CHAPTER 5

SOUND LEVEL MEASUREMENT ON POWER

TRANSFORMERS

5.1 Introduction

Sound pressure level measurements have been developed to quantify pressure

variations in air that a human ear can detect. The perceived loudness of a signal is
dependent upon the sensitivity of the human ear to its frequency spectrum. Modern
measuring instruments process sound signals through electronic networks, the
sensitivity of which varies with frequency in a manner similar to the human ear.

5.2 Methods for Sound Level Measurement

Sound power is the parameter, which is used for rating and comparing sound
sources. Sound power can be calculated from sound pressure or sound intensity as
discussed in chapter-1. Sound intensity measurements have the following advantages
over sound pressure measurements [14].

i) An intensity meter responds only to the propagating part of a sound field and
ignores any non-propagating part, for example, standing waves and reflections.

ii) The intensity method reduces the influence of external sound sources, as long as
their sound level is fairly constant.

The sound pressure method takes the above factors into account through
correction for background sound and reflections [14].
The sound intensity is the time-averaged product of sound pressure ’p’ and
particle velocity ’v’[14].
5.1

Where, I=Intenisty, p=pressure, v=velocity

49
 It is found that when converted in to sound power, the sound intensity
measurement leads to values of 2 to 3 dB less in comparison to the sound pressure
measurement.

As reported in [14], Measurements on 30 transformers in the range from 0.1 to

350 MVA, a mean difference of 3.8dB(A) was established between pressure and
intensity measurement values.

5.3 Procedure for Sound Level Measurement as per Standards

International standards IEC 60076-10 and IEEE standard C57.12.90-1993 lay

down the standard procedure for sound level measurements on transformers. Some of
the salient points are extracted from the standards and presented here to introduce the
subject.

Ambient Sound Pressure Level

The ambient sound pressure levels are measured immediately preceding and
immediately following the sound measurements with the transformer energized. The
ambient sound shall be measured at minimum four locations. Additional measurement
may be made if the ambient measurements vary by more than 3 dB around the
transformer. At least one of the locations for measuring ambient sound pressure levels
shall be on the center of each face of the measurement surface.

Reference Sound-Producing Surface

The reference sound-producing surface of a transformer is a vertical surface that

follows the contour of a string stretched around the periphery of the transformer or
integral enclosure. The contour shall include radiators, coolers, tubes, switch
components, and terminal chambers, but exclude bushings and minor extensions, such
as valves, oil gauges, thermometers, terminal boxes, and projections at or above cover
height.

First Measurement Position

The first microphone location shall coincide with the main drain valve. Additional
microphone locations shall be at 1m intervals in a horizontal direction, proceeding
clockwise on the measuring surface.

50
 Fig. 5.1 Microphone Location for Measuring Audible Sound on Transformer

Locations of Microphones from Sound-Producing Surface

The microphone shall be located on the measurement surface. As per IEEE

Standard C57.12.90-1993, the microphone shall be spaced 0.3m away from the
reference sound producing surface (principle sound radiating surface). When fans are
in operation, the microphone shall be located 2m away from any portion of the
radiator, coolers, or cooling tubes cooled by forced air. As per IEC 60076-10, for
measurements made with forced air cooling with auxiliaries out of service, the
prescribed contour shall be spaced 0.3m away from the principle radiating surface and
when these in service, the prescribed contour shall be spaced 2m away from the
principle radiating surface. There shall be a minimum of six microphone positions.

Generally, sound level is measured from sound-producing surface, apart at 0.3m

for ONAN and 2m for remaining cooling scheme. When the distance of microphone
becomes double, the drop in sound pressure level will be 6 dB. Following formula
is used for calculation of sound level at different distances from the sound emitting
surface.

5.2

Where, d=reference measurement distance (m), d1=desired measurement distance (m)

Height of Microphone Locations

As per IEEE Standard C57.12.90 -1993, and IEC-60076-10, for transformers

having an overall tank or enclosure height of less than 2.4m and 2.5m respectively,
51
measurements shall be made at half height. For transformers having an overall tank or
enclosure height of 2.4m or more, measurements shall be made at one-third and at
two-thirds height. Sound measurement of cooling system shall be made at locations
only at one-half height.

Correction Factors

Measurement should be made in an environment having an ambient sound

pressure level at least 5 dB (as per IEEE standard) or 3 dB (as per IEC-551 and CBIP
manual) below the measured sound pressure level of the transformer. When the
ambient sound pressure level is 3 dB or more below the measured transformer sound
level, the corrections shall be applied to the combined transformer and ambient
pressure level to obtain the transformer sound pressure level. The various correction
factors as defined in IEEE, IEC-551 and CBIP manual are reproduced below in Table
5.1. The measurements carried out on transformer with ambient sound level
difference of less than 3 dB are invalid.

Table 5.1 Sound level Correction Factors as per Standards

IEEE IEC 551 CBIP

Difference Difference Difference

Correction Correction Correction
between between between
to be to be to be
combined & combined & combined &
subtracted subtracted subtracted
Background Background Background
(dB) (dB) (dB)
Noise (dB) Noise (dB) Noise (dB)

5 1.6 3 3 3 3

6 1.3 4-5 2 4-5 2

7 1.0 6-8 1 6-8 1

8 0.8 9-10 0.5 - -

9 0.6 - - - -

10 0.4 - - - -

Above 10 0.0 - - - -

52
5.4 Scales for Sound Level Measurement

There are three types of scales for sound level measurement: (1) A-weighting
scale (2) B-weighting scale (3) C-weighting scale. The typical frequency response
waveforms for the above weighting scales are represented in Fig.5.2.

Fig. 5.2. Frequency Response for the A, B and C Weighting Networks

A-weighting scale

A-weighted scale is generally accepted for measurement of transformer sound

because the purpose is to measure the sound level frequency within the band, which
can be sensed by the human ear. Sound measurements made with the A-weighting
scale are designated as dB (A).

B-weighting scale

It is used to predict the performance of loudspeakers and stereos, but not industrial
sound.

53
C-weighting scale

The C-weighting scale includes the sound at low frequency range than the A and
B scales.
Table 5.2 Sound level Conversion Chart

Decible (dB)
Frequency (Hz)
A –Weighting B -Weighting C -Weighting

10 -70.4 -38.2 -14.3

12.5 -63.4 -33.2 -11.2

16 -56.7 -28.5 -8.5

20 -50.5 -24.2 -6.2

25 -44.7 -20.4 -4.4

31.5 -39.4 -17.1 -3.0

40 -34.6 -14.2 -2.0

50 -30.2 -11.6 -1.3

63 -26.2 -9.3 -0.8

80 -22.5 -7.4 -0.5

100 -19.1 -5.6 -0.3

125 -16.1 -4.2 -0.2

160 -13.4 -3.0 -0.1

200 -10.9 -2.0 0

250 -8.6 -1.3 0

315 -6-6 -0.8 0

400 -4.8 -0.5 0

500 -3.2 -0.3 0

630 -1.9 -0.1 0

54
 800 -0.8 0 0

1000 0 0 0

1250 +0.6 0 0

1600 +1.0 0 -0.1

2000 +1.2 -0.1 -0.2

2500 +1.3 -0.2 -0.3

3150 +1.2 -0.4 -0.5

4000 +1.0 -0.7 -0.8

5000 +0.5 -1.2 -1.3

6300 -0.1 -1.9 -2.0

8000 -1.1 -2.9 -3.0

10000 -3.5 -4.3 -4.4

12500 -4.3 -6.1 -6.2

16000 -6.6 -8.4 -8.5

20000 -9.3 -11.1 -11.2

For conversion of dB to A, B & C weighted scale, add factor from conversion

chart Table 5.2 with respect to frequency of that dB level.

5.5 Sound Level Measurement on Different Scales

In order to understand the difference and verify the explanation given earlier as to
why sound level measurements are specified according to A-weighted scale, a
measurements of sound level with A and C-weighted scales were carried out on a 10
MVA, 110/33/11kV, 3-Phase, 50 Hz Power transformer.

55
 Table 5.3 Sound level Measurement on Different Scales

Test Measured Sound level (Corrected)

Background 60.8 dB(A)

No-load Condition 63.4 dB(A)

No-load Condition 72.4 dB(C)

Load Condition 61.8 dB(A)

From the above experimentation, it is inferred that, (a) sound level is high on C-
weighted scale as compared to that on A-weighted scale, because C-weighted scale
includes sound at lower frequency, which can not be sensed by the human ear. (b) A-
weighted scale is generally accepted for measurement of transformer sound because
the purpose is to measure the sound level frequency within the band, which can be
sensed by the human ear. (c) The sound level under load when excited at impedance
voltage during works testing is significantly lower than no-load sound level.

5.6 Case Study - Load Sound Level Measurement

Load sound level measurement was carried out on four power transformers during
load loss measurement at factory in order to establish the influence of load sound
level on overall transformer sound.

Fig. 5.3. Load Test of Transformer at Industry

It is well known that during load loss measurement at works while the windings
carry full load current, the transformer is excited only partially at impedance voltage,
which is merely 10-12 percent of the voltage. As such negligibly small amount of flux
is produced in the core. Accordingly, the sound produced by the core would also be

56
negligible. However, it is noticed that transformer produces sound during load loss
measurement. This is mainly due to magneto-motive force due to current flowing
through the windings and causing vibrations therein. The measured load sound levels
along with transformer details are shown in Table 5.4 below.

The core sound level of the above transformers could not be measured at rated
excitation. Therefore this value was calculated using the modified empirical formulae.
Comparison of the measured load sound level and the calculated core sound level
suggest that the former is significantly lower and hence could be ignored, for all
practical purpose
Table 5.4 Load Sound level

Rating Measured Sound level Calculated Sound level

10 MVA, 110/30/11 kV, 52.49A 61.82 dB(A) 68 dB(A)

50 MVA, 132/33 kV, 218.7 A 63.48 dB(A) 71 dB(A)

12.5MVA, 132/34.5 kV, 54.67 A 66.10 dB(A) 70 dB(A)

31.5 MVA, 132/34.5kV, 137.77A 66.23 dB(A) 71 dB(A)

5.7 Case Study – No-Load Sound Level Measurement

During no-load test, primary winding is supplied with full voltage and other
winding is kept open. In this condition the transformer core carries rated flux, which
results in significant core vibrations thereby producing sound. Various experiments to
establish behavior of the sound were carried out by changing excitation voltage and
frequency.

Characteristic of Sound level as a Function of Flux density

In order to establish the behavior of sound level as a function of flux density,

following experiments were carried out,

A Change in excitation voltage keeping the supply frequency constant

B Change in frequency keeping the excitation voltage constant

57
A) Change in excitation voltage keeping the supply frequency constant

The sound level measurements were performed on the following power and
distribution transformers raging from 100MVA to 2.5MVA (see Table 5.5) by
changing excitation voltage keeping the frequency constant.
Table 5.5 Transformer Details-1

Transformer MVA kV Class CRGO Lamination Flux

No. Rating Thickness Density

T-1 100 220/132 MOH 0.27 mm 1.6 T

T-2 63 132/33 MOH 0.23 mm 1.59 T

T-3 2.5 20/0.725 M4 0.27 mm 1.7 T

The measured sound levels in dB(A) at varying flux density are reported in Table
5.6 below and presented graphically in Fig.5.3.

Table 5.6 No-load Sound level Vs. Flux Density

Sr. Flux Density Measure sound Measure sound Measure sound

No. (T) level of T-1 level of T-2 level of T-3

1 1.8 78.00 dB(A) 70.20 dB(A) 56.08 dB(A)

2 1.7 75.00 dB(A) 68.20 dB(A) 53.73 dB(A)

3 1.6 71.16 dB(A) 64.49 dB(A) 52.00 dB(A)

4 1.5 68.00 dB(A) 63.58 dB(A) 50.95 dB(A)

5 1.4 65.00 dB(A) 62.88 dB(A) 49.81 dB(A)

6 1.3 63.00 dB(A) 62.42 dB(A) 49.00 dB(A)

7 1.2 60.00 dB(A) 61.92 dB(A) 48.31 dB(A)

8 1.1 57.00 dB(A) 61.22 dB(A) 48.26 dB(A)

9 1 57.82 dB(A) 60.72 dB(A) 47.71 dB(A)

58
 Fig. 5.4 Sound level Vs, Flux Density

It is observed from the fig 5.4 that (a) Sound level decreases with decrease in flux
density and doesnt have linear relationship, (b) Sound level changes by 2-3dB(A) on
changing the flux density value by 0.1T in the flux density zone closer to rated flux
density.

B) Change in frequency with constant excitation voltage

The sound level measurement is carried out on Power transformers by changing

frequency (in-turn flux density) ranging from 48 Hz to 53 Hz with constant excitation
voltage. The transformer details are given in Table 5.7 below.

Table 5.7 No-load Sound level Vs. Frequency

Measure Measure Measure

Frequency
Sr. No. sound level of sound level of sound level of
(Hz)
T-4 T-5 T-6

1 48.00 84.10 dB(A) 70.62 dB(A) 71.10 dB(A)

2 49.00 83.50 dB(A) 68.82 dB(A) 70.50 dB(A)

3 50.00 83.00 dB(A) 66.79 dB(A) 69.90 dB(A)

4 51.00 83.50 dB(A) 66.20 dB(A) 69.40 dB(A)

59
 5 52.00 83.90 dB(A) 66.56 dB(A) 69.90 dB(A)

6 53.00 85.20 dB(A) 67.20 dB(A) 70.10 dB(A)

Fig. 5.5 Sound level Vs. Frequency

Sound level is increases at frequencies below and above rated frequency. Higher
sound level at lower frequency than rated frequency could be attributed to higher flux
density. Higher sound level at high frequency could be attributed to higher
magnetostriction frequency.

5.8 Case Study - Comparison between Load and No-load Sound level

Measurement of Load and No-load sound level was carried out on different
transformer to find out the influence of load & No-load sound level on the overall
transformer sound level.

Fig. 5.6 No-load and Load Test of Transformer at Industry

60
 Table 5.8 Measured Load and No-load Sound level

Measured Load Measured No-Load

MVA Rating
Sound level Sound level
50 MVA, 132/33 kV, 63.48 dB(A) 74.16 dB(A)

100 MVA, 230/110/11 kV 64.21 dB(A) 75.66 dB(A)

10 MVA, 110/33/11 kV 61.8 dB(A) 63.4 dB(A)

It is observed from the Table 5.8 that, Load sound level is lower than No-load
sound level and No-load sound level (Core sound) gives the major effect on overall
transformer sound.

5.9 Measurement of Magnetostriction

Several techniques have been successfully used for the measurement of

magnetostriction in silicon – Iron laminations, but probably the simplest method is a
technique used by Neurath to measure it in localized areas using displacement
transducers. The sensitivity and accuracy of the technique was improved by Browney
and Maples who measured the fundamentals component of magnetostriction using a
ceramic displacement transducer capable of measuring ac displacements in the range
10-4 to 10-8 cm.

Simmons and Thompson made a systematic study of the use ceramic transducers
and showed ways of avoiding errors produced by transverse or vertical vibrations.
This was done by either using three separately calibrated transducers or one stereo
cartridge whose output were connected in opposition to produce a signal independent
of vertical vibrations.

The magnetostriction of individual samples is best measured using the trnaducer

technique by clamping one end of the sample and measuring the displacement of
several points along its length. A plot of displacement against distance alog the
sample is almost linear, slight undulations are present due to localized variations in
magnetostriction. The method is accurate to within +/-0.02 x 10-6 strain. It is more
convenient to measure the displacement of two pints, say 20 cm apart, by subtracting

61
the outputs of two fixed transducers of the same sensitivity. In this way the average
magnetostriction of large samples can be quickly recorded in one reading.

In order to investigate the localized variations in magnetostriction, Simmons went

on to construct a “double pickup” built from two mono cartridges in which the pickup
points were only 0.3 cm apart, small enough to measure the magnetostriction in
individual grains. A commercial double pickup with an accuracy better than 10% is
now available constructed from a piezoelectric ceramic 1.1 cm long cemented to
pickup needled 2.3 cm apart.
The measurement of the harmonics of magnetostriction is more difficult because
their magnitudes are very small, but the same technique can be used, the output of the
device being filtered to obtain the appropriate harmonic. The harmonics like the
fundamental component of magnetostriction are extremely stress-sensitive and it is
this factor which undoubtedly controls the noise output due to magnetostriction.

5.10 Sound Level Determination Report as per IEC Standards

Typical report of sound level determination

Contract: Site:

Manufacturer: Place of manufacture:

………………………………………………………………………………………………..………………………………………

Details of Transformer

Serial number: MVA:

Tapping range: Voltage ratio: kV

Connections: Rated frequency: Hz

Rated current: kA Rated voltage: kV

……………………………………………………………………………………………………..…………………………………

62
Details of guaranteed level

Sound pressure level: dB(A) Measurement distance: m

…………………………………………………………………………………………

TRANSFORMER WITH/WITHOUT COOLERS

…………………………………………………………………………………………

Details of measurement method

Measurement standard:

Sound pressure

A-weighted

………………………………………………………………………………………

Test object

TRANSFORMER WITHOUT COOLERS

TRANSFORMER WITH COOLERS

COOLERS WITHOUT TRANSFORMER

…………………………………………………………………………………………

Plan of test object

Height of microphone above ground:

……………………………………………………………………………………………………………………………………………

Test conditions

Excitation voltage: kV

Frequency: Hz

63
Tap position:

Measurement distance: m

Length of prescribed contour(s), lm: m

Height of test object, h: m

Area of measurement surface, S : m2

……………………………………………………………………………………………………………………………………………

Sound pressure method is used

A-weighted sound pressure levels of the background noise

Plan position At start of tests At end of tests

1
2
3
4
5
6
7
8
9
10
Average, LbgA
A-weighted sound pressure levels

Plan position Height 1 Height 2

1
2
3
4
5
6
7

64
 8
9
10
Average, LpAO

LpAO --- LbgA (must be  3 dB(A)): dB(A)

Environmental correction (must be  7 dB): dB

Corrected average A-weighted sound pressure level, LpA dB(A)

5.11 Conclusion

Sound energy is cause and sound pressure is the effect. The sound power is rather
the unique descriptor of the noisiness of a sound source. Measurement of transformer
sound level in A-weighted scale with Decibel (dB) unit. Sound level measurement
procedure is as per International standards IEC 60076-10 and IEEE standard
C57.12.90-1993. Sound level decreases with decrease in flux density and doesn’t have
linear relationship. Sound level behavior with respect to frequency, no significant
change in sound level for supply frequency within a range of ± 3%.

65