Using Multiple Ultrasonic Sensors

The easiest way to do it is to fire the sensors sequentially.

You probably need to allow a time between sensors firing of at least 10 x transit time of the normal maximum detection distance. So if you can detect at 2m you need to allow 320/40 = 0.125 seconds between one sensor and the next sending pulses.

(I do mean to divide by 40 in the example because the distance is to the target and back again.)

MK

As Michael indicates, this is normally handled with sequential activation. This slows response time so sometimes it is not ideal. Best speed is achieved by using the return from one sensor to trigger the start of the next device.

Another method is to use different frequencies for each sensor direction. This requires different narrow band sensors for each direction.

It may be also possible to send from a single omni directional transmitter and have separate receivers for each direction. This could still result in a reflected signal arriving before a long return, depending on geometries involved - that needs to be assessed.

Many of the ultrasonic sensors that are easy to buy have a single circuit board with a transceiver pair.  I searched online for a simple way to change the frequency on such boards but didn't find one right away.  Is there a method that you've seen for this?

Also I like the sequential trigger idea.  An interrupt from receipt could read the timer and initiate the next device, in a ring.  Thank you!

Thanks for the 10x figure, michaelkellett!

The low cost sensor pairs with electronics use tuned transmitters and receivers so they can only work at the designed frequency,

I've never seen  a low cost system that used different frequencies to separate sensors.

The 10x figure is a good rule of thumb, based on trying to measure reversing sensors and arrays of them in lab conditions. Once you get your system working you can experiment with  a bit of tuning of the interval.

The problem with triggering in a ring is that depending on the target the first echo may well not be as big a a later echo from a different target and that the varying time between measurements may well cause problems at a system level.

MK

Ultrasonic sensors typically have a resonant frequency that can't be changed much.

A typical SRF04 or other sensor in the SRF series operates at 40 KHz.

The Maxbotix sensors operate at 42 KHz.

The URM06 (Sen151, Sen152) operate at 49.5 KHz.

I have used piezo buzzers as transmitters and receivers as well - they are available in a wide variety of frequencies, but tend to be in the audible spectrum.

I suspect the next thing to do is to actually write the software and tune it with a test rig. Maybe I'll get to it this weekend!

I'll post the Arduino or CircuitPy sketch up here for discussion when I do.

Thanks for the hint re: resonant frequencies. I think I'm stuck with the other methods we've discussed. I remember pursuing this on my own in the past and your post jogged my memory that Matrixbot are at 42KHz.

Let's see how a test system works.

Tony

Here is an article from MaxBotix on running multiple sensors.  Summary: the ring method is their preferred/most reliable way to handle multiple sensors. I have 3 Maxbotix HRLV-EZ4 devices, so I can chain them like the diagram!

michaelkellett and dougw  thank you for the fine discussion.

Good work. Waiting to see how this turns out.