Approved 2005-05

TR 603
REPRODUCIBILITY
OF MEASUREMENTS WITH ISO 140
AND CALCULATIONS WITH EN 12354

Christian Simmons

Nordic Innovation Centre ISSN 0283-7234

Stensberggata 25, NO-0170 OSLO
Telephone +47 47 61 44 00
Fax +47 22 56 55 65
[email protected]
www.nordicinnovation.net
 NT TECHN REPORT 603
Approved 2005-05
Authors: Nordic Innovation Centre project number:
Christian Simmons 04030 (1670-04)
Institution:
simmons akustik & utveckling AB
Title:
Reproducibility of measurements with ISO 140 and calculations
with EN 12354
Abstract:
Uncertainties of measurement and prediction of the sound insulation in a building is
an important issue to the building industry, because they put increased expenses on all
actors of the building process who have to keep safety margins to prescribed
requirements. This report deals with estimates of both types of uncertainty and some
recommendations on suitable safety margins are given.

Requirements on sound insulation are described in the Swedish sound classification

standards SS 25267 and SS 25268. New standards for the measurement and the
prediction (by calculation) of sound insulation have been developed over the last
years, and the frequency range of interest has been expanded to include low
frequencies (50-100 Hz). This project was initiated to examine how the new methods
apply to typical building cases with respect to the new requirements on sound
insulation (within the expanded frequency range).

An inter-laboratory comparison (round robin) was made in Mölndal, Sweden with the
support from Nordisk InnovationsCenter and the eight laboratories participating. The
results of this comparison may be considered as examples of uncertainty that may
occur in real consultancy work, but the numbers of experienced operators who
delivered results, and the complexity of the calculation examples, prevent an
interpretation of the results to reliable estimates of the uncertainty of the calculation
method.

Some practical aspects on measurements and calculations of sound insulation are

discussed, which may help improve the methods applied. The advice given may help
the building industry handle uncertainties in prediction and verification more
efficiently. The measurement uncertainty can be reduced to some extent by a careful
measurement work and an extended averaging procedure.

Technical Group:
Building (Acoustics)
ISSN: Language: Pages:
0283-7234 English 35p.; 1 enclosure
Key Words:
Sound insulation, measurement, prediction, uncertainty
Distributed by: Report Internet address:
Nordic Innovation Centre www.nordicinnovation.net
Stensberggata 25
NO-0170 Oslo
Norway
Contents
SUMMARY AND CONCLUSIONS .......................................................................................... ii
INTRODUCTION .................................................................................................................... 1
PART A. MEASUREMENTS OF AIRBORNE AND IMPACT SOUND INSULATION ............. 2
PARTICIPATING LABORATORIES ............................................................................................. 2
MEASUREMENT VENUE AND MEASUREMENT CASES ......................................................... 2
MEASUREMENT PROCEDURE ..................................................................................................5
MEASUREMENT RESULTS ........................................................................................................ 7
EXAMINATION OF SOUND FIELDS – CASE A ........................................................................10
IMPACT SOUND..........................................................................................................................15
PART B. CALCULATIONS ACCORDING TO EN 12354 (ISO 15712) ................................. 17
B.1 DETERMINATION OF PARTITION AREA AND RECEIVING ROOM VOLUME .................17
B.1 ANALYSIS .............................................................................................................................23
B.2 SOUND INSULATION CALCULATED ACCORDING TO EN 12354 ....................................24
B.2 ANALYSIS ............................................................................................................................26
PART C. EXPERIENCES OF AGREEMENT BETWEEN MEASUREMENTS AND
CALCULATIONS IN SITU ................................................................................................................ 27
INTRODUCTION..........................................................................................................................27
RECOMMENDATIONS OF SAFETY MARGINS.........................................................................28
HOLLOW CORE CONCRETE SLABS – LOSS FACTORS ....................................................... 29
RECOMMENDED PUBLICATION AND PRESENTATION OF THE RESULTS ................... 34
ACKNOWLEDGEMENTS ..................................................................................................... 34
REFERENCES ...................................................................................................................... 35
ENCLOSURE

i
Summary and conclusions
Uncertainties of measurement and prediction of the sound insulation in a building is an
important issue to the building industry, because they put increased expenses on all actors
of the building process who have to keep safety margins to prescribed requirements. This
report deals with estimates of both types of uncertainty and some recommendations on
suitable safety margins are given.

Requirements on sound insulation are described in the Swedish sound classification

standards SS 25267 and SS 25268. New standards for the measurement and the prediction
(by calculation) of sound insulation have been developed over the last years, and the
frequency range of interest has been expanded to include low frequencies (50-100 Hz). This
project was initiated to examine how the new methods apply to typical building cases with
respect to the new requirements on sound insulation (within the expanded frequency range).

An inter-laboratory comparison (round robin) has been made in Mölndal, Sweden with the
support from Nordisk InnovationsCenter and the eight laboratories participating. The
operators made sound insulation measurements on 5 partitions located in the same building.
The operators were instructed to follow the procedures of ISO 140 parts 4, 7 and 14, as well
as the guidelines in the informative annex H of the national standard SS 25267. The
differences of the measured sound insulation and its components (sound pressure level
difference, reverberation time, partition area and receiving room volume) are analysed in part
A of this report. The sample standard deviation of the weighted airborne and impact sound
insulations and the spectrum adaption terms R’w (C;C50-3150) and L’n,w + CI,50-2500 were

Standard 90% confidence Standard 90% confidence

Estimated deviation, all (1,6*Standard- deviation regular (1,6*Standard
uncertainties, in cases deviation), all spaces deviation),
decibels, dB: cases regular spaces
R’w 1,0 1,7 0,7 1,1
R’w + C 1,2 1,9 0,8 1,3
R’w + Ctr 1,3 2,2 0,9 1,5
R’w + C50-3150 1,3 2,1 0,7 1,1
R’w + Ctr,50-3150 1,7 2,7 0,8 1,3

Standard 90% confidence

Estimated deviation, all (1,6*Standard
uncertainties, in cases deviation), all
decibels, dB: cases
L’n,w 0,7 1,1
L’n,w + CI 0,7 1,2
L’n,w + CI,50-2500
0,8 1,3

Part B deals with variations of sound insulation related to the choice of the area S of the
common partition and the volume V of the receiving room. In part A (measurements), the
sample standard deviation of the factor 10lg(ST/0,16V) was determined to 0,6-0,8 dB,
including the variation of reverberation time T. In part B.1, a brief survey was made among
the operators on the choice of S and V for 5 schematic but still realistic cases (with a fixed
value of T). The sample standard deviation was 0,7 dB (if one outlier was removed from the
set of data, case B6, 1,2 dB including this case). It can be concluded, that work should be

ii
made to improve the instructions of the measurement standard to make the choice of S and
V less ambiguous.

Additionally, sound insulation of 7 hypothetical building constructions was calculated by 4

operators according to EN 12354. The wo of these operators were not experienced with the
standardised calculation procedures. The standard deviation of the calculated weighted
airborne and impact sound insulations were:

Standard deviation, 4 operators, all cases

R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
3,0 3,0 2,6 2,3
Standard deviation, 4 operators, case 1 horisontally excluded
R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
2,6 2,6 1,8 1,4

The results of this comparison may be considered as examples of uncertainty that may occur
in real consultancy work, but the numbers of experienced operators who delivered results,
and the complexity of the calculation examples, prevent an interpretation of the results to
reliable estimates of the uncertainty of the calculation method.

In part C, a large set of calculations of sound insulation between rooms according to EN

12354 (ISO 15712) has been compared to field measurements. The resulting differences
yield a more reliable estimate of the practical uncertainty that may be expected in practical
planning work.

The comparison between measured and calculated sound insulation, in the table below,
refers to vertical sound transmission through concrete slabs with different floorings. The
number of comparisons are given for each type of weighted value.

Difference calculated-
measured insulation:
R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
…between the -0,17 0,42 1,87 1,91
averages
…standard deviation 2,3 1,6 4,4 2,9

…90%-confidence 3,5 3,0 5,1 2,7

limits (5% risk of non-
conf.)

From the results presented in part A and C, the following safety margins are recommended
for practical planning work, applicable when sound data of building elements have been
tested and documented properly, and the quality of workmanship is high. Under these
conditions, the uncertainty may be assumed less than reported from the cases of part C.
Measured sound insulation may occasionally be less satisfactory than predicted, by 1-2 dB, if
the following safety margins are applied during the planning process:

Practical safety margin to

requirement (verified by R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
measurements):
2 3 2 3

iii
Some practical aspects on measurements and calculations of sound insulation are
discussed, which may help improve the methods applied. The advice given may help the
building industry handle uncertainties in prediction and verification more efficiently.

The measurement uncertainty can be reduced to some extent by a careful measurement

work and an extended averaging procedure. The uncertainty of calculations of the
performance in a building with specified products may be reduced by a continuous
comparison between predicted and measured sound insulation. Input data for the
calculations may initially be taken from laboratory values or theoretical calculations, and
adjusted after some time if empirical experience proves there are systematic differences that
need to be compensated for. The manufacturers could take on the responsibility for this data
maintenance, but all actors of the building industry could contribute, with an open mind to
exchange of experience.

iv
Introduction
Uncertainties of measurement and prediction of the sound insulation in a building is an
important issue to the building industry, because they put increased expenses on all actors
of the building process who have to keep safety margins to prescribed requirements. This
report deals with estimates of both types of uncertainty and some recommendations on
suitable safety margins are given.

Requirements on sound insulation are described in the Swedish sound classification

standards SS 25267 and SS 25268. These standards are also intended to facilitate to
developers and tenants, how to express their demands of a subjectively good sound climate
of a dwelling in terms of standardized quantities (i.e. airborne sound insulation R and
normalized impact sound pressure level Ln). The uncertainties of measurement and
prediction of these quantities have an immediate consequence on which sound climate that
could be guaranteed by a developer, or a contractor. In practice, this means building
constructions must be chosen with reasonable safety margins to meet a strict requirement. If
the margin is known and taken into account, the risk of being disapproved at the final
inspection of the building is reduced. If the margin is unknown, this implies a risk. Risks that
can not be handled efficiently within the planning process increase building costs.

New standards for the measurement and the prediction (by calculation) of sound
insulation have been developed over the last years, and the frequency range of interest
has been expanded to include low frequencies (50-100 Hz). This project was initiated to
examine how the new methods apply to typical building cases with respect to the new
requirements on sound insulation (within the expanded frequency range).

In the first parts (A and B), the results from a round robin study is reported, where 8
operators from well known laboratories in the Nordic countries have measured and
calculated sound insulation in a defined set of building cases. From the individual results,
the uncertainties have been calculated, which are thought to be applicable as global
estimates of the uncertainty of the methods when these are applied to common types of
buildings for dwellings, offices, schools etc. It should be noted however, there may exist
other building cases where the conditions are different, and where the uncertainty may be
greater than estimated by the comparisons made in this report. In such cases, additional
evaluations may be done by the operator to estimate the actual uncertainty. Additional
information is given in earlier studies at Delta Akustik & Vibration [refs. 5 and 9].

Some data have been collected for the purpose of being analysed in the ISO working group
TC 43/SC 2/WG 18 on measurement uncertainties and the CEN working group TC126/WG
2 on the uncertainty in calculation. The goal is to develop methods to estimate the
uncertainty in the individual case rather than relying on global estimates, which is a
requirement from the Swedish board of accreditation (SWEDAC) and the European
cooperation for Accreditation of Laboratories. ISO TC 43 has decided to initiate a work to
upgrade the relevant measurement standards with instructions on the expression of
measurement uncertainty (ISO/TC 43 document N1002 sept. 2003). This is however a
complicated task to accomplish, and much work must be done to find feasible methods. This
report hopefully adds some information that will support the committee works.

1
Part A. Measurements of airborne and impact sound
insulation
Participating laboratories
We acknowledge the participation of the following laboratories:
Gärdhagen akustik AB, Mölndal, Sweden
Ingemansson Technology AB, Gothenburg, Sweden
SP National testing and research institute, Borås, Sweden
VTT Technical Research Centre of Finland, Esbo, Finland
NBI Norwegian Building Research Institute, Oslo, Norway
Delta Akustik & Vibration, Århus, Denmark
WSP Acoustics, Gothenburg, Sweden
Icelandic Building Research Institute, Reykjavik, Iceland

The project leader contributed with airborne measurement results in the case A only. The
measured data was collected mainly for the purpose of extending the comparison and to
examine the room acoustics of the test rooms. The measurement data are not reported
per operator since there is no need to identify individual results. Where needed, the results
or operator are labelled with a randomly chosen ID-number OP1-9.

Measurement venue and measurement cases

All measurement spaces are located in the venue Krokslätts Fabriker 37, SE-431 37
Mölndal. The building is owned by Husvärden AB, who admitted the spaces for
measurements. The measurements were made in partly furnished office spaces, not in use,
to which we had a personal key. The spaces and the furniture were not altered between each
measurement session.

Measurement cases (pictures follow on next pages):

A. R, Ln
office 47535 5th floor => large conference room 47433 4th floor (mobile wall opened)

B. R, Ln
office 47535 5th floor => large conference room 47433 4 th floor (mobile wall closed)

C. R, - - -
office 47535 5th floor <= large conference room 47433 4 th floor (mobile wall opened)

D. R, Ln
telephone room 5th floor 47533 => office 47535 (same floor, # 47534 intermediating)

E. R, - - -
reception open space 47431 4th floor => office 47535 5th floor (uses case D rev.time)

F. R, Ln
large conference room 47425 => meeting room 47424, 4th floor (no furniture)

G. R, - - -
large conference room 47425 => meeting room 47424, 4th floor (with furniture)

Some pictures describe the venue briefly:

2
The large conference room 47433 on the 4th floor, with the mobile wall folded:
Part with sound absorbing ceiling (left), the more reverberant high part (right):

3
A part of the large conference room 47433 is located under the office, the other part is
covered by a glass light. A low wall up to the breast height, and the windows on top, closes
the office space on the 5th floor. The wall is common with the conference room, but not the
windows:

The reception space, entrance to the conference room and the walkway outside of the office
47535 (arrows show measurement case E):

The conference room 47425 (irregular shape) and meeting room 47424 (regular shape, with
furniture*) used for measurement cased F and G:

*) The meeting room 47424 was measured with furniture (case G) and without furniture (case F).

4
Each operator reported to the project leader (X denotes measurement case):

• Ls-X source room sound pressure levels, space and time averaged, in dB
• Lm-X receiving room sound pressure levels, space and time averaged, in dB
• Lb-X background sound pressure levels, in dB
• Tm-X receiving room reverberation times, in seconds
• S_X partition area, in m2
• V_x volume of receiving room, in m3

The types of room/space were considered by the operators as reasonably representative

for both dwellings and offices where sound requirements apply. Measurement conditions
were altered by means of

• room size, room shape (case A-B, case D, case E)

• reverberation time (case A-B, case F-G)
• sound insulation (case D, case E)
• background sound pressure level (all cases, time dependent air handling on/off)

Note: As mentioned in the introduction, there may exist building cases where the conditions are
different from those found at the measurement venue, and where the uncertainty may be greater than
estimated by the comparisons made in this report. In such cases, additional evaluations of measure-
ment uncertainty of the actual case may be done by the operator according to [ref. 1] and [ref. 3].

Measurement procedure
All operators were asked to bring their own measurement equipment and to verify calibration
and appropriate function prior to departure. The project leader verified calibration labels on
the equipment before the measurements were initiated.

The operators were asked to make sound insulation measurements in 7 room cases
with different acoustical conditions, as described above. The measurements should be
made according to
• EN ISO 140 parts 4 and 7, basic field measurement procedures [ref 3]
• EN ISO 140 part 14, additional guidelines for measurements in complicated spaces
[ref. 3]
• SS 25267 annex H, additional guidelines for measurements in the extended frequency
range 50-3150 Hz [ref. 4]

The operators were welcomed to the measurement venue and briefed about the various
measurement cases. Then, the measurement case “A” was executed under observation of
the project leader. After this first measurement, the project leader summarized all
observations made on the measurement procedure and whether it was in accordance with
the standards listed above. After correction of the procedure, the measurement case A was
repeated (if necessary). The data from the first part were stored but not used thereafter, only
the data from the latter part was used in the statistical calculations presented below. The
project leader made no further comments or corrections to the measurement procedure of
the operator.

There were some deviations from the standardised procedures given in the call to the
operators, that were commented and in some cases corrected:

5
1. The measurement standards were not available to some operators at the
measurement venue. This was corrected after our comment.

2. Insufficient space between positions of the measurement microphone(s), particularly

the variation of height above the floor. Microphone positions were placed too close to
the walls and did not sample the entire volumes. The loudspeaker was too close to
the floor (case A, vertically, top-down). The loudspeaker stand was not equipped with
vibration insulated pads. Some operators did not use diffusing objects as
recommended by SS 25267/H and told us they never used this option. Improved after
our comment.

3. Uneven scanning speed and path of the moving microphone gives a higher time
weight of some samples with respect to the other samples. Corrected.

4. Some operators used 2 channels and measured Ls and Lm in parallel, some used a
single channel and measured Ls and Lm in sequence. The latter method is sensitive to
the stability of the sound source. The project leader and one of the operators checked
this stability separately. Variations were within 0,3 dB after 20 loud noise sequences
with the equipment used. Other operators told they used to check the stability from
time to time. One operator told they have had problems with stability of the
microphone polarization voltage of a dual channel system, giving random errors of
both microphone signals which were very difficult to detect since they occurred
occasionally.

5. Irregular supervision of signal to noise ratio (S/N) of the signal from the receiving
room microphone. The most common way to check S/N was simply to turn the noise
source on and off, and to observe the increase/decrease of the sound pressure level
(SPL) in the receiving room. All operators noticed that the ventilation system, giving a
medium and low frequency sound, was operating in irregular intervals. The S/N was
checked again and the margin was verified >10 dB in all cases. Rainfall noise on the
roof caused problems to one operator, which was stated in his report. This operator
had an advanced tool that monitored the Lmax-Leq difference each second of the
measurement period, and outliers were immediately spotted. Some analyzers had a
Lmax/Leq peak-and-hold facility, which gave information about any occurrence of a loud
disturbance during the measurement period. Poor S/N is known to be an important
cause for deteriorated sound insulation results in situ. Both the standards and the
analyzers should be more developed with respect to automated supervision and alerts
of S/N problems during the measurement sequences.

6. Reverberation decays were not inspected in detail, except by one operator. It was
said to be too time consuming. Some operators inspected the decays occasionally,
but relied on the quality indicators of the analyzer, e.g. question marks or asterisks. In
case of poor quality, new decays were acquired and they replaced the first values.
Some of the operators told, they always get a few question marks in the reverberation
results, and they do not investigate the cause (because of time restrictions). One
operator was not sure whether T20 or T30 was evaluated by the analyzer.

Decays were measured in the conference room (case A) by the project leader and by
one of the operators, and each of these decays were inspected. The on-set/off-set of
the regression lines were corrected in several cases. The resulting reverberation
times resembled the average values for all operators within ½ standard deviation of all
measurements in the low and mid frequency range. It can be concluded, that
inspecting and correcting the reverberation decays (when the automated regression
analysis has failed) reduces uncertainty of the reverberation time measurement
considerably.

6
 In an earlier study made for a manufacturer of sound analyzers, unfortunately not
published, the project leader compared the traditional method (interrupted noise) with
the new methods (based on impulse response) according to ISO/DIS 3382-2. The
latter method was found to be much less prone to random errors caused by the
calculation of the regression line by the analyzer softwares. Impulse response
methods may replace interrupted noise methods, or at least be given the status of
reference method. Particularly at frequencies below 200 Hz, a general experience is
that the error caused by wrong regression calculations may be substantial.

7. Several of the operators told they did not have dedicated training and education with
respect to room acoustics, sound insulation measurements and the handling of the
measurement equipment. Several operators told that this was the first occasion where
they had made measurements under observation and had relevant advice on
improvements. The education and training level of operators should be checked and
improved by the responsible management at the laboratories, if needed.

Measurement results

The main parameters studied are variations about the ensemble average of the sound
pressure level difference, reverberation time, partition area and the receiving room volume.
The absolute values are not reported, since these are related to the actual types of
construction, which are not relevant with respect to the goals of this investigation. All
measured values obtained by the operators are tabulated in Enclosure A. (Impact sound
results follow after the airborne sound analysis.)

Results are shown in the figures below as third octave band data 50-3150 Hz, followed by
the weighted airborne sound insulation including the spectrum adaption terms, in decibels
(dB):

The overall result, including all cases, for the airborne sound insulation is shown in figure 1.
The solid (red) line with triangular marks shows the standard deviation about the ensemble
average1. The solid (blue) line with squared marks shows the standard deviation of the
sound pressure level difference between the source and receiving rooms. The solid (green)
line with circular marks shows the standard deviation of the sound absorption term 10
log(S/A). The dashed (red) line with triangular marks shows the standard deviation of sound
insulation according to ISO 140-2:1991.

It can be observed from the figure 1, that the results obtained in this study resemble the data
tabulated in ISO 140-2 and the inter-laboratory comparison study made by Delta Akustik &
Vibration in 1992 [ref. 5]. However, there is a noticeable reduction of uncertainty in this study
at the low frequencies which may be explained by improvements of the measurement
procedures. The main part of the uncertainty is pertinent to the sound pressure level
difference ∆L, which depend heavily on the amount of spatial averaging.

1
All standard deviations in this report are calculated as the sample standard deviation of the
measured values [ref. 1].

7
 Figure 1. Std dev of dL, 10lg(S/A) and R

Frequency (Hz), weighted values

In the figure 4 below, the standard deviations are plotted as in figure 1, but two measurement
cases have been considered as statistical outliers and these cases have been removed from
the data series. One is case D, where there were difficulties to interpret the applicable
partition area S since there was no real common partition (one office space was located
between the source “telephone” room and the receiving “office”. Compare from the pictures
shown above. Some operators then chose a reference area 10 m2, some used the V/7,5 m2
rule (62/7.5=8.2 m2) according to ISO 140-4. One operator (OP2) chose to report the
standardized sound pressure level difference DnT instead of R’w. This has the same effect as
choosing S=V/3,1, i.e. 21 m2. The impact on the 10lg(S/A) factor is shown in figure 2, where
case D values excluding OP 2 have been calculated as well (blue dashed lines with circular
marks):

Frequency (Hz), weighted values

8
The ISO 140-4:1999 gives the following instructions on the determination of S:

The instruction from the project leader was however to report R’ and not DnT.

The other outlier is the case E, measurement from the reception space to the office room.
The sound pressure level difference varies considerably more in both case D and E than in
the other cases, where the source rooms were large and well defined rooms with a higher
diffusivity. The variation of sound pressure level difference about the ensemble average is
shown in figure 3:
Figure 3. Std dev of dL, all cases A-G

Frequency (Hz), weighted values

See encl. A for tabulated values. Removing case D and E gives the following uncertainties
for well defined spaces (thin dashed lines are copied from figure 1, for comparison):

Figure 4. Std dev of dL, 10lg(S/A) and R - excluding the D (tel.room) & E (reception) cases

Frequency (Hz), weighted values

9
The figure 4 shows some interesting changes as compared to figure 1. The uncertainty of R
is still explained mainly by the variations of ∆L, but there is a significant improvement. The
uncertainty of the weighted sound insulation R’w is now only 0,7 dB. Even more interesting,
the uncertainty does not increase when the low frequency spectrum adaption term C50-3150 is
added. This conclusion is concordant with the opinion among some building acoustic
consultants, who have had a positive experience of the extension of the frequency range
from 100-3150 Hz to 50-3150 Hz made to the Swedish building code (BBR) in 1999. Most
consultants were by then already using this extended frequency range and supported the
changes of code, even though some were very sceptical. It is not clear however, if the
consultants also changed the measurement procedure accordingly, so this comparison with
practical experience should merely be interpreted as an indication.

Examination of sound fields – case A

Which parts of the measurement procedure could be improved to reduce uncertainty?

The instructions on how S and V should be determined by the operator have been clarified
to some extent by the new EN ISO 140-14. Still however, some operators considered the
partition area seen from the receiving room instead of the source room. The standards are
ambiguous in this respect, they refer to the “partition under test”, “common area” etc. It
appears from this round robin, that the operators have read the standards differently, e.g. as
an average of two areas or even as “the radiating area” (in the receiving room). In case A,
S-values reported by the operators ranged from 18 to 26 m2. One plausible reason for the
large variation in case A is the complicated shape of the ceiling and walls of one end of the
conference room, c.f. the pictures in the first section of this report. The area is considerably
larger seen from the conference room than from the office.

The standards may clarify that the area S should be determined from the interior
dimensions of the source room. S may be defined as the area of a partition exposed to
the sound source AND this part must be facing the receiving room side in case of
staggered or stepped rooms.
In case no common area exists, it should be stated clearly that R must be reported with
a hypothetical common area S that is calculated from the relation V/3,1 and with a
remark that R then equals DnT. The rule “max(S; V/7,5)” of the present edition of ISO
140-4 should be changed to “max(S;V/3,1) accordingly, since this makes R always equal
to or greater than DnT.

Compare with the equation 5b from EN 12354-1 (ISO 15712-1):

What influence does the diffusivity of the sound fields have on the uncertainty?

The project leader made some detailed measurements in the case A, office-to-conference
room. The sound pressure levels were measured at 38 fixed positions in the source room
(Ls) (office 47535), 0,3 m apart, 0,5 m and 1,5 m above the floor, >0,5 m from the walls and
ceiling. The same microphone grid was used with a second loudspeaker position, thus a total
of 76 SPL were measured in the source room. The increase of SPL close to the loudspeaker
was observed and positions closer than 1,5 m from the omnidirectional loudspeaker were

10
 excluded. In the conference room (47433, the folded wall fixed in an opened position), the
grid of measurement positions covered 2x22 positions at 0,7 m distance from each other, 0,5
m and 1,5 m above the floor, >0,5 m from the walls and ceiling (Lm). The same 2 loudspeaker
positions were then used in the source room.

In addition, source room SPL’s were determined with 2+2 hand-held microphone sweeps
according to the instructions of annex H of SS 25267, see figure 5. The radius of the
cylindrical sweeping movement was 1,1 m, the speed about 0,3 m/s. The measurement was
repeated but the microphone paths and loudspeaker positions were altered, thus 8 different
sweeps were made in the source room. The standard deviation and the room average SPL
was compared with those obtained by the 76 fixed microphone positions. The agreement is
satisfying, as listed in the following table:

Difference mean SPL 8 Frequency, Hz

sweeps – 76 fix-pos
50 63 80 100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150
-0,65 -1,27 -0,18 -0,76 -0,81 -0,65 0,00 0,13 -0,31 -0,75 -0,56 -0,48 -0,16 -0,64 -0,57 -0,01 -0,89 -0,99 -0,87
Diff. of standard dev. 8
sweeps – 40 fix-pos
0,03 0,40 -0,10 0,34 0,02 0,01 -0,21 -0,02 -0,09 -0,15 -0,09 -0,08 -0,07 0,10 -0,08 -0,01 0,06 -0,08 0,00

Figure 5. Measurements of sound pressure levels with a hand-held swept microphone

The reverberation time was measured in the receiving room using 2 loudspeaker positions,
5+5 fixed microphone positions (distributed over the entire space). The measurement was
repeated, thus 20 different microphone positions were used. 2 decays were recorded in each
position. Each decay was inspected by the project leader. Decays with a correlation to the
straight line less than about 0,96 were rejected or the on-set/off-set points were adjusted to
make the decay regression line fit the early part (-5,-20 dB) of the decay curve. At low
frequencies, erroneous regression lines were common and the measurements had to be
repeated several times. The following reverberation times (average and standard deviation)
were obtained in the conference room:

11
From the reverberation times, there is no reason to believe that there should be excessive
spatial variations of the sound pressure level in the receiving room. In the source room, there
are no sound absorbing tiles, nor furniture, but the shape of ceiling and diffusive effects of a
large table and some bookshelfs were judged to be sufficient to make the sound pressure
diffused. The standard deviation of Ls and Lm were then compared to the guidelines of ISO
140-14, for the purpose of checking the diffusivity of the measurement rooms. See figure A.1
from ISO 140-14 and figure 6:

Figure A.1 – Theoretical values of the spatial standard deviation of the sound pressure level measured at
ten microphone positions in empty rooms with different volumes

12
 Figure 6. Comparison of SPL std-dev with ISO 140-14 Annex A

Frequency, Hz

It appears from the figure 6, that the measurement spaces in case A (both the office and
conference room) are less diffuse than as anticipated from the reverberation times. This
may be explained by the irregular shapes of both spaces and the uneven distribution of
sound absorbing tiles in the receiving room.

Figure 7 shows a comparison, where the uncertainty of case A is estimated from the
measured standard deviation of the Ls, Lm and T. The variation of the ratio S/V was
estimated by the project leader from some plausible variations of geometrical dimensions, as
seen from the source room or the receiving room. A standard deviation of 0,5 dB is added to
the uncertainty budget to take variations in instrumentation sensitivity into account. This
value (0,5 dB) has been found by comparison of calibration data for sound analyzers at SP,
Sweden.

The agreement is satisfying at low frequencies, but at high frequencies the estimated
uncertainty exceeds the actual uncertainty (as determined by the operators). The reasons for
this deviation were not investigated further, but additional analysis of the data will be
proposed to the ISO TC 43/SC 2/WG 18.

13
 Figure 7. Case A: Standard deviation of R - All operators vs SAU estimate from (Ls, Lm, T, S/V)

Frequency (Hz)

From the figures presented, it may be concluded that the rooms used for measurement case
A do not give ideally diffuse sound field, that would be suitable for precise sound insulation
measurements. The other cases may also be considered as reasonably complicated. The
uncertainty determined by comparison using all cases in this study is likely to be on the safe
side if it is used as a general estimate of measurement uncertainty pertinent to the standar-
dised measurement methods applied in this study. If the cases D and E are excluded, the
standard deviation given in figure 4 may be relevant to “regular spaces” with reasonable
diffuse sound fields and well determined geometry. It seems feasible to estimate the uncer-
tainty of an individual case by analysis of the standard deviation of each parameter from
which an uncertainty budget may be established. Uncertainty of measurement equipment
sensitivity and area/volume factors should be added if the reproducibility (inter-laboratory)
uncertainty is estimated. This feature could be implemented in the software of most modern
analyzers, but this procedure should be analysed in detail (e.g. by the ISO WG 18) before a
general guideline is proposed for a ISO 140-revision.

Finally, based on the results reported above, the following uncertainties are suggested as
estimates of uncertainty of weighted airborne sound insulation in buildings (using the
standards), in case they are unknown in the individual case:

Standard 90% confidence Standard 90% confidence

Estimated deviation, all (1,6*Standard- deviation regular (1,6*Standard
uncertainties, in cases deviation), all spaces deviation),
decibels, dB: cases regular spaces
R’w 1,0 1,7 0,7 1,1
R’w + C 1,2 1,9 0,8 1,3
R’w + Ctr 1,3 2,2 0,9 1,5
R’w + C50-3150 1,3 2,1 0,7 1,1
R’w + Ctr,50-3150 1,7 2,7 0,8 1,3

14
Impact sound
The normalized impact sound pressure levels Ln, in decibels (dB), were determined by 4 of
the operators in four cases (A, B, D, F). The operators were not the same in all cases. The
overall variation of results (standard deviation) are demonstrated by figures 8 and 9 below,
where comparisons with the standard deviation of reproducibility taken from ISO 140-2 are
included:

Figure 8. Std dev of dLn, cases A, B, D, F

Figure 9. Overview of all dLn, all cases

15
Some of the tapping machines were calibrated by the manufacturer or by SP, a few were
only checked by the laboratory with respect to heights of foots. The rotational speed and the
shape of the hammers are also of importance, and tapping machines should be calibrated
when used in situ.

One operator made a mistake, where the volumes of the receiving rooms in cases A and C
were exchanged. The resulting error, almost 3 dB in both cases, was found by the project
leader and corrected. The figures 8 and 9 are based on the corrected volumes. This mistake
illustrates the importance of being concentrated on the task, above all that there should be
no disturbing events at the measurement site. There are many measures and decisions
taken by the operator, that are sensitive to errors.

Several operators noticed by listening to the tapping machine in the source room in case F,
that the parquet floor was not attached firmly to the slab floor at all points. The variation of
impact sound pressure level between the operators in case F increased from 1600 Hz to
about twice the standard deviation for all cases A, B, D, F. Thus, it is important to listen to the
tapping machine when it operates, and to study the variation of sound levels. The operator
should increase the number of positions whenever the spatial variation is higher than normal.

Finally, based on the results reported above, the following uncertainties are suggested as
estimates of uncertainty of weighted impact sound insulation in buildings in case they are
unknown in the individual case:

Standard 90 %
deviation, confidence
all cases (1,6*Standard
Estimated uncertainties, deviation), all
in decibels, dB: cases
L’n,w 0,7 1,1
L’n,w + CI 0,7 1,2
L’n,w + CI,50-2500 0,8 1,3

16
Part B. Calculations according to EN 12354 (ISO 15712)

B.1 Determination of partition area and receiving room volume

During the planning phase of this project, some operators mentioned that variations of
sound insulation may sometimes be explained by the choice of the area S of the common
partition and the volume V of the receiving room. In part A (measurements), the sample
standard deviation of the factor 10lg(S/A) was determined to 0,6-0,8 dB, including the
variation of reverberation time. In this part B.1, a brief survey was made among the
operators on the choice of S and V for some schematic but still realistic cases.

All operators were instructed to estimate the common partition area S and the receiving room
volume V for the following schematic cases B1-B5. If not stated elsewhere, the room height
is 2,6 m. The “drawings” below are not scaled correctly (on purpose), which is commented
below.
Case B1.

B11. Measurement direction: bedroom room to living room S =___m2, V =___m3

B12. Measurement direction: living room to bedroom S =___m2, V =___m3

17
B2.

B21. Measurement direction: bedroom to living room S =___m2, V =___m3

B22. Measurement direction: living room to bedroom S =___m2, V=___m3

Remark
One operator made extensive remarks on the choice of S and V: For wardrobes that are
empty and placed close to the wall, no reduction of S is made. For wardrobes filled with
clothes, S is reduced since it is the wall that is the test object. Wall surface above the
wardrobe is however included if it is visible from the source room.

18
B3.
Vertical sound insulation between a large open space (attic) and a bedroom below this
space:
B31. Impact sound level from the attic to the bedroom: V =____m3
B32. Airborne sound insulation measured from the attic to the bedroom: S=___m2, V=___m3
Point out where the tapping machine, the loudspeaker and the microphone should be placed.

Position of the tapping machine

All operators chose the option: Within the common partition area
Position of the loudspeaker
The operators chose a distance 2,5-0 m outside of the common partition, where the walls of
the bedroom limited the common partition.

B4.
B41. Measurement direction kitchen/living room (upper arrow), S=___m2, V=___m3
B42. Measurement direction kitchen/living room (lower arrow), S=___m2, V=___m3

19
B5
B5. Room 47431 reception space – Room 47424 small meeting room in the measurement
venue: S=___m2, V=___m3 Show how loudspeaker and microphones should be placed.

Most operators chose a more or less direct sound exposure of the two walls exposed to the
reception space, according to the illustration below. It is likely that this type of measurement
setup will be sensitive to near field effects when the source room sound pressure level is
determined. It appeared that the operators have not yet got used to the ISO 140-14, and the
practical application of the guidelines. In course of a training the staff at the laboratories,
practical examples of application of this part of the ISO 140 should be discussed.

A typical position of the loudspeaker (small circles) and source space microphone positions
(large circles). The loudspeaker is situated at 45 degrees facing the common partition:

20
B6
Measurement diagonally between two open-plan type of dwellings, both with two floors, with
openings for the internal stairs. Sound insulation from the living room at the first floor to the
upper part (2nd floor) of the living room in the opposite dwelling. The operators were given
additional information, that sound may be transmitted through the wall at the 1st floor and
vertically through the light in the floor of the receiving space. Sound may also be transmitted
through the stairs opening in the source space, and then horisontally through the wall to the
2nd floor of the receiving space. The WC room is located only on the upper floor, thus the
separating area of the walls are different on the two floors. S=___m2, V=___m3.

21
B7
On the upper floor, there is a large living room (open space). On the ground floor, there is an
entrance, a large wardrobe and a bedroom which belongs to another tenant. Determine S for
the partition between the living room and the bedroom, as well as V of the bedroom.

The operators chose to place the tapping machine within the walls of the sleeping room
below, not close to the ”matplats” nor the stairs. The loudspeaker should be placed close to
the bedroom walls, but not above it.

22
B.1 Analysis
The analysis of the cases B11-B7 was made with respect to the ratio 10*log(S/A) assuming
an fixed arbitrary reverberation time, i.e. it is the variation of the ratio 10*log(S/V) that is
analysed. The following average values and standard deviations were encountered:

CASE AVERAGE OF 10*LOG(S/A) , dB STANDARD DEVIATION OF 10*LOG(S/A),

dB
B11 -9,0 0,2
B12 -6,1 0,6
B21 -6,3 0,3
B22 -3,3 0,6
B32 -4,1 0,0
B41 -10,0 1,4
B42 -9,0 1,1
B5 -3,0 1,0
B6 -7,3 2,0
B7 -4,1 0,1
STANDARD DEVIATION, 1,2
ALL CASES
STANDARD DEVIATION, 0,7
EXCL. CASE B6

As mentioned above, the “drawings” B1-B7 were not scaled properly, as may occur in real
consultancy work, e.g. when a measurement or calculation is being prepared on the basis of
preliminary data. The uncertainty calculated and reported above would probably be reduced
somewhat if the drawings had been more detailed and correctly scaled, and if the operator
would have had the opportunity to get more information about the floor plans etc.

Some operators discussed ambiguous choices of S and V in some detail, especially whether
wardrobes, kitchen cupboards etc should be considered. However, there were no obvious
changes to the ISO 140-14 proposed by the operators. It seems that after some education
and training on the application of the new part -14, some examples published in a handbook
or similar publication would help the operators make consistent choices when the
circumstances on the test venue are difficult. In any case, a test report should state which
choices have been made and which paragraphs of the measurement standard were applied.

23
B.2 Sound insulation calculated according to EN 12354

In this part B2, the operators were asked to calculate the sound insulation of another set of
building examples according to EN 12354 (ISO 15712) parts 1 and 2. They were asked to
apply their own input data for the building elements first, then repeat the calculations using
input data from the built-in database. One operator preferred to report estimates based on
experience rather than calculations, but these estimates have not been included in the
calculated uncertainty.

The reported data are: R’w, R’w + C50-3150 and L’n,w, L’n,w + CI,50-2500 in each of the cases 1-4.

Case 1.
Hollow core concrete slab type HD/F 120/200, 295 kg/m2 including joints. Flooring with
wooden parquet 14 mm on a PE-foam 3 mm. Partition wall made from massive conrete 200
mm. External wall made from plasterboards 13 mm on wooden studs 170 mm, thermal
insulation with mineral wool, external plasterboard, wooden laths and wooden panel. Room
walls within dwellings built by plasterboards 13 mm on steel C-studs 70 mm. Room
dimensions WDH 4,5x3x2,6 m, excluding entrance space.

• Horizontal sound insulation, airborne and impact

• Vertical sound insulation, airborne and impact

24
Case 2.
As in case 1, but partition walls are built by 2x13 plasterboards, 70+70 wooden studs
(separated), 140 mineral wool, 2x13 plasterboards.

• • Horizontal sound insulation, airborne and impact

• • Vertical sound insulation, airborne and impact

Case 3. Concrete slabs and walls, cast in situ.

As for case 1, but slab is made from cast massive concrete 250 mm. Flooring with parquet
flooring 14 mm on PE-foam 3 mm. Partition wall built by massive concrete 200 mm cast in
situ. External wall: Sandwich concrete element by 150 mm concrete, 150 mm mineral wool,
70 mm external concrete. Room walls as in case 1.

• Horizontal sound insulation, airborne and impact

• Vertical sound insulation, airborne and impact

Case 4.
All rooms facing the partition walls are WDH 5x4x2,6 m.

• Horizontal sound insulation, airborne and impact, both upper and lower floors
• Vertical sound insulation, airborne and impact, from upper to the lower floor

25
B.2 Analysis
Four operators reported calculations, of which 2 are experienced with the standard and the
software used (BASTIAN). Two operators made calculations for the first time in this project.
The standard deviation of the calculated weighted airborne and impact sound insulations
were:

Standard deviation, 4 operators, all cases

R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
3,0 3,0 2,6 2,3
Standard deviation, 4 operators, case 1 horisontally excluded
R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
2,6 2,6 1,8 1,4

The results of this comparison may be considered as examples of uncertainty that may occur
in real consultancy work, but the low numbers of experienced operators who delivered
results, and the complexity of the calculation examples, prevent an interpretation of the
results to reliable estimates of the uncertainty of the calculation method.

For the purpose of estimating the uncertainty of calculations as compared to measurements

in buildings, a set of measurement data was supplied by the Spenncon (Consolis) company
in Norway. Further comparisons have also been made within another research work
sponsored by the SBUF, the Swedish research foundation of building contractors [refs. 6, 7].
These data are presented and discussed with respect to uncertainty, in the part C below.

26
Part C. Experiences of agreement between measurements
and calculations in situ

Introduction
When a calculation model of a building is established according to EN 12354 (ISO
15712), several decisions must be taken by the operator.
• Appropriate input data for each building element that enclose the transmission rooms
must be chosen
Note: Data for elements are published by some manufacturers and in the SAU/Delta database
that is connected to the software BASTIAN [ref. 10]. The database was used in this project and
presently it is used also by most building acoustic consultants in the Nordic countries. The data
for elements are mainly based on laboratory measurements, but some data are based on
theoretical analysis or field measurements. For the heavy elements (made from concrete), the
database content has been calculated by Delta Akustik & Vibration on the basis of theory and
extensive comparisons with both laboratory and field measurements.

• The operator has to define the size of partitions and room dimensions
Note: The size of the model may differ from the actual physical dimensions, particularly where
the geometry of the building is complicated, e.g. staggered rooms, open-plan spaces etc.

• The operator has to define the junctions between the building elements
Note: In BASTIAN, there are several predefined types of junction available, according to EN
12354. The user can not change the vibration attenuation at the chosen junction, this quantity is
calculated by the software.

• The actual performance of elements in the building depends on which elements it is

connected to, and the quality of workmanship (air leakages, structural sound bridges
etc.)
Note: Another important issue is the choice of loss factor, that relates to the structural vibration
energy transmission through external building elements (i.e., elements that are connected firmly
to the partition and flanking structure and therefore extract energy from the “system”, but do not
increase the flanking transmission to the receiving room). This loss factor is particularly
important to heavy building systems. Large concrete slabs, with room partitions built by light
weight plasterboard walls, will give substantially higher sound insulation vertically than buildings
with heavy partition walls. This is sometimes referred to as the “area factor” and it is treated in
EN 12354. Concrete slabs cast in situ are likely to get another loss factor than pre-cast concrete
slabs, as is discussed below.

• One may assume that calculation results may vary depending on the operators
experience and which building element data in the database is considered most
appropriate to model the real building construction.

• When comparisons are made with respect to the conformance to field measurements,
measurement uncertainty has to be taken into account (safety margin).

For these reasons, there was a need to compare measurements and calculations of real
buildings, to see whether there are systematic or random errors that need to be taken into
account. The sound insulations in a variety of building cases (mainly residential buildings)
have been analysed by the project leader, using the BASTIAN software and the SAU/Delta
database of building elements. Measurement data are taken from the SBUF study [refs. 6, 7]
and a set of data supplied by the Spenncon company in Norway (not published). The
comparison between measured and calculated sound insulation, in the table below, refers to
vertical sound transmission through concrete slabs with different floorings. The number of
comparisons are given for each type of weighted value.

27
Difference calculated-
measured insulation:
R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
…between -0,17 0,42 1,87 1,91
the averages
…standard 2,3 1,6 4,4 2,9
deviation
…90%-confidence limits 3,5 3,0 5,1 2,7
(5% risk of non-conf.)
Number of 26 36 30 43
comparisons
Measured average
insulations 59,4 57,6 54,1 51,3

Remark
The measured building cases were not documented with respect to all building products
used for the respective building. When the calculations were made, data for constructions
typical for the age and type of building were used when no other information was available.
Naturally, such assumptions increase uncertainty of the calculations. In spite of these
uncertainties, the table above shows that the 90% confidence limits agree reasonably well
with a common experience, that a 3 dB margin is sufficient for most practical applications.
The risk of a field measurement (performed according to all relevant standards) being non-
conformant to the requirement is then less than 5%. This confidence limit corresponds to a
coverage factor of k=1,6, assuming the random variations being normal distributed and the
influence of systematic errors being negligible. If well documented building products are
used, and the quality of workmanship is high, the margins should be possible to reduce,
according to the recommendation below.

Note: In the context of calibration, a 95% confidence interval (double sided) is recommended by EA
[ref. 1] and k=2 is adviced by EA. For practical planning work, this would imply a more strict safety
margin (2.5%) with increase of building costs as a consequence.

The large variation in L’n,w is likely to be caused by several factors. The average impact
sound pressure level is low (L’n,w – average 54 dB) which indicates that the elastic function of
the flooring is generally satisfactory. The high frequency impact sound insulation may vary
considerably with minor structural sound bridges, aging of elastic interlayers and difficulties
to measure the low sound pressure levels (disturbed by background noise).

Recommendations of safety margins

From the results presented in part A and C, the following safety margins are recommended
for practical planning work, applicable when sound data of building elements have been
tested and documented properly, and the quality of workmanship is high. Under these
conditions, the uncertainty may be assumed less than reported from the cases of part C.
Measured sound insulation may occasionally be less satisfactory than predicted, by 1-2 dB, if
the following safety margins are applied during the planning process:

Practical safety margin to

requirement (verified by R’w R’w + C50-3150 L’n,w L’n,w + CI,50-2500
measurements):
2 3 2 3

28
No comparison has been made of data for light weight slabs, but the practical experience is
that the margin must be increased compared to the uncertainty values given above. This
depends on the type of product and the quality of workmanship. Some indications are given
in [ref. 9].

Hollow core concrete slabs – loss factors

For hollow core concrete elements, airborne sound insulation may be treated as for massive
elements. Impact sound pressure levels (L’n) of elements differ from slabs cast in situ
(without joints). Empirical data (measurements) show a rapid growth of L’n at high
frequencies. This reason has not been examined thoroughly (to our knowledge), but it could
possibly be explained by the fact that the tapping machine makes impact on one element
only, and the vibration transmission could be hindered by poor adhesion of the mortar
between the elements.

This may be a reasonable explanation, because the joints, as cast in the normal case, do
not transmit moment forces efficiently. Shear forces could be expected to be transmitted
more efficiently. The input data of hollow core elements in the database are adjusted after
comparison with field measurements to take this effect into account.

For hollow core concrete elements, attached to each other with a joint cast concrete (mortar),
or a supporting beam, without a reinforced thick topping (cast in the building), there are
reasons to look more into detail to the energy dissipation to connecting structures (i.e. the
area factor), and how such concrete slabs should be modelled using EN 12354 (e.g. with the
BASTIAN software).

Is it realistic to correct the sound insulation with respect to structural losses to connected
slabs (area factor), as is illustrated in the figure 4.1 below (taken from [ref. 8]) ?

The Swedish institute ”Stålbyggnadsinstitutet” made a survey in 1994 [ref. 8], where the
propagation of vibrations across a joint with a rigid steel support beam was measured.
Different types of casting of the joint and filler inside the steel beam were tested. The
measurement positions on the elements are illustrated in the figure 4.2 from [ref. 8]. The
results show, that vibration levels on the opposite side of the beam (as seen from the tapping
machine) increase when the joint cast is altered to increase the stiffness. This means that the
joint reflects less energy back to the source when the stiffness of the joint is increased. On
the other side, the vibration levels are reduced if the joint stiffness is reduced.

29
Figure 4.1. The propagation of structure borne Figure 4.2. Measurement setup with tapping
sound in a building structure with load bearing machine (|____|) and accelerometers (O).
massive walls or steel columns and beams

From ref. 8, figure 4.1. Vibration levels. Solid line; cast with regular mortar. Dashed line; liquid mortar with
ABS polymers. Dotted line; polystyrene foam insulation of the beam webs.

30
 These results may be interpreted to predict how vibrations and impact sound in a building
would be influenced by the stiffness of the joints between the elements. Vertically, impact
sound would increase, but horizontally, sound transmission would be reduced if the stiffness
of the joints is reduced. Thus, the vibrations are “isolated” within the excited element.

Measurements have been made in a residential building with small dwellings, divided by
plasterboard walls, on a slab made by hollow core concrete element slab attached to a
structural steel frame, without a topping. The measurements show how vibrations are
attenuated with distance from the tapping machine in two directions: in parallel to the
elements and perpendicular. The results are shown in the figures 10 below. These
measurements confirm, that there is a significant attenuation between hollow core elements
when there is only a joint casting with regular mortar and no reinforced topping. Further
comparisons with field measurements have confirmed that the same tendency appears also
when a thin nonreinforced topping is cast on top of the concrete elements. From a
comparison with different settings in the BASTIAN software, it was concluded that the area
factor is taken care of without any special settings with respect to structural losses.

Note: For massive concrete slabs (cast without joints) and rooms divided by light weight walls, the
structural loss factor seems to be underestimated and there are field data that proves higher structural
loss factor than estimated by BASTIAN (default setting). In BASTIAN, the structural loss factor may be
forced by the user. Advice on how this can be done is described at the FAQ Advanced section of the
homepage www.simmons.se. At this section, advices of calculation settings are published
continuously.

Figure 10. Small dwellings. Ln measured horisontally, Small dwellings. Horisontal measurement.
from staircase.

Figure 10 Impact sound measurements on hollow core concrete elements, without a topping.
Left: Measurements at increasing distance (number of joints between elements) on the bare concrete slab
(without a flooring).
Right: Impact sound predicted according to EN 12354-2, using data for elements fitted to
vertical measurements (thin lines) compared to measured levels (thick line). With an efficiently sound
insulating floating access floor.

In the study by Stålbyggnadsinstitutet [ref. 8], there are also comparisons between laboratory
data and field measurement data for some typical residential buildings with hollow core or
cast concrete slabs, shown in the picture below. This comparison confirms, that the field
measured sound insulation (dark bars) is typically 3-10 dB better than the insulation in the
laboratory (circles).

31
This is explained by the authors as the influence on the loss factor of the slab that is due to
energy dissipation to other parts of the building structure (i.e. the area factor). In this
collection of data, there are no examples of hollow core concrete elements assembled on
steel beams without a topping. That building technique has only been used in residential
buildings during the last 10-15 years. The elements are then combined with floating access
floors that take care of the uneven floor level of the HD elements, and increase sound
insulation.

In 1997, Delta Akustik & Vibration conducted a study on the input data for calculations
according to EN 12354 [ref. 9]. About 200 field measurements from the Nordic countries
were compared with calculations according to EN 12354 parts 1 and 2, with respect to both
heavy and light weight structures. The main findings are concluded in the tables below. The
average difference between the calculated and the measured results were negligible. The
90% confidence limit value was about 3-4 dB for heavy constructions and >5 dB for light
weight constructions. In this study however, some measurements were not satisfactory
documented and the project members had to apply general assumptions on some details of
the constructions of the buildings where the measurements were made. Therefore, it is
reasonable to assume that the uncertainty would be reduced if there is good documentation
of all constructions. The tabulated uncertainties may therefore be regarded as conservative:

Table 4.1. Average, standard deviation, and 90% confidence limits for the difference between calculated
and measured R’w-values.

32
Table 4.2. Average, standard deviation, and 90% confidence limits for the difference between calculated
and measured L’n,w-values vertically.

The uncertainty of calculations may be reduced by a continuous comparison between

predicted and measured sound insulation. Input data for the calculations may initially be
taken from laboratory values, and adjusted after some time if empirical experience prove
there are systematic differences that need to be compensated. The manufacturer should
therefore keep an open dialogue with the contractors and their consultants, and make
adjustments to product documentation whenever needed.

33
Recommended publication and presentation of the results
The results of this project may be presented by means of several information activities,
as decided by Nordisk InnovationsCenter:

• as an official Nordtest (Nordisk InnovationsCenter) technical report

• an invited paper and an oral presentation at the conference “Uncertainties in noise
measurements and prediction”, arranged 27-29 June 2005 in Le Mans, France, by the
European division of the International Institute of Noise Control Engineering (I-INCE),
supported by the European Assoc. of Acoustics (EAA). A special invitation has been
made to members of the CEN/TC126 and TC43/SC2. Ref. www.uncertainty-noise.org
• an invited paper and oral presentation at the conference “Forum Acusticum” 29-31
August 2005 in Budapest, Hungary, in cooperation with the CEN/TC 126/WG 2
convenor (from TNO Holland). Ref. www.fa2005.org
• a contributed paper to the special issue of the general magazine “Bygg & Teknik” in
April 2005, which is widely spread among building consultants in the Nordic countries.
Ref. www.byggteknikforlaget.se
• a paper in the journal “LjudBladet” of the Swedish acousticians society (SAS)
• a copy to the convenors of CEN/TC126/WG 2 and TC43/SC2/WG 18, working on
procedures for the handling of uncertainties in calculations and measurements
• a link to the homepage of Nordisk InnovationsCenter or a PDF-copy of the report at our
homepage, available for direct download

Acknowledgements

We acknowledge the generous support of this study by

1. Nordisk InnovationsCenter (financial support)
2. SBUF Svenska Byggbranschens Utvecklingsfond (travel expenses Forum Acusticum)
3. Husvärden AB (admittance of the measurement venue)
4. Svenska Betongindustrin (Betongvaruindustrin, Fabriksbetongindustrin, Cementa)
5. Spenncon (Consolis)
6. The participating laboratories (time costs)
7. Henrik Olesen and Dan Brösted Pedersen at Delta Akustik & Vibration, Denmark
8. Klas Hagberg, WSP, Sweden
9. Hans Goydke, ISO TC 43/SC 2/WG 18.

34
References
1. EAL-R2-Sv Angivande av mätosäkerhet vid kalibrering. In Swedish, translated from
the original publication EAL-R2. European cooperation for Accreditation of
Laboratories. Styrelsen för ackreditering och teknisk kontroll SWEDAC
(www.swedac.se).
2. GUM Guide to the expression of Uncertainty in Measurement. International
organisation of standardisation, ISO. ISBN 92-67-10188-9. www.iso.ch.
3. EN ISO 140 parts 2, 4, 7 and 14. International organisation of standardisation, ISO.
www.iso.ch
4. SS 25267:2004 Acoustics – Sound classification of spaces in buildings – Dwellings.
SIS förlag, www.sis.se. In Swedish (a translation to English is under preparation)
5. Nordic Intercomparison Programme in the field of Acoustics. 3. Measurement: Field
measurements of airborne sound insulation. Delta Akustik & Vibration, Nordtest
project report No 1023-92. www.delta.dk
6. Simmons, C. Ljudisolering i bostadshus byggda 1880-2000. Indata för beräkningar
enligt SS-EN 12354. SBUF projekt 11254. Sveriges Byggindustrier FoU-Väst
RAPPORT 0405. ISSN 1402-7410. (www.bygg.org. In Swedish. Summary in English,
see ref. 6)
7. Simmons, C. Reliable building element sound insulation data for EN 12354
calculations facilitates analysis of Swedish dwelling houses. Inter-noise 2004 Prague.
(www.simmons.se)
8. Andersson, J. Stålbyggnadsinstitutet publikation 144 (1994). ISBN 91 7127 002 7. In
Swedish.
9. Pedersen, D.B. Delta Akustik & Vibration. Nordic basis of calculation of sound
insulation in buildings. Report Nordtest project 1346-97. www.delta.dk.
10. BASTIAN version 2.1. User manual. Datakustik GmbH 2003. www.datakustik.de

35
Enclosure A

Table A1. Sound pressure level ∆L differences about the average value for all operators, measurement case A
Frequency OP1 OP2 OP3a OP3b OP4 OP5 OP6 OP7 OP8 OP9 dL_A_stddev
50 2,57 0,80 -2,07 -2,47 -0,69 -6,37 2,33 1,93 0,93 3,03 2,94
63 1,70 -1,42 2,53 -1,97 -2,08 -3,47 2,43 5,03 -2,87 0,13 2,81
80 -2,63 -0,91 1,17 -2,63 0,12 -0,73 2,17 2,27 -0,33 1,47 1,79
100 0,29 2,48 -3,46 -3,96 1,94 0,54 -0,96 3,04 1,94 -1,86 2,48
125 2,57 -0,57 -1,38 -2,68 -0,48 0,32 -0,88 1,52 0,62 0,92 1,51
160 0,87 -3,39 2,08 -0,52 -1,32 -0,32 -1,92 3,08 0,78 0,68 1,91
200 -0,70 0,57 -0,87 0,13 -0,97 -0,37 -1,47 2,93 1,53 -0,77 1,35
250 0,26 -0,44 -1,39 -0,39 0,01 0,61 0,61 0,51 0,61 -0,39 0,66
315 -0,13 0,48 0,03 -1,27 0,33 -0,27 0,33 0,13 0,23 0,13 0,50
400 1,06 1,40 0,08 -0,62 0,48 -1,52 -1,02 0,68 -1,82 1,28 1,18
500 0,77 -0,72 -0,34 -0,04 0,46 -0,44 -0,94 1,06 0,06 0,16 0,64
630 1,21 -1,08 -0,60 0,90 -0,80 -0,20 -0,10 1,00 0,20 -0,50 0,80
800 1,01 -0,61 -0,69 0,71 -0,19 -0,59 -0,89 0,61 0,71 -0,09 0,70
1000 1,19 -0,63 -0,67 0,03 0,23 0,03 -1,17 0,43 0,33 0,23 0,67
1250 0,85 -1,00 -0,53 0,77 -0,33 0,37 -0,93 0,67 0,17 -0,03 0,68
1600 0,92 -0,88 -0,24 0,76 -0,84 0,36 -0,44 0,76 -0,34 -0,04 0,66
2000 0,56 -1,20 0,17 0,57 -0,43 -0,13 0,27 0,37 -0,43 0,27 0,56
2500 0,33 -0,81 0,22 0,42 -0,48 -0,28 0,12 0,12 0,82 -0,48 0,50
3150 -1,19 0,11 0,97 0,77 -0,03 -0,23 0,37 -0,73 0,27 -0,33 0,66

E1
Table A2. Reverberation time T differences about the average value for all operators, measurement case A
Frequency OP1 OP2 OP3a OP3b OP4 OP5 OP6 OP7 OP8 OP9 dT_A_stddev
50 0,31 0,07 -0,11 -0,01 0,09 -0,22 0,01 0,11 -0,12 -0,16 0,16
63 -0,11 0,05 -0,04 0,04 0,13 -0,19 0,34 0,19 -0,07 0,03 0,15
80 -0,04 0,15 0,02 -0,06 0,20 -0,15 0,01 -0,07 -0,16 0,09 0,12
100 0,04 0,01 0,02 0,07 -0,04 -0,10 0,03 0,02 -0,01 0,01 0,05
125 -0,03 -0,01 0,03 0,04 0,02 0,05 0,02 -0,01 -0,08 -0,03 0,04
160 -0,06 -0,02 0,01 -0,05 0,07 0,07 -0,02 0,01 -0,02 -0,01 0,04
200 -0,03 0,03 0,07 -0,03 0,01 0,01 0,00 -0,05 -0,02 -0,03 0,04
250 -0,01 0,06 0,04 -0,05 0,09 -0,06 0,09 -0,07 0,02 0,00 0,06
315 0,01 -0,01 -0,05 -0,01 0,03 -0,04 0,01 0,05 0,01 0,00 0,03
400 0,03 0,03 -0,06 0,00 0,08 -0,10 0,04 0,00 -0,02 0,00 0,05
500 0,05 -0,02 -0,10 -0,07 0,01 0,07 -0,05 0,06 0,02 -0,05 0,06
630 0,00 0,02 -0,07 -0,09 0,06 0,16 -0,02 0,01 0,00 -0,05 0,07
800 0,02 0,04 -0,02 -0,06 0,06 -0,03 0,04 -0,01 0,00 0,02 0,04
1000 0,01 -0,01 -0,04 -0,04 0,03 0,02 0,02 0,08 0,00 -0,01 0,04
1250 -0,03 0,06 -0,06 -0,09 0,04 0,07 0,06 0,04 -0,03 0,01 0,06
1600 0,01 0,05 -0,04 -0,06 0,05 0,01 0,03 0,02 0,02 -0,02 0,04
2000 -0,01 0,03 -0,05 -0,04 0,03 0,03 0,03 0,06 -0,02 -0,02 0,04
2500 -0,02 0,08 -0,07 -0,05 0,02 0,06 0,00 -0,03 0,02 -0,03 0,05
3150 -0,02 0,08 -0,06 -0,06 0,01 0,08 -0,02 0,01 -0,02 -0,05 0,05

E2
Table A3. Ratio of area S over volume V differences about the average value for all operators, measurement case A
OP1 OP2 OP3a OP3b OP4 OP5 OP6 OP7 OP8 OP9 dS/V_stddev
All freq 0,03 0,02 -0,02 -0,02 0,02 0,02 -0,01 -0,02 -0,02 -0,02 0,02

Table A4. Ratio 10log(S/A) differences about the average value for all operators, measurement case A
Frequency OP1 OP2 OP3a OP3b OP4 OP5 OP6 OP7 OP8 OP9 d10lgS/A_A_stddev
50 2,43 0,97 -1,09 -0,54 1,32 -0,57 -0,02 0,03 -1,13 -1,41 1,23
63 0,20 0,68 -0,93 -0,48 1,34 -0,71 1,31 0,24 -1,10 -0,55 0,89
80 0,80 1,49 -0,37 -0,90 2,00 -0,38 -0,01 -0,99 -1,67 0,03 1,14
100 1,30 0,60 -0,42 -0,03 0,42 -0,32 0,08 -0,44 -0,66 -0,52 0,62
125 0,81 0,48 -0,37 -0,31 0,93 0,97 -0,02 -0,64 -1,09 -0,77 0,75
160 0,63 0,44 -0,48 -0,87 1,23 1,09 -0,25 -0,50 -0,66 -0,62 0,78
200 0,86 0,73 -0,18 -0,70 0,88 0,73 -0,12 -0,83 -0,64 -0,72 0,73
250 0,90 0,79 -0,41 -0,87 1,18 0,29 0,24 -1,00 -0,50 -0,63 0,79
315 1,05 0,50 -0,80 -0,61 0,93 0,47 -0,10 -0,36 -0,51 -0,58 0,68
400 1,14 0,68 -0,84 -0,56 1,15 0,18 0,04 -0,58 -0,64 -0,58 0,76
500 1,26 0,50 -0,99 -0,84 0,88 0,99 -0,33 -0,28 -0,42 -0,77 0,83
630 1,00 0,63 -0,89 -0,98 1,06 1,31 -0,24 -0,54 -0,55 -0,81 0,90
800 1,06 0,69 -0,68 -0,86 1,03 0,50 0,00 -0,65 -0,58 -0,52 0,75
1000 1,02 0,48 -0,76 -0,76 0,90 0,71 -0,09 -0,28 -0,57 -0,65 0,71
1250 0,85 0,76 -0,85 -0,98 0,94 0,91 0,07 -0,44 -0,70 -0,56 0,80
1600 1,02 0,72 -0,77 -0,86 0,98 0,67 -0,05 -0,52 -0,49 -0,69 0,77
2000 0,94 0,66 -0,81 -0,76 0,91 0,77 -0,03 -0,34 -0,66 -0,68 0,74
2500 0,92 0,90 -0,88 -0,78 0,90 0,93 -0,13 -0,71 -0,45 -0,71 0,81
3150 0,94 0,93 -0,83 -0,83 0,87 1,04 -0,21 -0,51 -0,62 -0,79 0,83

E3
Table A5. Sound insulation index R differences about the average value for all operators, measurement case A
Frequency OP1 OP2 OP3a OP3b OP4 OP5 OP6 OP7 OP8 OP9 dR_A_stddev
50 5,03 1,73 -3,17 -2,97 0,63 -6,97 2,33 1,93 -0,17 1,63 3,46
63 1,94 -0,76 1,64 -2,46 -0,76 -4,16 3,74 5,24 -3,96 -0,46 3,13
80 -1,81 0,59 0,79 -3,51 2,09 -1,11 2,19 1,29 -2,01 1,49 1,97
100 1,54 3,04 -3,86 -3,96 2,34 0,24 -0,86 2,64 1,24 -2,36 2,64
125 3,42 -0,08 -1,78 -2,98 0,42 1,32 -0,88 0,92 -0,48 0,12 1,75
160 1,54 -2,96 1,54 -1,36 -0,06 0,74 -2,16 2,54 0,14 0,04 1,73
200 0,19 1,29 -1,01 -0,61 -0,11 0,39 -1,61 2,09 0,89 -1,51 1,22
250 1,12 0,32 -1,78 -1,28 1,22 0,92 0,82 -0,48 0,12 -0,98 1,07
315 0,92 1,02 -0,78 -1,88 1,22 0,22 0,22 -0,18 -0,28 -0,48 0,94
400 2,22 2,12 -0,78 -1,18 1,62 -1,38 -0,98 0,12 -2,48 0,72 1,62
500 2,00 -0,20 -1,30 -0,90 1,30 0,60 -1,30 0,80 -0,40 -0,60 1,13
630 2,18 -0,42 -1,52 -0,12 0,28 1,08 -0,32 0,48 -0,32 -1,32 1,09
800 2,05 0,05 -1,35 -0,15 0,85 -0,05 -0,85 -0,05 0,15 -0,65 0,94
1000 2,25 -0,15 -1,45 -0,75 1,15 0,75 -1,25 0,15 -0,25 -0,45 1,13
1250 1,69 -0,21 -1,41 -0,21 0,59 1,29 -0,81 0,19 -0,51 -0,61 0,96
1600 1,95 -0,15 -1,05 -0,15 0,15 1,05 -0,45 0,25 -0,85 -0,75 0,92
2000 1,51 -0,59 -0,59 -0,19 0,51 0,61 0,21 0,01 -1,09 -0,39 0,75
2500 1,22 0,12 -0,68 -0,38 0,42 0,62 0,02 -0,58 0,42 -1,18 0,71
3150 -0,26 1,04 0,14 -0,06 0,84 0,84 0,14 -1,26 -0,36 -1,06 0,78

R’w 1,70 0,70 -1,30 -1,30 0,70 0,70 -0,30 0,70 -0,30 -1,30 1,06
R'w+C 1,70 -0,30 -1,30 -1,30 0,70 0,70 -1,30 1,70 -0,30 -0,30 1,16
R'w+Ctr 1,70 -0,30 -1,30 -2,30 0,70 0,70 -1,30 1,70 0,70 -0,30 1,34
R'w+C50- 1,00 -1,00 -1,00 -1,00 1,00 1,00 -1,00 1,00 0,00 0,00 0,94
3150
R'w+Ctr50- 0,90 -0,10 -1,10 -2,10 0,90 -0,10 -0,10 1,90 -0,10 -0,10 1,10
3150

E4
Table B1. Sound pressure level ∆L differences about the average value for all operators, measurement case B
Frequency OP2 OP4 OP5 OP8 OP9 dL_B_stddev
50 2,42 0,37 -6,80 1,30 2,70 3,91
63 0,22 -1,82 -1,27 -0,77 3,63 2,17
80 -2,71 0,38 0,77 0,67 0,87 1,52
100 -0,92 1,77 0,79 -0,71 -0,91 1,22
125 1,38 0,18 -0,12 0,18 -1,62 1,07
160 0,52 -1,20 0,10 0,00 0,60 0,72
200 -0,05 -1,54 0,36 -0,04 1,26 1,01
250 0,88 -0,54 -0,24 0,86 -0,94 0,83
315 1,18 0,58 -1,12 -0,22 -0,42 0,90
400 0,00 1,05 -0,35 -1,35 0,65 0,93
500 0,84 -0,56 0,14 -0,56 0,14 0,59
630 1,65 -0,86 0,14 0,24 -1,16 1,11
800 1,77 -0,39 -0,49 -0,39 -0,49 0,99
1000 1,96 -0,64 -0,34 -0,14 -0,84 1,13
1250 1,84 -0,13 -0,43 -0,23 -1,03 1,08
1600 1,39 -0,35 0,15 -0,95 -0,25 0,87
2000 1,31 -0,40 -0,20 -1,00 0,30 0,87
2500 1,72 -0,73 -0,73 -0,03 -0,23 1,01
3150 0,99 -0,80 -0,20 -0,40 0,40 0,70

E5
Table B2. Reverberation time T differences about the average value for all operators, measurement case B
Frequency OP2 OP4 OP5 OP8 OP9 dT_B_stddev
50 -0,03 0,01 -0,03 -0,06 0,09 0,06
63 0,07 0,04 -0,09 0,02 -0,03 0,06
80 0,13 0,04 -0,04 -0,08 -0,07 0,09
100 0,06 -0,02 -0,02 -0,01 -0,03 0,04
125 0,01 -0,07 0,10 -0,04 -0,02 0,07
160 -0,04 0,03 -0,08 0,10 -0,01 0,07
200 -0,04 0,05 -0,04 0,02 -0,01 0,04
250 -0,02 0,10 -0,05 -0,03 0,01 0,06
315 0,01 0,07 -0,04 0,02 -0,06 0,05
400 -0,03 0,04 0,02 0,00 -0,05 0,04
500 -0,03 0,05 -0,01 0,00 -0,02 0,03
630 -0,04 0,04 0,02 0,05 -0,08 0,06
800 -0,03 0,04 -0,01 0,05 -0,05 0,04
1000 -0,05 0,04 0,04 0,02 -0,03 0,04
1250 0,04 0,03 0,00 0,02 -0,08 0,05
1600 0,02 0,04 0,03 -0,03 -0,07 0,05
2000 0,03 0,03 0,00 -0,02 -0,06 0,04
2500 0,02 0,02 -0,02 0,02 -0,04 0,03
3150 0,03 0,04 0,02 -0,04 -0,04 0,04

E6
Table B3. Ratio of area S over volume V differences about the average value for all operators, measurement case B
OP2 OP4 OP5 OP8 OP9 dS/V_B_stddev
All freq -0,01 -0,02 0,05 -0,01 -0,01 0,03

Table B4. Ratio 10log(S/A) differences about the average value for all operators, measurement case B
Frequency OP2 OP4 OP5 OP8 OP9 d10lgS/A_B_stddev
50 -0,40 -0,35 0,89 -0,61 0,47 0,65
63 0,17 -0,23 0,49 -0,11 -0,32 0,33
80 0,55 -0,17 0,86 -0,70 -0,54 0,68
100 0,30 -0,59 0,93 -0,27 -0,36 0,61
125 -0,13 -0,87 1,67 -0,44 -0,22 0,98
160 -0,48 -0,25 0,53 0,38 -0,18 0,43
200 -0,44 -0,16 0,85 -0,09 -0,17 0,50
250 -0,33 0,01 0,80 -0,39 -0,09 0,48
315 -0,18 -0,15 0,88 -0,14 -0,42 0,50
400 -0,35 -0,25 1,17 -0,21 -0,36 0,66
500 -0,37 -0,21 1,02 -0,22 -0,22 0,58
630 -0,40 -0,26 1,17 0,01 -0,51 0,68
800 -0,36 -0,29 1,02 -0,02 -0,36 0,59
1000 -0,47 -0,30 1,22 -0,16 -0,28 0,69
1250 -0,07 -0,34 1,05 -0,15 -0,49 0,61
1600 -0,13 -0,27 1,20 -0,35 -0,45 0,68
2000 -0,08 -0,31 1,08 -0,30 -0,40 0,62
2500 -0,13 -0,36 0,96 -0,13 -0,33 0,55
3150 -0,08 -0,26 1,16 -0,45 -0,36 0,66

E7
Table B5. Sound insulation index R differences about the average value for all operators, measurement case B
Frequency OP2 OP4 OP5 OP8 OP9 dR_B_stddev
50 2,02 0,02 -5,88 0,72 3,12 3,50
63 0,42 -2,08 -0,78 -0,88 3,32 2,06
80 -2,16 0,24 1,64 -0,06 0,34 1,37
100 -0,66 1,14 1,74 -0,96 -1,26 1,35
125 1,24 -0,66 1,54 -0,26 -1,86 1,40
160 0,02 -1,48 0,62 0,42 0,42 0,86
200 -0,48 -1,68 1,22 -0,18 1,12 1,21
250 0,56 -0,54 0,56 0,46 -1,04 0,74
315 1,04 0,44 -0,26 -0,36 -0,86 0,74
400 -0,36 0,84 0,84 -1,56 0,24 1,00
500 0,52 -0,78 1,12 -0,78 -0,08 0,83
630 1,24 -1,16 1,34 0,24 -1,66 1,37
800 1,42 -0,68 0,52 -0,38 -0,88 0,96
1000 1,48 -0,92 0,88 -0,32 -1,12 1,14
1250 1,74 -0,46 0,64 -0,36 -1,56 1,25
1600 1,22 -0,58 1,32 -1,28 -0,68 1,19
2000 1,20 -0,70 0,90 -1,30 -0,10 1,05
2500 1,62 -1,08 0,22 -0,18 -0,58 1,03
3150 0,94 -1,06 0,94 -0,86 0,04 0,95

R’w 0,60 -0,40 0,60 -0,40 -0,40 0,55

R'w+C 0,80 -1,20 0,80 -0,20 -0,20 0,84
R'w+Ctr 0,20 -0,80 1,20 0,20 -0,80 0,84
R'w+C50- 0,00 -1,00 1,00 0,00 0,00 0,71
3150
R'w+Ctr50- 0,00 -1,00 1,00 0,00 0,00 0,71
3150

E8
Table C1. Sound pressure level ∆L differences about the average value for all operators, measurement case C
Frequency OP2 OP6 OP7 OP9 dL_C_stddev
50 -2,99 3,80 -0,60 -0,20 2,82
63 -3,69 3,23 -2,07 2,53 3,40
80 -2,81 1,17 -1,43 3,07 2,63
100 -0,07 -0,01 -1,61 1,69 1,35
125 -0,92 -0,33 0,67 0,57 0,76
160 -0,60 -1,57 1,33 0,83 1,33
200 0,09 -1,03 1,47 -0,53 1,08
250 -0,42 0,27 -0,13 0,27 0,34
315 0,77 -1,26 -0,46 0,94 1,04
400 0,02 0,63 -0,37 -0,27 0,45
500 -0,48 -0,04 0,06 0,46 0,39
630 -0,34 0,48 -0,32 0,18 0,40
800 0,30 0,10 -0,50 0,10 0,35
1000 0,24 -0,28 -0,28 0,32 0,32
1250 0,29 0,27 -0,43 -0,13 0,35
1600 0,14 0,22 -0,38 0,02 0,27
2000 0,22 0,19 -0,51 0,09 0,34
2500 0,65 -0,12 -0,62 0,08 0,52
3150 -0,07 0,96 -0,84 -0,04 0,74

E9
Table C2. Reverberation time T differences about the average value for all operators, measurement case C
Frequency OP2 OP6 OP7 OP9 dT_C_stddev
50 0,07 -0,11 0,03 -0,10 0,09
63 0,05 -0,15 -0,04 0,00 0,09
80 0,15 -0,01 -0,07 -0,07 0,10
100 0,04 0,11 -0,04 0,01 0,06
125 0,06 -0,08 0,00 -0,07 0,07
160 0,05 0,00 -0,03 -0,02 0,04
200 0,00 0,01 0,01 0,00 0,01
250 0,02 0,02 -0,01 0,00 0,02
315 -0,01 -0,06 0,00 0,00 0,03
400 -0,01 -0,02 0,04 -0,02 0,03
500 -0,03 0,01 0,04 -0,02 0,03
630 0,02 0,03 -0,02 0,00 0,02
800 -0,01 0,03 0,01 0,01 0,02
1000 -0,02 0,01 0,02 0,00 0,02
1250 0,00 0,01 0,02 -0,01 0,01
1600 0,03 0,00 0,00 -0,02 0,02
2000 -0,01 0,02 0,02 -0,01 0,02
2500 -0,03 0,01 0,03 -0,01 0,03
3150 0,01 0,01 0,02 -0,02 0,02

E10
Table C3. Ratio of area S over volume V differences about the average value for all operators, measurement case C
OP2 OP6 OP7 OP9 dS/V_C_stddev
All 0,02 0,03 -0,03 -0,03 0,03
freq

Table C4. Ratio 10log(S/A) differences about the average value for all operators, measurement case C
Frequency OP2 OP6 OP7 OP9 d10lgS/A_C_stddev
50 0,88 -0,08 -0,01 -0,79 0,68
63 0,96 -0,55 -0,37 -0,05 0,67
80 1,46 0,39 -0,93 -0,92 1,16
100 0,42 1,11 -1,02 -0,51 0,95
125 1,19 -0,24 -0,08 -0,87 0,86
160 0,83 0,43 -0,69 -0,56 0,74
200 0,25 0,47 -0,31 -0,41 0,43
250 0,43 0,55 -0,55 -0,43 0,57
315 0,39 -0,04 -0,18 -0,17 0,27
400 0,23 0,24 0,06 -0,53 0,36
500 0,00 0,52 0,02 -0,55 0,44
630 0,43 0,64 -0,64 -0,42 0,63
800 0,10 0,61 -0,36 -0,35 0,46
1000 0,09 0,49 -0,20 -0,37 0,38
1250 0,26 0,46 -0,23 -0,49 0,44
1600 0,55 0,40 -0,39 -0,56 0,56
2000 0,17 0,55 -0,24 -0,48 0,45
2500 0,04 0,51 -0,11 -0,44 0,39
3150 0,35 0,46 -0,24 -0,57 0,49

E11
Table C5. Sound insulation index R differences about the average value for all operators, measurement case C
Frequency OP2 OP6 OP7 OP9 dR_C_stddev
50 -2,10 3,70 -0,60 -1,00 2,55
63 -2,73 2,68 -2,43 2,48 2,98
80 -1,35 1,55 -2,35 2,15 2,19
100 0,38 1,08 -2,63 1,18 1,79
125 0,30 -0,60 0,60 -0,30 0,55
160 0,17 -1,13 0,67 0,27 0,78
200 0,35 -0,55 1,15 -0,95 0,94
250 0,03 0,83 -0,67 -0,17 0,62
315 1,15 -1,25 -0,65 0,75 1,14
400 0,28 0,88 -0,32 -0,82 0,74
500 -0,50 0,50 0,10 -0,10 0,42
630 0,13 1,13 -0,98 -0,27 0,88
800 0,42 0,73 -0,88 -0,27 0,72
1000 0,33 0,23 -0,47 -0,07 0,36
1250 0,55 0,75 -0,65 -0,65 0,75
1600 0,73 0,63 -0,77 -0,58 0,78
2000 0,40 0,70 -0,70 -0,40 0,66
2500 0,68 0,38 -0,72 -0,32 0,64
3150 0,30 1,40 -1,10 -0,60 1,10

R’w 0,00 0,00 0,00 0,00 0,00

R'w+C 0,50 -0,50 0,50 -0,50 0,58
R'w+Ctr 0,00 0,00 0,00 0,00 0,00
R'w+C50- 0,75 -0,25 -0,25 -0,25 0,50
3150
R'w+Ctr50- 0,50 -0,50 -0,50 0,50 0,58
3150

E12
Table D1. Sound pressure level ∆L differences about the average value for all operators, measurement case D
Frequency OP2 OP3 OP4 OP6 OP7 OP9 dL_D_stddev
50 2,86 -1,82 -0,77 1,98 -1,92 -0,32 1,99
63 -4,16 0,43 3,94 1,53 3,03 -4,77 3,67
80 -3,94 -0,34 6,51 0,16 2,46 -4,84 4,18
100 -0,11 -1,40 5,10 1,00 -3,10 -1,50 2,86
125 0,11 -1,72 -1,52 2,98 -0,62 0,78 1,74
160 -3,80 1,20 -0,60 0,50 0,10 2,60 2,16
200 -3,88 1,49 -0,58 1,19 -1,11 2,89 2,39
250 -1,48 0,54 -1,58 2,64 -0,66 0,54 1,59
315 0,27 -0,98 0,74 -0,38 -0,18 0,52 0,64
400 -1,06 -0,06 -1,81 1,44 -0,86 2,34 1,59
500 -2,29 -1,46 1,94 0,64 -1,66 2,84 2,12
630 -2,64 -0,22 1,02 1,88 -1,42 1,38 1,76
800 -1,55 -1,35 1,46 1,85 0,05 -0,45 1,41
1000 -2,78 -2,22 2,08 3,58 0,28 -0,92 2,48
1250 -3,80 0,05 0,98 2,05 0,25 0,45 1,99
1600 -1,85 0,05 0,55 1,65 0,05 -0,45 1,15
2000 0,19 -0,30 0,70 0,40 -0,50 -0,50 0,51
2500 -0,97 -0,03 0,17 0,77 -0,13 0,17 0,57
3150 0,26 -0,09 0,21 0,11 0,11 -0,59 0,31

E13
Table D2. Reverberation time T differences about the average value for all operators, measurement case D
Frequency OP2 OP3 OP4 OP6 OP7 OP9 dT_D_stddev
50 0,02 0,12 0,02 -0,16 -0,02 -0,15 0,11
63 0,05 -0,07 0,05 -0,15 -0,04 0,00 0,08
80 0,14 -0,04 0,04 -0,02 -0,08 -0,08 0,08
100 0,02 0,03 0,04 0,09 -0,06 -0,01 0,05
125 0,05 0,02 0,02 -0,09 -0,01 -0,08 0,06
160 0,07 -0,06 0,00 0,02 -0,01 0,00 0,04
200 -0,01 0,02 0,01 0,00 0,00 -0,01 0,01
250 0,02 -0,01 0,02 0,02 -0,01 0,00 0,02
315 -0,02 -0,03 0,05 -0,07 -0,01 -0,01 0,04
400 -0,01 -0,02 -0,01 -0,02 0,04 -0,02 0,02
500 -0,03 -0,01 0,01 0,01 0,04 -0,02 0,03
630 0,01 -0,01 0,02 0,02 -0,03 -0,01 0,02
800 -0,02 -0,01 0,03 0,02 0,00 0,00 0,02
1000 -0,02 -0,04 0,02 0,01 0,02 0,00 0,02
1250 0,00 -0,04 0,03 0,01 0,02 -0,01 0,02
1600 0,03 -0,04 0,02 0,00 0,00 -0,02 0,03
2000 0,00 -0,05 0,03 0,03 0,03 0,00 0,03
2500 -0,02 -0,04 0,01 0,02 0,04 0,00 0,03
3150 0,01 -0,04 0,02 0,01 0,02 -0,02 0,02

E14
Table D3. Ratio of area S over volume V differences about the average value for all operators, measurement case D
OP2 OP3 OP4 OP6 OP7 OP9 dS/V_D_stddev
All freq 0,15 -0,01 -0,04 -0,04 -0,01 -0,04 0,07

Table D4. Ratio 10log(S/A) differences about the average value for all operators, measurement case D
Frequency OP2_DnT OP3 OP4 OP6 OP7 OP9 d10lgSA_D_stddev_OP2_21m2* d10lgSA_D_stddev_exOP2
50 3,17 0,81 -0,68 -1,72 0,08 -1,66 1,84 1,10
63 3,46 -0,30 -0,39 -1,98 -0,07 -0,72 1,82 0,76
80 3,99 -0,23 -0,57 -1,01 -0,61 -1,57 2,01 0,51
100 2,90 0,14 -0,77 -0,33 -0,74 -1,19 1,49 0,50
125 3,55 0,41 -0,58 -1,80 0,09 -1,68 1,96 1,00
160 3,59 -0,75 -1,00 -0,74 -0,13 -0,97 1,78 0,35
200 2,74 0,21 -0,89 -0,96 -0,02 -1,08 1,44 0,60
250 3,01 -0,16 -0,84 -0,81 -0,17 -1,02 1,52 0,40
315 2,83 -0,13 -0,36 -1,52 0,06 -0,89 1,50 0,64
400 2,84 -0,12 -1,01 -1,08 0,47 -1,09 1,53 0,71
500 2,57 -0,08 -0,88 -0,84 0,38 -1,15 1,38 0,64
630 2,97 -0,08 -0,78 -0,75 -0,30 -1,05 1,50 0,39
800 2,64 -0,12 -0,72 -0,78 -0,03 -0,98 1,35 0,43
1000 2,70 -0,36 -0,77 -0,83 0,21 -0,93 1,39 0,47
1250 2,86 -0,39 -0,72 -0,86 0,17 -1,06 1,47 0,48
1600 3,16 -0,35 -0,78 -0,93 0,02 -1,12 1,60 0,46
2000 2,82 -0,52 -0,77 -0,73 0,21 -1,01 1,44 0,47
2500 2,68 -0,37 -0,90 -0,78 0,33 -0,96 1,40 0,54
3150 2,96 -0,34 -0,81 -0,86 0,17 -1,13 1,52 0,51

2
* OP 2 calculated DnT instead of R, i.e. they used V/3.1=21 m instead of V/7,5 or 10 m2.

E15
Table D5. Sound insulation index R differences about the average value for all operators, measurement case D
Frequency OP2 OP3 OP4 OP6 OP7 OP9 dR_D_stddev
50 6,05 -1,05 -1,45 0,25 -1,85 -1,95 3,07
63 -0,73 0,17 3,57 -0,43 2,97 -5,53 3,25
80 0,03 -0,57 5,93 -0,87 1,83 -6,37 4,00
100 2,75 -1,25 4,35 0,65 -3,85 -2,65 3,17
125 3,68 -1,32 -2,12 1,18 -0,52 -0,92 2,11
160 -0,20 0,40 -1,60 -0,20 0,00 1,60 1,04
200 -1,10 1,70 -1,50 0,20 -1,10 1,80 1,47
250 1,55 0,35 -2,45 1,85 -0,85 -0,45 1,60
315 3,08 -1,12 0,38 -1,92 -0,12 -0,32 1,71
400 1,80 -0,20 -2,80 0,40 -0,40 1,20 1,60
500 0,23 -1,57 1,03 -0,17 -1,27 1,73 1,28
630 0,35 -0,35 0,25 1,15 -1,75 0,35 0,98
800 1,10 -1,50 0,70 1,10 0,00 -1,40 1,19
1000 -0,12 -2,62 1,28 2,78 0,48 -1,82 1,99
1250 -0,97 -0,37 0,23 1,23 0,43 -0,57 0,79
1600 1,27 -0,33 -0,23 0,77 0,07 -1,53 0,97
2000 2,97 -0,83 -0,03 -0,33 -0,23 -1,53 1,55
2500 1,70 -0,40 -0,70 0,00 0,20 -0,80 0,92
3150 3,25 -0,45 -0,55 -0,75 0,25 -1,75 1,72

R’w 2,33 -1,67 0,33 0,33 -0,67 -0,67 1,37

R'w+C 2,33 -0,67 0,33 0,33 -1,67 -0,67 1,37
R'w+Ctr 3,17 -0,83 1,17 0,17 -2,83 -0,83 2,04
R'w+C50- 2,17 -0,83 2,17 0,17 0,17 -3,83 2,23
3150
R'w+Ctr50- 1,50 -0,50 4,50 -0,50 0,50 -5,50 3,27
3150

E16
Table E1. Sound pressure level ∆L differences about the average value for all operators, measurement case E
Frequency OP2 OP4 OP5 OP7 OP9 dL_E_stddev
50 1,95 -2,29 4,31 -0,79 -3,19 3,10
63 3,35 -1,99 2,31 0,31 -3,99 3,02
80 5,47 -0,14 4,92 -2,78 -7,48 5,43
100 5,72 -0,71 2,76 -3,24 -4,54 4,24
125 5,22 0,48 2,13 -3,47 -4,37 3,97
160 3,25 -0,64 3,30 -2,90 -3,00 3,13
200 2,80 0,88 3,88 -0,92 -6,62 4,13
250 1,85 0,14 2,34 0,14 -4,46 2,69
315 1,36 -0,41 1,79 -0,31 -2,41 1,67
400 1,73 -0,93 2,97 -0,03 -3,73 2,58
500 1,00 0,57 2,47 -0,93 -3,13 2,13
630 1,60 -0,43 2,47 -1,73 -1,93 1,97
800 1,65 -0,61 2,69 -2,11 -1,61 2,09
1000 2,62 -1,78 3,12 -2,78 -1,18 2,69
1250 2,31 -2,03 3,17 -1,73 -1,73 2,52
1600 1,41 -0,58 2,82 -1,18 -2,48 2,11
2000 1,86 -0,42 2,38 -1,72 -2,12 2,05
2500 2,38 -1,02 1,98 -1,52 -1,82 2,01
3150 1,89 -0,12 2,48 -1,72 -2,52 2,18

E17
Table E2. Reverberation time T differences about the average value for all operators, measurement case E
Frequency OP2 OP4 OP5 OP7 OP9 dT_E_stddev
50 0,05 0,05 -0,01 0,01 -0,12 0,07
63 0,07 0,07 -0,16 -0,02 0,02 0,10
80 0,16 0,06 -0,09 -0,06 -0,06 0,11
100 0,04 0,06 -0,06 -0,04 0,01 0,05
125 0,04 0,01 0,05 -0,02 -0,09 0,06
160 0,08 0,01 -0,09 0,00 0,01 0,06
200 0,01 0,03 -0,07 0,02 0,01 0,04
250 0,01 0,01 0,00 -0,02 -0,01 0,01
315 -0,01 0,06 -0,06 0,00 0,00 0,04
400 -0,01 -0,01 0,00 0,04 -0,02 0,02
500 -0,04 0,00 0,04 0,03 -0,03 0,04
630 0,00 0,01 0,04 -0,04 -0,02 0,03
800 -0,03 0,02 0,05 -0,01 -0,01 0,03
1000 -0,03 0,01 0,03 0,01 -0,01 0,02
1250 -0,01 0,02 -0,01 0,01 -0,02 0,02
1600 0,02 0,01 0,01 -0,01 -0,03 0,02
2000 -0,02 0,01 0,01 0,01 -0,02 0,02
2500 -0,03 0,00 0,00 0,03 -0,01 0,02
3150 0,00 0,01 0,01 0,01 -0,03 0,02

E18
Table E3. Ratio of area S over volume V differences about the average value for all operators, measurement case E
OP2 OP4 OP5 OP7 OP9 dSA_E_stddev
All freq -0,01 0,00 0,01 0,01 0,00 0,01

Table E4. Ratio 10log(S/A) differences about the average value for all operators, measurement case E
Frequency OP2 OP4 OP5 OP7 OP9 d10lgSA_E_stddev
50 0,00 0,31 0,18 0,23 -0,72 0,42
63 0,27 0,59 -1,14 0,07 0,21 0,67
80 0,92 0,54 -0,59 -0,35 -0,52 0,69
100 0,06 0,55 -0,42 -0,26 0,07 0,37
125 0,12 0,16 0,73 -0,02 -0,99 0,62
160 0,56 0,14 -0,98 0,16 0,12 0,58
200 -0,18 0,36 -0,69 0,39 0,11 0,45
250 -0,19 0,13 0,22 -0,05 -0,11 0,17
315 -0,38 0,59 -0,40 0,17 0,01 0,41
400 -0,42 -0,10 0,21 0,54 -0,23 0,38
500 -0,71 0,01 0,58 0,43 -0,32 0,53
630 -0,29 0,13 0,60 -0,24 -0,20 0,37
800 -0,65 0,16 0,63 0,01 -0,15 0,46
1000 -0,62 0,08 0,46 0,22 -0,13 0,41
1250 -0,39 0,20 0,13 0,25 -0,19 0,28
1600 -0,14 0,10 0,29 0,05 -0,30 0,23
2000 -0,47 0,11 0,31 0,24 -0,18 0,33
2500 -0,56 0,02 0,22 0,41 -0,09 0,37
3150 -0,31 0,09 0,29 0,22 -0,29 0,28

E19
Table E5. Sound insulation index R differences about the average value for all operators, measurement case E
Frequency OP2 OP4 OP5 OP7 OP9 dR_E_stddev
50 1,92 -1,98 4,52 -0,58 -3,88 3,29
63 3,66 -1,44 1,16 0,36 -3,74 2,78
80 6,42 0,42 4,32 -3,18 -7,98 5,78
100 5,76 -0,14 2,36 -3,54 -4,44 4,21
125 5,36 0,66 2,86 -3,54 -5,34 4,43
160 3,80 -0,50 2,30 -2,70 -2,90 2,99
200 2,66 1,26 3,16 -0,54 -6,54 3,93
250 1,66 0,26 2,56 0,06 -4,54 2,74
315 1,00 0,20 1,40 -0,20 -2,40 1,48
400 1,28 -1,02 3,18 0,48 -3,92 2,66
500 0,28 0,58 3,08 -0,52 -3,42 2,34
630 1,30 -0,30 3,10 -2,00 -2,10 2,22
800 1,04 -0,46 3,34 -2,16 -1,76 2,25
1000 2,00 -1,70 3,60 -2,60 -1,30 2,66
1250 1,90 -1,80 3,30 -1,50 -1,90 2,43
1600 1,24 -0,46 3,14 -1,16 -2,76 2,27
2000 1,40 -0,30 2,70 -1,50 -2,30 2,05
2500 1,82 -0,98 2,22 -1,08 -1,98 1,89
3150 1,54 -0,06 2,74 -1,46 -2,76 2,22

R’w 1,60 -0,40 2,60 -1,40 -2,40 2,07

R'w+C 1,40 -0,60 3,40 -1,60 -2,60 2,41
R'w+Ctr 1,80 -0,20 2,80 -1,20 -3,20 2,39
R'w+C50- 1,40 -0,60 3,40 -1,60 -2,60 2,41
3150
R'w+Ctr50- 2,40 -0,60 3,40 -1,60 -3,60 2,88
3150

E20
Table F1. Sound pressure level ∆L differences about the average value for all operators, measurement case F
Frequency OP3 OP4 OP6 OP8 OP9 dL_F_stddev
50 0,88 2,18 -1,02 -0,72 -1,32 1,49
63 -0,08 -1,78 1,32 -1,08 1,62 1,48
80 0,00 -0,80 0,50 -0,70 1,00 0,77
100 0,28 -2,12 0,78 -0,42 1,48 1,37
125 0,96 -1,24 1,16 -1,94 1,06 1,47
160 2,48 -0,52 -2,12 0,98 -0,82 1,77
200 0,58 -2,82 1,18 0,38 0,68 1,60
250 0,98 -1,12 0,18 0,18 -0,22 0,76
315 2,20 -1,00 -1,30 0,50 -0,40 1,41
400 2,06 -0,54 -0,44 0,96 -2,04 1,57
500 0,40 -0,90 0,30 1,20 -1,00 0,94
630 0,94 -0,36 0,44 0,24 -1,26 0,84
800 0,62 -0,18 -0,08 0,62 -0,98 0,66
1000 0,76 -0,54 0,26 0,06 -0,54 0,55
1250 1,10 -0,20 -0,40 0,00 -0,50 0,64
1600 0,56 -0,44 -0,74 0,16 0,46 0,57
2000 0,92 -0,38 -0,18 -0,08 -0,28 0,53
2500 0,74 -0,56 -0,76 0,34 0,24 0,63
3150 0,40 -0,40 0,20 -0,20 0,00 0,32

E21
Table F2. Reverberation time T differences about the average value for all operators, measurement case F
Frequency OP3 OP4 OP6 OP8 OP9 dT_F_stddev
50 -0,10 0,32 0,36 -0,12 -0,09 0,24
63 0,09 0,02 0,17 -0,07 -0,03 0,10
80 -0,07 0,03 -0,05 0,08 -0,05 0,06
100 0,05 0,02 0,04 0,06 -0,12 0,07
125 -0,06 -0,03 0,10 0,04 0,04 0,06
160 0,01 0,12 -0,12 0,04 -0,16 0,12
200 -0,11 0,12 -0,11 -0,01 0,01 0,10
250 0,04 -0,03 0,04 -0,01 0,02 0,03
315 -0,10 0,11 -0,07 0,05 -0,07 0,09
400 -0,07 0,10 -0,03 0,03 -0,07 0,07
500 -0,06 0,11 -0,08 0,06 -0,11 0,10
630 -0,07 0,13 -0,01 -0,02 -0,05 0,08
800 -0,05 0,08 0,03 0,00 -0,02 0,05
1000 -0,09 0,17 0,06 -0,03 -0,05 0,10
1250 -0,10 0,20 0,12 -0,04 -0,05 0,13
1600 -0,12 0,19 0,12 -0,02 -0,04 0,13
2000 -0,13 0,20 0,09 -0,08 0,00 0,13
2500 -0,11 0,14 0,09 -0,03 0,00 0,10
3150 -0,06 0,16 0,09 -0,07 -0,04 0,10

E22
Table F3. Ratio of area S over volume V differences about the average value for all operators, measurement case F
OP3 OP4 OP6 OP8 OP9 dS/V_F_stddev
All freq 0,00 0,00 0,00 0,00 0,00 0,00

Table F4. Ratio 10log(S/A) differences about the average value for all operators, measurement case F
Frequency OP3 OP4 OP6 OP8 OP9 d10lgSA_F_stddev
50 -0,60 0,85 0,97 -0,69 -0,53 0,83
63 0,20 -0,02 0,57 -0,48 -0,26 0,41
80 -0,38 0,29 -0,18 0,49 -0,22 0,37
100 0,17 0,08 0,19 0,20 -0,65 0,37
125 -0,28 -0,12 0,30 0,03 0,07 0,22
160 0,11 0,56 -0,33 0,20 -0,54 0,44
200 -0,30 0,45 -0,23 -0,01 0,10 0,30
250 0,05 -0,09 0,12 -0,11 0,02 0,10
315 -0,27 0,38 -0,11 0,14 -0,15 0,26
400 -0,20 0,32 -0,01 0,06 -0,17 0,21
500 -0,14 0,36 -0,12 0,16 -0,25 0,25
630 -0,22 0,40 0,03 -0,09 -0,13 0,24
800 -0,20 0,24 0,11 -0,07 -0,08 0,17
1000 -0,31 0,46 0,18 -0,16 -0,17 0,31
1250 -0,36 0,48 0,29 -0,21 -0,20 0,36
1600 -0,40 0,43 0,28 -0,15 -0,16 0,35
2000 -0,41 0,49 0,24 -0,29 -0,03 0,37
2500 -0,41 0,38 0,25 -0,18 -0,05 0,32
3150 -0,26 0,46 0,27 -0,31 -0,17 0,34

E23
Table F5. Sound insulation index R differences about the average value for all operators, measurement case F
Frequency OP3 OP4 OP6 OP8 OP9 dR_F_stddev
50 0,26 3,06 -0,04 -1,44 -1,84 1,93
63 0,16 -1,84 1,86 -1,54 1,36 1,67
80 -0,40 -0,50 0,30 -0,20 0,80 0,54
100 0,48 -2,02 0,98 -0,22 0,78 1,22
125 0,72 -1,38 1,42 -1,88 1,12 1,52
160 2,58 0,08 -2,42 1,18 -1,42 2,00
200 0,30 -2,40 0,90 0,40 0,80 1,37
250 1,08 -1,22 0,28 0,08 -0,22 0,83
315 1,92 -0,58 -1,38 0,62 -0,58 1,29
400 1,86 -0,24 -0,44 1,06 -2,24 1,57
500 0,26 -0,54 0,16 1,36 -1,24 0,97
630 0,68 0,08 0,48 0,18 -1,42 0,83
800 0,44 0,04 0,04 0,54 -1,06 0,63
1000 0,48 -0,12 0,48 -0,12 -0,72 0,50
1250 0,78 0,28 -0,12 -0,22 -0,72 0,56
1600 0,16 -0,04 -0,44 0,06 0,26 0,27
2000 0,52 0,12 0,02 -0,38 -0,28 0,36
2500 0,36 -0,14 -0,54 0,16 0,16 0,35
3150 0,18 0,08 0,48 -0,52 -0,22 0,38

R’w 1,00 0,00 0,00 0,00 -1,00 0,71

R'w+C 0,60 -0,40 -0,40 0,60 -0,40 0,55
R'w+Ctr 1,20 -0,80 0,20 0,20 -0,80 0,84
R'w+C50- 0,60 -0,40 -0,40 0,60 -0,40 0,55
3150
R'w+Ctr50- 0,80 -0,20 -0,20 -0,20 -0,20 0,45
3150

E24
Table G1. Sound pressure level ∆L differences about the average value for all operators, measurement case G, and all cases
Frequency OP3 OP4 OP8 dL_G_stddev ∆L_av_all δL_stddev_allcases
50 3,07 -1,53 -1,53 2,66 0,37 2,58
63 1,53 -0,17 -1,37 1,46 0,38 2,57
80 1,07 -0,83 -0,23 0,97 0,37 2,74
100 3,17 0,07 -3,23 3,20 0,31 2,37
125 2,43 1,53 -3,97 3,46 0,24 1,93
160 1,17 -1,23 0,07 1,20 0,25 1,80
200 0,40 -0,60 0,20 0,53 0,26 1,92
250 -0,03 -0,53 0,57 0,55 0,16 1,20
315 0,00 0,10 -0,10 0,10 0,14 0,92
400 0,30 -0,50 0,20 0,44 0,19 1,35
500 -1,10 -0,20 1,30 1,21 0,15 1,20
630 -0,27 -0,27 0,53 0,46 0,16 1,12
800 -0,50 -0,80 1,30 1,14 0,14 1,06
1000 -0,20 -0,70 0,90 0,82 0,18 1,41
1250 -0,23 -0,33 0,57 0,49 0,17 1,26
1600 -0,57 -0,27 0,83 0,74 0,13 0,97
2000 -0,37 -0,67 1,03 0,91 0,11 0,87
2500 -0,13 -0,73 0,87 0,81 0,12 0,88
3150 -0,13 -0,53 0,67 0,61 0,11 0,89

E25
Table G2. Reverberation time T differences about the average value for all operators, measurement case G, and all cases
Frequency OP3 OP4 OP8 dT_G_stddev T_av_all T_all_stddev
50 -0,17 0,16 0,01 0,17 0,02 0,13
63 -0,06 0,02 0,05 0,06 0,02 0,10
80 0,04 -0,06 0,03 0,06 0,01 0,09
100 -0,03 0,07 -0,04 0,06 0,01 0,05
125 -0,01 -0,03 0,03 0,03 0,01 0,05
160 0,00 0,05 -0,05 0,05 0,01 0,06
200 -0,02 0,05 -0,02 0,04 0,00 0,04
250 -0,08 0,01 0,07 0,08 0,01 0,04
315 0,03 0,03 -0,05 0,05 0,00 0,05
400 0,06 0,00 -0,05 0,06 0,00 0,04
500 0,06 0,04 -0,10 0,09 0,00 0,05
630 0,09 0,06 -0,15 0,13 0,01 0,06
800 0,05 0,04 -0,10 0,08 0,01 0,04
1000 0,09 0,03 -0,13 0,11 0,01 0,05
1250 0,11 0,05 -0,16 0,14 0,01 0,06
1600 0,14 0,05 -0,18 0,17 0,01 0,06
2000 0,12 0,05 -0,18 0,16 0,01 0,06
2500 0,09 0,05 -0,15 0,13 0,01 0,05
3150 0,11 0,06 -0,16 0,14 0,01 0,06

E26
Table G3. Ratio of area S over volume V differences about the average value for all operators, measurement case G, and all cases
OP3 OP4 OP8 dSA_G_stddev dS/V_all_stddev
All freq 0,00 0,00 0,00 0,00 0,03

Table G4. Ratio 10log(S/A) differences about the average value for all operators, measurement case G, and all cases
Frequency OP3 OP4 OP8 d10lgSA_G_stddev 10lgS/A_av_all 10lgS/A_all_stddev
50 -0,77 0,72 0,05 0,74 0,15 1,01
63 -0,36 0,14 0,22 0,32 0,12 0,87
80 0,24 -0,40 0,16 0,35 0,15 1,04
100 -0,22 0,52 -0,31 0,46 0,11 0,74
125 -0,05 -0,11 0,16 0,14 0,14 0,94
160 -0,01 0,37 -0,37 0,37 0,11 0,84
200 -0,14 0,29 -0,16 0,26 0,10 0,69
250 -0,41 0,10 0,30 0,37 0,09 0,72
315 0,11 0,17 -0,29 0,25 0,09 0,69
400 0,22 0,04 -0,25 0,24 0,10 0,73
500 0,25 0,22 -0,47 0,41 0,10 0,72
630 0,41 0,34 -0,75 0,65 0,10 0,78
800 0,25 0,26 -0,51 0,45 0,09 0,68
1000 0,44 0,23 -0,66 0,58 0,10 0,70
1250 0,53 0,32 -0,84 0,74 0,10 0,74
1600 0,66 0,32 -0,98 0,86 0,10 0,78
2000 0,62 0,36 -0,97 0,85 0,10 0,73
2500 0,51 0,37 -0,89 0,77 0,10 0,72
3150 0,60 0,41 -1,01 0,88 0,10 0,77

E27
Table G5. Sound insulation index R differences about the average value for all operators, measurement case G, and all cases
Frequency OP3 OP4 OP8 dR_G_stddev R_av_all R_dev_all
50 2,30 -0,80 -1,50 2,02 0,40 2,80
63 1,17 -0,03 -1,13 1,15 0,36 2,49
80 1,33 -1,27 -0,07 1,30 0,36 2,77
100 2,93 0,63 -3,57 3,30 0,33 2,48
125 2,37 1,47 -3,83 3,35 0,27 2,14
160 1,13 -0,87 -0,27 1,03 0,21 1,58
200 0,23 -0,27 0,03 0,25 0,23 1,69
250 -0,43 -0,43 0,87 0,75 0,17 1,29
315 0,10 0,30 -0,40 0,36 0,17 1,11
400 0,53 -0,47 -0,07 0,50 0,21 1,49
500 -0,80 0,00 0,80 0,80 0,16 1,17
630 0,17 0,07 -0,23 0,21 0,17 1,15
800 -0,23 -0,53 0,77 0,68 0,15 1,09
1000 0,23 -0,47 0,23 0,40 0,18 1,36
1250 0,30 0,00 -0,30 0,30 0,15 1,11
1600 0,07 0,07 -0,13 0,12 0,15 1,07
2000 0,27 -0,33 0,07 0,31 0,15 1,06
2500 0,40 -0,40 0,00 0,40 0,13 0,90
3150 0,47 -0,13 -0,33 0,42 0,16 1,16

R’w 0,00 0,00 0,00 0,00 0,13 1,05

R'w+C 0,67 -0,33 -0,33 0,58 0,16 1,18
R'w+Ctr 0,33 0,33 -0,67 0,58 0,17 1,35
R'w+C50- 0,00 0,00 0,00 0,00 0,17 1,29
3150
R'w+Ctr50- 1,33 -0,67 -0,67 1,15 0,20 1,69
3150

E28
Table A6-Ln. Normalised impact sound pressure level Ln differences about the average value for all operators, measurement case A
Frequency OP2 OP3 OP4 OP5 OP6 OP7 OP8 OP9 dLn_stddev_A
50 -0,39 0,41 0,61 -1,59 1,2125 -0,8875 2,1125 -1,4875 1,317397
63 2,29 0,89 0,89 -0,21 -1,8125 -0,7125 0,8875 -2,2125 1,527311
80 -0,55 3,25 -0,65 -3,05 1,35 -1,45 3,15 -2,05 2,349468
100 -0,74 2,76 1,76 -2,94 0,8625 -1,7375 1,4625 -1,4375 1,995665
125 -1,05 0,95 1,75 -2,35 1,95 -1,05 1,05 -1,25 1,610679
160 1,06 1,16 2,26 -2,14 2,5625 -2,4375 0,0625 -2,5375 2,108444
200 0,14 2,34 0,34 -1,76 0,7375 -0,2625 -1,0625 -0,4625 1,233969
250 -0,90 0,00 0,40 -0,60 0,3 0,6 0,3 -0,1 0,518239
315 -0,27 -0,67 0,73 -0,57 0,725 0,025 -0,475 0,525 0,587367
400 -0,14 -0,64 0,26 -0,34 0,7625 -0,0375 -0,3375 0,4625 0,465794
500 0,07 -0,13 -0,03 -0,73 0,675 -0,325 -0,225 0,675 0,480327
630 0,30 0,20 -0,30 -0,60 1 -0,4 -0,4 0,2 0,526444
800 -0,04 -0,24 -0,44 -0,04 0,8625 0,0625 -0,1375 -0,0375 0,381491
1000 0,05 -0,05 -0,55 -0,25 0,65 0,15 0,15 -0,15 0,35051
1250 -0,81 -0,21 0,09 -0,31 0,9875 0,1875 0,1875 -0,1125 0,516686
1600 -0,58 -0,28 -0,48 -0,38 1,525 0,525 -0,475 0,125 0,71863
2000 -0,20 -0,10 0,10 -0,20 1 -0,5 0,2 -0,3 0,459814
2500 -0,86 -0,46 0,34 -0,36 0,8375 0,5375 0,3375 -0,3625 0,590248
3150 -1,06 -0,76 -0,06 0,34 0,8375 0,6375 0,6375 -0,5625 0,722965

L'n,w -0,38 0,63 0,63 -0,38 0,63 -0,38 -0,38 -0,38 0,52
CI 100-2500 0,00 0,00 0,00 -1,00 1,00 0,00 0,00 0,00 0,53
ISO717
CI 50-2500 0,63 -0,38 0,63 -1,38 0,63 -0,38 0,63 -0,38 0,74
ISO717
L'n,w -0,38 0,63 0,63 -0,38 0,63 -0,38 -0,38 -0,38 0,52
L'nw+CI -0,38 0,63 0,63 -1,38 1,63 -0,38 -0,38 -0,38 0,92
L'nw+CI 50-2500 0,25 0,25 1,25 -1,75 1,25 -0,75 0,25 -0,75 1,04

E29
Table B6-Ln. Normalised impact sound pressure level Ln differences about the average value for all operators, measurement case B
Frequency OP4 OP5 OP8 OP9 dLn_stddev_B
50 -0,65 0,45 2,45 -2,25 1,973998
63 -1,15 -0,45 1,25 0,35 1,034408
80 -0,625 -1,725 1,875 0,475 1,53921
100 -0,475 -1,875 1,325 1,025 1,477329
125 -0,325 -2,025 0,375 1,975 1,658061
160 -0,2 -0,6 0 0,8 0,588784
200 -0,525 -0,625 0,475 0,675 0,670199
250 0,5 -1 0,9 -0,4 0,860233
315 -0,45 -0,75 0,85 0,35 0,732575
400 -0,4 -0,6 0,3 0,7 0,60553
500 -0,225 -0,725 -0,125 1,075 0,763217
630 -0,35 -0,65 -0,15 1,15 0,793725
800 -0,7 0,1 -0,3 0,9 0,68313
1000 -0,225 -0,025 -0,525 0,775 0,556028
1250 -0,375 -0,075 -0,275 0,725 0,499166
1600 -0,55 0,15 -0,25 0,65 0,519615
2000 -0,45 0,25 -0,25 0,45 0,420317
2500 -0,575 0,425 0,225 -0,075 0,434933
3150 -0,775 0,425 0,525 -0,175 0,60208

L'n,w -0,5 -0,5 0,5 0,5 0,57735

CI 100-2500 0 0 0 0 0
ISO717
CI 50-2500 0,5 -0,5 -0,5 0,5 0,57735
ISO717
L'n,w -0,5 -0,5 0,5 0,5 0,57735
L'nw+CI -0,5 -0,5 0,5 0,5 0,57735
L'nw+CI 50- 0 -1 0 1 0,816497
2500

E30
Table D6-Ln. Normalised impact sound pressure level Ln differences about the average value for all operators, measurement case D
Frequency OP4 OP5 OP6 OP9 dLn_stddev_D
50 -0,675 -1,775 1,125 1,325 1,486327
63 -1,375 2,425 -1,275 0,225 1,774589
80 0,075 -0,925 -0,325 1,175 0,88459
100 -1,425 2,475 -1,825 0,775 2,007278
125 -0,625 -0,925 -0,525 2,075 1,393736
160 -0,7 0,4 0,2 0,1 0,483046
200 -0,3 1,4 -0,4 -0,7 0,948683
250 -0,55 1,55 -1,05 0,05 1,126943
315 -0,9 1 -0,1 0 0,778888
400 -0,125 -0,325 0,075 0,375 0,298608
500 -0,575 0,825 -0,675 0,425 0,741058
630 -0,5 0,6 -0,8 0,7 0,761577
800 -0,225 0,275 -0,725 0,675 0,607591
1000 -0,225 0,475 -1,525 1,275 1,187083
1250 0 0,8 -1,2 0,4 0,864099
1600 -0,325 1,175 -0,925 0,075 0,88459
2000 -0,65 0,15 -0,05 0,55 0,5
2500 0,275 0,775 -1,525 0,475 1,037224
3150 0,175 1,275 -0,725 -0,725 0,95

L'n,w -0,25 0,75 -1,25 0,75 0,957427

CI 100-2500 -0,5 0,5 0,5 -0,5 0,57735
ISO717
CI 50-2500 -0,5 0,5 0,5 -0,5 0,57735
ISO717
L'n,w -0,25 0,75 -1,25 0,75 0,957427
L'nw+CI -0,75 1,25 -0,75 0,25 0,957427
L'nw+CI 50- -0,75 1,25 -0,75 0,25 0,957427
2500

E31
Table F6-Ln. Normalised impact sound pressure level Ln differences about the average value for all operators, measurement case F
All cases and standard deviation according to ISO 140-2
Frequency OP4 OP5 OP8 OP9 dLn_stddev_F dLn_stddev_allcases dLn_stddevISO140
50 -1,525 -0,325 0,275 1,575 1,28938 1,37
63 0,6 -1,5 0 0,9 1,067708 1,31
80 0,55 -1,75 0,75 0,45 1,173314 1,66
100 0,125 -3,175 2,725 0,325 2,424012 1,84 2,5
125 0,575 -1,925 1,075 0,275 1,325079 1,40 2,14
160 -0,025 -0,425 0,275 0,175 0,30957 1,32 1,96
200 0,7 -0,4 -0,7 0,4 0,658281 0,92 1,79
250 -0,05 0,45 -0,55 0,15 0,420317 0,67 1,79
315 0,375 -1,025 -0,025 0,675 0,741058 0,63 1,79
400 0,325 -0,175 0,025 -0,175 0,236291 0,40 1,79
500 -0,25 -0,25 0,25 0,25 0,288675 0,53 1,79
630 -0,35 -0,75 0,35 0,75 0,675771 0,60 1,79
800 -0,425 -0,525 -0,025 0,975 0,684957 0,51 1,79
1000 -0,475 -0,775 0,125 1,125 0,838153 0,65 1,43
1250 -0,475 -0,575 0,725 0,325 0,629153 0,56 1,43
1600 -1,1 0 1,2 -0,1 0,94163 0,70 1,79
2000 -1,85 -0,55 3,15 -0,75 2,176388 0,95 2,14
2500 -1,775 -0,075 3,025 -1,175 2,136001 1,02 2,50
3150 0,025 0,225 1,325 -1,575 1,195478 0,79 2,50

L'n,w -0,5 -0,5 1,5 -0,5 0,674 0,674

CI 100-2500 0,75 -0,25 -1,25 0,75 0,550 0,739
ISO717
CI 50-2500 ISO717 0,25 0,25 -0,75 0,25 0,590 0,827
L'n,w -0,5 -0,5 1,5 -0,5 1 0,674
L'nw+CI 0,25 -0,75 0,25 0,25 0,5 0,739
L'nw+CI 50-2500 -0,25 -0,25 0,75 -0,25 0,5 0,827

E32
 Return address:
Nordic Innovation Centre,
Stensberggata 25
NO-0170 Oslo, Norway

NORDTEST
NORDTEST is a Nordic Innovation Centre
brand offering competence and expertice
in the field of harmonizing of norms and
methods,a large Nordic net-work of experts,
more than 650 recommended Nordic testing
methods and 550 published technical reports.
Nordic Innovation Centre
www.nordicinnovation.net
The Nordic Innovation Centre initiates and finances
activities that enhance innovation collaboration and
develop and maintain a smoothly functioning market in
the Nordic region.

The Centre works primarily with small and medium-

sized companies (SMEs) in the Nordic countries. Other
important partners are those most closely involved with
innovation and market surveillance, such as industrial
organisations and interest groups, research institutions
and public authorities.

The Nordic Innovation Centre is an institution under the

Nordic Council of Ministers. Its secretariat is in Oslo.

For more information: www.nordicinnovation.net