Converting 2 sonar transducer based circuit into 1 transducer? best ICs to use?

[hood_electron]

Found this reverse engineering of the popular HCSR04 on youtube, but i need to be able to use a single transducer for both the receiving and the transmitting. Planning to use this in a 7X1 array. The outputs of the MAX232 are inverting so im a bit unsure on how to receive from the same transducer im transmitting from without blowing up my delicate little op amp.


And speaking of the op amp, is there any other filtering needed for the transducer other than the simple RC filter? also should i use some other ic (op-amp) than the one in this picture, im looking to get a reading of atleast 1mm precision (0.1mm if possible). The transducers will be harvested from HCSR04. I will be using STM32F11s (black pills) for the ADC and calculations.


Im very new to analog circuits so please be kind on my mistakes and incompetence

[MasterT]

 Speed of sound 340 m/s
 Freq. 40 kHz
 Resolution =  340 ÷ 40,000 = 8.5 mm

[hood_electron]

isnt the resolution you can get from the transducer based off your sampling rate?

[ledtester]

Found this reverse engineering of the popular HCSR04 on youtube, ...

What's the link?

This looks like it matches what you have:



The MAX232 is used to drive the transmitter harder than what the MCU can do on its own - like +/- 9V instead of 0-5V.

This distance sensor looks like it uses just a single transducer:

https://www.diymore.cc/products/dc-5v-waterproof-ultrasonic-module-distance-measuring-transducer-sensor-module

[hood_electron]

this is the video.

And yes, the schematic is indeed the of the sonar sensor im working with. Where did you get this?

Any idea on how i can make this into a single transducer design?

Thanks for the help

[ledtester]

I guess I would first try something like this:



This doesn't use the MAX232 at all but drives the transducer directly from two of your MCU's GPIO pins.

R17 and C15 are part of the TL074 circuitry in the schematic I posted above.

The way it works is this:

- to send a chirp you configure the GPIO pins to be outputs and then toggle them at 40KHz to drive the transducer.
- to use the transducer as a receiver set GPIO B to an output with a LOW output and configure GPIO A as an input. The received signal will pass through the op-amp circuit via R17 and C15.

The emitted chirp won't be as strong as you can get with the MAX232 but you'll get something.

[ledtester]

Doing some more research I've found details of how the single-transducer device I linked to in reply #3 works...

- It's called a JSN-SR04T and you can find a similar product by the name of AJ-SR04M
- It uses a transformer to drive the transducer

Here is the link for the AI generated search result from Brave's search engine:

https://search.brave.com/search?q=JSN-SR04T+schematic&source=web&summary=1&summary_og=1032678a29d69f84d83556

Also, more details of how to drive the transducer from a transformer can be gleaned from the datasheet of the PW0268 which is a specialized ultrasonic ranging IC:

https://www.farnell.com/datasheets/1714673.pdf

In the upper right you can see how the transducer is driven. And for receiving the signal comes through the 3.9K resistor / 470p cap.

[MasterT]

isnt the resolution you can get from the transducer based off your sampling rate?

All of the ultrasonic distance meters I've seen based on comparator triggering, when reflected wave arrived. So resolution is limited by wavelenght.

 But I done a  project (~10 years ago) where I sampled using ADC, than performed DSP and get a phase. Phase measurements significantly more accurate, 0.1mm achievable. But my intuition tells me that results 'd be biased by surface properties of the object. Likely size of the object and angle to falling acoustic wave also 'd affect, so more research need to be completed.

Cheap SR04 sensors have mark "T" - transitting and "R"-receiving, probably there is some technological differencies.
If one transducer only nedeed that datasheet must state it could be used as transmit-receive.

[hood_electron]

would a high frequency H-bridge work? im really not wanting to tune 14 transformers without an oscilloscope. Would also allow me drive them as hard as the boost converter source

[Andy Chee]

If you want to drive 14 ultrasonic emitters, I would suggest winding a transformer like this (ignore the SCR, concentrate on the transformer):

https://www.daenotes.com/sites/default/files/article-images/scr-triggering-series-parallel.PNG

[hood_electron]

the transducers need to work independently, this wont work unfortunately

[Andy Chee]

Can you describe your entire project?  Why do you need 14 transducers to work independently?

[ledtester]

In this EEVBlog video:

EEVblog #1315 - Ultrasound Probe Extreme Teardown!
https://youtu.be/lmy8J8n9wPU

Dave mentions the LM96550 Ultrasound Transmit Pulser IC:

https://www.mouser.com/datasheet/2/405/lm96550-442390.pdf

and that has a reference to the LM96530 Ultrasound Transmit/Receive Switch:

https://www.mouser.com/datasheet/2/405/lm96530-312024.pdf

which contains this block diagram:



and a description of how it works on page 10.

[hood_electron]

I need to have two 7x1 arrays of transducers, each capable of transmitting and receiving signals. One of the arrays will be stationary on a car and another one directly on top of that one on a servo.

I want to achieve a few things with that. One is mapping the angles and contours of objects in front of the array. Planning on using the top array for transmitting and the bottom array for recieving. At any time only one of the transducers will be transmitting (and others listening) so i dont have to deal with phased array bs. Can also use the top array and rotate it to map objects around it 360 degrees.

Another is being able to track down objects sending out sonar pings. Use the top array to receive the pings and rotate and adjust till the middle transducer receives the signal the earliest (thus being closest to it and thus the whole array pointing at the object). Check if the pair around the middle one receive it second and the one next to those receive it next and so on.

Ill be using ESP32s/8826s and ESP-Now to communicate with the sonar pingers so i can tell which one to fire (can also get the moment it fires and use that time of flight to figure of the distance between the array and the pinger).

Everything else is just basic sonar project you find on youtube and such.

Ive also figured out how to do the whole not blow up your amplifier with 24v. There are T/R switchers made exactly for this purpose, Im planning on using the MD0100 by microchip.