Connect with us

Detecting Water Droplet Electronically

Discussion in 'Electronic Design' started by Newsgroups, Jul 12, 2004.

Scroll to continue with content
  1. Newsgroups

    Newsgroups Guest

    Could any point me in the right direction for detecting a falling water
    droplet electronically? This is being done for a photograph so the water
    droplet must free fall without hitting anything. Can't be distorted in any
    way. The droplet is not accelerated other then what gravity provides. Only
    needs to fall about 1 or 2 feet. This turned into a larger problem then I
    first expected. I've tried the following:

    1) Laser pointer into a Cadmium Sulfide photocell but the light pretty much
    travels straight through the clear water and the photocell doesn't react
    fast enough.

    2) Infrared beam has the same problem.

    3) Built a small two plated capacitor (with air as the dielectric) using
    copper clad boards. Measured 27 pf and it would change very nicely when I
    held a water droplet between the plates. But when the droplet is traveling
    through the air, again it was to fast to get a reaction larger enough to
    measure.

    I'm going to dismiss hall-effect thinking that water droplets have a slight
    electronical charge not a magnetic one right? Maybe I just need to redesign
    the detection circuits for the above (capacitor or photocell) to make them a
    ton more sensitive. In any case I would love the input of people much
    smarter than me :)

    Any help or pointers towards web pages, discussion groups, physics forums
    etc would help. I'm not a student here doing home work. Just a guy in my
    back room having fun with electronics and his new D70 camera.

    thanks
    Dan
     
  2. JB

    JB Guest

    You could try using hot water and detecting the droplets using a PIR
    sensor/switch (bog-standard intruder/motion sensor type device for security
    lighting etc..) These often have variable sensitivity and detection
    distance/angle.You will need to disable or cover the auxilliary photocell
    (for switching the unit off during the daytime). You could then use it in
    normal visible light.

    JB
     
  3. Fred Bloggs

    Fred Bloggs Guest

    Arrange for a steady stream of droplets into water and phase lock an
    oscillator to microphone input of the "plop"- then when you want to snap
    a pic or whatever, press a button asynchronously and fire the trigger at
    a manually adjustable delay into each PLL cycle until you get it right.
     
  4. A photocell is never fast enough. Use a photodiode.

    Rene
     
  5. Ben Bradley

    Ben Bradley Guest

    The problem here is more the slow reaction time of the photocell
    than anything else. Use a photodiode. The droplet will refract the
    light enough to cause a significant dip in the light seen by the
    diode. You may want to add a slight delay in triggering the camera
    with this, unless you want a pic with the laser shining on it.
    Capacitance would change instantaneously as the droplet falls
    through, so you just need a faster measurement technique. Use the
    capacitance as one of the frequecy-determining elements in an
    oscillator of about 1MHz, and have a circuit (perhaps a PLL) watch the
    frequency change as the droplet falls through.
     
  6. logized

    logized Guest

    Dan,
    Maybe you could use a motion detection camera to sense the water
    droplet appearing, see http://www.supervisioncam.com/
    It can make a sound or run a program to trigger your camera. Don't know if
    it would be fast enough though.

    Dave
     
  7. Yep. And try mounting the photodiode off axis from the source beam and
    detecting the scatter or reflection from the drop.
     
  8. Mike Page

    Mike Page Guest

    Though I favour an optical technique this one has me thinking. Other
    devices work on capacitance change, including condensor microphones
    (which are well written up). The idea is you maintain a constant charge,
    so the change in capacitance manifests as a voltage (Q=CV). Involves a
    high impedance buffer, which should be do-able with a CMOS opamp and
    some careful cabling.

    Regards,
    Mike.
     
  9. John Fields

    John Fields Guest

    ....

    ---
    get yourself an optical slotted switch with a wide gap, like this:

    http://www.optekinc.com/pdf/OPB800L.pdf

    Digi-Key either has some or some like it.

    Pull up the phototransistor's collector with whatever value is
    recommended in the data sheet, then force enough current through the
    LED to get a stable low out of the transistor's collector with nothing
    in the gap, then when a drop of water falls through the optical axis
    of the switch, the transistor's output will go high. If you get the
    drop exactly centered you may get two high peaks with a valley in
    between; the first peak when the droplet first breaks the beam and
    reflects and diffracts the IR away from the sensor, then the valley
    when the droplet is normal to the beam and lets some IR through, and
    then the final peak as the droplet is leaving the beam.
     
  10. Genome

    Genome Guest

    | Could any point me in the right direction for detecting a falling
    water
    | droplet electronically? This is being done for a photograph so the
    water
    | droplet must free fall without hitting anything. Can't be distorted
    in any
    | way. The droplet is not accelerated other then what gravity provides.
    Only
    | needs to fall about 1 or 2 feet. This turned into a larger problem
    then I
    | first expected. I've tried the following:
    |
    | 1) Laser pointer into a Cadmium Sulfide photocell but the light pretty
    much
    | travels straight through the clear water and the photocell doesn't
    react
    | fast enough.
    |
    | 2) Infrared beam has the same problem.
    |
    | 3) Built a small two plated capacitor (with air as the dielectric)
    using
    | copper clad boards. Measured 27 pf and it would change very nicely
    when I
    | held a water droplet between the plates. But when the droplet is
    traveling
    | through the air, again it was to fast to get a reaction larger enough
    to
    | measure.
    |
    | I'm going to dismiss hall-effect thinking that water droplets have a
    slight
    | electronical charge not a magnetic one right? Maybe I just need to
    redesign
    | the detection circuits for the above (capacitor or photocell) to make
    them a
    | ton more sensitive. In any case I would love the input of people much
    | smarter than me :)
    |
    | Any help or pointers towards web pages, discussion groups, physics
    forums
    | etc would help. I'm not a student here doing home work. Just a guy in
    my
    | back room having fun with electronics and his new D70 camera.
    |
    | thanks
    | Dan
    |
    |
    |
    |

    Mount the camera below the point where the drop is detected so it's
    passing by when the camera is triggered?

    DNA
     
  11. Jamie

    Jamie Guest

    dropping water passing the plates (not touching) should product an
    electrical charge. this is the bases of the home made thunder shower.
    i don't remember the exact construction but it can be found on the
    net in any case, one could use a FET input with diode suppression from
    the plates to detect this charge.
    another approach is to use U-sonic, since it can be narrowed down to a
    small stream and water will some what absorb it, i think i could make
    for a good absent detector when the water is passing through its path.
     
  12. mike

    mike Guest

    Mike's rule #2...Never measure anything when you can control it instead.
    Use an electromagnet to "shake loose a droplet". A little physics will
    tell you when the drop will be where you want it. Delay the camera
    trigger by that amount.
    mike

    --
    Return address is VALID.
    Bunch of stuff For Sale and Wanted at the link below.
    Compaq Aero floppy,ram,battery.
    FT-212RH 2-meter 45W transceiver.
    Toshiba & Compaq LiIon Batteries, Test Equipment
    30pS pulser, Tektronix Concept Books, spot welding head...
    http://www.geocities.com/SiliconValley/Monitor/4710/
     
  13. Rolavine

    Rolavine Guest

    Subject: Re: Detecting Water Droplet Electronically
    Yeah, or create the drop by opening and closing a solenoid? I went with the'
    control i't idea too, but the sensing ideas were all really pretty amazing.
    There are some real good engineers on this group!

    Rocky
     
  14. mike

    mike Guest

    There's an even simpler method. Make a stream of drops close together.
    Snap a few random pictures. Eventually, you'll get one positioned where
    you want it. Ain't statistics grand?
    mike

    --
    Return address is VALID.
    Bunch of stuff For Sale and Wanted at the link below.
    Compaq Aero floppy,ram,battery.
    FT-212RH 2-meter 45W transceiver.
    Toshiba & Compaq LiIon Batteries, Test Equipment
    30pS pulser, Tektronix Concept Books, spot welding head...
    http://www.geocities.com/SiliconValley/Monitor/4710/
     
  15. Dan

    Dan Guest

    Wow, thanks for feedback. The photodiode seems to win the vote and John's
    idea of using the optical slotted switch (which I have plenty of from old
    computer mice) is perfect. Maybe after my fun is over I'll play with the
    capacitor approach. I liked Ben's idea of using the 1 MHz osc, that would be
    fun to try. Sound would be an option too as Fred pointed out but I really
    want to try to get a fix on the current droplet-in-flight rather than one
    following
    for reasons I hadn't fully explained. Yes the plot thickens ..

    The rest of the story is I've already built a "Common Pin" launcher. Reminds
    me of a little rail gun if you could see it (I could always put a picture up
    at my site). Anyway its activated via a solenoid (you knew that was coming).
    My Nikon 995 has a serial port for control of all its functions and I have
    the protocol and commands from the net. I'm currently finishing up the
    little reservoir with the tiny tube to drop the water droplets. And the
    air intake is controlled by a floppy stepper motor to control the drops, the
    last
    piece was the water detection for feedback and mostly timing.

    The (hopeful) final results will be a macro picture of a common pin
    projectile hitting a water droplet in mid air. It's probably been done a
    million times but I've never done it and wanted the exercise of learning
    what it would take. Would be fun to have the pictures also :)

    - Dan

    P.S. You guys are great, thanks for the all the great ideas.
     
  16. John Fields

    John Fields Guest

    ....

    ---
    get yourself an optical slotted switch with a wide gap, like this:

    http://www.optekinc.com/pdf/OPB800L.pdf

    Digi-Key either has some or some like it.

    Pull up the phototransistor's collector with whatever value is
    recommended in the data sheet, then force enough current through the
    LED to get a stable low out of the transistor's collector with nothing
    in the gap, then when a drop of water falls through the optical axis
    of the switch, the transistor's output will go high. If you get the
    drop exactly centered you may get two high peaks with a valley in
    between; the first peak when the droplet first breaks the beam and
    reflects and diffracts the IR away from the sensor, then the valley
    when the droplet is normal to the beam and lets some IR through, and
    then the final peak as the droplet is leaving the beam.
     
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

-