SCHOOL OF MECHANICAL AND AEROSPACE ENGINEERING

2013-2014


MA2700 CONTINUAL ASSESMENT CASE STUDY REPORT










Name of Student: Tran Anh Thong
Matriculation Number: U1320125A
Case study subject: The manufacturing process of jet turbine blades


2

Table of Contents
1) Case Study Background ............................................................................................................. 3
2) Investment Casing Process ........................................................................................................ 4
2.1)Background ........................................................................................................................... 4
2.2)Investment Casting Process .................................................................................................. 5
2.3)Consideration of Investment Casting for Turbine Blades .................................................... 7
3) Single Crystal Forming Process .................................................................................................. 8
3.1) Purpose of making single crystal turbine blade .................................................................. 8
3.2) Single Crystal Production Process ....................................................................................... 9
4) Thermal Barrier Coating ........................................................................................................... 12
4.1) Introduction ....................................................................................................................... 12
4.2) Atmospheric Plasma Spray (APS) ....................................................................................... 12
4.3) Electron-beam physical vapor deposition (EBPVD) of TBC ceramic ................................. 13
5) Defects Testing ......................................................................................................................... 15
6) Conclusion ................................................................................................................................ 17
Preferences ................................................................................................................................... 17


3

1) Case Study Background:
Since the jet engine was invented in the 1930s, it has played a momentous part in the
development of aviation. The early designs of jet engines tend to have low life expectancy
because at that time, the materials used for turbo blades lacked the durability when working in
extreme condition inside the jet engine. Poor material limits the operational temperature and
thus limiting thrust as well as top speed of the engine. Nowadays, such problems are better
handled as more suitable materials are available and the manufacturing techniques used for
turbine blades are developed and refined. This report will attempt to reveal more about the
manufacturing methods of turbine blades and the working principals behind them.

The picture above displays turbine blades made by three different method: (a) conventionally
cast; (b) directionally solidified, with columnar grains, as can be seen from the vertical streaks;
and (c) single crystal.
In this report, we will discuss conventional techniques in producing jet turbine blades such as
investment casting, single crystal growing process, thermal barrier coating and a few testing
method to look for defects in newly made blades.

4

2) Investment casting:
2.1) Background:
There are many methods of casting alloys, nevertheless, turbine blades are usually made by
investment casting method or otherwise known as lost-wax casting method. Investment casting
is one of the oldest casting methods there are, dating around 5000 years ago. It is preferred in
making turbine blades since the method can create casting with intricate details.
In order to increase the thermodynamics efficiency of the jet turbine engine, operating
temperature of the engine must be sufficiently high, especially in the high pressure turbine
section. A limiting factor is the operational temperature of turbine blades in that section, which
rotate at extremely high speed and endure high stresses and temperature. Thus, any increase in
the working temperature can adversely affect the service life of those blades. As a result,
engineers have designed the turbine blade such that cooling is done internally by adding holes
and passages inside the blade itself. The network of passages allows cool air to be pumped
through the center of the blade and escape through holes at the surface, covering the blade in
a film of cool air. Machining the internal cooling passages from a single block of alloy would
prove to be impossible. Therefore, investment casting is used to create the complete detailed
structure of the turbine blade.


5

2.2) Investment Casting Process:
The process has the following steps. First, a
wax model is created. This is done by injecting
molten wax into a master mould to solidify
around ceramic cores which will form the
cooling passages in the completed turbine
blade. Pinning wire is pressed through the
wax to butt against the ceramic core in the
model. Next, several models can connect with
replicas of runners and risers to form cluster
should several blades be casted at a time.
Then the investment shell is produced by
dipping the model in ceramic slurries of
alumina, silica, zirconium and other binding agents. Next, the model is covered with larger
particles of the same materials as the slurries. This operation is repeated a few time until the
investment shell reach the required thickness. After that, the shell is heated to melt away the
wax inside while leaving the ceramic core intact. Then, it is heated again to strengthen the
structure of the shell.

Precise positioning of the ceramic core and maintaining such position throughout the mould
preparation and casting process is required. However, the core have high length-to-diameter
ratio and therefore can only be poorly supported in the mould. To solve this problem, pinning
wires are employed to hold the ceramic core in place for the whole process. Typically, seven to
ten pins are used for a 2-inch thick turbine blade.
6


After the mould is preheated and degassed, it is ready to be poured with molten superalloy.
The pouring process happen in a vacuum at around 1500 degree Celsius. The cooling process is
then controlled meticulously to create the precise microstructures. The performance and
service life of the blade is largely determined by the grains structure. Blades used in the hottest
parts of the engine are usually made up of directionally solidified grain or single crystal for best
7

performance.
After the turbine blade solidifies, the investment shell is broken off. Then, the ceramic core is
taken out by chemical solution. The turbine blade then goes through some minor machining to
acquire its final shape.

2.3) Consideration of Investment Casting for Turbine Blades:
Investment casting can help to create the complex structure of the turbine blade with some
modification and techniques. It also reduces the need of machining after casting, which is more
difficult since turbine blade materials are usually superalloy. Moreover, investment casting has
high level of accuracy and can be done automated in some part. Therefore, despite its
complicated process and high cost of production, investment casting is suitable for
manufacturing of the complex structure of turbine blades.



8

3) Single Crystal Production:
3.1) Purpose of making single crystal turbine blade:
Investment casting, which was discuss in the previous section, was first only able to produce
equiaxed turbine blades. After the molten metal is poured into the mold, the heat is cut off
completely and the final product is an equiaxed turbine blade. Such crystal structure,
nevertheless, cant perform well at the extreme temperature found in the jet engine.
Therefore, as turbine blades have to operate at high temperature and stress, it is necessary that
material can endure the heat without being deformed by creep is used. Furthermore, at high
temperature, grain boundaries of polycrystalline metal weakens faster than the internal of the
crystal itself since the boundaries are disordered. As such, the sliding between boundaries
further increases creep effect in the material. Also, there exists the diffusion of atoms along the
grain boundaries (Coble creep) which causes changes in grain shapes.

Therefore, turbine blades are made using metal with directionally solidified crystal structure to
limit the weakness at grain boundaries. Turbine blade which has columnar crystals with grain
boundaries parallel to its axis can limit or eliminate shear stress caused by tensile load while
rotating.

By making single crystal turbine blade, grains
boundaries are completely removed, which make
the material have better qualities compared to
equiaxed and columnar grains structure. With no
9

grain boundary, turbine blades are not weakened by grain boundary slipping and thus perform
better at high temperature. Moreover, since most impurities and disorders concentrate at
grains boundaries, single crystal turbine blades are more resistant to high temperature
corrosion.

3.2) Single Crystal Production Process:
From the 1960s, methods of making single crystal has been put to use in order to improve
critical properties of turbine blades: creep strength, oxidation resistance, thermal fatigue
strength and corrosion resistance. The technique to produce single crystal was developed
based on the existing method and experience of producing directionally solidified crystal
structure. The casting process for columnar-grain and single crystals are shown in the following
figure.

10

In both case, molten super alloy is poured into a mold placed on a water cooled copper chill. On
the chill surface, some grains will form and grow in one uniformed direction parallel to the
temperature gradient created by radiation heating and cooling. In the case of directional
solidified, the columnar grains will form the turbine blade itself. To prevent this from happen in
order to create a single crystal, Brigdeman-Stockbarger technique is employed. The mold
containing molten alloy is lowered in a furnace so that a single grain would form at the tip of
the cast and grow into one single crystal blade as it comes to the lower temperature region of
the furnace.
Since the original Brigdeman-Stockbarger technique is slow and
expensive, modification has been made to the method to
eliminate the need to lower the mold through the furnace, which
can reduce both time and cost of production. A helical grain
selector is being used. Only one grain from the columnar grain
block can emerge from the selector and grow into the whole
turbine.
For this process to be possible, the temperature must be
controlled precisely to achieve the suitable temperature gradient.
This is done by controlling the chill, the molten metal and mold
temperature. An investment casting of the turbine blade shape is
made as shown in the picture. In the process, the mold
temperature is kept high to prevent the nucleation of other
unwanted grains during the pour.
As solidification is allowed to happened, some grains are allowed
to form in the starter block. Of those grains, only few can grow
into the selector while other are physically blocked. After one or two turn of the helix, only one
grain survive to form the single crystal turbine blade. The molten alloy is let to grow into the
surviving grain. The helix diameter is typically about 3 to 5 mm depends on the mold and the
11

required heat flow. Also, a helix of circular cross section is preferred since other cross sections
may cause unwanted nucleation at sharp edges.
Furthermore, during the solidification process, heat
flow has to be constantly manage to keep the liquidus
and solidus isotherm close to the horizontal and close
to the baffle to prevent any unwanted grain creation.
This can be done through baffle design, mold wall
thickness and heat drain rate.





12

4) Coating:
4.1) Introduction:
Another improvement that can be made to the turbine blade is adding the thermal barrier
coatings (TBCs). TBCs are complex defected thick film made of zirconia-based refractory
ceramics oxides. The technology allows turbine blades to work at higher temperature and thus
can increase engines power as well as the blades service life.
Thanks to the development of advanced deposition technologies, TBCs production and
application is now more widespread. TBCs materials nature with extremely high melting point
(for instance, yttria partially stabilized zirconia has melting temperature about 3000K) requires
ultrahigh temperature processing capability. Thus, thermal plasma and electron beam sources
have become the method of choice to create TBCs. Both of those methods can also further
reduce the thermal conductivities of the materials used due to structure defects in the TBC
layer.
4.2) Atmospheric Plasma Spray (APS):
APS uses molten droplets deposition to deliver the ceramic particles to the surface of turbine
blades. In the process, tens of micrometers-sized particles of metal and ceramic in powder form
into an arc of plasma jet and get projected onto the turbine blade surface. The plasma is
formed from a high current arc with inert gas and powder particles. Before the spraying, the
substrate surface is usually roughen by an abrasive process such as grit blasting. The molten
particles settle on the destined surface quickly solidify and create a wall of splats of typically
100-150 microns in diameter and a few micrometers in thickness. Each splat is attached to the
substrate surface because of physical interlocking of the splat and irregularities on the surface.
Internal passages and cooling holes cannot be coated by plasma spray.
Since the solidification of each particle happens quickly and independently, the structure
13

formed as a result consists of splats interfaces and unfilled regions which creates porosity in the
coating. Therefore, the TBC has low thermal conductivity and non-linear stress-strain behavior.
As we can see, the processing parameters decide much of the characteristic of the deposited
microstructure. Such parameters include the characteristic of the spray stream (particles
trajectory, kinetic and thermal state and degree of melting); location and state of substrate
(roughness, temperature, position, geometry and movement); and speed of torch and part. To
sum up, properties of both the powder and torch play significant roles in determining the
characteristic of the TBC (porosity, interface, grain size, cracking, phase evolution, stability).
4.3) Electron-beam physical vapor deposition (EBPVD) of TBC ceramic:
EBPVD process utilizes the evaporation of molten ceramic with high vapor pressure when
overheated. A high energy electron beam scan over the ceramic material, melting and
evaporating the material in a vacuum chamber. A preheated substrate, in this case, a turbine
blade is placed in the vapor cloud so that the ceramic vapor would deposit on the blade at a
rate of microns per minute. For a defined stoichiometry of zirconia in the ceramic to be achieve,
oxygen is bled in to the chamber to make up for the deficit caused by dissociation. In the
process, the blade needs to be rotated in order to be coated evenly on all surfaces. Because of
the formation of the coating from vapor ceramic with combined actions of surface diffusion,
shadowing and crystal growth selection, the TBC formed will have columnar microstructure
with high capability of enduring stress. For the process to be continuous, ceramic material is
bottom fed into the crucible.

14

The coating process is the determining factor to many properties of the TBC layer such as intra-
columnar porosity characteristics (size, distribution, concentration, morphology). Therefore,
those characteristics can be manipulated by adjusting factors like deposition temperature,
rotational speed, chamber pressure, vapor incident pattern, condensation rate and partial
shadowing. Thus, the microstructure pattern can be adjust in the coating process to meet the
turbine blade property requirements.
To sum up, the EBPVD coating technique has the advantage of being able to produce coating
with columnar structure with strain tolerance and pseudo-plasticity. Compared to Plasma Spray
TBCs, TBC using EBPVD has better corrosion resistance and smoother surface, which creates
better aerodynamics properties for the turbine blades. Furthermore, the cooling channels will
stay open throughout the coating process. Nevertheless, the method still has certain
disadvantages in terms of limited material choice (due to vapor pressure problem), higher cost
and thermal conductivity and low utility rate of raw material.



15

5) Testing:
Before the turbine blade can be used, it must go through tests for defects in the material and
the blade structure itself.
First, the blade dimensions have to be checked to make sure it fits the design required
specifications. This task is carried out by a Coordinate-measuring machine. The machine will
take measurements of certain points on surface of the blade and check if it fits the design. Any
blade with high measurement deviation from the design will be rejected.

After that, the blade will be tested with dye penetrant to find for any cracks on the surface of
the blade.

16

Also, the interior of the blade can be inspected using fiber optic method. Before being put to
use, the blade will be put through an engine running test which they will be put into a real
working condition in a jet engine test.



17

6) Conclusion:
With advances in manufacturing methods, turbine blades are now made with higher quality
than ever before. By using Investment casting, we are able to create intricate turbine blade
designs with internal cooling passages and holes. Moreover, the turbine material processing
techniques also enable us to from turbine blades of single crystal structure which have better
capability to work in extremely high temperature of the jet engine. Thermal Barrier Coating
(TBC) is another improvement to the quality of the turbine blade. The use of TBCs can help
protect the turbine blade from the extreme heat of the engine and corrosion.


References:
Materials Science and Engineering Properties by Charles Gilmore
Roger Reed, University of Birmingham 2007
M. Gel& D. N. Duhl and A. F. Giamei, Pratt & Whitney, 1980
D. C. Power, Johnson Matthey Noble Metal, Royston, 1995
Sanjay Sampath, Uwe Schulz , Maria Ophelia Jarligo , and Seiji Kuroda, 2012
Manufacturing Processes for Engineering Materials, 5th ed, 2008
Henry L. Bernstein, Gas Turbine Materials Associates, San Antonio, Texas