updated 2010-05-27.
Contents:
Nifty things to do with ultrasonic transducers: ultrasonics and sonar
See also music.html
woefully incomplete ...
Acoustic stuff. Medical ultrasonics, simple distance measurement measurements ("sonar"), noise-cancellation, etc.
introductions to ultrasonics; beginner tutorials.
David Cary:
I've been playing with some Polaroid sonar modules (including the 6500 Sonar Module). They work for me.
I only needed to hooked up 4 wires to the module:
+5V----->[1 TL851 ] -init--->[14 9]-----> +echo gnd----->[3 ]
(all the relevant digital signals are on the "digital chip" TL851; I got the exact pin numbers from the free fax-back service Polaroid http://www.polaroid-oem.com has at 800 392 1170. . That chip is also in Texas Instruments _Data Book Volume 3: Linear Circuits:...Special Functions_. ) /* was some information at http://www.netwales.co.uk/ubiquity/polaroid/american/pdf.htm but this seems to be offline */
I hooked a square-wave function generator to the /init line, set it to a couple of pulses per second, and looked at everything with my o'scope.
When the box sees the -init line go high, the box sends out a pulse (I can hear the chirp). A short time later the box pulls the +echo line high. Then it is absolutely quiet. Then I pull -init low ... until the next time I pull -init high, telling it to chirp again.
Is your +5V power supply working OK ? Perhaps the sonar module pulled a spike of power so large that your power supply couldn't handle it and shut down, resetting your cpu.... over and over. Perhaps your -init signal is not properly connected, causing lots of oscillations that continuously ping the transducer.
A o'scope makes it a lot easier to debug any problems.
Subject: Re: [Q] POLAROID P6500 ULTRASONIC RANGING MODULE From: "Ronald M. Jackson" <fascinating@columbia-center.org> Date: Sat, 25 Jan 1997 09:19:53 -0800 ... We used the Polaroid Rangefinder in an article "Ultrasonic RADAR" that appeared in Electronics Now magazine in September 1993. We used the external blanking to shorten the blanking interval to 1 mS. So I know their blanking does work. The usual problem is that the module draws a modest amount of current (100 mA) most of the time, and then a whopping amount of current (2 A) when the transducer fires. This results in a voltage drop of 4 volts or so on the 5 volt supply. That is a DROP of 4 volts (ie 1 volt supply voltage). This causes all kinds of problems. When we sell one of their ranging boards we include two extra 470 uF capacitors. We suggest you install one capacitor on the ranging module. A good place is U1, with + to pin 5 and - to pin 16. We suggest that you install the other capacitor across pins 1 (gnd) and 9 (V+) on the circuit board where you connect up the flat cable from the ranging module. Don't extend the wires from the ranging module, just use the 6" flat cable that came with the module. If anyone is interested, we sell the ranging modules with transducer (with cables, capacitors, connectors) for $74.90 plus $3.90 S&H (US). You can reach me by email or call us at 1-800-683-5487. -- Ron Jackson
http://www.vnunet.com/News/1137599The world's top oil producer is building a cluster of 4,000 Intel-based Linux systems to run its oil detection systems.
State-owned Saudi Aramco, which oversees a quarter of the world oil reserves, uses algorithms known as Pre-stack Time Migration (PSTM) to enhance complex seismic data and create 2D and 3D images of oil and gas targets.
"This is a mission-critical application. It does some of the most important work of our business," said Mohammad Huwaidi, exploration systems analyst and PSTM deployment team leader, in a statement.
The PSTM application dates back to 1994 and previously ran on proprietary IBM hardware. Both IBM and Intel were involved in the migration, ...
...
The company is using 900 rack-mounted dual Intel Pentium III 1.4GHz processors, each with 2GB of memory, running Red Hat Linux 7.2.
Intel Solution Services helped to optimise the PSTM code, which was written in the Fortran mathematical programming language and uses the Intel Fortran for Linux 6.0 compiler and Intel VTune performance analyser to identify code improvement areas.
Further expansions will boost the number of processors to 4,000 during 2003.
...
typically in the range of 0.5 to 10 MHz ... has a low absorption rate in soft body tissue (~ 5 Np/m/MHz [FIXME:what kind of unit is this ???]) and a relatively small wavelength (millimeters). ... absorption ... limits how deep waves can pass into the body ... ... wavelength governs how precisely the wave can be focused into the tissue.
Comparatively, radio frequency (RF) waves have a low absorption rate but a very long wavelength (meters). Microwaves and lasers have shorter wavelengths (cm to mm) but much higher absorption rates. ...
Ultrasound works well in any anatomical location of the body that is not blocked by bone or air. ... ultrasonic imagers [are] safe and non-invasive ... ultrasound surgery uses high-power, continuous wave ultrasound to purposely cause controlled tissue damage. ...
... a focused beam ... ... can easily generate temperature elevations of 30 K to 40 K to coagulate tissue in just a few seconds.
... can cause the formation of gas bubbles. These small bubbles, known as cavitation, can resonate in an ultrasonic field. ... causing extremely high temperatures and ireversible mechanical damage ...
... tests indicate that the amount of ultrasonic energy absorbed by the body increases with the frequency: a higher frequency beam will cause a larger temperature rise than a lower frequency beam at the same intensity. ... [but] ... higher frequency waves must have much greater intensities than lower frequencies to cause significant cavitation. Therefore, higher frequencies can be used to avoid cavitation and maximize heating. Most therapeutic treatments using ultrasound use thermal effects since they are well understood ...
researchers have been investigating the use of ultrasonic imaging to guide, monitor, and control ultrasound therapy. ... ultrasound imaging measurements of low temperature elevations has been demonstrated in various materials. Ultrasonic temperature imaging, however, has not yet been demonstrated in live animals ...
... extensive research was performed using ultrasound in a low temperature treatment known as hyperthermia. In hyperthermia, the temperature of the tumor volume is elevated by 7 K to 10 K for 30 to 60 minutes. While this treatment is not sufficient to burn tissue, hyperthermia can enhance the tumor-shrinking effects of radiation therapy.
...
In the last 15 years, there has been renewed interest in using ultrasound to directly ablate (remove by vaporization) tissue noninvasively. ... non-invasive techniques such as lithotripsy (using focused shock waves to disintegrate kidney stones) ... minimally invasive procedures such as laproscopy ("camera" guided surgery through small incisions) ...
... ultrasound much be focused so that the target volume [the tumor] has a high intensity beam while the [healthy] tissue between the transducer and the target volume has much lower intensities. This focusing can be achieved by using a lens, a reflector, or a curved transducer ... [or] a phased array. ... For a single focus transducer, the 6 dB focal length (l) and focal width (w) are given by
l = K1(F/D)^2
w = K2(lambda F / D)
D = aperture diameter F = focal distance lambda = wavelength K1 = typically 8 to 14 mm, depending on aperture angle of array K1 = typically 1, depending on aperture angle of array...
At the Brigham and Women's Hospital laboratory, magnetic resonance imaging (MRI) guides, monitors, and controls ultrasound therapy. ... MR sequences have been designed to non-invasively detect and measure temperature in the body. These include molecular measurements such as the magnetic spin lattice delay time, the proton density diffusion of water, and the proton resonant frequency shift of hydrogen. ...
The proton resonant frequency shift has been shown to be linear with temperature. Thus, it can be used to quantify temperature increases in the body. ... eliminating the need [to implant] thermocouples. ...
... ultrasound can close blood vessels ...
... Ultrasonic phased arrays ... Diagnostic arrays transmit a ultrasonic pulse into the body and then record its echo. ... The elements are usually the size of one half wavelength of ultrasound in water (less than a millimeter) to improve beam forming and image resolution. ... Therapeutic ultrasound phased arrays, on the other hand ... radiate a continuous wave signal. ... [and] transmit much higher powers than diagnostic arrays [which] requires the array to have either larger or more numerous elements. ... While most radar or sonar phased arrays are designed to steer the direction of a beam ... a therapeutic phased array is designed to focus a beam to a particular location close to the array. ...
Just as a loudspeaker distorts music if driven too hard ... water will also distort high-intensity sound waves. Rather than being a problem, this nonlinearity offered sonar engineers a way to make the water itself generate low-frequency sound waves from high-frequency ones. ... And because of its small wavelength, an ultrasonic carrier can be sent from a physically small source in a tight beam.
... experts debated whether such parametric arrays would also work in air ... [until] Mary Beth Bennet and David T. Blackstock, both at the University of Texas at Austin, ... created an audible tone in air using ultrasonic waves. ...
F. Joseph Pompei ... [at] the Media Lab in the Massachusetts Institute of Technology ... Unlike earlier arrays built with piezoelectric elements, his apparatus employs wide-bandwidth electrostatic transducers ... he is able to project a three-degree-wide beam for some 200 meters.
Elwood G. Norris, an inventor at American Technology Corporation in San Diego ... [has] also been pursuing the prospect of using parametric arrays in air.
"Point-'n'-Shoot Sound Makes Waves" article by John Gartner 2002-02-21 http://www.wired.com/news/technology/0,1282,50483,00.html
...
The Audio Spotlight is one of two competing audio transmission systems that emit a one-foot square column of sound that can only be heard by people in its direct path. Joseph Pompei, a PhD student at the MIT Media Lab, decided to develop it while working at audio company Bose, which he joined at 16 as its youngest-ever engineer.
...
Pompei said the "non-linear" effect of air pressure modifies sound waves in a consistent fashion. He wrote algorithms that "reverse-engineered" the desired sound waves to determine the appropriate ultrasonic source signals.
According to University of Texas professor David Blackstock, high-frequency signals are easier to focus, and control like a flashlight, than sounds that are within the human range of hearing, which disperse in all directions. Blackstock said ultrasonic signals "decay more slowly than lower-frequency waves" so they are easier to send farther.
The Audio Spotlight emits sounds in the 60-kilohertz range, which, according to Blackstock, is well above the 20-KHz limit of human hearing.
...
An alternative to Pompei's invention, which also may be commercially available soon, is American Technology's Hypersonic Sound System. ...
American Technology president Terry Conrad ...
American Technology is also working on a stronger version of the technology called Directed Stick Radiator, Croft said. This "acoustic assault rifle" is shaped like a gun, but instead of bullets, it dispenses high-decibel sounds that would cause discomfort or even pain.
Holosonic Research Labs: Audio Spotlight by Dr. Joseph Pompei http://holosonics.com/
American Technology Corporation http://atcsd.com/
From: "Ronald M. Jackson" <fascinating@columbia-center.org> Newsgroups: comp.robotics.misc Subject: Re: how do I run robotics through a mac Date: Thu, 30 Jan 1997 12:01:35 -0800 Organization: Fascinating Electronics Inc. Jon Ladd wrote: > > I also use a Mac to run an Experimenter controller. It works as > advertised. The Experimenter uses an integer based 80C552 processor. > Step intervals can be changed by 1 ms through to 255ms/step - forward or > reverse. Max step count is 65,535. Both motors controlled will run > simultaniuosly. Call them up for more details. Radio Electronic did a > series of articles on the Experimenter. Check the library. The articles on the Experimenter appeared in the July and August 1993 issues of Electronics Now (formerly Radio Electronics). The September 1993 issue included an article on an "Ultrasonic RADAR" using the Experimenter and a Polaroid ranging system mounted on a stepping motor that may also be of interest to someone doing robotics.
where to buy ultrasonic stuff.
Includes "ultrasonic modules" (including drive circuit) and the raw transducers.
DAV has played a little with the "Polaroid 6500 Series Module and 600 Series Transducer" kit, which seems to be the de-facto standard ultrasonic system for hobbyists. Note that besides the standard transducer, there are also lots of other 600 Series Transducers that can plug directly into that board (some are a lot more rugged, designed for outdoor in-the-rain applications).
This is known to be incomplete; please help me complete it. Who sells the ultrasonic arrays used in medical applications ?
...
------------ From: "Ronald M. Jackson" <fascinating@columbia-center.org> Newsgroups: comp.robotics.misc Subject: Re: [Q] 6500 Polaroid Sonar Module? Date: Thu, 09 Jan 1997 22:05:41 -0800 Organization: Fascinating Electronics Inc. Chcotay wrote: > I was about to order a Polaroid 6500 Sonar Module......But We stock the Polaroid instrument grade transducer and ranging board here at Fascinating Electronics. Together they sell for $74.90. You can call us at 1-800-683-5487. ------------
Newsgroups: sci.electronics From: cgordon@vpnet.chi.il.us (gordon hlavenka) Subject: Re: Ques:anti-white noise Date: Sat, 25 Nov 1995 19:01:37 GMT Alex Gaal wrote: >The question that occurs to me is, how does this technology >sample the noise faster that any ear can hear it? I don't doubt that it >does I just don't understand it. I've got a pair of those noise-cancelling headphones from NCT. They work best at lower frequencies, which should tell you something :-) Basically there seem to be two factors at work: First, the "anti-noise" doesn't have to be _exactly_ 180 degrees out of phase -- that's just the most effective way to do it. If you are wiling to settle for just a 1% reduction in efficiency, you can allow yourself about 8 degrees of error. (Work out your cosines around 180 degrees) At 100Hz, 1 degree is about 28 microseconds. So your DSP can think for about 220 microseconds before generating the cancellation waveform. Now 220 usec may not sound like much, but a 8051-type processor running at 24MHz can execute 440 instructions in that time; DSPs can do even better (and run faster). The other thing you can do is to study the noise being sampled, and locate any _periodic_ components. You can then predict these, and update your predictions continuously. This obviously takes more DSP muscle and there's a response lag when the periodic noise changes, but it can allow you to work on some of the higher-frequency components. A "psychoacoustic" sidenote: I notice that when I'm wearing those noise-cancelling headphones, I get a very weird sensation. I can't describe it; it's just weird. I'm guessing that it has something to do with the phase shift on the anti-noise. I just think it's weird, but the first time I had my wife try them, as soon as I turned them on she turned white as a sheet and TORE the 'phones off. She won't go close to them now. And she can't explain why. My two-year-old just giggles. (But then, he does _that_ anyway :-) -- ------------------------------------------------------------- Gordon S. Hlavenka cgordon@vpnet.chi.il.us Warning! This post may contain a "stealth" .sig virus!
????
http://www.idec.com/usa/PDFs/Section%20H/SA1W.pdf
???
From: danjo@typhoon.xnet.com (Dan and Joanne Tucker) Newsgroups: comp.robotics.misc Subject: Re: Ultrasonic position measurement ? Date: 29 Jan 1997 16:03:52 GMT ... Mr.SO kin-ming (kmso@Comp.HKBU.Edu.hk) wrote: : My idea is using another transducer to pick up the ultrasonic signal : transmitted from a robot and find out the linear distance. The : original design of the ranging board is measure the time of flight : of a reflected echo using the same transducer, so this is why I think : a modification of the ranging board is needed. : : Mr.SO kin-ming : Email: kmso@Comp.hkbu.edu.hk Well the idea would work - so long as you know - to less than 1 ms - when the sender is pulsing. You can not use (a single) pick up device to get range/distance as there is no base point to time from. (RADAR) If you wanted to use more than one sender tho and these transmitters all fire at the same time, you could use the phase diference to locate a single receiver. (LORAN) Most sonar transducers are tuned to one specific frequency which leaves you only CW type transmission of data. Not very fast over such a low frequency. Also most sonar transducers have a large current draw and a long recovery time after firing. BC From: Kurt Rosenfeld <apt2d at worldnet.att.net> Newsgroups: comp.robotics.misc Subject: Re: ultrasonic interference? Date: Sun, 02 Feb 1997 11:30:17 -0500 Organization: CCNY Lines: 31 Reply-To: apt2d@worldnet.att.net NNTP-Posting-Host: s22-pm04.snaustel.campus.mci.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 3.01Gold (Win95; I) Reckless wrote: ... > (Onat Ahmet) wrote: > > >How about using an old video head? It would have two or four > >air-gap transformers, all shielded from each other. I'm not > >sure if it would work efficiently at the lower frequencies though... ... > A really good idea, one which will probably make it onto one of my > robots. The trick here is to use a high-frequency carrier through the > rotary transformer. I know that in vcr's the frequency is in the MHz > range through the transformer, but I don't know how efficient it will > be at 100kHz to 500kHz where switching power supply components seem to > top out. > > BTW some rotary transformers (lower drum assy) in vcr's can have as > many as 8 separate coils. > > -Reckless the VCR rotary transformer idea is a really elegant one. i will try to find out about the low frequency roll-off of these strange devices. otherwise, the idea might still be possible using some modulation scheme to get the data across the X-former. what range of voltages do ultrasonic transducers operate with? -kurt
From: "JohnS" Subject: Re: Seeking ultrasonic air speed sensor design Date: 24 Sep 1999 00:00:00 GMT Newsgroups: sci.electronics.design David Smith wrote in message news:7semq4$43i$1@brokaw.wa.com... > I know HANDAR manufactures a windspeed sensor based on piezo transducers and > time-of-flight calculations. Theirs is two dimensional using three sensors > and trigonometry to resolve wind direction. My application is only seeking > to determine windspeed in one direction using two transducers alternately > sending and receiving to compare flight times. I know the fundamental > principle is to send a pulse and time how long it takes to get there. At > 40kHz I have calculated that I need to know arrival time within about 5 > degrees of phase, but windspeed could also shift the arriving signal by more > than one wave cycle. Ideally I would like a circuit that can generate a > voltage proportional to phase angle difference. Any suggestions? Any idea > where I can learn more about this technology? > You have found a classic problem with ultrasonic flow measurement. The cheap 40KHz transducers have a high Q and will give you a burst of sound with a slow rise and fall time over a number of cycles even if you only apply one cycle of excitation. This makes it difficult to measure the time of flight because amplitude fluctuations due to turbulance and deflection of the sound will cause your measurement to jump cycles. On the other hand if you measure phase you have the problem of a phase shift less or greater than one cycle. Possible solutions are to go to a higher frequency, lower Q transducer if measuring time of flight of a burst, or a lower frequency if measuring phase. You can get transducers up to 200KHz for air from Murata or Nippon ceramics. There was a good reference book on ultrasonic measurements a while back, not sure if it is still current, 'Ultrasonic Measurements for Process Control - Theory, Techniques, Applications' written by Lawrence Lynnworth, published by Academic Press Inc. The other thing is that if you are considering a commercial use just about everything you can think of seems to be covered by a patent. We turned up hundreds when looking at ultrasonic flow measurement about five years back. -- JohnS
From: Mark Borgerson Subject: Re: Seeking ultrasonic air speed sensor design Date: 25 Sep 1999 00:00:00 GMT Organization: Oceanographic Embedded Systems Newsgroups: sci.electronics.design JohnS wrote: <<SNIP > > where I can learn more about this technology? > > > You have found a classic problem with ultrasonic flow measurement. The cheap > 40KHz transducers have a high Q and will give you a burst of sound with a > slow rise and fall time over a number of cycles even if you only apply one > cycle of excitation. This makes it difficult to measure the time of flight > because amplitude fluctuations due to turbulance and deflection of the sound > will cause your measurement to jump cycles. On the other hand if you measure > phase you have the problem of a phase shift less or greater than one cycle. > Possible solutions are to go to a higher frequency, lower Q transducer if > measuring time of flight of a burst, or a lower frequency if measuring > phase. You can get transducers up to 200KHz for air from Murata or Nippon > ceramics. > > There was a good reference book on ultrasonic measurements a while back, > not sure if it is still current, 'Ultrasonic Measurements for Process > Control - Theory, Techniques, Applications' written by Lawrence Lynnworth, > published by Academic Press Inc. > > The other thing is that if you are considering a commercial use just about > everything you can think of seems to be covered by a patent. We turned up > hundreds when looking at ultrasonic flow measurement about five years back. > Perhaps, in these days of cheap processors and DSPs, the way to go is to send a chirp (a burst of sound which is swept in frequency as it is emitted) then use DSP to recover the frequencies and use that data to find phase and flight time. A fairly common technique in sonar these days. I guess the key will be the bandwidth which your transducers will accept. Mark Borgerson
From: James Meyer
Subject: Re: Seeking ultrasonic air speed sensor design
Date: 24 Sep 1999 00:00:00 GMT
Organization: AT&T WorldNet Services
Newsgroups: sci.electronics.design
On Thu, 23 Sep 1999 19:14:39 -0700, "David Smith" wrote:
>sending and receiving to compare flight times. I know the fundamental
>principle is to send a pulse and time how long it takes to get there. At
>40kHz I have calculated that I need to know arrival time within about 5
>degrees of phase, but windspeed could also shift the arriving signal by more
>than one wave cycle. Ideally I would like a circuit that can generate a
>voltage proportional to phase angle difference.
Forget the phase angle idea. Turbulence in the air will make it very
difficult to make phase measurements. Send a single sharp edged pulse and
simply measure time of flight directly. Average the times to reduce the noise
from the turbulence. That will give you the best chance of making a measurement
that means something.
Jim
From: Bertrik Sikken Subject: Re: Seeking ultrasonic air speed sensor design Date: 24 Sep 1999 00:00:00 GMT Message-ID: <37EBE9CE.CA5950ED@cs.utwente.nl> Content-Transfer-Encoding: 7bit References: <7semq4$43i$1@brokaw.wa.com> <1XzrN1ABt9aJm2A7ziQrgqagle5O@4ax.com> <7sgmv8$pii$1@brokaw.wa.com> X-Accept-Language: en X-Server-Date: 24 Sep 1999 21:16:07 GMT Content-Type: text/plain; charset=us-ascii Organization: University of Twente, Enschede, The Netherlands Mime-Version: 1.0 Newsgroups: sci.electronics.design David Smith wrote: > > James Meyer wrote in message > news:1XzrN1ABt9aJm2A7ziQrgqagle5O@4ax.com... > > On Thu, 23 Sep 1999 19:14:39 -0700, "David Smith" > wrote: > > > > >sending and receiving to compare flight times. I know the fundamental > > >principle is to send a pulse and time how long it takes to get there. At > > >40kHz I have calculated that I need to know arrival time within about 5 > > >degrees of phase, but windspeed could also shift the arriving signal by > more > > >than one wave cycle. Ideally I would like a circuit that can generate a > > >voltage proportional to phase angle difference. > > > > Forget the phase angle idea. Turbulence in the air will make it very > > difficult to make phase measurements. Send a single sharp edged pulse and > > simply measure time of flight directly. Average the times to reduce the > noise > > from the turbulence. That will give you the best chance of making a > measurement > > that means something. > > > > Jim > > Thanks for responding James. I had considered this approach but was not sure > how generate or read a sharp pulse given the characteristics of the > inexpensive piezo tranducers (i.e. they don't instantly respond to a signal > as a transmitter, nor do they instantly respond to a pressure wave as a > receiver). Viewing and capturing the wave form in a simple test with > tranduscers about 12 inches apart, I noticed that the received wave form > begins to oscillate and then gradually increases over the course of perhaps > 20 wave cycles, thus making it difficult to even determine the first wave. > The transducers are designed for best response at 40kHz. I assume this is > because they have a natural resonance at this frequency. > Perhaps I'm starting with the wrong type of transducer? > Do you or does anyone else have any experience with this? The slow response of the transducer is caused by its small bandwidth. You can increase the bandwidth of the transducer by adding coils and resistors to the transducer. That should improve the transient response, although sensitivity goes down. Take a look at this picture: http://www.massaprod.com/datasheets/graphics/tr89_40_data.gif Bertrik
http://catless.ncl.ac.uk/Risks/21.02.html#subj11 [DAV: Most ultrasonics are above the audio frequency band of 20 Hz to 20 KHz, but this is within that band ... though one could argue that this is ultrasonic]Wake up, your TV is talking to your bracelet Wed, 16 Aug 2000 09:51:39 -0700 A new system called Whispercode, designed by a New Jersey company for monitoring the effectiveness of TV advertising, will involve the encoding of commercials with inaudible, identifying signals that can be picked up by a small device worn by a participant (perhaps in a bracelet or keychain) and relayed to a nearby recording box that records the fact that the wearer was in the room when the commercial was broadcast. [It should be noted, though, the system can't detect whether the participant is awake, attentive, and not bored to death.] The company's chief executive officer says, "With Whispercode, we will finally be providing our clients with a true accounting of where their advertising money is going." (*The New York Times*, 15 Aug 2000 http://partners.nytimes.com/library/financial/columns/081600tv-adcol.html; NewsScan Daily, 16 August 2000
``Flavio Noca, who heads the research effort at Jet Propulsion Laboratory (JPL) in Pasadena, Calif. ... A colleague, Michael Hoenk, and I ... Our group supervisor, Brian Hunt, was about to start a nanotechnology effort at JPL and had been in contact with Professor Jimmy Xu, who's now at Brown University. Xu told him that he had just been able to manufacture perfectly ordered nanotube arrays. These nanoscale rods protruding from a surface were just what we needed for our sensors.
It was only in the next few days, by early June, 1999, that I realized that these nanorod arrays already existed in nature in the form of stereocilia. I was even more surprised to find out that cilia are to be found almost everywhere. ...
Several research groups are attempting to make arrays of micromembranes and assemble them on a single silicon wafer. The U.S. Navy wants such a chip so that it can make acoustic cameras for divers to detect mines in turbid water and at night. ...
... the nanotube arrays are simple and cheap to make, and can be assembled into large and dense arrays. ...
Q: You said stereocilia were "found almost everywhere" in nature. Where, for example?
A: Stereocilia are found in the cochlea of all hearing animals. They're also present in the vestibular -- or balance -- system of animals, from our inner ear to lobsters. Stereocilia populate the lateral-line system of fish for the measurement of water flows along the animal body and, presumably, also for identifying the direction of sound sources. Even nonhearing organisms, such as hydra, jellyfish, and sea anemones, rely on stereocilia to detect swimming prey.
... among our surprising results is a calculation showing the feasibility of using artificial stereocilia in an air environment in contrast to biological stereocilia, which are always found in liquids.
... fish's lateral lines, which are long canals coated with stereocilia on the side of animal. We think that fish use these for prey detection, localization, and identification, and also as flow-control devices. In the past few years, there has been an incentive to develop autonomous swimming robots, so sensors for flow control would be very useful. It's still not well understood how fish detect prey with lateral lines, but the Navy would be very interested in having a similar device to detect foreign objects in the ocean in a passive way. ''
The Navy's Surveillance Towed Array Sensor System (SURTASS) uses underwater microphones to listen for submarines. These "hydrophones" play a passive role in sub detection.
The new and controversial addition to SURTASS involves ... 18 Volkswagen-sized speakers ...
... this Low Frequency Active (LFA) sonar enhancement to SURTASS ...
"The sonar sound field around the transmitting ship will be 180 decibels up to 1 km away ..."
Sensors and Software develops portable ground penetrating radar (GPR) systems http://www.cellcomputing.com/third/pdf3/c2i-casestudy-sensors-software.pdf
Subject: Detecting a flying mosquito From: Scott Stephens Bats use ultrasound. Beam around 100Khz and listen for the doppler modulated return their fluttering little wings create. I thought a kewl yuppy toy would be a little phased array panel that fires compressed vortex rings to blow the wings off da 'lil suckers. Or perhaps just a brief squirt of water would do. I have a feeling that between a couple watts of ultrasound and the firecracker-like pop of the vortex cannon, it would be too obnoxious to keep around. Scotthttp://www.piclist.com/techref/postbot.asp?by=time&id=piclist/2001/06/14/135032a&tgt=&key=ultrasound&from=
Abstract: Crosstalk is one of the most severe problems in conventional mobile robot sonar sensing. ... a new approach which enables a mobile robot to fire its sonar sensors simultaneously while totally eliminating misreadings caused by crosstalk or external ultrasound sources. ... using appropriate pseudo-random sequences together with a matched filter ...
...
... the Barker code is not applicable for our purpose since the number of different sequences having the same length is insufficiently small.
Instead, we simply use stochastic sequences, i.e. random noise. ...
...
... the advantages of our approach ...
1. Each sensor Si is able to identify its own echo. ... [rejecting] crosstalk or external ultrasound sources ...
2. Each sensor Si is able to identify the echos of all the other sensors. ... the possibility [of] triangulation in order to compensate for a sonar sensor's poor laterl resolution.
3. ... fire all sonars simultaneously ... speedup ...
...
This paper also has some interesting books in its list of references.
``The breakthrough, a programmable image processor, adds the power of a supercomputer to an ultrasound imaging system and provides the ability to use software instead of hardware to develop new diagnostic applications. The result will be clinical applications developed more quickly and cost effectively. ...
The ultrasound image processor, jointly developed by the UW Image Computing Systems Laboratory and by Siemens Ultrasound Group ... ''
Propagation of Sound in the Ocean http://www.beyonddiscovery.org/content/view.page.asp?I=223
...[in 1912] the Titanic tragedy motivated the Submarine Signal Company of Boston (now part of Raytheon Company ) and others to begin developing more active devices that would warn of icebergs and other navigational hazards. Within a week of the tragedy, L. R. Richardson filed a patent with the British patent office for echo ranging with airborne sound, following a month later with a patent application for the underwater equivalent. The first functioning echo ranger, however, was patented in the United States in 1914 by Reginald A. Fessenden , who worked for the Submarine Signal Company. Fessenden's device was an electric oscillator that emitted a low-frequency noise and then switched to a receiver to listen for echoes; it was able to detect an iceberg underwater from 2 miles away, although it could not precisely determine its direction.
...
Speaking generally, the ocean is divided into horizontal layers in which sound speed is influenced more greatly by temperature in the upper regions and by pressure in the lower depths. At the surface is a sun-warmed upper layer, the actual temperature and thickness of which varies with the season. At midlatitudes, this layer tends to be isothermal, that is, the temperature tends to be uniform throughout the layer because the water is well mixed by the action of waves, winds, and convection currents; a sound signal moving down through this layer tends to travel at an almost constant speed. Next comes a transitional layer called the thermocline , in which temperature drops steadily with depth; as temperature falls, so does the speed of sound. However, at a point roughly 600 meters to 1 kilometer (0.4 to 0.6 miles) below the surface, further changes in temperature are slight (the water the rest of the way to the bottom is effectively isothermal). Now the dominant factor influencing the speed of sound is the increasing pressure, which causes sound to speed up.
...
... because of the laws of refraction , sound waves can be trapped effectively in a narrow channel that straddled a region of minimum speed where the bottom of the thermocline met the top of the deep isothermic layer. ... a sound wave traveling obliquely through the thermocline ... Sound introduced into this sound channel thus could travel thousands of miles horizontally with minimal loss of signal. ...
... Christopher Clark is interested in whether whales, like dolphins and bats, echolocate. Rather than bounce sound off objects a few yards away, however, whales send their sound pinging off geologic structures hundreds of miles away. The theory that whales use their own sound to get their bearings has been around for some time; now the data from SOSUS tracking give Clark some compelling circumstantial evidence to support it. ... Clark hypothesizes that whales use sound not just to communicate, but to navigate, that is, they map the ocean acoustically to find their way around in it.
sound and hearing http://hyperphysics.phy-astr.gsu.edu/hbase/sound/soucon.html has data on the speed of sound through many materials, and equations for speed vs. temperature, speed vs. density, and other material properties.
The speed of sound in sea water http://www.akin.ru/spravka_eng/s_i_svel_e.htm W. D. Wilson's formula has a complicated-looking empirical equation that calculates c(S,T,P), the speed of sound in sea water given its salinity, temperature, and hydrostatic pressure. I think perhaps there's a simpler way to express this.
Underwater acoustics http://www.npl.co.uk/npl/acoustics/techguides/ has an even more complicated-looking equation that calculates c(S,T,P), and a simpler equation that calculates c(D,S,T), given depth, salinity, and temperature. There's also a formula for converting pressure to depth and back -- -- it includes g(theta), a formula for estimating the gravitational force given the latitude theta.
... Recent work indicates that a 3 dimensional array of some 512 microphones [ 8 x 8 x 8 ??? ] is required to exploit fully the potential for improved quality in signal aquisition. ... only with a real-time system can one truly understand the real issues.
...
... the Microphone Module... may be used as a standalone system that is set up and driven from an 82c55, 25-bit interface to a personal computer. In this mode ... a single ADSP21020, 33 MHz signal processor with up to 16 microphones of input. ... useful for other projects ... packaged as the Brown Megamike. ...
The Analog Devices AD1846 is a 16-bit, 2 channel A/D and D/A system ...
... the single ADSP21020 signal processing system can do source location and DSBF beamforming in real time for the 16 microphones.
...
... uninhibited parallelism ...
... 96 DSPs ...
[FIXME: can I buy the Brown Megamike ? Is there any other comparable off-the-shelf system I can buy ?] [FIXME: computer_architecture.html ?] [Given the improvements in DSP and CPU over the years ... what would I do differently if I needed to do something similar today ? Obviously, use faster DSP chips ... ... I'd at least consider daisy-chaining the 32 microphone modules ... perhaps consider FireWire bus ... ... ]
``Beamforming: A versatile approach to spatial filtering.'' article by B. D. Van Veen and K. M. Buckley in _IEEE ASSP Magazine_, April 1988. ``An overview of beamforming algorithms is provided by Van Veen and Buckley'' -- recc. Harvey F. Silverman [FIXME: read]
"design of DSP software and hardware in the field of smart sensors, depth sounding, sonar systems and underwater acoustics."
A simple way to prevent train crashes caused by buckled rails has been developed by engineers at the University of Illinois at Urbana-Champaign. The technique, developed by Richard Weaver and Vesna Damljanovic, gives an accurate measure of the stress in a rail simply by forcing it to vibrate and measuring the size of the vibrations.
...
The frequency of the induced vibration is related to the stress in the rail, Damljanovic says. "A length of track is like a violin string. It will change its pitch as you increase the tension." The team has identified five critical modes of vibration, the most important of which is side-to-side bending.
Subject:
Re: Ultrasonic sensors - what kind of range and accuracy is possible?
Date:
01 Dec 1998 00:00:00 GMT
From:
leafs at ozemail.com.au (Ian Wilson)
Reply-To:
leafs@ozemail.no.spam.com.au
Organization:
OzEmail Ltd.
Newsgroups:
sci.electronics.design
References:
1
edward.o.clapper at intel.com (Ed Clapper) wrote:
>Just need some general info about the practical range of ultrasonic sensors and
>the various factors that affect it.
>Other questions are:
>1. Can emitter/detector pairs be frequency-adjusted or are they fixed?
The small cheap devices are usually quite high Q so can not be used
much of freq (a few % and you really see degradation in returned
signal). More expensive devices are broader band (e.g. the unit used
in the Polaroid demo kit is quite broad band but make a click when
fired).
>2. What is the directional relationship of an ultrasonic sensor pair?
Very roughly the overlapping partion of their beamwidth - assuming no
highly reflective objects that provide multi-path returns. The
smaller the device relative to its frequency of operation the wider
the beamwidth - again roughly. a typical 12 mm device running at 40
kHz has beam width of about 40 degrees from memory.
>3. What is the variational effect of material properties (for the sensed
>distant object) for sensing distances? Is there a surface property that
>specifically affects the sensing capability?
The amplitude of the echo is affected but not its time delay. If you
look for a simple threshhold on your return then you may see
non-specular objects "dis-appear" faster than smooth plane reflectors.
>4. What is a typical ultrasonic frequency range for measuring distance?
I use a simple Murata pair of seperate TX and RX and easily get 5 ro 8
metres with out doing too much. This does use an MC1350 VGA to even
out the amplitude of the return and then goes into a ADC and a micro
for further filtering and processing. We have never needed to see how
much further we can go - and we are driving the TX at about 1/4 max.
>5. Accuracy? (%)
If the Q of the TX and RX is too high then you can not see the edges
of your burst very well and this introduces some error. But we find
that without any special processing we get 5 cm accuracy which is
heaps accurate for us. Never actually measured the accuracy - but I
suspect that it could be made much better.
Also note that for 40 kHz the 1/4 wave length in air is about 8 mm (I
think - something like this) and we foound that a small amount of sway
in the object being detected could easily change the amplitude of the
return by more than 15 dB. Not a problem in our case (and we have two
RX anyway) but might be an issue for some.
Ian Wilson
--------------------------------
Considered Solutions
considered@ozemail.no.spam.com.au
(do the no spam thing to make a valid email address)
Original Author: David Cary.
This page started: 1996 Aug 8 and has backlinks
David Cary feedback.html
Return to index // end http://rdrop.com/~cary/html/ultrasonic.html