![Reorganization of creative activity [15]](https://www.researchgate.net/profile/Antti-Salminen-2/publication/228442593/figure/fig1/AS:650777290895361@1532168805360/Reorganization-of-creative-activity-15_Q320.jpg)
Content uploaded by Antti Salminen
Author content
All content in this area was uploaded by Antti Salminen on Dec 13, 2013
Content may be subject to copyright.
67
ISSN 1392 - 1207. MECHANIKA. 2010. Nr.4(84)
The filler wire - laser beam interaction during laser welding with low
alloyed steel filler wire
A. Salminen
Lappeenranta University of Technology, Skinnarilankatu 34, 53850 Lappeenranta, Finland,
E-mail: antti.salminen@lut.fi
Machine Technology Centre Turku Ltd, Lemminkäisenkatu 28, 20520 Turku, Finland
1. Introduction
1.1. Laser welding
Laser welding is usually performed in the key-
hole-welding mode. The high power density required for
keyhole welding of steels is achieved by focusing the laser
beam power (>1 kW) to a small focus spot of diameter 0.1
– 0.5 mm. The heat absorbed from the laser beam to the
workpiece heats the material, melting it and finally vapor-
izing it. The use of laser welding has so far been popular in
applications were high weld quality or production through-
put is required. The industrial utilization of laser welding
has suffered from the stringent joint requirements. The use
of filler material with laser welding makes the joint toler-
ances less severe. Owing to the restricted amount of infor-
mation about the welding process with filler wire, the laser
welding with filler wire is use seldom.
The major drawback of the laser welding process
is the stringent joint requirements. In the case of a typical
butt joint the widest acceptable air gap for autogenous laser
welding is usually 10% of the material thickness. This re-
quires usually the joint edges to be machined or laser cut.
The tolerances can be fulfilled when components are rela-
tively small and manufactured with machining or laser
cutting. Even in these cases for large components it is dif-
ficult to achieve the required accuracy when positioning
them. The geometry of a joint, air gap and part mismatch
varies from joint to joint and between products and produc-
tion batches. Filling of the air gap requires additional mate-
rial that should exist in the joint area during welding. The
filler material is usually fed as a wire during welding.
Laser welding with filler wire is usually consid-
ered to be too a difficult process for industrial application,
having too many parameters and too stringent requirements
for wire positioning. According to literature the studies
have usually concentrated on the effect of filler material on
the metallurgy and properties of the weld metal to solve the
problems of one particular material and/or application.
1.2. Process efficiency
Energy input is different in laser welding com-
pared with conventional welding. The high intensity of the
laser beam creates a deep narrow weld. When the laser
energy is absorbed to the material through the wall of the
keyhole, the heat is distributed evenly throughout the
whole penetration depth. The cooling time for the critical
temperature range of 800 - 500°C, t8/5, is considerably
shorter for laser welding than for arc welding, which arises
from low energy input caused by the nature of the keyhole
welding mode and small focal point. The small concen-
trated energy input of laser welding creates a deep, narrow
weld and heat affected zone (HAZ). This energy input
mode results in a smaller grain size in the weld metal and
HAZ and minimal distortions. Due to a high cooling rate
the hardness of HAZ in C-Mn steels becomes a higher than
in the arc welding.
Fig. 1 The distribution of laser power during the welding
process, Pa is conducted loss from the melt zone,
Pb is power used in boiling weld metal and ionizing
the vapor, Pc is power prevented from interacting
with the workpiece by absorption or reflection from
the plasma cloud, Pe convective thermal loss from
the weld zone, Pf is irradiative loss from the weld
zone, and Pg is reflecting loss from the weld zone
[1]
The total efficiency of the welding method de-
pends on the efficiency of the welding machine and proc-
ess. The efficiency of a CO2 laser is typically between
10 - 15%. The efficiency of the laser welding process is
difficult to measure because of the high thermal gradient. It
can be divided into two different factors: absorptivity and
melting efficiency. Absorptivity depends on the material,
laser light wavelength and welding system. The proportion
of the energy that is not absorbed by the material is lost by
any of the following mechanisms [1]:
• conductive and convective losses from the melt
zone;
• laser power used in boiling the material and ioniz-
ing metal vapor;
• laser power absorbed by metal vapor;
• laser power reflected from metal vapor;
• laser power reflected from base material surface;
• laser power reflected from the keyhole;
68
• heat radiation losses from the weld zone;
• power passing through the keyhole.
Fig. 1 illustrates these different losses. They de-
pend on material properties, laser beam properties and in-
tensity. The power passing through can be minimized by
optimization of parameters. Power losses through the key-
hole increase with an increase in air gap in a butt joint [1].
Fig. 2 illustrates the fraction of power losses by
reflection from the keyhole and base material depending
on the welding speed [2].
Fig. 2 The change of keyhole and melt pool shape and la-
ser beam reflection with increasing welding speed
[2]
1.3. Review of status of laser welding with filler wire
The problems caused to the welding process by a
feed position error, transverse to the welding direction (WY)
are obvious. In case of 2 mm wire diameter, the position
error of WY = 0.25 mm will decrease the melting efficiency
by 30% and 36% for wire diameters of 1.0, respectively
[3]. This positioning error, WY, causes also asymmetrical
weld bead [4].
Fig. 3 describes the geometrical parameters of the
wire feeding and laser beam in comparison butt joint. The
wire is usually fed to the leading edge of the keyhole (lead-
ing feed, LF), which offers more evenly distributed filler
Fig. 3 The parameters of filler wire feed relative to the
laser beam, workpiece surface and joint during laser
welding
material in the weld metal [4]. However the dilution is with
both leading (LF) and trailing feed (TF) when using high
welding speeds (base material: mild steel, filler wire: aus-
tenitic stainless steel) [5]. Also the requirements set by
optical seam tracking may justify the use of trailing feed
[6].
The optimum feed angle (
α
W) is within a wide
range, 30°-75°, compared to the workpiece surface. The
geometrical dimensions of the welding set up often define
this range.
1.4. The phenomena affecting to use of filler wire
There are some aspects in the laser welding and
laser - material interaction that can be considered to affect
on the process performed with filler wire.
Angle of incidence of the laser beam is an impor-
tant beam parameter affecting the absorptivity. The absorp-
tion reaches its maximum when the angle of incidence is
about 90°, but decreases with the angle of incidence, see
Fig. 4, a. Absorption typically rises also with temperature
like shown in Fig. 4, b. Iron alloys start to vaporize when
the laser beam intensity exceeds 106 W/cm2 and a keyhole
starts to form [7]. A keyhole with molten edges is formed
by the equilibrium of forces caused by the laser beam,
metal vapor, material surface tension and gravity. When
the intensity reaches 106 W/cm2 the keyhole is formed and
reflectivity decreases to a value close to zero [8].
a b
Fig. 4 The effect of the incidence angle and temperature on
reflectivity: a - dependency of the reflectivity of the
surface on the angle of incidence of laser beam and
plane of polarisation; b - reflectivity as a function of
surface temperature for steel at various wavelengths
[9]
When considering the laser adding a filler wire
feeding to the laser welding process, the filler wire is fed in
nonoptimum way to the process. The feeding must usually
be performed in such a manner that absorption is not in its
maximum in respect of interaction angle and power inten-
sity. The wire and the laser beam interaction occur usually
outside of focal point and where the power intensity is
lower. Also the wire fed is in room temperature and is fed
so fast that there is no time for wire heating prior interac-
tion with the laser beam.
1.5. Laser welding – filler wire interactions
It has been shown that the filler wire may reflect
the laser beam during wire feed in open air. The published
results so far have been from single source. These experi-
ments had been carried out with higher laser power, but
explain quite nicely how the beam is reflected from the
wire.
During welding the filler wire is fed into the melt
pool or into a focal cylinder. When the laser beam meets
the filler wire, several phenomena occur. The phenomena
involved with the process itself are not fully clarified yet.
In most cases research has concentrated on the properties
of the weld rather than the process optimization. The filler
69
wire should be fed towards the focal point of the laser
beam [10]. Feeding takes place in the direction of the joint,
at the trailing or leading edge of the keyhole. When the
laser beam impinges on the filler wire, it absorbs part of
the power, part of it passes through the wire and part of it
is reflected away [3]. The fraction of the reflected laser
power depends on the beam power, the wire feed rate, the
wire - beam interaction point and the power density. The
beam melts the wire, which flows into the gap and fills it.
With very low wire feed rates the power is absorbed to the
melt drop, not to the unmelted wire, causing the wire to
melt by heat conducted from the melt drop [11].
With a low wire feed rate the shape of the re-
flected power is different, having a mushroom shape. The
leg of this reflection pattern is formed by a beam reflected
from the melt in the wire front edge and a cap is a reflec-
tion from the unmelted wire edge. The molten material,
which flows into the gap, melts the walls of the joint to-
gether with a part of the beam reflected into the groove [3].
The energy used to melt the base material is
claimed to be equal to autogenous welding of similar pene-
tration depth [12]. However, the amount of energy required
to heat and melt the wire should be added, together with
the laser energy loss via the joint air gap [12]. The laser
beam showed parameter-dependent melting behavior as
well as achievable maximum wire feed rates. Wire melting
can be divided into three characteristic melting procedures
[11]. The first procedure is such that the laser power is
convected from the melt drop to the wire. The second pro-
cedure is caused by direct contact of the laser beam to solid
wire. The third mechanism is basically similar to the sec-
ond but the plasma cloud generated is preventing direct
contact with the laser beam and filler wire.
At low rate the wire fusion front is positioned al-
most horizontally below the wire tip. The wire melts not by
direct contact with the laser beam but as a consequence of
heat convection via the droplet contacting the laser beam.
Because of an increase in wire feed rate the fusion front at
the wire end is turned into a vertical position. During this
phase of melting, metal fume plasma flowing off the fusion
front forms. Since the flow of metal fume is directed along
the normal line of the source surface, the estimation of the
average inclination of the surface of the wire can be carried
out. The slope angle of the fusion zone into the laser beam
is αw = 85° [6]. The fraction of reflected laser beam has
been calculated to be 8.6% of the incoming laser beam
based on reflectivity and size of the beam – wire interac-
tion point [12]. The beam reflection from the wire surface
may logically be compared with the reflection of a laser
beam from a flat steel surface, which is demonstrated [8].
The studies show that the reflectivity of a surface
is at its minimum when the incidence angle of the laser
beam to the metal surface is close to 90°. The absorptivity
decreases steeply when the angle of incidence is changed
from 90 degrees (absorptivity 80%) to an angle of 70° (ab-
sorptivity 20%) [13]. The angle of the wire front edge is
reported to change to 55° [11]. If the wire feed rate is too
high a part of reflected beam is directed towards the top
surface of the workpiece and inside the keyhole.
If the wire feed position misses the plane charac-
terized by the laser beam optical axis and the welding di-
rection, the weld cross-section is not symmetrical. For ex-
ample a misalignment of 0.4 mm in the horizontal direc-
tion transverse to the plane defined by the laser beam and
filler wire (FY) has been shown to lead to an asymmetrical
weld [3]. A similar result is also reported by the author [14]
showing that the weld volume increases on the opposite
side of the wire misalignment position transverse to the
welding direction. During welding the filler wire is fed into
the melt pool or into a focal cylinder. When the laser beam
meets the filler wire, several phenomena occur. The phe-
nomena involved with the process itself are not fully clari-
fied yet. In most cases research has concentrated on the
properties of the weld rather than the process optimization.
The filler wire should be fed towards the focal point of the
laser beam [10]. Feeding takes place in the direction of the
joint, at the trailing or leading edge of the keyhole. When
the laser beam impinges on the filler wire, it absorbs part
of the power, part of it passes through the wire and part of
it is reflected away [3]. The fraction of the reflected laser
power depends on the beam power, the wire feed rate, the
wire - beam interaction point and the power density. The
beam melts the wire, which flows into the gap and fills it.
With very low wire feed rates the power is absorbed to the
melt drop, not to the unmelted wire, causing the wire to
melt by heat conducted from the melt drop [11].
1.6. The justification of laser welding
Even though laser welding is utilized so far only
too few areas of industry, it has typically major effect on
the product structure and performance. Laser welding is a
typical new production technology suitable for, e.g. think-
ing models for creative working like shown in Fig. 5.
Fig. 5 Reorganization of creative activity [15]
This welding technology typically can be ex-
ploited full only when the creative problems solution and
new working manners is taken into use. Another indirect
advantage on laser welding is that the digital nature of it
makes the welding process predictable. The weld can be
considered as a module with function to join. This typical
feature of reproducibility makes it possible also to utilize
advanced means to study the quality of product develop-
ment like showed in case of different product design cases
[16].
2. Experimental procedure
The reflection experiments were performed to ex-
amine the laser power – filler wire interaction. The pres-
ence of base material hinders the measurement of both the
intensity and form of the laser beam reflection during
welding. These experiments were therefore carried out
without base material. The laser used in the experiments
was a Rofin-Sinar RS 6000 CO2 laser with a TEM02* beam
mode. The diameter of the raw laser beam was 40 mm and
70
the diameter of focal point 0.26 mm.
The process parameters used, laser beam set up
and shielding gas arrangement, were those typically used
to weld steel thickness in the range 3 - 6 mm, with filler
wire for bridging a butt joint air-gap. The shielding gas
was introduced through a nozzle coaxial with the laser
beam. Shielding was performed with helium gas, with 20
l/min flow rate. Initial trials concerned the investigation of
the reflection angle by placing a cylinder of heat sensitive
paper around the beam - wire interaction point coaxial with
the incoming laser beam. A short laser energy pulse
(1 W/10 ms) was then focused on the surface of the filler
wire and the angle of reflection detected by the heat sensi-
tive paper. The filler wire was a commercially available
G3Si copper coated GMAW wire, of diameter 0.8 mm,
suitable for welding of low alloyed steel.
The literature review and earlier experiments had
shown that the direction of reflected laser beam is typically
aimed symmetrically in a plane defined by laser beam op-
tical axis and filler wire central axis. The reflection is fol-
lowing this plane and the beam is reflected to opposite side
of the beam in comparison to wire feeding. Based on this
information the position of measurement means of reflec-
tion was defined.
2.1. Reflections during laser welding
The welding experiments were performed to es-
tablish suitable welding parameters for laser welding with
filler wire. A schematic illustration of the experimental set-
up is shown in Fig. 6. The beam was focused using a para-
bolic off-axis focusing mirror of 150 mm focal length.
Shielding gas was introduced coaxially with the vertical
laser beam and plasma control gas at a 45° angle to the
laser beam, directed to the beam-wire interaction point.
Wire
Welding direction
Laser beam
Body and
heat paper
Welding direction
Body and
heat paper
Wire
Laser beam
a b
Fig. 6 The arrangements of welding experiments the reflec-
tion definition experiments are also shown as the
body and heat paper: a - is case of trailing feed and
b -case of leading feed
The nozzle diameter was 5 mm. The laser beam
was focused on the surface of the workpiece, in the middle
of the joint, and the filler wire was fed to the interaction
point of the laser beam and the workpiece (Fig. 6).
The parameters used were selected such that the
welding and acceptable weld quality was obviously
reached. The weld quality was evaluated according to valid
laser weld quality standard EN ISO 13919-1, where laser
weld has quality levels B, C, D and rejected quality [17].
A set of simulating experiments of reflected
power measurements were performed. The simulation was
required since the measurement was impossible due to
dimensions of the measuring equipment.
Table 1
The parameters used in experiments
Laser
power
Feed
dir.
Feed
angle
Wire
dia.
Focal
pos.
Feed
pos.
Focal
length
Wel-
ding
speed
Wire
feed
rate
kW o mm mm mm mm m/min m/min
1 5 LF 60 0.8 -2 -2 300 0.60 7.2
2 5 LF 60 1 0 0 150 0.75 6
3 5 LF 60 0.8 2 2 150 0.90 10.7
4 5 LF 45 0.8 0 0 150 0.90 10.7
5 5 LF 45 1 2 2 300 0.60 4.7
6 5 LF 45 0.8 -2 -2 150 0.75 9.2
7 5 LF 30 1 -2 2 150 0.90 7.1
8 5 LF 30 0.8 0 -2 150 0.60 7.2
9 5 LF 30 0.8 2 0 300 0.75 9.2
10 5 TF 60 0.8 2 0 150 0.60 7.2
11 5 TF 60 0.8 -2 2 150 0.75 9.2
12 5 TF 60 1 0 -2 300 0.90 7.1
13 5 TF 45 1 2 -2 150 0.75 6
14 5 TF 45 0.8 -2 0 300 0.90 10.7
15 5 TF 45 0.8 0 2 150 0.60 7.2
16 5 TF 30 0.8 0 2 300 0.75 9.2
17 5 TF 30 0.8 2 -2 150 0.90 10.7
18 5 TF 30 1 -2 0 150 0.60 4.7
The simulating experiments were performed such
that a plate was placed in front of beam-wire interaction
point with a approximated distance of keyhole such that
the plate edge was blocking the reflection aimed towards
keyhole back wall. This “mask” plate was cutting the lower
reflections away letting the beam reflected above the
workpiece surface to proceed and be couched by the power
meter place on top of the plate, see
Fig. 7. This measurement showed how much en-
ergy is aimed such that it is missing the welding process.
Reflected
laser beam
Laser beam
“Workpiece”/
mask plate
Heat sensitive paper
/power gauge Filler wire
Fig. 7 The arrangements of simulating experiments for
power measurement
2.2. Shape of the melting front of filler wire
The shape of wire tip was investigated by photo-
graphic experiments. The wire was again fed into the focal
cylinder in open air and the wire edge was photographed.
The arrangements are shown below in Fig. 8. The laser
welding process generates a metal vapor that can be seen
as a very bright light. The effect of process brightness was
compensated by the experimental arrangement. A 1 kW
diode laser with wavelength of 840 nm was used to illumi-
nate the process. The camera was equipped with a filter
transparent only for the wavelength of illumination laser.
71
The light was aimed to process with short light pulses
making short exposure times possible. The wire feed pa-
rameter combinations used were according to those of the
welding and reflection experiments.
Laserbeam
Filler wire
Camera
Diode laser
Fig. 8 The experimental arrangements of photographic
study
3. Results
The reflection experiments showed clearly that the
filler wire may reflect a considerable fraction of incoming
laser beam. The reflection effects as well on the weld qual-
ity as the energy lost during welding. The shape of wire
end is affected by the wire feeding speed and positional
parameters of laser beam in comparison to filler wire.
3.1. Reflections
The reflection was aimed in each experiment to
the same direction dictated by the filler wire feed direction.
The reflections were aimed to opposite direction of the
wire feed direction (Fig. 9).
The laser power reflected was measured from the
direction shown in Fig. 10. The measured reflected powers
with different parameters are shown in Table 2.
Reflecte d
laser beam
Laser beam
Fil le r wir e
Laser beam
Reflecte d
laser beam
Fill er w ire
Fro n t v ie w
Top view
Side view
Fig. 9 Principle directions of the laser beam reflection from
wire
Table 2
Measured reflected power in open-air experiments
Exp.
Reflected
power, kW
Exp.
Reflected
power, kW
1 1.90 10 0.55
2 0.25 11 1.20
3 0.75 12 1.10
4 0.15 13 0.20
5 1.65 14 0.80
6 1.50 15 2.00
7 1.30 16 2.40
8 0.25 17 0.25
9 2.45 18 0.15
The laser power reflected during welding was
aimed partly to the welding process, partly above the
workpiece. Fig. 10 shows the directions of the reflected
laser power during welding. From the reflection angle it
can be seen that the feed angle and feed position both have
effect on the direction of the reflection. The experiments
marked with W present the welding experiments. The max-
imum measured reflection angles vary from 0 to 43 de-
grees. The highest reflection directions were measured in
case of lowest wire feeding angles (feeding angle 30 de-
grees to workpiece surface).
Fig. 10 Direction of reflected beam during welding. Q, P
and H are experimental series. Three columns,
marked with 30°, 45° and 60°, are showing the
wire feed angle
3.2. Weld quality
The weld quality reached with various parameters
varied according to the parameters used. The parameters
were selected in such a manner that it was expected that
part of the experiments do not produce acceptable quality.
The quality was evaluated according to EN ISO 13919-1
standard taking into account the material thickness and the
13 weld fault criteria and each weld was given a weld qual-
ity class. The quality varied from the best quality B in case
of experiments 1, 6, 8 and 10 to the quality not qualifying
the lowest level D.
Illumination
laser
Camera
Process
72
Table 3
The quality levels reached in experiments
Exp.
Weld qual-
ity
Exp.
Weld
quality
1 B 10 B
2 D 11 F
3 F 12 F
4 F 13 C
5 - 14 F
6 B 15 F
7 - 16 -
8 B 17 F
9 F 18 C
The achieved quality class B, C, D is shown with
equivalent letter. Letter F means the part failed the quality
testing and – means that the part was not possible to evalu-
ate.
The cross-section examples of different quality
levels are shown below in Fig. 11.
a
b
Fig. 11 Examples of weld cross-sections reaching different
quality levels: a - weld class B in case of experi-
ment 1; b - failed weld in case of experiment 2
3.3. Wire melting
The photography of the phenomena during melt-
ing of the wire in open air resulted acceptable quality pho-
tographs. The wire, droplet and wire beam interaction zone
are all clearly visible. The disturbing plasma cloud can be
seen in some photographs as a pale mist escaping from the
interaction zone to the direction about parallel to the mol-
ten surface.
The illustrations of the melting process show
clearly that the reflection theory and the shape of the filler
wire melting edge causes measured reflections. The shape
of the melting front is cylinder and it looks like the half of
keyhole formed by laser beam. When the shape is evalu-
ated on the other direction it can be seen that it is round
showing upper part to be more horizontal and turning to
vertical when looking further down. This roundness de-
pends on the wire feed rate and position it will obviously
reflect the laser beam like seen during reflection experi-
ments. Fig. 12 shows the melting edge of filler wire in case
of backlight when using three different wire feed rates. In
case of 6 m/min the feed rate is high enough to occasion-
ally turn the melting edge further horizontal, but when
reaching feed rate of 8 m/min and further we can see that
the melting edge is in almost 45° angle to vertical, the wire
still melting totally.
a b
c
Fig. 12 The shape of wire edge with welding speed:
a - 4 m/min; b - 6 m/min; c - 8 m/min. Laser beam
is introduced in a vertical direction and the wire in
a 45° angle to it
The melting edge shape can be seen from the
Fig. 12. The shape of the edge changes with the feeding
rate since the time the wire remains longer time unmelted
while entering the focal cylinder is changing. The front
edge of the wire is concave. The droplets leaving the wire
edge are quite large in diameter and there seem not to be
extremely dramatic changes in the drop size when compar-
ing the different wire feed rates.
4. Discussion
The experiments carried out showed that like as-
sumed the filler wire entering the laser welding process
causes some reflections of laser beam. This is happening
for two reasons first because the wire is cold and secondly
the beam position in feeding is not optimum geometrically
to focal point neither if feeding angle nor if feeding posi-
tion is considered. Due to this nonoptimized position the
absorption and the laser beam intensity do not reach their
maximum in the point of interaction. The wire can and will
reflect part of the incoming laser beam. This reflection is at
its minimum when feeding parameters are correct and es-
pecially when laser power in comparison to feeding rate is
sufficient. When proceeding towards the maximum melt-
ing capacity and exceeding by increasing the wire feed
rate, in case of free air melting tests, it was noticed that
73
total melting of the wire was possible only to some extent
and after that the filler wire does not anymore melt but is
only heated. In this case the solid wire hits the laser beam
in non-optimum positions and the fraction of reflected
beam almost reached 50% (2.45 kW of the laser power of
5 kW was reflected in worst case) of incoming laser power.
Since the reflective surface is partly formed by the solid
wire, the direction of this reflected beam is aimed above
the workpiece during welding if the wire is fed in a low
angle compared to the workpiece surface. The direction of
this reflection is surprisingly accurate and aimed by the
reflecting surface angle to the laser beam. The most risky,
low feed angle must usually be avoided in practice because
the arrangement requires lot of room in front of welding
nozzle, which is not typically available during practical
welding. In practical applications this low feeding angle is
not used due to the dimensional problems of feeding
equipment and complexity increase in case of 2-3 D appli-
cations.
The photographs of wire end in melting show that
the assumed shape of the wire end is equivalent with sur-
prisingly high reliability to the assumptions of wire edge
quality gained by the reflection experiments. The photo-
graphs showed also that the high wire feed rate will that
block the keyhole. As is shown the higher the wire feed
rate the deeper in enters the focal point prior melting.
When the critical speed is exceeded (5 kW laser power and
0.8 mm wire diameter the critical speed is 8 m/min) the
wire will pass the focal point, still melting but generating
large molten droplets. The photographs of the molten drop-
let show also the fact the molten material moves below the
wire – away from the laser beam. This caused by the pres-
sure of the plasma cloud that introduces forces exceeding
the force of gravity and the feeding movement.
There is a lower risk of weld imperfections when
the energy input is increased even though the feeding posi-
tion parameters are not optimized. In practice this can be
done by lowering the welding speed from the absolute
maximum. The higher energy input generates a larger
cloud of metal vapor cloud and larger weld pool. In this
case the power intensity required for melting and vaporiz-
ing the material can be reached. The formed large plume of
metal vapor will melt the wire even if it is fed inaccurately;
the plume also absorbs the laser beam reflected from the
wire. If wire feed angle is steep, i.e. more than 45°, most of
the reflection is directed into the keyhole. Steep feeding
angle is on the other hand risky if the wire does not hit the
joint.
a b
Fig. 13 The effect of wire feeding position to laser beam
behavior in case: a - poor feeding parameters;
b - optimum feeding parameters
The fraction of laser energy reflected from wire
should be taken into account in the energy balance. The
amount of lost laser energy can be in worst case quite a
large fraction from the incoming laser energy. If the wire
position and feed angle is correct this energy is, however,
aimed towards the keyhole back wall and utilized totally to
welding process. In this case the multiple reflections start
from the wire melting edge and continue inside the key-
hole. Fig. 13 illustrates the behavior of laser beam in case
of poor wire feeding parameters (position and angle) and in
case of optimum wire feeding parameters.
5. Conclusions
The experiments carried out showed that laser
welding with filler wire can be utilized with the correct
parameter combinations. The parameters can be defined
from the gap volume, base material and laser power avail-
able. Bridging the air gap will lead to a wider weld, requir-
ing a larger volume of material to be melted. Increasing the
energy input can achieve this. The welding parameters can
be optimized within system limits and the lowest required
energy input can be established.
The test methods showed new phenomena in the
welding process. The melting edge of the wire can under
some conditions a surface that will reflect a small amount
of the laser beam. This surface changes with a change in
the filler wire speed, or rather in energy input vs. wire vol-
ume. This phenomenon has an effect on weld quality. The
experiments carried out showed clearly that the wire posi-
tioning accuracy is in the order of 1mm without any dra-
matic problems. The optimization of the process requires
more research concerning the phenomena involved.
The fraction of the beam reflected by the wire in-
creases with an increase in the filler wire feed rate, a de-
crease in the laser power, or an increase in the focal length
of the focusing mirror.
The best weld quality can be achieved when the
wire feed angle is between 45° and 60° to the horizontal.
The wire diameter should be less than the air gap width
and the wire – laser beam interaction point should be on
the level of workpiece top surface or slightly below it.
Filler wire feed combined with optical seam track-
ing and adaptive adjustment of the wire feed volume ac-
cording to gap volume will increase the use of laser weld-
ing especially in cases where a hybrid welding process,
combination of laser and arc welding, cannot be used.
Acknowledgements
The author wishes to show his gratuity to financ-
ing partner National Technology Agency of Finland. The
author also wants to wish his gratuity to co workers in the
laboratory Mr. Pertti Kokko for his assistance in practical
experiments, Mr. Antti Heikkinen for his work in metal-
lography, Dr. John Ion for fruitful discussions and last but
not least Dr. Taito Alahautala from company Cavitar Oy
for his assistance in high speed photography.
References
1. Lampa, C., Powell, J. Ivarson, A., Magnusson, C.
Factors affecting efficiency of laser welding. -Lasers in
Engineering, 1995, No:4, p.73-83.
74
2. Lampa, C. Laser Welding; Energy redistribution and
weld geometry. Doctoral Thesis. Lulea University of
Technology. 1997:33. ISSN:1402-1544. ISRN: LTU-
DT—1997:33—SE.-137p.
3. Arata, Y., Maruo, H., Miyamoto, I., Nishio, R. High
power CO2 laser welding of thick plate. multipass
welding with filler wire. -Trans. Jwri 15, (2), 1986,
p.199-206.
4. Watson, M.N., Dawes, C.J. Laser welding of
structural steel with wire feed. -The Welding Institute
research report 264/1985, The Welding Institute,
Cambridge, March 1985. -36p.
5. Nielsen, S.E., Hansen, L.E., Kristensen, J.K. High
power laser welding of C/Mn steel. -The effects of
using different filler materials and plasma control
gases. Proc. 5th int. Conf. on Welding and Melting by
Electron and Laser Beams, 14-18 June, 1993, La Baule,
France, Organized by Institut de Soudure and TWI,
vol.1, p.203-210.
6. Dilthey, U., Hendricks, M., Huwer, A., Jacobskötter,
L., Scneegans, J. Adaptive Drahtzuführung für das
Laserstrahlschweissen bei variable Fügespaltweiten.
Proc. Int. Conf.Beam Technology, Karlsruhe, Germa-
ny, 13.-14.3.1991, DVS-Berichte Band 135, p.71-74.
7. Bagger, C. Investigations in On-Line Process Control
of the Laser Welding Process. Doctoral thesis. -
Technical University of Danmark, 1991, APP.91-
02/PI.91.1-A.-280p.
8. Cleemann, L. Schweissen mit CO2-Hochleistungsla-
sern. -Düsseldorf: Technologie Aktuell 4, VDI-Verlag
GmbH, 1987.-256p.
9. Hügel, H., Dausinger, F. Interaction Phenomena and
Energy Coupling in Laser Treatment Processes.
-Process Fundamentals par 2/3. Euro Laser Academy,
1996.-57p.
10. Panten, M., Schneegans, J., Hendricks, M., Huwer,
A., Jacobskötter, L. Laser Welding with Filler Wire.
IIW-DOC. IV-545-90, 1990.-17p.
11. Dilthey, U., Schneegans, J. Studies in laser beam
welding with filler wire addition of unalloyed and low
alloyed steels. Schweissen und Schneiden 46, 1994,
Heft 3, p.119-123.
12. Carlson, K.W., Gregson, V.G. A parametric study of
narrow gap laser welding. -Materials Processing Confe-
rence ICALEO'86, Arlington, Virginia, USA, 10.-
13.11.1986, Laser Institute of America, p.211-222.
13. Beyer, E., Behler, K., Herziger, G. Influence of laser
beam polarization in welding. -Proc. 5th Int. Conf.
Lasers in Manufacturing, September 1988, p.233-240.
14. Salminen, A.S. Laser Welding with Filler Wire.
Licenciate thesis. -Lappeenranta University of
Technology. Lappeenranta 1992.-103p. (in Finnish).
15. Povilionis, A., Bargelis, A. Structural optimization in
product design process. -Mechanika. -Kaunas:
Technologija, 2010, Nr.1(81), p.66-70.
16. Čikotienė, D., Bargelis, A. Research of quality impact
to the product design properties and characteristics.
-Mechanika. -Kaunas: Technologija, 2009, Nr.5(79),
p.63-67.
17. EN ISO 13919-1: Welding – Electron and laser-beam
welded joints – Guidance on quality levels for
imperfection – Part 1: Steel.-9p.
A. Salminen
UŽPILDANČIOS VIELOS SĄVEIKA SU LAZERIO
SPINDULIU SUVIRINANT MAŽAI LEGIRUOTĄ
PLIENĄ
R e z i u m ė
Lazerinis suvirinimas vis plačiau taikomas įvai-
riose pramonės šakose. Suvirinimas naudojant užpildančią
vielą dažnai vertinamas kaip įprastinis metodas. Šis darbas
parodė, kad nustačius tam tikrus apribojimus užpildanti
viela gali būti naudojama pramoniniam suvirinimui. Laze-
rio spindulio atspindys nuo vielos paviršiaus gali būti dide-
lis, tačiau jį galima kontroliuoti pagal normalius atspindžio
dėsnius. Suvirinimo procese gali būti panaudota ir dalis
spindulio atspindžio nuo vielos paviršiaus.
A. Salminen
THE FILLER WIRE - LASER BEAM INTERACTION
DURING LASER WELDING WITH LOW ALLOYED
STEEL FILLER WIRE
S u m m a r y
Laser welding is gaining new applications in vari-
ous industries. Often the use of filler wire is not considered
due to reputation of complicity. This study has, however
gained almost no interest of research groups after millen-
nium. The study showed that when certain rules are fol-
lowed the filler wire feeding can be used and applied to
industrial applications. The reflection of the laser beam
from wire surface can be considerable, but I can be con-
trolled due to its behavior according to the normal reflec-
tion laws. Even the fraction of beam reflecting from the
wire surface can be utilized to the process.
А. Салминен
ВЗАИМОДЕЙСТВИЕ ПРИСАДОЧНОЙ ПРОВОЛОКИ
С ЛАЗЕРНЫМ ЛУЧОМ ПРИ ЛАЗЕРНОЙ СВАРКЕ
НИЗКО ЛЕГИРОВАННОЙ СТАЛИ
Р е з ю м е
Использование лазерной сварки в промыш-
ленности расширяется. Применение присадочной про-
волоки часто оценивается как обычный метод. Данное
исследование показало, что при введении некоторых
ограничений использование присадочного материала
может быть применено в промышленности. Отражение
лазерного луча от поверхности проволоки может быть
значительным, но его можно контролировать по нор-
мальным законам отражения. В процессе сварки может
быть использовано даже часть отраженного луча от
поверхности проволоки.
Received February 23, 2010
Accepted June 21, 2010
