Beam width measuring

[epilot]

Hello,

I have some ultrasonic sensors (T & F 40kHz),
Unfortunately there is not any info, mark or specification on them, the seller did not know any thing about them, but it seems they are made in China.
Now I have to measure the beam width of them (specially the transmitters).
Can anyone advice me how to do so please?

Thanks for any input.

 

[dknguyen]

Because they are 40kHz, assume 30 degrees radial bandwidth.

 

[epilot]

Because they are 40kHz, assume 30 degrees radial bandwidth.

Why 30 degrees? I have seen 50,60 and 70 degrees 40kHz ultrasonic transducers too.
did you make any math for it?

 

[audioguru]

It isn't difficult for you to measure the beamwidth.

 

[JimB]

To measure the beamwidth, you will need an oscillator to drive the transmitter and a meter of some kind to measure the output of the receiver.

Place the transmitter and receiver so that they are facing each other, make the distance between them about 1 metre (my best guess as a starting point).
Turn on the oscillator and measure the output from the receiver.
Physically turn the transmitter by 10degrees and measure the receiver output.
Turn another 10 degrees and measure again.
Turn...... you get the idea.

Draw a graph of receiver output against rotation of the transmitter and you will be able to measure the beamwidth.

Beware of reflections from other objects in the room, you could get some strange results.

JimB

 

[audioguru]

If the seller didn't know anything about them and thinks they are made in China then they probably don't work. They are rejects thrown away by a real manufacturer. E-Bay is full of stuff like that. Some surplus joints too.

 

[dknguyen]

Are you sure that they aren't just measuring the beamwidth from different points/distances away from the arpature? Because I'm pretty sure the beam width is something mostly dictated by frequency, and the only things that really affect it are the width of the transducer which causes diffraction which makes the signal messier and wider. The beamwidth is also not a perfect triangle and is messy with lobes and stuff so depending where you measure it you may get different readings.

Have you ever seen a polar coordinate graph of beamwidth measurement?

 

[JimB]

I am not sure if you are replying to my post, but I guess so.
Anyway here goes.

Are you sure that they aren't just measuring the beamwidth from different points/distances away from the arpature?

Given that beamwidth is usually quoted in degrees (in my experience) it seems reasonable to do the measurements by rotating the transducer.
There was a thread here some months ago where hydrophone resopnses were produced by moving a metal object in a cartesian coordinate manner wrt the transducer.
Whatever works for you I guess.

Because I'm pretty sure the beam width is something mostly dictated by frequency, and the only things that really affect it are the width of the transducer which causes diffraction which makes the signal messier and wider.

My experience of ultrasonic transducers is minimal, but "gut feel" tells me that the physical construction of the transducer will affect the beam pattern.

The beamwidth is also not a perfect triangle and is messy with lobes and stuff so depending where you measure it you may get different readings.

I am not sure what you are getting at here, nobody has said that the beam pattern would be a perfect triangle. It is usual for beamwidth to be specified at the "half power points" (-3dB) wrt the peak response of the beam.
If you want to specify the beamwidth at some other response value, it needs to be stated.

Have you ever seen a polar coordinate graph of beamwidth measurement?

Yes, frequently. It is normal to present the response of a radio antenna as a polar plot. Always for azimuth and sometimes for elevation.

See the attachment.
I did a plot of an 40khz ultrasonic transducer under less than ideal conditions (transmitter and receiver were attached to a hard bench using Blu-Tac).
Although the measurements were done by rotating the transducer, I have graphed the data in a cartesian manner, mainly because I was not able to batter the Excel Chart Wizard into submission! I did not want to produce a polat plot using may data. (I have never used it to produce a polar plot before).
My results show a 3dB beamwidth of about 40 degrees.

JimB

 

[philba]

Nice post, Jim.

I've got a bunch of no-name us-transducers and have thought about measuring them. I guess I'm more interested in the measurement than the actual transducers. Did you do anything to try to reduce reflections? I thought about putting baffles around the receiver to discourage non-line-of-sight paths.

by the way, I think you can use the radar chart format to produce a polar plot. just pad out the full circle with what ever min value you want. It's not perfect but it's useful.

 

[JimB]

Did you do anything to try to reduce reflections?

No, this was just a "quick and dirty" try-out to support this thread.
If I were to do this seriously, I would probably mount the transducers on stalks above the supporting surface. I would also cover the mounting surface with a soft foam mat.

by the way, I think you can use the radar chart format to produce a polar plot. just pad out the full circle with what ever min value you want. It's not perfect but it's useful.

I started to use the radar chart, but the results were anything but what I was expecting. I have not used that plot format before and I think the problem could be in the way I presented the data.

JimB