> This is not really a PIC question, but the project > itself will be controlled by one. > > I need to measure a distance from a 'sensor' to a wall > which will have a rough surface. The measuring > distance should not be any more than 5 inches. The PIC > will be used to control the 'sensors' movement across > the wall and to gather the distance data, and thus I > should be able to map the surface contour. > > It would seem a reasonable problem except that the > wall will be under murky water. Also the wall may have > mud sticking to it at places so this means I could not > have a sensor which actually touched the wall. If this > happened then mud may attach itself to the sensor > giving me inaccurate data. > > Any leads on a type of sensor would be appreciated. > > eg Microwave, Laser etc. > > Tony ------------------------------------------------------------------- Tony, I see that you probably need a more thoroughgoing response to your question. As has been pointed out, the most appropriate vehicle for this application is ultrasound. This is due to the ability of ultrasound to penetrate the various layers and media you mention and still return a usable echo. The trick is to be able to sort out all the different echos due to the different layers and, ultimately, to be able to distinguish the one layer you are looking for, the surface behind the layer of mud. To do this, you need an array of ultrasonic transducers or, alternatively, a pair of transducers whose separation distance can be varied and, depending on your needs and the geometry of the target, that can be rotated about the line of position between the sensor and the target. The most powerful way to understand how the target information is extracted or deduced is in terms of the abstract two-dimensional Fourier representation of the target (with intervening mud layers, etc.). What you need to do is essentially the same process that is used, for example, in long baseline interferometry, an example of which is the radio interferometer (VLBA) in Socorro, NM. If you are familiar with it, you will recall that it consists of an array of antennas that can be combined electronically in different spacings and at different angles around the line of observation. In your problem, the ultrasonic transducers take the place of the radio antennas, but the underlying mathematics is the same. The various combinations of spacings and angles produce samples of the Fourier representation of the target. With sufficient samples, and with due regard for the equivalent of the Nyquist criterion (for a spatial, rather than a temporal, function), the Fourier representation can be inverse transformed to reconstruct the target with its detailed structure (here the ability to discriminate the various confounding layers). Essentially the same process is used for image reconstruction in the medical CAT scan. Beyond this sketch, I can only say that this is a very complex problem and that to be any more specific one would have to take a serious look at the details and the level of performance that is required. Once these are established, one would have to do some serious mathematics to derive an algorithm, akin to the VLBA or the CAT scan, but appropriate to your particular application. If you conclude that this is the kind of "horsepower" that you need to bring to bear, I can offer you some help. This is an area in which I specialize. It would be appropriate at that point if you contacted my by private e-mail to discuss it further. I hope this is helpful. --- Warren F. Davis ================================================ Davis Associates, Inc. 43 Holden Road West Newton, MA 02165 U.S.A. Tel: 617-244-1450 FAX: 617-964-4917 Visit our web site at: http://www.davis-inc.com ================================================