Bone Conduction Transducer: Output Force Dependency on Load Condition
The aim of this investigation is to determine the output force variability of the new bone conduction transducer, called Balanced Electromagnetic Separation Transducer (BEST), and of the conventional transducer from Radioear B71. The loads used in this study were the mechanical impedance of the skull of thirty test persons and the mechanical impedance of the artificial mastoid (AM) 4930 from Brüel&Kjær. The hypothesis is that the BEST has less variability since it has inherent damping of the spring elements, while the damping with the B71 is mainly due to to damping in the skin and the soft tissue. Hence, the damping of the B71 transducer is basically dependent on the patients individual properties.The output force of the transducers were estimated in three different ways:
1. Transducer modelled as a four pole equivalent.
2. Transducer modelled as an impedance analogy lumped parameter model.
3. Direct measurement using the AM (not completed in this study).
The models of the transducers were determined from frequency
As expected the variation of the output force was less for BEST than for B71. For BEST, the standard deviation of the force is approximately -3 to +2 dB at 316 Hz, while for B71 the deviation at 200 to 500 Hz is up to +4 to -10 dB. Both transducers show greater load dependency at higher frequencies. BEST has a standard deviation ranging from +0.8 / - 1 dB to +
3 / - 5 dB for the frequencies between 0.8 and 3 kHz. The B71 have a standard deviation in output force level of -6 to +4 dB from 1.2 to 10 kHz.
The results also show that the output force of the transducers differ less than the mechanical impedances of the test objects. The impedance of the artificial mastoid are very different about 7 dB relative to 1 Ns/m between the mean and AM. Comparing the output forces from the BEST loaded with the two impedances (mean human skull and AM), lower force magnitudes is found for the test objects up to 10 dB for frequencies from 1.3 to 10 kHz. The B71 shows also lower force magnitude with the impedance of the test persons than with the impedance of AM at even lower frequencies than the BEST, from 6 to 10 dB for 0.4 to 10 kHz.
Further, in the low frequency range 100 to 400 Hz, the BEST shows approximately 5 dB higher output force levels than the B71 with the same amount of stimulation. This is true for both types of mechanical impedances, which is an important advantage for the BEST due to the non-linear distortion at low frequencies for B71. When the output force is calculated with the mean value of the impedances of the subjects, the B71 gives higher amplitudes than the BEST, about 10 dB, in the frequency region from 1.3 to 10 kHz. When the force is determined by the impedance of the AM, higher force level of the B71 starts at 400 Hz becoming equal to that of BEST at 1 kHz to increase again the remaining frequency range.
Since the main resonance of the variables Z21 and Z12 did not occur at the same frequency it gives some artifacts that should be disregarded. It is of interest to calculate the value of lumped parameters of the transducers and make new calculations of the output force to be sure that the artifacts vanish.