Transducer Frequency Response 1

[Electricliff]

This relates to my 50KHz ultrasonics problem, but presumeably relates to all acoustics.

1. A transducer can convert electrical energy to acoustic energy.
2. Another transducer converts acoustic energy to electrical energy.

I can build a driver with a flat frequency response for the first transducer (measured with readily available instruments)
I can build an amplifier with a flat frequency response for the second transducer.(measured with the same instruments)

If I drive 1. with the driver amp and receive the acoustic energy with 2. amplified by my receiver amp, how do I determine from the frequency response of the overall signal path what each part of the path contributes to that response?

The obvious answer would seem to be get a transmitting transducer or a receiving transducer with a flat frequency response. But how do the manufacturers of such devices know that the response is flat and how do they measure it?

What is the standard to which all acoustic transducers can be referred?

At the moment I'm floundering. Can anyone explain or point me to a good text book on the subject.

Thanks

 

[GregLocock]

I don't /know/ but they probably use reciprocity at some point in the calibration process.

Cheers

Greg Locock

 

[MikeyP]

I'm not sure if this answers your question exactly...

There are 2 approaches to microphone calibration. 1) Substitution in a field. 2) Reciprocity.

1) Place the transducer in a constant steady state acoustic field (a pistonphone for example). Place a transducer of known calibration in the same field and compare the two results. If the same mic amp is used in both cases, then the effects of the rest of the measurement chain can be subtracted out. For best results the transducers must be of the same shape and size so that they modify the acoustic field in exactly the same way. Of course this leads to the obvious question of "How do you calibrate the reference transducer?". Well, a known potential field is relatively easy to produce accurately. we can place the mic in this field and use it to very precisely "push" the diaphragm. In that way we get an absolute calibration for our reference mic.

2) Condenser microphones are "reciprocal transducers". An acoustic stimulus pushes the diaphragm which causes a chance in capacitance which in turn produces a change in voltage across the device. However the reverse is equally true. That SAME voltage applied across the transducer causes the diaphragm to move and generate the SAME acoustic pressure in front of the mic as was measured originally.

Note that this transducer reciprocity is not exactly what we mean by "calibration by reciprocity". The reciprocity in this sense is "acoustic reciprocity", ie. if you swap a transmitter and receiver locations you get the same answer. So we know that our mic can be used as both a transmitter and receiver. By applying acoustic reciprocity (a form of the Maxwell-Betti reciprocity theorem) we can calibrate a pair of mics in a specially designed coupler. By using one as transmitter and one as receiver and the swapping them over.

(See

http://www.bksv.com/pdf/Bp1697.pdf

for example)

This type of calibration is also absolute as it is related to physical quantities which are easily measureable, ie. mass, length, time. (It is quite easy to make a capacitor whose value is known very precisely)

Now only very few laboratories will have this type of equipment so we use "tracability" to ensure that you and I, the grunts in the lab, can make accurate measurements. Before we take an acoustic measurement, we calibrate our humble mic with a field calibrator. This is calibration by substitution in a sound field, the "substitution" having been done in a NAMAS calibration laboratory with a fancy reference microphone the last time we sent our calibrator to be calibrated (about 5 years ago if most labs I have worked in are anything to go by!). So where does the NAMAS calibration lab get its reference mic from? It is calibrated by reciprocity (or by electrostatic calibration) at one of the standard central measurement labs (eg NPL in the UK or NIST in the US).

Hope this was useful

M

 

[MrTransistor]

You might check out "Ferroelectric Transducers and Sensors" by J.M. Herbert. It's an older book but I have found it useful for my work with piezoelectric structures.