17 / 40
March 2016
AFRICAN FUSION
15
Figure 4: Vickers hardness tests close to surface.
Figure 5: Vickers hardness values close to the root.
Welding parameters
In order to establish appropriate parameters for the experi-
ments and simulation, a series of tests were done. This section
introduces some of these variations. Figure 3 a-g is a compila-
tion of metallographic sections fromsamples welded at differ-
ent speeds and currents, while Table 2 gives a list of all these
different welding current, welding speed and processes used
to produce the samples.
Steel grade Outer diameter [mm] Wall thickness [mm]
S355-TT
73.5
4
S355-TT
101
6
20MnV6-TT
168.3
8
20MnV6-TT
168.3
16
S890QL1
168.3
8
S890QL1
168.3
16
Table 1: A list of materials and dimensions used.
Themetallographic sections are necessary to characterise
theweldpenetrationandweld formation, (Figure 3a-g). At first,
all micro-sections look similar, but looking into the detail a
more homogeneousweld formation andweld penetration can
be seen when using pulse GMAW (Figure 3e and 3f).
However, the advantage of pulse welding cannot be
modelled in common FE welding simulation software and is
therefore only an option for welding experiments.
In addition to weld penetration and weld formation, it is
important to consider hardening. These investigations were
done according to DIN EN ISO 9015-1:2011-05 [4]. Figure 4 and
Figure 5 show the results of standardised testing, which is
divided into a measurement close to the surface and another
close to the root.
Sample
number
Welding
current [A]
Welding speed
[cm/min]
Process
a
320
70
standard
b
320
90
standard
c
320
100
standard
d
320
110
standard
e
320
70
pulse 3 [Hz]
f
320
70
pulse 10 [Hz]
g
270
50
standard
Table 2: A list of parameter variations for the metallographic
sections shown below.
Figure 3 a-g: Metallographic
sections of joints welded at different
welding speeds and welding
currents, as per Table 2.
Considering the area close to the surface it is noticeable that all
standard welds are very close to each other and have a similar curve.
The pulsed welding is at the same level of hardness as standard base
metal S355-TT. In contrast to standardwelding, thepulsedweldingdoes
not tend to increase the hardness in the heat-affected zone. Vickers
hardness values close to the root (see Figure 5), do not deviate from
this trend. The pulsed welding is again at a lower level of hardness
compared with standard welding at a welding speed of 70 cm/min.
Welding simulation
To save money and time during the development of a K- joint truss