5
\$\begingroup\$

enter image description here

I understand how these ultrasonic sensors work on the basic level: They send out ultrasound, which is then reflected by an object and is then sensed by the receiver.

But the receiver and the transmitter are placed side by side. Won't the receiver immediately pick up the signal as it is sent? How does ignore the signal as it is sent, and record it only as it returns after being reflected by the target?

\$\endgroup\$
  • 1
    \$\begingroup\$ The receiver is 'blanked' until the transmitted pulse is finished typically so the receiver is basically out of circuit until that point. That is for a pulsed system; other techniques for very short range include modulating the signal. \$\endgroup\$ – Peter Smith Mar 25 at 16:17
  • 1
    \$\begingroup\$ The signal sent as pulses. These sensors are not good for ultra-short ranges, so the signals received "immediately" are filtered out. \$\endgroup\$ – Eugene Sh. Mar 25 at 16:17
  • \$\begingroup\$ @Peter Smith Interesting, what kind of modulation is used for short range? \$\endgroup\$ – S. Rotos Mar 25 at 16:30
  • \$\begingroup\$ The Polaroid onestep autofocus camera used 50KHz sin-bursts, if I recall rightly. Whereas human-body internal-sonograms use 2MHz-5MHz bursts. \$\endgroup\$ – analogsystemsrf Mar 25 at 16:51
  • \$\begingroup\$ @analogsystemsrf Really? They use acoustic sensors to measure distance? Well, this won't perform well through a window... \$\endgroup\$ – Eugene Sh. Mar 25 at 16:56
9
\$\begingroup\$

But the receiver and the transmitter are placed side by side. Won't the receiver immediately pick up the signal as it is sent? How does ignore the signal as it is sent, and record it only as it returns after being reflected by the target?

In a pulsed system (the most common type), the receiver is blanked during the period of the transmit pulse so it does not respond during that time.

This is the same in pulsed Sonar and Radar and determines the minimum distance that can be resolved.

\$\endgroup\$
5
\$\begingroup\$

There are monostatic and bistatic ultrasonic transceivers.

Monostatic

Monostatic transceivers are single transceivers that do the transmitting and receiving. As long as the transmission stops before the reception starts, you can do measurements. The minimum distance that you can measure is limited in this configuration, but it's smaller and cheaper to implement. Speed of sound is roughly 343m/s. If we assume there is an object 1 meter in front of the sensor, it will take the signal 6 milliseconds to bounce back. That's the window that you're allowed to drive the transceiver and stop it from vibrating at that distance. At 0.1 meters we have 0.6 milliseconds etc. It becomes extremely hard and frankly impossible to drive a proper signal in that timeframe.

If you want to get technical: while transmitting the membrane of the transceiver is being vibrated at a specific frequency. If you'd try to read(receive) anything while it's doing this, you're just going to get a oversaturated signal back. When you stop actively driving the transceiver it will stop ringing in a short time, at which point you can start listening to what it's receiving.

Bistatic

A bistatic transceiver(the one in your picture) uses a separate transmitter and receiver, which means it can receive while transmitting. The minimum distance isn't limited like the monostatic configuration. If you have the space you should go for this configuration. It's easier, better performing and also widely available. The parking sensors on cars use the monostatic configuration for example.

Your question

You're asking why the receiver isn't immediately picking up the signal from the transmitter. While it's true that cheap implementations basically shut the receiver off while transmitting, this is not required. And I'd argue that if you do that sort thing in your design, you should go for a monostatic configuration. Otherwise you're just wasting space and parts.

The directivity of ultrasonic transceivers is fairly precise and won't affect a receiver that's next to it as long as you physically separate the vibrations from each other(A PCB won't effectively transfer kHz frequencies) and don't point it at each other. Most transmitters cap out at ~90 degrees field of view@40kHz(most common) which is enough to keep it from affecting the receiver.

For example, this one has a FoV of 80 degrees with the following graph in the datasheet:

enter image description here

See how it tapers off at the extreme edges?

\$\endgroup\$
  • 2
    \$\begingroup\$ Sorry, but a lot of your beliefs here are practically mistaken - crosstalk between transducers on a little PC board is a very real problem, remember the transmit pulse is many orders of magnitude stronger than the receive one. Also the kinds of power setups and circuitry in general used on the cheap ones provide another source of crosstalk. Further, the term "bistatic" does not really apply to situations where the transmitter and receiver are effectively co-located, but rather to those where there is substantial spacial separation. \$\endgroup\$ – Chris Stratton Mar 25 at 19:19
  • \$\begingroup\$ @ChrisStratton I did mention that the cheap boards tend to compensate for cross-talk by simply shutting off the receiver when transmitting. It's an implementation detail that isn't required if you have a proper setup that isolates the transceivers on a mechanical level. Furthermore, the distance between the transducers is relative to the actual distance you're measuring. A few cm is quite a lot when you're only measuring a few meters of distance at the power required to drive the transmitter. \$\endgroup\$ – Tryphon Mar 25 at 19:37
  • 1
    \$\begingroup\$ Again, no. Actually look at one of these and you'll see that the distance between the transducers is negligible at the useful ranges. Actually try to build a variation which can operate without blanking for the duration of the transmission pulse and transmit transducer ringing and you'll realize that is a lot harder than you imagine as well. \$\endgroup\$ – Chris Stratton Mar 25 at 19:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.