R. CELIN et al.: A COMPARISON OF AS-WELDED AND SIMULATED HAZ MICROSTRUCTURES
455–460
A COMPARISON OF AS-WELDED AND SIMULATED HEAT
AFFECTED ZONE (HAZ) MICROSTRUCTURES
PRIMERJAVA MIKROSTRUKTURE TOPLOTNO VPLIVANEGA
PODRO^JA VARJENEGA IN SIMULIRANIH VZORCEV
Roman Celin1, Jaka Burja1, Gorazd Kosec2
1Institute of Metals and Technology, Lepi pot 11, 1000 Ljubljana, Slovenia
2Acroni d.o.o., Cesta Borisa Kidri~a 44, 4270 Jesenice, Slovenia
roman.celin@imt.si
Prejem rokopisa – received: 2016-01-05; sprejem za objavo – accepted for publication: 2016-01-29
doi:10.17222/mit.2016.006
The high-strength steel grade S690QL and a filler welding wire Mn3Ni1CrMo were the materials chosen for welding a
V-shaped butt weld. In order to prevent the weld’s cold cracking, a multi-pass welding technique was applied. A metallographic
investigation revealed microstructure variations in different areas of the weld’s heat-affected zone. A reverse-engineering
approach was used to test a dilatometer’s capabilities to simulate different HAZ microstructures. Hollow steel-cylinder
specimens were subjected to several weld thermal cycles in order to generate similar microstructures as in the real weld’s HAZ.
The microstructures of the as-welded and simulated heat-affected zone specimens were investigated. Good agreement was found
between the dilatometer-simulated HAZ microstructures and those in a real HAZ weld.
Keywords: welded joint, microstructure, high-strength low-alloy steel, simulation, dilatometer, heat-affected zone
Visokotrdno jeklo S690QL in varilna `ica Mn3Ni1CrMo kot dodatni material, sta bila uporabljena pri varjenju so~elnega
V-spoja. Za prepre~itev pokanja v hladnem je bila uporabljena tehnika ve~varkovnega varjenja. Metalografska preiskava je
odkrila razli~ne mikrostrukture v razli~nih predelih toplotno vplivanega podro~ja zvara. Za preizkus delovanja dilatometra je bil,
za simulacijo mikrostruktur razli~nih podro~ij toplotno vplivanega podro~ja, uporabljen princip povratnega in`enirstva.
Razli~na mikrostruktura toplotno vplivanega podro~ja je bila simulirana z izpostavitvijo votlih cilindri~nih vzorcev razli~nim
toplotnim ciklom. Opravljena je bila metalografska preiskava realnega zvarjenega spoja in simuliranih vzorcev. Primerjava
rezultatov je pokazala dobro ujemanje simuliranih in realnih mikrostruktur.
Klju~ne besede: zavarjen spoj, mikrostruktura, visokotrdno jeklo, simulacija, dilatometer, toplotno vplivano podro~je
1 INTRODUCTION
The decision to use high-strength steel depends on a
number of application requirements, such as thickness
reduction, corrosion resistance, formability and weld-
ability. The quenched and tempered low-alloy (QTLA)
steels, usually containing less than 0.25 % carbon and
less than 5 % alloying elements, are strengthened prima-
rily by quenching and tempering to produce micro-
structures containing martensite and bainite.1 S690QL is
such a steel grade with high strength and toughness.2
Any common welding procedure can be used to join
QTLA steels.3 For any given steel, the welded joint’s
microstructures and the mechanical properties in the
weld metal (WM) and the heat-affected zone (HAZ) are
influenced mainly by the welding thermal cycle.
Due to the welding thermal cycle and the associated
peak temperature, a change of the parent metal’s micro-
structure and the mechanical properties happens in the
HAZ. With increasing distance from the fusion line the
peak temperature decreases, thus forming different
microstructures. The width of the HAZ depends on the
welding procedure, the thermal conditions and the
physical properties of the parent metal.
Figure 1 shows a simplified presentation of different
HAZ regions of a multi-pass welded joint. Weld pass 2
has a portion of HAZ that can be treated as a single weld
pass and can be divided into four regions. These regions
are defined by the peak temperature to which the region
was exposed during the weld thermal cycle:4,5
• Coarse-grain region (CG HAZ) – material adjacent to
the fusion line that reaches a peak temperature TP
above 1300 °C,
• Fine-grain region (FG HAZ) – TP is lower, but still
above AC3,
• Inter-critical region (IC HAZ) – TP is between AC1 <
TP < AC3,
• Sub-critical region (SC HAZ) – TP is lower than AC1.
However, the welding of thick steel components
usually requires the application of a multi-pass welding
procedure. In this case, the first pass (weld pass 1) HAZ
regions are reheated to different peak temperatures
during the second weld pass thermal cycle (weld pass 2).
Figure 1 also shows a simplified presentation of the
different reheated first weld pass HAZ regions:4,5
• region 1: Super-critically reheated coarse-grain HAZ
– SCR CG HAZ,
Materiali in tehnologije / Materials and technology 50 (2016) 3, 455–460 455
UDK 67:017:669.14.018.298:621.791.05 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(3)455(2016)
• region 2: Inter-critically reheated coarse-grain HAZ –
IR CG HAZ,
• region 3: Sub-critically reheated coarse-grain HAZ –
SC CG HAZ,
• region 4: Super-critically reheated fine-grain HAZ –
SCR FG HAZ,
• region 5: Inter-critically reheated fine-grain HAZ –
IR FG HAZ,
• region 6: Sub-critically reheated fine-grain HAZ –
SC FG HAZ.
In a real weld the HAZ regions are narrow in relation
to the weld. To achieve regions of uniform micro-
structure suitable for investigations, weld simulators are
used.6-12 Usually, these weld thermal simulations are
made in conjunction with weldability investigations to
determine the proper welding parameters.
In the presented investigation a dilatometer with a
controlled heating and cooling fixture was used to simu-
late the weld thermal cycles. The goal of the investi-
gation was to test the dilatometer’s capability to simulate
the microstructures that correspond to the real weld’s
HAZ microstructures using a reverse-engineering
approach.
2 EXPERIMENTAL PART
The chemical analysis of the Micral 690 sample with
an ICP spectrometer was made prior to any further inve-
stigations. The determination of the welding parameters
for the Micral 690 welding procedure was carried out by
considering the EN 1011-2 recommendations, the steel
manufacturer’s specifications and the data from several
papers.13–16
The two plates, with a thickness of 15 mm, a length
of 500 mm, a width of 150 mm and a V-groove joint
geometry, were welded with 8 passes using the sequence
shown in Figure 2. A manual MAG welding procedure
in a flat (PA) position was used. The filler material was
grade Mn3Ni1CrMo welding wire with a diameter of
 1.2 mm, according to SIST EN ISO 1683417. The
shielding gas used was a mixture M21 (82 % Ar + 18 %
CO2). The welding was carried by a skilled, certified
welder and without preheating.
The welding current, voltage and time were recorded
during the procedure. The weld pass 5 thermal cycle was
recorded by dipping a type-S thermocouple (Pt Rh–Pt)
directly into the weld’s molten bead. The approximate
position of the thermocouple is marked with a dot.
Standard18 non-destructive examinations and mechanical
tests of the weld joint were also carried out and are
described in a previous paper.19
Seven hollow, 10-mm-long cylinders with a 4-mm
outer diameter and 1.5-mm inner diameter were ma-
chined from the weld’s parent material (base metal)
plate. Holes were bored in order to obtain a uniform
heating and cooling of the sample during the thermal
simulation.
The simulations were carried out with a TA instru-
ments DIL805A/D quenching dilatometer in a vacuum
atmosphere, with argon used as a coolant. Control of the
thermal cycle was maintained via a type-S thermocouple
that was spot welded directly onto the hollow cylinder
sample.
R. CELIN et al.: A COMPARISON OF AS-WELDED AND SIMULATED HAZ MICROSTRUCTURES
456 Materiali in tehnologije / Materials and technology 50 (2016) 3, 455–460
Figure 2: Sketch of the welded joint
Slika 2: Skica zavarjenega spojaFigure 1: Simplified representation of the different regions of the
HAZ
Slika 1: Poenostavljena predstavitev razli~nih obmo~ij toplotno vpli-
vanega podro~ja
Figure 3: Schematic diagrams for the thermal cycle simulations
Slika 3: Diagrami za potek simulacij termi~nih ciklov
Figure 3 shows schematic diagrams with the parame-
ters for seven thermal cycle simulations used to generate
the corresponding microstructure:
• single weld pass CG, FG, IC and SC regions of the
HAZ,
• reheated weld pass SCR CG, IC CG, SC CG regions
of the HAZ, as in Figure 1.
For the reheated simulated thermal cycles the same
parameters were used as for the single-cycle simulation.
The time–temperature dependence (Figure 4) recorded
during weld pass 5 was used to determine the simulation
heating rate, the peak temperatures, the holding times
and the cooling rates.
During a single coarse-grain (CG) thermal cycle
simulation the specimens’ phase-transition temperatures
AC1 and AC3 were recorded as well as the dilatation and
temperature data from each simulation. The specimen
was prepared by grinding and polishing with diamond
paste, followed by a chemical etching with 5 % Nital.
Macro- and microscopic examinations20–22 were per-
formed using a light microscope (LM) to characterize the
microstructures of the deposited weld metal, the heat-
affected zone, the parent metal and the dilatometer-simu-
lated hollow cylinder specimens of the parent metal.
A comparison between the as-welded and the dila-
tometer-simulated microstructures was carried out.
3 RESULTS AND DISCUSSION
The results of the parent metal’s quantitative chemi-
cal analysis are presented in Table 1. The steel contains
strong carbide-forming elements such as Nb, Mo, Cr and
Ti, which ensure high strengths, even at elevated tempe-
ratures.
Table 1: Chemical composition of parent metal in mass fractions, w/%
Tabela 1: Kemi~na sestava jekla v masnih dele`ih, w/%
C Si Mn P S Cr Ni
0.164 0.29 0.72 0.006 0.001 0.54 0.20
Cu Mo Ti Nb Al B N
0.25 0.263 0.02 0.028 0.028 0.001 0.011
The applied welding parameters and the calculated
welding speed and heat input were:
• weld length, 500 mm
• welding time, 132–422 s
• voltage, 21–22 V
• current, 100–165 A
• weld inter-pass temperature, 130 °C
• welding speed, 1.8–2.89 mm/s
• heat input, 1.03–1.63 kJ/mm
Lower-value welding parameters were applied for the
root pass welding.
After the temperature–time data acquisition from the
temperature-measurement instrument’s memory card and
data analysis for the welding pass 5, a cooling time t8/5 of
7.6 s was determined and the peak temperature TP was
1370 °C (Figure 4).
The shape of the cooling curve (Figure 4) is affected
by the inter-pass temperature. Also, the weld’s cooling
time is prolonged due to the reduced temperature diffe-
rence between the weld material and the surrounding
parent metal.
Figure 5 shows the ICR CG HAZ recorded thermal
cycle simulation with the temperature and the dilatation
curve. From the dilatation data analysis of the phase
transformation during the CG thermal cycle simulation
the transition temperatures AC1 and AC3 were found to be
830 °C and 885 °C, respectively. During the rapid heat-
ing of steel, as in the case of welding, the phase-trans-
formation temperatures are increased, the observed
transformation temperatures AC1 and AC3 are well above
equilibrium, i.e., about 710 °C and 785 °C respectively.
The increase of the transformation temperatures is
attributed to the diffusion process of the transformation
from ferrite to austenite, which is time and temperature
dependent. The dilatation curve in Figure 5 clearly
indicates that a transformation took place below 500 °C,
R. CELIN et al.: A COMPARISON OF AS-WELDED AND SIMULATED HAZ MICROSTRUCTURES
Materiali in tehnologije / Materials and technology 50 (2016) 3, 455–460 457
Figure 5: ICR CG HAZ thermal simulation with temperature-time and
dilatation-time dependence
Slika 5: Potek temperature in raztezka v odvisnosti od ~asa pri simu-
laciji interktiti~nega grobozrnatega TVP
Figure 4: Weld pass 5 – recorded temperature – time dependence
Slika 4: Polnilni varek 5 – zabele`ena odvisnost temperatura – ~as
during the first cooling, and that another transformation
took place during the second cooling, but it was not as
intense as the first one, thus proving that the inter-critical
temperature was indeed reached.
A good fit was found between the theoretical (Figure
3) and dilatometer-simulated thermal IR CG HAZ cycle
(Figure 5). The heating rates, cooling rates, and holding
times were very close to those programmed, with minor
deviations due to the response of the dilatometer control
and the regulation system. The first thermal cycle peak
temperature was 1360 °C, with an inter-critical peak
temperature of 866 °C.
Figure 6 shows the welded joint’s macro section with
each weld pass and the corresponding HAZ. Between
subsequent weld passes an unaffected region of the
coarse-grain HAZ microstructure can be distinguished.
The reheated pockets of the CG HAZ regions are small
and discontinuous, which makes their microstructure
difficult to identify and investigate. During metallo-
graphic investigations the dimensions of the IR CG HAZ
microstructure region were estimated to be 0.8 mm long
and 0.3 mm wide.
Figure 7 shows the weld pass 5 HAZ area not
affected by a subsequent weld pass thermal cycle. It is an
area with a longitudinal microstructure transition from
the weld metal (WM) through the heat-affected zone
(HAZ) microstructures to the parent metal (PM).
The weld metal (WM) consists of columnar dendrites
with a bainite microstructure. Some individual marten-
sitic grains are also present in the middle of the deposi-
ted weld metal. The microstructure of the HAZ consists
of martensite and bainite with a transition to the
unaffected tempered martensitic microstructure of the
parent metal (PM) (Figures 8 and 9).
Figure 8.1 shows the coarse grains (CGs) that are
present in the HAZ adjacent to the fusion boundary of
the weld. The fine-grain (FG, Figure 8.3) region follows
due to the peak temperature above AC3 but lower than in
the CG region, the temperature and time were not
sufficient to cause severe grain growth. With increasing
distance from the fusion boundary there are inter-critical
(IC, Figure 8.5) and subcritical (SC, Figure 8.7) regions
of the HAZ.
R. CELIN et al.: A COMPARISON OF AS-WELDED AND SIMULATED HAZ MICROSTRUCTURES
458 Materiali in tehnologije / Materials and technology 50 (2016) 3, 455–460
Figure 8: Real and simulated HAZ microstructures
Slika 8: Realne in simulirane mikrostrukture TVP
Figure 6: Welded joint macro section
Slika 6: Makroposnetek zavarjenega spoja
Figure 7: Transition from weld metal (WM) to parent metal (PM)
Slika 7: Prehod iz vara (WM) v osnovni material (PM)
The reheated inter-critical region (Figure 9.11 and
9.12) is very susceptible to failure, due to the fact that
the phase transformation into austenite began on the
grain boundaries; these small areas were then quickly
cooled and are in turn hard and brittle. The subcritical
regions are mainly tempered bainite and martensite with
precipitated carbides and therefore represent no danger
to the structural integrity of the weld. The simulated
microstructures in Figure 8 and Figure 9 are labelled
with even numbers. Although the grain size of the simu-
lated specimen is slightly larger than that of the real
welded joint when comparing the identical thermal
cycle, we can distinguish a similarity between the
welded HAZ and the dilatometer-simulated microstruc-
tures. The reason for the larger grains is that the thermal
pinning is not considered in the thermal cycle simulation
process.6
4 CONCLUSIONS
Normally, a dilatometer is used to observe a speci-
men’s dimensional changes under a controlled heating or
cooling rate. It can also be used to construct a conti-
nuous-cooling- transformation (CCT) diagram or an
isothermal time-temperature-transformation (TTT) dia-
gram. The TA DIL805A/D dilatometer with controlled
heating and cooling fixtures was tested to simulate a real
weld’s HAZ microstructure. Based on the investigation
the following can be concluded:
• The use of a simulated HAZ microstructure is a con-
venient way to study the weldability of a given steel.
• The presented investigation was limited to micro-
structure, due to the specimen’s size and geometry.
• The simulation of a weld’s HAZ microstructure is
possible within the limits of the dilatometer’s
capabilities and the specimen’s size.
• Hollow cylinder samples had a better response to the
heating and cooling rate change than a solid cylinder.
• The inter-critical temperature during welding of
S690QL is between 830 °C and 885 °C.
Acknowledgment
The authors are thankful to the Acroni d.o.o. and
Elektrode Jesenice for the technical support related to
work presented in this paper.
5 REFERENCES
1 F. J. Winsor, ASM Handbook, Vol. 6: Welding, Brazing and Solder-
ing, ASM, International, Materials Park, OH 1993, 662
2 SIST EN 10025-6:2005+A1:2009, Hot rolled products of structural
steels – Part 6: Technical delivery conditions for flat products of high
yield strength structural steels in the quenched and tempered condi-
tion
3 S. Kou, Welding metallurgy, 2nd ed., Wiley Interscience, Hooboken
2003, 406
4 M. Hamada, Control of strength and toughness at the heat affected
zone, Welding International, 17 (2003) 4, 265-270
5 T. Lolla, S. S. Babu, S. Lalam, M. Manohar: Understanding the Role
of Initial Microstructure on Intercritically Reheated Heat- Affected
Zone Microstructures and Properties of Microalloyed Steels, Proc. of
the 9th Inter. Conf., June 4–8, 2012 Chicago, 34-42
6 Y. Shi, Z. Han, Effect of weld thermal cycle on microstructure and
fracture toughness of simulated heat-affected zone for a 800 MPa
grade high strenght low alloy steel, Journal of Materials Processing
Technology, 207 (2008), 30–39, doi: 10.1016/j.jmatprotec.2007.
12.049
7 G. R. Goodall, J. Gianetto, J. Bowker, M. Brochu, Thermal simu-
lation of HAZ regions in modern high strength steel, Canadian
Metallurgical Quarterly, 51 (2012), 1, 58–66, doi:10.1179/
1879139511Y.0000000023
8 I. Samard`i}, A. Stoi}, D. Kozak, I. Kladaric, M. Dun|er, Applica-
tion of Weld Thermal Cycle Simulator in Manufacturing Engineer-
ing, Journal of Manufacturing and Industrial Engineering, Journal of
Manufacturing and Industrial Engineering, 12 (2013), 1–2, 7–11,
doi: http://dx.doi.org/10.12776/mie.v12i1-2.177
9 9G. L. Liang, S.W. Yang, H. B. Wu, X. L. Liu, Microstructure and
mechanical performances of CGHAZ for oil tank steel during high
heat input welding, Rare Metals 32 (2013) 2, 129–133, doi:10.1007/
s12598-013-0036-y
10 Y. Adonyi, Heat-Affected Zone Characterization by Physical
Simulations, Welding Journal, 10 (2006), 42-47
11 J. Wang, S. Lu, L. Rong, D. Li, Thermal cycling, microstructure and
mechanical properties of 9Cr2WVTa steel welds, Journal of Mate-
rials Processing Technology, 222 (2015) 8, 434–443, doi:10.1016/
j.jmatprotec.2015.03.017
12 M. Dun|er, I. Samard`i}, T. Vuherer, Dependence of hardness and
impact energy on cooling time t8/5 and temperature for S960QL,
Metalurgija, 54 (2015) 3, 539–542, doi:10.12776/mie.v12i1-2.177
13 Welding – Recommendations for welding of metallic materials – EN
1011-2 Part 2: Arc welding of ferritic steels
R. CELIN et al.: A COMPARISON OF AS-WELDED AND SIMULATED HAZ MICROSTRUCTURES
Materiali in tehnologije / Materials and technology 50 (2016) 3, 455–460 459
Figure 9: Real and simulated HAZ microstructures; reheated regions
Slika 9: Realne in simulirane mikrostrukture TVP; ponovno pregreta
podro~ja
14 I. Hajro, O. Pa{i}, Z. Burzi}, Karakterizacija zavarenih spojeva na
visoko~vrstom konstrukcionom ~eliku S690QL, Zavarivanje i zava-
rene konstrukcije, 58 (2010) 4, 123–129
15 M. Shome, Effect of heat-input on austenite grain size in the heat
affected zone of HSLA-100 steel, Materials Science and Engineering
A, 39 (2007) 13, 454–460, doi:10.1016/j.msea.2006.09.085
16 Acroni Micral 690 Datasheet, https://www.metalandsteel.com/docu-
mentserver/store/17606/2b66fa1b-adaa-4b42-9700- 1ab07d115e60.pdf
17 SIST EN ISO 16834: – Welding consumables – Wire electrodes,
wires, rods and deposits for gas shielded metal arc welding of high
strength steels – Classification
18 SIST EN ISO 15614-1: Specification and Qualification of Welding
Procedures for Metallic Materials – Welding Procedure Test – Part 1:
Arc and Gas Welding of Steels and Arc Welding of Nickel and
Nickel Alloys
19 R. Celin, J. Berneti~, D. A. Skobir Balanti~, Welding of the steel
gradeS890QL, Mater. Tehnol., 48 (2014) 6, 931–935
20 S. Ko`uh, M. Goji}, I. Ivani}, B. Kosec, Microstructure of welded
austenitic stainless steel after annealing at 900 °C, Zavarivanje i
zavarene konstrukcije, 61 (2013) 4, 149–156
21 F. Tehovnik, B. Arzen{ek, B. Arh, D. Skobir, B. Pirnar, B. @u`ek,
Microstructure evolution in SAF 2507 super duplex stainless steel,
Mater. Tehnol., 45 (2011) 4, 339–345
22 M. Pournavari, S. P. H. Marashi, H. L. Jaber, DP780 dual-phase-steel
spot welds: critical fusion-zone size ensuring the pull-out failure
mode, Mater. Tehnol., 49 (2015) 4, 579–585, doi:10.17222/
mit.2014.184
R. CELIN et al.: A COMPARISON OF AS-WELDED AND SIMULATED HAZ MICROSTRUCTURES
460 Materiali in tehnologije / Materials and technology 50 (2016) 3, 455–460