Assessment of Sound Emission Phenomena at a Boundary Layer during

Steel Quenching

Franc RAVNIK and Janez GRUM

University of Ljubljana, Faculty of Mechanical Engineering, Aškerčeva 6, 1000 Ljubljana

Phone: +386 1 4771765, Fax: +386 1 4771225; e-mail: franc.ravnik@fs.uni-lj.si; janez.grum@fs.uni-lj.si

Abstract The final stage in the production of machine components in the manufacturing process often represents quench-

ing and tempering. Mechanical properties, such as residual stresses and hardness layer of final part, therefore

depend on optimum parameters of a quenching process chosen and monitoring of the process itself. The paper

treats an experimental setup comprising detection of sound emission in the course of quenching process. Due to

heat transfer from a specimen’s surface to a quenching medium, film boiling and nucleate boiling occur round a

heated specimen, which strongly affects quenching. An investigation of sound emission in the quenching process

was carried out with cylindrical specimens made of chrome-molybdenum heat-treatable steel 42CrMo4

quenched in different quenching media. Sound-pressure signals demonstrated by different amplitudes depending

on time at different frequencies are shown in 3D diagrams. It has turned out that an analysis of sound-emission

signals are connected with wettings kinematics and can provide useful information that confirms differences

occurring in quenching with different quenching media, different specimen’s shapes and under different quench-

ing conditions. Analysis of sound emission signals can give useful information and confirm the differences

caused by different quenching conditions. The results lead to the applicability of the new approach to the control

of the hardening processes of steels.

Keywords: sound emission, acoustic spectrograph, cooling rate, polymeric water solution, quenching, sound

pressure level, vapor film

1. Introduction

In order to use material as much as possible, most machine parts manufacturing concluded

with quenching to obtain the desired hardness profile. A purpose of quenching is to provide a

cooling rate higher than the upper critical cooling rate in a certain depth of a machine compo-

nent in order to obtain desired mechanical properties. Critical cooling rates with carbon steels

and low-alloy steels can be achieved by cooling a machine component from the austenitic mi-

crostructure, which is obtained by immersion into quenching media such as water, oil, poly-

meric water solutions, and emulsions [7]. The rate of heat transfer during quenching depends

on the wettability of the liquid quench media. Wettability can be defined as the tendency of a

liquid to spread on a solid substrate and can be characterized by the degree and the rate of

wetting [10].

During the quenching process in fluids, the occurrence of three wetting phases of heat transfer

to quenching media with Leidenfrost temperature between 100 and 200°C and consequently

the varying of the heat transfer coefficient α, is characteristic [13, 14]:

� In the initial phase of immersion, a stable vapor film of a fluid quenching medium

forms around the workpiece, so that the heat transfer to the quenching medium is con-

siderably slowed down (Fig. 1) [13]. Heat transfer coefficient varies. aFB=100 to 250

W/m2K when quenching in water.

� In further cooling the vapor film gradually decays so that a phase of nucleate boiling

occurs at some parts of the workpiece surface having a temperature lower than the

Leidenfrost temperature. In nucleate boiling, a surface section wets with the liquid

quenching medium and the heat transfer from the workpiece increases. Vaporizing of

30th European Conference on Acoustic Emission Testing & 7th International Conference on Acoustic Emission University of Granada, 12-15 September 2012

www.ndt.net/EWGAE-ICAE2012/

media at simultaneous vertical moving of specimen enables heat transfer coefficient

anb=10 to 20 kW/m2K when quenching in water.

� At lower temperatures, the third phase of convective cooling occurs at the interface

with heat transfer coefficient aconv= approx. 700 W/m2K when quenching in water.

Spreading velocity of the wetting front depends on several physical properties of the specimen

and the quenching medium like heat transfer coefficient, surface roughness, shape of the spec-

imen, temperature of the quenching media and forced convection and other. Specimen cooling

is therefore liable to a great local variation, which influences the microstructure and mechani-

cal properties of the specimen obtained [5].

The three phases are characterized by different modes of heat transfer, which contributes to

different cooling intensities. With exacting workpiece shapes it often occurs during heat

treatment, i.e. during workpiece cooling, that all the three phases occur simultaneously on dif-

ferent workpiece surface areas, which produces considerable thermal stresses and also an ex-

plicit influence on microstructural stresses leading to distortion and residual stress, or even to

distortion and cracking, of the workpiece. Therefore, a distribution of heat transfer coefficient

is an important parameter for the distortion of the whole part [14]. Consequently, the meas-

urement of the wetting behavior of the workpiece is essential for the control of the quenching

process. Steel quenching starts at the austenitizing temperature of a given steel, the quenching

process in fluid quenching media always being accompanied by an evaporation phase of the

medium in contact with the workpiece surface, since a boiling point of most of quenching

media ranges between 100 °C and 300 °C.

Formation and liberation of vapor bubbles, their oscillation and disappearing in the liquid

generate noise, which is strongest in the transition layer between film boiling and nucleate

boiling of the quenching medium. Detection of emitted sound signals and their analysis can

therefore provide useful information on the quenching process [4]. Sound generated by bub-

bles in the water was examined by Minnaert [6]. The Minnaert model of bubble formation

frequency during immersion indicates a relation between the bubble size and the frequency of

bubble formation (Eq. 1.). In the initial phase of nucleate boiling, bubbles of smaller diame-

ters are formed and their frequency is higher, and vice versa when nucleate boiling of the

quenching medium is nearing its end.

Nucleateboiling stage

Vapourblanket stage

Convectiveheat transfer stage

Maximumcooling rate

Time t [sec] / Cooling rate v [°C/s]c

Tem

pe

ratu

re T

[°C

]

Fig. 1. Nukiyama boiling curve and dif-

ferent stages of boiling (heat

transfer).

Fig. 2. Cooling rate and temperature dur-

ing quenching.

Minnaert's frequency of oscillation bubble:

ρ

χω 031 p

RO

M

⋅= [1]

where:

p0 [N/m2] .... is hydrostatic pressure of the liquid around the bubble in conditions of stat-

ic equilibrium,

χ [ ] ............. polytrophic coefficient,

RO [m] ........ radius of the bubble in the equilibrium and

ρ [kg/m3] .... the density of the liquid

2. Experimental procedure

A measuring setup should be adapted to the expected frequency of sound phenomena. Thus it

should be known which frequencies of bubble formation and decay occur predominantly in

the range of hearing and which slightly is below this threshold [11] (Fig. 3).

The conditions at which the signals

were detected should be monitored

by temperature measurement of the

specimen during the quenching pro-

cess. The phenomena occurring at

the workpiece/medium interface

should then be logically interrelated

in film boiling, including additional

environment sound effects [4].

Narazaki [9] concluded that the

sound pressure level is a function of

the temperature of the quenching

media and cooling stage, respective-

ly. Furthermore, bubble formation

and sound generation phenomena

should be interrelated with material properties obtained.

2.1 Experimental setup

Figure 4 shows the experimental setup for detecting sound signals during wetting processes.

The setup is to be independent from the quenching medium type used and of the quenching

mode. Although the quenching process takes some seconds or even up to several minutes, the

experimental system should register individual events sensed up to 0.1 second or less.

Experimental setup used in the experiment comprises two independent lines, i.e.:

� a setup for monitoring the temperatures at the surface and in the core during quench-

ing and monitoring of the temperature of the medium itself;

� a setup for detection and processing sound pressure level (Lp).

The measuring setup for detecting and processing of sound signals includes:

� A high sensibility hydrophone in frequency ranges from 0.1 Hz to 180 kHz in a wide

range of working temperatures between 40°C and 80°C. The hydrophone is piezoelec-

tric transducer, that is, it uses piezoelectric ceramics as sensing elements permanently

bonded into sound-transparent rubber.

0 2.52.01.51.00.5Bubble s diameter ’ [mm]

Fre

qu

ency [

kH

z]

20

15

10

5

0

depth in water 100 mm

depth in water 150 mm

frequency range

Fig. 3. Expected frequency range depends on

the size of the bubble and immer-

sion depth.

� A third-octave and octave band-pass filter in combination with multi-channel measur-

ing amplifier/pre-amplifier. It enables frequency response measurements on electro-

acoustic transducers with 50 third-octave filter bands, centre frequencies from 2 Hz to

160 kHz.

� A two-channel Sound Blaster card enables an analog-to-digital conversion of the sig-

nal captured and permits 16- or 24-bit sampling of velocity up to 96 kHz.

� Specially designed program package, for the detection, recording and processing of

sound signals in their original and digitized form, permits subsequent evaluation and

monitoring of events in the specimen during quenching.

� PC with operating system MS Windows XP professional.

Simultaneously to sound emission, temperature variations at the workpiece surface (Tw) and

in the core, and quenching-medium temperatures (Tqm) prior to, during and after quenching

were monitored. Specimens were equipped with two thermocouples, the first being 2 mm be-

low the surface and the second in the core centre.

The measuring setup for monitoring temperature variations at the specimen surface and in its

core consisted of:

� Thermocouples, mounted near the surface and in the core, for measuring temperature

variations during quenching;

� USB I/O platform for signal processing of temperature signals with a 16-bit resolution

and sampling rate of up to 200 kHz;

� Software for monitoring time-dependent temperature.

� Digital thermometer for temperature of the quenching medium measuring, prior to,

during, and after quenching.

Fig. 4. Experimental setup for temperature measurement and experimental setup for detection

of sound signals during quenching

2.2 Specimen material, form and quenching conditions

The specimens of different cylindrical forms were made of low-alloy Cr-Mo heat-treatment

steel (EN - 42CrMo4). It is characterized by good through-hardenability, i.e. even up to 57

HRC and high strength after heat treatment. Therefore it is widely used in the production of

statically and dynamically loaded machine components.

For the steel chosen an adequate quenching medium and a suitable quenching mode ensuring

a cooling rate slightly higher than the critical one should be used [2, 3, 8, 15]. In accordance

with the steel manufacturer's recommendations, austenitizing temperature between 820 and

850°C followed by water quenching and between 830 and 860°C for oil quenching was cho-

sen.

2.3 Measuring procedure

The specimen was gradually heated to the austenitizing temperature of 860°C and when the

temperature was stable, the specimen was quickly moved to a quenching bath to be quenched

in different quenching media. The initial quenching-medium temperature was about room

temperature, whereas the final temperature varied due to the different specific heats of the

quenching media used. The captured sound signals were subsequently evaluated with the pro-

gram, SpectraLAB, and the temperature signals were processed in MS EXCEL.

In the acoustic theory, sound motion in liquids and gases is always longitudinal in a form of

compression wave motion. This means that particles oscillate in longitudinal direction and

produce periodic compression and extension, which propagate through the media with the

sound velocity. Therefore material density and the sound pressure increase in the compres-

sion. The pressure changes are caused by the longitudinal wave motion and defined by the

sound pressure. The total sound pressure consists of positive and negative compression dis-

turbances measured from the equilibrium value of the sound pressure. The sum of all devia-

tions from the equilibrium position equaling zero, in practice the term effective sound pres-

sure pRMS is used and is expressed in case of sinusoids by:

( )[ ]ctxkPppRMS −⋅== 22

1

2 cos [Pa] [2]

The sound-wave motion is composed of several superimposed types of wave motions having

different frequencies, amplitudes, and phases. A change of the sound pressure can be deter-

mined at an optional point in the space in which sound propagates. Sensing sound changes

should measure the variation of timely sound pressure. Recorded acoustic signal should per-

mit the determination of sound frequency and sound intensity given in synoptic logarithmic

scale. A representation of level values with a logarithmic ratio of the acoustic quantity chosen

to its reference value can be described as a time variation. The unit of sound power level in

acoustics is expressed in decibels [dB]. Consequently, results of a given acoustic characteris-

tic shall always be given not only as a measured value but also with its reference value [1].

The sound power at a certain point is thus a difference between the instantaneous pressure and

the ambient pressure. The sound pressure (p) is expressed in [Pa]. A reference value of the

sound-pressure for liquids pref is equal to 10-6

Pa. The results of spectral analysis are graph-

ically represented in the diagrams where the complex sound pressure level Lp is a function of

frequency. The sound-pressure level Lp is defined as:

refref

pp

p

p

pL log20log10

2

2

⋅=⋅= [dB] [3]

where:

p [Pa] ........ sound pressure

pref [Pa] ..... reference value of the sound-pressure

P1 [Pa] ...... instantaneous sound pressure amplitude

K [ ] .......... is the wave number:

c [m/s] ...... is the velocity of the sound-wave motion

t [s] ........... time

3. Results

The results of spectral analysis are diagrams showing the complex sound pressure level Lp as

a time depend function of frequency. The sources of the sound pressure are consequences of

the occurrence of the phase transformation of' steel and of the boiling process at the interface

during the cooling process. The recorded diagram of the sound pressure shows a three-

dimensional variation of the sound pressure level as a function of frequency and time in dif-

ferent diagrams, such as the time variation of the pressure amplitude, a phase diagram, a pow-

er spectrum of the signal etc. The displays of the sound-pressure levels in the diagrams pro-

vide the best displays of the dependence between the acoustic signal captured and the events

at the specimen/quenching-medium interface.

Film

bo

iling

Tra

nsi

tio

n b

oilin

g

Nu

clea

te b

oili

ng

Fre

e c

on

vect

ion

Fig. 5. Typical sound pressure signal. Fig. 6. Spectrogram of sound pressure signal.

� Time series diagram represents the amplitude ratio during quenching process. A typi-

cal raw sound signal captured during quenching, recorded from the hydrophone in the

quenching media, is presented in Figure 5. The amplitude ratio between maximum

level at the start of the quenching process and background noise level at the very end

of the process is sufficient to identify different stages of the quenching process.

� A spectrogram is a calculated time record of sound signals, in which frequency is seen

as the ordinate value and signal intensity as color intensity: cold - blue color - low in-

tensity; hot - red color – strong intensity (Figure 6).

3.1 Analysis of the acoustic spectrum

A spectrogram is a time record of sound signals, from which a frequency is evident as a value

at the ordinate and signal intensity as color intensity (cold, blue color - low intensity, hot, red

color – strong intensity). The comparison of the duration of the vapor blanket stage calculat-

ed from the ultrasonic and the reference data shows a good agreement with respect to the

measurement uncertainty [12].

The spectrogram of the quenching process shows the presence of low frequencies at the be-

ginning of the process, i.e. in the first 10 seconds, when using water as a quenching medium.

Intense sound pulses occur due to the formation and decay of the vapor film. Their frequency

gradually reduces due to a decreasing cooling rate.

The analysis of the sound spectrum obtained during quenching comprises the detection of the

sound-pressure level and transformation into a voltage signal. It provides detailed information

on individual events in the quenching medium used. Fig. 7 shows two records of sound pres-

sure level signals detected with the hydrophone and shown in the spectrogram giving the fre-

quency of events during nucleate boiling in the water. Analyzing both records, the following

may be concluded:

12

3

5

4

7

Time (seconds)

Fre

qu

en

cy (

Hz)

SP

L

(dB

rm

s)

6

Perc

ent

Full S

ca

le

Time (100 sec/div) Fig. 7. Acoustic signal obtained in quenching in water.

� The beginning of quenching is described by the signals of different frequencies, which

in their amplitude differ from the environmental noise (Fig. 7, point 1);

� The end of quenching is described by the signal having intensity comparable to that of

the environmental noise, which means that the number and intensity of the signals

with regard to the frequency and their intensity reduces. Determination of the end of

cooling is of subjective nature and cannot be made by means of the spectrogram. (Fig.

7, point 2);

� The process duration is the difference between the beginning and end of quenching;

wherefrom an average cooling rate can be established (Fig. 7, area 3);

� From the voltage signal detected with the hydrophone and from the spectrogram, two

characteristic areas can be determined:

* an area of lower frequencies, i.e. up 1 kHz in water quenching, and up to 2 kHz in

oil quenching, where only the amplitude is changing (Fig. 7, area 4), and

* an area of higher frequencies, i.e. up 18 kHz in water quenching, and up to 20 kHz

in oil quenching, where the signal frequency is changing as well (Fig. 7, area 5).

� Individual peaks of higher sound-emission intensity given by a voltage signal of high-

er frequencies (up to 6 kHz in water quenching, and up to 14 kHz in oil quenching)

due to surface oxidation and oxide cracking (scale) at the surface (Fig. 7, points 6) are

identified. In sound emission there are also short-time peaks of high intensity with fre-

quencies of up to 12 kHz in water quenching as a result of too high a cooling rate (Fig.

7, area 7).

Generally it can be concluded that from the detected sound emission, characteristic changes at

the specimen/quenching-medium interface during quenching could be determined.

3.2 Analysis of the sound signals generated at quenching process

The time series sound pressure signals for

the specimens quenched in water (Fig.

9A) and in salt solution (Fig. 9B) are

graphically represented in the diagrams

below. Salt solution was chosen on pur-

pose to obtain too high internal stresses,

which cause the cracking of the specimen

(Fig. 8B).

10,0 20,0 Time t [s]

0,0

50,0

-50,0

Perc

en

t F

ull

scale

0,0 10,0 20,0 Time t [s]

0,0

50,0

-50,0

Perc

en

t F

ull

scale

Peak amplitude

A quenched in water B quenched in salt solution

Fig. 9. Time series sound pressure signal

Comparing sound pressure diagrams obtained indicates important characteristics shown with

differentiation in sound intensity and frequency ranges. Analysis provides detailed infor-

mation on individual events in the quenching medium used. Figures 9, 10 and 11 show sound

pressure level signals detected with the hydrophone and calculated spectrograms, giving the

frequency of events during the quenching of the specimen in pure water and salt solution. By

comparing signals of both quenching media, the following may be concluded:

Fig. 8. Specimen shape

A Before quenching B After quenching

� Significant change of the amplitude can be determined when quenching in salt solu-

tion: after approx 16 seconds of quenching process duration significant amplitude

peak shows the moment when specimen cracking occurs (Fig. 9B).

� Typical events are recognized: background noise, the beginning of the cooling process

and specimen cracking (Fig. 10A). Time zooming shows that the duration of cracking

signal is less than 0.05 seconds when four peaks follow one after another (Fig. 10B).

0,0 10,0 20,0 Time t [s]

0,0

50,0

-50,0

Pe

rcen

t F

ull

sca

le

Beginig of the process

Background noise

Specimen cracking

Process start

0,0

50,0

-50,0

Pe

rcent F

ull

sca

le

16,2 16,3 16,4Time t [s]16,5

1. peak

2. peak

3. peak

4. peak

Crack appearance period

Crack decay period

A typical events B “Time zooming” signal

Fig. 10. Time series sound pressure signal when quenched in salt solution

� Significant change of the amplitude shown as sudden color change can be also deter-

mined in spectrogram of sound pressure signal when specimen cracking occurs (Fig

11B). Spectrogram shows that the change occurs in all frequency ranges for a very

short time shown as more intensive red colour.

Fre

quency

[Hz]

-55

-65

-75

-85

-105

-115

-95 Rela

tive

Am

plit

ude [dB

]

0,17 24,45Time [1 sec/div]

1,0k

500

400

300

600

800

2,0k

3,0k

4,0k

5,0k6,0k

8,0k

10,0k

20,0k

Fre

quency

[Hz]

-55

-65

-75

-85

-105

-115

-95 Rela

tive

Am

plit

ude [dB

]

0,17 24,45Time [1 sec/div]

1,0k

500

400

300

600

800

2,0k

3,0k

4,0k

5,0k6,0k

8,0k

10,0k

20,0k

Pe

ak

am

pli

tud

e

A quenched in water B quenched in salt solution

Fig. 11. Spectrogram of sound pressure signal

� Figure 12 shows the calculated average spectrum for both quenching media. Figure 12B

shows in the cracking moment significant change of the intensity in frequency range be-

tween 6 and 7 kHz.

� Detailed investigation of sound pressure signal enables evaluation of frequency signals.

Figure 13 shows the course of high frequency signals 19, 20, 21.2, 22.4 and 23 kHz,

during time. In the moment of cracking of the specimen, between 16.2 and 16.25 se-

conds, a significant increase of the signal amplitude can be recognized by frequencies.

300 500 700 1,0k 2,0k 3,0k 5,0k 7,0k10.0k 20.0k

Frequency [Hz]

-50

-60

-70

-80

-90

-100

Rela

tive

Am

plit

ud

e [dB

]

300 500 700 1,0k 2,0k 3,0k 5,0k 7,0k10.0k 20.0k

Frequency [Hz]

-50

-60

-70

-80

-90

-100

Rela

tive

Am

plit

ude

[dB

]

Peak amplitude

A quenched in water B quenched in salt solution

Fig. 12. The average spectrum of samples

-140

-120

-100

-80

-60

-40

-20

0

0,1

1,3

2,5

3,6

4,8

6,0

7,2

8,3

9,5

10

,7

11

,9

13

,0

14

,2

15

,4

16

,5

17

,7

18

,9

20

,1

21

,3

22

,4

23

,6

Am

plitu

de

[dB

]

Time [sec]

23,0 kHz 22,4 kHz21,2 kHz

20,0 kHz19,0 kHz

Fig. 13. The course of frequency signals during time

3. Conclusions

This article describes the signals captured by measuring sound emission caused during

quenching. The sound emission occurring during quenching is found predominantly in the

audible frequency spectrum; therefore, detecting the sound pressure signal is efficient since

the phenomena at the interface can be monitored. The sound pressure signal, which exposes

as the amplitude and duration of the signal during quenching, was captured by the hydro-

phone. During recording of the sound pressure, a change of the signal shape appears, corre-

sponding to the formation of a vapor film phase on the specimen surface and to the nucleate

boiling phase on the specimen/quenching-medium interface. Significant change of the signal

followed when specimen cracking occurs. This change was connected with the high internal

stresses during quenching process, which caused rupture of the specimen in axial direction.

The signal was audible during the entire quenching process and the capturing of the sound

emission with hydrophone was satisfactory. Significant change of the signal was well visible

and could be recognized in the cracking of the specimen. On the basis of the experiment, the

conclusion can be drawn that the experimental setup for the capturing of the sound emission

was reliable and has given sufficient results. The analysis of the results offers an interesting

new approach to evaluation and, more importantly, to monitoring, controlling and optimizing

of the quenching process.

References

1 Čudina M.: Technical Acoustics - Measuring, evaluation and decreasing noise and vi-

bration (in Slovene: Tehnična akustika – merjenje, vrednotenje in zmanjševanja hrupa in

vibracij), University of Ljubljana, Faculty of Mechanical Engineering, 2001.

2 Grum J., Božič S., Lavrič R.: Influence of Mass of Steel and a Quenching Agent on Me-

chanical Properties of Steel, Heat Treating, Proceedings of the 18th Conference, Includ-

ing the Liu Dai Memorial Symposium, ASM International, The Materials Information

Society, Eds.: R.A. Wallis, H.V. Walton, pp 645-654, 1998.

3 Grum J., Bozič S. Zupančič M.: Influence of quenching process parameters on residual

stresses in steels, ASM Heat Treating Society, The 3rd International Conference on

Quenching and Control of Distortion, Eds.: G.E. Totten, B. Liščić, H.M. Tensi, Prague,

Czech Republic, pp 530-541, 1999.

4 Grum J., Ravnik F., Investigation of sound phenomena during quenching process. Inter-

national Journal of Materials and Product Technology, Vol. 27, No. 3/4, pp. 266-288,

2006.

5 Howes M. A. H.: Steel Heat Treatment Handbook, Marcel Decker, Chicago, 997.

6 Leighton T. G.: The Acoustic Bubble, Academic Press, Inc., San Diego, California,

USA, pp 613, 1997.

7 Liščić B., H. M. Tensi,L. C. F. Canale and G. E. Totten: Quenching theory and technol-

ogy, second edition, CRC press, Taylor&Francis Group, Boca Raton, USA, 2010.

8 Macherauch E., Vohringer O.: Residual Stresses After Quenching, Eds.: B. Liščić, H.M.

Tensi, W. Luty: Theory and Technology of Quenching, A Handbook, Springer-Verlag

Berlin, Heidelberg, pp 117-181, 1992.

9 Narazaki, M., I. Takatsudo and S. Fushizawa: Boiling noise in the water quenching of

hot metals, Netsu Shori (Heat Treatment), Vol. 28, No.2, pp 93-99, 1988.

10 Prabhu N. K. and Fernandes P.: Heat Transfer During Quenching and Assessment of

Quench Severity—A Review, Journal of ASTM International, Selected Technical Papers

STP1523, Quenching and Cooling, Residual Stress, and Distortion Control, Eds.:

Lauralice C.F. Canale and Narazaki, M., ASTM International, 100 Barr Harbor Drive,

PO Box C700, West Conshohocken, PA 19428-2959, pp 40-62, 2010.

11 Prezelj J., Ravnik F., Grum J., Čudina M.: Sound signature of steel hardening process.

V: 3rd Congress of the Alps Adria Acoustics Association, Graz, Austria. Proceedings.

Graz: Joanneum Research, pp 27-28, September 2007.

12 Stoebener D., Goch G.: Wet-state Ultrasonic Measurements of Cylindrical Workpiece

during Immersion Cooling, IDE 2005, Bremen, 2005.

13 Tensi, H. M.: Wetting kinematics. Eds: Liščić, B.; Tensi, H. M.; Luty, W.: Theory and

Technology of Quenching, A handbook, Springer-Verlag, Berlin, pp 481,1992.

14 Tensi, H.M., and G.E. Totten: Development of the Understanding of the Influence of

Wetting Behaviour on Quenching and the Merits in these Developments of Prof. Imao

Tamura', Proceedings of the 2nd

International Conference on Quenching and the Control

of Distortion, Nov. 4-7, PP 17-27, 1996.

15 Totten G., Canale I. C. F.: Overview of Distortion and Residual Stress Due to Quench

Processing: Part I – Factors Affecting Quench Distortion, International Journal of Mate-

rials and Product Technology, Inderscience enterprises Ltd., 2005.