Mounting ultrasonic sensor, issues with resonance

[Jackster]

Hi all

We have a pair of 40kHz ultrasonic sensors working in TX and RX.
Currently, they are soldered directly onto the PCB which is screwed into our aluminium (currently testing with 3D printed plastic) case.

What we are finding is that depending on what is attached to this case, what horns we have and if we have extension tubes on those horns etc we change the initial echo feedback.
Even holding onto the PCB changes this initial echo feedback into our sensor.
We are looking to get the best performance out of these sensors as possible so I need to help to understand what is going on here and how I can negate the issue.

Here are some examples. The echo is the blue line.

Bare PCB

With our narrow horns

Horns with small extension tubes on


This is the style of transducer we use.

It is not physically touching the case but is hard mounted to the PCB which in turn is mounted to the case.


From looking at a few off the shelf sensors. The ones you put on the car bumper are mounted in rubber.
But most others are just hard mounted.

This would not be such a problem if we were not after being able to range down to 10cm which going off the graph above, is seeing resonance at 10-30cm which is not good.
We use a threshold to omit this issue but ideally, I want to reduce this issue as much as possible.

We did try early on putting the two sensors on some long wires but even then we saw some feedback.

Am I chasing dragons here trying to omit something that can't be fixed?
Do I just need to extend my threshold a little more or is there something that can be done here?

Thanks!

[KaneTW]

This is unavoidable. Ultrasonic transducers react very strongly to the rest of the mechanical system. Reduce the Q of the system to make the effects less apparent.