I’m not sure whether it is appropriate to post this thread in this section, since most of the topics are related to XBee and other wireless systems.
I’m working on a project which requires to use an ultrasonic system. I have to objects and want to measure the distance between the two using ultrasound waves. One object has an emitter which sends out an ultrasound wave, and the other object has the receiver. The travel time will be recorded. The product of the travel time and speed of sound will yield the distance.
I was trying to find ultrasonic transducers online to accomplish this task. However, most of the products have certain beam patterns so they can only focus in an area. My requirements are:
(1) The emitter (receiver) can send (receive) ultrasound waves in 360 degrees.
(2) The emitter has a wide range, ideally 100 ft. The receiver should be able to detect the sound wave 100 ft away from the emitter.
(3) The result should be accurate. The calculated distance should be actual distance +/- 5 in.
I’m not sure where I can buy an emitter or receiver without a beam pattern. I only found one transducer with a range of 60 ft, but it is still not good enough. This is not my major study area so I have to learn everything from scratch, but the project has to be done as soon as possible. I’d really appreciate any help and suggestions.
Thanks!
I don’t think you’ll find an ultrasonic “speaker” that has an onmi-directional pattern. If they exist, they’ll be pretty rare as there’s not much call for such a device. You can get close to the pattern you want by at least one of 2 methods (that come to mind). Either you rotate a transducer mechanically to scan the 360 deg desired or you can use mulitple transducers in a phased array to cover the 360 deg. I guess you could try to make a transducer yourself with the desired onmi pattern but I think that’ll be hard to do quickly. The former may cost less but has the disadvantage that rotation rates limit how often you can determine range. The latter costs more (more transduscers) and perhaps requires more power (unless you sequence the transducers which gets you back to the the limitation as scanning). If each transducer had a beamwidth of 30 deg, you’d need 12 of them to cover 360 deg if you did sequential lobing. I’ll have to think a bit to figure out if some other arrangement (a phased array) could reduce the count from 12. And you could use a hybrid system, a active phased array on a rotating platform to cut down on the “blind” time.
You’re “lucky” in that separating the transmitter and receiver allows you to optimize each. I’l guess that 100 ft is not a problem. Think about the “big ears” that are used to collect sound at sporting events. The large reflector collects more sound than a small one would, and directs it to a microphone, thus increasing the system’s sensitivity. You could use the same concept. How large can you make your “ear” ? How accurately can this “ear” be aimed at the transmitter ? Note that most of the hobby transducers are intended for a monostatic range finding system. You are a bistatic case and so aren’t dependant on the, perhaps poor, reflectivity of the target. Be careful in how you interpret the specs.
One thing you haven’t mentioned is how you’re going to synchronize the transmission with the timing of the reception. That is, how will you know at the receiver when the transmission was initiated. What you measure with the reciever is when the sound was received vs some timebase. Either you need synchronized clocks at the transmitter and receiver, kept to better than 0.5 msec synchronicity, and a schedule of transmission times … or some other way to know the time the transmission was sent (ie - a pulse of light or RF or ???).
Thank you very much for your reply!
We may use multiple transmitters to cover the angular range. 360 deg is probably more than enough. 180 deg should be acceptable. The ultrasonic transducers are not that expensive. The problem is whether the board could power multiple transmitters at the same time. If the transmitters are in parallel, it is critical to have enough voltage on them.
The tricky part about the receiver is that it is moving while receiving signals. There is no good way to set the receiver aiming the transmitter, so I don’t know how to position the reflector. This will be the limiting factor to optimize the receiver. We have two op-amps on the receiver board to amplify the signal.
We have Zigbee on board to transfer data from the transmitter to the receiver. I am mainly in charge of the hardware, and a team of programers are working on this issue. Do you have and suggestions on this?
Thanks!
On the Xmit side you can sequence the burst to come from only 1 speaker at a time. If you had 6 speakers for 180 deg coverage, your send out 6 bursts to get 1 distance measurement. For onmi coverage on the receive side you have some options depending on the physical size you can have at the “target”. The simplest approach would be to arrange a number of microphones to give you 360 deg coverage (not unlike the Xmit side) and then combine their outputs in a summing ciruit with your op-amp. This can be a little tricky in that when 2 mics receive the same burst you want to be sure their outputs combine constructively and not cancel each other. I don’t know how phase matched the mic outputs can be or how stable they’ll be over time/temp/etc such that you could phase match w/circuitry prior to combining. I guess you could sequence the listening too, perhaps combining the outputs only from non-overlapping mics and having 2 bursts from each speaker before going to the next speaker.
How often must you obtain a distance measurement ? How likely is your environment to provide strong echos ? How fast is your “target” moving and/or turning ? If transducers are cheap $s then I’d get a few and do some experiments to see how far they work together and how matched speakers and mics might be. You might get lucky and be able to implement a simple solution.