Echo ranging transducer selection /ultrasonic

[dark]

Hi ,

If we use an ultrasonic transducer to detect level of liquid inside a container (echo ranging method with a transceiver sensor ). Then can someone help me with selecting transducer frequency and bandwidth to attain a range resolution of +-1mm or lower. I need to measure a distance of about 800cm.

Rgds

 

[misterT]

You need a frequency at least 174 kHz and the sensor needs to be temperature compensated.. maybe even humidity and pressure compensated depending on your environment.

Something like this: https://www.bannerengineering.com/e...-Sensors/498/U-GAGE-T30UX-Ultrasonic-Sensors/

 

[dark]

Hi, I need to know whats responsible for resolution in level metering , is it freq or circuit. Thanks

 

[misterT]

Hi, I need to know whats responsible for resolution in level metering , is it freq or circuit. Thanks

Both. Also temperature, humidity and air pressure affect the accuracy (speed of sound). 1 mm resolution is not easy to achieve with ultrasound. You won't find many commercial products and the ones you find will be expensive. Most of them won't measure up to 80cm range.

"Measurements are affected by factors such as the wavelength of the sound, the Q of the transducer, the reflecting characteristics of the target, the operation of the target detection electronics in the sensor, and the uncertainty in the assumed value of the speed of sound."

https://www.sensorsmag.com/sensors/...-sensor-proximity-or-distance-measurement-838

 

[misterT]

Oops.. I read your original post wrong. I thought you need the range to be up to 800mm, but I see it is actually 800cm.

Forget that. There is no way an ultrasonic sensor will measure that distance. Especially with frequencies capable of measuring 1mm differences.

A 200 kHz soundwave attenuates ~10 dB/m:
Total beam reflection is equivalent to a virtual source at twice the distance. Therefore, the spreading loss for the sound reflected from a large flat surface is equal to 20*log(2*8) dB, and the absorption loss is equal to 2*8*10 dB.
So the reflected beam from a very large flat hard surface at 8m will be attenuated ~185 dB when it arrives back to the receiver. That means that the sound pressure of the signal is about 1/2000000000 (that is 0.00000005 %) of what left the transducer. That signal is long lost in the movement of air molecules.

 

[atferrari]

Look for radar

A very practical detail: if surface waves exist, your measurment will be seriously hampered to get that precision.

I do not know what experience you have on that but just a simple vibration will make nice concentric waves.

Much worse if the tank is in a truck...or in a vessel in motion (oh yes!).

Even with a metric tape in a (very) careful process, in shore tanks of big size, the mm is a matter of bonafide.

Google for radar in tank gauging.

 

[dark]

Both. Also temperature, humidity and air pressure affect the accuracy (speed of sound). 1 mm resolution is not easy to achieve with ultrasound. You won't find many commercial products and the ones you find will be expensive. Most of them won't measure up to 80cm range.

"Measurements are affected by factors such as the wavelength of the sound, the Q of the transducer, the reflecting characteristics of the target, the operation of the target detection electronics in the sensor, and the uncertainty in the assumed value of the speed of sound."

https://www.sensorsmag.com/sensors/...-sensor-proximity-or-distance-measurement-838

Sorry for the mistake , I mean 800mm . Isnt the resolution dependant on the sensor bandwidth?

 

[dark]

A very practical detail: if surface waves exist, your measurment will be seriously hampered to get that precision.

I do not know what experience you have on that but just a simple vibration will make nice concentric waves.

Much worse if the tank is in a truck...or in a vessel in motion (oh yes!).

Even with a metric tape in a (very) careful process, in shore tanks of big size, the mm is a matter of bonafide.

Google for radar in tank gauging.

Radar because of operating in RF region will require regulations and approvals process it is time consuming. Thanks.

 

[misterT]

Sorry for the mistake , I mean 800mm . Isnt the resolution dependant on the sensor bandwidth?

Ok, then. 800mm and 1 mm resolution is achievable, but not easily.

I don't know what you exactly mean by bandwidth. If you mean signal frequency, then a narrow bandwidth is better in "traditional" sensor. That is what "Q of the transducer" means. There are some research done on multi-frequency systems that can achieve high-accuracy, but I have not seen any commercial products that use that kind of technique.

https://en.wikipedia.org/wiki/Q_factor

 

[alec_t]

Is the transducer in air, aimed down at the liquid surface; or is it in the liquid and aimed up at the surface? (A transducer in liquid would normally be operated at a higher frequency, and would thus have greater resolution, than one in air).

 

[dark]

Hi , Here are the details for your quick review;

**broken link removed**

It is +/-4% , but manufacturer is not at all replying regarding Q factor I mailed them twice.

Rgds

 

[misterT]

It is +/-4% , but manufacturer is not at all replying regarding Q factor I mailed them twice.

If that is a transducer element and you are going to build the electronics yourself, the Q factor depends on how your electronics interact with the sensor.

Is that a transducer element, or a fully working commercial sensor?

EDIT: The "Best Oparating Frequency" is +/- 4%. That is not in any way related to sensing accuracy or resolution.

EDIT2: Sorry, I was wrong about the Q factor. Of course mechanical resonators also have a Q factor. But, attached electronics do affect the whole systems Q factor. Smaller Q factor means broader bandwidth and less ringing. So actually broad bandwidth might actually be better than narrow. But, small Q means usually less sensitivity at the resonant frequency.
Can you post a link to full datasheet of that transducer?

 

[dark]

If that is a transducer element and you are going to build the electronics yourself, the Q factor depends on how your electronics interact with the sensor.

Is that a transducer element, or a fully working commercial sensor?

EDIT: The "Best Oparating Frequency" is +/- 4%. That is not in any way related to sensing accuracy or resolution.

EDIT2: Sorry, I was wrong about the Q factor. Of course mechanical resonators also have a Q factor. But, attached electronics do affect the whole systems Q factor. Smaller Q factor means broader bandwidth and less ringing. So actually broad bandwidth might actually be better than narrow. But, small Q means usually less sensitivity at the resonant frequency.
Can you post a link to full datasheet of that transducer?

The specs I attach is for transduce and without any conditioning ,from the above specs I cant figure out Q of above Xducer . What I am confused is what Q factor is the best and which is poor? Rgds

 

[misterT]

What I am confused is what Q factor is the best and which is poor?

1) What you want is good sensitivity to the frequency you are using. High Q is an implication of sensitivity, but it does not actually tell anything about it.
2) You also want as less ringing as possible. Low Q gives less ringing. Broad bandwidth means faster rise and fall times. Usually low Q means less sensitivity at the resonant frequency.

So there is no simple answer. You have to figure out a balance that works best for you on your application.