Shop OBEX P1 Docs P2 Docs Learn Events
Attempt at a Propeller lie detector : Need truthful input — Parallax Forums

Attempt at a Propeller lie detector : Need truthful input

Oldbitcollector (Jeff)Oldbitcollector (Jeff) Posts: 8,091
edited 2009-04-17 05:01 in Propeller 1
I'm working on an idea for a Propeller based Lie Detector using
uController's Game Stacker/Spin Studio board.

The idea is to make use of his RGB led, blue for normal, red when lie is detected.

I thought I *might* be able to use the RC time code from the PEkit Labs,
starting with RC Resistance Meter using prop-plug pins 30,31 for reading
the leads (pictured below) and then reporting the data to the screen.

The RC Delay circuit calls for a 0.01uf cap between the two I/O pins, but
it looks like there is too much "skin resistance" for this to work reliably.

Someone with a little more EE knowledge care to help me sort out what
I can add to this circuit to make this work?

Thanks
OBC

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
New to the Propeller?

Visit the: The Propeller Pages @ Warranty Void.
576 x 384 - 101K

Comments

  • Oldbitcollector (Jeff)Oldbitcollector (Jeff) Posts: 8,091
    edited 2009-04-17 03:58
    Upon further reflection and Google, it looks like human skin resistance
    is between 500 and 1500 ohms. (Depending on skin moisture and cleanness)

    Perhaps I need to add an additional resistor in line to get the levels
    closer to intended values for the spin code. (10k)

    Still open for expert opinions. [noparse]:)[/noparse]

    OBC

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    New to the Propeller?

    Visit the: The Propeller Pages @ Warranty Void.

    Post Edited (Oldbitcollector) : 4/17/2009 4:19:50 AM GMT
  • PhilldapillPhilldapill Posts: 1,283
    edited 2009-04-17 04:14
    OBC, if you're depending on the resistance of a person's skin to figure out if they are lying(based on persperation, clenching, etc.), I don't think that's the best way to do it. Here's an example. Let's say you ask the person a series of easy questions and they are relaxed. Their resistance will be fairly high because of the lack of perspiration and clenching. Now, you ask them if they ever look at the same sex with lust(just a WEIRD out of the blue question), and they wil either clench up, laugh, or what have you. There will probably be a response that you can detect, but if you ask another normal question(like if they like onions or not), and they will most likely still be a little tense from the last question, skewing the results.

    I'm sorry I can't come up with anything better than to say "nope, that's not it", but I just see this approach causing problems.

    Maybe it's how you ask things, and duration in between tough questions that is needed. Maybe it's not so much a harware question afterall...
  • Bob Lawrence (VE1RLL)Bob Lawrence (VE1RLL) Posts: 1,720
    edited 2009-04-17 04:46
    The resistance of our skin varies from person to person and fluctuates between different times of day. In general, dry skin is a poor conductor that may have a resistance of around 100,000 Ω, while broken or wet skin may have a resistance of around 1,000 Ω.

    Info Source: en.wikipedia.org/wiki/Electric_shock

    And that's the truth turn.gif honest.............. lol.gif

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Aka: CosmicBob

  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2009-04-17 05:01
    Oldbitcollector,
    ·
    Not sure about the lie detecting part, but here is something you can try in lieu of using RCtime.· I tried this with so-so results; I think it needs a little experimentation.
    ·
    The basic principle is to set up one of the counters to operate in an "inverter" mode.· Adding a capacitor from the input of the inverter to ground in combination with a resistor (your skin) across the input and the output of the inverter you have a basic oscillator that varies in frequency depending on what the resistor·(or capacitor) value is.
    ·
    From here, you just Zero the Phase accumulator for that counter, wait for a consistent time interval, and read the Phase accumulator.
    ·
    CON     ''General Constants for Propeller Setup
      _CLKMODE = XTAL1 + PLL16X
      _XINFREQ = 5_000_000
    RCpinDrive = 4
    RCpinFeedback = 5
      
    OBJ     ''Setup Object references that make this demo work
        Ser       : "FullDuplexSerial"
    PUB Main_Program
        Ser.start(31, 30, 0, 19200)     '' Initialize serial communication to the PC
        DIRA[noparse][[/noparse]RCpinDrive]~~                                  'Make BPin an Output
        CTRA :=  %01001 <<26 | RCpinDrive <<9 | RCpinFeedback 'Setup counter for POS detector with Feedback mode
        FRQA := 1                                           'Start Counter
    
        repeat
          PHSA := 0
          waitcnt(cnt + 1000000)
          Ser.dec(PHSA)
          ser.tx(13)
    

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
    652 x 502 - 65K
Sign In or Register to comment.