Previewpdf
Previewpdf
SOLIDIFICATION PROCESSING
MATERIALS ENGINEERING
Ill
lv Foreword
Merton C. Flemings
Toyota Professor of Materials Processing
Department of Materials Science and Engineering
Massachusetts Institute of Technology
Cambridge, Massachusetts
Preface
v
vi Preface
Kuang-0 (Oscar) Yu
Contents
1 Introduction
Kuang-0 (Oscar) Yu
3 Stress Analysis 55
Umesh Chandra and Alauddin Ahmed
4 Defects Formation 95
Vijay Suri and Kuang-0 (Oscar) Yu
vii
viii Contents
Index 695
Contributors
*Current affiliation: Center for Product Design and Innovation, ITESM Campus Monterrey,
Monterrey, Mexico
xi
xil Contributors
I. WHY MODELING?
The production of almost all metallic components involves melting and solidi-
fication processes. When a molten metal is poured into a mold to make a
product with a specific shape, the process is called casting. However, in pro-
cesses such as water, gas, vacuum, and centrifugal atomization, the molten
metal is first disintegrated into small molten droplets which then solidify as
powder.
Casting and solidification processing involves many physical phenomena
such as fluid flow, heat transfer, electromagnetic force, thermal stress, defect
formation, and microstructure evolution. The quality of the final product
depends on the mechanisms of defect formation and microstructure evolution,
which are controlled by heat transfer, fluid flow, and thermal stress. How to
control processing parameters, such as metal superheat, pouring/casting speed,
and mold preheat temperature/cooling condition, to provide proper solidifica-
tion conditions and satisfactory quality castings, has been the subject of inten-
sive investigations.
In production environments, process engineers typically use the trial-and-
error method based on empirical relationships between processing parameters
and the quality of the resultant castings. This method usually leads to long
process development times and high production costs. Process modeling, on
the other hand, enables process engineers to make virtual castings using com-
puter techniques. As a result, the effects of processing parameters on the qual-
ity of the resultant castings can be evaluated without incurring the cost of
actually making castings. The processing parameters can easily be modified
until a set of processing parameters that will result in castings with satisfactory
1
2 Yu
quality is found. By applying casting process modeling, the time and cost of
developing new, and enhancing existing, processes can be significantly reduced.
As mentioned earlier, casting and solidification processing involve many
complex physical phenomena. Developing a comprehensive model to represent
all these phenomena is a very challenging task. From an engineering point of
view, it is not always possible or even necessary to have a comprehensive model
which simulates all the involved physical phenomena. In general, each casting
and solidification processing process has its own unique characteristics which
have a dominant effect on the quality of the resultant product. Thus, develop-
ing a model that provides an effective way to simulate these unique character-
istics is often not only technically sufficient, but also cost effective. In this
introduction chapter, a general description of the various metal manufacturing
processes and their applications is first presented. The modeling approach used
to effectively simulate the unique characteristics of each process is then dis-
cussed.
A. Cast Products
Many different casting processes are used to make shaped components; the
following sections briefly describe some of the most common.
1. Sand Casting
Sand casting is the most widely used shape casting process. It uses bonded sand
as the mold material and can produce castings that weigh from only a few
grams to more than a hundred tons. The sand casting process is applicable to a
wide range of metals including aluminum, steel, cast iron, etc. Sand cast pro-
ducts are used by almost all industries, from the high volume, cost sensitive
automobile industry to the high unit cost and top quality aerospace industry.
Sand casting is always performed in air atmosphere, with the sand mold at
room temperature. As a result, sand casting usually results in a relatively rough
surface product; it also has only a limited capability to make thin wall compo-
nents.
2. Investment Casting
The investment casting process uses ceramic molds and can be carried out in
vacuum as well as in air. The ceramic molds may be preheated to very high
temperatures (e.g., up to 1550°C for nickel-base superalloys), allowing for the
producing of thin wall castings. Because of the high ceramic mold preheat
temperatures, radiation heat loss from the mold surfaces strongly affects soli-
dification conditions. The use of the vacuum environment enables the invest-
ment casting of superalloys and titanium alloys, which have a chemical
composition otherwise difficult to control in air. On the other hand, aluminum
alloys, steels, and cobalt alloys are typically cast in air. Investment castings are
mostly used for aerospace and medical implant applications, which tend to
have a relative low production volume but high unit cost. Recently, investment
cast golf club heads have become an important nonaerospace application for
titanium alloys.
investment casting, but at a cost close to that of sand casting. Lost foam
castings (aluminum alloys and cast iron) are mostly used in the automobile
industry.
6. Semi-Solid Metalworking
Semi-solid metalworking (SSM) also relies on a metallic die/mold; it bears some
similarities to the die casting process. The two most important features of the
SSM process are its reliance on input materials with a unique fine grain micro-
structure, and an operating temperature between the melt liquidus and solidus
temperatures, i.e., in the mushy region. The unique fine grain microstructure of
the input material is largely maintained in the final casting, resulting in
Introduction 5
B. Wrought Products
The majority of metal components are made by making wrought products. For
a long time, the input material for wrought processing was made by the con-
ventional ingot casting process. Since the 1960s, continuous and semicontin-
uous casting processes have been gradually introduced into production. Now,
with very few exceptions, most of input materials for wrought products are cast
by either continuous or semicontinuous casting processes.
2. Continuous Casting
The advantages of continuous casting in primary metals production have been
recognized for more than a century. The dramatic growth of this technology,
however, has only been realized since the 1960s. The principal advantages of
the continuous casting process are high productivity, high material yield, good
product quality, and low energy consumption. The primary purpose of con-
tinuous casting is to bypass conventional ingot casting and to cast a form that
is directly rollable on finishing mills. The cross-sectional shapes of continuously
cast blooms/billets/slabs can be round, square, or rectangular. The principle of
6 Yu
turbine disk materials, although some tool steels, high strength aluminum
alloys, and titanium alloys are also amenable to this process.
3. Spray Forming
Although the advanced PM process can produce satisfactory products, its
processing steps are complex and its production cost is high. Spray forming
has the potential to make products with mechanical properties that are equiva-
lent to those of advanced PM products, but at lower cost. The principle of
spray forming is to use a mandrel or drum to catch molten metal droplets,
produced by gas atomization, before they are completely solidified. The metal
droplets hit the surface of the mandrel or drum, are flattened, and accumulate
layer by layer to form billets or hollow cylindrical tubes/preforms. The billets
can be used as input material for close die forging to make jet engine turbine
disks. The tubes can be used in the as-sprayed condition whereas hollow
cylindrical preforms can be ring-rolled to form engine frame components.
Although superalloy components are the primary applications for spray form-
ing, other high performance alloy components have also been produced by this
process.
tion. It is obvious that one who wants to perform casting process modeling
needs a good understanding of the physical meaning as well as the mathema-
tical representation of all these phenomena. In addition, one needs some back-
ground in the numerical analysis of differential equations and computer
programming. However, it should be emphasized that, from an application
point of view, process engineers should concentrate their efforts on under-
standing the problem, establishing an appropriate model to represent that
particular problem, making sense out of the model prediction, and developing
a strategy to solve the problem. Thus, process engineers should first have a very
good understanding of the production process they are working on. They need
just enough background in mathematical equations, numerical analysis, and
computer programming techniques to allow them to effectively perform their
own tasks. It is not necessary for process engineers to have a deep technical
background in differential equation solving and computer code writing in
order to perform casting process modeling.
ingots. In addition, critical defects are also quite different for shape and ingot
castings. For example, porosity (macroshrinkage and microporosity) is the
most important defect in shape castings, but macrosegregation, which impacts
ingot chemistry uniformity and formability in subsequent forging and rolling
operations, is of primary concern for most ingots. Finally, shape castings are
cast one by one, and always in a transient condition. As a result, true three-
dimensional transient models are required. On the other hand, ingots are cast
either in truly steady state conditions (continuous casting) or in quasi-steady
state conditions (semicontinuous casting processes). In most cases, two-dimen-
sional steady state models are adequate.
Thus, a transient model is needed to simulate the hot top procedure. However,
in the middle portion of the ingot, where a quasi-steady state condition is
reached, the shapes of the liquid metal pool and the mushy zone are relatively
constant. Thus, a steady state model can be used to predict the liquid pool and
mushy zone profiles, as well as their impact on the macrosegregation pattern in
the resultant ingot.
In the continuous casting and DC casting processes, the molten metal
superheat and ingot casting speed are not related and can be specified inde-
pendently. On the other hand, in the VAR and ESR processes, electrode melt-
ing and ingot casting rates are coupled and usually cannot be controlled
independently. This situation leads to narrow processing windows to produce
ingots with a desirable structure. In addition, the strong electromagnetic field
in VAR, ESR, and PAM processes has important effects on fluid flow behavior
and macrosegregation formation tendency in these ingots. Because VAR and
ESR processes are primarily used to melt high performance and high segrega-
tion tendency alloys, such as superalloys, titanium alloys, and tool steels,
macrosegregation has a major effect on the quality of the resultant ingots.
Thus, developing a model which can provide an accurate way to evaluate
the ingot macrosegregation formation tendency has very practical benefits
for the VAR and ESR processes.
IV. SUMMARY
Stress Analysis 91
32. B Dorri, U Chandra. Determination of thermal contact
resistance using inverse heat conduction procedure.
Proceedings of the Seventh International Conference on
Numerical Methods in Thermal Problems, RW Lewis, JW Chin,
GM Homsy (eds.), Swansea UK: Pineridge Press, 1991, pp.
213-223.
184 Stefanescu
186 Stefanescu
262 Cheng
494 Tims
496 Tims
498 Tlms
536 Thomas
538 Thomas
690 Liu
694 Llu