K. KOLAØÍK et al.: USING THE BARKHAUSEN-NOISE ANALYSIS AND METAL-MAGNETIC-MEMORY METHOD ...
437–441
USING THE BARKHAUSEN-NOISE ANALYSIS AND
METAL-MAGNETIC-MEMORY METHOD FOR MATERIAL
CHARACTERISTICS UNDER FATIGUE DAMAGE
UPORABA METODE BARKHAUSNOVEGA HRUPA IN
MAGNETNEGA SPOMINA ZA KARAKTERIZACIJO
UTRUJENOSTNIH PO[KODB MATERIALA
Kamil Kolaøík1, Jiøí [ime~ek1, Antonín Køí`1, Jiøí ^apek2
1University of West Bohemia, Faculty of Mechanical Engineering, Univerzitení 8, 30100 Plzeò, Czech Republic
2Czech Technical University, Faculty of Nuclear Sciences and Physical Engineering, Trojanova 13, 120 00 Prague, Czech Republic
kamil.kolarik@email.cz
Prejem rokopisa – received: 2016-03-30; sprejem za objavo – accepted for publication: 2016-06-09
doi:10.17222/mit.2016.056
The article deals with process monitoring and the subsequent diagnosis of fatigue damage during material testing of steel S355.
The study was performed by means of the metal-magnetic-memory method, Barkhausen-noise analysis and X-ray diffraction.
This combination could significantly enhance the inspection quality of critical mechanical-unit parts during their lifespan. X-ray
diffraction was applied for the description of the state of residual stresses, in combination with the Barkhausen-noise analysis,
which allows us to immediately determine the changes in the surface layer resulting from the residual-stress redistribution,
hardness and potential microcracks. Additionally, a new unconventional method for the determination of the stress-concen-
tration-zone location and for the detection of material defects and imperfections, the so-called metal-magnetic-memory method,
will be introduced.
Keywords: Barkhausen noise, metal-magnetic-memory method, X-ray diffraction, fatigue damage
^lanek obravnava proces kontrole, in iz tega naknadno diagnozo utrujenostne po{kodbe, med preizkusom materiala, jekla S355.
[tudija je bila izvedena s pomo~jo metode magnetnega spomina kovine, z Barkhausnovo analizo hrupa in z rentgensko
difrakcijo. Ta kombinacija lahko ob~utno izbolj{a kvaliteto kontrole kriti~nih mehanskih komponent med njihovo `ivljenjsko
dobo. Rentgenska difrakcija je bila uporabljena za opis stanja zaostalih napetosti, v kombinaciji z analizo Barkhausnovega
hrupa, kar omogo~a takoj{nje dolo~anje sprememb v povr{inski plasti, kar je posledica prerazporeditve zaostalih napetosti,
trdote in potencialnih mikrorazpok. Dodatno bo predstavljena nova neobi~ajna metoda za dolo~anje podro~ij koncentriranja
napetosti ter za odkrivanje napak in nepravilnosti, imenovana metoda magnetnega spomina kovine.
Klju~ne besede: Barkhausnov hrup, metoda magnetnega spomina kovin, rentgenska difrakcija, utrujenostna po{kodba
1 INTRODUCTION
Characteristics of all machine components depend on
the final surface quality1 and surface material layers,
namely the residual-stress (RS) distribution and the
roughness. These properties involve functionality and
durability, especially fatigue life. The surface of an
engineering component is a sensitive part, especially
because of the interface between the component and the
external environment. The surface layers of the material
are damaged by the transmitted working load and other
degradative effects such as vibrations, radiant energy,
chemical influences (rust), etc. In general, it is known
that more than 90 % of the operating-part fractures are
caused by fatigue.2
Material fatigue is a process (a change in mechanical
properties, creation of a microcrack net, growth of the
maximum crack and workpiece fracture), which
gradually deplete the utility properties of a product until
a fracture of the workpiece and irreversible damage
occur.3 Currently, the process of crack formation can
only be observed if we use non-destructive testing me-
thods such as ultrasonic testing, magnetic particle
inspection, industrial radiography and capillary testing.
These methods detect only incurred cracks and defects,
but they are not able to register the initialization of
cracks.4
Recently, it was suggested that this drawback could
be eliminated with a combination of the Barkhausen-
noise analysis (BNA)5,6 and the metal-magnetic-memory
method (MMM).7,8 Based on different signal inception
depths, a combination of these methods can provide a
substantial financial reduction thanks to a faster response
and cheaper control of a potentially critical point
achievement. In addition, for a description of particular
surface stages (especially the level of residual stresses in
test samples before and after the loading), the mentioned
magnetic methods should be supplemented with the
results obtained with the X-ray diffraction technique.9,10
The principle of the BNA has been known for many
years and used in diagnostics of machined layers.11 The
MMM method is based on the magneto-elasticity and
magneto-mechanical effect theory.12,13 Due to the
Materiali in tehnologije / Materials and technology 51 (2017) 3, 437–441 437
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
UDK 620.1:67.017:620.17 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 51(3)437(2017)
self-magnetization phenomenon of ferromagnetic ma-
terials, during the operation of equipment, the magnetic
field is kept down when the work load is off. The
residual magnetic field is related to the mechanical
stress, that is, the residual magnetic field density Hp is
characterized by the polarity varying along the normal.
Thus, using the MMM testing instrument, the most
serious zones of a stress concentration (SZC), defects
and metal-structure non-uniformities can be found.14
Moreover, it does not require special magnetization
devices and the tested surface does not require a special
treatment. Therefore, the MMM technology has been
widely applied to inspect pressure vessels,15 piping16 and
other equipment in service. The first pilot results of a
fatigue-damage analysis obtained with the BNA and
MMM methods have been already reported.17,18
In the light of these facts, we performed the BNA and
MMM testing on the samples of the chosen unalloyed
construction steel, monitoring the future fatigue damage
even before the initialization of the magistral crack.
Subsequently, we compared the obtained results with
theoretical predictions and, finally, we discussed the
applicability of these methods.
2 SAMPLES UNDER INVESTIGATION
Experimental samples (Figure 1) were made from
the unalloyed construction steel S355J2G3 with a
chemical composition in mass fractions (w/%): 0.2C,
0.55Si, 1.6Mn, <0.035P and <0.035S. It is a commonly
used material for welded constructions, machine parts
and electrical-apparatus parts that are not strained by
high temperature. Samples were manufactured from
6-mm thick plates with the water-jet-machining method
to avoid a thermal influence and consequent deforma-
tion. The surfaces chosen for the analysis were then
pre-treated with metallographic papers in a range of
240-400-600-800-1200-2500.
In order to restrain the notch effect19, sample edges
were sharpened and, moreover, polished with a diamond
paste with the average grain size of 3 μm.
3 EXPERIMENTAL TECHNIQUES
3.1 XRD analysis
XRD measurements were performed with a PROTO
iXRD COMBO diffractometer in the -goniometer or
iso-inclination mode with Cr-K	 radiation. The line
{211} of the 	-Fe phase was measured with interplanar
lattice spacings computed from the maxima of Pearson
VII functions fitted to Cr-K	1 profiles after the Cr-K	2
stripping carried out with the Rachinger method. The
stresses were computed presuming the biaxial state of
macroscopic residual stresses (RS) using the Winholtz-
Cohen method and X-ray elastic constants ½s2 = 5.76
TPa–1, –s1 = 1.25 TPa–1.
3.2 Barkhausen noise analysis
Variations of microstructural changes during the
loading were also studied using the micro-magnetic
testing method.20 Magneto-elastic parameter mp was
chosen as the characteristic of the surface and subsurface
layers. This parameter was determined with the root-
mean-square method involving all the measured signal
amplitudes and it is given in volts. It was calculated for a
specified analysis frequency range. The measurements
were performed using a commercial unit Stresstech
MicroScan 600-1 magneto-elastic analyser with standard
sensors, the S1-138-1501 type for mp changes during the
fatigue test and the S1-171-25-11 type for the surface
characterization after the crack initialization. The main
parameters of the applied method had a sinusoidal shape
of the magnetic signal, a magnetic voltage of 3.5 V and a
frequency of 75 Hz. The obtained results are the mean
values from 10 measurements. The estimation of the
penetration depth  of the excitation signal depends on
the used frequency f and the analysed material (per-
meability μ and conductivity ), in Equation (1).21 In
practice, a typical expectable effective penetration depth
is in a range of up to 20 μm for this experimental
arrangement in Equation (1):


=
1
f
(1)
3.3 MMM method
A TSC-3M-12 instrument was used for the analysis.
Probe 1-8E with four sensors was employed.
K. KOLAØÍK et al.: USING THE BARKHAUSEN-NOISE ANALYSIS AND METAL-MAGNETIC-MEMORY METHOD ...
438 Materiali in tehnologije / Materials and technology 51 (2017) 3, 437–441
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 1: Shape and dimensions of the specimens for fatigue test
Slika 1: Oblika in dimenzije preizku{ancev za preizkus utrujanja
3.4 Fatigue loading
The bending fatigues tests were performed in a
SCHENCK PWYG machine. The loading frequency was
17 Hz. The load ratio was R = –1 and the loading force
was from 220 MPa to 370 MPa. Cycles were required to
initiate cracks.
4 RESULTS AND DISCUSSIONS
Results of the fatigue test after different loads of the
samples are given in Table 1; they are in accordance
with the theoretical presumptions.22
Table 1: Results of the fatigue test – the number of cycles to crack
initialization
Tabela 1: Rezultati preizku{anja utrujanja – {tevilo ciklov do nastanka
razpoke
Sample load (MPa) number of cycles
1 220 4.518.600without a crack
2 250 698.500
3 320 600.300
4 330 288.400
5 370 82.000
6 370 67.400
Variations in the magneto-elastic parameters mp at
three different loads are shown in Figure 2.
We can see a similar development of the mp-function
independently of the load forces. This behaviour should
be the subject of further study. Generally, the mp
development depends on the loading cycles reflecting
primarily the changes in the RS. Changes in the
mechanical properties during the loading are determined
by the movement of dislocations and their mutual
interactions. Dislocation movement is further influenced
by the presence of precipitates, foreign particles, grain
boundaries, etc. It is, therefore, clear that changes in the
configuration and density of dislocations and also
changes in the distribution and morphology of other
types of obstacles will occur during the cyclic deforma-
tion. A greater loading force leads to a higher density of
dislocations, 1010 to 1013 cm–2, resulting in an even more
significant drop of mp. Moreover, the changes in mp
reflect the responses of material properties, such as
hardening and softening, vacancy concentration and also
the mentioned formation of dislocations.
At the start, loading cycles result in a fluctuation of
dislocation density, which can be manifested in the
changes in the RS as well as in the cyclic softening and
hardening of the material.23 The latter effects influence
the domain wall movement, which leads to increases and
decreases of mp for all the loading forces during the
K. KOLAØÍK et al.: USING THE BARKHAUSEN-NOISE ANALYSIS AND METAL-MAGNETIC-MEMORY METHOD ...
Materiali in tehnologije / Materials and technology 51 (2017) 3, 437–441 439
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 3: Fatigue crack on the surface of sample 4, the distance
between lines is 10 mm
Slika 3: Utrujenostna razpoka na povr{ini vzorca 4, razdalja med
linijami je 10 mm
Figure 2: Changes in Barkhausen noise, i.e., parameter mp, measured
in the middle of a sample (Figure 1), marked with "X" symbol, in the
direction parallel to the loading axis as a function of the loading cycles
of the tested specimens
Slika 2: Spreminjanje Barkausnovega hrupa, to je parameter mp,
merjen na sredini vzorca (Slika 1) ozna~eno z "X" simbolom, v smeri
vzporedno z osjo obremenjevanja, v odvisnosti od ciklov obremenje-
vanja preizku{ancev
Figure 4: Gradient of mp obtained from the area of crack determined
with BNA
Slika 4: Gradient mp dobljen iz podro~ja razpoke, dolo~ene z BNA
fatigue test. At the end of the fatigue life, mp drops
rapidly because of the crack growth (Figures 3 and 4).
Figure 4 shows the mp measured after the cycle
loading around the crack area. Because of a high level of
plastic deformation and the crack acting as a dielectric,
inhomogeneities of the magnetic field and changes in the
domain structure of the crack are formed24, which lead to
a decrease of mp. From Figure 4, it is evident that a
higher loading force results in larger deformation zones,
a larger amount of cracks and, consequently, in a lower
value of mp. Because of the crack initialization, the RS
starts to relax and it reaches its minimum if the cracks
are fully developed, as can be seen in Figure 5.
During the cycling loading, dislocations move in the
material and can gather, which leads to an accumulation
of stress in the given areas. Increasing the stress
accumulation results in weak changes of the magnetic
field.18 The final state of this process is an initialization
of cracks. These conclusions are consistent with the
results presented in Figures 6 and 7, where magneto-
grams of the samples without and with a crack are
presented, respectively. In addition, in Figure 7, a high
value of the magnetic-field gradient (dHp/dx) is evident,
which corresponds well with the fatigue-crack presence
(Figure 3). The physical nature of the gradual dHp/dx
increase during the cycling of the analysed surface is
related to the secondary domain structure of the ferro-
magnetic sample. This is produced in the vicinity of
defects in the lattice. The increase in the dHp/dx
amplitude is associated with an increase in the density of
the closing domains compensating the dispersive
magnetic field in the vicinity of the accumulating defects
during the cycling. The sharp increase in dHp/dx during
the cracking is related to the creation of the closing
(secondary) domains, which compensate the dispersive
field of the new surface, minimizing the magnetostatic
energy of the magnetic sample.
5 CONCLUSIONS
The results of the Barkhausen-noise analysis and
metal-magnetic-memory testing correspond well with
the established theoretical predictions. Moreover, it was
found that
• dependence of the magneto-elastic parameter mp on
the loading cycles is not monotonous. Nevertheless,
mainly before the cracking, it oscillates (Figure 2);
• cracks tend towards an inhomogeneous magnetic
field of the material (due to the interruptions of the
magnetic line of the force), which leads to an mp re-
duction (Figure 4). The presence of the cracks causes
a relaxation of residual stresses (Figure 5).
• using the metal-magnetic-memory method, it should
be possible to analyse fatigue damage before the
initialization of the magistral crack (Figures 6 and
7). On the other hand, it is necessary to know the
history of the material.
K. KOLAØÍK et al.: USING THE BARKHAUSEN-NOISE ANALYSIS AND METAL-MAGNETIC-MEMORY METHOD ...
440 Materiali in tehnologije / Materials and technology 51 (2017) 3, 437–441
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 6: Magnetogram obtained with the MMM method from the
surface of sample 1 after the cyclic loading. The lower lines (dHp/dx)
are the partial derivation of the measured upper lines (Hp)
Slika 6: Magnetogram, dobljen z MMM metodo iz povr{ine vzorca 1
po cikli~nem obremenjevanju. Spodnje linije (dHp/dx) so parcialna
derivacija izmerjenih zgornjih linij (HP)
Figure 7: Magnetogram obtained with the MMM method from the
surface of sample 4 after the cyclic loading; the lower lines (dHp/dx)
are the partial derivation of the measured upper lines (Hp)
Slika 7: Magnetogram, dobljen z MMM metodo iz povr{ine vzorca 4
po cikli~nem obremenjevanju, spodnje linije (dHp/dx) so parcialna
derivacija izmerjenih zgornjih linij (Hp)
Figure 5: Gradients of RS and mp from the area of crack (Figure 3)
determined with XRD and BNA
Slika 5: Gradient RS in mp iz podro~ja razpoke (Slika 3) dolo~ena z
XRD in BNA
To summarize, we can conclude that the metal-mag-
netic-memory technique and Barkhausen-noise analysis
are advantageous in certain circumstances, for example,
for an in-service inspection of structures with complex
geometries.
Acknowledgments
This paper draws on the outcomes of the SGS – 2015
– 016 project: Analysis of Surfaces of Assemblies and
Tools by Surface Integrity Method and Impacts on
End-Use Properties.
6 REFERENCES
1 J. P. Davim, Surface integrity in machining, Springer, London 2010,
doi:10.1007/978-1-84882-874-2
2 J. Schijve, Fatigue of structures and materials, Kluwer Academic
Publisher, Dordrecht 2001, 513, doi:10.1007/0-306-48396-3
3 R. G. Govindaraju, A. Strom, D. C. Jiles, S. B. Biner, Evaluation of
low-cycle fatigue damage in steel structural components by magnetic
measurement technique, Review of Progress in Quantitative
Nondestructive Evaluation, New York, 1993, 1839–1846,
doi:10.1007/978-1-4615-2848-7_236
4 M. Witos, M. Zokowski, Passive Magnetic Observer in NDE & SHM
Applications, EWSHM-7th European Workshop on Structural Health
Monitoring, France, 2014, 213–220
5 Y. Tomita, K. Hashimoto, N. Osawa, Nondestructive estimation of
fatigue damage for steel by Barkhausen noise analysis, NDT & E
International, 29 (1996) 5, 275–280, doi:10.1016/S0963-8695(96)
00030-8
6 K. Tiitto, Use of Barkhausen noise in fatigue, Nondestructive Testing
and Evaluation, 5 (1989) 1, 27–37, doi:10.1080/02780898908952952
7 V. T. Vlasov, A. A. Dubov, Development of theory of ferromagnetic
domain structure by the example of iron, Proceedings of the third
international conference Diagnostics of equipment and structures
with usage of metal magnetic memory, Moscow, 2003
8 A. A. Dubov, S. M. Kolokolnikov, Method of metal magnetic
memory (MMM) and inspection devices: Training handbook, Tisso
Co, Moscow 2003, 320
9 Q. Xin, D. Shu, L. Hui, W. Wei, J. Chen, Magnetic Barkhausen
noise, metal magnetic memory testing and estimation of the ship
plate welded structure stress, Journal of Nondestructive Evaluation,
31 (2012) 1, 80–89, doi:10.1007/s10921-011-0123-7
10 B. Hu, G. Gan, S. Gongtian, L. Luming, C. Xing, Study on Magnetic
Memory Method (MMM) for Fatigue Evaluation, Proc. of 17th
World Conference on Nondestructive Testing, China, 2008, 25–28
11 J. ^í`ek, M. Neslu{an, M. ^illiková, A. Mi~ietová, O. Melikhova,
Modification of steel surfaces induced by turning: non-destructive
characterization using Barkhausen noise and positron annihilation,
Applied Physics, 47 (2014) 44, 445301
12 A. A. Dubov, A study of metal properties using the method of
magnetic memory, Metal Science and Heat Treatment, 39 (1997) 9,
401–405, doi:10.1007/BF02469065
13 A. A. Dubov, Principle features of metal magnetic memory method
and inspection tools as compared to known magnetic NDT methods,
CINDE Journal, 27 (2006) 3, 16–20
14 X. Hai Yan, X. M. Qiang, Y. Zhijun, Z. Lihong, Stress State Analysis
of Failure Blade with MMM Method, Daqing Petroleum Institute
China, 2006
15 O. Stupakov, T. Toshiyuki, K. Kolaøík, Barkhausen Noise Testing of
Residual Stresses Introduced by Surface Hardening Techniques,
Electromagnetic Nondestructive Evaluation (XVII), 39 (2014),
288–295, doi:10.3233/978-1-61499-407-7-288
16 J. W. Wilson, G. Y. Tian, S. Barrans, Residual magnetic field sensing
for stress measurement, Sensors and Actuators A: Physical, 135
(2007) 2, 381–387, doi:10.1016/j.sna.2006.08.010
17 G. Pluvinage, Notch effect in fatigue and fracture, Springer,
Netherlands 2001, 22, doi:10.1007/978-94-010-0880-8_1
18 W. A. Theiner, Physical Basis of Micromagnetic Methods and Sensor
Systems and their Application Areas, Proceedings of the 1st
International Conference on Barkhausen noise and Micromagnetic
Testing, Hannover, 1998, 197–218
19 W. A. Theiner, E. Brinksmeier, E. Stücker, Stress Measurements on
Components with Nondestructive Ferromagnetic Methods, Residual
Stresses in Science and Technology, 1 (1987), 167–174
20 M. Lindgren, T. Lepistö, Effect of cyclic deformation on Barkhausen
noise in a mild steel, NDT & E International, 36 (2003) 6, 401–409,
doi:10.1016/S0963-8695(03)00031-8
21 S. Tiitto, R.Säynäjäkangas, Spectral damping in Barkhausen noise,
IEEE Transactions on Magnetics, 11 (1975) 6, 1666–1672,
doi:10.1109/TMAG.1975.1058960
22 W. Guan, P. Guo, C. Xuedong, Metal Magnetic Memory Testing
Technique for Typical Defects in Pressure Vessels and Pipelines,
ASME 2015 Pressure Vessels and Piping Conference, American
Society of Mechanical Engineers, 2015, doi:10.1115/PVP2015-
45826
23 L. Faturík, L. Tr{ko, S. Hr~ek, O. Bokùvka, Comparison of Struc-
tural Design in High and Ultra-High Cycle Fatigue Regions,
Transactions of FAMENA, 38 (2015) 4, 1–12
24 M. Neslu{an, J. Majerík, P. Kejzlar, M. ^illiková, A. Mi~ietová,
Barkhausen Noise Emission in Hard Milled Surfaces of Steel C55,
Transactions of FAMENA, 39 (2016) 4, 55–64
K. KOLAØÍK et al.: USING THE BARKHAUSEN-NOISE ANALYSIS AND METAL-MAGNETIC-MEMORY METHOD ...
Materiali in tehnologije / Materials and technology 51 (2017) 3, 437–441 441
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS