ultrasonic transmitter receiver design

Thread Starter

aliyesami

Joined Sep 12, 2009
21
Last edited:

DickCappels

Joined Aug 21, 2008
10,152
The transmitter part looks like it should work to some extent. There is a question as to how much drive you need. Most likely it will take experimentation with a complete system to find out.
 

MisterBill2

Joined Jan 23, 2018
18,176
Do you have any words relating to the application of the circuits in the link? Like the size and the material of the pipe used, and the flow rate in that application. Also, what sort of transducers were usedin that application. Because it connects to an arduino package, it is probably a hobby type of application and may not be very accurate. Of course, the post did state to detect water flow, not measure water flow. So it may be quite adequate. Possibly.
 

Thread Starter

aliyesami

Joined Sep 12, 2009
21
the size is 2 inch dia , the material is PVC and the flow rate is 60 Gal/min , clean water with no solids .
I only want to detect the flow on or off , i dont want to calculate the rate of flow . If the water flowing i want to detect it.
 

Thread Starter

aliyesami

Joined Sep 12, 2009
21
is there a ultrasonic transducer that acts as transmitter and receiver both? upon reading water flow meters I think i need to send signals in both direction , upstream and downstream using same set of sensors .
the other option is to just bounce the signal perpendicular down the pipe and read the reflected wave ,but I am not sure if that will change with water flowing or not.
 

MisterBill2

Joined Jan 23, 2018
18,176
I think that there are two transducers and the water velocity alters the phase shift of the signal, sort of like doppler radar.
I suggest visiting the website of a company that makes the sensors and understanding the principle before going very far with any hardware. But if all you need to do is detect a flow or no-flow condition, that could be done by listening for turbulence noise. That could utilize a much simpler scheme.
Weknow the pipe diameter and the mass flow rate but we do not know the pressure or the temperature or how much the flow rate may vary and still be "OK". Also, is the flow driven by a pump, or gravity, or air pressure on top of fluid in a tank? That affects the noise level.
 

Thread Starter

aliyesami

Joined Sep 12, 2009
21
i dont want to use noise based sensor as its prone to false positives , biggest one is the lawnmowers .
i want to detect flow due to pump pumping water at rate of 60 gal/min.
I have these two designs in mind which one is more feasible ?
but for these I would need a ultrasonic sensor that would act both as transmitter and reciever .
 

Attachments

MrChips

Joined Oct 2, 2009
30,714
What research have you done to find an ultrasonic sensor which can detect fluid flow?
What physical and scientific principles would be required for such a sensor to be successful?
 

Thread Starter

aliyesami

Joined Sep 12, 2009
21
I have read that I need to send a signal in the flow direction and measure the delay and then send the signal in opposite direction and measure the flow. The difference in the time will give me the indicator if water is flowing or not.
Then there is another method of doppler sensor which is more suitable for water with solid particles which is not applicable for my case as I am pumping clean water .
my confusion/question more is on how to mount these sensors on the pipe at an angle and still be able to send the signal through as illustrated by my drawing.
if i mount them flush to the surface as shown by dia B then i have to put some kind of angled surface inside the pipe to direct the sound waves which I want to avoid.
I am good in micro controller interface so once i figure out these sensors i can interface them easily to esp32.
 

MisterBill2

Joined Jan 23, 2018
18,176
Certainly there are quite a few ultrasonic flowmeter systems available, but none of them are cheap. The big challenge with ultrasonics is the transducers.
What has not been mentioned at all is the pressure of the water in the pipe. Not that it will affect the sound velocity, but that it will certainly affect the mounting of the transducers.
So now I am taking the liberty of suggesting a more classical approach, which is to measure the pressure difference as the water flows past some constriction in the pipe. This would be similar to measuring the voltage developed across an ammeter shunt in an electrical circuit.
In fact, the TS could even monitor the pressure increase across the pump, and not need to do any installation of ultrasonic transducers at all. When the pump is not pumping and the flow is not much, or possibly none at all, the pressure across the pump is zero. That is quite different from the gage pressure inside the pipe. The differential pressure can be measured either with a differential pressure transducer or with two gage pressure transducers, one before and one after the pump.. Much simpler, much easier, and the information will also be useful for other purposes as well.
 

Janis59

Joined Aug 21, 2017
1,834
Let make a look into LM567 datasheet. Seems it is rather near what You want, yet it is designed for IR frequencies but in sound range. Use a smaller caps and be in US frequencies.
 

MisterBill2

Joined Jan 23, 2018
18,176
Interesting but not informative as far as just how it is done, as to what power level is needed.. Also it gives no details at all as to how the transducers work.
So does the TS really think that is going to be an easy way to do it???
Will it be easier than two pressure taps and two pressure transducers?? Will it possibly cost less??????
 

richbrune

Joined Oct 28, 2005
126
Interesting but not informative as far as just how it is done, as to what power level is needed.. Also it gives no details at all as to how the transducers work.
So does the TS really think that is going to be an easy way to do it???
Will it be easier than two pressure taps and two pressure transducers?? Will it possibly cost less??????
I'd try an adxl335 accelerometer, an lm358 dual op amp with one amp gain of two, second stage as a comparator. I'd use a pot to adjust vref. The subtle shaking should easily provide the type of output that the 335 is commonly used for. Possibly a 555 tomer ic to latch (debounce) the output during flow. Old school, but all the parts are within reach of my bench. A picaxe 08m2 (<$5) could probably take the place of the amps and timer, as well as sending a serial message to some system, if needed. Some duct tape and soldering and you're done with a few bucks of parts.
 

MisterBill2

Joined Jan 23, 2018
18,176
Does the TS even have the ultrasonic transducers on hand? And the hardware to couple them to the pipes? It seems that the interface circuitry will be the ,east of the challenges.
 

MisterBill2

Joined Jan 23, 2018
18,176
Why is "Ultra-Sonic" detection required ???
https://www.mcmaster.com/2766K112/
.
.
.
View attachment 295729
My impression is that the motivation is only in that direction, possibly due to some claims of how wonderful it is to do it using an arduino package. That is typical"Nuts and Volts" policy. Only one concept considered, no matter how many better approaches are available. In this case there are many options..

The benefits of using a pressure transducer arrangement, especially using two gage type sensors, is totally ignored.
The mechanical flow switch option shown does require a larger penetration of the piping and is subject to mechanical damage and failure. The pressure sensing scheme can function with much smaller penetrations, depending on the installation resources available.
 

MisterBill2

Joined Jan 23, 2018
18,176
It is not like the ultrasonic scheme is so new, although it has undoubtedly become less expensive. But for applications where extreme accuracy and freedom from contamination are vital, ot may be the proper choice. But I am suspecting that the clamp-on versions will require some calibration, although possibly if only fluid velocity data is required. I had not been aware of the clamp-on variety, only the very accurate in-line versions. Certainly the "cool factor" makes it seem like the only choice. And reading the pressure across a pump by using two "gage" sensors will allow determination of both flow and pump effectiveness. I did learn a few tricks during those years of designing testing machinery.
 
Top