1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Speed of Sound in Water

  1. Jun 14, 2013 #1
    I hope this is applicable to this section.

    Basically I have two hydrophones immersed in a water bath, approximately 12cm apart. One is connected to a function generator, the other to an oscilloscope.

    I'm comparing both the transmitted and received waveforms on the scope, but I'm finding it difficult to explain my results. Basically my observed waveform is at best in phase with the input, and at high freq (~100kHz) actually leads the input signal on the scope!

    Doing a basic v = s/t, and taking the accepted value of v=1482m/s, I reckon I should be getting a Δt of 80μs between the waveforms.

    Have I broken physics??
  2. jcsd
  3. Jun 14, 2013 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    I doubt it.
    How big is your water bath?
    What material is used to construct the water bath container?
    Why are your two hydrophones so close together?
    Any possibility of the HF sound setting up resonance in the container itself?
    Are the sound waves reflecting of the sides of the water bath?
  4. Jun 14, 2013 #3


    User Avatar
    Gold Member

    When sinusoidal waves are displayed on a dual-trace o'scope and they appear to be "in phase" that would mean within a few degrees of one another. Have you considered that the received waveform may be say, 355 degrees later?
  5. Jun 17, 2013 #4
    The bath is approximately 30*16*20cm in dimension internally.

    The water bath is of metal (stainless steel I think) construction, but I have wrapped the walls in bubblewrap, placed pebbles in the bottom and various bits of foam floating on top to try and minimise reflections from the container.

    The hydrophones are close together due to the small size of the bath.

    I'd hope with the padding being used that reflections wouldn't be a problem, but it is a possibility.

    I'm going to measure the time delay between the two waves assuming 355* or greater phase difference to see if that explains things, but I think I tried this last week and it didn't really explain it. I'll post back with results though.

    Thanks guys.
  6. Jun 17, 2013 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Welcome to Physics Forums.

    The wave period for ~100 kHz is only ~10 μs. A delay of 80 μs (that you calculated correctly, btw) will show up many periods later, and so be indistinguishable from a delay that is less than 1 period. Bobbywhy was pretty much on the right track, but the problem is a lot worse than being ~360°, it's more like ~8*360°.

    The easy thing to try is to move the hydrophones slightly closer or farther, by a few mm, and see if the waveform delay changes by the appropriate amount. Best if you can observe the waveforms while moving the hydrophones.

    Another thing to try is a much lower frequency, with a period considerably longer than 80 μs. This would mean considerably lower than 10 kHz.

    If possible, try a signal burst that lasts only several periods @ ~100 kHz and then turns off, and look for when the receiver picks up the burst. May or may not be possible with the generator you have. The off time should be considerably longer than 80 μs before sending the next burst.
    Last edited: Jun 17, 2013
  7. Jun 20, 2013 #6
    Years ago, I had the telecom students build a simple TDR, Time Domain Reflectometer.
    I had them create the pulse with a d flip-flop, D tied high, and -Q tied to clr.
    clk is tied to a 1 khz clock. The output of Q will be about a 100 ns pulse occurring at 1 khz.
    You can make the pulse longer by adding a resistor and capacitor.
    Something like this should let you see your exact delay, as the pulse rep rate is much slower
    than travel time.
    I expect you will be able to see the secondary reflections from the sides of the tank,
    but they may get lost in the general pulse dispersion.
    You can increase the clock input rate to more closely match the scope scan.
    have fun, sounds like an interesting experiment.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook