Connect with us

Reading input to parallel port

Discussion in 'Electronic Basics' started by Peter Andersen, Nov 14, 2005.

Scroll to continue with content
  1. Hi,

    I am trying to capture an input with a status pin on the standard IBM
    parallel port.

    I looks like all the status pins are set high default when there's no
    external voltage source connected.
    If I measure the voltage drop across pin 13 (status 4) and pin 25 (ground) I
    get 5,1 V. Shouldn't it be zero?

    I am using FreeBSD and if I read the value of the status pins with nothing
    connected I get
    Status: 0x7f
    Nothing happens if I connect a 5 V voltage source to pin 13 and pin 25.

    If I short circuit pin 13 and pin 25 I get
    Status: 0x6f

    Isn't it supposed to be the other way around or am I doing something wrong?

    Best regards.
     
  2. Guest

    Unconnected inputs are much more likley to read high than low; in some
    logic families it's almost reliable.

    Some of the parallel port signals are inverted between their electrical
    levels on the bits accessable to software.
     
  3. I suspect it has something to do with a pull-up resistor inside the port.
    Is it possible to force the unconnected input-pin to logic 0 and logic 1
    when a voltage source is connected (in this case +5V)?
     
  4. Undriven inputs, will 'float' to levels, depending on the logic family
    involved. Historically, the TTL inputs used on printer ports, had a small
    internal current, that drove them high. If you look at the equivalent
    circuit for the original inputs, they have a resistance of about 4KR,
    between the input, and the 5v supply rail. Latter chips using CMOS logic,
    still emulate the original inputs. Note that the connection to pin 11, is
    inverted, while the other default status inputs are non-inverted, which
    allows a 'disconnected' status to be detected (this line is called
    'Busy/Off Line' for this reason).

    Best Wishes
     
  5. John - kd5yi

    John - kd5yi Guest


    http://www.lvr.com/files/ibmlpt.txt
    http://www.lvr.com/jansfaq.htm
     
  6. wrong?

    The printer port has rather strong pull-up resistors to VCC, so yes, it is
    correct that you measure 5 V on the inputs and read all ones when left open.

    Meindert
     
  7. Sjouke Burry

    Sjouke Burry Guest

    Printer port input and outputs act like ttl , and the status pins
    are normally high,because older printers could just use a switch
    to pull the line to ground(no supply/electronics needed).
     
  8. Bob Monsen

    Bob Monsen Guest

    Yes. You use a 'pull-down' resistor to gnd, and then a voltage source to
    overcome that.

    Pullups on PC parallel ports can be as small as 1k, so I'd go for a 220
    ohm to ground. However, you can measure it easily by attaching a known
    value resistor r to ground, and then Rpullup = r*(5/vmeasured - 1)

    Then, choose a pull-down resistor that pulls it down enough to be seen as
    a 0 by the port input driver.

    ---
    Regards,
    Bob Monsen

    Even the greatest of creations start from small seeds.
    - Unknown
     
  9. Chris Jones

    Chris Jones Guest

    TTL inputs tend to float high when they are not connected. The newer chips
    would be made to act like the old TTL chips too. This might explain what
    you see.
     
  10. Andrew Holme

    Andrew Holme Guest

    It is working correctly. The inputs have pull-up resistors to 5V. They are
    designed to be pulled down by open-collector or open-drain outputs e.g.


    VCC
    +
    |
    |
    .-.
    | | Pull-up
    | | Resistor
    '-'
    |
    Pin |
    13 | |\
    Apply .-----> >------o------| >O-
    your | |/
    5V |
    here ___ |/
    |>
    10K |
    |
    ===
    GND

    View in fixed-pitch font.
    created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de
     
  11. DaveM

    DaveM Guest


    The status inputs on most IBM parallel ports are TTL (or derivatives
    thereof) and are active low. This means that an unterminated device input
    will float to a logical "1" unless specifically brought to a low state by
    external circuitry. A TTL "1" is a voltage of 2.4V or greater (up to a max
    of 5V).

    Your port and your software are working properly.

    --
    Dave M
    MasonDG44 at comcast dot net (Just substitute the appropriate characters in
    the address)

    Never take a laxative and a sleeping pill at the same time!!
     
  12. Guess you'd better enlarge your knowledge about the parallel port interface.
    One place - amongst others - is:
    http://www.beyondlogic.org/
    It lacks hardware details. Some more can be found at:
    http://www.epanorama.net/circuits/parallel_output.html

    Although pin description and register usage are well defined, hardware
    details of the inner port are not very common. One of the reasons may be
    that since the times of the first PC manufacturers used a variation of
    components to implement its funcion. An average printerports hardware was
    build like this:
    - The datalines were TTL-outputs like LS273 or LS374. Some had smal
    seriesresistors in the outputlines like 22E or 33E (E for Ohm). A capacitor
    of about 470pF to GND was common although I often saw its space on the board
    left empty.
    - The controllines used open collector outputs like LS06 or LS07. The had
    pullup resistors of let's say 4k7 and also capacitors to GND.
    - The statuslines had LS14 Schmidtriggered inverters for input and low
    (150E) pullup resistors. Place for capacitors was often available but left
    empty.

    It did not take long before special I/O chips took over the role of the
    common TTL-components but the function was assumed to remain the same. I did
    not always fit. I remember a parallelport chip build in CMOS that was blown
    whenever the PC was switched off before the printer was. Nevertheless a
    common parallel printerport will still be able to provide (about) the same
    performance the old things did.

    petrus bitbyter
     
  13. Thanks for all the replies. They were all very helpful.
     
  14. scada

    scada Guest

    No, it's working as expected. The status port is 5 bits, S7 is normally low
    (inverted), S6, S5, S4, & S3 are all normally high. The open circuited byte
    in binary would look like 01111111 , or 0x7F. The 3 least significant bits
    (LSB S0-S2) are not there, they will allways be high. When you grounded Pin
    13 (S4) you got 01101111 or, 0x6F. To change bit S7 you need to pull it high
    to +5V (through a resistor - 10K should do it). To change bits S6-S3 you
    pull them low (to ground, negative return). Good luck.
     
  15. Jasen Betts

    Jasen Betts Guest

    you're doing it right.the pins are high by default, you need to connect the
    pins to ground to change their status.

    If you want it to be the other way round you can connect them to ground
    using a 1K resistor and then they'd be low by default and applying +5v
    would change thier state.


    Bye.
    Jasen
     
  16. Rich Grise

    Rich Grise Guest

    Actually, that sounds about right - they're pulled up internally, so you
    have to pull them low to get a 0:
    http://www.epanorama.net/documents/pc/parallelport_schematic.html

    Have Fun!
    Rich
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-