Assessment of the effect of Nd:YAG laser pulse operating parameters on the metallurgical characteristics of different tool steels using DOE software

Abstract

To ensure the reliability of repair welded tool surfaces, clad quality should be improved. The relationships between metallurgical characteristics of cladding and laser input welding parameters were studied using the design of experiments software. The influence of laser power, welding speed, focal point position and diameter of welding wire on the weld-bead geometry (i.e. penetration, cladding zone width and heat-affected-zone width), microstructural homogeneity, dilution and bond strength was investigated on commonly used tool steels 1,2083, 1,2312 and 1,2343, using DOE software.

Full text

Page 1

T. MUHI^, L. KOSEC, G. LIEDL, M. PLETERSKI
ASSESSMENT OF THE EFFECT OF ND:YAG
LASER PULSE OPERATING PARAMETERS
ON THE METALLURGICAL CHARACTERISTICS
OF DIFFERENT TOOL STEELS USING DOE SOFTWARE
Received – Prispjelo: 2010-04-04
Accepted – Prihva}eno: 2010-06-10
Original Scientific Paper – Izvorni znanstveni rad
INTRODUCTION
Polymer injection molds often require some form of
repair cladding during the operation and whole life cycle
due to the different mechanisms of failure in the molds
?1-3?. Special processes and welding procedures are re-
quired for repair cladding due to the considerable metal-
lurgical concerns and low weldability of tool steels.
Among other known processes, laser cladding is becom-
inganincreasinglyattractivetechniqueduetotheproper-
ties of laser radiation: high input energy, low distortion,
minimum dilatation between the base material and clad,
aswellasamorerefinedmicrostructureduetohighcool-
ing rates ?4?. Despite the fact that the necessary equip-
mentandfillermaterialsarereadilyavailableonthemar-
ket, there is a lack of published scientific work on pulse
laser repair cladding of tool steels. The input parameters
of laser welding determine the shape of laser clad-bead
and the metallurgical properties of clad ?5?. The primary
objective of this work is to establish and statistically ana-
lyze the optimal conditions of pulsed laser input parame-
ters for cladding of tool steels using preplaced wires.
EXPERIMENTAL PROCEDURE
Material
Thesteelsusedinthepresentstudyarecommercially
available tool steel types 1,2083 (X42Cr13Mo4),
1,2343 (X38CrMoV51) and 1,2311 (X40CrMnMo7),
which are commonly used for dies for die casting and
plastic injection molding. Test plates from steels 1,2083
and 1,2343 were vacuum-quenched and tempered to
level of hardness 54 HRC and 47 HRC, while the plate
made of steel 1,2311 was used in a pre-hardened state
(33 HRC). The filler wires used in the experiments are
commercially available wires usually used for MIG
welding, having similar nominal chemical composition
to the base material. In order to investigate the influence
of wire diameters on the clad shape, we used two differ-
ent wire diameters of 0,35 and 0,5 mm.
Laser cladding
Thecladdingoftestspecimenswasperformedwitha
pulsed Nd:YAG laser (LASAG model SLS200 CL60)
by melting of the preplaced filler wire. Argon gas 5,0
with flow rate of 5 l/min was used for shielding. The la-
METALURGIJA 50 (2011) 2, 101-105
101
To ensure the reliability of repair welded tool surfaces, clad quality should be improved. The relationships bet-
ween metallurgical characteristics of cladding and laser input welding parameters were studied using the de-
sign of experiments software. The influence of laser power, welding speed, focal point position and diameter
of welding wire on the weld-bead geometry (i.e. penetration, cladding zone width and heat-affected-zone
width), microstructural homogeneity, dilution and bond strength was investigated on commonly used tool ste-
els 1,2083, 1,2312 and 1,2343, using DOE software.
Key words: laser cladding, tool steel, DOE software
Procjena utjecaja operativnih parametara Nd:YAG pulsnog lasera na metalur{ka svojstva raz-
li~itih alatnih ~elika pomo}u DOE softvera. Pobolj{anje kvalitete navarenog sloja nu`no je za osiguranje
pouzdanosti povr{ina alata, obra|enih postupkom reparaturnog navarivanja. Odnos izme|u metalur{kih oso-
bina navarenog sloja i ulaznih parametara laserskog zavarivanja istra`en je pomo}u softvera Design of experi-
ments. Utjecaj snage lasera, brzine zavarivanja, polo`aja `ari{ta i promjera `ice za zavarivanje na geometriju
navara (t.j. na penetraciju, {irinu navara i {irinu zone utjecaja topline), mikrostrukturna homogenost, mije{anje
i snaga spoja istra`eni su na uobi~ajenim alatnim ~elicima 1,2083, 1,2312 i 1,2343 pomo}u softvera DOE.
Klju~ne rije~i: reparaturno navarivanje, alatni ~elik, DOE programska oprema
ISSN 0543-5846
METABK 50(2) 101-105 (2011)
UDC – UDK 621.791.725:669.14:004.4=111
T. Muhi~, TKC d.o.o., Ljubljana, Slovenia, L. Kosec, Faculty of Natural
Sciences and Engineering, Ljubljana, Slovenia, M. Pleterski Faculty of
Mechanical Engineering, Ljubljana, Slovenia, G. Liedl Faculty of me-
chanical and industrial engineering, Vienna, Austria

Page 2

ser beam was moved in automatic mode; only the wire
waspreplacedbyaoperator.Theinfluenceofoperator’s
skills on the results of manual laser cladding is minimal,
so it can be neglected ?6?. The focal distance of lenses
was160mmandwascontrolledbypreciselasertriangu-
lation. Preliminary trials using different values of laser
peak power (Pp), laser feed rate (v), pulse frequency and
pulseduration(?)wereundertakentodeterminefeasible
test conditions (Table 1).
Design of experiment
Lasercladdingtestswereperformedtodeterminethe
laser operating parameters having the greatest influence
on the clad geometry and dilution ratio, and to examine
the correlation between the operating parameters and
the metallurgical properties of the clad. The analysis of
experimental results was based on the analysis of vari-
ance (ANOVA) method. A 23design was executed. The
experiment was carried out according to the designed
matrix in a random order to avoid any systematic error
(Table2).Thetransversallysectionedspecimensofclad
were measured using an optical microscope and ade-
quate software. The average of three weld profiles was
recorded for each response. Responses ?a, ?e, DR and
fraction P/R were calculated according to the equations
shown in Figure 1. A test of significance of the regres-
sion models, a test of significance of individual model
coefficientsandthelack-of-fittestwerealsoperformed.
Step-wise regression method was used to automatically
eliminate the insignificant terms of the model. The re-
sulting ANOVA (Table 3) summarizes the analysis of
variance for each response and shows the significant
terms of model.
Geometrical and microstructural analysis
After laser cladding, the specimens were sectioned
transversally to the clad track, polished with SiC paper
and diamond paste. The samples were than etched with
Nital 2 % etchant. The geometry and microstructure of
the clad track were than studied by optical microscopy
and analyzed with PRIMO software.
RESULTS AND DISSCUSION
Theanalysisofvariancedeterminestherelativeinflu-
ence of each factor. The analysis of variance (ANOVA)
on the measured data suggests the following findings.
Reinforcement
OnthebasisofANOVAresults,wecanconcludethat
Pp, ?, ? and ? are significal model terms associated with
reinforcement. In the Figure 2 it is clear to observe, that:
– lowerreinforcementisattainedwithhighervalues
ofPp,?and?.Wethereforegethighermeltingloss
at higher values of laser caldding parameters;
– larger wire diameter on both materials gives clad
with higher reinforcement.
Penetration
On the basis of interaction plot (Figure 3) we can
conclude that the penetration ratio is similar for smaller
diameters of the filler wires. For larger wire diameters
we can observe greater penetration in the material
1,2343. The lower thermal diffusivity of 1,2343 ex-
plains the higher penetration of clad in the base material
compared to 1,2312.
According to the performed tests, we can conclude
that the most significant laser processing parameters of
pulse cladding are duration of pulse and peak power.
The wire diameter is a less significant parameter. In or-
der to investigate further, we designed a more detailed
experiment on three different types of tool steels:
102
METALURGIJA 50 (2011) 2, 101-105
T. MUHI^ et al: ASSESSMENT OF THE EFFECT OF ND:YAG LASER PULSE OPERATING PARAMETERS...
Figure 1 Principal clad geometry in cross section and re-
sponse values
Table 1. Mean values of control factors at each level
FactorP
Mean
Level 1Level 2
APp/ kW1,52
B
? / ms812
C
? / Hz510
Dv / mm/s1 1,25
E
? / mm0,330,5
Fmaterial1,23431,2312
Figure 2 Plot of main effects on the reinforcement re-
sponse

Page 3

1,2083,1,2343and1,2312(Figure4).Therefore,athree
level and three factor DoE has been considered. In this
way, ANOVA results with a quadratic model. This
showsthebestfittotheexperimentaldata.Theresulting
response surface for the adjusted model is shown in Fig-
ure 5, where the values of fraction P/R vary. Other pa-
METALURGIJA 50 (2011) 2, 101-105
103
T. MUHI^ et al: ASSESSMENT OF THE EFFECT OF ND:YAG LASER PULSE OPERATING PARAMETERS...
Table 2 Design matrix with process factors and experimentally measured responses
Std. O.
FactorsResponses
ABCDEFW/ µmP/ µmR/ µm
á/ ° DR/ %
öe
öa
P/R
1111111827430160152 30,22 5,1 1,922,69
2211112 1 21084596 16417,5912,61,438,8
3121112 1 242 62386 16818,46 14,4 1,997,24
42211111 025 94411416113,338,91,098,28
51121121 054532 11216122,079,41 1,984,75
6212111814701 621658,1613,11,1611,31
7122111 1 10149860167 14,3718,32,218,3
82221121 411 947441725,48 32,0 1,49 21,52
9111212 1 06242512215636,328,07 2,5 3,48
10211211 1 033842 10115712,44 10,21,23 8,34
11121211947602120 16016,16 7,891,575,02
122212111 291 94465169 10,9119,8 1,3714,52
13112211 837821104 148 13,198,051,02 7,89
142122121 23982783171 11,52 14,931,5 9,96
15122212 1 195558 83165 19,214,4 2,146,72
16222211 821 83252165 6,23 15,790,99 16
17111122 1 010399264138 53,95 3,832,531,51
1821112199 577242 13242,65 4,131,732,38
191211211 007334 25614751,46 3,393,01 1,3
202211221 388 861 180158 25,937,711,61 4,78
21112121 968 250245 142 61,963,95 3,87 1,02
222121221 192 675200 14531,745,961,773,38
231221221 195 52918514437,19 6,462,26 2,86
242221211 171639136 164 19,09 19,091,83 4,7
25111221 835200297127 75,452,81 4,180,67
26211222 1 202 767 20315632,675,92 1,573,78
271212221 263 545211151 40,575,99 2,322,58
28221221 1 046707 22413928,84,671,48 3,16
291122221 043 355 250141 53,084,17 2,941,42
30212221994446 23413948,514,252,231,95
311222211 023 347219143 48,124,67 2,95 1,58
32122222 1 427814143 16222,42 9,98 1,755,69
Figure 3 Main effect plots and interaction plot on process response penetration in µm

Page 4

rameters remain at fixed values: ?=10 Hz; v=1,25 mm/s
and ?=0,5 mm).
On contour graphs we can identify three zones as a
function of P/R fraction. These three zones are in corre-
lation with the clad bond strength. More importantly, no
defects such as cracks, pores and incomplete fusion
were present in the II. zone (2 < P/R < 3).
Weld metal microstructural characteristics
Presence of defects due to incomplete fusion on the
fusion line was noticed in some deposits (zone I).
Cracks and micro cracks in the clad layer are one of the
major concerns in laser pulse cladding, as they can af-
fect the clad application. Cracks were found at high
pulse energy (Figure 6a).
The cracks starts to emerge at a value of fraction P/R
higher than 3,5 (zone III).
Macro-cavities are also generated during pulsed la-
ser cladding. These cavities are initiated at the root of
clad during laser cladding in deep penetration mode
(P/R>4). They are located mostly at the edges and at the
root of the bead (Figure 6b).
104
METALURGIJA 50 (2011) 2, 101-105
T. MUHI^ et al: ASSESSMENT OF THE EFFECT OF ND:YAG LASER PULSE OPERATING PARAMETERS...
Table 3
ANOVA partial sum of squares for reinforce-
ment
Source Sum of squaresd.fMean squareF
Model1,59 x105
439 772,6693,3
Pp
11 063.281 11 063,2826,8
?
11 137.81111 137,7826,9
?
8 745,031 8 745,0321,1
?
1,281 x105
1 1,28x105
310
Residual11 140,8427 412,62
Cor total1,702x105
61
Note: R2=0,9346; predicted R2=0,9081; adjusted R2=0,9249;
adequate precision= 29,185
Figure 4 Clad shapes in the cross section
Figure 5 Contour graph shows the effect of Ppand ? on the fraction P/R for different types of steels

Page 5

CONCLUSIONS
The main findings of the presented research, within
the factor limits, can be summarized as follows:
– input laser parameters related to pulse energy af-
fectthecladshape.Cladheightisreducedwithin-
creasing pulse energy;
– the depth of penetration into the substrate in-
creases with increasing pulse energy, pulse dura-
tion and pulse frequency, but decreases with in-
creasing wire diameter;
– penetration depth is also strongly affected by the
typeofmaterial.Steelwithlowerthermalconduc-
tivity (1,2083) shows a higher penetration rate;
– laserinputparametersshouldbechosenaccording
to the physical properties of material, especially
thermal conductivity;
– higher values of fraction P/R bring an increased
occurrence of porosity and cracking of clad. This
can be eliminated by slowing the cooling rate, by
reducing the cladding speed or by conventional
preheating.
REFERENCES
?1?
A. Persson, S. Hogmark, J. Bergström: Failure modes in
field-tested brass die casting dies, Journal of Materials Pro-
cessing Technology, 148 (2004) 1, 108-118
M. Muhi~, J. Tu{ek, F. Kosel, D. Klob~ar, M. Pleterski:
Thermal fatigue cracking of die-casting dies, Metalurgija
49 (2010), 9-17
D. Klob~ar, J. Tu{ek: Thermal stresses in aluminium alloy
die casting dies,Computational Materiales Science, 43(
2008) 4, 1147-1154
A. Kaldos, H. J. Pieperb, E. Wolf: Laser machining in die
macking - a modern rapid tooling process, Journal of Mate-
rials Processing Technology, 155-156 (2004), 1815-1820
F.MalekGhaini,M.Hamedi,M.Torkamany,J.Sabbaghza-
deh: Weld metal microstructural characteristics in pulsed
Nd: YAG laser welding, Scripta Materialia, 56 (2007) 11,
955-958
E. Capello, B. Previtali: The influence of operator skills,
process parameters and materials, Journal of Materials
Processing Technology, 174 (2006) 1-3, 223-232
?2?
?3?
?4?
?5?
?6?
Note: The responsible for English language is Marko Ore{kovi~,
Ljubljana, Slovenia
METALURGIJA 50 (2011) 2, 101-105
105
T. MUHI^ et al: ASSESSMENT OF THE EFFECT OF ND:YAG LASER PULSE OPERATING PARAMETERS...
Figure 6 a) clad solidification crack
b) porosity in clad root