Efficiency of Inaudible Piezo Tweeter?

Adequate for what? If there's a receiver, can't you use that to determine if it's adequate?

Measuring sound level is a bitch. Microphone calibrated in the region of interest is a good thing. Then you get to start worrying about how it couples to the air, reflects off the walls, gets absorbed by the surroundings etc.

I have an ultrasonic leak detector that mixes ultrasonic down into the audio range. Has an LED bar graph. Not calibrated, but could be used as a relative indicator.

You're also probably interested in how well the drive frequency matches the resonant frequency of each and every transducer/enclosure. And the resonant frequency of the receiver. Did I mention Measuring sound level is a bitch.

An industrial ultrasonic cleaner I was just troubleshooting has a piezo feedback element bonded to the surface (along with a whack of transducers).

Interestingly, it has an (expensive) 4-quadrant multiplier chip in it.

Best regards, Spehro Pefhany

Point it at your dog. (I'm not kidding.)

Degree of distress = strength of output.

LV

We used to use a 1/4 inch B&K measurement microphone, hooked up to the appropriate sound level meter. Try renting appropriate test equipment for a day, or find an acoustics lab that would take on a small job. Note that most commercial sound level meters have deliberately crappy high frequency response, because they are meant to reflect the response of the human ear.

I don't know if this still works: My old boss, who had two high quality Starkey hearing aids, could not hear anything as we passed under the ultrasonic motion detector transducers outside out lab. The Starkeys had wide frequency response, as well as overload protection. The sound pressure level coming out of the transducer was both picked up by his microphones, and high enough to make the amplifiers cut out.

use another transducer to receive the output?

measure heating of the transducer, subtract from input power?

for what?

That's very dependent on geometry of the setup, pattern of radiation, distances etc.

Very difficult unless you go the whole hog and set up a standardised test rig using (very expensive) equipment from companies like Bruel & Kjaer. Measuring relative output isn't too hard, you just need an acoustic-electric converter that will respond to some degree at your operating frequency and in your medium (air? water? solid?) - such as a piezo microphone or a transducer. If you then have a setup that you already know is working correctly, then you can calibrate your rough measuring devices against that.

Thanks for the replies so far.

Is there no rudimentary way to estimate the efficiency/intensity by applying a signal, or sweep, to the tweeter and monitoring it on the CRO?

Ken Gillmore

The answer is critically dependent on your definitions of "rudimentary" and "estimate". There are a lot of very creative people here who might be able to help you if you'd bother to disclose WTF you're trying to accomplish.

A question like, "How can I determine whether a piezo element, maybe a tweeter, operating at some unknown frequency that may or may not be fixed is adequate for my undisclosed application" is unlikely to get you much help.

My rudimentary estimate of the answer is 42.

As per my previous post, in an absolute sense, not without specialised equipment. Sweep testing is a useful technique in picking the optimum frequency of operation, but an uneven response in your pickup device will give misleading results, obviously.

One possible approach would be to do an end-to-end test with two identical devices such as piezo speakers, and assume that the response is the same for each. That way, you have the frequency response for each individually, and if the curve is reasonably smooth then it will make a reasonable measurement device, possibly with some compensation applied.

However, IME not much at ultrasonic frequencies is free from peaks and dips.

I once worked with a physicist trained in acoustics. He worked in a lab where they built and tested underwater transducers. Early in his employment he tested and found he had a 200% efficient transducer. Being quite a bright fellow, this lead him to discover the labs calibration standard had a 10db error.

You mentioned measuring sound level is a bitch.

Mikek

PS. His naive manager complimented him on his very efficient transducer. He never did any tongue in cheek with that manager again!

First, I know squat about piezo's. (But I've got a stack of ultrasonics books next to me :^) What frequency? Can you monitor I and V into transducer as a function of frequency and figure something out where you can hear it? Can the same piezo be used as a microphone? You need a dectector (obvioulsy). Some random ideas: Something to absorb the sound and measure temperature rise. Make a laser interferometer and observe modulation of air density.

George H.

Just install a second piezo transmitting a fixed level sound at a frequency x + 1KHz. Measure the beat note with a normal sound pressure meter.

Get Smart's "Cone of silence"? ;-)

I don't think you can measure sound output without actually measuring sound output. But if measuring sound output seems impossible to you, I would try to get hold of the manufacturer's data sheet. This would provide a rudimentary way to estimate the efficiency/intensity of the piezo device, because you could pick the number from the data sheet.

Then the only mystery would be if the piezo device was damaged. Can you see if the piezo bender is attached to some sort of diaphragm? If so, does that joint look good? Then you would need to check if the piezo bender was still good, and still electrically connected to the device terminals.

What I did back in the day was use an HP audio oscillator, an HP audio voltmeter, and a GR decade resistance box, to plot the impedance vs frequency. If the piezo device looked like the capacitor it should, I would assume all systems were go.

If not, I would buy a new tested good piezo tweeter, with data sheet.

Ahh, would that work? With light, a photodiode is a square law detector and you get mixing in the photodiode. But a microphone responds to the amplitude so no mixing. (it's linear unless perhaps you really over drive it.) But perhaps a sound pressure meter is not a microphone?

George H.

you're going for intermodulation, and using the amplitude of that to derive the amplitude of the two sound sources.

-- Les Cargill

Intermodulation is still a mixing process and, if the detectors are linear, no intermodulation (mixing) will occur. George is correct.

If the sound level detectors are non-linear, then you can get intermodulation.

nd

In this case, I would advise that the OP use his head:

He would need a tweeter of known sound output, and an oscillator for each tweeter, so that the beat frequency would be in the audible range.

es?

and

ure

Hide quoted text -

Wow, This is getting to be fun! First I don't know how the ear works. Is it an amplitude detector like a microphone, or is it a power detector? But let me leave the ear for a minute and just assume one has a microphone and a 'scope. Then the *interference* beats that one would see on the 'scope are just changes in amplitude. (The sound gets loud and soft.) But if one was to do an FFT of the 'scope time record there would not be any sum or difference frequencies present. (assuming no non-linearity in the microphone.) At least that's my understanding... one could always do the experiment. Back to the ear, I've heard (NPI) about these ultra-sonic 'speakers' where two ultra sonic signals are combined in the ear and the difference frequency is heard. Perhaps that is using the non- linearity of the ear?

George H.