VSEBINA – CONTENTS
IZVIRNI ZNANSTVENI ^LANKI – ORIGINAL SCIENTIFIC ARTICLES
A fatigue characterization of honeycomb sandwich panels with a defect
Utrujenostna karakterizacija satastih sendvi~nih panelov z napako
B. Keskes, Y. Menger, A. Abbadi, J. Gilgert, N. Bouaouadja, Z. Azari . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Fatigue properties of a high-strength-steel welded joint
Utrujenostne lastnosti zvara visokotrdnega jekla
Z. Burzi}, V. Grabulov, S. Sedmak, A. Sedmak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Mehanske lastnosti zvara iz jekla maraging po izlo~evalnem `arjenju
Mechanical properties of maraging steel welds after aging heat treatment
D. Klob~ar, J. Tu{ek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
An experimental verification of numerical models for the fracture and fatigue of welded structures
Eksperimentalna verifikacija numeri^nih modelov za prelom in utrujenost zvarjenih struktur
S. Sedmak, A. Sedmak, M. Arsi}, J. Vojvodi~ Tuma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Izra~un parametrov Weibullove porazdelitve za oceno upogibne trdnosti valovitih stre{nih plo{~
Computation of the parameters of the Weibull distribution for estimating the bending strength of corrugated roofing sheets
M. Ambro`i~, K. Vidovi~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
STROKOVNI ^LANKI – PROFESSIONAL ARTICLES
Investigation of the influence of the melt slag regime in a ladle furnace on the cleanliness of the steel
Raziskava vpliva re`ima `lindre v ponov~ni pe~i na ~istost jekla
Z. Adolf, I. Husar, P. Suchánek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
The influence of illite-kaolinite clays’ mineral content on the products’ shrinkage during drying and firing
Vpliv vsebnosti glin ilinit-kaolinit na kr~enje pri su{enju in `ganju
M. Krgovi}, N. Marstijepovi}, M. Ivanovi}, R. Zejak, M. Kne`evi}, S. \urkovi} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
The application of spheroidal graphite cast iron in Bosnia and Herzegovina
Uporaba nodularne grafitne litine v Bosni in Hercegovini
D. Pihura, M. Oru~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
15. KONFERENCA O MATERIALIH IN TEHNOLOGIJAH / 8. – 10. oktober, 2007, Portoro`, Slovenija
15th CONFERENCE ON MATERIALS AND TECHNOLOGY / 8–10 october, 2007, Portoro`, Slovenia . . . . . . . . . . . . . . . . . . . . . . . 197
ISSN 1580-2949
UDK 669+666+678+53 MTAEC9, 41(4)155–195(2007)
MATER.
TEHNOL.
LETNIK
VOLUME 41
[TEV.
NO. 4
STR.
P. 155–195
LJUBLJANA
SLOVENIJA
JULY-AUGUST
2007

B. KESKES ET AL.: A FATIGUE CHARACTERIZATION OF HONEYCOMB SANDWICH PANELS WITH A DEFECT
A FATIGUE CHARACTERIZATION OF HONEYCOMB
SANDWICH PANELS WITH A DEFECT
UTRUJENOSTNA KARAKTERIZACIJA SATASTIH SENDVI^NIH
PANELOV Z NAPAKO
Boualem Keskes1,2 , Yves Menger1, Ahmed Abbadi1, Joseph Gilgert1,
Nourredine Bouaouadja2 , Zitouni Azari1
1LFM Université de Metz ENIM Île du Saulcy, F-57045 Metz cedex 01, France
2 Mechanic and Optic department, University of Setif Algeria, Algeria
b_keskes univ-metz.fr
Prejem rokopisa – received: 2006-12-11; sprejem za objavo – accepted for publication: 2007-02-19
Honeycomb sandwich panels are used because of their high stiffness, good fatigue resistance and low weight. These panels are
used in a variety of applications, but particularly in the aerospace industry. When this is the case a simple knowledge of the
static properties is not sufficient and additional information about the fatigue properties is required. In real situations these
panels can be affected by manufacturing defects and impacts, and it is important to know the effects of these defects and the
behaviour of the damaged panel; it also important to determine the location of the defect. In our investigation these defects will
be simulated by a blind hole in the centre of the lower face sheet. Static and fatigue tests (four-point bending) with
acoustic-emission monitoring were carried out on sandwich panels with defects. The load/displacement and the S-N fatigue
curves are presented and analyzed.
Key words: sandwich, honeycomb, four-point bending, fatigue, defect, acoustic emission
Satasti sendvi~ni paneli se odlikujejo po veliki togosti, dobri odpornosti proti utrujenosti in nizkem razmerju mase. Te panele se
uporablja za razli~ne namene, posebno v letalski industriji. V tem primeru ni dovolj poznanje stati~nih lastnosti, zato so
potrebne dodatne informacije o utrujenosti. Pri uporabi lahko na lastnosti panelov vplivajo napake pri izdelavi in po{kodbe, zato
je treba poznati u~inek napak in razvoj po{kodb, pa tudi znati dolo~iti mesto po{kodbe. Napake so v tem delu simulirane s slepo
izvrtino v sredini spodnje povr{ine panelne plo{~e.
Izvr{eni so bili stati~ni in utrujenostni preizkusi s {tirito~kovnim upogibom z akusti~no emisijo na panelih z napako.
Predstavljene in analizirane so odvisnosti obremenitev – pomik in utrujenostne krivulje S-N.
Klju~ne besede: sendvi~, satje, 4-to~kovni upogib, utrujenost, napaka, akusti~na emisija
1 INTRODUCTION
The main benefits of using the sandwich concept in
structural components are the high stiffness, the good
fatigue resistance and the low weight. Recent advances
in materials and manufacturing techniques have resulted
in further improvements and the increased uniformity of
the properties of sandwich composites. In order to use
these materials in different applications, a knowledge of
their static properties alone is not sufficient, and
additional information about their fatigue properties is
required. However, many difficulties are encountered,
mainly in forecasting their fatigue life, which reduce the
utilisation of such materials in various industrial and
aerospace applications.
Investigations of the bending-fatigue behaviour of
sandwich beams were performed by Olsson and Lönnö 1,
Echtermeyer et al. 2, Allen and Shenoi 3, Lagunegrand et
al. 4, Burman and Zenkert 5. It was found that under
fatigue cycling of constant amplitude, the nucleation
phase of fatigue damage extends over the major part of
the fatigue life, while the phase of defect propagation is
very short. The fatigue life of a component will also be
adversely affected by damage, though the magnitude of
this reduction in fatigue life is often more difficult to
establish 6,7. The fatigue of damaged (initial defects)
structures may be determined by the extensive testing of
specimens with various defects at different load levels 8.
Static overload or fatigue damage may cause
significant degradation of the core or skin, which is not
easily detectable by a visual inspection or conventional
non-destructive evaluation techniques. Damage in a
sandwich structure is not only caused by in-service
loads; in the manufacturing processes used, defects
within the sandwich, such as skin/core interface
disbanding and stress concentrations at joints between
core materials, can also occur 9,10.
Acoustic emission (AE) provides the possibility to
monitor, dynamically in real time, the response to a
discontinuity under an imposed structural stress, and has
a significant advantage over other non-destructive testing
methods. AE is a quality-control and non-destructive
evaluation (NDE) technique that has proven to be most
useful in metals and sandwich-composite structures 11,12.
Basically, AE employs a transducer, fixed to the
structure, which registers emitted sounds from the loaded
structure. The emitted sounds are then quantified and
compared to a database of known sound-defect relation-
ships, and the degree of damage from which the sounds
are emitted can be quantified 13. The analysis of the AE
signals in the time and frequency domains can allow AE
monitoring to be used to identify failure modes.
Materiali in tehnologije / Materials and technology 41 (2007) 4, 157–161 157
UDC/UDK 539.4:620.17 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 41(4)157(2007)
Attempts to correlate our results with those from other
researchers are also complicated by the effect that
acquisition parameters, such as filtering, threshold
settings and sensor response, may have on the processed
results. Of primary interest when performing mechanical
testing utilising AE equipment, is the amount of activity
(e.g. hits, events or counts), when it occurs (relative to
load and/or time), where it originates (multiple sensor
arrays allow location detection), and the characteristics
of the signals in both the time and frequency domains.
The signal amplitude is widely used as the first stage of
damage characterisation. The main disadvantage of AE
is that it requires a knowledge of the signal-propagation
characteristics and a history of typical failures for the
material and the structure under investigation. In practice
it is common to determine experimentally the velocity by
measuring the time taken by a known signal to travel a
defined distance. The sound velocity in the sandwich
panel was found to be approximately 2500 m/s 12,14.
To establish a robust fatigue-life model, a better
understanding of the various failure mechanisms during
cyclic loading is necessary. Fatigue tests (four-point
bending) have been carried out on sandwich panels with
and without a defect. To study the growth of the damage
near the hole, AE was used, since it was expected that
the different stages of the failure mode would be
revealed. The expected result is a diminution of the
remaining fatigue-life time, in proportion to the size of
the hole.
2 MATERIAL AND EXPERIMENTAL
TECHNIQUES
Sandwich panels consist of an aluminium core and a
pure aluminium face sheet. The honeycomb core is an
open cell with a density of 82 kg/m3 of aluminium core.
The cell size is 9.6 mm. The dimensions of these panels
are shown in Table 1.
Table 1: Sample dimension
Tabela 1: Dimenzije preizku{anca
L (mm) b mm h mm hc (mm) tf mm L2 mm L1 mm
500 250 10 8.80 0.60 420 210
The skin material is AlMg3 5754. The properties of
the core are shown in Table 2 15.
Table 2: Core mechanical properties
Tabela 2: Mehanske lastnosti jedra panela
Material
Aluminium
–
Aluminium
Core ECM
Cell dimension /mm 6.4
Density /(kg/m3) 82
Shearing Strength (configuration L) /MPa 2.4
Shearing Modulus (configuration L) /MPa 430
Shearing Strength (configuration W) /MPa 1.40
Shearing Modulus (configuration W) /MPa 220
Compression Strength /MPa 4.5
The beams with defects were tested with four-point
bending and were monitored with AE, as shown in
Figures 1 and 2. The tests were carried out with a
servo-hydraulic Instron 8501 universal testing machine
with a 10-KN capacity and a 2-mm/min crosshead
velocity (Figure 3).
Cyclic flexural tests were also performed. The tests
were carried out under load control at a load ratio R =
0.1 using a sinusoidal wave form. The beams were
cycled at a frequency of 2 Hz. The fatigue data were
generated at load levels of 100 %, 90 %, 80 %, 70 % and
60 % of the ultimate static load. The fatigue life of the
specimens is defined as the number of cycles to ultimate
failure. The normalised applied load is plotted against
the number of cycles on a log-log scale.
The AE system used was a Vallens AMSY-5,
data-acquisition unit with Vallens SE-45-type trans-
B. KESKES ET AL.: A FATIGUE CHARACTERIZATION OF HONEYCOMB SANDWICH PANELS WITH A DEFECT
158 Materiali in tehnologije / Materials and technology 41 (2007) 4, 157–161
Figure 1: Honeycomb sandwich panel showing L and W configu-
rations 15
Slika1: Satasti sendiv~ni panel v L- in v W-konfiguraciji 15
Figure 2: Schematic of sandwich beam for four-point bending and
hole geometry
Slika2: Shema sendvi~ne grede pri 4-to~kovnem upogibu in geome-
trija izvrtine
ducers 16. These sensors have sensitivity in the range 25
kHz to 120 kHz and a secondary range of sensitivity
from 120 kHz to approximately 450 kHz. The AE signal
was band-pass filtered with a 30 kHz to 1 MHz pream-
plifier and the total system amplification maintained at
40 dB allowed processing of the preamplifier input
signal up to 99.9 dB above 1 mV (±99 mV peak). The
AE sensors were mounted directly on the lower skin
without any preparation of the contact surface using
petroleum jelly as the acoustic coupling. The contact
pressure was maintained with elastic tape.
3 RESULTS AND DISCUSSION
3.1 Static tests
AE was used to detect the damage and the crack
mechanisms in the structure. Figures 4 to 5 show the
applied load vs. the displacement. However, the same
observation can be made with the energy or the
amplitude of the events. The same test had already been
performed on the same sample, but without a defect 10.
We observed that the sandwich structures present a
maximum ultimate load and exhibit ductile behaviour.
The material shows more resistance in the L than in the
W cell configuration (Figures 4 and 5). At the begin-
ning of the plasticity domain, and during the catastrophic
failure, there is intense AE activity. Moreover, the hole
does not have an influence on the static behaviour of the
honeycomb sandwich panels. The modes of collapse
were identified: the face yield, the cell buckling, and the
indentation beneath the loading rollers, as shown in
Figure 6.
These modes of collapse are confirmed by a number
of investigations 6,8 The final failure for all the static tests
occurred in the top skin and the core by a local
indentation in the vicinity of the loading points, as
illustrated in Figure 6. The localization of the damage in
all cases was in the region close to the support, between
the support and the adjacent load application.
B. KESKES ET AL.: A FATIGUE CHARACTERIZATION OF HONEYCOMB SANDWICH PANELS WITH A DEFECT
Materiali in tehnologije / Materials and technology 41 (2007) 4, 157–161 159
Figure 3: Test setup
Slika 3: Shema preizkusne naprave
Figure 6: Failure mode showing local indentation skin and cell
buckling
Slika 6: Na~in po{kodbe z lokalnim vdorom in izlo~anjem alumi-
nijastih celic
Figure 4: Load-deflexion Alu-Alu direction W
Slika 4: Obremenitev upogiba Alu-Alu v smeri W
Figure 5: Load-deflexion Alu-Alu direction L
Slika 5: Obremenitev upogiba Alu-Alu v smeri L
3.2 Fatigue tests
To investigate the effect of the defect (hole) on the
fatigue life, flexural tests were performed in the cell
direction L for an aluminium-aluminium sandwich panel
with 82 kg/m3.
Both curves (Figures 7 and 8) show the AE activity
(Energy) vs. time and the minimal displacement vs. time.
AE makes it possible to locate the crack initiation.
Indeed, when the crack begins to grow there is a peak in
the energy consumed. The displacement vs. time
dependence shows that the crack initiation occurred at
the same time.
The Wöhler curves were plotted with all the results
of these tests and compared to the results for the same
panels without a defect (Figure 9).
The results show that the fatigue life is the highest
for the panels without a defect for the same applied load.
Indeed, for an applied load of 65 % of the ultimate static
load the number of cycles to failure is about of 5.105
cycles (Figure 9) for the panel with a defect, while for
the panels without a defect it is greater than 106 cycles.
The final fractures for the panels with and without a
defect are shown in Figure 10 and Figure 11,
respectively. The crack started at the hole and grew in
terms of its width (Figure 10).
4 CONCLUSION
Defect effects in static and fatigue studies of
honeycomb core sandwich panels were investigated and
the following observations were made.
B. KESKES ET AL.: A FATIGUE CHARACTERIZATION OF HONEYCOMB SANDWICH PANELS WITH A DEFECT
160 Materiali in tehnologije / Materials and technology 41 (2007) 4, 157–161
Figure 11: Fatigue-failure mode of the panel without a defect
Slika 11: Na~in preloma panela brez napake
Figure 8: Results for sample 2. AE activity (Energy) and the minimal
displacement vs. time (70 % ultimate load)
Slika8: Rezultati AE aktivnosti za preizku{anec 2 in minimalni
upogib v odvisnosti od ~asa (70 % kon~ne obremenitve)
Figure 9: Wöhler Curves Aluminium-Aluminium
Slika 9: Wöhlerjeve krivulje aluminij-aluminij
Figgure 7: Results for sample 1. AE activity (Energy) and the
minimal displacement vs. time (80 % ultimate load)
Slika7: Primer AE aktivnosti za preizku{anec 1 in minimalen upogib
v odvisnosti od ~asa (80 % kon~ne obremenitve)
Figure 10: Fatigue-failure mode showing cracking skin through the
defect
Slika 10: Utrujenostna po{kodba, ki prikazuje razpoke skozi ko`o
No defect effect was observed in the case of the static
study, in contrast to the case of the fatigue behaviour of
the sandwich honeycomb panels. In the fatigue study, the
fatigue life of the defect panels decreased rapidly when
the applied load increased, compared to the panels
without a defect. The results of the acoustic emission
show the crack initiation and propagation and can be
used as a reliable survey method for the damage
mechanisms. The analysis of the acoustic signals
confirmed that the majority of the damage growth occurs
at peak-load levels and demonstrated the significant
effect of the loading level on the progress of the damage.
5 REFERENCES
1 Olsson KA, Lönnö A., Test procedures for foam core materials, In:
Olsson KA, Reichard RP, editors. Proceedings of the First
International Conference on Sandwich Constructions. Solihull, UK:
EMAS, UK, 1989, 293–318
2 Echtermeyer AT, Buene L, McGeorge D, Sund OE. Four point bend
testing of GRP sandwich beams – Part 1. Der Norske veritas
VR-91-P0013. 1991
3 Allen HG, Shenoi RA. Flexural fatigue tests on sandwich structures,
In: Weismann-Bermann D, Olsson KA, editors. Proceedings of the
Second International Conference on Sandwich Constructions.
Solihull, 1992, 499–517
4 Lagunegrand L, Lorriot Th, Harry R, Wargnier H. Design of an
improved four point bending test on a sandwich beam for free edge
delamination studies, Composites: Part B 37 (2006), 127
5 Burman M, Zenkert D. Fatigue of undamaged and damaged honey-
comb sandwich beams. J Sandwich Struct Mater (2000) 2, 50–74
6 Kapil Mohana, Yip Tick Hon, Sridhar Idapalapati, Hong Pheow
Seow Failure of sandwich beams consisting of alumina face sheet
and aluminum foam core in bending Materials Science and
Engineering A 409 (2005), 292–30
7 A. Abbadi, Z. Azari, S. Belouatar, J. Gilgert, S. Dominiack Static
and fatigue characterization of honeycomb sandwich panels JNC14
Compiègne, French (2005)
8 Demelio G, Genovese K, Pappalettere C, An experimental investi-
gation of static and fatigue behaviour of sandwich composite panels,
Composites B Eng 32 (2001) 4, 299–308
9 H. Chen, R. Bai, Postbuckling behavior of face/core debonded com-
posite sandwich plate considering matrix crack and contact effect,
Compos Struct 57 (2002) 1, 305–313
10 Arnaud pollien, Yves Conde, Laurent Pambaguiam, Andreas
Mortensen, Graded open Cell aluminium foam core sandwich beams
Materials Science and Engineering A 404 2005)
11 S. Mechraoui, S. Benmedakhene, S. Amami, A. Laksimi – Damage
crack analysis of glass/epoxy composite laminates by acoustic
emission, JNC15, Compiègne (2005)
12 B. SHAFIQ, A. QUISPITUPA – Fatigue characteristics of foam core
sandwich composites International Journal of Fatigue, 28 (2006),
96–102
13 Nondestructive Testing Handbook-Volume 5, Acoustic Emission,
Testing, American Society for NDT 5asnt), 1987
14 Prosser W, Advanced AE techniques in composite materials
research, Journal of Acoustic Emission 14, (1996) 3/4, 1–11
15 EURO-COMPOSITES, ECM-Honeycomb Technical Manuel, Euro-
composites S.A, Luxembourg, 1998
16 Vallen System, www.vallen.de.com
B. KESKES ET AL.: A FATIGUE CHARACTERIZATION OF HONEYCOMB SANDWICH PANELS WITH A DEFECT
Materiali in tehnologije / Materials and technology 41 (2007) 4, 157–161 161

Z. BURZI] ET AL.: FATIGUE PROPERTIES OF A HIGH-STRENGTH-STEEL WELDED JOINT
FATIGUE PROPERTIES OF A HIGH-STRENGTH-STEEL
WELDED JOINT
UTRUJENOSTNE LASTNOSTI ZVARA VISOKOTRDNEGA JEKLA
Zijah Burzi}1, Vencislav Grabulov1, Stojan Sedmak2, Aleksander Sedmak3
1Military Technical Institute, Katani}eva 15, 11000 Beograd, Serbia & Montenegro
2Faculty of Technology and Metallurgy, Karnegijeva 4, 11000 Beograd, Serbia & Montenegro
3Faculty of Mechanical Engineering, Kraljice Marije 16, 11000 Beograd, Serbia & Montenegro
zijah_burzic rvkds.net
Prejem rokopisa – received: 2006-05-17; sprejem za objavo – accepted for publication: 2007-03-09
In addition to the strength and toughness properties, the fatigue properties of welded joints are necessary for the design of
high-strength steel structures exposed to variable loading. Wöhler curves were determined for smooth tensile specimens tested
with variable loading. The fatigue behaviour of a welded joint can be improved by removing the overfill; however, the fatigue
properties are still lower than the values for the parent metal. Tests with pre-cracked specimens have shown the difference in the
fatigue-crack growth-rate properties to be less significant than other crack parameters.
Keywords: HSLA, welded joint, low-cycle fatigue, high-cycle fatigue, crack growth rate
Utrujenostne lastnosti so potrebne poleg trdnosti in `ilavosti za na~rtovanje struktur iz visokotrdnih jekel, ki prena{ajo
spremenljivo obremenitev. Wöhlerjeve krivulje so bile dolo~ene za gladke preizku{ance in spremenljivo obremenitev.
Utrujenostno vedenje zvarjenega spoja se lahko izbolj{a z odstranitvijo nadvi{enja, vendar ostanejo utrujenostne lastnosti ni`je
kot pri osnovnem materialu. Preizkusi na vzorcih z razpoko so pokazali, da je manj izrazita razlika v hitrosti rasti razpoke kot
pri drugih parametrih razpoke.
Klju~ne besede: HSLA, zvarjeni spoj, malocikli~na utrujenost, velikocikli~na utrujenost, hitrost rasti razpoke
1 INTRODUCTION
When selecting a material for a particular
application, the ease and cost of welding must be
considered, and the material selected should give a
welded product with adequate properties for the
minimum cost. The steels developed for heavily loaded
structures and their welded joints have to be resistant to
variable loading, in addition to having adequate strength
and toughness properties 1. The benefits of a strength
increase can be expressed in terms of reduced com-
ponent dimensions, followed by a significant reduction
in welded-joint cross-sections, the consumption of
welding electrodes and the time necessary to produce the
welded joints.
Since defects are frequently involved in welded
structures, it is necessary to design against low-stress
failure by fatigue. By maintaining an adequate level of
control by non-destructive testing, it is possible to ensure
that cracks exceeding some maximum size, as the most
dangerous defects, will be avoided and the stress
concentration in components be reduced as a result. This
is important for the welded joints of high-strength steel
that are exposed to fatigue. The data about the fatigue
behaviour of HSLA steel welded-joint constituents are
necessary, starting with the design stage, since the
benefits gained by having a high strength could be lost
under variable loading.
2 PREPARATION OF SAMPLES
The experiments were performed with the
high-strength steel NIONICRAL-70 (NN70), with the
nominal yield-stress class of 700 MPa and its welded
joints, produced with metal manual arc welding (MAW).
This steel is designed for the manufacturing of pressure
vessels and in shipbuilding, e.g., for submarines, but is
also applicabble for other heavy-duty sructures. The
chemical composition and the mechanical properties are
shown in Tables 1 and 2, respectively.
Two plates of NIONICRAL-70, 18 mm thick, were
prepared by edge machining for asymmetric 2/3 X and
welded in 6 passes with a Tenacito-80 electrode, which
produced a slightly undermatched welded joint.
3 TESTING FOR FATIGUE-ENDURANCE
DETERMINATION
Low-cycle fatigue initiation can be expected in the
region of welded joints, because the yield stress can be
achieved locally by stress concentration. The smooth
specimens for testing with variable loading are presented
in Figure 1. Four sets of smooth specimens were
prepared: OM, from the parent metal and from welded
samples; XN, in the as-welded condition; XB, with the
overfill removed by grinding; and XO, with both sides
machined to 15 mm and so the rough layer of rolling was
removed together with the overfill. The specimens were
Materiali in tehnologije / Materials and technology 41 (2007) 4, 163–166 163
UDC/UDK 539.4:669.14.018.298 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 41(4)163(2007)
tested on a hydraulic machine, with the lower grip fixed
and the upper grip oscillating with the frequency f = 9
Hz to 15 Hz, depending on the maximum load in the
cycle,  g , at the stress ratio R =  d/ g = 0.1 ( d is the
lower stress in the cycle). Local plastic deformation
ahead the crack tip is typical for crack initiation and
growth in the early stage of low-cycle fatigue, followed
by shear lips. But, if the frequency is low, in low-cycle
fatigue of high-strength steel the fracture appearance is
similar to that obtained in high-cycle fatigue, with no
shear lips. In some cases the loading level was so low
that the specimens did not fracture, even after more than
one million cycles. As a result of this they were tested in
a high-cycle regime.
For a high stress level (691 MPa), close to the yield
stress (Table 2), the as-welded XN specimens fractured
with low-cycle fatigue after only 6700 cycles. The
fatigue crack initiated in the region of stress
concentration, in the transition from the overfill to the
heat-affected zone (HAZ), and developed through the
HAZ’s coarse-grain region, followed by significant
contraction of the cross-section. The crack grows on
both sides of the specimen, and the final fracture
occurred in the weld metal in a reduced ligament size. At
a stress of 415 MPa the specimen fractured after 66,300
cycles. The high-cycle fatigue crack in the third XN
specimen initiated in the same region and propagated in
the plane normal to the load direction, completely
through the HAZ’s coarse-grain region of reduced
ductility. The stress level was 274 MPa and the number
of cycles at fracture was 143,700. The stress con-
centration in the XB specimens was reduced by
removing the overfill. This increased the number of
cycles to fracture: (a) for stress 685 MPa to 13,700
cycles, (b) for 228 MPa to 690,800 cycles. The crack
initiated in the HAZ’s coarse-grain region in both
low-cycle (a) and high-cycle (b) fatigue and developed
mostly through the HAZ. The best results for the welded
samples were obtained with machined sides of the XO
specimen in the absence of the stress concentration due
to geometry. The crack initiated in the weld metal’s
critical microstructure. In low-cycle fatigue (a) the crack
developed under 45° at a load of 586 MPa, up to 49,100
cycles. In high-cycle fatigue (b) the crack path is normal
to the applied load of 443 MPa, and the fracture occurred
after 565,300 cycles. In some specimens the crack
initiated from embedded defects.
The results are summarized in the upper part of the
relationship applied load   vs. number of cycles N
(Wöhler) (Figure 2). The fracture stress is satisfactory
for the parent metal (OM)   = 625 MPa and for the
machined specimens (XO)   = 530 MPa, but the results
for the XN specimens in the as-welded condition (  =
370 MPa) and for the XB specimens with the overfill
Z. BURZI] ET AL.: FATIGUE PROPERTIES OF A HIGH-STRENGTH-STEEL WELDED JOINT
164 Materiali in tehnologije / Materials and technology 41 (2007) 4, 163–166
Table 1: Chemical composition of NIONICRAL-70 steel, w/%
Tabela 1: Kemi~na sestava jekla Nionicral 70, w/%
C Si Mn P S Cr Ni Mo V Al
0.1 0.2 0.23 0.009 0.018 1.24 3.1 0.29 0.05 0.08
Table 2: Mechanical properties of NIONICRAL-70 steel
Tabela 2. Mehanske lastnosti jekla Nionicral 70
Specimen
orientation
Yield stress Tensilestrength Elongation Contraction Charpy V impact energy, E/J
Rp0.2/MPa Rm/MPa A/% Ζ/% +20 °C −60 oC −100 °C
Parallel 780 820 19 66 126 117 93
Perpendicular 770 810 20 74 81 76 49
Figure 2: Upper part of the relationship applied load   vs. number of
cycles N
Slika 2: Zgornji del odvisnosti obremenitev – {tevilo amplitud
Figure 1: Smooth specimens for testing by variable loading
Slika 1: Gladki preizku{anci za preizkus pri variabilni obremenitvi
ground away (  = 415 MPa) are not acceptable. This is
confirmed by the value of the coefficient m – the Wöhler
curve slope in the initial part, up to 100,000 cycles,
which is 5.90 for the OM specimens, 5.80 for the XO
specimens, 4.23 for the XB specimens and 4.17 for the
XN specimens. The recommended design values are 5 to
7 2.
4 TESTING OF FATIGUE-CRACK GROWTH
RATE
Welded structures can contain small pre-existing
cracks, which will propagate under repeated loads up to
the size critical for fracture. Since in this case the zone
ahead of the crack tip, exposed to the cyclic plasticity, is
small, a plane-strain state is formed, even for a small
thickness, and generally the data obtained with small
specimens can be applied.
The testing of the fatigue-crack growth rate is
performed using the ratio R = 0.1, with precracked
SE(B) specimens (width W = 16 mm, thickness B = 12
mm, span S = 4W) of parent metal, weld metal and HAZ,
on a CRACKTRONIC dynamic testing device. The
number of cycles was registered for each 0.1 mm of
crack growth, as presented in the crack length a vs.
number of cycles of N relation (Figure 3). The curve on
the right-hand side is for the parent metal, the middle
curve is for the weld metal, and the curve on the
left-hand side is for the HAZ. From these curves the data
necessary for Paris law are derived:
d
d
ma
N
C K= ( )∆ (1)
Here, da/dN is the crack size a growth per unit cycle,
N is the cycle number, C and m are constants obtained
from experiments and given in Table 3,  K = Kmax – Kmin
is the stress-intensity factor range in the loading cycle.
The obtained relationships da/dN vs.  K are given in
Figure 4. It is interesting that the difference in the
fatigue threshold value, the value of the stress-intensity
factor range,  Kth, at which existing crack will not grow,
is not significant: for the parent metal it is 10.22
MPa m1/2, for the weld metal it is 9.11 MPa m1/2 and for
heat-affected zone it is 8.51 MPa m1/2.
Z. BURZI] ET AL.: FATIGUE PROPERTIES OF A HIGH-STRENGTH-STEEL WELDED JOINT
Materiali in tehnologije / Materials and technology 41 (2007) 4, 163–166 165
Figure 4: Diagram da/dN –  K
Slika 4: Odvisnost da/dN- K
Figure 3: Crack length a vs. number of cycles N for parent metal
(right), weld metal (in the middle) and HAZ (left)
Slika 3: Odvisnost dol`ine razpoke – {tevilo ciklov za osnovni
material (desno), deponirani material (sredina) in HAZ (levo)
Table 3: The values for parameters C and m in the Paris equation
Tabela 3: Vrednosti parametrov C in m v Parisovi ena~bi
Specimen C m Specimen C m
Parent metal HAZ
I 3.98  10 14 4.139 I 1.90   10 20 10.259
II 1.67   10 13 3.765 II 4.63   10 12 2.667
Weld metal III 2.90   10 16 6.403
I 8.38   10 15 4.798 IV 7.87   10 13 3.560
II 3.30   10 19 8.462 V 1.48   10 16 6.505
III 7.93   10 15 5.078 VI 1.74   10 14 4.929
5 CONCLUSIONS
The importance of reducing the stress concentration
for fatigue life can be easily seen from Figure 3. The
critical load for N = 105 cycles for the smooth parent-
metal specimen is 625 MPa, reduced in the machined
specimen of the welded joint to 530 MPa, with the
overfill ground away it is reduced to 415 MPa, and down
to 370 MPa in the as-welded condition.
In regime I the crack-growth rate is low since the
threshold for the crack Kth is approached. In regime II
the Paris law is obeyed, while in regime III the crack-
growth rate increases above that predicted by the Paris
relation.
The fatigue resistance of the weld metal and the HAZ
is reduced, compared to the parent metal, and this has to
be taken into account when a welded structure is
designed with high-strength steel.
Acknowledgement
The paper was prepared with the financial support of
the Serbian Ministry of Science and Environment
Protection – project ON 14027 "Special topics in
fracture mechanics of materials".
6 REFERENCES
1 W. S. Pellini: Guidelines for fracture-safe and fatigue-reliable design
of steel structures, The Welding Institute, Abington, Cambridge,
1983
2 H. W. Munse: Fatigue of welded steel structures, Welding Research
Council, 1964
Z. BURZI] ET AL.: FATIGUE PROPERTIES OF A HIGH-STRENGTH-STEEL WELDED JOINT
166 Materiali in tehnologije / Materials and technology 41 (2007) 4, 163–166
D. KLOB^AR, J. TU[EK: MEHANSKE LASTNOSTI ZVARA IZ JEKLA MARAGING PO IZLO^EVALNEM @ARJENJU
MEHANSKE LASTNOSTI ZVARA IZ JEKLA MARAGING
PO IZLO^EVALNEM @ARJENJU
MECHANICAL PROPERTIES OF MARAGING STEEL WELDS
AFTER AGING HEAT TREATMENT
Damjan Klob~ar, Janez Tu{ek
Fakulteta za strojni{tvo, Univerza v Ljubljani, A{ker~eva 6, 1000 Ljubljana, Slovenija
Prejem rokopisa – received: 2006-09-21; sprejem za objavo – accepted for publication: 2007-06-06
V industriji tla~nega litja aluminija so bile narejene pilotne raziskave, ki naj bi potrdile mnenje, da orodja iz jekel maraging
dose`ejo dalj{o trajnostno dobo in s tem ni`jo ceno ulitka. Zato smo se odlo~ili, da jekla maraging uporabimo za reparaturno
varjenje orodij. Ta raziskava je del tiste, s katero smo ugotavljali primernost tega jekla za reparaturno varjenje orodij za tla~no
litje aluminija. Cilj dela je bil ugotoviti optimalne parametre izlo~evalnega `arjenja varov iz jekla maraging, ki bodo zagotovili
optimalne mehanske lastnosti. Posebej pripravljene varjence z utori v obliki ~rke U smo po postopku TIG navarili z jeklom
maraging UTP A702. Varjenje je potekalo z minimalnim vnosom energije v za{~itni atmosferi plina argona. Tako pripravljene
varjence smo izlo~evalno `arili pri ~asih od 0,7 h do 10,8 h ter temperaturah od 445 °C do 515 °C. Analizirali smo tudi varjence
v navarjenem stanju in tiste, katerih varki so bili med varjenjem kovani. Iz posameznega varjenca smo izdelali 3 epruvete
ISO-V za `ilavost, rezino za analizo mikrostrukture in trdote ter 7 miniaturnih epruvet za natezno trdnost. Navar iz jekla
maraging dose`e optimalne mehanske lastnosti z 2-urnim izlo~evalnim `arjenjem na temperaturi 480 °C. Izlo~evalno `arjenje
pri temperaturi 515 °C naj bi trajalo 1 h, pri temperaturi 445 °C pa 3 h. Kovanje varkov pove~uje `ilavost zvara.
Klju~ne besede: jeklo maraging, izlo~evalno `arjenje, mehanske lastnosti, mikrostruktura, varjenje TIG, tla~no litje aluminija
Investigations have shown that tools from maraging steels achieved longer tool life in high pressure die casting of aluminum.
We have decided to check the use of maraging steel for repair welding of tools. A part of more extensive investigation aimed to
establish the suitability of maraging steel for repair welding of high pressure die-casting toolings is done. In this work the
results of the investigation used to establish optimal precipitation hardening parameters of maraging steel welds, which provides
optimal mechanical properties is presented. Specimens with a “U” shaped grove were weld cladded with maraging steel UTP
A702. GTA welding was carried out in protective atmosphere of argon gas and with minimal heat input. The test specimens
were than precipitation annealed from 0.7 h to 10.8 h and temperatures from 445 °C to 515 °C. Specimens in as-welded
condition and specimens with hammered welds were checked parallely. From same specimen 3 ISO-V specimens for toughness
test, a slice for microstructure analysis and measurement of hardness, and 7 miniature specimens for tensile test were
manufactured. The results showed that maraging steel welds achieve optimal mechanical properties after precipitation annealing
for 2.5 h at 480 °C. Precipitation annealing at 515 °C should last for 1 h and at 445 °C for 3 h. The hammering of welds
increases weld toughness.
Keywords: maraging steel, precipitation annealing, mechanical properties, microstructure, GTA welding, high pressure
die-casting, response surface modelling
1 UVOD
Tla~no litje aluminijevih zlitin je velikoserijski po-
stopek za proizvodnjo izdelkov zahtevnih geometrijskih
oblik v ozkih dimenzijskih tolerancah. Med tla~nim
litjem se v orodje vliva talino aluminijeve zlitine s
temperaturami do 700 °C, pri ~emer je hitrost taline od
30 m/s do 100 m/s ter polnilni tlak od 50 MPa do 80
MPa 1. Te obremenitve kraj{ajo trajnostno dobo orodij s
a) termi~nimi cikli, ki povzro~ajo utrujenostne razpoke
na povr{ini gravure, b) s korozijo ali sprijemanjem taline
aluminija na gravuro, c) z erozijo povr{ine zaradi
pretoka aluminija in ~) lomom orodja 2-4. Kon~na cena
ulitka je odvisna od parametrov tla~nega litja, ki vplivajo
na trajnostno dobo orodja. Namen raziskave je podalj{ati
trajnostno dobo orodij za tla~no litje aluminija z repa-
raturnim varjenjem, pri ~emer bi kot dodajni material
uporabili jekla maraging zaradi njihovih izjemnih
mehanskih lastnosti.
Prednost jekel maraging v primerjavi s klasi~nimi za
delo v vro~em se ka`e v manj{em modulu elasti~nosti in
manj{em linearnem temperaturnem razteznostnem
koeficientu (tabela 1). To povzro~a manj{e napetosti v
orodju med temperaturnimi obremenitvami. Ve~ja
toplotna prevodnost jekla maraging dodatno zmanj{a
temperaturo povr{ine in posledi~no napetosti. Maraging
jekla imajo v primerjavi s klasi~nimi jekli za delo v
vro~em ve~jo `ilavost in trdnostne lastnosti, kar vpliva
na bolj{o odpornost proti termi~nemu utrujanju. Za
uspe{no podalj{anje trajnostne dobe orodja morajo imeti
materiali za orodja za tla~no litje aluminija dobro
stabilnost mehanskih lastnosti pri povi{anih tempera-
turah. To dosegajo z jekli maraging z vsebnostjo niklja
14 % 5, 12 % (Thyrotherm 1.2799) ali celo 2 % 6. Manj{a
vsebnost Ni premakne transformacijo ferita v avstenit k
vi{jim temperaturam. Slabost manj{e vsebnosti Ni je
slab{a `ilavost 6.
Nasprotno od klasi~nih orodnih jekel za delo v
vro~em imajo jekla maraging omejeno vsebnost C, H in
N, svoje odli~ne lastnosti pa dose`ejo z legiranjem s/z
Co, Mo, Ti in Al 7-9. Dodatna prednost je njihova odli~na
varivost, saj jih pri varjenju ni treba predgrevati ali
Materiali in tehnologije / Materials and technology 41 (2007) 4, 167–171 167
UDK/UDC 669.14:620.17:621.791.05 ISSN 1580-2949
Izvirni znanstveni ~lanek/Original scientific article MTAEC9, 41(4)167(2007)
pogrevati, da bi dobili duktilen in `ilav martenzit. Tudi
mehanska in EDM-obdelava se izvajata la`e kot pri
klasi~nih orodnih jeklih.
Toplotna obdelava jekla maraging zajema topilno in
izlo~evalno `arjenje. Topilno `arjenje poteka pri
temperaturah 815–915 °C eno uro, pri tem pa se v
avstenitu topijo legirni elementi. Po ohlajanju na zraku
se tvori mehak martenzit, nasi~en z legirnimi elementi.
Pri varjenju z manj{im vnosom energije kot 1,8 kJ/mm
topilno `arjenje ni potrebno in varjenju sledi izlo~evalno
`arjenje 10. Slednje navadno poteka pri temperaturah od
470 °C do 550 °C pribli`no tri ure, dejanski parametri pa
so odvisni od kemi~ne sestave jekla. Med tem `arjenjem
se v materialu tvorijo izlo~ki, ki povzro~ajo deformacije
kristalnih re{etk. To povzro~a linearno spremembo
dimenzij od 0,05 % do –0,1 %, velikost spremembe pa je
odvisna od parametrov izlo~evalnega `arjenja 8.
Kljub odli~nim mehanskim lastnostim se jekla
maraging le malo uporabljajo kot material za orodja.
Vzrok je vi{ja cena. Na{a ideja je bila, da bi uporabili
jekla maraging za reparaturno popravilo orodij oz. za
navarjanje na povr{ino. S tem bi dobili cenovno sprejem-
ljiva visokokakovostna orodja. Na povr{ino orodja za
tla~no litje aluminija je treba navariti dovolj debelo plast
jekla maraging, da prepre~imo akumulacijo toplote na
prehodu med osnovnim jeklom in navarjeno plastjo.
Uspe{en odvod energije s povr{ine orodja bomo
zagotovili, ~e bo povr{inski sloj jekla maraging segal do
hladilnih kanalov.
Namen {tudije je bil karakterizirati jekla maraging (z
18 % Ni) glede na mehanske lastnosti in mikrostrukturo.
Narejena je bila analiza varjenja TIG jekla maraging
glede na vnos energije, temperaturo predgrevanja in ko-
vanje varkov, analiziran pa je bil tudi vpliv parametrov
izlo~evalnega `arjenja na razvoj mikrostrukture in
mehanske lastnosti.
2 EKSPERIMENTALNI DEL
Pripravljena je bila serija vzorcev iz jekla 1.2344 z
izdelanim utorom U po sliki 1. Pred navarjanjem so bili
vzorci pobolj{ani na trdoto HRc 46. V posamezen utor U
je bilo po postopku TIG navarjenih 14 varkov, dolgih 75
mm. Varjenje je potekalo z varilnim tokom 150 A,
varilno napetostjo 12 V, s hitrostjo varjenja 5 cm/min in
vnosom energije 2,1 kJ/mm v za{~iti plina argona s
pretokom 10 L/min. Navarjali smo dodajni material jeklo
maraging 1.6356 (tabela 2), ki je bilo v obliki palic
premera 2,5 mm in dol`ine 1000 mm. Tik pred
varjenjem smo s povr{ine varilnih `ic odbrusili bakreno
za{~itno plast ter `ico o~istili z acetonom. S tem smo
prepre~ili kontaminacijo vara z bakrom. Teme varka smo
pred varjenjem novega varka o~istili z `i~no {~etko in
acetonom. Varjenje ve~ine vzorcev je potekalo pri
temperaturi predgrevanja 100 °C, en vzorec je bil varjen
s predgrevanjem pri temperaturi 400 °C, varki enega
vzorca pa so bili med varjenjem ro~no kovani. S
kovanjem spremenimo natezne zaostale napetosti v
zvaru v tla~ne in zmanj{amo verjetnost pokanja med
termi~nim utrujanjem.
Navarjeni vzorci so bili izlo~evalno `arjeni po na~rtu
izlo~evalnega `arjenja, ki je bil izdelan s statisti~nim
na~rtovanjem eksperimentov. Uporabili smo sredi{~no
zasnovan na~rt eksperimentov in metodologijo povr{in
odziva, ki omogo~a razvoj regresijskih modelov vi{jega
reda. Parametri izlo~evalnega `arjenja so bili dolo~eni
tako, da zajamejo {ir{e obmo~je toplotne obdelave.
D. KLOB^AR, J. TU[EK: MEHANSKE LASTNOSTI ZVARA IZ JEKLA MARAGING PO IZLO^EVALNEM @ARJENJU
168 Materiali in tehnologije / Materials and technology 41 (2007) 4, 167–171
Tabela 1: Mehanske in fizikalne lastnosti jekel 5,12
Table 1: Mechanical and physical properties of steels 5,12
Lastnost 1.2344 UTP A7021.6356
Marlok
C1650
Gostota /(kg/dm3) 7,8 – 8,09
Modul elasti~nosti /GPa 210 191 186
Natezna trdnost /MPa 1430 1763 1600
Meja plasti~nosti /MPa 1230 1688 1500
Trdota /HRc 43–54 40–50 47–51
@ilavost /J
20 °C 15–20 15–21 25
200 °C – – 35
400 °C – – 45
Temperaturni koeficient dol`inskega
raztezka /(10–6 mm/(mm °C))
20°C–400°C 12,5 – 10
20°C–600°C 13,1 – 5,6
Toplotna prevodnost /(W/(m °C))
20 °C 25 – 28
500 °C 28,5 – 32
600 °C 29,3 – 33
Tabela 2: Kemi~na sestava jekel v masnih dele`ih (%) 5,12
Table 2: Chemical composition of steels in % 5,12
Element 1.2344 1.6356 Marlok C1650
C 0,3–0,4 0,02 <0,008
Cr 4,8–5,5 – <0,30
Mo 1,2–1,5 4,0 4,5
Ni – 18,0 14,0
Co – 12,0 10,5
Ti – 1,6 0,2
Si 0,8–1,2 – <0,10
Mn 0,2–0,4 – <0,10
Al – 0,1 –
Slika 1: Shema gradnje varkov v vzorcu z utorom U
Figure 1: Schematic of U grove specimen weld filling
Temperatura izlo~evalnega `arjenja je bila izbrana v
obmo~ju od 445 °C do 515 °C, ~as pa v obmo~ju od 0,7
h do 10,8 h. Centralna to~ka sredi{~no zasnovanega
na~rta eksperimentov je bila izbrana po priporo~ilu
proizvajalca pri 480 °C in treh urah.
Varjenci so bili po varjenju in toplotni obdelavi
razrezani po shemi na sliki 2. Iz vsakega varjenca so bile
izdelane tri epruvete ISO-V za preizkus `ilavosti, rezina
za analizo mikrostrukture in sedem miniaturnih epruvet
za natezni preizkus. Natezni preizkusi so bili narejeni na
ra~unalni{ko krmiljeni napravi Zwick Z050, ki je imela
vgrajen senzor sile GTM 50 kN in ekstenziometer tip
66607. Preizkusi so bili izvedeni po standardu EN
10002-1. Preizkus udarne `ilavosti po Charpiju je bil
opravljen v skladu s standardoma EN 10045-1:2000 in
EN 10045-2:1992. Iz posameznega varjenca so bile
preizku{ene tri epruvete ISO-V in izra~unane povpre~ne
vrednosti `ilavosti.
Na rezinah vzorcev je bila narejena metalografska
analiza in meritev trdote po Vickersu. Vzorci so bili
polirani in jedkani v 4-odstotni raztopini nitala (4 %
HNO3 + 96 % C2H5OH) ter pregledani na opti~nem
mikroskopu.
3 REZULTATI
Diagram na sliki 3 prikazuje trdoto in natezno trdnost
temenskih varkov iz jekla maraging v navarjenem stanju
ter `ilavost ve~varkovnega vara iz jekla maraging.
Najvi{jo natezno trdnost 1170 MPa je dosegel var,
narejen pri temperaturi predgrevanja 100 °C. Kovan
varek je imel nekoliko ni`jo natezno trdnost (1046 MPa),
medtem ko se je pri varku, narejenem pri temperaturi
predgrevanja 400 °C, drasti~no zmanj{ala natezna
trdnost na 953 MPa. Trdota temenskega vara v
navarjenem stanju je bila HV 400, kar je ve~ kot trdota
jekla maraging v homogeniziranem stanju, ki ima od HV
305 do HV 339. Trdota se je pri tem povi{ala zaradi
izlo~evalnega `arjenja med varjenjem TIG ter zaradi
temperature predgrevanja. Trdota kovanih varkov se je
zvi{ala na HV 410 zaradi kovanja, ki varke mehansko
utrdi. Trdota temenskega varka, narejena pri temperaturi
predgrevanja 400 °C je bila HV 390. Padec trdote je
posledica ve~jega vnosa energije med varjenjem, ki se
pojavi zaradi vi{je temperature predgrevanja. Najve~jo
`ilavost zvara   86 J smo dobili pri varjencu s kovanimi
varki. @ilavost vara v navarjenem stanju je bila 80 J,
medtem ko je bila `ilavost vara, narejena pri temperaturi
predgrevanja 400 °C, le 56 J.
Slika 4 prikazuje potek trdote in natezne trdnosti
temenskih varkov iz jekla maraging ter `ilavost ve~var-
kovnega vara v navarjenem stanju in po izlo~evalnem
`arjenju z razli~nimi parametri. Iz diagrama je razvidno,
da se z vi{anjem temperature in dalj{anjem ~asa izlo~e-
valnega `arjenja natezna trdnost in trdota pove~ujeta,
medtem ko se `ilavost zmanj{uje. Ob~uten padec `ila-
vosti se pojavi `e po kratkotrajnem izlo~evalnem
`arjenju (480 °C/ 0,7 h), ko se v mehkem martenzitu
oblikujejo izlo~ki, mikrostruktura pa je {e premalo
starana (slika 5a). Pri tem dose`e var `ilavost 27,7 J,
trdoto HV 410 in natezno trdnost 1306 MPa. S
pove~anjem temperature in ~asa izlo~evalnega `arjenja
dobimo dobro staran var, katerega mikrostruktura je
prikazana na sliki 5b (445 °C/ 3 h). Za dobro staran var
je zna~ilna `ilavost okoli 20 J, trdota okoli HV 505 ter
natezna trdnost 1383 MPa. ^e ~as izlo~evalnega `arjenja
D. KLOB^AR, J. TU[EK: MEHANSKE LASTNOSTI ZVARA IZ JEKLA MARAGING PO IZLO^EVALNEM @ARJENJU
Materiali in tehnologije / Materials and technology 41 (2007) 4, 167–171 169
0
200
400
600
800
1000
1200
1400
1600
1800
varjeno stanje 480 °C/0,7 h 505 °C/1,1 h 445 °C/3 h 480 °C/10,8 h
T
rd
o
ta
H
V
in
R
m
/M
P
a
0
10
20
30
40
50
60
70
80
90

il
a
v
o
s
t
/J
Povr{ina - trdota /
Povr{ina - /MPa
@ilavost /J
HV
Rm
Slika 4: Trdota, `ilavost in natezna trdnost varov v varjenem stanju ter
po izlo~evalnem `arjenju z razli~nimi parametri
Figure 4: Hardness, toughness and tensile strength of a surfacing
weld after different precipitation hardening
Slika 2: Iz varjenca smo izrezali: a) 7 miniaturnih epruvet za natezni
preizkus, 3 standardne epruvete ISO-V za `ilavost ter rezino za
meritev trdote in analizo mikrostrukture. b) Shemati~en prikaz
epruvete za natezni preizkus.
Figure 2: From the welded specimen were manufactured: a) 7 tensile
test specimens, 3 toughness test specimens and a slice for
microstructure analysis. b) Scheme of tensile test specimen.
0
200
400
600
800
1000
1200
varjeno s tanje varjeno s t., kovanje varjeno st., T = 400 °C
T
rd
o
ta
H
V
in
R
m
/M
P
a
0
20
40
60
80
100
120
p
Povr{ina - trdota /
Povr{ina - /MPa
@ilavost /J
HV
Rm
Slika 3: Trdota, `ilavost in natezna trdnost temenskih varov v varje-
nem stanju v odvisnosti od tehnologije varjenja
Figure 3: Hardness, toughness and tensile strength of surfacing weld
in as-welded condition of different conditions welding
{e podalj{amo in/ali povi{amo temperaturo (npr. 480 °C/
10,8 h), dobimo preve~ starano mikrostrukturo (slika
5c). Zanjo je zna~ilna trdota HV 540, natezna trdnost
1758 MPa in `ilavost 14,6 J.
Pri izbiri parametrov toplotne obdelave izlo~evalnega
`arjenja je potrebna posebna pozornost. Izbrati je namre~
treba parametre, ki dajo najbolj ustrezne mehanske
lastnosti za posamezno aplikacijo. S tem namenom je
bila narejena statisti~na analiza mehanskih lastnosti
varov iz jekla maraging z metodologijo povr{in odziva.
Izdelani so bili modeli za napoved trdote, `ilavosti in
natezne trdnosti v odvisnosti od parametrov izlo~eval-
nega `arjenja. Modeli so prikazani v grafi~ni obliki na
slikah 6a–c). Pove~anje temperature in/ali ~asa toplotne
obdelave izlo~evalnega `arjenja pove~a trdoto in natezno
trdnost ter zmanj{a `ilavost vara iz jekla maraging.
Mikrostrukture varov so primerjane na sliki 5.
Mikrostruktura na sliki 5a je premalo starana in je
tipi~na za var v varjenem stanju oz. za var, izlo~evalno
`arjen pri prekratkih ~asih ali prenizkih temperaturah.
Mikrostruktura, prikazana na sliki 5b, je dobro starana in
jo dobimo pri izlo~evalnem `arjenju z optimalnimi
parametri. Slika 5c je zna~ilna za preve~ starano
mikrostrukturo. Na mejah med celi~nimi dendriti se
pojavijo zna~ilna bela podro~ja, ki so bogata z Ni. Na teh
mestih se pretvorba v povratni avstenit pojavi `e pri
ni`jih temperaturah. Taka mikrostruktura ni `elena, ker
ima majhno `ilavost. Pri bolj izrazitem prestaranju se
zni`ata tudi trdota in natezna trdnost, kar posledi~no
vpliva na ve~jo obrabo in slab{o odpornost proti
termi~nemu utrujanju.
Na osnovi analize mikrostrukture je bil izdelan
grafi~ni prikaz stanja mikrostrukture glede na parametre
izlo~evalnega `arjenja (slika 6d). Premalo staran var
dobimo pri izlo~evalnem `arjenju pri ni`jih temperaturah
kraj{i ~as. Dobro staran var dobimo z izlo~evalnim `arje-
njem s parametri iz osrednjega podro~ja, preve~ staran
var pa z izlo~evalnim `arjenjem pri visokih temperaturah
in dolgih ~asih. S slike 6d) lahko ugotovimo take
parametre izlo~evalnega `arjenja, pri katerih bomo v
izbranem jeklu dobili `eleno mikrostrukturo. ^e te
rezultate kombiniramo z razvitimi modeli za napoved
mehanskih lastnosti, dobimo dodatno informacijo, ki
omogo~a la`jo izbiro parametrov izlo~evalnega `arjenja
za aplikacijo.
4 DISKUSIJA
Priporo~en vnos energije med varjenjem je po lite-
raturi 1,8 kJ/mm 10, mi pa smo varili z vnosom energije
2,1 kJ/mm. Mikrostruktura vara v navarjenem stanju je
premalo starana in ima visoko `ilavost ter relativno
nizko trdoto in natezno trdnost. Glede na trdoto jekla
maraging v homogeniziranem stanju je trdota v navarje-
nem stanju ve~ja. Zaradi manj{ega pove~anja trdote in
premalo starane mikrostrukture topilno `arjenje ni
D. KLOB^AR, J. TU[EK: MEHANSKE LASTNOSTI ZVARA IZ JEKLA MARAGING PO IZLO^EVALNEM @ARJENJU
170 Materiali in tehnologije / Materials and technology 41 (2007) 4, 167–171
a)
b)
Slika 6: Spreminjanje a) `ilavosti /J, b) trdote HV, c) natezne trdnosti
/MPa in d) stanja mikrostrukture v odvisnosti od parametrov izlo~e-
valnega `arjenja
Figure 6: Change of weld a) toughness, b) hardness, c) tensile
strength and d) microstructure after different precipitation annealing
µ
µ
µ
Slika 5: Mikrostrukture temenskih varkov po izlo~evalnem `arjenju:
a) 480 °C / 0,7 h, b) 445 °C / 3 h, 480 °C / 10,8 h
Figure 5: Microstructures of surfacing welds after precipitation
hardening: a) 480 °C / 0,7 h, b) 445 °C / 3 h, 480 °C / 10,8 h
potrebno. Kljub temu pomeni varjenje z vnosom energije
2,1 kJ/mm zgornjo vrednost vnosa energije.
Kovanje varkov med varjenjem je zamudno. Pri
varjenju trdih in krhkih orodij je pogosto neizogibno, da
prepre~imo pokanje. Kovanje spremeni natezne zaostale
napetosti, ki nastanejo v varu zaradi varjenja, v tla~ne in
posledi~no prepre~uje {irjenje razpok. Varom iz jekla
maraging pove~a `ilavost in trdoto. Vi{ja `ilavost je
najverjetneje posledica drobljenja ve~jih kristalnih zrn v
manj{a, vi{ja trdota pa je posledica mehanske utrditve.
Kovanje varkov iz jekla maraging je priporo~ljivo.
Predgrevanje na temperaturo 400 °C negativno
vpliva na mehanske lastnosti vara iz jekla maraging, saj
zmanj{uje trdoto, natezno trdnost in `ilavost, najverjet-
neje zaradi segrevanja varkov v obmo~ju temperatur od
700 °C do 1000 °C, to je v obmo~ju krhkosti. Pri tem se
na mejah avstenitnih zrn in celi~nih dendritov pojavijo
Ti(C, N)-izlo~ki, ki povzro~ajo krhkost jekla 7. Zaradi
pomanjkanja Ti v preostanku mikrostrukture sta manj{a
tudi natezna trdnost in trdota zvara. Predgrevanje na
temperaturo 400 °C se zato ne priporo~a. Najvi{ja
temperatura predgrevanja, ki jo priporo~a literatura, je
200 °C 12. Jekla maraging so dobro variva tudi pri sobni
temperaturi, ker se pri ohlajanju na zraku pojavi duktilen
in `ilav martenzit.
Pove~anje trdote in natezne trdnosti ter zmanj{anje
`ilavosti s pove~evanjem ~asa in temperature izlo~e-
valnega `arjenja je `e poznano iz literature. Pri tem je
zanimivo, da se pove~anje trdote in natezne trdnosti
sklada s padcem `ilavosti. To je v skladu z objavljenimi
podatki v literaturi za jekla maraging 8,12. Pri tem velja
poudariti, da literatura navadno ne navaja podatkov za
natezno trdnost, niti podatkov za mehanske lastnosti
zvarov. Prednost razvitih modelov za napoved mehan-
skih lastnosti po izlo~evalnem `arjenju je, da lahko
parametre dolo~imo na osnovi `elenih mehanskih
lastnosti. S tem zagotovimo, da ima material optimalne
mehanske lastnosti za dolo~eno uporabo `e vnaprej in
posledi~no vplivamo na trajnostno dobo orodja.
Grafi~en prikaz stanja mikrostrukture varov iz jekla
maraging glede na parametre izlo~evalnega `arjenja
predstavlja pomembno informacijo pri izbiri optimalnih
parametrov za posamezno aplikacijo. ^e ta model
kombiniramo z modeli za napoved mehanskih lastnosti
ter z zahtevami orodjarjev, se nam zo`ijo mo`nosti izbire
parametrov izlo~evalnega `arjenja. Kljub temu pa lahko
izberemo take parametre, ki bodo omogo~ali vi{jo trdoto
ali natezno trdnost ali vi{jo `ilavost.
5 SKLEPI
Analiza varjenja TIG jekla maraging z 18 % Ni ter
analiza izlo~evalnega `arjenja varov glede na mikro-
strukturo in mehanske lastnosti je pokazala naslednje:
• varjenje z vnosom energije do 2,1 kJ/mm je prakti~no
in povzro~a le delno izlo~evalno `arjenje vara, zato
topilno `arjenje po varjenju ni potrebno;
• temperatura predgrevanja 400 °C je previsoka za
varjenje jekel maraging z vsebnostjo 18 % Ni ali ve~.
Priporo~a se predgrevanje do 200 °C;
• kovanje varkov je za`eleno, saj zvi{uje `ilavost in
trdoto varkov ter spreminja natezne zaostale nape-
tosti v tla~ne;
• izlo~evalno `arjenje poteka hitreje pri vi{jih tem-
peraturah. S pove~evanjem temperature in ~asa
izlo~evalnega `arjenja se natezna trdnost in trdota
pove~ujeta, `ilavost pa pada;
• dobro starano mikrostrukturo dobimo npr.: pri
izlo~evalnem `arjenju na temperaturi 515 °C eno uro,
pri temperaturi 445 °C tri ure oz. z uporabo razvitih
modelov, ki so prikazani na sliki 6;
• premalo starano jeklo maraging ni primerno za
povr{inski sloj orodij, ker ima prenizko trdoto, pri
preve~ staranem materialu pa se pojavi ob~uten
padec `ilavosti.
6 LITERATURA
1 A. Srivastava, V. Joshi, R. Shivpuri, Computer modeling and
prediction of thermal fatigue cracking in die-casting tooling, Wear
256 (2004), 38–43
2 A. Persson, S. Hogmark, J. Bergstrom, Simulation and evaluation of
thermal fatigue cracking of hot work tool steels, International Journal
of Fatigue 26 (2004), 1095–1107
3 A. Persson, S. Hogmark, J. Bergstrom, Thermal fatigue cracking of
surface engineered hot work tool steels, Surface and Coatings
Technology 191 (2005), 216–227
4 W. Young, Why Die Casting Dies Fail, Paper No. G-T79-092, 10th
SDCE International die casting exposition & congress, St.Louis,
Missouri, North American Die Casting Association, 1979
5 Metso Powdermet, Materials Technology Solutions, Marlok, Longer
die life Better quality, www.metsopowdermet.com/, (2005)
6 Y. Wang, A study of PVD coatings and die materials for extended
die-casting die life, Surface and Coatings Technology 94–95 (1997),
60–63
7 R. F. Decker, S. Floreen, Maraging steels – The first 30 years,
Wilson, R. K., Maraging steels – recent developments and applica-
tions, The Metals & Minerals Society, Warrendale, PA, 1988
8 J. Grum and M. Zupan~i~, Suitability assessment of replacement of
conventional hot-working steels with maraging steel, Part I:
Mechanical properties of maraging steel after precipitation hardening
treatmen, Zeitschrift fuer Metallkunde/Materials Research and
Advanced Techniques 93 (2002), 164–170
9 J. Grum, M. Zupan~i~, Suitability assessment of replacement of
conventional hot-working steels with maraging steel, Part II:
Microstructure of maraging steel after precipitation hardening
treatment, Zeitschrift fuer Metallkunde/Materials Research and
Advanced Techniques 93 (2002), 171–176
10 D. A. Canonico, Gas Metal-Arc Welding of 18 % Nickel Maraging
Steel, Welding Research (1964), 433–442
11 D. Klob~ar, J. Tu{ek, B. Taljat, G. Scavino, Influence of thermal
fatique on materials for die-casting tooling, Rosso, M., Actis Grande,
M., in Ugues, D., Tooling materials and their applications from
research to market: proceedings of 7th International Tooling
Conference, 2.5.2006 Torino, Politechnico di Torino, 479–485
12 A. T. Minotto, B. Taljat, J. Tu{ek, Development of weld-cladding
processes for surface enhancement of hot-work tooling, 5th Inter-
national Conference on Industrial Tools, ICIT 2005, 12.4.2005
Velenje, TECOS Celje, 207–214
D. KLOB^AR, J. TU[EK: MEHANSKE LASTNOSTI ZVARA IZ JEKLA MARAGING PO IZLO^EVALNEM @ARJENJU
Materiali in tehnologije / Materials and technology 41 (2007) 4, 167–171 171

S. SEDMAK ET AL.: AN EXPERIMENTAL VERIFICATION OF NUMERICAL MODELS ...
AN EXPERIMENTAL VERIFICATION OF NUMERICAL
MODELS FOR THE FRACTURE AND FATIGUE OF
WELDED STRUCTURES
EKSPERIMENTALNA VERIFIKACIJA NUMERI^NIH MODELOV
ZA PRELOM IN UTRUJENOST ZVARJENIH STRUKTUR
Stojan Sedmak1, Aleksander Sedmak2, Miodrag Arsi}3, Jelena Vojvodi~ Tuma4
1Faculty of Technology and Metallurgy, Karnegijeva 4, 11000 Beograd, Serbia & Montenegro
2Faculty of Mechanical Engineering, Kraljice Marije 16, 11000 Beograd, Serbia & Montenegro
3Institute for Material Testing, Bulevar vojvode Mi{i}a 43, 11000 Beograd, Serbia & Montenegro
4Institute of Metals and Technology, Lepi pot 11, 10000 Ljubljana, Slovenia
Prejem rokopisa – received: 2006-05-17; sprejem za objavo – accepted for publication: 2006-12-19
By allowing a more detailed analysis, numerical modelling has an important role to play in the development of structures.
However, the results of a numerical model’s analysis have to be verified with an experimental analysis for the application. Two
full-scale welded-structure tests were used for the verification of the proposed numerical models. In the first case a pressure
vessel pre-cracked in weld metal and instrumented for a J-integral measurement was tested. In this way the crack’s driving
force, expressed as the J integral, could be determined directly. The crack’s driving force was calculated by applying the
Ratwani-Erdogan-Irwin (REI) numerical model. The experimental results were compared to the values obtained with the REI
model. In was shown that the application of the model produces conservative results and that the model can be applied for a
structural integrity assessment of cracked pressure vessels. The second case was the welded structure of a rotor excavator. The
critical welded components were modelled in full scale. The loading was recorded with strain gauges at critical locations,
analyzed and elaborated in the form of the stress spectrum applied in the testing. The experimental results were compared to
those obtained with numerical models. The comparison revealed that the model significantly overestimates the real cycle
number for the fatigue limit. These two examples clearly demonstrated the complexity of the problem. In the case of the
pressure vessel the model results are conservative, while those obtained for the fatigue are too optimistic.
Keywords: welded joint, numerical model, experiment, fracture, fatigue
Numeri~no modeliranje je zelo pomembno za razvoj struktur, ker omogo~a bolj natan~no analizo, vendar je potrebno rezultate
numeri~ne analize eksperimentalno verificirati. Dve varjeni strukturi sta bili uporabljeni za verifikacijo predlo`enih numeri~nih
modelov. V prvem primeru je bila preverjena posoda pod pritiskom z razpoko v zvaru. Posoda je bila instrumentirana za
neposredno dolo~itev J-integrala. Gonilna sila za propagacijo razpoke je bila izra~unana z numeri~nim Ratwani-Erdogan-
Irwinovim (REI) modelom. Eksperimentalne rezultate smo primerjali z REI-vrednostmi. Pokazalo se je, da so rezultati modela
konservativni pri oceni integritete v primeru razpoke v posodi pod pritiskom. Drugi analiziran primer je bila zvarjena struktura
dela rovokopa~a. Kriti~ni zvarjeni del je bil modeliran v pravi velikosti. Obremenitev je bila registrirana z merilnimi trakovi na
razli~nih mestih, nato pa analizirana in predstavljena v obliki spektra napetosti. Rezultati preizkusa so bili primerjani z rezultati
iz numeri~nega modela. Primerjava je pokazala, da model pomembno preceni realno {tevilo obremenitev do utrujenostne
trdnosti. Ta dva primera ka`eta na kompleksnost problema. V primeru posode pod pritiskom so rezultati modela konservativni,
v primeru utrujenosti pa optimisti~ni.
Klju~ne besede: zvarni spoj, numeri~ni model, preizkus, prelom, utrujenost
1 INTRODUCTION
The impressive developments in fracture mechanics
and the numerical modelling of structures during the past
40 years has enabled the improvement of existing and
the introduction of new methods for the evaluation of
residual life and the assessment of structural integrity 1.
By applying these methods the service safety of
structures has been increased and their life has been
extended, resulting in significant cost savings. Today,
numerical modelling is an invaluable tool for the design
of different structures, as well as the manufacturing and
use of steel structures, power and petrochemical plants,
aircraft, machinery and vehicles. Welded structures have
a very important role in many of these sectors and,
therefore, require special attention. The reason for this is
the possibility of cracks occurring in welded joints,
which can endanger the structural integrity and nega-
tively affect the service safety. Many standards and
documents consider this structural integrity, and
probably the most important is the SINTAP (Structural
INTegrity Assessment Procedure) 2,3, which is based on
fracture-mechanics analyses and experience with cracks
in a welded structure 4. The fracture-mechanics approach
was applied with success and formally accepted in the
case of the Trans Alaska Crude Oil Pipeline 5, for a
"fitness-for-purpose" approach. The maximum allowed
crack size was verified experimentally before the
proposal for when and how to repair the cracks in
welded joints. The most important conclusion in this
investigation was that "fracture-mechanics analysis is an
acceptable basis for an allowable exception from valid
standards under circumstances, under the condition that
this analysis provides a clear and conservative structural
integrity assessment".
Materiali in tehnologije / Materials and technology 41 (2007) 4, 173–178 173
UDC/UDK 539.42:621.791.05 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 41(4)173(2007)
Numerical models are developed based on the
available data, which do not necessarily coincide in all
cases with the real situation. Two important require-
ments before the acceptance of a proposed model are:
1. the model has to be conservative in order to cover
different circumstances,
2. the model must be experimentally verified.
Experiments with full-scale welded structures are not
frequently performed, because they are connected with
many problems in their organization and realisation, and,
not surprisingly, they are very expensive. The obtained
results are, however, generally valid for the tested
structure and the applied testing conditions. For this
reason, numerical modelling has an important role to
play in the development of structures since it allows a
more detailed analysis. In any case, it is questionable
whether the results of the numerical model analysis
represent the real situation of the considered structure,
because a large number of influential parameters have to
be considered. The answer to this question can only be
obtained from a proper experimental analysis.
2 CRACK-GROWTH ANALYSIS WITH THE
J-INTEGRAL
Two aspects of the J integral have to be considered
for a structural integrity assessment. In the first the J
integral is an elastic-plastic fracture-mechanics para-
meter that defines the cracked body’s geometry and the
loading (the crack driving force – CDF), and in the
second it represents the crack resistance of the material
(JIc and J–R curve) 6. The fundamental difference bet-
ween these two aspects is the crack-growth behaviour. In
the first case the crack size is not variable and it is used
as a parameter with the stress. In the second case,
however, the crack growth is included. The criterion for
the initiation of stable crack growth is mathematically
defined as:
J a J( , )σ ≥ Ic (1)
where J( ,a) is the crack driving force (CDF), depen-
ding on the remote stress,  , and the crack length, a,
whereas JIc is the material’s resistance to the initiation
of stable crack growth. In the case of elastic-plastic
fracture mechanics the crack growth analysis is not
restricted to the application of Equation (1), but also
involves the stable crack growth and the condition for
the initiation of its unstable spreading, e.g., the J–R
curve, which is compared to the CDF using a con-
venient graphical method. The initiation of unstable
crack growth is given by:
∂
∂
∂
∂
J a
a
J
a
( , )σ ≥  
indicating that the increase in the CDF must be greater
than the increase of the material’s crack resistance for
the same crack extension. When the value J( ,a)
reaches the value JIc (1), (the intersection of the CDF
curve and the J–R curve), the stable crack growth is
initiated, and it continues up to the instant at which
J( ,a)/a becomes greater than J/a (2), (the tangent
of the CDF curve to the J–R curve) producing unstable
crack growth. The extent of the stable crack growth,  a,
is the difference (on the abscissa) of the marked points
A and ao, including the crack-tip blunting.
A theoretical–analytical model, such as the Ratwani-
Erdogan-Irwin (REI) model 7, can be applied for the
CDF determination. The crack resistance can be
determined experimentally, e.g., according to ASTM
E1820 8 or by applying the J-integral direct-measure-
ment method 9.
3 VERIFICATION OF THE STRUCTURAL
INTEGRITY ASSESSMENT MODEL
The conservative prediction of the structural integrity
of a cracked welded structure means that the material’s
crack resistance is greater than the maximum CDF.
Thus, it should be proved that the applied model for the
CDF and the residual strength prediction is conservative.
With this aim the experimental pressure vessel (Figure
1) was welded from high-strength low-alloyed SM80P
steel (700 MPa yield strength class) 16-mm steel plates,
applying a qualified welding procedure, and prepared for
S. SEDMAK ET AL.: AN EXPERIMENTAL VERIFICATION OF NUMERICAL MODELS ...
174 Materiali in tehnologije / Materials and technology 41 (2007) 4, 173–178
Figure 1: Experimental pressure vessel: I, shape and dimensions; II,
detail of crack; III, integration path; IV, distribution of strain gauges.
A notch. B fatigue pre-crack. C,D stable crack growth. E final fatigue
Slika 1: Eksperimentalna posoda pod pritiskom: I oblika in mere. II
Detail razpoke. III pot integracije. IV Porazdelitev merilnih trakov za
napetost. A zareza. B utrujenostna pred-razpoka. C, D Stabilna rast
razpoke. E Kon~na utrujenost
the verification of the REI model 10. This was performed
using a J-integral direct measurement 9 on the fatigue
pre-crack positioned in the WM (Figure 1.II). To
prepare the fatigue pre-crack, a segment of size 180 mm ×
380 mm, containing the SAW weld metal, was cut from
the welded prototype. After machining the notch in the
centre of the weld metal, a fatigue pre-crack was produced
according to the standard procedure, and the segment was
than re-welded in the pressure vessel. The properly selec-
ted contour DCBAB’C’D", given in Figure 1.III, was
covered by regularly distributed strain gauges, Figure
1.IV, and the clip gauge applied for the crack-opening
displacement enabled a direct evaluation of the J
integral. It is interesting to note that during the test the
crack increased in length, from the initial value of 2c =
64.25 mm up to 72 mm after the first stage and up to the
final value of 80 mm measured after the experiment. The
crack did not grow in terms of depth, as can also be seen
in Figure 1.II.
The shell parameter , necessary in the next calcu-
lation for the crack length 2c = 64.25 mm for the
mid-thickness shell radius 2R = 1184 mm, the wall thick-
ness W = 16 mm, and the Poisson’s ratio  = 0.3, is:
[ ] [ ]λ = − = − ⋅ =12 1 12 1 0 3
32125
1184 0 016
0 62
1 4 2 1 4( ) ( . )
.
.
.
/ /
ν (3)
The crack driving forces for the axial surface crack in
the pressure vessel were calculated using the REI model
for this value of the shell parameter and expressed by the
set of lines in Figure 2, depending on crack ratio, a/W,
and the normalized pressures pR/WRp0.2 (p is the applied
pressure, Rp0.2 is the weld metal’s yield stress). Point "A"
is experimentally obtained at a pressure of 100 bar for
the crack depth, measured after testing (a = 11 mm,
crack ratio a/W = 0.69). For the same pressure CDF,
calculated with the REI model for a/W = 0.69, is 40 %
higher than the point "A", indicating that the model is
conservative and can be used for a structural integrity
assessment.
For the J-R curve the determination of the SEN (B)
specimen was selected (Figure 3.I) for the single-
specimen technique (ASTM E1820), and instrumented
tensile panels with a surface crack (Figure 3.II) for the
S. SEDMAK ET AL.: AN EXPERIMENTAL VERIFICATION OF NUMERICAL MODELS ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 173–178 175
Figure 4. Typical plots of load, P, vs. crack-opening displacement
(COD) for a crack in the HAZ (I – tensile panel; II – three-point bend
specimen – arrow indicates "pop-in")
Slika 4. Tipi~ne odvisnosti P od premika odprtja razpoke (COD) za
razpoko v HAZ (I – natezna plo{~a; II – trito~kovni upogibni preiz-
ku{anec; pu{~ica je ozna~ba to~ke "pop-in")
Figure 2: Crack driving forces with marked point "A" for the
experimentally obtained J integral
Slika 2: Gonilna sila razpoke s to~ko A za eksperimentalno dolo~en
J-integral
Figure 3: Pre-cracked specimens (I – three-point bend specimen; II –
tensile panel, III – details of surface cracks on tensile panels)
Slika 3: Preizku{anci s pred-razpoko (I – trito~kovni upogibni
preizku{anec; II – natezna plo{~a; III – detalj; s povr{inskimi
razpokami na nateznih plo{~ah)
J-integral direct-measurement method 9. The cracks
located in the HAZ of the tensile panel are presented in
Figure 3.III.
From the extended experimental program 10 the
specimens precracked in the HAZ were selected as the
most critical for the structural integrity assessment. The
two plots of load vs. crack opening displacement are
typical, one of the uniform form (tensile panels) and the
other with "pop-in", indicating an arrested, rapidly
growing crack (some of the SEN(B) specimens) (Figure
4.II). The derived corresponding J-R curves with critical
SEN(B) specimens are presented in Figure 5.
The pop-in occurred only for the small ratio a/W =
0.21 and a/W = 0.20 in SEN(B) – TPB specimens of
HAZ (Figures 4 and 5) indicating that a through crack
can develop easily, which was not observed for tensile
panels. In a real pressure vessel the crack is similar to
that in the tensile panel and pop-in is not expected. The
lower bound (LB) curve covers the pop-ins found during
testing (Figure 6) and the obtained relations indicated
that the structural integrity was not critical because of
the crack ratio a/W = 0.548, the crack depth is ao = 8.77
mm, the pressure of p = 120 bar, significantly higher
than the operating pressure, can produce an unstable
crack growth of an additional 0.8 mm, before the crack
tip reaches the region of tougher material in which the
crack can continue to grow under increasing pressure in
a stable manner. A crack of that size is not probable in a
real pressure vessel because much smaller cracks can be
detected with non-destructive testing.
Thus, the structural integrity assessment of the
cracked pressure vessel with the considered REI model
is conservative and can be reliably applied.
4 VERIFICATION OF THE FATIGUE-CRACK
GROWTH MODEL
Service fatigue cracks frequently occur in the welded
components of rotor excavator bearing structures
(Figure 7). The program of the performed investigation
included the identification of critical regions, the
measuring of strains with strain gauges in different
service conditions, the determination of the stress
spectrum, the model design of a critical welded joint and
the model testing by constant amplitude load and using a
defined stress spectrum. The obtained experimental
results are compared to numerical solutions obtained by
applying the linear fatigue-damage accumulation
hypothesis by Palmgren-Miner 11, and a modified linear
hypothesis, Corten-Dolan 12, Serensen-Kogaev 13 and
Haibach 14.
S. SEDMAK ET AL.: AN EXPERIMENTAL VERIFICATION OF NUMERICAL MODELS ...
176 Materiali in tehnologije / Materials and technology 41 (2007) 4, 173–178
Figure 7: Cracks in critical regions of the welded components of a
rotor excavator
Slika 7: Razpoke v kriti~nih podro~jih zvarjenih komponent rotorja
rovokopa~a
Figure 5: Scatter band J–R curves set of SEN(B) – TPB specimens for
HAZ with two cases of pop-ins for a/W values 0.21 and 0.20
Slika 5: Trosenje obmo~ja krivulj J–R za SEN(B) – TPB-preizku-
{ance za HAZ z dvema primeroma "pop-in" za vrednosti a/W 0,21 in
0,20
Figure 6: Structural integrity assessment for HAZ J–R curves lower
bound (LB), including specimens with pop-in
Slika 6: Ocena integritete strukture za HAZ J–R-krivulje spodnjega
praga (LB) z vklju~enim "pop-in"
The specimen for experimental fatigue testing was
modelled as a critical cross-welded joint to simulate the
bearing structure component exposed to cracking
(Figure 8). The applied variable loading of the bearing
structure consisted of transport (4 %), average operation
(60 %) and full cut (36 %), as part of the digging process
and by low-frequency self-vibrations. The time-
dependent complex stresses of the structural elements
are presented in Figure 9, as discriminated according to
the stress origin in order to establish a stress spectrum
simulating real-service loading 15. The total spectrum of
stress ranges also involved the pick loadings, repre-
senting 0.05 % to 10 % of the total operating time, which
are so high that they can arrest the excavator operation.
The experiments were performed at constant amplitude
on a high-frequency resonant pulsator and by a defined
spectrum on a servo-hydraulic closed-loop device. The
obtained relations of stress, S, vs. the number of cycles,
N, are presented in the double logarithmic diagram in
Figure 10, designed as a F. S. for the constant amplitude
loading (S6.7N = const), and as an A. S. for the simulating
spectrum. The numerically obtained results for all the
applied hypotheses are also presented for comparison.
The comparison of the fatigue testing results shows
that the effect of the constant amplitude is more obvious
(Figure 10), indicating that the structure in real
operating conditions with a different amplitude can
exhibit a longer fatigue life than expected, based on
laboratory tests using a constant amplitude. The four
hypotheses considered revealed a too optimistic fatigue
life compared to the experimental results, overestimating
the number ND for the fatigue limit. For this reason they
should be applied with caution. It should be noted that
the Haibach hypothesis is proposed for welded joints.
5 CONCLUSION
When applying the experimental results it is clear
that the numerical models in the structural analysis have
to be applied with a necessary caution because their
efficiency is dependent on the considered influencing
factors, since all the relevant factors and their real effect
on the situation must be considered. It was found that in
the case of a welded vessel exposed to internal pressure
the applied model produces conservative – and thus
acceptable – results, and that in the case of the fatigue of
the rotor excavator’s welded components the numerical
model results overestimate the experimental values and
cannot be applied as sufficiently conservative.
Acknowledgement
The paper is prepared with the financial support of
the Serbian Ministry of Science and Environment
Protection – project ON 14027 "Special topics in
fracture mechanics of materials".
6 REFERENCES
1 R. W. Nichols: The use of fracture mechanics as an engineering tool,
The 1984 ICF Honour Lecture, ICF 6, New Delhi, India, 1984
2 SINTAP: Structural Integrity Assessment Procedure, Final Report,
EU-Project BE 95-1462. Brite Euram Programme, Brussels, 1999
3 N. Gubeljak, U. Zerbst: SINTAP- Structural INTegrity Assessment
Procedure, in From fracture mechanics to structural integrity
S. SEDMAK ET AL.: AN EXPERIMENTAL VERIFICATION OF NUMERICAL MODELS ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 173–178 177
Figure 10: Comparison of S–N curves, obtained experimentally with
constant (F. S.) and a variable amplitude loading spectrum (stress
spectrum) (A. S.) and calculated by linear accumulation hypothesis
(Corten-Dolan – C-D, Haibach – H, Serensen-Kogaev – S-K and
Palmgren-Miner – P-M)
Slika 10: Primerjava eksperimentalnih odvisnosti S-N s konstantno (F.
S.) in spremenljivo amplitudo spektra obremenitve (spekter napetosti)
(A. S.) in izra~unane s hipotezo linearne akumulacije (Corten – Dolan
– C-D, Haibach – H, Serensen-Kogaev – S-K in Palmgren-Miner –
P-M)
Figure 9: Stress variations measured in rotor excavator operation
Slika 9: Variacije napetosti, izmerjene pri delu rovokopa~a
Figure 8: Specimen for fatigue testing, model of cross-welded joint of
rotor excavator arrow
Slika 8: Preizku{anec za preizkus utrujenosti, model kri`no varjenega
spoja konice rotorja rovokopa~a
assessment, ed. Z. Radakovi} and S. Sedmak, DIVK, TMF, Beograd,
2004
4 F. M. Burdekin: The assessment of critical crack size in pressure
vessels, the monograph Modern aspects of desing and manufacturing
of pressure vessels and penstocks. ed. S. Sedmak, GO[A, TMF,
Beograd, 1982
5 R. P. Reed, H. I. MacHenry, M. B. Kasen: A fracture mechanics
evaluation of flaws in pipeline girth welds, Welding Research
Council Bulletin, No 245, 1979
6 B. Bozi}, S. Sedmak, B. Petrovski, A. Sedmak: Crack gowth
resistance of weldment constituents in a real structure, Bulletin T. Cl
de l’Academie serbe des Sciences at des Arts, Classe des Sciences
techniques, No 25, Beograd, 21–42, 1989
7 M. M. Ratwani, F. Erdogan, G. R. Irwin: Fracture propagation in
cylindrical shell containing an initial flaw, Lehigh University,
Bethlehem, 1974
8 ASTM E 1820-99: Standard Test Method for Measurement of
Fracture Toughness
9 D. T. Read: Experimental method for direct evaluation of the J
contour integral, ASTM STP 791, 1983
10 US-Yugoslav joint project Fracture mechanics of weldments, Annual
reports, Faculty of Technology and Metallurgy, Beograd, Principal
investicator S. Sedmak, 1982-1992
11 M. A. Miner: Cumulative damage in fatigue, J. Appl. Mech. Trans.
ASME, 12, 1945
12 H. T. Corten, T. J. Dolan: Cumulative fatigue damage, Proc. Int.
Conference on Fatigue of metals, ASME and IME, 1956
13 S. V. Serensen, V. P. Kogaev, R. M. Snejderovic: Bearing capacity
and calculation of machine elements resistance, in Russian,
Mashinostroenie, Moskva, 1975
14 E. Haibach: Modifizierte lineare Schädenakkumulationshypothese
zur Berücksichtigung des Dauerfestigheitsabfals mit fortschreitender
Schädigung, Laboratorium fur Betriebsfestigheit, TM, No 50/70,
Darmstadt, 1970
15 M. Arsi}: Correlation between endurance limit and fatigue threshold
of welded joints, Ph. D. thesis, in Serbian, Mechanical Engineering
Faculty, Pri{tina, 1995
S. SEDMAK ET AL.: AN EXPERIMENTAL VERIFICATION OF NUMERICAL MODELS ...
178 Materiali in tehnologije / Materials and technology 41 (2007) 4, 173–178
M. AMBRO@I^, K. VIDOVI^: IZRA^UN PARAMETROV WEIBULLOVE PORAZDELITVE ...
IZRA^UN PARAMETROV WEIBULLOVE
PORAZDELITVE ZA OCENO UPOGIBNE TRDNOSTI
VALOVITIH STRE[NIH PLO[^
COMPUTATION OF THE PARAMETERS OF THE WEIBULL
DISTRIBUTION FOR ESTIMATING THE BENDING STRENGTH
OF CORRUGATED ROOFING SHEETS
Milan Ambro`i~1, Krunoslav Vidovi~2
1Institut "Jo`ef Stefan", Jamova 39, 1000 Ljubljana, Slovenija
2Esal, d. o. o. Anhovo, Vojkova 9, 5210 Deskle, Slovenija
milan.ambrozic ijs.si
Prejem rokopisa – received: 2006-11-20; sprejem za objavo – accepted for publication: 2007-02-21
V ~lanku je opisana uporaba Weibullove porazdelitve pri vrednotenju ve~kratnih meritev nekaterih mehanskih veli~in valovitih
stre{nih plo{~ iz vlaknocementa, ki so bile izdelane v redni proizvodnji podjetja Esal, d. o. o. Anhovo. Tu se omejimo na
zlomno silo pri pre~ni upogibni obremenitvi plo{~e in zlomni moment pri vzdol`ni upogibni obremenitvi. V vsakem primeru
smo izra~unali oba Weibullova parametra, od katerih je pomemben predvsem Weibullov modul, ki podaja {irino porazdelitvene
funkcije merjene veli~ine.
Klju~ne besede: vlaknocementi, valovite stre{ne plo{~e, mehanske lastnosti, Weibullova statistika
In this paper the application of the Weibull distribution for the evaluation of repeated measurements of some mechanical
quantities on corrugated roofing sheets made from fibre-cement composites in the serial production of the company Esal d.o.o.
Anhovo is described. The focus is on the breaking force in the transversal bending, loading and breaking moment during the
longitudinal bending loading of the plate. For all cases the two Weibull parameters were calculated; especially important is the
Weibull modulus, which gives the width of the distribution function of the measured quantity.
Key words: fibre-cement composites, corrugated roofing sheets, mechanical properties, Weibull statistics
1 UVOD
Vlaknocementi (VC) so kompoziti iz cementa in
oja~itvenih vlaken, ki pove~ajo natezno in upogibno
trdnost materiala; znano je namre~, da sam hidratizirani
cement zdr`i veliko ve~je tla~ne obremenitve kot
natezne. Zaradi nevarnosti za zdravje so azbestna vlakna
v VC nadomestili z drugimi: z naravnimi (npr. lesno
celulozo iz drevesnih vrst, ki so raz{irjene na podro~ju
uporabe vlaknocementnih izdelkov) in sinteti~nimi
(steklenimi, ogljikovimi, polivinilalkoholnimi itd.).1-6 Od
sinteti~nih organskih vlaken so med najustreznej{imi
tista iz polivinil alkohola (PVA). V podjetju Esal, d. o.
o., v Anhovem, ki je me{ana dru`ba Salonita Anhovo in
Eternita iz [vice, uporabljajo PVA-vlakna za izdelavo
vlaknocementov za valovitne stre{ne plo{~e. Glede na
{tevilo celih valov v plo{~i, 5 ali 8, ozna~ujemo plo{~e
na kratko V5 ali V8. Pri razvoju novih VC gradbenih
elementov in tudi med velikoserijsko proizvodnjo je
treba s standardnimi preizkusi preveriti razli~ne mehan-
ske lastnosti materiala in izdelkov, tudi glede na namen-
skost in na klimatske razmere okolja, kjer naj bi izdelke
vgrajevali7-9.
Veliko je proizvodnih parametrov, s katerimi lahko
izbolj{amo kakovost vlaknocementnih izdelkov5,10-13. Kar
se ti~e samih oja~itvenih vlaken v cementni matrici, so
pomembni vrsta, volumenski dele`, dol`ina in poravna-
nost vlaken5. Pri izbiri vrste vlaken je treba med drugim
upo{tevati njihov elasti~ni modul, natezno trdnost in
povr{inske lastnosti, ki omogo~ajo dober spoj med
vlakni in cementno matrico. Poleg optimalnih mehanskih
lastnosti izdelkov je treba gledati tudi na proizvodne
stro{ke, saj so sinteti~na vlakna relativno draga. Tako je
najugodnej{i volumenski dele` PVA-vlaken nekaj
odstotkov.
Izmerjene vrednosti zna~ilnih mehanskih lastnosti
kon~nih izdelkov, npr. zlomne sile, navadno ustrezajo
Weibullovi porazdelitvi, posebno pri krhkih materialih,
kot sta keramika in cement14-19. Weibullovo porazdelitev
so uspe{no uporabili na {tevilnih podro~jih, npr. v
strojni{tvu, gradbeni{tvu, pri in`enirski keramiki in
biokeramiki20-24. Navadno se uporablja 2-parametri~na
Weibullova porazdelitev, ki bo podrobneje opisana v
nadaljevanju, v nekaterih primerih pa je ustreznej{a
uporaba 3-parametri~ne Weibullove porazdelitve.
Pri vsaki seriji izdelanih plo{~ izmerimo v Esal-u
nekatere mehanske lastnosti na nekaj vzor~nih plo{~ah,
to je navadno od 12 do 15 preizkusnih plo{~ na teden.
Tako se je nabralo `e veliko {tevilo meritev in v tem
prispevku bomo spoznali, da se dajo izmerjene mehan-
ske koli~ine na plo{~ah dobro opisati z 2-parametri~no
Weibullovo porazdelitvijo. Opisana bo koristnost upo-
rabe Weibullove porazdelitve pri napovedi mehanskih
Materiali in tehnologije / Materials and technology 41 (2007) 4, 179–184 179
UDK/UDC 539.4:691.54 ISSN 1580-2949
Izvirni znanstveni ~lanek/Original scientific article MTAEC9, 41(4)179(2007)
lastnosti izdelkov v velikoserijski proizvodnji. Za
in`enirja in za uporabnika vlaknocementnih izdelkov so
Weibullovi diagrami nazoren prikaz statisti~ne poraz-
delitve vrednosti merjene veli~ine in s tem mehanske
zanesljivosti vlaknocementnih izdelkov.
2 SESTAVA IN DIMENZIJE VALOVITIH
STRE[NIH PLO[^ V5
O sestavi in izdelavi vlaknocementnih izdelkov po
Hatzschekovemu postopku je v tej reviji `e bil objavljen
prispevek11.Tu omenimo le, da so vhodne surovine za
izdelavo stre{nih plo{~ portlandski cement, voda, polnila
in vlakna. Poleg oja~itvenih PVA-vlaken se uporabljajo
tudi celulozna vlakna, med drugim zaradi olaj{anja
proizvodnega postopka. Skupni masni dele` celuloznih
in PVA-vlaken v navadni redni proizvodnji stre{nih
plo{~ je 6,1 % glede na trdne sestavine (brez vode in
zraka); od tega je 1/3 masnega dele`a PVA-vlaken in 2/3
celuloznih vlaken. Pri opisu dimenzij se omejimo na
plo{~e V5, ki jih v Esalu izdelajo ve~ kot plo{~ V8.
Geometrijo plo{~ V5 dolo~ajo parametri: {irina W = 920
mm, dol`ina L = 1250 mm, valovna dol`ina profila   =
177 mm, vi{ina profila (dvojna amplituda vala) H = 51
mm, debelina T  6 mm.
3 MERJENJE MEHANSKIH LASTNOSTI
PREIZKUSNIH PLO[^
Neposredno pred mehanskimi preskusi se plo{~e
namakajo 24 h v vodi. S tem simuliramo slab{e vremen-
ske razmere; po namakanju se namre~ poslab{ajo
mehanske lastnosti plo{~.
Izmerili smo razli~ne mehanske lastnosti plo{~ V5,
od katerih v tem prispevku opi{emo le zlomno silo pri
pre~ni upogibni obremenitvi (glede na valove plo{~,
slika 1a) in zlomni moment pri vzdol`ni obremenitvi
(slika 1b). Za merjenje zlomne sile smo uporabljali
laboratorijsko merilno napravo BP-10, Walter+Bai AG,
[vica, ki ima merilno obmo~je od 2 kN do 10 kN. Glede
eksperimentalnih pogojev, kot so geometrijski parametri,
smo upo{tevali standarda EN 49425 in DIN 274/126.
Razdalje na sliki 1a so podane v milimetrih. Naprava
neposredno izmeri silo FT pri pre~nem (transverzalnem)
zlomu, medtem ko zlomni moment ML pri vzdol`nem
(longitudinalnem) zlomu izra~unamo iz zlomne sile FL in
geometrijskih parametrov:
M
F L
LL
L s=
4
(1)
kjer je Ls razmik med sredinama podpor, L pa dol`ina
plo{~e. Enota za moment je sicer N m, vendar pa ra~u-
namo zlomni moment na dol`insko enoto plo{~e; da
poudarimo to renormalizacijo zlomnega momenta,
bomo zanj pisali enoto N m/m. Relativne napake pri
meritvah so pribli`no 0,5 % za silo in 0,1 % ali manj za
dol`inske dimenzije. Zato lahko iz ena~be (1) ocenimo
{e relativno napako za zlomni moment, to je 0,7 %.
Omenimo {e, da lahko pri obeh na~inih upogibne
obremenitve izra~unamo tudi druge veli~ine, npr.
efektivno upogibno trdnost materiala. Zaradi valovite
geometrije plo{~ je treba upogibne trdnosti ra~unati
numeri~no, in merilna naprava je povezana z ra~unal-
nikom, ki ima ustrezni ra~unalni{ki program.
4 STATISTI^NA OBDELAVA PODATKOV
Velikokrat se pri statisti~ni obravnavi izmerjenih ali
izra~unanih podatkov zadovoljimo z izra~unom pov-
pre~ne vrednosti in standardne deviacije veli~ine, ki pa
nam ne povesta vse informacije o statisti~ni porazdelitvi
vrednosti merjene veli~ine. Zato je priporo~ljivo najprej
ugotoviti (~e je to mogo~e!), za katero statisti~no
porazdelitveno funkcijo v danem primeru sploh gre,
potem pa najti proste parametre te funkcije. ^eprav se
verjetno v ve~ini primerov pri statisti~ni obravnavi za
ve~je mno`ine podatkov uporabi Gaussova porazdelitev,
je za nekatere mehanske lastnosti konstrukcijskih mate-
rialov (kovine, keramika, cement in beton) ustreznej{a
Weibullova porazdelitev.
M. AMBRO@I^, K. VIDOVI^: IZRA^UN PARAMETROV WEIBULLOVE PORAZDELITVE ...
180 Materiali in tehnologije / Materials and technology 41 (2007) 4, 179–184
Slika 1: Geometrija pri pre~ni (a) in vzdol`ni (b) upogibni
obremenitvi plo{~e V5 glede na evropska standarda EN 494 in DIN
274/1
Figure 1: Geometry for transversal (a) and longitudinal (b) bending
loading of the plate V5, in agreement with the European standards EN
494 and DIN 274/1
Ozna~imo merjeno veli~ino z x. Njeno porazdelitev
lahko opi{emo s katerokoli od naslednjih dveh funkcij.
Prva je navadna porazdelitvena funkcija ali verjetnostna
gostota p(x), tako da pomeni njen dolo~eni integral
P a x b p x x
a
b
( ) ( )≤ ≤ = ∫ d (2a)
verjetnost, da bo izmerjena vrednost veli~ine x le`ala
med vrednostima a in b. Druga funkcija je kumulativna
porazdelitvena funkcija
P x p x x
x
x
( ) ( ' ) '
min
= ∫ d (2b)
ki pomeni verjetnost, da bo izmerjena vrednost dane
veli~ine le`ala med teoreti~no najmanj{o mo`no vred-
nostjo xmin in variabilno vrednostjo x.
Tako sta 2-parametri~ni Weibullovi porazdelitveni
funkciji naslednji:
p x
m
x
x
x
x
x
m m
( ) exp=
⎛
⎝
⎜
⎞
⎠
⎟ −
⎛
⎝
⎜
⎞
⎠
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
−
0 0
1
0
(3a)
P x
x
x
m
( ) exp= − −
⎛
⎝
⎜
⎞
⎠
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟1
0
(3b)
Weibullova parametra sta Weibullov modul m in
umeritveni parameter x0 (angle{ko scale parameter).
Parameter x0 v literaturi imenujejo tudi karakteristi~ni
parameter: ~e je merjena veli~ina x na primer sila F,
potem ta parameter ozna~imo z F0 in ga imenujemo
karakteristi~na sila. Weibullov modul je brezdimenzijski
in je za zna~ilne krhke snovi veliko ve~ji od 1, karak-
teristi~ni parameter pa ima dimenzijo spremenljivke x.
Oba parametra dolo~ata {irino krivulje p(x): ~im ve~ji je
m in ~im manj{i x0, tem o`ja je krivulja, hkrati x0 podaja
tudi pri~akovano vrednost spremenljivke x (ena~bi 5).
^eprav funkciji p(x) in P(x) dajeta ekvivalentno
informacijo o statisti~ni porazdelitvi, je za relativno
majhno {tevilo izmerjenih vrednosti spremenljivke x
primernej{a direktna uporaba funkcije P(x). Funkcijo
P(x) bomo na kratko imenovali Weibullova funkcija.
Njeni mejni vrednosti sta P(0) = 0 in P() = 1, pri ~emer
je teoreti~no najmanj{a mo`na vrednost veli~ine x kar
enaka ni~. Omenimo {e, da dobimo 3-parametri~no
Weibullovo statistiko iz 2-parametri~ne tako, da dodamo
{e parameter premika xmin in v desnih straneh ena~b (3)
naredimo transformacijo x  x – xmin; to pomeni, da je
teoreti~no najmanj{a mo`na vrednost veli~ine x enaka
xmin namesto ni~. Sliki 2 prikazujeta 2-parametri~no
Weibullovo funkcijo za m = 10 v naravni skali in v skali,
kjer je graf lineariziran. Sliki ponazorita tudi geome-
trijski pomen obeh parametrov: Weibullov modul m je
smerni koeficient premice v lineariziranem grafu, pri x =
x0 pa je verjetnost P enaka 1 – 1/e  63,2 %, temu pa
ustreza to~ka (0, 0) v lineariziranem grafu (kro`ca na
obeh slikah).
Pri danem statisti~nem vzorcu imamo N izmerjenih
ali izra~unanih vrednosti spremenljivke x, ki jih ozna-
~imo z xi. Cilj je najti Weibullova parametra, ki najbolj
ustrezata statisti~nemu vzorcu16,27-32. Vrednosti xi najprej
uredimo po velikosti od najmanj{e do najve~je. Nato
vsakemu (i-temu po vrsti) izmerku priredimo {e ocenje-
no verjetnost Pi; za kar obstaja ve~ na~inov, najve~krat
pa se uporablja preprosta ena~ba:
P
i
Ni
= −
+
0 3
0 4
,
,
(4)
Tako dobimo N urejenih parov (xi, Pi), ki jim
priredimo Weibullovo funkcijo, tako da se jim najbolj
prilega. Pri tem si pomagamo z linearizacijo ena~be (3b),
kot prikazuje slika 2b. Povle~i moramo premico, ki se
najbolj prilega mno`ici to~k v transformiranem kordi-
natnem sistemu. Namesto prikaza na sliki 2b raje upo-
rabljamo posebne Weibullove diagrame z originalnima
spremenljivkama in nelinearno skalo, kot je prikazano na
sliki 3.16
Ko sta Weibullova parametra znana, lahko izra~u-
namo razli~ne statisti~ne veli~ine, kot sta pri~akovana
vrednost <x> in standardna deviacija σx veli~ine x:
< >= ⋅ +⎛
⎝⎜
⎞
⎠⎟
x x
m0
1
1Γ (5a)
s x
m m
x = ⋅ +
⎛
⎝⎜
⎞
⎠⎟
− +⎛
⎝⎜
⎞
⎠⎟0
21
2
1
1Γ Γ (5b)
kjer je  gama funkcija.
Ker sklepamo o statisti~ni verjetnosti iz omejenega
{tevila podatkov, sta izra~unana parametra m in x0
pravzaprav le oceni in ne natan~ni teoreti~ni vrednosti,
pa ~eprav bi bila sama porazdelitev res natan~no
M. AMBRO@I^, K. VIDOVI^: IZRA^UN PARAMETROV WEIBULLOVE PORAZDELITVE ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 179–184 181
0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4
0,0
0,2
0,4
0,6
0,8
1,0
(1, 0,632)
(a)
P
x/x
0
-1,0 -0,5 0,0 0,5 1,0
-10
-5
0
5
10
(0, 0)
(b)
ln
(l
n
(1
/(
1
-P
))
)
ln(x/x
0
)
Slika 2: Graf funkcije P(x/x0) za m = 10; a) naravni spremenljivki, b)
linearizirani graf s prirejenima spremenljivkama. Obmo~je neodvisne
spremenljivke v grafu (b) ustreza intervalu x/x0 od 1/e do e, v mejnih
to~kah pa je verjetnost P prakti~no 0 in 1.
Figure 2: Diagram of the function P(x/x0) for m = 10; a) natural
variables, b) linear diagram with transformed variables. The range of
the independent variable in diagram (b) corresponds to the interval
x/x0 from 1/e from e, while the probability P is practically 0 and 1,
respectively, for these limit points.
Weibullova.27 Zato navadno podajamo interval 90 %
zaupanja (na kratko 90 % IZ); npr. ~e je za parameter m
ta interval enak 90 % IZ = 8–12, to pomeni, da lahko
pri~akujemo z verjetnostjo 90 %, da je prava vrednost m
res med 8 in 12.
5 REZULTATI IN RAZPRAVA
Kot zgled vzemimo izmerjene vrednosti FT in ML za
nebarvane plo{~e V5 z zgoraj navedenimi merami iz
redne proizvodnje v letih 2003 in 2004, kar nam da
okrog 400 meritev za obe leti skupaj za vsako mehansko
veli~ino. Vrednosti smo ovrednotili z Weibullovo
statistiko, kjer je spremeljivka x sila FT ali moment ML.
Rezultate za N = 50, 100, 200 in 400 podatkov za obe
veli~ini prikazujeta tabeli 1 in 2; pri tem za prve tri
vrednosti N vzamemo po vrsti podatke iz leta 2003, za N
= 400 pa podatke obeh let. Po standardu EN 494 naj bi
se sicer sila FT prera~unala na 1 m {irine plo{~e, vendar
je zaradi nazornosti v tabeli 1 prikazana izmerjena sila
za dejansko {irino 920 mm; prera~un za {irino 1 m bi
nam dal nekaj ve~je vrednosti, kot so v tabeli. Weibullov
modul m je v vseh primerih reda velikosti 10, kar je
zna~ilno za krhke konstrukcijske materiale, kot so
keramika in cementni kompoziti. Korelacijski koeficient
 v zadnjem stolpcu tabele pove, kako dobro se Wei-
bullova funkcija prilega eksperimentalnim podatkom; pri
tem  = 1 pomeni popolno ujemanje. Zaradi ve~je
nazornosti je v tabeli zapisan v odstotkih. Ugotavljamo,
da je  v vseh primerih nad 96 %, torej Weibullova
porazdelitev zelo dobro opisuje podatke. Poleg ocenjenih
vrednosti Weibullovega modula m in parametra FT0 ali
ML0, ki ustreza parametru x0 v ena~bah (3), prikazujeta
tabeli tudi ustrezne 90-odstotne intervale zaupanja.
Intervali zaupanja za m so za N = 50 dokaj {iroki, ker je
to {e vedno premajhen vzorec za zares zanesljivo stati-
stiko. Z nara{~ajo~im N se vsi intervali zaupanja
postopoma o`ajo, in za m je 90 % IZ okrog desetine
ocenjene vrednosti m {ele pri N = 400. Za preizkus smo z
uporabo naklju~nega generiranja {tevil izvedli tudi
numeri~no simulacijo Weibullove porazdelitve za dan
par parametrov m in x0 in izra~unane „naklju~ne”
vrednosti xi statisti~no obdelali podobno, kot da bi bili
eksperimentalni izmerki. Ugotovili smo podobno o`anje
90 % IZ za m in x0 kot pri obdelavi pravih eksperimen-
talnih podatkov. V tabelah sta podane tudi ocene za
pri~akovano vrednost (PV) in standardno deviacijo (SD)
veli~in, izra~unane iz ena~b (5); te vrednosti so blizu
vrednostim, dobljenimi s standardnimi statisti~nimi
obrazci, npr.:
< >=
=
∑x N x ii
N1
1
Tako kot za oba Weibullova parametra lahko izra~u-
namo tudi 90%IZ za PV in SD zlomne sile in momenta,
vendar jih tu ne navajamo. Sliki 3 prikazujeta prila-
goditev Weibullovih diagramov v nelinearni skali ekspe-
rimentalnim podatkom (N = 50) za zlomno silo in
moment pri obeh na~inih obremenitve. Narejeni sta bili s
komercialnim programom za Weibullovo porazdelitev
Reliasoft’s Weibull ++.16 ^im ve~ja je strmina premice,
tem o`ja je statisti~na porazdelitev in manj{a je
verjetnost, da bodo v vsakdanji rabi izdelki odpovedali
pri relativno majhnih mehanskih obremenitvah. Pri
znanih Weibullovih parametrih lahko s preureditvijo
ena~be (3b) izra~unamo, kolik{na je pri podani verjet-
nosti, npr. P = 10 %, mejna vrednost veli~ine (sile ali
momenta), tako da pri~akujemo z verjetnostjo P, da bo
meritev koli~ine dala vrednost, manj{o od mejne vred-
nosti. Nekaj zgledov je prikazanih v tabeli 3 za
Weibullova parametra, ki ustrezata tabelama 1 in 2 pri
N = 400. Tabelo 3 je treba pravilno razumeti: na primer,
podatek za mejno silo 3518 N pri verjetnosti 0,1 %
M. AMBRO@I^, K. VIDOVI^: IZRA^UN PARAMETROV WEIBULLOVE PORAZDELITVE ...
182 Materiali in tehnologije / Materials and technology 41 (2007) 4, 179–184
Tabela 1: Parametri Weibullove porazdelitve za pre~no zlomno silo FT
Table 1: Weibull parameters for the transversal breaking force FT
N
m FT0/N PV, SD/N
 /%
ocena 90 % IZ ocena 90 % IZ <FT> FT
50 11,63 9,86−13,72 6625 6475−6778 6341 661 97,0
100 12,71 11,40−14,16 6452 6355−6551 6196 594 97,5
200 11,70 10,83−12,65 6167 6093−6242 5904 612 96,7
400 12,29 11,63−12,98 6172 6124−6219 5919 586 98,3
Tabela 2: Parametri Weibullove statistike za vzdol`ni zlomni moment ML
Table 2: Weibull parameters for the longitudinal breaking moment ML
N
m ML0/ (N⋅m/m) PV, SD/(N⋅m/m)
 /%
ocena 90 % IZ ocena 90 % IZ <ML> ML
50 10,74 8,90−12,95 99,81 97,57−102,10 95,23 10,71 99,0
100 11,32 9,93−12,89 98,52 97,01−100,05 94,20 10,08 98,6
200 11,04 10,11−12,06 96,54 95,44−97,65 92,22 10,10 98,6
400      −    − 	
	   
pomeni, da pri~akujemo, da bo po~ila komaj ena od tiso~
plo{~ pri pre~ni sili manj kot 3518 N.
Tabela 3: Mejne vrednosti zlomne sile in zlomnega momenta pri dani
verjetnosti
Table 3: Limiting values of the breaking force and breaking moment
for a given probability
P/(% FT /N ML/(N⋅m/m)
10 5139 79,97
1 4245 64,85
0,1 3518 52,79
6 SKLEP
Dvoparametri~na Weibullova funkcija dobro opisuje
porazdelitev zlomnih sil in momentov pri pre~ni in
vzdol`ni upogibni obremenitvi valovitih stre{nih plo{~
V5 iz vlaknocementa. Vizualizacija podatkov z grafi
(sliki 3) daje nazoren prikaz ujemanja med meritvami in
Weibullovo porazdelitvijo. ^im ve~ja je strmina pre-
mice, kar pomeni ve~ji Weibullov modul m, tem ve~ja je
mehanska zanesljivost izdelkov, tj. manj{a je verjetnost
(pri istem parametru x0), da se bodo plo{~e zlomile pri
majhnih obremenitvah. To je posledica dejstva, da
pomeni ve~ji Weibullov modul manj{e nihanje zlomnih
obremenitev – manj{o standardno deviacijo. Ugotovitev
ponazorimo z nekaj {tevilkami. Vzemimo zaokro`eno
vrednost karakteristi~ne zlomne sile F0 = 6000 N,
Weibullov modul pa naj bo 10 ali 15. Vrednost m ne
vpliva bistveno na povpre~no zlomno silo <F0>: ta je
enaka 5708 N pri m = 10 in 5794 N pri m = 15. Mo~no
pa se spremeni verjetnostna porazdelitev za manj{e sile.
Za zgled vzemimo mejno silo 4000 N: pri m = 10 je
verjetnost, da se izdelek zlomi pri manj{i sili od dane
vrednosti 4000 N, enaka 1,72 %, pri m = 15 pa je ta
verjetnost samo {e 0,23 %.
Pri oceni statisti~nih parametrov se moramo zavedati,
da se lahko ta ocena zelo odmika od dejanske vrednosti
parametrov. Iz tabel 1 in 2 je razvidno, da je za 400
podatkov pri~akovani interval (z 90-odstotnim zaupa-
njem) za resni~no vrednost Weibullovega modula {irok
okrog 10 % izra~unane vrednosti tega parametra,
medtem ko je izra~un drugega parametra Weibullove
porazdelitve (FT0 ali ML0) zanesljivej{i.
7 LITERATURA
1 J. B. Studinka, Asbestos substitution in the fibre cement industry.
The International Journal of Cement Composites and Lightweight
Concrete, 11 (1989), 73–78
2 H. Savastano, P. G. Warden, R. S. P. Coutts, Potential of alternative
fibre cements as building materials for developing areas, Cem.
Concr. Compos., 25 (2003), 585–592
3 R. S. P. Coutts, A review of Australian research into natural fibre
cement composites, Cem. Concr. Compos., 27 (2005), 518–526
4 V. Agopyan, H. Savastano, V. M. John, M. A. Cincotto, Develop-
ments on vegetable fibre-cement based materials in Sao Paolo,
Brasil: an overview, Cem. Concr. Compos. 27 (2005), 527–536
5 Y. P. Ma, B. R. Zhu, M. H. Tan, Properties of ceramic fiber rein-
forced cement composites, Cem. Conc. Res., 32 (2005), 296–300
6 A. Peled, B. Mobasher, Pultruded fabric-cement composites, ACI
Mater. J., 102 (2005), 15–23
7 S. A. S. Akers, J. B. Studinka, Ageing behaviour of cellulose fibre
cement composites in natural weathering and accelerated tests, The
International Journal of Cement Composites and Lightweight
Concrete, 11 (1989), 93–97
8 S. A. S. Akers, Micromechanical studies of fresh and weathered fibre
cement composites, The International Journal of Cement Composites
and Lightweight Concrete, 11 (1989), 117–131
9 P. Purnell, J. Beddows, Durability and simulated ageing of new
matrix glass fibre reinforced concrete, Cem. Concr. Compos., 27
(2005), 875–884
10 K. Vidovi~, B. Lovre~ek, M. Hraste, Influence of surface charge on
sedimentation and filtration behaviour of fibrous material, Chem.
Biochem. Eng. Q, 10 (1996), 33–38
11 K. Vidovi~, Vlaknocement – Lastnosti materiala in tehnologija
proizvodnje, Mater. tehnol. 38 (2004), 197–203
12 C. Negro, A. Blanco, I. S. Pio, J. Tijero, Methodology for flocculant
selection in fibre-cement manufacture, Cem. Concr. Compos., 28
(2006), 90–96
13 Beaudoin JJ. Handbook of Fiber-Reinforced Concrete – Principles,
Properties, Developments and Applications. Noyes Publications,
New Jersey, US, 1990
14 Weibull W. A statistical representation of fatigue failure in solids.
Transactions of the Royal Institute of Technology 1949, No. 27,
Stockholm
15 Weibull W. A statistical distribution function of wide applicability. J
Appl Mech. 18(1951), 293–297
M. AMBRO@I^, K. VIDOVI^: IZRA^UN PARAMETROV WEIBULLOVE PORAZDELITVE ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 179–184 183
1,00
5,00
10,00
50,00
90,00
99,00
3000,00 8000,00
Zlomna sila
FT /
/
/
N
P
%
1,00
5,00
10,00
50,00
90,00
99,00
200,00100,00
Zlomni moment
ML / (Nm/m)
P
%
a)
b)
Slika 3: Weibullova grafa P(FT) (a) in P(ML) (b) za N = 50 meritev
Figure 3: Weibull diagrams P(FT) (a) and P(ML) (b) for N = 50
measurements
16 ReliaSoft’s Weibull ++, Life Data Analysis Reference. ReliaSoft
Publishing, 1992
17 Kosma~ T, Oblak C, Jevnikar P, Funduk N, Marion L. The effect of
surface grinding and sandblasting on flexural strength and reliability
of Y-TZP zirconia ceramic. Dental Mater. 15 (1999), 426–433
18 Setien VJ, Armstrong SR, Wefel JS. Interfacial fracture tougness
between resin-modified glass ionomer and dentin using three
different surface treatment. Dent Mater. 21 (2005) 6, 498–504
19 Lewis G, van Hooy-Corstjens CSJ, Bhattaram A, Koole LH.
Influence of the radiopacifier in an acrylic bone cement on its
mechanical, thermal, and physical propertires: Barium sulfate-
containing cement versus iodine-containing cement. J. Biomed Mater
Res B; 73B (2005) 1, 77–87
20 Anton N, Ruiz-Prieto JM, Velasco F, Torralba JM. Mechanical
properties and wear behaviour of ceramic matrix composites based
on clinker portland doped with magnesia. J Mater Processing Tech;
78 (1998), 12–17
21 Toutanji HA. Evaluation of the tensile strength of cement-based
advanced composite wrapped specimens. Comp Sci Tech; 59 (1999)
15, 2261–2268
22 Caliskan S. Aggregate/mortar interface: influence of silica fume at
the micro- and macro-level. Cem Concr Compos; 25 (2003) 4–5,
557–564
23 Li QS, Fang JQ, Liu DK, Tang J. Failure probability prediction of
concrete components. Cem Concr Res; 33 (2003) 10, 1631–1636
24 Huang JS, Cheng CK. Fracture toughness variability of foamed
alumina cements. Cem Concr Res; 34 (2004) 5, 883–888
25 EN 494, Fibre-cement profiled sheets and fittings for roofing –
Product specification and test methods. December 2004
26 DIN 274/1, Asbestzement-Wellplatten – Masse, Anforderungen,
Prufungen. April 1972
27 Quinn G. Flexure strength of advanced structural ceramics: A round
robin. J. Am Ceram Soc; 73 (1990) 8, 2374–2384
28 Ritter JE, Bandyopadhyyay N, Jakus K. Statistical reproducibility of
dynamic and static fatigue experiments. Ceram Bullet; 60 (1981),
798–806
29 Johnson LG. The median ranks of sample values in their population
with an application to certain fatigue studies. Industrial Mathematics
1951; 2
30 Lloyd DK, Lipow M. Reliability: Management, Methods and Mathe-
matics. Prentice Hall, Englewood Cliffs, New Jersey,1962
31 Li GQ, Cao H, Li QS, Huo D. Theory and its Application of
Structural Dynamic Reliability. Earthquake Press, Beijing 1993
32 Wu D, Zhou J, Li Y. Unbiased estimation of Weibull parameters
with the linear regression method. J Eur Ceram Soc; 26 (2006),
1099–1105
M. AMBRO@I^, K. VIDOVI^: IZRA^UN PARAMETROV WEIBULLOVE PORAZDELITVE ...
184 Materiali in tehnologije / Materials and technology 41 (2007) 4, 179–184
Z. ADOLF ET AL.: INVESTIGATION OF THE INFLUENCE OF THE MELT SLAG REGIME ...
INVESTIGATION OF THE INFLUENCE OF THE MELT
SLAG REGIME IN A LADLE FURNACE ON THE
CLEANLINESS OF THE STEEL
RAZISKAVA VPLIVA RE@IMA @LINDRE V PONOV^NI PE^I NA
^ISTOST JEKLA
Zden k Adolf, Ivo Husar, Petr Suchánek
V[B-Technical University of Ostrava, Department of Metallurgy, 17. listopadu 15/2172, Ostrava-Poruba, 708 33, Czech Republic
zdenek.adolf vsb.cz
Prejem rokopisa – received: 2006-10-16; sprejem za objavo – accepted for publication: 2007-01-15
An experimental investigation of the influence of the ladle slag’s chemical composition, the ladle slag’s mass and the furnace
slag’s mass (during the tapping flow in the ladle) on the sulphur content in steel was carried out. The parameters of the slag and
the steel were obtained in operational conditions of oxygen LD converter melts and from the subsequent ladle furnace LF
processing.
Key words: ladle furnace, steel processing, effect of slag composition, refining effect
Eksperimentalna raziskava vpliva kemi~ne sestave, koli~ine (pri izlivu jekla v ponovco) na vsebnost `vepla v jeklu. Parametri
jekla in `lindre so dolo~eni pri izdelavi jekla v kisikovem konvertorju in obdelavi taline v ponov~ni pe~i.
Klju~ne besede: ponov~na pe~, procesiranje jekla, sestava `lindre, rafinacijski u~inek
1 INTRODUCTION
Steel refined in a ladle furnace should have the
following characteristics prior to casting:
• a chemical composition within the prescribed interval
of alloying-element content and limited amounts of
impurities,
• the required metallographic purity in terms of com-
position, magnitude, number and density of inclu-
sions,
• the required casting temperature, depending on the
steel’s liquidus temperature.
In order to achieve these parameters the steelmaker
usually has at his or her disposal a reheating ladle fur-
nace, aluminium wire and a cored wire feeder, the
mixing of steel with argon blowing or with induction
stirring and an oxygen activity measurement system.
The slag is the most important factor for ensuring the
quality of molten steel in the ladle. Ladle slag is formed
from the products of de-oxidation of the steel, the added
mixtures and from the corrosion products of the ladle
lining, particularly at the slag line. Admixtures added
intentionally to the slag ensure the required chemical
composition of the slag, its fluidity and its ability to
refine steel, i.e., for the absorption of inclusions and
unwanted elements from the steel. The admixtures
consist mostly of lime, fluorspar, calcium carbide and
fireclay. Also, synthetic slags usually containing Al2O3,
CaO, MgO, SiO2, and a minimum of iron oxides, MnO
and sulphur are used more and more often. Slags are
often prepared from waste materials, either by simple
mixing or by sintering, while the most expensive slags of
the highest quality are manufactured by re-melting the
input raw materials. These slags are usually used as a
replacement for fluorspar and for the preservation of the
required fluidity and for obtaining a sufficient refining
effect in the ladle.
In this article the slag regime in a ladle furnace
during steel refining, partly under fluorspar and partly
under alumina slags, is compared. The slag regime of
heats was evaluated with respect to the type and mass of
the additions to the ladle, the ladle slag mass, the mass of
converter slag overflowed to the ladle, the mass of
corroded ladle lining and the slag desulphurisation
capacity.
2 CHARACTERISTICS OF THE HEATS
The investigated steel I (intended for rail production)
has an increased content of carbon (0.70–0.76 %),
manganese (0.85–0.85 %) and silicon (0.30–0.40 %),
controlled amounts of sulphur (0.01–0.02 %) and limited
amounts of aluminium (0.003 %). It is, therefore,
produced according to a technology without the use of
aluminium and the deoxidation of steel in the ladle with
coke, FeSi and MnSi. At the same time, during the
tapping from the LD converter slag-forming materials
are added to the ladle, e.g., lime (1200 kg), fluorspar
(300 kg) and fireclay (150 kg), or lime, fluorspar and
synthetic slag CCA with high contents of Al2O3 (see
Table 1).
Materiali in tehnologije / Materials and technology 41 (2007) 4, 185–188 185
UDC/UDK 669.041:669.14 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 41(4)185(2007)
Table 1: Chemical composition of the synthetic slag CCA
Tabela 1: Kemi~na sestava sinteti~ne `lindre CCA
w/% Al2O3 CaO MgO SiO2 S Granularity
average 55.05 26.60 6.56 3.21 0.0295
5–50 mmmin 53.60 26.10 6.21 3.21 0.0290
max 56.50 27.10 6.91 3.21 0.0300
The refining of the rail steel was compared to that of
the structural steel II with 0.18–0.20 % C,
0.41–0.51 % Mn, 0.20–0.30 % Si, with limited sulphur
content (0.020 %) and a specified aluminium content
(0.015–0.030 %). During tapping to the ladle this steel is
deoxidised by aluminium and alloyed with FeMn and
FeSi. At the same time, lime (1200 kg) and fluorspar
(200 kg), or lime (900 kg) and synthetic slag CCA (700
kg), replacing the fluorspar, are added to the ladle.
The average chemical composition of ladle slags
based on lime and fluorspar prior to the exit from the LF
stand is given in Table 2.
Alumina slags were formed during the making of
steel I for rails with synthetic slag CCA (700 kg) with
800 kg of lime (variant A), or with 600 kg of lime
(variant B), and during the making of structural steel II
with the synthetic slag CCA (700 kg) with 900 kg of
lime. Table 3 gives their average chemical composi-
tions.
The average content of FeO in the fluorspar slags
was 0.55 %, and in the alumina slags it was 2.0 %. If we
compare both types of slags, then fluorspar slags have
approximately 8 % to 11 % more CaO, higher basicity
and a significantly higher content of CaF2, which is
added to the ladle during tapping in the form of
fluorspar. On the other hand, alumina slags have higher
contents of Al2O3 (by 9 % to 13 %) and MgO (by 1.5 %
to 3.5 %). The synthetic slag CCA is, apart from
converter slag and lime, also a source of MgO and
Al2O3.
3 RESULTS AND DISCUSSION
The mass of the slag in the ladle was calculated from
the balance of CaF2 (Steel I - fluorspar slags) and from
the balance of Al2O3 (Steel I and II - alumina slags). The
mass of converter slag that overflowed into the ladle was
calculated from the balance of CaO, and the extent of the
wear of the ladle from the balance of MgO in the ladle
slag. These data are for fluorspar slags and for two
variants of alumina slags (rail steel I or structural steel
II) summarised in Table 4.
The data in Table 4 shows that ladle slags differ
primarily in terms of their mass. Steel I fluorspar slags
are formed, apart from SiO2, from deoxidising silicon,
lime, fluorspar and fireclay. The sum of the masses of
these three components in the charge was approximately
1720 kg. The converter slag that overflowed into the
ladle during the tapping affects the mass and the
Z. ADOLF ET AL.: INVESTIGATION OF THE INFLUENCE OF THE MELT SLAG REGIME ...
186 Materiali in tehnologije / Materials and technology 41 (2007) 4, 185–188
Table 2: Chemical composition of the fluorspar slags (CaF2)
Tabela 2: Kemi~na sestava fluoridnih (CaF2) `linder
w/% CaO SiO2 Al2O3 MgO CaF2 S Basicity
Steel I - rails
average 52.9 23.5 4.2 4.9 13.29 0.64 2.25
min 50.9 21.9 3.5 4.5 11.74 0.42 2.12
max 55.2 25.2 5.1 5.7 14.48 0.75 2.52
Steel II - structural
average 56.41 19.34 12.94 7.02 4.22 0.41 2.92
min 50.30 17.40 10.90 5.30 2.84 0.28 2.52
max 59.80 21.70 14.60 14.00 6.11 0.66 3.23
Table 3: Chemical compositions of alumina slags (CCA)
Tabela 3: Kemi~na sestava aluminatnih `linder (CCA)
w/% CaO SiO2 Al2O3 MgO CaF2 S Basicity
Steel I - rails
Variant A
average 45.2 22.3 15.4 8.11 1.0 0.23 2.02
min 45.0 20.9 15.0 7.88 0.6 0.19 1.93
max 45.5 23.6 15.7 8.50 1.8 0.20 2.15
Variant B
average 44.7 20.1 17.9 8.5 0.9 0.23 2.22
min 44.0 18.9 17.4 7.7 0.2 0.14 2.13
max 45.3 21.3 18.3 9.3 1.5 0.31 2.44
Steel II - structural
average 45.23 17.89 22.05 8.51 0.10 0.20 2.53
min 41.10 15.50 18.00 6.80 0.00 0.13 2.12
max 48.30 21.10 25.30 11.00 0.27 0.31 2.95
chemical composition of the ladle slag. The mass of the
slag in the heats with fluorspar slag was approximately
250 kg lower than the average for heats with alumina
slag, for which a slightly lower overall mass of ladle slag
was used. Alumina slags are formed of lime and
synthetic slag CCA with an average mass of 1510 kg in
variant A, and 1310 kg in variant B (again, apart from
the SiO2 from the deoxidising silicon).
For steel II the fluorspar slags are formed, apart from
SiO2 and Al2O3, from deoxidising silicon and aluminium,
of lime and fluorspar with a total mass of 1400 kg; for
alumina slags fluorspar is replaced with the slag CCA
and the new mass of lime and slag CCA is 1600 kg.
The wear of the lining is expressed as the mass of
MgO released from the ladle wear into the slag. The
smallest lining wear for steel I was observed in the
variant B, while the greatest wear was observed for the
variant A of alumina slag. This is apparently related to
the considerable differences in the total mass of slag in
the ladle. The fluorspar slag shows – contrary to
expectations – on average a low wear of lining; however,
with a large scatter of values of approximately 30 kg of
MgO.
The mass of converter slag overflowed into the ladle
was, for the structural steel, three to five times greater
then for the steel I. The share of converter slag in
approximately 35 % of the total slag mass in the ladle is
the source of the relatively high content of MgO in the
ladle slag. The wear of the lining is practically the same
for both types of ladle slags, and it is more influenced by
the slag’s mass then by the slag’s chemical composition.
The parameters of the steel II desulphurisation with
fluorspar and alumina slags are given in Table 5.
Table 5: Parameters of steel II desulphurisation by fluorspar and
alumina slags
Tabela 5: Parametri raz`veplanja jekla II s fluoridno in z aluminatno
`lindro
c's w SBOF/% w SLF/% /% Ls
Fluorspar slag
average 0.009945 0.026 0.011 68.769 40.24
min 0.006744 0.016 0.006 47.619 15.56
max 0.015171 0.043 0.018 80.000 55.00
Alumina slag
average 0.002481 0.030 0.019 36.131 11.42
min 0.001426 0.018 0.011 20.000 4.48
max 0.003582 0.038 0.029 52.778 20.71
Note:
c's= sulphide capacity of slag, 1
SBOF = sulphur content in steel during tapping from the converter,
w/%
SLF = sulphur content in steel after treatment in the ladle furnace,
w/%
  = rate of steel desulphurisation after treatment in the ladle furnace,
%
Ls = distributing coefficient of sulphur (Sslag /Ssteel), 1
It can be concluded from Table 5 that fluorspar slags
have a greater desulphurisation effect than alumina slags
(see , Ls), due to their higher basicity and their higher
value of sulphide capacity given by the CaO content,
which is on average higher by 10 %.
It was observed that fluorspar slags had a greater
fluidity than alumina slags. This means that a small
content of CaF2 (up to 5 %) in the fluorspar slag
increases the fluidity of this slag in a wide range of
chemical compositions, especially of the CaO content. In
contrast to this, the liquidity of the alumina slags can
only approach that of the fluorspar slags in a relatively
narrow range of content of the more active oxides (CaO,
Al2O3, SiO2, MgO). It is necessary to increase the CaO
content to at least 50 %, the Al2O3 content to 25–30 %,
and preserve the SiO2 content at the level of 15–20 %
and the MgO content at the level of 7 % in order to
increase the sulphide capacity of the alumina slag to the
level of the fluorspar slag.
4 CONCLUSION
The slag regime of heats during the making of rail
steel I with a limited content of aluminium, increased
amounts of C, Mn and Si and with controlled amounts of
sulphur and structural steel II deoxidised by aluminium
with a limited sulphur content are compared. The
original fluorspar slag with low amounts of Al2O3 was
substituted with an alumina slag in which the fluorspar
was replaced by synthetic slag CCA.
Z. ADOLF ET AL.: INVESTIGATION OF THE INFLUENCE OF THE MELT SLAG REGIME ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 185–188 187
Table 4: Parameters of ladle slags
Tabela 4: Parametri ponov~ne `lindre
m/kg LS CS MgO lining
Steel I - rails
Fluorspar slags
average 2411 325 71.5
min 2319 291 52.6
max 2508 380 82.9
Alumina slags – variant A
average 2569 564 99.1
min 2519 431 93.4
max 2637 745 111.0
Alumina slags – variant B
average 2210 590 69.0
min 2161 521 64.9
max 2273 719 75.6
Steel II – structural
Fluorspar slags
average 3374 1495 88.4
min 2632 153 57.3
max 4092 2800 99.8
Alumina slags
average 3650 1693 86.9
min 2743 323 59.3
max 4645 3308 101.0
Note:
mass LS = mass of ladle slag, m/kg
mass CS = mass of converter slag that overflowed into the ladle, m/kg
MgO lining = wear of ladle lining, m/kg
The following conclusions from the comparison of
both slag regimes of these heats (based on 200-tonne
heats) are proposed:
a) Alumina slags formed by lime, synthetic slag CCA
and the products of steel deoxidisation have a lower
desulphurisation capacity then the fluorspar slags due
to their lower CaO content.
b) The main advantage of fluorspar slags is in their
great desulphurisation effect over a wide range of
slag chemical compositions, even for a low CaF2
content.
c) The liquidity of alumina slags is limited to a narrow
range of CaO, Al2O3, SiO2 and MgO content.
d) The mass of ladle slags is markedly affected by the
volume of furnace slag overflowed into the ladle that
has to be minimised.
e) The wear of the lining depends on the mass of the
ladle slag. The effect of the slag’s chemical
composition is of minor importance.
This investigation was carried out in the frame of
Grant project Reg. No. 106/04/0029 with the financial
support of the Grant Agency of the Czech Republic and
the project EUREKA Reg. No. OE214, with the finan-
cial support of the Ministry of Education, Youth and
Sports of the Czech Republic.
Z. ADOLF ET AL.: INVESTIGATION OF THE INFLUENCE OF THE MELT SLAG REGIME ...
188 Materiali in tehnologije / Materials and technology 41 (2007) 4, 185–188
M. M. KRGOVI] ET AL.: THE INFLUENCE OF ILLITE-KAOLINITE CLAYS’ MINERAL CONTENT ...
THE INFLUENCE OF ILLITE-KAOLINITE CLAYS’
MINERAL CONTENT ON THE PRODUCTS’ SHRINKAGE
DURING DRYING AND FIRING
VPLIV VSEBNOSTI GLIN ILINIT-KAOLINIT NA KR^ENJE PRI
SU[ENJU IN @GANJU
Milun Krgovi}1, Nada Marstijepovi}1, Mileta Ivanovi}1, Radomir Zejak2,
Milo{ Kne`evi}2, Sne`ana \urkovi}1
1University of Montenegro, Faculty of Metallurgy and Technology, Cetinjski put bb, 81000 Podgorica, Montenegro
2University of Montenegro, Faculty of Civil Engineering, Cetinjski put bb, 81000 Podgorica, Montenegro
milun cg.ac.yu
Prejem rokopisa – received: 2007-01-19; sprejem za objavo – accepted for publication: 2007-03-18
In this paper an investigation of the influence of the mineral content of illite-kaolinite clays on the products’ shrinkage during
drying and firing is presented. Under the same conditions for preparing raw materials and ceramic mass, as well as under the
same firing regime, the products’ shrinkage during drying and firing is mostly influenced by the amounts of quartz and the illite
and kaolinite clay minerals.
Keywords: clay, linear shrinkage, total shrinkage, sintering, porosity
Predstavljena je raziskava vpliva vsebnosti glin ilinit-kaolinit na kr~enje pri su{enju in `ganju. Pri enakih pogojih priprave
surovega materiala in keramike in enakemu re`imu `ganja na kr~enje najbolj vpliva razmerje med kremenom ter glino
ilinit-kaolinit.
Klju~ne besede: glina, linearno kr~enje, skupno kr~enje, sintranje, poroznost
1 INTRODUCTION
The illite-kaolinite clays differ significantly in
mineral composition from illite, kaolinite, quartz,
feldspar, Fe2O3 and CaCO3 1,2. The presented raw
material composition, depending on the sintering
temperature, produces solid-state reactions, polymorphic
transformations of quartz and liquid-phase formation3.
Besides the sintering temperature of the ceramic mass,
the raw material mineral content also has an important
role for the relations between the microstructure
constituents4. The appearance of the liquid phase
accelerates the solid-state reactions5. The mineral content
of illite-kaolinite clays, besides other factors, determines
the formation of new crystal phases during the sintering
process, and the polymorphic transformations of quartz
caused by volume changes6.
During the drying of shaped ceramic products, the
simultaneous transfer of mass and heat in the homo-
geneous polydispersive system material with water
occurs7. The water is transported from the internal area
of the material through capillaries to the surface of the
products, where it evaporates. The process depends on
the diffusion and evaporation rates. The difference in the
water content on the surface and in the inside layers of
the material enables humidity to transfer continuously
from the inside to the surface by diffusion8. The volume
changes during the drying induce internal stresses.
2 EXPERIMENTAL
Two types of illite-kaolinite clays were used for the
preparation of samples (the clays are marked "PV" and
"BP"). The samples were formed by plastic shaping in a
mould of parallelepiped shape with dimensions 7.7 cm ×
3.9 cm × 1.6 cm and marked with the numbers 1, 2, 3, 4,
5 … 15.
The clays’ analyses consisted of a determination of
the mineral content with x-ray analysis, as well as che-
mical and granulometric analyses with a determination
of the particle size distribution. The grain shape was
determined microscopically. The linear and volume
shrinkage of the samples during the drying in air to a
constant mass and during drying in a dryer at 110 °C
were determined too.
The samples were fired at 800 °C, 900 °C, 1000 °C,
1100 °C and 1200 °C. The total porosity, the mineral
content by x-ray analysis, and the microscopic analyses
were assessed for the sintered samples.
3 RESULTS AND DISCUSSION
On the basis of the mineral and chemical compo-
sitions shown in Figure 1, Figure 2 and Table 1, it was
concluded that the investigated specimens were
illite-kaolinite clay types containing -quartz, carbonate
and Fe2O3. In the "BP" clay the content of -quartz was
Materiali in tehnologije / Materials and technology 41 (2007) 4, 189–192 189
UDC/UDK 553.61:663.3.041 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 41(4)189(2007)
lower and there was a larger content of illite and
kaolinite. The granulometric analysis shows a higher
average content of "PV" clay grain (Table 2), although
the pulverization of the clays was performed under the
same conditions. This is due to the difference in mineral
content, particularly the higher percentage of quartz in
the "PV" clay.
The higher content of -quartz in the "PV" clay and
the higher average content of grains explain the lower
average value of the volume shrinkage for the 15
samples during the air drying to a constant mass (Figure
3, Figure 4 and Table 3). The results of the volume
shrinkage, as well as the average value for 15 samples
during drying in a dryer to a constant mass, show higher
values of the volume shrinkage for the samples based on
"PV" clay (Figure 5, Figure 6 and Table 4). The higher
the content of -quartz, the lower the content of clay
minerals and the higher average value of the grains in
this clay enable the easier transport of water from the
internal area of the samples through the capillaries to the
surface at a drying temperature of 110 °C. The different
values of the volume shrinkage of the samples during
drying are the consequence of an unequal pressure
during plastic shaping in a mould.
The volume shrinkage during firing is smaller for the
samples based on the "PV" clay (Figure 7). The X-ray
M. M. KRGOVI] ET AL.: THE INFLUENCE OF ILLITE-KAOLINITE CLAYS’ MINERAL CONTENT ...
190 Materiali in tehnologije / Materials and technology 41 (2007) 4, 189–192
Table 1: Chemical composition of clay
Tabela 1: Kemi~na sestava gline
Oxides SiO2 Fe2O3 Al2O3 CaO MgO Na2O K2O SO3 Ig. loss
Clay "BP"
Percentage by weight, w/%
72.68 5.70 10.98 0.48 0.70 0,31 1.12 – 8.03
Clay "PV"
Percentage by weight, w/%
71 5.51 10.55 1.42 0.62 0.45 1.86 0.25 8.34
Table 2: Average grains value of the investigated clay types
Tabela 2: Povpre~na velikost zrn v preiskanih glinah
Clay type Average grains value (MV), d/µm
"BP" 17
"PV" 27
Table 3: Average values of volume shrinkage during drying in air to
the constant mass
Tabela 3: Povpre~no volumensko kr~enje pri su{enju na zraku do
konstantne mase
Clay type Average values of volume shrinkage (for15 samples) /%
"BP" 18.4
"PV" 15.8
Table 4: Average values of volume shrinkage during drying in dryer
to the constant mass
Tabela 4: Povpre~no volumensko kr~enje pri su{enju v su{ilniku do
konstantne mase
Clay type Average values of volume shrinkage (for15 samples) /%
"BP" 0.98
"PV" 1.65
Figure 3: Volume shrinkage during drying to a constant mass in air
("PV" clay)
Slika 3: Volumensko kr~enje pri su{enju do konstantne mase na zraku
("PV"-glina)
Figure 2: X-ray diffractogram of "BP" clay
Slika 2: Rentgenski difraktogram "BP"-gline
Figure 1: X-ray diffractogram of "PV" clay
Slika 1: Rentgenski difraktogram "PV"-gline
M. M. KRGOVI] ET AL.: THE INFLUENCE OF ILLITE-KAOLINITE CLAYS’ MINERAL CONTENT ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 189–192 191
Figure 9: X-ray diffractogram of sintered product ("PV" clay, T =
1200 °C)
Slika 9: Rentgenski difraktogram za sintrani produkt ("PV"-glina, T =
1200 °C)
Figure 6: Volume shrinkage during drying to a constant mass in dryer
("BP" clay)
Slika 6: Volumensko kr~enje pri su{enju do konstantne mase v
su{ilniku ("BP"-glina)
Figure 5: Volume shrinkage during drying to a constant mass in dryer
("PV" clay)
Slika 5: Volumensko kr~enje pri su{enju do konstantne mase v
su{ilniku ("PV"-glina)
Figure 4: Volume shrinkage during drying to a constant mass in air
("BP" clay)
Slika 4: Volumensko kr~enje pri su{enju do konstantne mase na zraku
("BP"-glina)
Figure 8: X-ray diffractogram of sintered product ("BP" clay, T =
1200 °C)
Slika 8: Rentgenski difraktogram za sintrani produkt ("BP"-glina, T =
1200 °C)
0
2
4
6
8
10
12
14
16
18
20
800 900 1000 1100 1200
F
BP
PV
iring temperature, T / ºC
T
o
ta
l
sh
ri
n
k
ag
e
%
,
Figure 7: Volume shrinkage during sintering
Slika 7: Volumensko kr~enje pri `ganju
analysis of the sintered products shows a higher content
of the minerals sillimanite, Al2O3·SiO2, and mullite,
3Al2O3·2SiO2, formed during the sintering of the samples
based on "BP" clay (Figure 8). The higher content of the
-quartz in the "PV" clay and the higher total porosity
explain the lower volume shrinkage during sintering,
when the quartz is partly transformed into tridimite
(Figure 9). The microstructures of the sintered products
are shown in Figures 10 and 11.
4 CONCLUSION
The investigation of the influence of illite-kaolinite
clays’ mineral content on the linear and volume
shrinkage during drying and firing shows that the
shrinkage depends on the following factors:
• Using the same conditions for the preparation of raw
materials and ceramic mass during drying in air and
in a dryer to a constant mass, the most important
factor is the influence of the relation between the
quartz and clay’s minerals content
• the values of the volume and linear shrinkage of
samples during the firing process depend also on the
relation between the quartz and the clay’s minerals
content. This influence is particularly strong at
temperatures above 1000 °C.
5 REFERENCES
1 M. Tecilazi}-Stevanovi}, Principles of Ceramic Technology, Faculty
of Technology and Metallurgy, University of Belgrade, Belgrade
(1990)
2 R. E. Grim, Clay Mineralogy, New York: McGraw-Hill Book Co.,
(1983)
3 M. M. Krgovi}, N. Z. Blagojevi}, @. K. Ja}imovi}, R. Zejak, Res. J.
Chem. Environ. 8 (2004) 4, 73–76
4 J. Griffiths, Ind. Min., 272 (1990), 35–40
5 M. M. Krgovi}, Z. K. Ja}imovi}, R. Zejak, Tile & Brick Int.
17 (2001) 3, 178–181
6 M. Krgovi}, M. Ivanovi}, N. Z. Blagojevi}, @. Ja}imovi}, R. Zejak,
M. Kne`evi}, Interceram, 55 (2006) 2, 104–106
7 Lj. Kosti}-Gvozdenovi}, R. Ninkovi}, Inorganic Technology,
Faculty of Technology and Metallurgy, University of Belgrade, Bel-
grade (1997)
8 B. @ivanovi}, R. Vasi}, O. Janji}, Ceramic Tiles, Institute of Mate-
rials in Serbia, Belgrade, (1985)
M. M. KRGOVI] ET AL.: THE INFLUENCE OF ILLITE-KAOLINITE CLAYS’ MINERAL CONTENT ...
192 Materiali in tehnologije / Materials and technology 41 (2007) 4, 189–192
Figure 11: Microstructure of sintered product ("BP" clay, T = 1200
°C)
Slika 11: Mikrostruktura sintranega produkta ("BP"-glina, T = 1200
°C)
Figure 10: Microstructure of sintered product ("PV" clay, T = 1200
°C)
Slika10: Mikrostruktura sintranega produkta ("PV"-glina, T = 1200
°C)
D. PIHURA, M. ORU^: THE APPLICATION OF SPHEROIDAL GRAPHITE CAST IRON ...
THE APPLICATION OF SPHEROIDAL GRAPHITE CAST
IRON IN BOSNIA AND HERZEGOVINA
UPORABA NODULARNE GRAFITNE LITINE V BOSNI IN HERCEGOVINI
Dervi{ Pihura, Mirsada Oru~
University of Zenica, Metallurgical institute "Kemal Kapetanovi}", Travni~ka cesta 7, 72000 Zenica, Bosnia and Herzegovina
pihura yahoo.com
Prejem rokopisa – received: 2006-07-20; sprejem za objavo – accepted for publication: 2007-02-16
Modern technology ensures the quality and the economy of machine parts manufactured from spheroidal graphite iron (SGI)
castings. Although this technology has been in widespread use in the EU countries for half a century and the production
volumes continue to increase, in Bosnia and Herzegovina (B&H) there is no interest in the production and use of SGI. From an
analysis of industrial facilities we have concluded that the foundry industry in B&H could also be competitive in the market for
SGI.
Key words: Spheroidal graphite iron castings, machine parts
Moderna tehnologija zagotavlja kakovost in ekonomi~nost strojnih delov, izdelanih iz nodularne grafitne litine. ^eprav je ta `e
pol stoletja veliko uporabljena v dr`avah EU in v ZDA, obseg proizvodnje pa stalno raste, v Bosni ni zanimanja za proizvodnjo
in uporabo nodularnih odlitkov. Iz analize industrijskih kapacitet sklepamo, da bi lahko bila livarska industrija v BiH
konkuren~na tudi na trgu nodularnih odlitkov.
Klju~ne besede: nodularna siva litina, strojni deli
1 INTRODUCTION
Spheroidal graphite iron (SGI) castings are not
produced and used in the metallurgical industry in
Bosnia and Herzegovina (B&H), which is the opposite
situation to most industrial countries, where the use of
SGI castings increases at rates of up to 10 % per year
(Figure 1). It is reasonable to expect that such industrial
branches as energy, transport, agriculture, etc. will be
forced into an increasing use of SGI, especially for
moving and rotating machine parts, for example:
• The increasing speeds used on railways requires the
substitution of some steel cast parts, for example,
gears, with gears made from SGI. These gears are
used to advantage when operating railways at speeds
from a hundred up to a few hundred kilometres per
hour.
• The automotive industry needs high-quality machine
cast parts. The B&H foundry industry needs to set up
for casting machine blocks from SGI, as well as other
cast parts.
• The use of wind energy is a solution that B&H will
soon exploit. When generating energy from wind
machines, a great number of the parts are made from
SGI.
2 INVESTIGATIONS AND DEVELOPMENT OF
SGI IN B&H
The development work for the production of parts
from SGI castings began approximately 30 years ago at
the "Kemal Kapetanovic" Metallurgical Institute in
Zenica. The technology for the manufacture of specific
SGI machine parts was developed at the institute, after
which it was transferred to the former foundry of the
Energoinvest company in Sarajevo, the foundry in Ilija{,
and other foundry companies.
The investigations were carried out to answer
different questions, relating to:
• the economics of the production of SGI in B&H;
• the influence of properties due to the presence of
residuals and impurities in the melting charge on the
process of nodulation and the final degree of
nodularity of the solidified SGI castings in relation to
the extended and the conventional chemical analysis;
Materiali in tehnologije / Materials and technology 41 (2007) 4, 193–195 193
UDC/UDK 669.13:669.1(497.15) ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 41(4)193(2007)
0
5000
10000
15000
20000
25000
1950 1960 1970 1980 1990 2000 2010
Years
P
ro
d
u
c
ti
o
n
(t
x
1
0
0
0
)
Figure 1: Growth in the world production of SGI from 1950 to 2000 1
Slika 1: Rast svetovne proizvodnje nodularnih odlitkov v obdobju
1950 do 2000 1
• the relation between the degree of nodularity and the
microstructure homogeneity and the mechanical
properties of SGI parts;
• the effect of the optimal addition of alloys and of the
processing parameters on the nodulation;
• the kinetics of nodulation.
Many of the findings and the accumulated field
experience could still be used for the renewal of the
production of SGI and various castings 2,3, especially
since customers may inquire about mutually exclusive
properties for some specific castings.
3 CHARACTERISTICS OF SGI CASTINGS
It is clear that the production of castings from SGI of
different quality levels could be revived in B&H in order
to replace the supply or imports from abroad. In this way
it is possible to produce castings from SGI for specific
uses and for the substitution of steel castings. The
melting temperature for SGI is about 300 °C lower than
that for steel, and this in itself is a significant advantage
in terms of saving energy and ecology. Apart from the
better castability, the easier mechanical machining and
the lower friction explain the continuous increase in the
production of machine parts from SGI in the countries of
the EU and the USA. It seems natural to expect that also
in B&H SGI should again achieve the position it had in
the foundry industry two decades ago, particularly since
a large part of the production was delivered to customers
abroad because of its high quality.
The increase in the world production of SGI (Figure
1) asks the question, which casting iron will dominate
the production of castings: spheroidal or lamellar
graphite iron, malleable cast iron or cast steel? Malleable
cast-iron production started to decrease in the early
1980s, while SGI production increases continuously,
partly because of the substitution of other cast irons and
partly because of new uses for the castings. The increase
in the production of SGI becomes clear from a
comparison of the casting technology and the cast’s
properties in Figure 2 and 3 4. The comparison accounts
for several factors, such as, the chemical analysis, the
mechanical properties, the section and size of the cast
parts, the processing time, the assortment, the energy
consumption, the total costs, the sampling, the testing
and the examination and control 2.
The effect of graphite nodules on the stress-strain
relations can be calculated from the equation
Rm = (1 – e
– Eo–t)/
where Eo is the modulus elasticity,  is a factor of the
dependence of Eo from the stress up to the value of Rp0.2
and can be expressed as  = 1/Rp0.2, and t is the
elongation.
The as-cast microstructures of SGI castings depend
on the chemistry and the cooling rate. The micro-
structure consists of ferrite, pearlite and free carbides,
and it can be modified by subsequent heat treatments.
When compared to steel castings, SGI castings are less
expensive, show a greater yield, and thus a greater
weight of the final cast part versus the weight of the used
melt. The consumption of energy for the production of
SGI parts is one-third lower than that for cast steel parts.
Also, the investment costs for the SG foundry are lower
than those for a steel foundry 2,4.
The basic advantages of SGI are accurate
dimensions, better uniformity of the strength properties,
less hot and cold cracking, and an easy heat treatment.
Additional advantages of the SGI castings are a lower
coefficient of thermal expansion, less shrinkage and
piping, better damping properties, better fluidity, better
machinability and reduced surface scaling. Also, with
SGI castings the effect of the wall thickness on the
mechanical properties is smaller (2).
4 FURTHER DEVELOPMENT
The increasing market demand for a better and more
uniform quality of castings and improved mechanical
properties as well as the pressure to lower prices require
that foundries develop and use improved technologies
and new grades of castings. They have to do this even
though the optimal combination of different properties
can be obtained with SGI (Figure 3).
Accordingly, the world production of SGI is
increasing rapidly; an increase of about 50 % was
achieved in the past decade, a growth rate that is only
rarely met with other major industrial products.
However, methods for the production of parts competing
for the same applications as SGI castings are also
constantly improving and the SGI foundry industry is
forced to compete, not only with other cast-iron alloys,
but also other materials that are potential substitutes. In
this competition SGI castings seem to be in better
position than other iron castings because of the more
D. PIHURA, M. ORU^: THE APPLICATION OF SPHEROIDAL GRAPHITE CAST IRON ...
194 Materiali in tehnologije / Materials and technology 41 (2007) 4, 193–195
Figure 2: Mechanical properties of cast irons, cast steels and rolled
steels for the period 1950–1960 (left-hand side) and 2000 (right-hand
side) 2,5; (LG – gray iron, MG – malleable iron, SG – spheroidal
graphite iron, RS – rolled steel, CS – cast steel), (Rm – Tensile
strength, Rp0.2 – Yield strength, A5 – elongation)
Slika 2: Mehanske lastnosti `elezovih litin, litega `eleza in valjanega
jekla za obdobje 1950–1960 (levo) in 2000 (desno) 2,5; (LG – siva
litina, MG – temprana litina, SG – nodularna litina, RS – valjano
jeklo, CS – jeklena litina), (Rm – raztr`na trdnost, Rp0.2 – meja
plasti~nosti, A5 – raztezek)
ecologically friendly production process and its lower
costs.
With the introduction of new technological processes
such as ADI and ESR better mechanical properties of
SGI (Figure 2 and 3) are being achieved and the com-
petitiveness of castings, especially for use in more severe
conditions, is strengthened.
5 CONCLUSION
The overall competitiveness involving the economy
of production and the properties make SGI castings very
suited to a number of applications in machine parts. The
increased production and use of SGI castings is justified
if advanced technology and organisation are achieved in
the production and delivery of castings with qualities
that satisfy export-market requirements.
6 REFERENCES
1 www.ductile.org/didata/Section 2/figures
2 R. Hummer et al.: Giesserei, 88 (2001) 9–11, 49–55
3 World Steel production, Worldsteel News, January (2005), 5
4 F. Kritschtner: Livarski vestnik, (1996), 4, 1–15
5 Gusseisen mit Kugelgraphite: Giesserei Kalender, 1985, 72–81
6 P.M. Cabanne, Hommes & Fonderie, 306 (2000), Aout/Septembre,
18–22
D. PIHURA, M. ORU^: THE APPLICATION OF SPHEROIDAL GRAPHITE CAST IRON ...
Materiali in tehnologije / Materials and technology 41 (2007) 4, 193–195 195
Properties Cast Iron
Gray Malleable White Steel Nodular
Fluidity
Machining NA
Damping
Surface hardening NA
Modulus off elasticity NA
Impact energy NA
Corrosion resistant
Strength/Mass NA
Abrasion
Costs
The best The worst
Figure 3: Properties of cast ferrous materials 5,6
Slika 3: Lastnosti litih `eleznih materialov 5,6