Calibration Procedures

-from Beynon et al (1993)

Sound field calibration has two components: physical calibration which establishes the relationship between the sound pressure level (SPL) at a particular point in the sound field and the corresponding dial reading of the audiometer; and biological calibration which establishes the SPL corresponding to normal thresholds.  

Physical calibration, also known as the substitution method,  involves measuring the SPL at the test point in the sound field, and noting the difference between the audiometer dial reading and the measured value. This is done for each frequency and each stimulus type to give a set of correction figures for the particular test set-up used.  These correction figures can then be incorporated in calibrating the audiometer.  

In a substitution method, a microphone positioned where the subjects head is supposed to be located during testing.  The microphone measures the SPL at this location and theoretically captures the signal intensity corresponding to what a subject should be hearing at the entrance to the ear.  The microphone hence substitutes for a human subject. 

Biological calibration is the process of finding the SPL at the test point in the sound field that corresponds to audiometric threshold when a subject is present.  This is also known as the HATS or head and torso method of calibration.  This is the sound field equivalent of finding the levels in a coupler that correspond to threshold for earphone testing.  As for earphone calibration these levels are called reference equivalent threshold sound pressure levels (RETSPL); that is, they are the SPL measured at the reference point that correspond to (or are equivalent to) threshold.  Once the RETSPL values are known the sound field can be calibrated.  The audiometer is set to produce the RETSPL at the test point when the dial reading is zero, and any measurement of a subject's threshold above this level gives the threshold in dB HL, as for earphone testing.  Setting the audiometer in this way acts as a physical calibration and implements the results of the biological calibration.  The difficulty in sound field calibration is that of establishing what RETSPL figures should be used.  

According to Walker et al. (1984) and Beynon et al. (1995), when using the substitution method, the SPL produced by the loudspeaker at a position of 0.15 m on the left/right and up/down axis should be within +/- 2dB from the SPL at the test point.  These authors also state that the SPL at 0.1 m in front or behind of the test point should not deviate by more than +/- 1 dB from the theoretical value given by the inverse square law.  A 0.1 m criteria may be difficult to achieve with young children.  Recognizing this, Walker et al (1994) recommended a distance of 0.3 m to account for the possibility of movement by children.   

One problem with the substitution method of calibrating a sound field is the effects of the human head and torso which is not accounted for by the microphone.  The variation measured by the substitution method assumes that a microphone accurately reflects the variation heard by a human subject.  

A study by Beynon et al (1995) set out to measure the SPL variation in the substitution and head and torso calibration methods.  What they found was a great variation in SPL with the subject in the room.  95% of the variation was within +/- 3.4 dB for the microphone only situation and +/- 4.6 dB with the subject present situation.  Since both situations are within the +/- 5 dB step clinical criteria used in practice, it appears that both methods of calibrating the sound field (substitution and head and torso method) are clinically acceptable.  

Note:  This site is best viewed using IE 4.X or higher under 800x600 resolution