Variability in Switching Transducers

A study by Valente et al (1994) compared two transducers (TDH 39P earphones and ER-3A insert phones) by looking at their real-ear SPL performance across a frequency range of 500-4000 Hz.  The authors found that the real-ear SPL differed by 2.4 dB at 1500 Hz to as much as 10.4 dB at 500 Hz between the two transducers.  The SPL generated using the TDH 39P earphone was significantly great than the real ear SPL generated by the ER-3A insert phone.  

Other studies have also found output differences between the two styles of transducers.  A study by Frank and Vavrek (1992) found that when the output of the TDH-49 and the ER-3A were measured in a coupler, the following difference were observed.

The TDH 49 was measured in a NBS-9A coupler and the ER-3A in a HA-2 coupler.  

The significant difference in measured SPL between the two transducers points out the effect that transducer type can have on real ear measured SPL.  From this problem, one can deduce the potential problem of switching transducers from different audiometers for which the transducer was not originally calibrated.  Despite this variability across transducers, switching earphones is a common practice.  According to Lilly and Purdy (1993), doing this can create significant differences between the input impedance of the ER-3A insert phone and the specified load impedance of the output to the audiometer.

Audiologists should be aware that the ER-3A insert phones come in different impedance values (10, 50, 300 ohms).  In the Valente et al (1994) study, an insert phone of 10 ohms was calibrated for an audiometer rated for 10 ohm transducers.  A 50 ohm TDH style earphone can also be used on this audiometer, but a 6 dB average increase in output measurements can be expected compared to the 10 ohm device.  Audiologists need to apply a correction factor to account for this increase when switching transducer among different audiometers.  

The above example outlined how a transducer of greater impedance can be used for an audiometer rated with lower impedance.  The reverse is not true.  That is, it is not recommended that transducers of 10 ohms be used for audiometers rated for 50 ohms.  Doing so can lead to excessive distortion and damage to the audiometer.   

The following table from Valente et al. (1994) further illustrates the problem of switching transducers among different audiometers.  

The ER-3A insert phones were calibrated for audiometer A (see below).  ANSI (1989) standards are listed on the top of table 2.  As you can see, the output values from audiometer A are all within 3 dB of the ANSI (1989) standard.  Great amount of variability existed at audiometer B-C relative to A and to the standard.  

Taken from Valente et al 1994 

The critical point to note here is that when the same ER-3A insert phone is used on different audiometers, it is possible to obtain different SPLs measured in the ear canal when the attenuator is fixed at a predetermined level.

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