought the Design of Everyday Things, which essentially talks about the same concept in terms of physical objects. I am currently taking an industrial design class as well, which has further emphasized the importance of form and function and its symbiosis.

Fast forward to this weekend. My boyfriend Erik gives me his extra iPod. Not being a Mac fan, I am still excited. I was talking on the phone and in my multitasking I connected the WRONG dock cover to the docking station. After wasting 40 minutes on tech support while alternately googling trying to see if the fucking cover was removable, my patience evaporated and I had to destroy the cover in order to get it off. While the rest of the iPod maybe a piece of brilliant design, the docking design leaves A LOT to be desired. I will admit tho that I did enjoy the destruction.

Week FIVE

2005-10-06T13:15:00.000-07:00

I spent quite a bit of time in the lab Wednesday (instead of doing my ID project)...basically lit up a 10 segment bar, got the servo to do some weird unintended things and made some noise with the piezo buzzer. See code below, its simple but might save you a few seconds if you pick it up from here:

* BUZZER CODE *


DEFINE OSC 4

main:
  HIGH portd.2
  pause 500 ' change up pauses and see what happens
  LOW portd.2
  pause 500
goto main

*BUZZER CODE with ARRAY

DEFINE OSC 4

start:

  TRISA = %11111111 ' Set PORTA to all input
  OUTput portd.2

  toneRange VAR WORD[10] ' define array with 10 elements
  toneRange(1)=200
  toneRange(2)=400
  toneRange(3)=600

main:

  i var byte
' loop allows me to build up a range of pauses with a few lines of code
' change 3 to array length as array items are added

  FOR i = 1 TO 3
    HIGH portd.2
    pause toneRange(i)
    LOW portd.2

    pause toneRange(i)
    next i

goto main

Observation Project TWO, Phase 1.

Our group had lots of ideas...involving pool cues, rotating tables, something we dubbed the Jeopardy Meat Buzzer, vacuum tube delivery. All good, but we decided to go simple: Ping Pong. So we got some paddles and did some um, research. See the video here:

http://itp.nyu.edu/ ~sj695/pingpong/

Week FOUR

2005-10-04T10:57:00.000-07:00

Observation Project Continued...

Meeting again, Charles and I decided that the dog run mister would be the most beneficial of our ideas. The mister's features:

1) A long mister at waist level in a shaded area
2) When the temperature/humidity/heat index go above a certain point the mister will automatically come on
3) Has human override so it can be turned on/off whenever wanted
4) Has a motion detector so if there is no one in the run the mister will not turn on

Serials and Pots

After soldering the header to the serial connector, I got to programming. I had lots of ideas, but eventually I just wanted output that moved along the screen as the potentiometer was turned.

TIP: I found that soldering the header was not necessary. Just clip the pins off the connector, clipping all the way down to the hole and jam the header in the holes. I tested it and the connector worked great this way.

The code:



 DEFINE  ADC_BITS  10     ' Set number of bits in result
 DEFINE  ADC_CLOCK 3      ' Set clock source(3=rc)
 DEFINE  ADC_SAMPLEUS 50  ' Set sampling time in uS
 
 
 ADCvar  VAR WORD    ' Create variable to store result
 smallRange VAR WORD             ' holds smaller range of pot
 decRange VAR WORD
 prevRange VAR WORD
 TRISA = %11111111       ' Set PORTA to all input
 ADCON1 = %10000010      ' Set PORTA analog and right justify result
 Pause 500               ' Wait .5 second
 
 
 serout2 PORTC.6, 16468, ["Avani Patel"]
 main:     
     ADCIN 0, ADCvar       ' Read channel 0 to adval
 '    serout2 PORTC.6, 16468, [DEC ADCvar, 13, 10]  
 ' print it to serial out, 
 ' with linefeed and carriage return (10, 13)    
 
     smallRange = ADCVAR/20
     decRange=smallRange
     
     serout2 PORTC.6, 16468, [DEC smallRange,13,10]
     
     
     while decRange > 0    ' inserts spaces based on POT
         serout2 PORTC.6, 16468, [" "]
         decRange=decRange-1  
         wend
         
     if smallRange > prevRange THEN  ' checks if pot is increasing or decreasing
      SEROUT2 PORTC.6, 16468, ["increasing ---->"]
      SEROUT2 PORTC.6, 16468, [10, 13]
     ENDIF
 
     if smallRange = prevRange THEN
      SEROUT2 PORTC.6, 16468, ["------ same -------"]
     ENDIF
      
     if smallRange <>
         SEROUT2 PORTC.6, 16468, ["<------- decreasing"]
         SEROUT2 PORTC.6, 16468, [10, 13]
     ENDIF
     prevRange=smallRange
     
 GoTo main

Took a bit of debugging and ugly math to pad the spaces, but it worked.

Week THREE

2005-10-04T10:37:00.000-07:00

The world of row column scanning.

Back in undergrad we made an 8 or 16-bit adder in my logic and design class. The adder display results on our theoretical 7 segment LED display. So I thought I would do a 3- switch puzzle that when completed flashes the correct number code. The first step was just wiring the 7 seg LED. That wasn't as easy as I had thought. Carlin came by and explain row/column scanning, which conceptually I understood and even wired it properly, but it was messy as you can see!

I spend so much time on wiring that I ran out of time to actually execute the puzzle. But its something I want to keep in mind to do at a later date.

Week TWO Observations

2005-09-22T14:45:00.000-07:00

Charles and I visited the dog run at Washington Sq Park on Wednesday. We talked to Lenore, who brings her dog Tony on a regular basis. We asked her what she thought could be improved at the run. A few ideas / complaints she had (all weather related too):

- the idea of a wading pool and increasing the shade in the run, which around 3 PM had shade along the perimeter, but had mostly directly sun in the center of the run. Mostly a concern on hot days.

- the poor drainage when it rains, the run turns into a lake.

- when it snows the park is too heavily salted (salt is very bad for dog paw pads).

She told us to talk to Pat, the president of the dog run association. Pat had a lot to say. The rules of the run:

The things that stand out

- even though there is fencing, some of the more athletic dogs have been known to jump the fence

- the rules state the females in heat or aggressive dogs are not allowed, aggressive dogs have injured/killed other dogs fairly recently in the run

- the run will most likely be moved soon. Pat expressed concern about this
because its close to the street, so if a dog jumps over the fence, they
would jump right onto a busy road. Also the firetrucks run along that same
stretch of road and the noise tends to spook many of the dogs. Also the
shade is considerably less.

- since the run is completely self maintained, volunteers/money are always needed. The bulletin board keeps owners up to date on developments.

Some observations I made on my own (and partly because i have a dog).

- Dogs love to chase tennis balls.

- Poop pickup is always a problem, esp since the run is a public place its essential.

Ideas based on observations:

1. A sensor that can detect a dog in heat and set off a signal upon detection
2. A ball dispensor (sort of like the tennis ball kind) that shoots balls around the park, if high dog activity is detected
3. A water mister that goes off if the temp reaches a certain temperature, adjusts amount of water/time on according to heat index, humidity.

Week TWO

2005-09-20T10:24:00.000-07:00

This week's assignment called for programming the PIC chip for the first time. Before I began any programming, I hooked up my ON/OFF switch so I know what state I am in. The first switch I had hooked up was a single button switch which wouldn't suffice. I also added an input terminal to more securely fasten my power and ground from fromo the power supply to the board.

I set up the wiring as directed, but ran into a minor problem when I used a 1M capacitor instead of a 22pico one. Once that was straightened out and I added a pull up resistor to the master reset, my LED started blinking.

Todd suggested I use the b ports for my output, so I first added two LEDs to the b ports and once I got that to work, proceeded to add the rest of the wiring for a total of 8 LEDs (+1 initial from portd.1).

I got the whole row blinking and messed around with the timing of the blinking a bit for a different effect.

For the initial switch code I had one LED in my LED row blink when the single button switch was pressed.

Puzzle Idea
My idea for the puzzle was to hook up three single button switches to input terminals and randomly generate the sequence in which they must be pressed. There are only 6 possibilities, so its simple enough to try out. When you have the correct answer the green LED will light up, else the red LED will light up. The 8 panel LED will also tell you how many tries you have had.

Of course this is just my test plan, we'll see how far I actually get when it comes to coding....

Observation Project Ideas
I am very interested in the notion of public spaces, it was one of the main reasons I decided to go to school in NY, just because of the sheer number of posssibilities. My ideas were to scope out Washington Sq Park, Bobst Library and the Rec Center, all different environments with an interesting mix of people. Observations and Charles ideas to follow...

Week ONE

2005-09-15T12:16:00.000-07:00

My first entry. I went into the shop, did some soldering and after reading up on how breadboards and wiring work, I went to work and was very happy to get my first LED to light up.

I did a bit of rewiring, since I realized some of it was extraneous, like the wiring on the left side of the board. Then I went on to light up two LEDs in series (3 LEDS, actually 4, since I have a control LED on the bottom of my board, would not light up since 5V wasn't sufficient to power all)

And then in parallel....

And then I replaced my plain vanilla resistor with a photoresistor.

I have video, but I'm not sure how to put it up just yet. Till next week!