Begginer questions: Infrared chronometer

Discussion in 'The Projects Forum' started by congo blue, Jun 27, 2006.

  1. congo blue

    Thread Starter New Member

    Jun 27, 2006
    Hello to everyone, this is my first post here.

    I am trying to build a device to measure the speed of objects falling trough a pipe, which can be pressurized or depressurized at will, for physiscs demonstrations.

    My plan is to use 2 infrared (could be visible light if that is easier) emitter/detector pairs, connected to a chronometer. When the first beam is blocked by the falling object, the chronometer starts, when the second beam is blocked, the chronometer stops, and I can get a reading.

    I figure I can divide the project in the following parts: 2 photo triggers, a clock signal generator with a known frequency and start and stop 'switches', a counter for the clock's 'ticks' and a display.

    So far I have found tutorials on the phototriggers using an emitter, a phototransistor, a general purpose transisitor and some resistances, and tutorials on making 'clocks' based on a 555 (but not one with simple start and stop switches). I also found a way to count the clock's 'ticks', but I am not sure on how to divide it by a reasonable number so that I can get my final display in the 100's using three 7 segment LED displays.

    My question for you, is how to bring this all together. I am asking for pointers to material that can help me figure it out myself, but if you could take the tame to draw a full schematic, or schematics for the different elements, I will be in your debt. :)

    Thank you for your attention.
  2. mozikluv

    AAC Fanatic!

    Jan 22, 2004

    will your display show result in seconds or milliseconds?
  3. congo blue

    Thread Starter New Member

    Jun 27, 2006
    mozikluv, thanks for your fast response.

    Milliseconds is what I am looking for, I want to see the speed difference in droping wood, lead and styrofoam spheres of the same size down a pipe at normal atmospheric pressure, at as closse as a vacumm as i can safely get with clear PVC pipe, and at a pressure of about 40 psi, so the difference will be small.

    I've been doing some research, and I have found similar projects that skip the 555 timer, and use a crystal for a pulse generator, connected to a divider and then to cascading counters, the display logic and the display.

    It also seems that what I need for the start/stop is a flip/flop, whose output is ANDed with the clock signal before going into the counter. The photogates would be NOTed, and connected to the trigger and reset respectively. The first shadow would set the flip/flop to "on", letting the clock pulses get to the counter, and the second shadow would reset it to "off".

    Using all those big words makes me feel very smart, it's a shame i have absolutely no idea how to build that stuff.
  4. beenthere

    Retired Moderator

    Apr 20, 2004
    Hi,You might want to look into either a cmos or ttl clock oscillator. They are inexpensive, and possibly more accurate than a home-built time base. Do you have any test equipment to use, especially an oscilloscope?
  5. Gadget

    Distinguished Member

    Jan 10, 2006
    Yep, a couple of IR senders and receivers should work fine for sensing.
    Perhaps the top one going to the SET input and the Bottom going to the RESET input of a basic SR flip flop... a couple of NAND or NOR gates will do this nicely. That will give you an output pulse the same length as the time the object is Between the sensors.
    Use that pulse to run the clock (enable)
    If your using a clock with separate Start/stops, then simply buffer the outputs from each sensor and send direct to the clock.
    You could place the In/Out pins of a CD4016 across the start, and one across the stop switches with the IR receivers gating them if you wern't sure of the interfacing details... so long as everything is Low Voltage, and the negative rails of the clock and trigger circuits are tied together.
  6. hgmjr

    Retired Moderator

    Jan 28, 2005
    Greetings congo blue,

    You might need to consider the effect of the curvature of the sphere in your apparatus. If the diameter of the sphere is quite a bit smaller than the diameter of the tube in which it is dropped then a potential error in the time of flight measurement could occur. This error would occur if the sphere falls in such a manner that its lowest point is offset from the center of the pipe when it blocks either the top and/or the bottom beam.

    One way to minimize this error would be to provide two beams at the top and two at the bottom. The two beams at either end would be on the same plane but they would be aligned such that one beam shot north-to-south while the other beam shot east-to-west. The output of the two beams at the top would then be logically OR'd together such that if either of the two beams were blocked your timer would be triggered to start. Similarly the two beams at the bottom would be OR'd together and used to stop the timer.

    A small error would still be present but that too could be further reduced by introducing another beam so that you would have three beams at the top and three at the bottom. The three beam solution would have the beams spaced 120 degrees apart.

    I only mention this for your consideration as I imagine that the differences in the time-of-flight with and without the vacuum may be small requiring a high degree of repeatability in the measurement to permit you to have confidence in the outcome of the experiment.

  7. congo blue

    Thread Starter New Member

    Jun 27, 2006
    Thanks for your answers, I have been busy building the non electronic parts.

    Beenthere: Could you elaborate on the ttl clock? Do you have a specific one you recommend?

    Gadget: Thanks, I will look into the CD4016, I am correct that it is so that I don't have to worry about the specifics of the sensors, and get a clean 'on' and 'off'?

    hgmjr: Thanks for your suggestion. I think it would be simpler to use cylindrical weigths instead of speheres, but anyway I want the users to perform several dozen measurments, and some simple statistical analisys.

    I will try to figure out the details during the weekend, and will post a schematic for review.

    Thanks again for your attention.
  8. beenthere

    Retired Moderator

    Apr 20, 2004

    If you are going to use cmos logic, than the CTS MXO45 oscillators will probably do. They're limited to 5 volts, but cover a range from 1 MHZ up to 80 MHz. Digi-Key carries them @ $2.78.