Article citation info: 51

Dybaáa J, Nadulicz KM. Diagnostics of welded joints using passive and active magnetic methods. Diagnostyka. 2017;18(4):51-59

ISSN1641Ͳ6414
DIAGNOSTYKA,2017,Vol.18,No.4
eͲISSN2449Ͳ5220

DIAGNOSTICS OF WELDED JOINTS

USING PASSIVE AND ACTIVE MAGNETIC METHODS
Jacek DYBAàA1, Krzysztof Marcin NADULICZ2
1
Institute of Vehicles, Warsaw University of Technology, Narbutta 84, 02-524 Warsaw, Poland,
[email protected]
2
Laboratory of Non-Destructive Testing, Military Institute of Armament Technology, WyszyĔskiego 7,
05-220 Zielonka, Poland, [email protected]

Abstract
As the quality of welded joints transfers to the load carrying capacity of structures, welded joints
naturally become the object of diagnostic tests. Welded joints are mainly tested in order to find out the places
with structural or mechanical non-homogeneities. It is caused by the fact that the existence of welding
imperfections significantly decreases mechanical properties of the welded joint and especially its fatigue
strength.
The paper presents some results of diagnostic tests for welded joints of steel components carried out
with the use of passive and active magnetic methods. Faults of the welded joint occurring within a tested
object cause the phenomenon of magnetic anomalies existing over a tested surface and this phenomenon has
been used to detect welding imperfections. A specially designed measurement matrix consisting of 768 one-
axial miniature magnetic field sensors has been used to record the magnetic flux leakage. The paper includes
a comparison of the effectiveness of diagnosing using the passive and active magnetic method. Moreover, it
was shown that the magnetic diagnostics provides rapid inspection of quality of welded joints with no need
for special preparation of joints for testing which is relevant for its practical applications and results in low
costs of such tests.

Keywords: magnetic diagnostics, passive methods, active methods, welded joints

DIAGNOSTYKA POàĄCZEē SPAWANYCH Z WYKORZYSTANIEM

PASYWNYCH I AKTYWNYCH METOD MAGNETYCZNYCH

Streszczenie
PoniewaĪ jakoĞü záączy spawanych przekáada siĊ na moĪliwoĞci przenoszenia obciąĪeĔ przez
konstrukcje, poáączenia spawane stają siĊ w naturalny sposób obiektem badaĔ diagnostycznych. Gáównym
celem badaĔ poáączeĔ spawanych jest wykrywanie miejsc, w których wystĊpuje niejednorodnoĞü strukturalna
lub mechaniczna záącza. Taki cel badaĔ wynika z faktu, Īe wystĊpowanie wad spawalniczych w
decydującym stopniu pogarsza wáaĞciwoĞci mechaniczne poáączenia spawanego, a w szczególnoĞci jego
wytrzymaáoĞü zmĊczeniową.
W niniejszej pracy przedstawiono wyniki badaĔ diagnostycznych poáączeĔ spawanych stalowych
elementów z wykorzystaniem pasywnych i aktywnych metod magnetycznych. Do wykrywania defektów
poáączenia spawanego wykorzystano zjawisko powstawania anomalii magnetycznych nad badaną
powierzchnią wywoáanych wystĊpowaniem wad spawalniczych w diagnozowanym elemencie. Zaburzenia
strumienia magnetycznego byáy rejestrowane za pomocą specjalnie skonstruowanej matrycy pomiarowej
skáadającej siĊ z 768 jednoosiowych miniaturowych czujników pola magnetycznego. W pracy dokonano
porównania skutecznoĞci diagnozowania pasywną i aktywną metodą magnetyczną. Wykazano równieĪ, Īe
diagnostyka magnetyczna umoĪliwia wykonanie szybkiej kontroli jakoĞci poáączeĔ spawanych bez
koniecznoĞci specjalnego przygotowania poáączeĔ do badaĔ diagnostycznych, co ma duĪe znaczenie
uĪytkowe i przekáada siĊ bezpoĞrednio na niski koszt takich badaĔ.

Sáowa kluczowe: diagnostyka magnetyczna, metody pasywne, metody aktywne, poáączenia spawane

1. INTRODUCTION Making welded joints is associated with melting,

subsequent solidification, and also introducing heat
Welded joins are widely used to connect steel energy (heat input) into the welded material which
components. The common use of welded joints is changes its structure in the Heat Affected Zone.
caused by significant advantages of this method of Welding often leads to structural inhomogeneity
joining represents as they can be made easy and fast (gas bubbles, metallic or non-metallic inclusions)
and provide strong connections and possibilities for and mechanical inhomogeneity (cracks and
complete automation of component joining process. delaminations) of the joint. Hot cracks occur during
52 DIAGNOSTYKA, Vol. 18, No. 4 (2017)
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

crystallising of weld metal, and usually such cracks Level of the magnetic flux leakage depends on
are formed in the middle line of the weld. The cold the inhomogeneity localisation in the material and
cracks happen at the final phase of the weld cooling its orientation relative to the magnetic field flux
or after some time when the welding was finished passing through the material (Fig. 2). Magnetic flux
and they usually exist on the edge of the weld leakage is the strongest in the case of transverse
within heat affected zone. In addition to cracks inhomogeneities located on the tested surface or at
resulting from welding processes, welded joints are a small depth under this surface.
exposed to the appearance of fatigue cracks a) c)
resulting from the mechanical loads of the joint
during operation [14]. As the quality of welded
joints transfers to the load carrying capacity of
structures, welded joints naturally become the
object of diagnostic tests. Welded joints are mainly
tested in order to find out the places with structural
or mechanical inhomogeneities. It is caused by the b) d)

fact that the existence of welding imperfections

significantly decreases mechanical properties of the
welded joint and especially its fatigue strength.
Contemporary technical diagnostics is focused
most of all on the development of non-invasive
methods which allow to determine the technical Fig. 2. Influence of material’s inhomogeneity location
condition of diagnosed objects without any and orientation on the magnetic flux leakage effect:
disassembling, turning off or even disturbing the a) transverse surface discontinuity,
functioning. Non-invasive diagnostic tests are based b) transverse subsurface discontinuity,
c) transverse internal discontinuity,
on assessment of quality of a tested object while
d) longitudinal internal discontinuity [10]
preserving its original properties and without its
destruction. Non-destructive methods of detecting
There are different ways used for detection of
heterogeneity in welded joints include primarily
the magnetic flux leakage outside of tested element.
visual, penetrating, X-ray, and ultrasonic methods.
In a magnetic particle method, the magnetic flux
The quality of welded joints for steel components
leakage is displayed by using ferromagnetic powder
may be also assessed using magnetic methods [1,
[11]. The disadvantage of such approach is caused
12].
by a need for accurate cleaning and preparing the
Magnetic diagnostic methods may be divided
tested surface which requires increased time and
into passive and active ones. The use of external
financial resources. In the magnetographic method
magnetic field acting against a tested component is
magnetic flux leakage is recorded by means of a
a common feature of active magnetic diagnostic
special tape that contains small magnetisable
methods and the evaluation of object’s technical
ferromagnetic particles. The magnetic tape contacts
condition is made by analysing recorded magnetic
the surface of the tested component and if, due to
parameters. The phenomenon of anomalies existing
the existence of inhomogeneity in the material,
in distributions of magnetic field intensity
magnetic flux leakage occurs on the material
components over a tested surface is used for
surface, the tape will be magnetised. The image
magnetic detection of inhomogeneities within a
fixed on the tape can be read by a special head and
tested object [8, 9]. The magnetic field flux passing
displayed as a distribution of the magnetic field. In
through the material is dispersed in a place of
this way, the surface magnetic phenomena
inhomogeneity which produces magnetic flux
connected with the existence of inhomogeneities in
leakage outside the tested object because on the
the tested element can be identified [2, 4]. The
edges of discontinuity the specific “magnetic poles”
magnetic flux leakage may be also measured by
are created, which in turn generates magnetic field
using magnetic field sensors. The measurements of
above this discontinuity (Fig. 1).
magnetic field intensity above the surface of a
magnet tested element can detect local magnetic anomalies
existing above the surface of material with
structural or mechanical inhomogeneities.
H Passive magnetic diagnostic methods deploy an
analysis of the self-magnetic flux leakage
distribution on the surface of the tested
ferromagnetic material. The self-magnetic flux
leakage is a magnetic field appearing on the surface
of an element made of ferromagnetic material under
Fig. 1. Magnetic flux leakage at the location the influence of the working loads or residual
of a crack stress, and as a result of the inhomogeneities and
 DIAGNOSTYKA, Vol. 18, No. 4 (2017) 53
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

discontinuities of the ferromagnetic material for diagnostic tests of welded joints. The welded
structure. joints were made using different methods of
Local magnetic anomalies existing on welding:
ferromagnetic surface in the zones of stress a) 111 - MMA welding (Manual Arc Welding –
concentration are produced by changed magnetic welding with coated electrode);
parameters of areas where the stresses are b) 135 - MAG welding (Metal Active Gas – arc
concentrated [5, 6]. On the other hand, welding by metal electrode in chemical active
inhomogeneities within ferromagnetic material gas shield);
structure cause the dispersion of the weak magnetic c) 136 - MAG welding (Metal Active Gas – arc
field flux, which is the magnetic field of the Earth, welding by powder wire in chemical active gas
passing through the tested element. Existing self- shield);
magnetic flux leakages outside the tested element d) 141 – TIG welding (Tungsten Inert Gas – arc
also cause the formation of local surface magnetic welding by non-melting electrode in inert gas
anomalies. shield);
A method of metal magnetic memory (MMM e) 311 – Acetylene-oxygen welding (Gas).
method) may be distinguished among passive The welded joints were made in two variants as
diagnostic methods [3, 16]. An interpretation of the reference joints (properly made, without faults –
changes in values of derivative of a selected reference samples) and faulty joints (with an
component of magnetic field intensity measured at introduced fault – slag, air, and non-ferromagnetic
the surface of the tested object has a principal inclusions – faulty samples). Fig. 3 and 4 show
meaning for the methodology was developed by exemplary reference samples of welded joints for
Dubov and formulated in the ISO 24497 standard diagnostic tests. For comparative purposes, a
[17]. The analysis of changes in values of homogeneous sample was also made, without
derivative of component of magnetic field intensity welded joints.
can be used to detect the zones of stress
concentration and inhomogeneities in the tested
object. The MMM method can be used for
diagnostic tests of steel components and welded
joints [7, 13]. The main difficulty using passive
magnetic diagnostic methods is caused by the fact
that the diagnostic process is based on the
measurements of weak magnetic fields and the Fig. 3. The welded joint made by MAG
results of measurements are burdened with many method
external factors and internal material properties
[15].
This paper presents results of diagnostic tests
for welded joints of steel components using passive
and active magnetic methods. The effect of
magnetic anomalies appearing above the tested
surface and caused by the presence of
inhomogeneities within a tested component was
used to detect the faults of a welded joint. The Fig. 4. The welded joint made by TIG method
magnetic flux leakage was recorded by means of a
specially designed measurement matrix consisting 3. MEASUREMENT MATRIX
of 768 one-axial miniature magnetic field sensors.
A comparison of the effectiveness of diagnosing Magnetic flux leakage was recorded by the
using the passive and active magnetic method has specially designed matrix consisting of 768 one-
been made. It was also proved that the magnetic axial miniature sensors produced by Allegro
diagnostics can be used for rapid examination of the MicroSystems with the measurement range
quality of welded joints without any special between 0 and 5 Gauss, whose principle of
preparations of joints for diagnostic tests, which is operation is based on the Hall effect. The
relevant for its practical applications and results in measurement matrix has the dimensions 125 x 110
low costs of such tests. mm and includes 256 three-dimensional
measurement points (16 columns and 16 lines) (Fig.
2. TESTED OBJECTS 5, 6). Each measurement point consists of three
mutually perpendicular sensors measuring
A few samples with dimensions of 125 x 100 x independently three orthogonal components of the
6 mm (length x width x thickness) made of steel flat magnetic field:
bar (steel ST3) by EnergomontaĪ Póánoc Technika
Spawalnicza i Laboratorium Sp. z o. o. were used
54 DIAGNOSTYKA, Vol. 18, No. 4 (2017)
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

a) Hx – component perpendicular to the tested sensor’s output signal. This preprocessing

weld and tangential to the surface of the simplified the interpretation of the sensor's
sample; indications, because then the sensor’s zero output
b) Hy – component vertical to the surface of the voltage signaled the absence of magnetic field.
sample;
c) Hz – component parallel to the tested weld and
tangential to the surface of the sample.

Fig. 6. A view of the back side of the

Fig. 5. A view of the front side of the measurement matrix
measurement matrix

The matrix power consumption was limited. A

sequential power supply switching for particular
columns (48 sensors simultaneously) was applied.
In order to reduce the impact of applied power
switching for columns of sensors (multiplication)
the signals from particular sensors were initially
amplified in acoustic low-passing band filter
limiting the signal band to 300 Hz. In the next step,
the signal was amplified again after de-
multiplication and before sending the measurement
signals to A/C converter of the microcontroller. The
raw measurement signals were amplified by 200
times in total.
An initial calibration of sensors was carried out
before each measurement cycle because a spread of Fig. 7. The touchscreen monitor
parameters of sensors was observed at testing. The
matrix was calibrated by sending into the input of 4. DIAGNOSTIC TEST OF WELDED JOINTS
each sensor a specific value of reference voltage USING THE PASSIVE MAGNETIC
from the converter of Raspberry 3 controller to get METHOD
the output level of 2.5 V at 1.0 mV accuracy
corresponding to zero level of magnetic field In the first stage of the experiment the samples
intensity. were tested using the passive magnetic method.
Selection of individual sensors, controlling the Therefore, the tested elements were not subjected to
power supply, measurements of received signals, as the action of an external artificial magnetic field
well as their registration, were all performed using and only the Earth's magnetic field influenced the
the Raspberry 3 microcomputer. A 7” touchscreen elements. Fig. 8 shows a set of measurement
monitor was used to display preliminary test results instruments used for testing.
(Fig. 7). The values of output voltages of
magnetometers were recorded on the microSD
memory card and then displayed on a colour scale.
Because the value of the sensor's output voltage,
proportional to the intensity of the measured
magnetic field, ranged from 0 to 5 V, the analysis
was preceded by subtraction of 2.5 V from the
 DIAGNOSTYKA, Vol. 18, No. 4 (2017) 55
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

Fig. 8. Measurement instruments used for Fig. 11. Distribution of Hx component of the
passive magnetic tests magnetic field on the surface of the reference
sample 136/02
The samples were X-rayed before starting the
magnetic tests. Exemplary X-ray images of samples
are presented in Fig. 9 and 10.

Fig. 9. X-ray image of the reference sample

136/02
Fig. 12. Distribution of Hy component of the
magnetic field on the surface of the reference
sample 136/02

Fig. 10. X-ray image of the faulty sample

136/01

The measurement sensors were calibrated

before starting each cycle of magnetic tests. The
reference samples (without any faults) were tested
three times. The results of measurements were
averaged before recording on the memory card. In Fig. 13. Distribution of Hx component of the
the next step, the magnetic field was measured in magnetic field on the surface of the faulty
the same way on the surface of the faulty samples. sample 136/01
Fig. 11-14 show some exemplary
magnetographs made at this stage of the
experiment.
56 DIAGNOSTYKA, Vol. 18, No. 4 (2017)
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

magnetograms were taken and the

defectoscope magnet coil was powered by
current of 3 A. Fig. 17 shows the set of
instruments used at testing.

Fig. 14. Distribution of Hy component of the magnetic

field on the surface of the faulty sample 136/01

Magnetograms of the sample surfaces

performed by the passive magnetic method show
that the intensity of measured magnetic field was Fig. 16. Distribution of Hy component of the
very low and in practice constant on the whole magnetic field on the surface of the
surface. Therefore, it is difficult to use them for homogeneous sample (without welded joint)
assessing the quality of performed welded joints, as
the joint is not detected on the magnetograms.
Magnetograms of samples with welded joints
actually cannot be distinguished from the
magnetogram of the homogeneous sample without
any welded joint (Fig. 15 and 16).

Fig. 17. Instruments used for active magnetic

Fig. 15. Distribution of Hx component of the tests
magnetic field on the surface of the
homogeneous sample (without welded joint)
The samples were X-rayed before starting the
5. DIAGNOSTIC TEST OF WELDED JOINTS magnetic tests. Exemplary X-ray images of samples
USING THE ACTIVE MAGNETIC are presented in Fig. 18 and 19.
METHOD

In the second stage of the experiment the

samples were tested using the active magnetic
method. Tested components were subjected to
the action of an external artificial magnetic
field generated by the defectoscope DMS-11
of the TechPan company that was powered
with the stabilised power source P335
providing continuous regulation of voltage to
35 V and current to 6 A. Magnetisation of Fig. 18. X-ray image of the reference sample
samples was carried out directly before the 141/02
 DIAGNOSTYKA, Vol. 18, No. 4 (2017) 57
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

Fig. 19. X-ray image of the faulty sample

141/01

The measurement sensors were calibrated Fig. 22. Distribution of Hx component of the
before starting each cycle of magnetic tests. The magnetic field on the surface of the faulty
reference samples (without any faults) were tested sample 141/01
three times. The results of measurements were
averaged before recording on the memory card. In
the next step, the magnetic field was measured in
the same way on the surface of the faulty samples.
Fig. 20-23 show some exemplary
magnetograms made at this stage of the experiment.

Fig. 23. Distribution of Hy component of the

magnetic field on the surface of the faulty
sample 141/01

Magnetograms of the surfaces of samples taken

by the active magnetic method indicate that the
Fig. 20. Distribution of Hx component of the intensity of magnetic field changes significantly on
magnetic field on the surface of the reference the surface of samples.
sample 141/02 Component Hx of the magnetic field
(perpendicular to the tested weld and tangential to
the surface of the sample) has reached the greatest
value above the root of the weld. In the case of a
faulty sample the area of maximal values of Hx
component of the magnetic field was greater than
for the reference sample. The mere existence of a
distinct maximum of Hx component of the magnetic
field informs about the inhomogeneity of the
sample caused by the presence of the weld which
may be confirmed by the comparison of
magnetograms for the welded sample and the
homogeneous sample without any welded joint
(Fig. 24).
Component Hy of the magnetic field (vertical to
the surface of the sample) changes its sense above
Fig. 21. Distribution of Hy component of the the root of the weld. In the case of the faulty sample
magnetic field on the surface of the reference the change of the sense of component Hx of the
sample 141/02 magnetic field was sharper than for the reference
sample. In the lack of any inhomogeneity caused by
the presence of the weld the sense of component Hx
58 DIAGNOSTYKA, Vol. 18, No. 4 (2017)
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

of the magnetic field changes more smoothly and inspection of the originality of identification
steadily which may be observed on the marking of cars.
magnetogram of the homogeneous sample without It is also possible to assess the quality of the
any welded joint (Fig. 25). weld workmanship or the technical condition of the
used weld. The evaluation has a rough character of
course as the value of magnetic flux leakage can
provide only general information on the size of the
inhomogeneity. On the other hand, the magnetic
tests are rapid (short time for receiving the
magnetograms and lack of special preparations of
the tested surface) and, as a consequence, not
expensive. The active magnetic method may be
used as a “screening” method for detection of
defects in welded joints.
It is worth to notice that magnetograms may be
used in diagnostics of investigated surfaces not only
for the welded joints but also for the detection of
other types of inhomogeneities in tested
components caused e.g. by forging cracks, fatigue
Fig. 24. Distribution of Hx component of the cracks, hardening cracks, grinding cracks, tearing
magnetic field on the surface of the and dents, and corrosive losses. Magnetic method
homogeneous sample (without the welded may be used to test elements made of ferromagnetic
joint) material: ferrite steels, cast iron and cast steel. It is
possible to detect inhomogeneities and
discontinuities on the material surface and
relatively large subsurface faults, but located close
to the tested surface.
In the future, the research work will be focused
on developing a matrix consisting of AMR
(Anisotropic MagnetoResistance) magnetometers
which use the anomaly of the magnetoresistance
phenomenon. Building the matrix of sensors is
planned with changeable dimensions and curved
surfaces to carry out rapid diagnostics for welded
joints of rocket propellant burning chambers and
tanks storing the compressed air.

SOURCE OF FUNDING
Fig. 25. Distribution of Hy component of the
magnetic field on the surface of the The paper has been prepared basing on tests
homogeneous sample (without the welded carried out under the project No DOB-
joint)
BIO6/21/8/2014 financed by the National Centre of
Research and Development (NCBiR) in the
6. CONCLUSIONS Competition No 6/2014 for performance of projects
on scientific research and development work for
Results indicate the benefits of the active state security and defence.
magnetic method. Magnetograms received by
means of the passive magnetic method have proved REFERENCES
to be useless in practice whereas the magnetograms
taken by the active magnetic method provided 1. Augustyniak B. About the causes of magnetic
relevant information about investigated surfaces. properties of steel and its use in non-destructive
Only the component Hz of the magnetic field testing. Zakopane, XIV Seminar on Non-destructive
(parallel to the tested weld and tangential to the Material Testing, 04.03y07.03.2008. (Polish)
surface of the sample) was useless. The analysis of 2. Czuchryj J, WnĊk T. Non-destructive testing of
distribution of two remaining components of the welded joints. II-nd Edition Corrected, Gliwice,
magnetic field above the tested surface can be Instytut Spawalnictwa, 1998. (Polish)
useful at detection of material inhomogeneity and 3. Dubov A. Principal features of metal magnetic
memory method and inspection tools as compared to
evaluation of its extension. It creates a possibility
known magnetic NDT methods. Montreal, 16th
for detection of a “camouflaged” weld (e.g. ground World Conference on Nondestructive Testing,
and painted) which may be significant regarding WCNDT 2004, 30.08y3.09.2004.
 DIAGNOSTYKA, Vol. 18, No. 4 (2017) 59
DYBAàA J, NADULICZ KM: Diagnostics of welded joints using passive and active magnetic methods

http://www.ndt.net/article/wcndt2004/html/htmltxt/35 Jacek DYBAàA, Ph.D., D.Sc.

9_dubov.htm Eng. – graduate of the Faculty of
4. Drop M, Litwinowicz L, Wáoczyk E. Material Automotive and Construction
diagnostics of energy equipment components and Machinery Engineering, Warsaw
prediction of durability. Part II. Warszawa, Biuro University of Technology (1988)
Gamma, 2011. (Polish) and the Institute of Organization
5. Jakubovics JP. Magnetism and magnetic materials. of Production Systems, Warsaw
London, Institute of Materials, 1994. University of Technology
6. Kaleta J. SMART magnetic materials. Design, (1990). He received Ph.D. and
manufacture, property testing, application. Wrocáaw, Habilitation Doctorate (D.Sc.)
Oficyna Wydawnicza Politechniki Wrocáawskiej, degrees, all in mechanical
2013. (Polish) engineering (machine design and
7. Kolokolnikov S, Dubov A, Steklov O. Assessment of maintenance), from the Warsaw
welded joints stress–strain state inhomogeneity University of Technology in
before and after post weld heat treatment based on the 2000 and 2010, respectively. He
metal magnetic memory method. Welding in the is an associate professor at the
World, 2016, vol. 60, issue 4, pp. 665-672. Laboratory of Mechatronics,
https://doi.org/10.1007/s40194-016-0335-7 Diagnostics and Technical
8. LewiĔska-Romicka A. Magnetic Tests. Volume I Safety, Institute of Vehicles,
Warszawa, Biuro Gamma, 1998. (Polish) Warsaw University of
9. LewiĔska-Romicka A. Magnetic Tests. Volume II Technology.
Warszawa, Biuro Gamma, 1998. (Polish) His major research interests are
10. Lipka M. Magnetic-powder testing. Gliwice, focused on diagnostic systems,
Politechnika ĝląska w Gliwicach, 2004. (Polish) machine condition monitoring
http://www.newimiue.ipnet.pl/dokumenty/proszek.pdf and diagnostics, non-destructive
11. Pashagin AI, Benklevskaya NP. Magnetic-powder testing, vibroacoustic
testing of components using magnetic indicator diagnostics, signal processing
packages. Russian Journal of Nondestructive Testing, and analysis, data selection and
2000, vol. 36, issue 9, pp. 640–646 data mining; pattern recognition,
12. Piech T. Magnetic Tests Warszawa, Biuro Gamma, neural networks and genetic
1998. (Polish) algorithms.
13. Roskosz M. Metal magnetic memory testing of Member of the Polish Society of
welded joints of ferritic and austenitic steels. NDT & Technical Diagnostics.
E International, 2011, vol. 44, issue 3, pp. 305-310.
https://doi.org/10.1016/j.ndteint.2011.01.008 Capt Krzysztof Marcin
14. Totten G, Howes M., Inoue T. (Eds.) Handbook of NADULICZ MSc Eng. –
Residual Stress and Deformation of Steel. Ohio, graduate of Faculty of Arms and
ASM International, 2002. Aviation at the Military
15. Wac-Wáodarczyk A. Magnetic materials. Modeling University of Technology in
and application. Lublin, Politechnika Lubelska, 2012. Warsaw (1996). He works as an
(Polish) assistant at the Military Institute
16. Wáasow WT, Dubow A. The physical basis of the of Armament Technology in
Metal Magnetic Memory method. Warszawa, Zielonka in Laboratory of Non-
Energodiagnostyka, 2008. (Polish) Destructive Testing. His main
17. BS ISO 24497-2:2007, Nondestructive Testing - research interest is non-
Metal Magnetic Memory - Part 2: General destructive testing of welded
Requirements. joints in rocket equipment such
as engines, combustion chambers
Received 2017-08-05 and compressed air tanks by
Accepted 2017-11-03 using passive and active
Available online 2017-11-06 magnetic methods.