Ultrasonic Transducers - design & construction

Discussion in 'Electronics Resources' started by Bazz, Nov 22, 2011.

  1. Bazz

    Thread Starter New Member

    Nov 22, 2011
    10
    0
    Hi,
    can anyone point me in the direction of good books, papers or websites to look at to pick up the basics of how ultrasonic transducers work?

    My top-level interest is in how ultrasonic flow-meters work with liquids.

    Thanks
    Bazz
     
  2. musicinhills

    New Member

    May 27, 2011
    1
    0
    Hello all
    I would like to know this also, i dont know for sure but some
    have to be wire wrapped as to much heat from soldering iron
    destroys them.
    I want to use them to detect sparking on power lines.
    And for all the fascinating sounds from insects and birds.
    it's a whole new world.
    Thank you all
    Kind regards.
     
  3. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    Ultrasonic flow measurement is an extremely complicated topic. The density of the fluid, particulate matter, pipe type and diameter, and type of fluid, all have an effect on the reading.

    The consensus was high resolution fast ADC in addition to Digital Signal Processing was needed to even get started, the price of which made the commercial ultrasonic flowmeters were found to be preferred, as they had more features as well as presets and calibration information to fit most installations.
     
  4. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    That may need a new topic, ultrasonic isn't very useful at a distance. If you can re-word your requirements in a new topic, members with more experience in power line transmission may be able to assist.
     
  5. Bazz

    Thread Starter New Member

    Nov 22, 2011
    10
    0
    Are you referring to piezo transducers? If so, one approach to soldering them is to use Wood's metal rather than conventional solder. It has a meting point of around 80'C; needs a flux. Works well I think.
     
  6. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    Ultrasonic transducers work the same as any other acoustic transducer.
    It's just that the dimensions are different to suit the smaller wavelengths.

    For example early echo sounders used small metal diaphragm Tx, similar to but much smaller and flatter than loudspeaker cones.

    Solid state TX make use of Raleigh or surface waves modes of excitation rather than body (shear or pressure waves) in general. Again as these move less mass they can reach a higher frequency.

    For materials look up Barium titanate.
     
  7. Bazz

    Thread Starter New Member

    Nov 22, 2011
    10
    0
    OK, but being a bit more specific, what sort of dimensions would be needed to work successfully at 2.8MHz? Although I have found reference to transducers working at frequencies as high as 48MHz, the detail tends to be highly proprietary! Also, in order to place a piezo element some distance from a very hot fluid, solid waveguides need to be used, which of course are rather different to a diaphragm.

    So far, the best info that I'm finding is in PhD theses.
     
  8. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    The textbook

    Ultrasonics by

    Dr P Vigoureux of the Royal Naval Scientific Service

    Has a chapter on transducers and a chapter on ultrasonics in various (inc organic ) gases and a similar chapter on liquids.
    go well
     
  9. Bazz

    Thread Starter New Member

    Nov 22, 2011
    10
    0
    OK, thanks for that- will check it.
     
  10. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    Be aware it is an old book.
     
  11. Bazz

    Thread Starter New Member

    Nov 22, 2011
    10
    0
  12. polychromeuganda

    New Member

    Dec 12, 2011
    8
    0
    I last looked at this 6 or 7 years ago. There's a good bit of information to be found from a few hours with Google. The transducers i saw discussed in the last 30 years were all based on the piezoelectric effect, only the means of coupling the wave to the propagation medium changed. The choice of bulk vs surface wave in the piezo material and the piezo material seemed to settle on a 2 or 3 materials and all the really whizzy stuff was coupling surface waves. The simplest (cheapest and most available) transducers are simple "benders" - small sheet metal disks using the least whizzy material in a bulk mode, with a rather low cutoff frequency no matter how small or thin. The disk is cupped or oilcanned by the bulk shrinkage and expansion of the material. The next level of transducers use a small patch of material bonded to a metal bar to induce a surface wave on the bar. The bar always ends at an angle to medium, the design details (and papers and marketing materials) focus on launching the wave from the bar to the medium. The more sophisticed transducers go beyond coupling waves, and act as directional arrays. A few papers are kind enough to expose the proprietary, secret formulation of the piezo materials, which seemed more like a choice of 3 basic materials.

    The investment and degree of difficulty to make transducers looks to be more on the order thick film circuits than ICs.

    Unless you're using a lot of them and can gain a competetitive advantage with a proprietary design, you buy them.

    BTW as you read, SAW (surface acoustic wave) resonators, filters, and other slow wave structures are a related but seperate topic.
     
  13. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,392
    1,605
    I've worked for two companies heavily devoted to ultrasonic devices, these were specifically bubble detection inside tubes (useful in medical applications) and level measurement (which got me aboard a couple of nuclear submarines). While I don't have a good physical reason for the frequencies used, the level measurements were done at 1MHz while the bubble detection was done at 3MHz.

    The transducers were simple housings we made on bench tops out of common materials. The crystal material was purchased in bulk, scribed and snapped to rough size. Apertures of certain size were created by layering masking tape (somewhere between 4 to 6 layers, I forget exactly how many) and applied to the crystal. The crystal was then adhered to the housing with epoxy. The entrapped air in the masking tape acted as a barrier so the wave front was only emitted between the strips.

    Sizing was used to create a defined length along a tube to detect bubbles above a certain threshold. Level transducers needed no such sizing, but was added for the mechanical reasons of spacing the crystal off the surface so a lead wire could be attached.

    There are two ways to drive such a thing, either continuous wave or with a "ping" pulse. The ping was typically 6 cycles. Pinging was a big advantage as it separates in time the transmit burst and the receive signal; as the receive signal was very low amplitude and required a 60dB amp to resolve the signal transmission noise was an issue.

    There was no physical difference between a transmitting and receiving crystal. In fact, they were almost randomly assigned at construction. It is also possible to use a single crystal as both transmit and receive as was sometimes done with the level measurement devices.

    I did work some with flow measurement, and while I got decent results the concept was declared "a dead issue" by management and I could not continue development. While we were typically working with tubing of 1/4" ID and flow rates below 1 liter/min our fearless leader made a set-up of 2" PVC pipe fed by a spare pool pump sitting in a garbage can of water. It was fairly easy to see the signal thru that thing and see a time shift for no flow vs flow on.

    Sorry I do not know what vendor we used for this crystal material. I believe they were in Connecticut but was not able to track them down. Boston Piezo-Optics looks to manufactures such crystals.

    The NDT sight has had some very useful study material on ultrasonics, but it seems to be down today (12/15/11). Yahoo cached it as recently as 12/11/11 so it may be back again soon.
     
Loading...