Canale L.C.F., Mesquita R.A., Totten G.E. Failure Analysis of Heat Treated Steel Components

Подождите немного. Документ загружается.

The ﬁnal analysis concluded that the tem-

perature of the trimming operation (immediately

following forging) was excessive and triggered

the initiation of a small crack. During further

trials in the shop, the time between forge and

trim was increased. This made the workpiece

have a higher strength because of the lower

temperature, and no cracking occurred even

though there was evidence of material banding.

Other changes were made, such as decreasing

the clearance between the punch and die from

0.127 to 0.076 mm (0.005 to 0.003 in.), which

made the trimming operation cleaner and more

robust.

Case Study 5: Upset Forging. One of the

most important operations in the forging process

is the upsetting of a billet of material with ratios

of axial length to diameter of less than 3 to 1 and,

preferably in some cases, 2.5 to 1 using ﬂat dies.

The material displacements are primarily radial,

extending out from the billet center and forming

the outside diameter. Material radial displace-

ments come from the decreasing axial length of

the starting billet. Material displacements during

upsetting have been well analyzed in the past by

numerous investigators who have identiﬁed the

steel alloy; axial, radial, and tangential strain;

strain rate; temperature of the workpiece and

die; along with the frictional effect at the

workpiece-die interface. During an extended

working stroke, there is a period of time where

the entire reduction operation is considered as

one of nonsteady state. During this operation,

there is one plasticity law that states that the sum

of the principal strain at any time is equal to 0.

This has proven to be a signiﬁcant benchmark to

Edge characteristic Type 1 Type 2 Type 3 Type 4 Type 5

Fracture angle 14–16



8–11



7–11



6–11



...

Rollover(a) 10–20% t 8–10% t 6–8% t 4–7% t 2–5% t

Burnish(a) 10–20% t(b) 15–25% t 25–40% t 35–55% t(c) 50–70% t(d)

Fracture 70–80% t 60–75% t 50–60% t 35–50% t(e) 25–45% t(f)

Burr Large, tensile plus

part distortion

Normal,

tensile only

Normal,

tensile only

Medium, tensile

plus compressive(g)

Large, tensile

plus compressive(g)

(a) Rollover plus burnish approximately equals punch penetration before fracture. (b) Burnish on edge of slug or blank may be small and irregular or even absent. (c) With

spotty secondary shear. (d) In two separate portions, alternating with fracture. (e) With rough surface. (f) In two separate portions, alternating with burnish. (g) Amount of

compressive burr depends on die sharpness.

Fig. 11 Piercing of low-carbon steels. Source: Ref 6

Fig. 12 Micrographs of crack formation at ﬂash edge

Failure in Steel Forging / 143

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 143

evaluate the effects of the lubrication at the die-

material interface.

Many forged circular components and also

circular components with off-center port bosses

begin as a billet of material of a given diameter

and are then forged between ﬂat dies to a larger

diameter and a reduced axial dimension. Sub-

sequent forging operations on the forged initial

preform create cylindrical walls along with

various conﬁgurations, depending on the ﬁnal

forging design.

Thus, an initial ﬂattening (or sometimes

referred to as “pancaking”) between ﬂat dies is a

common initial forge operation for a number of

steel components, for example, missile cases

and bowl shapes. A typical steel upset forged

between ﬂat dies made on a screw press is shown

in Fig. 13. However, during the upsetting

operation, cracking on the barrel of the upset

piece often occurs because of higher strain levels

than the material can sustain and, in particular,

higher effective strain levels when each of the

principal strains—axial, tangential, and radial—

are considered.

The ﬁx for a problem of this nature is to cal-

culate the strain ﬁeld in advance when plans are

made to create the upset piece. A numerical or

computer simulation of the process can provide

a “go” or “no-go” on the planned process. In the

past, there have been cases reported of increased

upset temperature within the workpiece, and

the solution to that problem is a reduction in the

strain ﬁeld. Most of the cases fall into the

situation of an unanticipated strain-rate effect,

requiring the rate of forging to be reduced.

Table 5 (with Fig. 14) shows some selected

results obtained from an MSC superforge

simulation of forging a ﬂattened disc (pancake)

between two ﬂat dies (Fig. 15). The analyses are

based on plasticity laws. The damage variable is

deﬁned as the ratio of the total cavity area over

the total area found in a representative volume

element. Therefore, the damage variable is a

dimensionless quantity between 0 and 1, where

D=0 describes the undamaged representative

volume element, and D=1 is the failure due to

rupture.

A critical value of damage for multiaxial

states of stresses may be deﬁned as a quantity

that describes the occurrences of measurable

cracks in the material. When the critical value of

damage reaches a certain magnitude, one can

Fig. 13

Steel upset forged between ﬂat dies made on a screw

press

Table 5 Analysis and simulation of upset forging a disc between ﬂat dies

Analytical analysis Numerical simulation (superforge)

Starting billet size: 101.6 mm (4 in.) diameter and 101.6 mm (4 in.) length

Ending ﬂattened disc size: 198.2 mm (7.8 in.) diameter and 26.67 mm (1.06 in.) length (axial thickness)

Strain: Equivalent plastic strain:

Axial?ln [1.05/4.00]=1.337 1.389

Radial+tangential?2 (ln[7.8/4.0])=2 (0.668)=+ 1.336 1.389

Analysis: axial strain (1.337)+radial strain (+ 0.668)+tangential strain (+0.668) ~0

Plasticity law: The algebraic sum of the principal strains equals 0.

Z stress average (disc center-to-edge readings): determined by maximum force

divided by disc (after working stroke) plan form area (PVA)

1.894 · 10

Pa (27,469 psi)

1.575 · 10

Pa (22,843 psi)

1.240 · 10

Pa (17,984 psi)

Z stress=Forging force /PVA=1.081 · 10

/47.78=1.56 · 10

Pa (22,624 psi) (Disc thickness center-to-barrel-edge readings)

Ratio of the average stress exerted during the ﬂattening operation to the yield

strength is 19,809 psi/16,215 psi=1.16, which is in the range of Z stresses

reported.

1.2554 · 10

Pa (18,207 psi)

1.211 · 10

Pa (17,563 psi)

1.225 · 10

Pa (17,766 psi)

Average of all Z stresses taken is 20,305 psi.

This ratio agrees closely with the graph of the average stress/yield strength when

plotting the dimensionless parameter of coefﬁcient of friction disc · diameter/part

thickness when forging a ﬂat disc (Fig. 14)

(Disc center-to-barrel-edge readings)

1.162 · 10

Pa (16,853 psi)

8.247 · 10

Pa (11,961 psi)

4.466 · 10

Pa (6,477 psi)

Yield strength: 1.118 · 10

Pa (16,215 psi)

144 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 144

deduce that the material is irreversibly damaged.

A typical result plot is shown in Fig. 16.

Case Study 6: Avoidance of Flow Thr ough,

Lap, and Crack. During the forging of an “H”-

shaped cylinder of a ﬂat web with projecting ribs

and ﬂanges, material displacements are required

to turn 90



in the direction from the web within

the die cavity to ﬁll the external/internal ribs and

ﬂanges. There have been numerous cases in the

past where insufﬁcient thickness of the material

front moving horizontally caused the front to

contact the die wall and upset on itself, enabling

the ﬁlling of the outer ﬂange.

Following the ﬁlling of an outside ﬂange, as

shown in Fig. 17, material displacements con-

tinue to move outward toward the ﬂash opening

and underneath the ﬁlled ﬂange. This combina-

tion of material displacements causes a ﬂow

through at the base of the ﬂange.

Another common case of underﬁll coupled

with lap formation in many forging designs

occurs at the intersection of outer ﬂanges and

cross ribs, where material displacements are

primarily 3-D. Additional material is required to

ﬁll the top of the ﬂange and rib intersection

because of the volume required. That additional

volume is provided by a preform shape with

increased ﬁllet radii or taper at the base of

the web.

The term lap describes a defect that forms

whenever material folds over itself during the

forging of a new shape, using a previously

designed preform as the input to follow on the

set of dies. Laps occur when both vertical and

Fig. 14 Pressure multiplication factors for forging of thin panels. Source: Ref 7

Fig. 15

MSC superforge simulation. Disc upset forged

between ﬂat dies, showing (a) start position and

(b) end position after 74.93 mm (2.95 in.) stroke

Failure in Steel Forging / 145

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 145

horizontal sections intersect. When this occurs,

it is an indication that a preliminary shape is

required as input to the next die in the forg-

ing sequence to provide material to ﬁll inter-

secting elements. Also, these types of problems

are analyzed by a technique referred to as

streamlining. The series of shapes are tracked

backward from the ﬁnish shape to blocker and

preblocker shapes to the cogged, rolled, or upset

piece.

A series of “part-way” downs were forged

using aluminum alloy 7075 on a hydraulic press

at the temperature of 399



C (750



F) in a set of

preform (blocker) dies that had been designed to

show the material displacement ﬁeld when

attempting to ﬁll a ﬂange from a web surface.

The samples were approximately 101.6 mm

(4.0 in.) in diameter and 203.3 mm (8.0 in.)

long to minimize the end effect. The strain rate

during the trials was selected so that the dis-

placement ﬁeld could also be applied to other

alloy systems, such as carbon steels, so that the

same level of forging pressure would be used

with similar results. The results are shown in

Fig. 18.

In the ﬁfth and sixth “part-way” down, there is

evidence that the material front is moving away

from the entrance to the ﬂange, and die contact is

lost until an additional stroke is applied. Then,

the outer die design adjacent to the ﬂash gutter

creates sufﬁcient back pressure to displace

material into the ﬂange. Thus, a shape is made

that will be suitable as an input to the ﬁnish dies,

Fig. 16 Typical state of strain in hot upset forging of steel showing fracture criteria in MSC superforge simulation

Fig. 17

Typical forging defect caused by excessive natural

ﬂow through the forging of a rib (ﬂange)-and-web

part. Flow through is the tendency of a metal to ﬂow naturally past

the rib (ﬂange) opening.

146 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 146

meeting the web thickness tolerances and ﬁlling

the ﬂange.

The process is to reverse-integrate from the

ﬁnish design to determine a more generalized

shape with precise volume distribution along the

three principal axes. Preforms in the series must

satisfy subsequent shapes of the ﬁnish product

design and local volume requirements, provid-

ing die materials for intersecting product fea-

tures and other geometric attributes—round

Fig. 18 Creation of a streamlined preform serving as an input to a ﬁnish die rib (ﬂange) and web design, avoiding ﬂow through

Failure in Steel Forging / 147

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 147

(simple and compound), elliptical, and tapered

shapes must all be accommodated.

Admissible criteria and speciﬁc tasks for

preform designs include:

 Speciﬁc alloy characteristics

 Microstructure requirements of a ﬁnished

product in terms of percent reduction at each

forging operation

 Match areas and volumes along principal

axes and location of centroids

 Perform reverse-integration, streamlining

the ﬁnish shape to more generalized fea-

tures—lower ribs and rails (ﬂanges) coupled

with increased web thickness and connect-

ing radii

 Calculate principal strains when comparing

ﬁnish cross sections to preform cross sec-

tions

 Generate an overall preform shape to obtain

uniform deformation in the ﬁnish die

 Examine the nature of material displace-

ments over the die contour for unsupported

material fronts as the working stroke pro-

gresses

 Examine preform locations in the ﬁnish (or

subsequent) die and initial die contacts

 Examine unsupported webs at die contact to

prevent buckling

 Make short plots of the displacement ﬁeld in

terms of the material contacting ﬁxed die

boundaries and the change of shape of the

material front being generated as the defor-

mation progresses

 Follow the continuous trace of the dis-

placement ﬁeld in terms of the material

contacting ﬁxed die boundaries and the

change of the shape of the materials front

being generated as the deformation pro-

gresses

 Determine the amount of energy expended

for each preform evaluated and then the

entire shape sequence

REFERENCES

1. G.E. Dieter, H.A. Kuhn, and S.L. Semiatin,

Ed., Handbook of Workability and Process

Design, ASM International, 2003

2. Metalworking: Bulk Forming, Vol 14A, ASM

Handbook, ASM International, 2005

3. T. Altan, G. Ngaile, and G. Shen, Cold and

Hot Forging: Fundamentals and Applica-

tions, ASM International, 2005

4. J.E. Johnson, Ed., Forging Industry Hand-

book, Forging Industry Association, Cleve-

land, OH, 1966

5. H. Gegel, G. Huang, and S. Manna, “Pre-

cision Forging—Quality—Productivity—

Equipment—A Technical Article,” UES

Software Inc., Dayton, OH

6. Piercing of Low-Carbon Steel, Metalwork-

ing: Sheet Forming, Vol 14B, ASM Hand-

book, ASM International, 2006, p 159

7. M.D. Stone, The Design and Construction

of Large Forging an d Extrusion Presses

for Light Metals, United Engineering and

Foundry, Pittsburg, PA

SELECTED REFERENCES

 J. Burke and V. Weiss, Advances in Defor-

mation Processing, Sagamore Army Mate-

rials Research, Vol 21, Army Materials and

Mechanics Research, Massachusetts and

Syracuse University, NY

 H. Chandler, Metallurgy for the Non

Metallurgist, ASM International, 1998

 G.E. Dieter, Mechanical Metallurgy, 3rd

ed., McGraw-Hill Book Co., 1986

 D.D. Fuller, Theory and Practice of Lubri-

cation for Engineers, John Wiley and Sons,

Inc., Chapman and Hall, Ltd., 1956

 C.G. Johnson, Forging Practice, American

Technical Society Publisher, Chicago, IL,

1954

 S. Kalpakjian, Manufacturing Processes of

Engineering Materials, 3rd ed., Addison-

Wesley, 1977

 A. Kannappan, “Wear in Forging Dies—A

Technical Paper,” Swedish Institute of

Product Engineering Research, Goteborg,

Sweden, 1969

 C. Lipson, “Wear—Consideration in

Design, Residual Stresses and Contact

Stresses—A Technical Paper,” University of

Michigan, Ann Arbor, MI

 F.A. McClintock and A.S. Argon, Ed., An

Introduction to the Mechanical Behavior of

Metals, School of Engineering, Massachu-

setts Institute of Technology, Cambridge,

MA, 1962

 “Research Report: Work at IIT Research

Institute,” Committee of Hot Rolled and

Cold Finished Bar Products, American Iron

and Steel Institute, New York

 A.M. Sabroff, F.W. Boulger, and H.J. Hen-

ning, Forging Materials and Practic es,

148 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 148

Battelle Memorial Institute, Columbus, OH,

Reinhold Book Company, 1968

 J.A. Schey, Introduction to Manufacturing

Processes, McGraw-Hill Book Company,

New York, 1977

 J.A. Schey, Ed., Metal Deformation Process,

Marcel Dekker Inc., 1970

 J.A. Schey and P.W. Wallace, Research

Report: Metal Flow in Closed Die Forging

of Steel, Part 2: Speed and Lubrication

Effects, American Iron and Steel Institute,

New York, 1966

 J.A. Schey, P.W. Wallace, and F.A. Shunk,

Research Report: Metal Flow in Closed Die

Forging of Steel, Part 1: Fundamental

Study, American Iron and Steel Institute,

New York, 1966

 T.M. Silva and T.A. Dear, “Wear in Drop

Forging Dies—A Technical Paper,” De-

partment of Mechanical Engineering, Uni-

versity of Birmingham, 1969

 J.W. Spretnak, “Technical Notes on Forg-

ing,” Forging Industry Education and

Research Foundation, Cleveland, OH,

1976

 “Technical Notes: Mechanical and Physical

Properties of Ferrous Forging,” Committee

of Hot Rolled and Cold Finished Bar Pro-

ducts, American Iron and Steel Institute,

New York

Failure in Steel Forging / 149

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_133-149.pdf/Chap_04/ 18/8/2008 3:06PM Plate # 0 pg 149

Failures from the Casting Process

Omar Maluf and Luciana Sgarbi Rossino, Instituto de Materiais

Tecnolo´ gicos do Brasil Ltda.

Camilo Bento Carletti, Celso Roberto Ribeiro, Clever Ricardo

Chinaglia, and Jose´ Eduardo May, Universidade Federal de

Sa˜o Carlos

THE HEAT TREATMENT of a steel com-

ponent is often the last step or near the end of

a somewhat complex manufacturing process.

Finished products require attention to each step

of the long operation chain from raw material to

ﬁnished product.

Early in-service failures of components after

heat treatment may result from improper plan-

ning, lack of required equipment, nonqualiﬁed

personnel, not enough time to execute the

expected operations, or even a combination of

some or all of these deﬁciencies (Ref 1). How-

ever, most of the early failures that happen

during the heat treatment process are the result

of features generated in previous manufacturing

stages. A component lifetime basically depends

on the following factors:

 Global component project

 Materials selection

 Material quality

 Processing methods, such as casting and

machining operations prior to the heat

treatment

 Heat treatment

 Final ﬁnishing operations

 Mechanical solicitation of the component

and the service environment

However, it is not a simple task to identify

which of these items is responsible for the early

failure of a component during heat treatment or

in service. The failure analyst uses these seven

items as a guide in a failure analysis. This

chapter deals speciﬁcally with improper casting

projects and those features that originated in

the casting process itself, including porosity

(generated by the presence of gas as well as by

shrinkage pores), decarburization, cold joint,

and inclusions. These features may not be

called defects because, according to ASTM

E1316-2005, “Standard Terminology for Non-

destructive Testing,” components have defects

only when they fail to meet their speciﬁcation

requirements. If a component has a large amount

of porosity, for example, it is not a defect unless

(1) an inspection porosity is speciﬁed, (2) its

amount exceeds the required acceptance criter-

ion, or (3) the component fails because of this

porosity. This chapter describes cast steel fea-

tures that may be identiﬁed or attributed to

component failure during heat treatment or

subsequent processing or service. As such, these

casting features are referred to as defects in this

chapter.

Failures due to Improper Cast Design

The engineer’s designing job may face

inevitable weak points due to some inherent

component characteristic in use. The engineer

should, however, try to overcome these incon-

veniences by looking for alternative solutions

and ﬁnding the middle ground between the

component functionality and the manufacturing

difﬁculties. The project aspects that should be

avoided at all costs in the cast component pro-

duction are (Ref 2–7)

 Sharp edges, sharp corners, and nonround

edges

 Abrupt section changes

 Holes, especially when located near the

external wall of the component

 Sections with cross connections

 Unfavorable length/width relationship

Rounding the corners, as shown in Fig. 1,

should always be performed in order to avoid the

stress concentration that can originate from

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_151-176.pdf/Chap_05/ 18/8/2008 3:09PM Plate # 0 pg 151

Failure Analysis of Heat Treated Steel Components

L.C.F. Canale, R.A. Mesquita, and G.E. Totten, editors, p 151-176

DOI: 10.1361/faht2008p151

www.asminternational.org

cracks formed during casting solidiﬁcation

in the mold, during heat treatment, especially

quenching and tempering, or even during heat-

ing for austenitization (Ref 2).

The risks will be reduced if steel with

increased temperability is chosen, which re-

quires a less severe quenching medium, such as

oil or air. Other strategies in the design of cast

tooling are to avoid creating components with

right angles or to machine the corners to make

them round. Another option is to quench and

temper the component and then remove the

exceeding material to give the component sharp

corners if they are required for its function. This

last strategy requires a steel with good temper-

ability. Otherwise, when the exceeding material

is removed, that region in the component will

present a surface with lower hardness and less

resistance to abrasion than the previous one.

The quenching of components with abrupt

section variations in a liquid environment

always represents a serious problem due to the

associated stress concentrations, even if the

transitions are made using the apparently correct

concordance radius resource (R ef 4, 7). In this

case, the solution is to create the component

in different parts, treat them separately, and

assemble later on. However, if the component

must be made as one unit due to a functional

imposition, the solution is to choose an air-

quenchable steel that presents a lower crack

probability.

The existence of holes raises a problem

mainly in high-carbon steels and/or alloying

elements. The abrupt section variation and other

speciﬁc aspects of the holes (Fig. 2) must be

considered. The accumulation of quenching

liquid in the interior of blind holes leads to an

improper heat loss of the internal walls, thus

lowering the hardness. In open-ended holes, the

heat removal may not be as effect ive as in the

rest of the component, which is more exposed to

the quenching liquid (Ref 5). Therefore, when

the chosen steel is quenched in a liquid medium,

the components containing holes must be

quenched in specially designed devices so that

they receive strong gushes of liquid in the

interior (of the holes), or they must be arranged

so that all of the set is subject to a strong stirring.

In cases where the holes do not need to be

hardened, they can be made of low-alloy low-

carbon steel components already inserted during

molding for the casting, for example, in tool

steel components. Another possibility during

heat treatment is to ﬁll the holes from casting

with any material that can totally inhibit contact

with the quenching ﬂuid that would result in

hardening of this region. Regard ing stress con-

centration, it is preferable to have the existence

of a completely quenched hole than the presence

of a mixed structure (hardened and soft). Both

methods, particularly the ﬁrst, make the

achievement and ﬁnishing of holes and threads

easier.

Cotter holes, especially the rectangular sec-

tion ones, are places with high stress con-

centration (Ref 2). Therefore, whenever

possible, they should be eliminated or sub-

stituted by channel sections, whose locking

efﬁciency is equivalent, but the stress-

concentration factor is three to four times lower.

Another geometry that should be avoided during

project design of the cast tooling component is

the cross type, such as the furnace grid and

heating equipment shown in Fig. 3(a). It causes

serious crack problems in the cross area during

the solidiﬁcation process while still inside the

sand mold or duri ng heat treatment (Ref 3). The

solution is to use node dislocation, as illustrated

in Fig. 3(b).

Lengthy components with very thin sections

or small diameters show serious bending pro-

blems during heat treatments, mainly quench-

ing, even when the steel is favorable to less

severe environments. The problems start from

the moment the compo nent is put into the fur-

nace for austenitization. If it is secured only in

two extremities, there is the risk of deﬂection

Fig. 1

Sharp edge elimination. (a) Sharp corners create high

strain concentration. (b) Exaggerated relief causes a

shrinkage cavity. (c) Ideal relief

Fig. 2

Types of holes. (a) Blind hole with a parallel bottom.

(b) Blind hole with a steeple bottom. (c) Passing hole—

the most economic

152 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_151-176.pdf/Chap_05/ 18/8/2008 3:10PM Plate # 0 pg 152

due to its own weight. If it is supported on the

furnace hearth, its heating will not be homo-

geneous, making the component subject to

bending and/or to the appearance of soft spots

(Ref 6). The correct heating method, in this case,

consists of hanging the component by one of its

extremities and using a furnace that allow s the

austenitization of the component hung in the

vertical position.

Also to be avoided is the man ufacturing of

too thin, lengthy components. They should be

split into components whose length-hei ght ratio

is more favorable (Ref 3). For example, high-

alloy, high-carbon, steel sheets used in the

guillotine have been replaced by shorter sheets

that, after quenching and temper ing are assem-

bled in a chassis to present a continuous edge.

An advantage of this solution is that only the

damaged component of the edge can be replaced

when there is an in-service failure of one of the

sections, leading to an easier and cheaper

operation.

Very big components with a circular section

larger than 25 cm (10 in.) in diameter, or rec-

tangular with equivalent mass, also present

problems during quenching. When carried out in

a liquid medium, the surface reaches the starting

martensitic transformation temper ature long

before the central region, with the generation of

stress that may cause internal cracks as a con-

sequence. It is recommended that big, bulky

components be replaced by sets of smaller

pieces, despite the adjustment problems that

result from this operation.

Figures 4 to 6 illustrate failures that happened

during the heat treatment operation due to poor

design considerations. Figure 4 is a crack that

occurred during the normalization heat treat-

ment of an AISI 1045 steel cast hull. A prior

thinning of the wall (due to deﬁciency of the

tooling or core alignment) promoted the crack-

ing during normalization. Figure 5 is a crack that

happened in the normalization heat treatment,

caused by stress buildup in the sharp edge

region. In Figure 6, poor design of an edge led to

cracking after quenching.

Fig. 3

Grid crossings. (a) Crossed node region of crack formation. (b) With dislocated nodes, the occurrence of cracks is less likely if

the distance between nodes, d, is larger than 2r+ e, where e is the thickness, and r is the curvature radius.

Fig. 4

Crack resulting from the normalization heat treatment

of an AISI 1045 steel cast hull caused by thinning of the

wall due to deﬁciency of the tooling or the core alignment

Failures from the Casting Process / 153

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_151-176.pdf/Chap_05/ 18/8/2008 3:10PM Plate # 0 pg 153