How Do I Measure The Amplitude of Ultrasonic Vibrations?

Ha! A bit of inspiration from one of the previous questions.

A shadowgraph...projecting an image...if the light is controlled stroboscopically you may be able to see the oscillation and measure it on the projected image...do a bit of arithmetic.

Bish bash bosh...done!

Del

Probe tip displacement can be very accurately measured using a specialized laser interferometer. The laser is strobed at a rate (f_S) held very near the probe frequency (f_P, within 1 Hz or less of f_S), and the maximum amplitude of the shift in the interference pattern (IP) is noted. The IP displacement will change at the beat frequency (f_S ± f_P), and the peak IP displacement amplitude will be a function of the probe amplitude and laser wavelength. With these in hand, the peak-to-peak probe amplitude can be determined. Alternatively, the strobe rate is held at the probe frequency, and phase between the two is varied over 360 degrees. One caveat is that the measured surface must be relatively smooth in order to maintain a clearly-defined interference pattern. No bumps, roughness or small, complex curves.

-e Hi!

Thank you for the in-depth and very interesting explanation (we learn something new every day). I guess that I am interested in the P waves.

More than being able to measure the correct absolute amplitude of these waves I am interested in measuring the relative distribution of the amplitude along the contact surface of a rectangular Sonotrode with the aim of achieving uniform distribution of the amplitude and thus getting uniform weld strength along the full length of the weld. At the moment I notice that the weld is stronger in the middle and weaker at both ends.

What I am actually looking for is the simplest measuring set up and equipment that will enable me to measure the relative distribution of the amplitude along a rectangular Sonotrode.

Regards,

Kobi.

Shalom Kobi,

Let me guess, you are trying to classify different plastics VS different ultrasonic densities in a welding operations? or am I totally off?

Laser interferometer will be you best bet, However if only relative measurements are needed, than any ultrasonic detector with measuring capabilities will suffice.

Wangito.

Why is this thread mentioned here?

What is the link between this science blogdig and CR4?

Can anybody shine some light?

It depends on the configuration of your sonotrode. If it is a rod that is being vibrated extesionally, the measuring method may be different than for a rod vibrating transversly. The laser interferometer is a good method, but sometimes difficult to implement in a production environment. Directional microphones can also be used. A miniature accelerometer attached to the sonotrode has been used and provides good data. Also, be aware that even if the vibration is uniform across the sonotrode, the effects of the vibration may be different at different parts of the joint, hence variations in weld strength for uniform input.

Would you mind telling me which accelerometer did you use? I am also looking to measure vibrations on a sonotrode with a relatively cheap method Thanks!

Look into laser vibrometry for mems. Also, you may save some time by attacking the problem from a structural acoustics standpoint that includes simpler bending and longitudinal waves instead of the zoepplitz equations as they are more for geophysics, but I may be wrong here.

Hi,

if the movement is perpendicular to your surface then it is cheapest and easyest to try a capacitive probe. (Metallic plate with small distance to the to be measured surface)

If you don't have one then build one with a 555timer or a CMOS SN74AC14 schmitt-trigger inverter where you take the distance to your probe to be a variable capacitor. Set this as a 500KHz RC oscillator and detect frequency as the output signal that will change with capacitance equivalent to distance.

For ultrasonic frequencies up to 40 KHz this will work. If higher frequencies are to be measured then it is easyer to put a high DC voltage on one of the surfaces coupled to it via a high resistance so that any change in distance will give a change in voltage.

It is good to metalise the surface but it works without as the material -if nonconducting- has some dielectric constant different from 1.

RHABE

i would just like to ask, what if the movement is not perpendicular and moving at 80kHz.example, at the bonding process the bonding tool presses the ball of gold in a horizontal direction at a very high frequency(80kHz). is your suggested method of using a high dc voltage and coupled to it via a high resistance possible, since a sidewards direction would also increase the distance, assuming that instead of a surface we would bias a wire placed under the vibrating method?

thanks,

paul

Having done this before, I can advise that the easiest and lowest cost method is to use a miniature accelerometer. This device consists of a piece of polarized PZT ceramic with fine wire leads attached to the two active surfaces. This little ceramic chip (perhaps 1 mm square) is bonded to the sonotrode surface. If the thickness of the ceramic is on the order of 0.5mm, its resonant frequency will be well above your 20 khz driving frequency and will provide faithful reproduction of mechanical vibration parameters. The attachment point of the accelerometer to the sonotrode should be close to a vibration antinode (e.g. either end of the sonotrode) but because of the poisson's ratio, you will obtain usable output anywhere along the length of the vibrating sonotrode. If your sonotrode is tapered or stepped, the reason is that it is also used as an impedance converter to increase the output amplitude of the drive transducer. If you have trouble locating a suitable piece of piezoelectric crystal, one can be fashioned from a Mallory sonalert. Its little noise maker crystal can be sliced to make hundreds of tiny accelerometers.

Hi,

you have to be pretty good in surface preparation and very small bonding gap and nonfilled epoxi to have the accelerometer coupled with high enough stiffness to give reliable measurements.

RHABE

Ordinary surface prepeparation, like solvent wipe, is sufficient for aluminum or titanium sonotrodes. Accelerometers are often attached with cyanoacrylate adhesives that provide good adhesion in minutes. Epoxy works well also, but the cure time is longer. You will find that because a wavelength in metals at this frequency is in the order of inches, and the accelerometer is very low mass, the joint characteristics are not critical.

Hi Kobi,

Have you seen if different sonotrode tip sizes and geometries (square, round, etc.) produce similar or different results?

Maybe ask the question...

Is that really what you want to measure...?

... or do you want to measure energy transfer? In which case maybe you could do some sort of calorific experiment in water?

Del

I'm a little unclear: and do not have the expertise already demonstrated by others:

Are you moving your fabrics to be welded along the welder surface, or is this a "spot-welding" process. Your comment about the weld being stronger in the middle than at the ends is the source of my confusion.

What are the dimensions of your sonotrode tip?

And being a consumate "dick around" it would seem to me that if the surface was polished to specularity, or if a mirror could be affixed, that a simple laser pen could be arranged in such a way as to have it's beam reflected off at a shallow angle,

and the spot on yonder far wall:

hmm. OK Have to think about this more.

however, you COULD by this method see if the vibration was constant across the surface.

(is there a "Stupid" response below the "off topic"??)

One phenom you may discover as you study ultrasonic welding, is that even if the displacement of the sonotrode can be made absolutely uniform, the respnse of the material being welded is not uniform across the area of the weld. This is often not an easy concept to accept, but if you look at the stress distribution across two normally clamped surfaces subjected to shear stress, you will find the stress to be non-uniform. As a result, non homogenious properties of the welds may have more to do with the material's response to the forces it is subjected to, than to the uniformity of the imposed vibrations.

Simple method: Put some white paint on the surface, mark some small spots or scratch fine marks. Then simply use a microscope and graticule or eyepiece with a built-in scale! Not expensive, though slightly fiddly. Then once you have some provisional results you can think of what you actually need to measure next and get the right kit.

If you want to map the vibration amplitude, you can coat the end of the sonotrode with lycopodium powder and operate the sonotrode briefly. The powder will be ejected from the highest amplitude portions and you will be left with a map of amplitude distribution.