direct laser fabrication of nickel alloy samples
TRANSCRIPT
Direct laser fabrication of nickel alloy samples
Li Penga,*, Yang Taipingb, Li Shengb, Liu Dongshengb, Hu Qianwub,Xiong Weihaoa, Zeng Xiaoyanb
aState Key Lab of Die and Mould Technology, Huazhong University of Science and Technology, Wuhan 430074, ChinabState Key Lab of Laser Technology, Huazhong University of Science and Technology, Wuhan 430074, China
Received 6 September 2004; accepted 13 January 2005
Available online 4 March 2005
Abstract
Direct laser fabrication (DLF) is an advanced manufacturing technology which can build full density metal components directly from
CAD files without using any modules or tools. An open-loop controlled hardware system and associated control software for the DLF process
of nickel alloy samples was constructed in our work. The hardware system is consisted of a CO2 laser, a 4-axis CNC table, a coaxial powder
nozzle and a powder recycler. In order to achieve the maximum flexibility and extensibility for the fabrication of metal parts, path plug-ins
was introduced into the control software. The effect of the specific energy on the cross-section shape of nickel alloy cladding was studied by a
single-track cladding experiment with different laser processing parameters. The comprehensive effect of the optimized laser processing
parameters was also studied by an orthogonal experiment. The experimental results showed that the specific energy for laser processing is
the most important factor, which controls the component qualities. There is an appropriate range for the specific energy in which the nickel
alloy samples can be fabricated layer by layer with a uniform height. If the specific energy is too low, the inner height of a sample is
lower than its contour height. Banding structure can be observed in the microstructure of nickel alloy samples. The gain size in the light zones
of the bandings is much smaller than those in the dark zones.
q 2005 Elsevier Ltd. All rights reserved.
Keywords: Laser; Nickel alloy; Direct Laser Fabrication (DLF); Specific energy
1. Introduction
Direct Laser Fabrication (DLF) is an advanced manu-
facturing technology which can build full density metal
components directly from CAD files without using any
modules or tools [1,2]. With one step manufacturing ability,
DLF technology can greatly reduce the lead-time and
investment cost for modules and dies design, hard or rare
metal components fabrication, components repair, etc.
Laser Engineering Net Shaping (LENS), Directed Light
Fabrication (DLF) [3], Direct Metal Deposition (DMD) [4],
Shape Deposition Manufacturing (SDM) [5], etc. are all
successful examples of DLF technology.
The principle of DLF technology is based on the Rapid
Prototyping (RP) and laser cladding. First the CAD model
of component is sliced into a series of parallel layers with
0890-6955/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijmachtools.2005.01.014
* Corresponding author. Tel.: C86 27 87541780; fax: C86 27 87542427.
E-mail address: [email protected] (L. Peng).
a certain build-height. Then, the solid areas in a layer are
filled by well-designed fill path (or named ‘tool path’) which
are translated into CNC instructions such as G-codes.
Controlled by the instructions, a focused laser beam moves
on a metal deposition substrate in x–y plane to create a
molten pool into which metal powder or wire is simul-
taneously transported to deposit a layer along with the pool
solidification. And then, an assembly of lens and nozzle
raise a build-height alone z-axis to repeat the process again.
Finally, the metal component with desired geometry can be
fabricated layer by layer.
As a particular application, DLF is a very attractive
processing technique to the aerospace industry because a
great amount of aerospace components with complex shape
are made of metals such as titanium, nickel alloys, which are
very difficult to process for conventional methods. Nickel
alloy provides high wear and corrosion resistance at ambient
and high temperature environment. Some researches on the
DLF process of titanium alloy such as Ti–6Al–4V were
International Journal of Machine Tools & Manufacture 45 (2005) 1288–1294
www.elsevier.com/locate/ijmactool
Fig. 1. Coaxial powder nozzle designed for DLF process.
L. Peng et al. / International Journal of Machine Tools & Manufacture 45 (2005) 1288–1294 1289
reported [3,6,7], however, there was very little literature
published on the DLF process research of nickel alloy
samples. The present work describes our researches on this
subject.
2. Experimental procedure
2.1. Hardware description
The hardware system for our DLF process consists of
a 5 kW ROFIN TR050 CO2 laser, a CNC system, a
powder feeder, a powder nozzle and a powder recycler.
The laser is a DC-excited, fast axial flow laser which
low TEM mode and excellent stability are very suitable
for the requirements of DLF process. The CNC system is
a 4-axis HNC-1 CNC unit and stage. The powder feeder
is a HGL-III Powder Feeder made from Huazhong
University of Science and Technology (HUST) for
general purpose such as laser cladding, laser welding,
laser repairs, etc. During the fabrication process, the Ni
alloy powder was injected into the molten pool through
the nozzle to deposit a layer along with the pool
solidification. A coaxial powder nozzle with four tubes
shown in Fig. 1 was taken in this experiment. The
unmelted powder can be recycled by a powder recycler
for later use. The CNC unit is the core device in the
hardware system because it controls the start/stop of
Fig. 2. Engineers can choose an appropri
the laser, powder feeder, powder recycler and the
movements of the powder nozzle.
2.2. Control software description
The control software of DLF process can be divided
into several rudimentary tasks such as CAD model
loading and viewing, model slicing, fill path generation
and path-to-instruction translation among which the fill
path design is the most important task of the whole
software development. Because of the influence of
thermal stress and surface tension, the fill path design
of the DLF process is more complicated than the
traditional RP processes such as SL, LOM, etc.,
especially if the process is open-loop controlled. An
excellent path design can not only improve the qualities
of final components but also increase the manufacturing
efficiency of the process. During the development of fill
path module, the effects of component shape on the path
scheme should be considered very carefully. For most
normal metal parts, the raster fill is a simple and efficient
method because the solid areas in each layer are much
wider than the laser spot diameter. For parts with thin-
walled structures, however, the width of solid areas in
each layer is nearly same as the diameter of the laser
spot, and in these situations the contour-offset method
maybe a good choice to fabricate nice thin-walled
structures. Furthermore, engineers sometimes need to
adjust and optimize the fill path once and once again for
some special parts. In order to achieve the maximum
flexibility and extensibility for the fabrication of metal
components with various shapes, a control software
named HGL-RP was developed in this work and path
pattern plug-ins were introduced with which engineers
can choose a suitable fill pattern for a certain component.
Moreover, engineers can even design particular path by
programming a customized plug-in, as described in Fig.
2. Three fill patterns for a hexagon with seven holes by
using different plug-ins are exhibited in Fig. 3. The
optimization of the process parameters and the fabrica-
tion of sample components are all instructed under the
G-code translated from the fill path by HGL-RP.
ate path scheme via path plug-ins.
Fig. 3. Three fill patterns for a hexagon with 7 holes. (a) Raster path.
(b) Offset path. (c) Raster & Offset path combination.
Table 2
The designed array for the orthogonal experiment
No. Laser
power
(W)
Scan
velocity
(mm sK1)
Powder
feed rate
(g minK1)
Overlap
(%)
Specific
energy
(J mmK2)
1 700 4 3.0 33.3 175
2 700 8 4.4 50.0 87.5
3 700 12 5.8 66.7 58.3
4 1000 4 4.4 66.7 250
5 1000 8 5.8 33.3 125
6 1000 12 3.0 50.0 83.3
7 1300 4 5.8 50.0 325
8 1300 8 3.0 66.7 162.5
9 1300 12 4.4 33.3 108.33
L. Peng et al. / International Journal of Machine Tools & Manufacture 45 (2005) 1288–12941290
2.3. Optimization of the process parameters
In order to understand the effects of the process
parameters such as laser power, scan velocity, overlap,
powder feed rate, etc. on the DLF process, a series of
experiments were carried out to get the optimal parameters.
(1)
Tabl
Para
Lase
Scan
Diam
Ni p
Pow
Carr
Subs
Now it is well known that the component qualities are
strongly influenced by the molten pool sizes and the
residual stress, which are mainly controlled by the laser
energy input [8]. Hence it is significant to study the
effect of the specific energy on the cross-section shapes
of the Ni alloy cladding layer on low carbon steel
substrate by single-track cladding with different laser
processing parameters. The parameters used in this
experiment are shown in Table 1.
(2)
An orthogonal experiment was carried out to reveal thecomprehensive effects of the optimized laser processing
parameters. During the experiment, an orthogonal
parameters array as shown in Table 2 was applied
respectively to fabricate nine hexagon prisms whose
dimension is shown in Fig. 4(a). Each prism was built six
layers with the specific combination of process par-
ameters. In order to avoid the possible defects appearing at
the same place in each layer, the hatch orientation in each
layer changed 1208 with the previous layer, as shown in
Fig. 4(b). Other parameters are shown in Table 3.
(3)
Ni alloy samples were manufactured with optimalparameters to examine the process feasibility.
3. Results and discussion
3.1. Influence of the specific energy on the cladding shape
Fig. 5 is the result of the single-track cladding
experiment. The laser energy input of the process can be
e 1
meters for single-track cladding
r power 1500, 2000, 2500, 3000, 3500 (W)
velocity 3, 5, 7, 9, 11 ( mm sK1)
eter of laser spot 2 mm
owder, character Ni–24Cr–4Si–2.5B–3Fe–2.5Mo–0.6C (wt%)
K140/C320 mesh
der feed rate 5.8 g minK1
ying gas, flow rate Ar, 8 L minK1
trate Low carbon steel plate, 10 mm thickness
described by the specific energy E (J mmK2) which can be
calculated by the simplified Eq. (1),
E ZP
DS(1)
where P is the laser power (W), D is the diameter of the laser
spot (mm), S is the scan velocity (mm sK1). The cladding
shape can be described with the cladding width W, the
cladding height H and the cladding depth h which are
defined in the cladding profile as shown in Fig. 5(a).
Fig. 5(b) shows the influence of the specific energy on the
cladding shape.
According to Fig. 5, the width, height and depth of the
cladding layer increased with the increase of the specific
energy. It should be pointed out that the ratio of clad depth
to clad height h/H is an important criterion to estimate
whether the specific energy is appropriate for the process or
not. If the ratio is too large, it refers to that too much
material of previous layers were remelted by the subsequent
laser scanning, causing irregularities in component shape or
even the collapsing in the top portion of the component. On
the other hand, the too small ratio will decrease the bonding
strength between the neighboring layers because the molten
Fig. 4. Prism dimension and hatch orientation designed for orthogonal
experiment. (a) 3D-view and dimension description of experimental prisms.
(b) Hatch orientation of each layer varies 1208 to improve component
qualities.
Table 3
Other parameters for the orthogonal experiment
Diameter of laser spot 1 mm
Ni powder, character Ni–24Cr–4Si–2.5B–3Fe–2.5Mo–0.6C (wt%)
K140/C320 mesh
Carrying gas, flow rate Ar, 8 L minK1
Substrate Low carbon steel plate, 10 mm thickness
L. Peng et al. / International Journal of Machine Tools & Manufacture 45 (2005) 1288–1294 1291
pool energy is insufficient to re-melt enough existing
material. The subsequent orthogonal experiment will
determine the suitable range of specific energy.
3.2. Influence of process parameters on component quality
The appearances of the orthogonal experimental speci-
mens are shown in Fig. 6. We can see the comprehensive
effects of the process parameters on the shape precision of
the components fabricated by DLF processes. According to
Fig. 6, the flatness (i.e. the height uniformity) of the
specimen is mainly controlled by the specific energy of
the laser. When the specific energy is 100–200 J mmK2, the
inner height and the contour height of the specimen is nearly
same and the top layer of the specimen is more flatter than
others, as shown in Fig. 6(a), (e), (h) and (i). When the
specific energy is less than 100 J mmK2, the inner height of
the specimen is lower than its contour height and the top
layer of the specimen is concave, as shown in Fig. 6(b), (c)
and (f). The lower the specific energy is, the more concave
the specimen is. Especially, Fig. 6(c) is concave at the most.
Oppositely, when the specific energy is greater than
Fig. 5. Influence of the specific energy on the cladding shape. (a) Cladding
width, height and depth in a cladding profile. (b) Influence of the specific
energy on the cladding width, height and depth.
200 J mmK2, the inner height of the component is higher
than its contour height and the top layer of the specimen is
convex, as shown in Fig. 6(d) and (g). The higher
the specific energy is, the more convex the specimen is.
Fig. 6(g) is convex at the most.
The verticality of the specimen contours is another
important factor to evaluate the specimen quality. The
morphologies of Fig. 6 demonstrated that the specimen
contours are mainly controlled by the laser power of the
process, even though the specific energy was the same.
When the laser power is 700 W, Fig. 6(a)–(c) exhibit good
verticality. With the increase of the laser power, the collapse
of contours becomes more and more evidence. As a
contrast, although the specific energy of Fig. 6(a) and (h)
is almost the same (specimen 1# in Fig. 6(a) is 175 J mmK2,
specimen 8# in Fig. 6(h) is 162.5 J mmK2), the contour of
specimen 8# is obviously collapse whereas specimen 1# has
a good contour.
The macroscopic optical micrographs which parallel to
the slicing direction of the deposit are shown in Fig. 7.
We can see that banding structures can be observed in all
specimens. From the microscopic optical micrographs
shown in Fig. 8, we can find that the gain size in the
light zones of the bandings is much smaller than those in
the dark zones. Because there are overlaps in the scan
tracks, some zones can be reheated by the laser,
which maybe causes the dendrites in the reheating
zones grow up. We can also find that banding interface
became more and more blurry with the increase of the
specific energy. It maybe caused by enlargement of
the reheating zone due to more laser energy input. The
composition and microstructure change caused by
the heat affected zone (HAZ) will be studied in detail
in the next work.
Usually, the cracks and the porosity are the main defects
found in the sampes by DLF. Although the parameters range
chosen for the orthogonal experiment is very wide, the
cracks are found in all specimens in this experiment. The
cracks may be caused by the thermal residual stresses and
the high oxygen content in the specimens. How to fabricate
crack-free components will be further researched. The
porosity can be eliminated by choosing appropriate process
parameters.
3.3. Process feasibility and Ni alloy samples fabrication
Although the DLF process is open loop controlled
currently in this experiment, the experimental results
showed that there is an appropriate range for the specific
energy in which the Ni alloy components can be fabricated
stably layer by layer with a uniform height. If the specific
energy is too small or too high, the difference between the
contour height and the inner height of the component will
become more evident which will result in the process
failure. Since the specific energy is controlled by laser
power, scan velocity and diameter of laser spot, the increase
Fig. 6. Appearances of the orthogonal experimental specimens. (a) Specimen 1#. (b) Specimen 2#. (c) Specimen 3#. (d) Specimen 4#. (e) Specimen 5#.
(f) Specimen 6#. (g) Specimen 7#. (h) Specimen 8#. (i) Specimen 9#.
L. Peng et al. / International Journal of Machine Tools & Manufacture 45 (2005) 1288–12941292
of laser power can increase the scan velocity while
the specific energy keeps constant. This will improve
the fabrication efficiency of the DLF process. Nevertheless,
it will cause the component contours to collapse if
Fig. 7. Macroscopic optical micrographs of the orthogonal experiment specimens (O
(b) Specimen 2#. (c) Specimen 3#. (d) Specimen 4#. (e) Specimen 5#. (f) Specime
the laser power is too high. In other words, both the laser
power and scan velocity should be restricted in a certain
range. The process parameters capable for building up the
components stably with uniform height are: laser power,
bservations parallel to the slicing direction of the deposit). (a) Specimen 1#.
n 6#. (g) Specimen 7#. (h) Specimen 8#. (i) Specimen 9#.
Fig. 8. Grain size of dendrites in light zone is much small than those in dark zone. (a) Dendrites in light zone. (b) Dendrites in dark zone.
Fig. 9. Some nickel alloy samples fabricated by DLF process.
L. Peng et al. / International Journal of Machine Tools & Manufacture 45 (2005) 1288–1294 1293
600–900 W; scan velocity, 4–8 mm/s; powder feed rate,
3.0–8.0 g/min. Fig. 9 shows several samples which include
normal parts and thin-walled structure parts.
4. Conclusions
The DLF process of Ni alloy samples was studied on
hardware construction, control software development and
process parameters optimization. The following conclusions
can be drawn up:
(1)
The DLF process of Ni alloy components is feasibleunder open loop control.
(2)
The specific energy is the most important factor thatcontrols the component qualities. A suitable specific
energy can control the size of the molten pool very well
to get a good bonding strength and prevent the
component collapse. The ratio of cladding depth to
cladding height can be used for the criterion to estimate
whether the specific energy is suitable or not.
(3)
There is an appropriate range for the specific energythat can fabricate the Ni alloy components stably
layer by layer with a uniform height. If the specific
energy is too small or too high, the difference
between the contour height and the inner height of
the component will become more evident and even
make the process fail.
(4)
A banding structure can be observed in the microstruc-ture of Ni alloy components. The gain size in the light
L. Peng et al. / International Journal of Machine Tools & Manufacture 45 (2005) 1288–12941294
zones of the bandings is much smaller than those in the
dark zones.
(5)
The cracks and gas porosity are main defects found inthe Ni alloy components fabricated by DLF process.
References
[1] D.T. Pham, R.S. Gault, A comparison of rapid prototyping
technologies, International Journal of Machine Tools and Manufac-
ture 38 (10–11) (1998) 1257–1287.
[2] P.A. Kobryn, E.H. Moore, S.L. Semiatin, The effect of laser power and
traverse speed onmicrostructure, porosity, and build height in laser-
deposited Ti–6Al–4V, Scripta Materialia 43 (4) (2000) 299–305.
[3] G.K. Lewis, E. Schlienger, Practical considerations and capabilities for
laser assisted direct metal deposition, Materials and Design 21 (4)
(2000) 417–423.
[4] J. Mazumder, D. Dutta, N. Kikuchi, A. Ghosh, Closed loop direct metal
deposition: art to part, Optics and Lasers in Engineering 34 (4–6)
(2000) 397–414.
[5] C. Adams, R. Brown, P. Freeman, A. Lewis, S. Barnes, Instant tools,
Manufacturing Engineer 78 (6) (1999) 264–268.
[6] P.A. Kobryn, E.H. Moore, S.L. Semiatin, The effect of laser power and
traverse speed on microstructure porosity and built height in laser-
deposited Ti–6Al–4V, Scripta Materialia 43 (4) (2000) 299–305.
[7] X. Wu, J. Liang, J. Mei, C. Mitchell, P.S. Goodwin, W. Voice,
Microstructures of laser-deposited Ti–6Al–4V, Materials and Design
25 (2) (2004) 137–144.
[8] J. Beuth, N. Klingbeil, The role of process variables in laser-based
direct metal solid freeform fabrication, JOM 53 (9) (2001) 36–39.