The purpose of this article is to shed some light on the issues related to the use of ultrasound as a wireless means to solve a design problem. In addition, the author found that the development systems available at reasonable cost (under $250) did not provide enough "hooks" to control pulse width, frequency, and sampling intervals.

I will use a real-world design using these issues as an example. The same principles apply to many applications of ultrasound. For example, it is obvious that audio and digital communications could be implemented with ultrasound in some applications just as a link designed using RF. For that matter, there have been several attempts (albeit unacceptable ones) from a consumer standpoint to produce "virtual speakers." My understanding is that the encounter with the "speaker" area can be somewhat painful to the listener if he or she is not situated exactly right. This is probably because of the relatively high power levels necessary to have node/antinode molecular pressure interaction in the audio spectrum produced by two independent ultrasound beams.

I will list several vendors, parts, and information sources at the end of this article. Keep in mind that this is a concrete example for a specific goal. Issues such as the speed of sound as a function of temperature, the transducer characteristics (focus, center frequency, etc.) are not addressed, but may be important in your particular application.

Some Basics

If we measure the time of flight of a pulse—which can be sound, RF, light, etc.—with reference to another object separated by some distance, then D = R × T. If a round trip to the object and back is necessary, then D = (R × T)/2. Sound travels approximately 1100 feet/second. An actual measurement will show that sound travels about 0.89 ft./ms. Therefore, if an object is 1 ft. away from our transducer, the round trip takes about 1.78 ms. It would seem that we could then sample about every 1.78 ms, or at 561 Hz.

It is simple to log the launch time of a pulse, log its return time, and calculate the distance.

Location Accuracy as a Function of Velocity

However, if the object is moving, things become more interesting. Obviously we would like to launch our pulse as frequently as possible to get the highest accuracy. The problem is that as the distance increases, so does the round-trip time. With most basic ultrasound systems the maximum distance to an object and its TOF for echo return sets the limit as to how fast we can sample, and therefore the accuracy of the distance measurement IF the object has a velocity.

Example 1

If we increase the distance from 1 ft. to 5 ft., the round trip time becomes 8.9 ms (0.89 ft/ms x 10 ft.).

The maximum sampling interval must now be greater than about 112 Hz (1/0.0089 ft./s = 112 Hz).

Increasing the distance to 10 ft.:

D = 10 ft.
T = 17.8 ms (2 x 8.9 ms)

Then,

Max. Sampling Interval = 56 Hz (1 /0.0178 ms).