Hi-Res Ear Simulator for

Measurements at Very High

Frequencies
By Kristian Gøtsche-Rasmussen, Morten Wille and Per Rasmussen
August 2019
 Hi-Res Ear Simulator for Measurements
at Very High Frequencies
By Kristian Gøtsche-Rasmussen, Morten Wille and Per Rasmussen

The new GRAS Hi-Res Ear Simulator is designed to extend the useful frequency range
up to 50 kHz. The extended frequency range is obtained by using a 1/4” pressure
microphone instead of a 1/2” microphone. The new ear simulator is compatible with
the specifications in IEC 60318-4.

It uses a resonance damping system similar to that used in the RA0401 High
Resolution Ear Simulator, but extends the useful frequency range to 50 kHz and
therefore meets the 40 kHz requirement of the Hi-Res standard for high performance
headphones (2014) as described in for example Tsunoda, Hara & Nageno, 2015,
Tsunoda, Hara & Nageno, 2016, and Tsunoda, Hara & Nageno, 2017.

The damped resonance and small microphone footprint result in improved repeat-
ability and extended frequency range and makes it possible to evaluate and compare
headphones in a wide frequency range with frequency response, THD, noise floor,
etc.

The Hi-Res Ear Simulator is part of a number of pre-configured, ready-to-use test

solutions from GRAS such as the 43AG, the 45CA, and the 45BB and 45BC KEMAR.
Visit gras.dk for further information.

The GRAS Hi-Res Ear Simulator is available in two versions:

RA0403 is externally polarized, RA0404 is prepolarized.

2 | Whitepaper Hi- Res Ear Simulator

Introduction
The IEC 60318-4 Ear Simulator is widely applied in the verification of fre-
quency response of acoustic products, e.g. in the audio (consumer) indus-
try, but also for measurements on hearing aids. It is recognized for its ability
to simulate the acoustic load as presented by the ear. Placing a headphone
on a standardized ear simulator, e.g. a GRAS RA0045 Ear Simulator accord-
ing to IEC60318-4 ensures that the headphone is facing the same acoustic
input impedance as a real human ear would provide.
An ear simulator based on the IEC 60318-4 standard, e.g. the GRAS RA0045
Ear Simulator is therefore necessary to obtain a realistic acoustic transmis-
sion line from the output of the transducer (e.g. a headphone or a hearing
aid) placed in a known measurement plane (e.g. the ear entrance point) to
the ear-drum point in the ear simulator.
However, the traditional IEC60318-4 (GRAS RA0045) ear simulator has an
undamped length resonance at 13.5 kHz. This is related to the half-wave-
length distance from the ear simulator input reference plane to the micro-
phone position. As the resonance is related to the main volume length, the
resonance frequency will shift if the length is altered.
In most practical situations the Device Under Test (DUT) will not be posi-
tioned precisely at the reference plane of the ear simulator. Figure 1 shows
a typical setup, where the main volume of the GRAS RA0045 Ear Simulator
is extended with an ear canal and exterior pinna (e.g. the GRAS KB5000
Anthropometric Pinna). This will change the ear canal length and thereby
the resonance frequency so that both the half-wave and full-wave reso-
nances can be seen below 20 kHz.

Figure 1
Reference Plane Half wave resonance
40 GRAS RA0045 Ear Simula-
35
tor volume extended with
Transfer impedance x freq. (dB re. 500 Hz)

anthropometric pinna
30
(GRAS KB5000)
25

20

15

10

-5
100 1000 10000
Frequency [Hz]

Ear Entrance Plane Full wave resonance

40

35 Half wave resonance

Transfer impedance x freq. (dB re. 500 Hz)

30

25

20

15

10

-5
100 1000 10000
Frequency [Hz]

GRAS | 3
The GRAS RA0401 High-Frequency Ear Simulator is designed to minimize
the effect of the undamped resonance at 13.5 kHz to accurately acquire
and verify frequency responses. This is obtained by using a built-in reso-
nance damping system in the ear simulator. This attenuates the resonance
peak by approximately 14 dB and reduces the effects of the shifts caused by
changes in DUT positioning in the ear canal. For more information, see the
white paper about GRAS RA0401 (Wille, 2017). It can be downloaded from
gras.dk.

The problem: Measurements of higher frequencies

New audio formats with wider frequency range have been introduced to
provide the users with higher sound quality and better listening experience.
This has led to the demand for precise and relevant test methods to effec-
tively validate and optimize designs and concepts.
Modern hi-res audio formats offer frequency ranges above 50 kHz and as
the internationally standardized ear simulators were primarily designed for
use below 10 kHz these are not adequate.

The solution: The new GRAS RA0403/04 Hi-Res Ear Simulator

The Hi-Res Ear Simulator RA0403/04 is based on the standardized IEC
60318-4 ear simulator and the RA0401 High-Frequency Ear Simulator, but
extends the frequency range of interest to 50 kHz. The extended frequency
range has been obtained by using a high frequency ¼” microphone in the
ear simulator together with a resonance damping system and at the same
time carefully adjusting the ear simulator impedances so that they are the
same as in the IEC 60318-4 ear simulator for frequencies below 10 kHz.
The RA0403/04 Hi-Res Ear Simulator simulates the acoustic transmission
line (resonances in the ear) which can be further extended by including the
pinnae and thus the diffractions in the exterior sound field.

4 | Whitepaper Hi- Res Ear Simulator

Specifications and Data
In Table 1, the specifications for the RA0403 are compared to those of
RA0045 and RA0401. The sensitivity is lower for the RA0403 Ear Simulator
which results in a higher noise floor. This means that the RA0403 Ear Simu-
lator is not suited for low-level measurements and a GRAS 43BB Low-noise
Ear Simulator System would be a better option. The resonance frequency for
the three ear simulators is the same.

Table 1
RA0045 RA0401 RA0403 Specifications of RA0403
compared to RA0045 and
Microphone 40AG (1/2”) 40AG (1/2”) 40BP (1/4”) RA0401.
Specifications for RA0404
Frequency range (HZ) 100-10k 100-20k 100-50k are the same as for RA0403,
except that a 40BD microp-
hone is used and the upper
Sensitivity (mV/Pa) 12.5 12.5 1.6
dynamic limit is 166 dB.

Volume (mm3) 1260 1260 1260

Dynamic range (dB) 25-164 25-164 44-169

Ref. plane resonance freq. (Hz) 13500 13500 13500

Figure 2 shows the transfer impedance curves for the new RA0403 Hi-Res
Ear Simulator compared to the traditional RA0045 and the RA0401. It can be
seen that the three ear simulators are identical in the frequency range up to
10 kHz, as specified in IEC 61038-4.

Above 10 kHz, the effects of the damping system used in RA0401 and
RA0403 can clearly be seen. The RA0401 extends the frequency range to
20 kHz, while the frequency range of the RA0403 extends up to 100 kHz,
with efficient dampening of resonances up to 50 kHz.

Figure 2
55 The extended frequency ran-
RA0045
50 ge obtained by RA0403/04
Transfer impedance x freq. [dB re 500 Hz]

RA0401
45
compared to frequency ran-
RA0403
ges of RA0045 and RA0401
40 IEC Tolerance

35

30

25

20

15

10

-5
100 1000 10000 100000
Frequency [Hz]

GRAS | 5
Tolerances of the RA0403 Ear Simulator
In Figure 3, the transfer impedance and tolerances of the new RA0403 Ear
Simulator are shown. Tolerances for the RA0403 Ear Simulator are the same
as for RA0401 below 20 kHz, i.e. following IEC 60318-4 tolerances for fre-
quencies 0.1-10 kHz and the GRAS specified tolerances (± 2.2 dB) from 10-
20 kHz. From 20-50 kHz the tolerances have been specified to ± 3.2 dB.

Figure 3
Transfer impedance
55
Typical response RA0403 of RA0403 with IEC
50 tolerances and GRAS
GRAS RA0403 Tolerance
Transfer impedance x freq. [dB re 500 Hz]

45 IEC Tolerance
tolerances. The GRAS
tolerances are specified
40 up to 50 kHz.
35

30

25

20

15

10

-5
100 1000 10000 100000
Frequency [Hz]

References
Standard IEC 60318-4: Electroacoustics -Simulators of human head and
ear - part 4: Occluded-ear simulator for the measurement of earphones
coupled to the ear by means of ear inserts
Hi-Res Audio Standard. Japan Electronics and Information Technology
Industries Association (JEITA). Announced June 12th 2014. Link: www.
jas-audio.or.jp/english/hi-res-logo-en
Tsunoda, N., Hara, T., & Nageno, K. (2015). A Headphone Measurement Sys-
tem for Audible Frequency and beyond 20 kHz. Audio Engineering Society.
Tsunoda, N., Hara, T., & Nageno, K. (2016). A Headphone Measurement Sys-
tem Covers both Audible Frequency and beyond 20 kHz (part 2). Audio En-
gineering Society.
Tsunoda, N., Hara, T., & Nageno, K. (October 2017). A Headphone Measure-
ment System Covers both Audible Frequency and beyond 20 kHz (Part 3).
Audio Engineering Society.
Wille, M. (October 2017). High Resolution Ear Simulator. White Paper (GRAS).
Can be downloaded at gras.dk

6 | Whitepaper Hi- Res Ear Simulator

GRAS Sound & Vibration A/S
Skovlytoften 33, 2840 Holte, DK

[email protected]
+45 4566 4046
gras.dk