Hardenability and Heat Treatment

.-.
-+t~wRiO;$! t I till' "i19
2~i
?H"ij'td n 'e: :alii it'll

~~_fI
!1JI. fllt!l~
ria
CHAPTER 24
Hardenability and Heat Treatment

Page
Introduction
24-2
Carbon and Low Alloy Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Transformation During Heating . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Transformations During Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Austenite Transformation Diagrams . . . . . . . . .
. . . . . . . . . . . .
Isothermal Transformation Diagrams ... . . . . . . . . . . . . . . . . . . . .
Continuous Cooling Diagrams . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
Types of Heat Treatment . . . . . . . . . . .
. . . . .
. . . . . . . . . . .
24-2
24-2
24-3
24-4
24.4
24.5
24-6
High Strength Low Alloy Steels
. . . . . . . . . . . . . . . . . . . . . . . . . . . 24-11
Austenitic Manganese Steels . . . . . . . . . . . . . . . . . . . .
. . . . .. 24-11
Martensitic Stainless Steels . . . . . . . . . . . . . . . . . . . . .
. . . . 24-12
Austenitic Stainless Steels . . . . . . . . . . . . . . . . .
. . . . 24-13
Precipitation Hardening Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 24-14
f:
[I.. .
rt
~c
l
t
,~~.
~
i.
I
Quench Hardening of Carbon Steel . . . . . . . . . . . . . . . . . . . . . . .

Hardenability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jominy Procedure . . . . . . . . . . . . . . . . . .
Ideal Critical Diameter . . . . . . . . . . . . . . . . . . . . . . . ..
Calculated End Quench Curves. . . . . . . . . . . . . . . . . . .
Application of Ideal Critical Diameter . . . . . . . . . . . . . . . . . . . .
Cooling Rate Equivalence . . . . . . . . . . . . . .. . . . . . . . .
Specifying Hardenability . . . . . . . . .
. . . . .. . . . . . . . .
Hardenability Bands . . . . . . . . . .
. . . . .
. . . . . . . .
Quenching to Produce Martensite. . . . . . . . . . . . .
. . . . . . . . . . . . . . .
Quenchants . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
Quenching Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Measuring Quench Severity . . . . . . . . . . . . . . . . . . . . . . .
Cooling Curves . . . . . . . . . . . . . . . . . . . . . . . . .
Quench Factors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hardenability of Cast 4130 Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
TTP Curves for Cast 4130 Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Microstructure and Properties . . . . . . . . . . . . . . . . . . . . .
I
Residual Stress, Distortion, and Cracking . . . . . . . . . . . . . . . . . . . . . . . . . . .
24-15
24-15
24-15
24-20
24-20
24-22
24-23
24-24
24-25
24-25
24-26
24-31
24-32
24-33
24-34
24-34
24-35
24-37
24-38
References .'. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24-42
STG:.Gt-
C/l-ST/AlG-'S
)ljIff#hfJOtU~, &
72z GJ;UIl'O.v.
/99~
,~
,(""
24-2 "Part 58: Processing Properties

r---~--.----r--~----r---~---r---r~nE
1100
2000
1900
\~
UPPER
TRANSFORMATION
LOWER
TRANSFORMATION
lIJ
a::
..........
..... ........
S>
WATER
QUENCH
,,
,,
1000
(y)
cr:
:::l
900 I
<t
cr:
1600
l.IJ
IL
~
l.IJ
l.IJ
IL
~
l.IJ
l.IJ
~ 1700
:::l
G
~
cr:
'\NORMALIZE
(AIR COOL)
:E
IBOO
....
a::
lIJ
CI..
u..
0
UJ
...... <FULL ANNEAL
\
\
:::>
l-
AUSTENITE
.....
1500
+ Fe 3 C
BOO I
1400
......
.. IK
1300fP
700
a
Fe
+ PEARLITE
0.2
0.4
0.6
PEARLITE
O.B
1.0
1.2
+ Fe3 C
1.4
1.6
CARBON-%
TIME--..
Fig. 24-1 A generalized sketch of the steps in the heat treating of

steel castings
Introduction
The overwhelming majority of carbon, low alloy, and high alloy
steel castings produced today are heat treated prior to being put into
service. Castings are heat treated to enhance particular properties
such as hardness, toughness, corrosion resistance, etc. and to im
prove uniformity ofproperties throughout the casting. The particular
type of heat treatment depends on both the alloy type and the
intended service conditions. It should be noted that heat treatment
procedures for cast steels are similar to those employed for wrought
steels.
In general, heat treatment is an operation involving heating to an
elevated temperature followed by controlled cooling for the purpose
of obtaining particular microstructures and combinations of proper
ties. The utility of steel lies largely in the relative ease with which
properties can be altered by controlling the rate at which parts are
heated and then cooled. The changes in properties occur principally
because of the variety of microstructural constituents obtained with
various cooling rates.
The essential elements of any heat treatment are the heating cycle,
holding temperature and time, and the cooling cycle. A heating cycle
and three different types of cooling cycles are schematically illus
trated in Figure 24-1. The length of time that a casting is held at a
particular temperature and the rate at which it is then cooled are
important factors. The holding time at a definite temperature should
be long enough to complete the desired microstructural transforma
tions.
Carbon and Low Alloy Steels

Transformations During Heating
Most carbon and low alloy cast steels contain less than 0.77%
carbon and as a consequence, the as-cast microstructure is a two
phase mixture consisting of ferrite (alpha iron) and pearlite. Ferrite
is a bOdy-centered cubic phase, and pearlite is a two-phase eutectoid
Fig.242 Simplified iron-carbon diagram. Shaded band is

annealing or normalizing temperature range.
microstructure offerrite and cementite (Fe)C) existing as alternating
lamella. Most heat treating begins by heating the steel into the
austenite phase region where the room-temperature microstructure
is transformed to the austenitic face-centered cubic structure. If such
a steel is heated to a moderate temperature below 1340 OF (727C),
internal stresses resulting from the casting process may be relieved
without substantially altering the as-cast microstructure or physical
properties.
A portion of the iron-carbon phase diagram is illustrated in Figure
24-2. The iron-carbon diagram illustrates the phases present as a
function of carbon concentration versus temperature. At carbon
concentrations below 0.77% and temperatures below the eutectoid
value of about 1330 OF (721C), the stable phases are ferrite and
pearlite. At carbon concentrations above 0.77%, the stable phases
below 1340 OF (727 0c) are pearlite and cementite.
If a steel containing less than the eutectoid carbon concentration
(hypoeutectoid) is heated to a temperature just above the eutectoid
temperature (P-S line in Figure 24-2), pearlite decomposes to pro
duce face-centered cubic austenite (gamma iron) resulting in a
mixture of ferrite and austenite phases. If heating is continued to
higher temperatures, the ferrite progressively decomposes to form
more austenite. At a sufficiently high temperature, the ferrite and
any carbides present are completely decomposed to produce a ho
mogeneous austenitic microstructure. The process of heating for the
purpose of forming a homogeneous austenite microstructure is re
ferred to as austenitizing or solution treating. Most strengthening
heat treatments begin by heating the steel fully into the gamma
region. The upper critical temperature for a steel containing less than
0.77% carbon is represented by the line G-S.
Heating a carbon steel containing more than 0.77% carbon (hy
pereutectoid) to a temperature above the S-K line in Figure 24-2
again results in a decomposition of the pearlite to form a structure
containing both austenite and cementite. At higher temperatures, the
cementite is progressively decomposed until at a sufficiently high
temperature, the structure consists of homogeneous austenite. The
upper critical temperature for a steel containing greater than 0.77%
carbon is represented by the line S-E. The lower critical temperature
for both hypoeutectoid and hypereutectoid steels is represented by
the line P-S-K.
l.
......
~""
~o
300 1
'\.
/Ms
""'"
'
t
E 200
{!!.
a.
100
0'
Fe
0.2
0.4
0.6
!,
'\
0,8
1.0
1.2
1.4
Weight per cent carbon
Fig.24-3 Influence of carbon on Ms and Mf temperatures (1)
Transformations During Cooling

If an austenitized solution treated steel containing less than 0.77%
carbon is slowly cooled from the austenite temperature region, the
first phase to precipitate from the austenite is alpha-ferrite. This
transformation results in a change in the atomic arrangement from
the face-centered cubic austenite to the body-centered cubic alpha
ferrite. Since alpha-ferrite can accomodate only a small amount of
carbon in solution, the concentration ofcarbon in austenite increases
during cooling, until at or below the eutectoid temperature of 1340
OF (727C); the remaining austenite transforms to pearlite. Slow
cooling a hypereutectoid steel containing more than 0.77% carbon
is similar to cooling a lower carbon steel except that the first
constituent to precipitate from austenite is cementite (FeJC). Ce
mentite usually develops at the austenite grain boundaries and forms
a network structure that can be observed in steel at room temperature
after the transformation is complete.
At the eutectoid composition (0.77% C), austenite decomposes
into pearlite without the formation of either primary ferrite orpri
mary cementite. Since both ferrite and pearlite formation require
some diffusion of carbon, higher cooling rates generally reduce the
thickness of the ferrite and cementite phases in pearlite. In general,
as the transformation temperature decreases, the ferrite and Fe3C
phase distribution formed becomes finer, which results in a progres
sive increase in hardness and strength.
At even highd cooling rates, the transformation of austenite to
ferrite and pearlik is delayed or may even be terminated in many
carbon and low' alloy steels. The transformation of austenite at
temperatures between 1000 OF (538C), and about 400 OF (204 C)
produces a structure composed of ferrite and cementite, but the
phases are not arranged in a lamellar form characteristic of pearlite.
Instead, the ferrite and cementite have a feathery or acicular appear
ance in which neither ferrite nor cementite can be clearly resolved
with an optical microscope. These structures are known as bainite.
When the transformation occurs in the range of about 700-1000
OF (371-538 C), bainite has a feathery appearance and is called
upper bainite. When transformation occurs at temperatures in the
range of about 400-700 OF (204-371 C), the bainite has a more
acicular appearance and is called lower bainite. The hardness of
bainite varies depending on the exact bainite structure achieved but
typically ranges from about Rockwell C 45 to Rockwell C 60. For
medium carbon steels, bainite can form during extended holding of
the austenite in the temperature range of 400-1000 OF (204-538 C)
24-3
or during continuous cooling of allloyed steels at certain cooling

rates.
Transformation of austenite at even lower temperatures, below
about 400 OF (204C), results in the formation of a body centered
tetragonal structure. This structure is referred to as martensite and is
found in fully hardened steels. Microstructurally, martensite has the
appearance of needles. Since martensite occupies a greater volume
than austenite, steel expands when martensite forms, and this results
in dimensional changes which may cause distortion. In severe situ
ations distortion may cause parts to crack during quenching.
In many cases, some austenite may remain in a severely quenched
steel after cooling. When this occurs, martensite needles are ob
served to exist in a matrix containing some austenite. This untrans
formed austenite is referred to as retained austenite. Retained
austenite is relatively soft and its presence detracts from martensitic
properties when a steel requiring full hardening is desired. Retained
austenite is also undesirable because it can subsequently transform
to martensite during machining or after a part is put into service. The
expansion occurring when retained austenite transforms to marten
site creates internal stress which may cause distortion and a loss of
tolerances in machined parts. The resulting untempered martensite
is also brittle which may lead to cracking in service. As a conse
quence, the presence of retained austenite is generally objection
able.
The transformation of austenite to martensite begins at some
temperature generally referred to as the martensite start (Ms) tem
perature and continues with decreasing temperature to some lower
temperature referred to as the martensite fmish (Mr) temperature. At
the Mr, the reaction of austenite to martensite ceases even though
some retained austenite may be present. The Ms and Mr tempera
tures are influenced by the concentrations of alloying elements
present. carbon concentration, and alloy segregation that may have
occurred during solidification. The influence of carbon content on
the Ms and Mr temperatures is illustrated in Figure 24-3 (1).
The Mr temperature in some alloy steels is below room tempera
ture. More complete transformation of austenite in such steels may
be obtained by cooling to the Mr temperature followed by reheating
to room temperature and cooling to the Mf temperature a second or
third time. With each cooling cycle a smaller percentage of retained
austenite is converted to martensite.
The Ms temperatures ofmany steels have been determined experi
mentally, and can be approximated using several empirical formu
lae. One empirical formula developed by Grange and Stuart (2) is
given as Eq 1:
MseF) = 1000- 650 * %C -70 * %Mn- 35 * %Ni -70* %Cr- 50 * %Mo
(EqI)
All elemental concentrations are expressed in wt% in this equa

tion, and this formula applies when all carbides are dissolved in the
austenite.
Another empirical formula reported by Atkins is given in Eq 2 (3);
Ms("C) =539-432 * o/oC - 30.4 * %Mn-17.7 * %Ni-12.1 * %Cr-7.5 * %Mo
(Eq2)
The Mf temperature is important when attempting to minimize

retained austenite in quenched steels. The Mf temperature generally
ranges from 300-475 OF (165 to 245 C) below the corresponding
Ms temperature. The influence of carbon concentration on the pro
portion of martensite formed at a given temperature below the Ms is
illustrated in Figure 24-4 (2).
The most common method of hardening steel involves quenching
from the austenitizing solutioning temperature in a liquid medium
that rapidly extracts heat. The rate of cooling in the part depends on
the heat removal characteristics of the cooling medium, the thermal
...
;..""\
24-4
Part 58: Processing Properties
Ms 0% martensite
0
20
40
60
l!!
80
Q)
100
i,
'I
..
.e
::i!
.c
It-
160
a.
Q)
200
I-
220
E
Q)
- ............
...
......
140
180
10% martensite
-------,------------------------
120
2!
:::>
-----__
............
.... ......
......
..........
' ,..., 99~o
.........
..... ......
240
...... .....
.......
......
260
.........
...... ......
280
....
300
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Carbon, %
lr'
characteristics of the alloy, and the section thickness of the part.

Cooling curves superimposed on "C curves" showing the beginning
and end of transformation are illustrated in Figure 24-5. The points
where the cooling curves cross the beginning and ending curves
represent the beginning and end of transformations for the indicated
cooling rates, and the hardness values given are approximate for a
steel containing 0.55% carbon. Curve A shows a slow cooling
condition such as in a furnace (annealing), with transformation
taking place at relatively high temperatures thus producing thick
lamellar pearlite. The cooling condition illustrated by curve B rep
resents conditions that might be encountered during air cooling
(normalized) or fan cooling and would result in the formation of
finer pearlite and higher hardness than observed in condition A. The
cooling characteristics illustrated by curve C, represent an initial
formation of fine pearlite, but there is insufficient time in the upper
temperature region to allow complete transformation to pearlite. As
a consequence, some of the remainder of austenite transforms to
bainite or martensite during continued cooling. Under this condi
tion, the transformation occurs to produce two or more microconsti
tuents. The formation of two or more types of microconstituents
such as pearlite and bainite, or bainite and martensite, is referred to
as a split transformation and is generally avoided.
At a sufficiently high rate of cooling, such as indicated by curve
D, transformation does not occur until the Ms temperature is reached
and martensite formation begins. The cooling rate indicated by
curve E is sufficient to produce a fully hardened steel. The cooling
rate that will accomplish this is called the critical cooling rate. A
cooling rate slower than the critical value will allow formation of
some pearlite, and therefore, will not produce full hardening.
Austenite Transformation Diagrams

Transformation diagrams were developed to illustrate the time
dependence of microstructural transformations. There are two
widely used diagrams for presenting transformation data. The fIrst
is the time-temperature-transformation (TIT) diagram, sometimes
referred to as the isothermal transformation diagram (IT diagram).
Fig. 24-4 Relationship between percent carbon in steel and

proportion of martensite formed at a given temperature. (2)
Isothermal Transformation Diagrams. TIT
800r-----~-----r----~------~----,_----~--~
Eutectold temperature
500
~
or
400
0.
E
,!
300
MarfenSlte
01
0.1
Martensite
and pearlite
Fine
pearlite
10
100
10!
Coarse
pearlite
10
105
T.me. in seconds
Fig. 24-5 Typical continuous cooling transformation curves
5.10!
diagrams
are developed by solution treating small samples of steel at the
appropriate temperatures and then rapidly cooling from the austeni
tizing temperature to a temperature between the Ae3 and the marten
site start temperature (Ms), holding for a predetermined period of
time, completing the quench, and analyzing the resulting micro
structure for the transformation products.
Time-temperature-transformation (TIT) diagrams for carbon
steels containing about 0.20% and 0.30% carbon are illustrated in
Figures 24-6 and 24-7 (4). A TIT diagram for a chrome molybde
num steel containing approximately 0.30% carbon is illustrated in
Figure 24-8. While carbon shifts the start of transformation curve to
longer periods of time at a given temperature, as may be seen by
comparing Figures 24-6 and 24-7, the unalloyed carbon steels are
very difficult to quench to fully hard microstructures because trans
formation ofaustenite to ferrite and pearlite begins at such short time
intervals over the temperature range of 800-1200 oF. The very fast
transformation of austenite to ferrite and pearlite makes it virtually
impossible to achieve fully martensitic structures except in very thin
sections.
The addition of chromium and molybdenum however, retards the
transformation and allows hardening to occur with a less severe
quench.
The TIT curve for 4130 steel, illustrated in Figure 24-8, is usually
interpreted to mean that the steel must be cooled past 1000 Op and to
the start of the martensitic transformation (Ms) in less than about 1.5
seconds to produce a fully martensitic structure and to the Ms in
about 10 seconds to produce a 50% martensitic structure. Slower
cooling rates may result in less than 50% hardening. Several books
containing TIT diagrams have been published (4,5).
I,
~.
~;
i'
24-5
~'
'C
800
I I I
I I I
I I I
I I
800
700
700
~,
.:i'
f
f
0::
~~
It
~,
f~
600
600
g:;
500
e:[
W 400
Go
I 80
500
e:[
ffi
400
fl~
I- 300
fI
200
zoo
100
100
:::;;
I- 300
....
f
~'
0.5 I 2
0,5 I 2
5 10
5 10
TIME - SECONDS
TIME - SECONDS
Flg.24-6 Carbon Steels: 1021; Chemical composition. AISI and

UNS: Nominal 0.18 to 023 C, 0.60 to 0.90 Mn, 0.040 P max, 0.050
S max. Legend. A =Austenite; F Ferrite; C =Carbide; M =
Martensite; B = Bainite; P == Pearlite
Continuous Cooling Diagrams. The second method of

studying and reporting transformations involves examining samples
of steel that are continuously cooled at specified rates. The resulting
diagrams are referred to as Continuous Cooling Transformation
(CCT) diagrams. CCT curves provide data on the temperature re
gime of each phase formation. the amount of transformation occur
ring at a given cooling rate with time, and the cooling rate necessary
to produce martensite.
A time-transformation-temperature (1TI) diagram with a super
imposed continuous cooling transformation (CCT) diagram is illus
trated in Figure 24-9 (6). Cooling curves associated with selected
positions on a Jominy bar are also superimposed.
CCT diagrams have some characteristics not seen in TTT dia
grams, such as a depressed Ms temperature at slow cooling rates,
and the prevalence of bainite in the microstructure. The CCT dia
grams are particularly helpful when devising heat treatments for
high hardenability-thick section steels (6-10).
The continuous-cooling diagram for a 1038 carbon steel contain
ing approximateIr. 0.38% carbon is illustrated in Figure 24-10 (3)
which indicates ~he difficulty of achieving full hardening in an
unalloyed low carbon steel. This diagram suggests that cooling rates
in excess of 10,000 C (18,000 oF) per minute are required in order
to achieve full hardening.
The continuous cooling transformation diagram for a chromium
molybdenum (AISI 4130) steel, illustrated in Figure 24-11, shows
how the required cooling rates for hardening are reduced with these
alloy additions (3). Achieving a 50% martensitic microstructure
requires a cooling rate of about 120 "F per second at 1300 OF, and a
fully martensitic microstructure requires a cooling rate above 300 OF
per second at 1300 oF. The discrepancies in the required cooling
rates needed to produce 50% martensite in 4130 steel as determined
from the TTT diagram (Figure 24-8) and from the CCT diagram
(Figure 24-11) is principally a result of the different methodologies
involved and possibly to differences in the composition and grain
size of the steels used to develop the transformation curves. Con
tinuous cooling diagrams are generally thought to indicate the prob
able quenched transformation products more accurately than iso-
Flg.24-7 Carbon Steels: 1029; Chemical composition. AISI and

UNS: Nominal 0.25 to 0.31 C, 0.60 to 0.90 Mn, 0.040 P max, 0.050
S max. Legend. Same as Figure 24-6
I lll.jilr
r i IllIdi
t t [11m
1 f 111,/1
i I I !tllii
.e
t i t ,lttH'
800
-'*""FI I
I I I I I I I
Ii
700
600
W
g:;
500,
le:[
ffi
400
Go
:E
W
I- 3001
200l 4 oO
lOOt- 200
a
0.5 I 2
106
5 10
TIME - SECONDS
Fig. 24-8 Chromium-Molybdenum Steel 4130; Chemical

composition. AISI and UNS: Nominal 0.28 to 0.33 C, 0040 to 0.60
Mn, 0.035 P max, 0040 S max, 0.15 to 0.30 Si, 0.80 to 1.10 Cr,
0.15 to 0.25 Mo
thermal transformation diagrams, but there is little or no definitive

data on the subject.
These and other published diagrams present approximate cooling
rates needed during quenching various alloys to obtain desired
microstructures (4,11). The actual rates required in parts made of
different alloys depend to some extent on the specific compositions
of the steels being quenched.
"
::.
.,.'
24-6
Austenitizing temperature
=860 C
Fig. 24-9 Experimentally

determined CT diagram
(continuous lines) for steel with
German designation 42 CrMo 4
(0.38% C, 0.99% Cr, and 0.16%
Mo). IT diagram is also shown
(dashed lines). (3)
800
700
Pearlite formation
600
500
:::l
!!
@)
Ql
c.
400
LMs \
300
0= Hardness, HRC,
Martensite
start
after cooling to
room temperature
2001
100 I
-CTdiagram
___ IT diagram
- Cooling curves
0'
1
101
30
60
I
I!
4 68
Hours
60,
105
!
1624
1 day
~
::c:
gf 40
~
'E
as
::c:
\n
30r
)
20
""'
!
10
20
30
40
Distance from quenched end
Types of Heat Treatments

It is apparent from the foregoing discussion that a range of
structures and properties can be achieved in cast steel by controlling
the composition, solution treating temperature, and the conditions
used to cool the steel from the austenitizing temperature back to
room temperature. The purpose of this section is to describe some
heat treatment cycles that may be used to achieve particular micro
structures and properties in cast products.
Stress Relief. Stress relief refers to the process of heating steel

to a temperature in the range of 800-1200 OF (427-649 C) for the
purpose of relieving internal stresses without substantially altering
the microstructure or properties of the part. Some stress relief can
occur at relatively low temperatures of 400-500 OF (205-260 0c) as
illustrated in Figure 24-12, but about 90% of the residual stress can
be removed at higher temperatures of 1000 OF (540 ot) or higher.
Sometimes a stress relief treatment is required after operations such
as welding, induction hardening, or grinding are performed after the
primary heat treatment. The maximum temperature for stress relief
is generally limited to about 50 OF (30 "C) below the tempering
temperature used after quenching. Higher stress relief temperatures
cause softening.
Solution Treating (Austenitizing). Solution treating or

austenitizing is an elevated-temperature process intended to achieve
a uniform carbon distribution in the austenite. The object of the heat
treatment is to dissolve second phases such as carbides and produce
a homogenous structure with uniformly distributed carbon.
Homogenization. Homogenization treatments are usually the
same as solution treatments and are intended to achieve a uniform
distribution ofcarbon in austenite. However, in some circumstances,
homogenization refers to efforts to achieve a homogenous distribu
tion of all elements in the austenite. Homogenization temperatures
may range from 2000-2300 OF (1093-1260 0c) or higher in efforts
to achieve uniform distributions of nickel, chromium, molybdenum,
and carbon in the matrix. However, the diffusion coefficients for
these elements are so low that the time required is long, often a
hundred hours or more. The results do not generally warrant the
costs associated with extended homogenizing treatments.
Annealing. The purpose of annealing is to soften the metal,
increase its ductility, relieve internal stresses caused by previous
treatments, and enhance machinability. In full annealing, a steel is
heated to at least 100 OF (38C) above the upper critical tempera
ture, as illustrated in Figure 24-2, and held for sufficient time to
"Part 58: Processing Properties
-24-8
achieve a uniform distribution of carbon in austenite. This is fol

lowed by slow cooling, usually in a furnace. When castings have
cooled 25 OF (14 0c) below the lower critical temperature, the
transformation ofaustenite is usually complete. The transformations
associated with annealing are not section size dependent unless the
sections are sufficiently large that macro segregation occurred dur
ing solidification. The castings can then be removed from the fur-
TEMPERING TEMPERATURE - C
1000,
100
200
300
400
500
SOO
nace and air cooled. Furnace cooling to lower temperatures wastes

furnace time and requires extra heat to reheat the furnace for the next
load. The strength of annealed castings is low, and the ductility is
high. Some low allay steels are annealed for machinability and
given a final heat treatment after rough machining.
There are variations in annealing treatments for specialized pur
poses, for example to achieve spheroidized pearlite structures.
Spheroidized pearlite has excellent machinability, but the thermal
cycle is long and expensive and should not he specified unless the
last increment in ductility and machinability is actually required.
700
,240
ARMCO IRON WITH 0.025 % C.

AvERAGE BRlNEll HARDNESS
lONGITUDINAL STRESS
...-- -- ... TANGENTIAL STRESS
~-~ RADIAL STRESS
8>- - - TANGENTIAL AND RADIAL
STRESS
0-----0
800
90, 0600
a.
::E
200
1900,
AUSTENITE
(r)
120 m
:x::
u.
IBOO'
(f)
a::
~
"' 0
lJ.J
a::
~90 ~600
a::
~8001' I S~EEL IWITH I 0.3 !/o C
300C
i '"
260
a..
0::
!;:i
ffi
lJ..I
1700
::J
900
lJ..I
a..
~60
lJ..I
1500
I
BOO
220
~_ _ _ _ _ _ _ _ _ _~~w
::E30
180
200
1300
0'
,
200
T'
400
600
800
.=to
1000
TEMPERING TEMPERATURE -
A~~
Fe
lJ..I
U
0.2
0.4
0.6
MARTEM PERING
Irl~
I
MjRTENSI,E
Ae l
a:
::J
M .m
S
~ Mf
CR.T.
ffi
~ Ael
lJ.J
"OTHERMAL
_Q~EN:~~G_ANc:. ~~P~~N~ __
0::
TEMPERED
+ Fe3 C
O.B
1.0
1.2
1.4
1.6
Fig. 24-13 Simplified iron-carbon diagram. Shaded band

represents hardening temperature range of carbon steels.
;t
lJ.J
700
PEARLITE
CARBON
jARTEN\ITE
+ PEARLITE
Fig.24-12 The effect of tempering on residual stresses, in

quenched cylinders (11)
CONV ENTI ONAl

QUENCHING AND TEMPERING
IK
, 140
1200
0
!JJ
AUSTEMPERING
!7,-
!;:i
0::
1600
::E
::E
a::
1000 U
S20t-_____~ _40 i

l80
.;.400
lJ..I
~ 0
,1100
160
60
::30
2000
@
( ,-",<::,,~Ae,
Fig. 24-14 Types of quench

cycles. A. Conventional quench
and temper; B. Martempering;
C. Isothermal quenching and
tempering; D, Austempering

Normalizing. Nonnalizing refers to solution treating a steel at a
temperature of at least 100 of (38C) above the upper critical
temperature followed by cooling in air. The nonnalizing tempera
ture depends on the carbon concentration as illustrated in Figure
24-2. The purpose of normalizing may be to refine the grain struc
ture, harden the steel slightly. or reduce carbon segregation that may
have occurred during solidification.
Nonnalized castings must be placed on trays so air can circulate
freely around every casting in the load as they are cooled. If air flow
is restricted, the operation will be more like annealing. Accelerated
cooling by fans or forced-air flow may produce a result more like
quenching.
The microstructure that results from normalizing is a mixture of
ferrite and pearlite, usually resulting in low residual stresses and
almost no distortion. Some castings are tempered after being nor
malized to achieve a slight additional softening and stress relief.
Tensile strengths up to 95 ksi (655 MPa) can be obtained by
normalizing low alloy steels containing above about 0.30% carbon.
Normalizing and tempering is used to meet a number of standard
casting specifications. Machinability of cast parts is usually good
because of the uniform structure obtained.
The cost of normalizing makes this heat treatment attractive. It
requires less furnace time than annealing; the cnoling cycle is less
expensive than quenching; and minimal problems are encountered
with distortion.
Quench Hardening. It is possible to accelerate cooling from
the solution treating temperature and control the transformation of
austenite to bainite and martensite for the purpose of achieving
higher strengths and hardnesses than can be obtained with annealing
or normalizing t:reatments. In hardening by quenching. the work is
austenitized at temperatures illustrated in Figure 24-13 and then
rapidly cooled to prevent the formation of ferrite and pearlite. A
schematic quench cycle is illustrated in Figure 24-14(A).
The maximum attainable hardness of a quenched steel is control
led almost exclusively by the carbon concentration and is obtained
by cooling at a rate equal to or greater than the critical cooling rate
for the alloy. Quenching media include water. brine solutions, oil,
water-polymer solutions, and in some cases inert gases.
Water and oil are the media most commonly used for quenching
steel castings. Wateris used whenever possible, but high-carbon and
deep-hardening steels require oil quenching. Some complicated
shapes also require oil quenching to minimize quench cracking.
Under almost all conditions. oil cools steel a little more slowly than
water.
Certain organic polymers can be added to water to provide a
quenching solution which resembles oil in its heat removal charac
teristics. The main advantage ofthese solutions is that they remove
heat more slowly ptan water without the fire hazard of oil. Their
greatest disadvantpge is that they require careful concentration,
temperature. and ,agitation control in order to achieve consistent
quench behavior. The quench severity provided by these baths can
vary widely with the type of polymer, polymer concentration. bath
temperature, and extent ofagitation during quenching. Proper quen
chant control is necessary and sometimes more difficult than antici
pated.
Interrupted or Slack Quenching. In some situations it is

desirable to quench steel in water or brine to produce a high surface
hardness. However, water or brine quenching may cause cracking
or excessive amounts of distortion if used to fully harden parts. If it
is not necessary to harden the part throughout the cross section, an
interrupted quench may be used. Interrupted or slack quenching
usually involves quenching in water for a predetermined time and
then transferring the part to an oil bath to complete the transforma
tion. This is not a commonly used procedure, but it is sometimes
used on steels containing 0.30% or more carbon where through
24-9
TEMPERING TEMPERATURE-oC
. ~~
I
u601
~~~
300
I
0::
:r 551-
400
500
600
en 50
fn
is0:: 45
~40
a
W35~
:r
___
--1-------- - - __ _
Z30
:::>
'! 25[
!--AS- QUENCHED HARDNESS
en
<1: 20 0
200
400
600
BOO
1000
..
TEMPERING TEMPERATURE - OF
70r!--------------------------------------,
65~~.::- -
l~ "'
~~
601- \\ \"
a:::
7 55
en
en
~
0::
<1:
:r
......
\\ \"
50
""
\,
,
\\
\,
45
40
35
30~1--------~-------L------~L-
to
SEC
10
___ L_ _ _ _~_ _L __ _ _ __J

1/2
MIN
25
HR
TIME AT TEMPERATURE
Fig.24-15 Effect of tempering temperature and time on the

softening characteristics of steel
hardening is not required. Quenching of carbon and low alloy steels

is always followed by tempering.
Tempering. Tempering is the process of heating a hardened steel

to a temperature below the lower critical temperature in order to
achieve some softening. and then cooling the steel back to room
temperature. The objective of a tempering treatment is to reduce the
hardness and relieve some of the stresses in order to obtain higher
ductility than available in as-quenched parts. Tempering slightly
modifies the structure of martensite, and this change can be used to
adjust strength, hardness. toughness. and other mechanical proper
ties to specified levels.
Carbon and low alloy cast steels are usually tempered in the range
of350 to 1300 OF (175-700 0c). The time of holding at temperature
may vary from 30 minutes to several hours. A longer tempering time
at a given temperature, or a higher temperature for a given time,
increases the extent of tempering. Martensite softens more than
pearlite at a given tempering temperature, and the steel composition
affects the rate of tempering. Some effects of temperature are illus
trated in Figure 24-15 where hardness is used to measure ~e re
sponse to tempering. In general. the presence of carbide-forming
elements causes steel to be more resistant to tempering.
-24-10 - Part 58: Processing Properties
lJ
TIME - hr.
TEMPERING
TEMPERATURE - C
2
.t:J
250
100
I
ch
....J~
ul~
~a:
a:~
mI
200 f-500
200
300
400
500
600
(f
~ 1-,
.::
4320
16
32
~--~o~------~----
~OO~~
END VALUE
OBTAINED IN
ACCELERATED
AGING
.
ctU
1200~120
:E
SO
..>r: 1501- 400
ii 1000+100
V)
V)
IJJ
a::
a:
RED. AREA
50
soo..f-so
0~
V)
100 f-300
~
40
CHARPY
IMPACT
2~ ~ELONG;TION~
10
:t:
l-
500..f-50
a::
~
!.
t AS
3
NORMALIZED
/
!
14
<
30
! I
60 90
END VALUE
OBTAINED IN
ACCELERATED
AGING
-
I'
180270575
Fig.24-17 Change in ductility with time for 1 in. (25.4 mm)

coupons aged at room temperature and at 400 OF (204C). Values
plotted are average of three tests.
c::i
:i!
50 f-200
20
400..f-40
<0
c: 200J- 20
ELONGATION
C\J
Table 24-1 Temperature Range of Martensite

Formation in Several Carbon and Alloy Cast Steels
(14)
t!)
01L--L__L--L__~~__~__L--L__L-~~ 0
1000
1200
400
600
SOO
200
50%
....J
UJ
TEMPERING
I in.(25.4 mml COUPONS AGED

AT 400F (204C)
100>
YIELD STRENGTH
!II
64
b
r
-r.N<;II r= STRENGTH
40 ~ED. OF AREA
?IL QUENCHED
1/2 In.( 12.7mml SECTION
60 I
TEMPERATURE - OF
Fig.24-16 Loss in room-temperature toughness due to temper

embrittlement, illustrated for wrought Ni-Cr-Mo steel (13)
Tempering below 1100 of (595C) may cause temper embrittle

ment in certain steels. The decrease in notch toughness caused by
temper embrittlement is illustrated in Figure 24-16 for a Ni-Cr-Mo
AISI 4320 steel tempered in the range of about 500-700 of (13).
Usually tempering in the temper embrittlement range is avoided,
and when higher tempering temperatures are used, the work may be
quenched from the tempering temperature to minimize the time in
the embrittling temperature zone.
Martempering. In the usual quench hardening process, parts are

cooled at a rate greater than the critical to form martensite. In some
parts prone to cracking or distortion, however, it is desirable to use
a practice where the parts are initially cooled faster than the critical
rate, but the cooling cycle is interrupted above the Ms temperature
until the temperature becomes uniform across the section. The parts
are then cooled to room temperature in air. This cycle is referred to
as martempering, and a martempering cycle is schematically illus
trated in Figure 24-14(B). The martemper cooling cycle favors
uniform martensite formation throughout the section and helps
minimize distortion and cracking.
In martempering, the work is usually quenched in a molten salt
bath to a temperature just above the temperature where martensite
begins to form (Ms) and held long enough for the temperature to
equalize in the heaviest sections. The work is then removed from the
salt and air cooled. Martensite forms during cooling from the Ms to
the Mf. Table 24-1 lists Ms and Mf temperatures for a number of cast
carbon and low alloy steels (14).
In modified martempering, the work is quenched into a bath at a
temperature between the Ms and Mf temperatures. When the tem
perature at the surface and center of the work has equalized, the
work is removed from the bath and either air cooled or placed in a
Steel
SAENo.
1030
1065
1090
1335
2340
3140
4130
4140
4340
4640
5140
6140
8630
9440
Martensite
start(M,)
OF
C
650
530
425
650
580
635
715
650
550
650
650
620
690
625
343
277
219
343
304
335
380
343
288
343
343
327
366
330
99%
Martensite
fonned
OF
C
560
425
315
560
560
550
650
570
480
570
570
560
630
540
293
219
157
293
293
288
343
299
249
299
299
293
332
282
Martensite
finish(Mr)
OF
C
450
300
180
450
405
430
550
440
370
480
450
450
530
405
232
149
82
232
207
221
288
227
188
249
232
232
277
207
tempering furnace. This thermal cycle is schematically illustrated in

Figure 24-14(C);
Most steels that can be oil quenched can. be martempered. The
maximum section size that can be through-hardened by martemper
ing is appreciably less, however, than the section size that can be
hardened by conventional quenching.
Austempering. The austempering process requires that a steel

be cooled from the solution treating temperature sufficiently quickly
to miss the nose of the C curve, but cooling is stopped and the
temperature held at some value below the nose of the C curve to
produce a bainitic structure. This cooling cycle eliminates the for
mation of pearlite as well as martensite. Minimizing the volume
expansion associated with martensite formation helps eliminate
cracking, but some sacrifices are made in hardness and toughness
with bainitic structures compared to fully martensitic structures. The
austempering cycle is schematically illustrated in Figure 24-14(D).
Upper or lower bainite can be formed depending on the temperature
at which the part is transformed.
24-"
35.,------------------------------------------------~
Fig. 2418 Hardness changes as a

function of temperaging time and
temperature for Nand T and Q and T
heat treatments after austenitizing at 950
C (1740 OF)
30
25
U
cr: 20
J:
iii
gj
"E
.-
o N & T (620C)
o N & T (580C)
A Q &T
Q &T
(580C)
(620C)
15
ttl
J:
10
5
0
0.01
0.09% C 1.7% Mn 0.3% Mo 0.06% V 0.03% Nb
0.1
10
100
Temperaging time, hrs
Aging for Hydrogen Removal. Hydrogen in steel can sub

stantially reduce the toughness and cause low elongation and reduc
tion of area values. If the steel contains 4 or 5 parts per million of
hydrogen, the ductility may be reduced to only 20% of that of a
hydrogen free steel.
Hydrogen in steel does diffuse reasonably rapidly. If held for a
sufficient time at room temperature, hydrogen will diffuse out of
steel and ductility will be restored.
Hydrogen removal can be accelerated by heating steel to 400 to
600 OF (2()()"315 C). This heat treatment is commonly referred to
as aging, and the required aging time is proportional to section
thickness. Generally, a thickness of I in. (25.4 mm) requires about
20 h for hydrogen removal. For heavy sections, 10 in. (254 mm) or
more, hydrogen removal by aging may become impractical due to
time requirements. The effect of aging is illustrated in Figure 24-17.
Age Hardening. The age hardening reaction most commonly

observed in carbon or low alloy steels results from the precipitation
of copper at temperatures of 800-900 Of' after the steel has been
austenitized and cooled as quickly as possible to room temperature.
The steel is then reheated to about 900 OF to cause copper precipita
tion with an associated increase in both hardness and strength.
In higher alloy materials containing substantial amounts ofnickel,
aluminum and titanium additions can be made that will promote the
formation of nickel-aluminide and nickel-titanium compounds. The
formation of these compounds is favored by first solution treating
the steel at temperatures in the range of 1800-2100 OF (980-1150
C), rapid cooling to room temperature, and then aging in the range
of 900-1100 OF (480-590 0c).
Age hardening has some advantages because the precipitation
occurs uniformly across the section and results in a minimum
amount of distortion. All age hardening steels require an initial
solution treatment to dissolve the hardening compounds and ele
ments prior to being quenched and aged. The age hardening tem
peratures are usually sufficiently low that a minimum amount of
surface discoloration or oxidation occurs during the hardening proc
ess.
High Strength Low Alloy Steels

High strength low alloy (HSLA) cast steels are similar in compo
sition and properties to the widely used wrought HSLA steels. This
family of cast materials has good strength and low-temperature
toughness, combined with excellent weldability. The carbon content
of these steels is kept low to promote toughness and weldability.
Strength is maintained by adding manganese for solid solution

strengthening and by the addition of microalloying elements vana
dium andlor niobium that form strengthening carbonitride precipi
tate during heat treatment. Desirable properties are achieved for
both quench and temper and normalize and temper heat treatments.
During the heat treatment of the HSLA cast steels, both the
conventional ferrous phase transformations and the simultaneous
precipitation of carbonitride precipitates take place. The kinetics of
each reaction must be controlled. During the initial austenitizing
heat treatment, the material is transformed to austenite and carboni
tride precipitates are solutionized. Temperatures as high as 1900 OF
may be necessary to fully solutionize niobium carbonitrides that
may be present. During quenching or air cooling from the austeni
tizing temperature, the ferrite/pearlite matrix or a martensite matrix
is developed but the precipitation of carbonitrides is prevented due
to the rapid cooling. Upon [mal tempering (aging), the matrix
hardness decreases due to the conventional tempering reactions, but
at the same time strengthening carbonitride precipitates form by an
age hardening reaction. The net result is that the [mal tempering step
does not result in the significant decreases in hardness as is observed
for conventional steels. The hardness changes observed during the
tempering (aging) of an HSLA cast steel are shown in Figure 24-18
for different heat treatments and tempering temperatures (15). Good
impact toughness is achieved for these materials when the material
is tempered to a slightly overaged condition. Special intercritical
heat treatments in particular have been observed to increase the
toughness of the HSLA cast steels without saCrificing strength (16).
Austenitic Manganese Steels

Manganese steels are brittle in the as-cast condition because they
contain a network of grain boundary carbides precipitated from
austenite during solidification and cooling. In carbon-depleted areas
along grain boundaries, some austenite may have transformed to
pearlite. The purpose of heat treating manganese steel is to produce
a uniform austenite structure supersaturated with dissolved carbon.
In this condition, the steel is wear resistant because it is very work
harden able on the wear surface but it maintains a ductile and tough
core. Austenitic manganese steels are commonly solution treated at
temperatures between 1925 and 1975 OF (1050-1075 0c) for best
toughness although higher temperatures are necessary for the higher
carbon and higher alloy grades (17).
Manganese steel castings should be held at the austenitizing
temperature long enough to ensure full carbide solution. A rule of
thumb is to hold the steel at temperature for 1 hour for each inch of
24-12
800
1
..,,"
I
I
I
I
\
700~
I
\
0
0
e
.
\
\
\
\
1
\
Austenite
:::J
500
' ... ......
...... ......
-11200
j'~
....
Austenite +
\ intergranular carbides
8.
{!!.
",
-11400
Austenite +
carbides +
fine pearlite
600
...
800
400
LL.
0
e
:::I
'@
Ql
c.
{!!.
-1600
3001
Me:
E
~
U)
200 1
10
ccc!if:
E
'E 'E 'E ~
I"' ... ..,

102
1to!
1:
-~
1()3
.c
'"
1:
i:
I"" '"
....
1Q4
.......
s::
.c:
l!ll
1()5
;Z;
1:
~
""-1400
1Q6
Time in seconds
(b)
section thickness. After complete solution treatment, the casting is

quenched in highly agitated cold water until the part reaches the
water temperature.
TIT curves for 2 manganese steel compositions are illustrated in
Figure 24-19(A) and (B) (18). Manganese steels are reasonably
harden able if properly solution treated but they need to he rapidly
quenched, primarily to minimize grain boundary carbide precipita
tion which can reduce the toughness.
in corrosion resistance because of the associated chromium deple

tion along the grain boundaries.
Whether the microstructure of a stainless steel is ferritic, marten
sitic, or austenitic is determined principally by the chromium,
nickel, and carbon concentrations. Generally the chromium steels
are either martensitic or ferritic depending on the chromium and
carbon concentrations.
The chromium steels are predominately martensitic when the
following inequality is met:
Martensitic Stainless Steels
% Cr-07x%C) < 12.5%
Martensitic and higher alloy stainless steels are solution treated

just as carbon, low alloy and manganese steels to achieve a uniform
carbon distribution in the austenite. Producing a "stainless" steel
with the ability to resist corrosion depends to a great extent on
Obtaining a uniform distribution of chrOlnium in the matrix. Chro
mium concentrations above about 12 % result in a passive surface
that maximizes corrosion resistance. However, precipitation of
chromium carbides at grain boundaries can lead to a significant loss
The amount of ferrite in chromium steels increases as the chro

mium content increases or as the carbon content decreases.
It is usually desirable to solution treat most stainless castings to
dissolve chromium carbides or sigma phase formed during cooling
in the mold. The solution of these phases helps produce a uniform
chromium distribution that favors corrosion resistance. It is simi
larly desirable to solution treat welded components to eliminate
grain boundary carbides that may have formed either during weld
(Eq3)
~.
t
t
r
1800'
Fig. 2420 Isothermal transformation diagram for

various 12% chromium steels
,1000
------ ---
--
1600
24-13
0.1% C-12% Cr
0.1% C-12% Cr -1/2% Me
0.1% C-12% Cr 1% Ni
0.1 % C-12% Cr -1% Ni 1/2% Me
0.1% C-12% Cr- 3% NI
800
1400
1200
Showing start of
transformation only
lI
o
1000
800
:;
{!!.
600
()
o
ef
ae
!
E
Q)
400
800 I--
M1 temperatures
..--------------------------------__ ...... __ .. _-_._-------_ .... ---------_ .. _-----
............ n .................................................. u ................. n ..... n .......... u ....................
400
200
200
1 min
5 10
1 hr
1000
100
10,000 100,000
Time, s
Fig. 24-21 Effect of ferrite on harden ability

1500 !
Approx.Ac,
1400
lI-
A+F
Austenite
1300
.ef
Steel A
78
1200
1100
83
5 10
102
103
-,......
1()4
ex::
Q)
Q)
'E
Q,
{!!.
.0
::f
105
1500
1400
Approx. Ac,
Steel B
--~:s~~~---<2:---------~~;'~c::------+
Ferrite
5% AF
1300
1200
A+F+C
1100
5 10
102
' ...
...... _
F+C
--------j74
78
50%~_~
"'_..l
103
1()4
:!2
..... .... ,
-184
105
Time, sec.
ing or during cooling of a welded part to room temperature. The

object is to dissolve second phases and produce a homogeneous
structure. A quench or other accelerated cooling is usually required
to retain carbon in solution and prevent precipitation of carbides.
The hardenability of the 12% chrome martensitic stainless steels
is quite high as illustrated by the TIT curve shown in Figure 24-20
(19). The presence of excessive amounts of ferrite in the microstruc
ture can result in some loss of hardenability as illustrated in Figure
24-21 (20,21).
Austenitic Stainless Steels

Austenitic stainless steels are also solution heat treated to opti
mize corrosion resistance. The principal object of the heat treatment
is to dissolve carbides and sigma phase if present to produce a

homogeneous chromium solid solution. Accelerated cooling is usu
ally required to retain carbon in solution and prevent precipitation
of chromium carbides which results in chromium depletion and
localized loss of corrosion resistance.
Corrosion-resistant stainless steels usually have a low carbon
concentration, normally less than 0.08%, and often less than 0.03%.
For best corrosion resistance, alloys should be solution treated to
dissolve grain boundary carbides prior to being put into service.
Solution treating temperatures range from 1900 to 2100 OF (1040
1150 C), depending on alloy content, section size, and corrosion
resistance requirements.
The solubility of chromium carbides in an 18% chrome-8% nickel
steel is illustrated in Figure 24-22 (22), The solubility of carbon in
austenite is about 0.03% at 1200 OF (650 QC), increases to about
24-14
TOe
uso
ITOF
1830
1000
1650
1470
1'2
1290
- -- -- -
6001
1'2
a'
...............
X,R
1010
5001
1
I
400'
~01
6 8 10 MIN
I II III
7'>-'
~1
10
TIME, hr
Fig. 24-23 Temperature-time precipitation curves for various phases observed in alloy U50 (9)
Several precipitates can form during casting, fabrication, or use of

stainless steel at moderately elevated temperatures. Tune-tempera
ture precipitation curves for grain boundary carbides and several
other phases in U50 stainless have been developed by Solomon and
Devine and are illustrated in Figure 24-23 (23). U50 contains about
21 % chromium and 7% nickel. The formation of M23C6 carbides
can occur quickly as a casting is cooled to room temperature or after
welding. Carbides can be avoided by solution treating followed by
rapid cooling often by water quenching to room temperature.
u..
,,{+C m
AUSTENITE + CARBIDES
Precipitation Hardening Steels

o
0.8
1.0
74% Fe
18% Cr
8%Ni
Flg.24-22 Phases in 18% Cr-8% Ni steel for carbon content

between 0 and 1% (8)
0.1 % at 1600 OF (870C), and then sharply increases to a value of

about 0.5% at 2500 OF (l370 C).
Because of higher carbon solubility at elevated temperatures, it is
possible to solution treat castings and weldments that may contain
smail amounts ofgrain boundary carbides, dissolve the carbon in the
matrix, release the chromium, and produce a uniform chromium
concentration throughout the structure, even at grain boundaries.
This condition promotes maximum corrosion resistance.
It has been known for many years that precipitation hardening can
occur in low carbon iron alloys containing titanium, molybdenum,
nickel, copper, and tungsten. However, since it is possible to
strengthen low alloy iron-base materials containing carbon with a
simple quench and temper treatment, the possibility of strengthen
ing alloys by precipitation hardening has not been developed to any
great extent, probably because of the higher cost of the elements
required compared to carbon. The recent development of maraging
steels, however, and the possibility of obtaining high strengths
without distortion that often accompanies quench and temper heat
treatments, has resulted in some interest in precipitation hardening
alloys.
The iron-copper system has a simple eutectoid transformation at
1560 OF and a copper solubility in austenite just above the eutectoid
temperature of approximately 4% (24). On transformation to ferrite
during cooling, the solubility ofcopper decreases to about 1.9% and
continues to decrease with decreasing temperature. The solubility at
room temperature is not precisely known, although it is quite low.
.,

Because of the increased solubility of copper in austenite, more
copper can be retained in solid solution by quenching from this
phase field than can be retained on quenching from the ferrite phase
field (25).
Solution treating creates a relatively soft condition in which the
precipitation hardening steels can be machined with greater ease,
compared to the fully hardened condition achieved after aging. The
cooling rate from the solution treating temperature must be fast
enough to prevent the precipitation hardening reaction from taking
place during quenching.
Age-hardenable steels can be aged to harden at a controlled rate.
One advantage of this type of hardening process is the absence of
distortion when compared to that normally occurring in steels hard
ened by the martensite reaction. Another advantage is the ability to
machine components in the soft, solution treated condition, fol
lowed by aging at relatively low temperatures, without significant
scaling.
A very fine dispersion of copper spherical precipitates formed
during age hardening produces the maximum internal stress and,
consequently, the highest strength attainable in copper-bearing
irons. Hardness measurements indicate that by the time the spheres
begin to appreciably increase in size, the hardness decreases. If rod
shaped precipitates begin to form, the material is well overaged (24).
The manganese-copper cast steels represent a family of low alloy
cast steels which can be precipitation hardened. These low carbon
steels normally contain 0.90 to 1.50% manganese and 1.50 to 1.80%
copper. The usual heat treating practice is to normalize from 1700
OF (925C) and precipitation harden for about two hours in the
range of 900 to 950 OF (480-510 C). This hardening reaction
produces an increase of 20 ksi (138 MPa) in tensile strength with a
minimum decrease in ductility.
CB-7Cu and CD-4MCu are stainless precipitation hardenable
steels. CB-7Cu, the equivalent of wrought 17-4 PH and 15-5 PH,
can be hardened by heating to 1925 OF (1050 0C) followed by
cooling to room temperature and aging at temperatures in the range
of 900-1150 OF (480-620 0C) depending on the desired properties.
The solution treating temperature for CD-4MCu is 2050 OF (1120
0C) with quenching from 1900-1750 OF (1040-955 0C). For appli
cations involving corrosion, this alloy is usually employed in the
solution treated and quenched condition. For structural applications,
aging treatments can be performed between 900-950 OF (480-510
C).
24-15
220
200
180
~
'8,
,...
.r.
160
140
//
Quenched
and tempered
120
0.
c::
100
"0
80
1ii
<Ii
):
60
40
20
0
220
260
Tensile strength (1000 psi)
Fig. 24-24 Tensile and yield strengths of normalized and

quenched and tempered carbon and low alloy steels of various
compositions.
Steels vary in their response to the quenching operation because the

depth below the surface that can be hardened depends on the
composition of the steel and the severity of the quench. Since the
cooling rate during a quench is fastest at the surface and slowest at
the center of a part, and since the hardening reaction is composition
dependent, hardenability is an important consideration in alloy
selection for the production of high strength parts.
Improved quenching practices may permit the use of a less har
denable, less expensive steel while still achieving desired proper
ties. The quenching practices employed. however, are usually a
compromise between a practice that could harden to the greatest
possible depth and the requirement that quenching be done without
excessive distortion or cracking (27).
Hardenability
Quench Hardening of Carbon

Steel
The best combination of properties in cast carbon and low alloy
steel can be oblfrlned by quenching to a martensitic microstructure
followed by tempering. A considerably higher yield strength can be
obtained at a given tensile strength by quenching and tempering
compared to normalizing and tempering, as illustrated in Figure
24-24.
Successful hardening depends on the geometry of the part, the
hardenability of the steel employed, and the quenching practices
employed. Quenching to develop a martensitic structure without
warping or cracking requires a good part design, a steel with ade
quate hardenability, properly maintained heat treating furnaces and
quench facilities, the proper quenchant, knowledgeable employees,
and experience.
Some heat treating problems are related to poor part design. Long
parts with small cross sections, those lacking symmetry, and those
with holes or recesses are difficult to heat treat without distortion or
cracking. Excessively rapid or nonuniform cooling ofparts can also
cause distortion or cracking (26).
The principal method of hardening carbon and low alloy steels
consists of quenching the steel from the austenitizing temperature.
Hardenability refers to the ability of a steel to be transformed

partially or completely from austenite to martensite at a given depth
when cooled under prescribed conditions. Hardenability should not
be confused with hardness. Hardness refers to the resistance of a
material to penetration as measured with a Rockwell, Brinell. or
other indentation test. Hardenability reflects the ability of a steel to
be hardened when cooled in a particular way. Hardenability is
empirically determined, and several types ofexperiments have been
devised to measure it in various kinds of steel.
Jominy Procedure. The most familiar and commonly used

procedure for measuring hardenability is the Jominy bar end-quench
test. This test has been standardized and is described in SAE-J406
and ASTM-A-255 specifications. In this test, a 4 in. long x I in.
diameter bar is austenitized at the proper temperature, dropped into
a fixture, and one end rapidly cooled with 75 OF water issuing from
a l/2 in. orifice under prescribed conditions. The austenitizing
temperature is selected according to the specific steel and may be as
low as 1500 OF (815C) or as high as 1700 OF (925 C). However
most steels are heated in the range of 1600 to 1650 OF (870-900 C).
Special precautions are taken to minimize scaling during solution
treating.
The end quenched bar is subjected to a continuous stream of cold
water and is cooled quickly under agitated water quenching condi
"'
24-16
Fig. 24-25 (a) Standard form of end-quench test

specimen (Jominy), support, and water-quench re
quirements, (b) End-quench hardenability of a sin
gle heat of steel, tested in nine laboratories, showing
reproducibility (Jominy)
112 in.
<12.7 mm)
1/8 in.
(3.2 mill)
~~-;---~
T
Unimpeded
Water Jet
4 in.
1102 111m)
l-in. 125.4 mm)
l-in.125.4-111m)
Round SDecimen
2-1/2 in.
(b4I11m)
... I
rwater at 75 F + 5 F
124 C + 2.8 cf
-L-
1/2 in.
113 111m)
(a)
F rom Quick -Opening

Valve
1/2-in.
Orifice
III
Distance from quenched end, mm

10
::[~'
0
a:
501
gf
CD
40 I-
30
20
40
50
Hardenability range, based on

tests by nine laboratories on
an SAE 4068 steel
\\
\\
c::
"E
:!p.
End-quenched method
30
20
10
(b)
12
16
20
24
28
Distance from quenched end, sixteenths in.
32
HardenabiJity and Heat Treatment
I
r
f..
tions at the lower end. The cooling rate is much slower and approxi
mates air cooling at the upper end. Rates comparable to still water
quenching, oil quenching, and forced air cooling are represented at
locations between the water quenched and air cooled ends ofthe bar.
After quenching, parallel flats are ground on opposite sides of the
bar and hardness measurements are made at 1116 in. intervals along
the bar length. The standard specimen, specimen support, water
quench requirements, and a composite of hardness-distance results
obtained on a particular steel in nine laboratories are illustrated in
Figure 24-25a and b.
In general, the Jominy test is considered to provide valid data on
steels having an ideal diameter (Dr) in the range of from about one
to six inches. The Dr can be less than 1 inch, but this usually requires
that Vickers hardness readings be taken closer to the quenched end
and closer together than possible using more conventional Rockwell
hardness measuring equipment.
The austenitizing time and temperature, extent of oxidation or
surface decarburization during austenitizing, care and consistency
of surface flat preparation, and bar positioning prior to making
hardness measurements are important procedural factors that can
influence test results. Water temperature, water-jet height, orifice
diameter, and transfer time from the furnace to the quench fixture
are also important but not as critical in obtaining consistent data.

However, all tests should be conducted in compliance with the
standards set forth (8,28).
The hardness vs. distance representation of Jominy data provides
hardness data on the steel at various cooling rates. The hardness
readings are plotted against distance from the quenched end to
produce the hardenability or end quench curve. Figure 24-26 illus
trates schematic end quench curves for four steels to show various
shapes that may be assumed by different alloys. The carbon steel is
shallow hardening and the chromium-molybdenum steel is fairly
deep hardening.
Some variation in composition occurs between heats of a given
grade. Therefore, the end quench curves can (within limits) vary
from heat to heat. When enough heats within a given grade have
been tested, the end quench curves may be plotted to form a har
denability band. Several end quench bands are illustrated in Figures
24-27 through 24-45.
DISTANCE FROM
10
30
20
40
50
20
QUENCHED END - mm
40
30
50
60
70
DISTANCE FROM QUENCHED END - mm

10
24-17
60
60
U
a:: 50
50
J:
0::
J: 40
Cr - Mo
~ .............. _ _
CIl
if)
if)
Ni"Cr- Mo
I.LJ
----
~ 30
B 30
a::
~ 20
o
~20
40
Mn-8
J:
10
10
16
24
10
40
50
1.0
1.5
2.5
2.0
Fig. 24-27 Hardenability curves for carbon cast steels
QUENCHED END - mm
30
20
0.5
DISTANCE FROM QUENCHED END - in.
Fig. 24-26 Hardenability curves for C, Mn-B, Ni-Cr-Mo, and Cr-Mo

cast steels
70
~h 102~0./~/.</~Q
40
32
DISTANCE FROM QUENCHED END - ~ in.
DISTANCE FROM
DISTANCE FROM
60
QUENCHED END
mm
60r'------r------r----~------~----_.------~
60
a::
~ 50
J:
I
CIl
CIl
40
CIl
if)
I.LJ
~ 30
a:: 20
a::
40
20
J:
10
O~I
0
0
0.5
1.0
1.5
2.0
2.5
______
________
0.5
1.0
______
______
1.5
______
2.0
2.5

Fig. 24-28 Hardenability curves for carbon cast steels
Fig. 24-29 End-quenched hardenability of manganese (1330) cast

steel
24-18

DISTANCE
FROM
QUENCHED
50
60
DISTANCE
END - mm
60
60
FROM QUENCHED
20
30
40
END - mm
50
60
U
0::
U
0::
::c 40
::c 40
(f')
(f')
(f)
(f)
l.rJ
Z
Cl
0::
10
l.rJ
Z
Cl
0::
20
<:(
<:(
20
::c
::c
0
0
1.5
1.0
0.5
2.5
2.0
Flg.24-30 End-quench hardenability of nickel (2320 and 2330) cast

steel
FROM
QUENCHED
1.5
1.0
2.0
2.5
DISTANCE
0.5
Flg.24-31 End-quench hardenability of nickel-chromium (3130)

cast steel
DISTANCE
END - mm
FROM QUENCHED END - mm
60
60
U
::c 40
::c 40
0:::
0:::
(f')
(f)
(f)
(f)
l.rJ
l.rJ
~20
EE20
<:(
<:(
J:
J:
0
0
1.5
1.0
0.5
DISTANCE
10
60
FROM
20
40
50
FROM
1.5
2..0
2.5
QUENCHED END - in.
Flg.24-33 End-quench hardenability of molybdenum (4030) cast

steel
DISTANCE
QUENCHED END - mm
30
1.0
0.5
DISTANCE
FROM QUENCHED END - in.
Fig. 24-32 End-quench hardenability band for nickel-chromium

(3140) cast steel
DISTANCE
2..5
2..0
60
10
60
FROM
2.0
QUENCHED END - mm
30
40
0:::
0:::
J: 40
::c 40
I
(f)
(f)
(f)
(f)
l.rJ
l.rJ
~ 2.0
EE2.0
<:(
<:(
J:
J:
0
0
0.5
DISTANCE
1.0
FROM
1.5
QUENCHED
2..0
END - in.
Fig. 24-34 End-quench hardenability of chromium-molybdenum

(4130 and 4140) cast steel
2.5
0.5
DISTANCE
1.0
1.5
2..0
Fig. 24-35 End-quench hardenability of

nickel-chromium-molybdenum (4330) cast steel
2..5
Hardenability and Heat Treatment I

DISTANCE
FROM
END -mm
QUENCHED
24-19
DISTANCE FROM QUENCHED END - mr

40
60
50,
60
60ri-----r----~----~-----,-----,----~r-
(/)
(/)
(/)
(I)
0::
0::
, 40
li.I
li.I
0::
0::20
O~i
0.5
1.0
DISTANCE
FROM
1.5
QUENCHED
END - in.
10
60
QUENCHED
______
______
0.5
______
1.0
______
J.5
DISTANCE FROM
Fig. 24-36 End-quench hardenability of nickel-molybdenum (4620

and 4640) series cast steel
DISTANCE FROM
2.5
2.0
DISTANCE
20
2.5
QUENCHED END,in.
Fig. 24-37 End-quench hardenability of chromium (51

steels
END - mm
____
2.0:
) cast
QUENCHED ENOl mm
FROM
20
30
40
50
60
60rl-----.-----,r-----,-----,------.~---.-,
0::
0::
, 40
40
(/)
(I)
(I)
(I)
li.I
li.I
0:: 20
I
0
0
10
60
"'
~
<
:!
0::
:~
,,""~
I ,t'
i
~
:i
~~
I\~
40
(/)
(/)
FROM QUENCHED
20
30
40
1.5
2'1
DISTANCE FROM QUENCHED ENDI- in.

nickel-chromium-molybdenum (8600) series cast steel
DISTANCE FROM
END - mm
50
J.O
0.5
DISTANCE
.~
~~
2.5
2.0
Fig. 24-38 End-quench hardenability of manganese-molybdenum

(8400) series cast steels
i
G.~
.~
1.5
1.0
0.5
DISTANCE
20
20
60
QUENCHED
0::
40
Cr - Mo CAST STEEL
C - 0.30 %
Cr - 2.50 - 3.50 %
Mo- 0.40 - 0.60 %
(/)
(I)
~20
020
0::
60
li.I
Z
li.I
END -i mm
30
i.-J:
H
ij
~
i
iP
.~
~,
o1
o
01
I
0.5
I
1.0
1.5
DISTANCE
FROM
QUENCHED
2.0
END - in.
Fig. 2440 End-quench hardenability of

manganese-nickel-chromium-molybdenum (9530) cast steel
2.5
0.5
1
1.0
DISTANCE FROM
I
/.5
QUENCHED
1
2.0
I
2.5
END - in.
Fig. 2441 End-quench hardenability of copper-manganese-silicon

cast steel.
24-20
DISTANCE
,I
iI,!
10
60
FROM
20
30
'j
~.
~m
I
40
50
60
10
60
COPPER-MANGANESE- SILICON
CAST STEEL
C - 0.20 %
Cu - 1.50-1.80 %
Mn- 0.90-1.25 %
Si - 0.85-1.10 %
(f)
(f)
0::
J:
DISTANCE
END- mm
It
QUENCHED
40
I
(f)
(f)
FROM
QUENCHED
END -mm
20
0::
J: 40
I
UJ
UJ
<{
J:
0:: 20
J:
20
<{
a
DISTANCE
FROM
QUENCHED
2.5
2.0
1.5
1.0
0.5
cast steel.
FROM QUENCHED END
10
60
30
20
40
50
0.5
1.0
1.5
2.0
2.5
END - in.
Fig. 24-42 End-quench hardenability of copper-manganese-silicon
DISTANCE
Fig. 24-43 Endquench hardenability of manganese-molybdenum

(8030) cast steel
DISTANCE
mm
WI
60
FROM QUENCHED END - mm
0::
0::
~40
J:
(f)
(f)
(f)
(f)
UJ
Z
UJ
Z
0
<{
J:
0::20
0:: 20
<{
J:
0
0
0.5
DISTANCE
1.0
FROM
1.5
2.0
2.5
QUENCHED END - in.
Ol~----~------~------~------~------~
0.5
1.0
2.5
o
2.0
1.5

manganese-molybdenum-boron (80630) cast steel

nickel-chromium-molybdenum-boron (86830) cast steel as
compared to 8630 cast steel
Ideal Critical Diameter. Jominy hardness curves can be used
opment, or where the end quench test is not practical such as on thin
section parts or stock. In these cases, the end quench curve may be
computed from the chemical analysis.
The hardenability of a steel is largely controlled by the carbon
content, grain size, and alloy element concentrations (8). For most
cast steels, the grain size is small and does not vary widely. Har
denability of cast steels, therefore, is determined largely by the
chemical analysis.
Hardenability is generally calculated in terms of the ideal critical
diameter, and equations most often used for this prediction contain
multiplicative factors for the alloy elements. The base Dr is deter
mined by the carbon concentration and grain size. The base Dr is
then multiplied by the various alloy factors, which take into account
the hardenability effects of the other elements present. Equation 4 is
one expression for calculating the ideal critical diameter.
to estimate the ideal critical diameter (DI) of an alloy. The "critical

diameter" (D) refers to the largest bar diameter that has 50% marten
site at the center after being quenched. The "ideal critical diameter"
is the largest diameter bar that can be quenched to produce 50%
martensite at its center after being given an "infinite" or "ideal"
quench. The ideal quench is one that lowers the surface temperature
of an austenitized steel to the bath temperature instantaneously.
Under these conditions, the cooling rate at the bar center depends
only on the thermal diffusivity of the steel.
The ideal critical diameter can be estimated from a Jominy curve
using Figure 24-46. For example, a steel with a 50% martensite
point 8116th of an inch from the end of the Jominy specimen can be
quenched to 50% martensite at the center of a 2.7 in. diameter bar
under ideal quenching conditions. The ideal critical diameter has
been correlated with the 50% martensite point in many alloys to
develop this figure.
Calculated End Quench Curves. There are situations

where end quench curves are not available, such as in alloy devel-
OJ = DI (carbon & grain size) * fMn * fSi fer * fMo * fv *fcu
(Eq4)
where fx = multiplicative factor for the particular alloying element
24-21
Table 24-2 Tabulated Hardenability Factors for Carbon-Grain Size and Selected Alloying Elements in Steel
Percent
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
#6
Carbon-Grain Size
#7
#8
0.0814
0.1153
0.1413
0.1623
0.1820
0.1991
0.2154
0.2300
0.2440
0.2580
0.273
0.284
0.295
0.306
0.316
0.326
0.336
0.346
0.0750
0.1065
0.1315
0.1509
0.1678
0.1849
0.2000
0.2130
0.2259
0.2380
0.251
0.262
0.273
0.283
0.293
0.303
0.312
0.321
0.0697
0.0995
0.1212
0.1400
0.1560
0.1700
0.1842
0.1976
0.2090
0.2200
0.231
0.241
0.251
0.260
0.270
0.278
0.287
0.296
Mn
Si
1.167
1.333
1.500
1.667
1.833
2.000
2.167
2.333
2.500
2.667
2.833
3.000
3.167
3.333
3.500
3.667
3.833
4.000
4.167
4.333
The base Dr and one set of alloy factors are presented in Table
24-2 (29). The alloy factors were developed based on data from
medium carbon steels of medium hardenability. These factors and
others for particular composition range steels have been incorpo
rated into calculators for quickly making calculations and plots of
Jominy and quench data (30,31).
The procedure for calculating the hardenability of a steel from the
composition includes the following steps:
1. Determine the ASTM grain size.
2. Determine the chemical composition.
3. Determine base Dr from carbon content and grain size (such as
from Table 24-2) (27).
4. Determine alloy factors (such as from Table 24-2).
5. Multiply the factors according to Eq 4 to calculate the ideal
diameter DI.
Table 24-2 does not provide alloy factors for calculating the
effects of vanadium or boron. There is some uncertainty and incon
sistency about the behavior ofthese elements. Vanadium up to about
0.1 % increases hardenability, but higher concentrations may pro
duce vanadium-rich carbides that are difficult to dissolve at standard
austenitizing temperatures.
Boron incre.bes hardenability of steels with carbon contents be
low about 0.3% but has little effect in higher carbon steels. Boron is
not widely used in cast steels because it is difficult to achieve
consistent hardening effects. When boron does raise hardenability,
it has an effect greater than in shallow-hardening steels. It is appro
priate to make a few heats when dealing with vanadium and boron,
to determine their effects in the alloy of interest.
Once the DI has been determined, the end-quench curve can be
determined using Table 24-3 and Figure 24-47 (32,33). The flISt
column of Table 24-3 gives the maximum hardness of quenched
steel as a function of carbon content. Subsequent columns provide
the hardness at reduced martensite contents of 95%,80% and 50%.
The maximum hardness is the initial hardness value often plotted
1/16 in. (1.59 mm) from the quenched end of a Jominy specimen. It
is also used in the ratio of initial hardness to distance hardness in
Figure 24-47. The numbers on the curves of Figure 24-47 indicate
the number of 1116 in. (1.59 mm) increments from the quenched
end, for example, 4/16,8/16, 12116 in. (6.35, 12.70, 19.05 mm), etc.
Ni
1.035
1.070
1.105
1.140
1.175
1.210
1.245
1.280
1.315
1.350
1.385
1.420
lA55
1.490
1.525
1.560
1.595
1.630
1.665
1.700
1.018
1.036
1.055
1.073
1.091
1.109
1.128
1.146
1.164
1.182
1.201
1.219
1.237
1.255
1.273
1.291
1.309
1.321
1.345
1.364
Cr
Mo
1.1080
1.2160
1.3240
1.4320
1.54
1.6480
1.7560
1.8640
1.9720
2.0800
2.1880
2.2960
2.4040
2.5120
2.62
2.7280
2.8360
2.9440
3.0520
3.1600
1.15
1.30
1.45
1.60
1.75
1.90
2.05
2.20
2.35
2.50
2.65
2.80
2.95
3.10
3.25
3.40
3.55
3.70
7.0 ""'- - - - - - - - -_ _ _ _ _ _ _ _ _-,
6.0
en
(I)
.s::;
5.0
.5
.5
(5
~E
3.0
'B
"iii
(I)
:2
2.0
1.0
or
4 6
8 10 12 14 16 18 20 222426 28 30 32
End quenched distance in 1/16 in. to

50% martensite point
Fig. 24-46 Relationship between end-quenched distance and

hardenability (DI)
Using the OJ already calculated from Eq 4, proceed horizontally

across Figure 24-47 and note the ratio given by the intersection with
each curve. The hardness at the indicated distances from the
quenched end can then be calculated by dividing the initial hardness
by the ratio given on the abscissa and plotting the resulting values
to draw the end-quench curve.
Since the curve is drawn based on carbon content and Dr, tables
have been published which give the hardness ratio at various dis
tances from the quenched end for various combinations of Dr and
carbon content (34,35).
It is possible to construct the end-quench hardness band for a
given alloy by preparing a curve for the maximum OJ band on the
Ii
24-22
8.0
'200
ro
III
c::
il
7.0
Is:
1 180
<!
120 ..J
<!
4.0
100
::J
..J
5.0
..J
<!
..J
<!
~6
140 :!:
<!
a::
3.0
1.0
~~8"'1 ,~40
2.5
2.0
1.5
3.0
DISTANCE
CRITICAL THICKNESS
4
8 10
20
40
IH -mm
60
100
200
10
8
a::
200W
IW
:!:
JOO ~
:!:
80
60<1
..J
<!
i=
ii:
40t:
a::
<1
..J
20<!
W
.~
0.1
0 1 VALUES
0.2
0.4 0.6
Hardness, HRC
Carbon,
....
<!
14
Table 24-3 Effect of Carbon Concentration and %

Martensite on the As-Quenched Hardness of Steel
HARDNESS
Fig. 24-48 Relationships among ideal critical size, 01, actual

critical size, 0, and severity of quenching, H
4.0
3.5
Fig. 24-47 Curves used tor the determination of hardness for

certain distances from the quenched end (2)
a::
w
INITIAL HARDNESS
RATIO
20
00
32
1.0
lW
a::
2.0
>4
o
80 U
..J
<!
60 ~
c:
71> 71
6.0
<!
i=
160 E
:!:
lJP"'> ;"},;;;
10
CRITICAL THICKNESS TH - in.
Fig. 24-49 The relation between ideal critical diameter, 01, and the
criticel thickness. TH. that can be fully hardened using a quenching
medium with severity H (6)
99%M
9S%M
9O%M
8O%M
5O%M
0.10
0.12
0.14
0.16
0.18
0.20
0.22
0.24
0.26
38.5
39.5
40.6
41.8
42.9
44.2
45.4
46.6
47.9
49.1
50.3
51.5
52.7
53.9
55.0
56.1
57.1
58.1
59.1
60.0
60.9
61.7
62.5
63.2
63.8
64.3
32.9
34.5
36.1
37.6
39.1
40.5
41.9
43.2
44.5
45.8
47.0
48.2
49.3
50.4
51.4
52.4
53.4
54.3
55.2
56.0
56.8
57.5
58.2
58.9
59.5
60.0
30.7
32.3
33.9
35.3
36.8
38.2
39.6
40.9
42.2
43.4
44.6
45.8
46.9
47.9
49.0
50.0
50.9
51.8
52.7
53.5
54.3
55.0
55.7
56.3
57.0
57.5
27.8
29.3
30.8
32.3
33.7
35.0
36.3
37.6
38.8
40.0
41.2
42.3
43.4
44.4
45.4
46.4
47.3
48.2
49.0
49.8
50.6
51.3
52.0
52.6
53.2
53.8
26.2
27.3
28.4
29.5
30.7
31.8
33.0
34.2
35.3
36.4
37.5
38.5
39.5
40.5
41.5
42.4
43.4
44.3
45.1
46.0
46.8
47.7
48.5
49.3
50.0
50.7
0.2~
0.30
0.32
0.34
0.36
0.38
0.40
0.42
0.44
0.46
0.48
0.50
0.52
0.54
0.56
0.58
0.60
M, martensite
highest allowable alloy element concentrations and another for the

minimum allowable alloy concentrations. Computing a DJ: for a heat
with all alloy elements at the top of the specification range and
another with all elements at their lower limits results in a band that
is excessively wide however. Since these extreme conditions are
unlikely to occur, establishing the band becomes one of probability
mathematics. Another approach to establishing a hardenability band
is to determine the DJ: for 50 til ~OO heats and statistically determine
the distribution of OI and thus the maximum and minimum OI
values.
Table 24-4 presents calculated DJ: values for various steels cov
ered by ASTM Specification A487. The compositions used for
computing the OI values are included in the table.
Application of Ideal Critical Diameter. The definition of

OI assumes that parts are cooled in an ideal quench. Real quenching
operations can only approach these ideal conditions. The depth of
hardening obtained in practice depends not only on the hardenability
of the particular steel but also the rate of cooling or quench severity.
Quench severity is often expressed by a number which is desig
nated as the "H value." The larger the H value, the more severe the
quench. The H value depends mainly on the quenching medium, its
temperature, and the degree of agitation during quenching. Table 24-5
lists H values for several quenchartts and quenching conditions (36).
9,
\ili;
~1{
24-23
:1
>4
Table 24-4
-4"
Class
Mn
IN-IQ
2N-2Q
4N-4Q-4QA
6N-6Q
7Q
0.25
0.25
0.25
0.33
0.15
0.80
1.25
0.80
1.50
0.80
0.60
0.60
0.60
0.60
0.60
0.15
0.15
0.60
0.60
0.60
0.25
0.25
0.60
0.60
0.85
0.10
0.20
0.23
0.35
0.50
8N-8Q
9N-9Q
ION-IOQ
llN-llQ
12N-12Q
13N-13Q
14Q
A-AN-AQ
B-BN-BQ
C-CN-CQ
DN
0.15
0.28
0.25
0.15
0.15
0.25
0.50
0.20
0.25
0.20
0.45
0.70
0.80
0.60
0.65
0.55
0.95
0.95
0.65
0.90
1.10
0.70
0.60
0.60
0.60
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.40
2.40
0.90
0.75
0.65
0.70
0.20
0.20
0.20
0.20
0.20
0.20
0.25
1.70
0.90
0.80
1.60
1.60
0.25
0.25
0.25
0.25
1.00
0.23
0.30
0.55
1.05
0.25
0.25
0.10
0.10
0.10
0.10
'g
~~~
:j:"ii"
-.~
-it
.::;;0:
I"
-~
II:1
il
iii
Calculated D1 for ASTM A487 Low Alloy Steels

AnaI;t:sis used in calculation, %
Si
Cr
Ni
Calculated Dl
mm
in.
Mo
VO.OS
V 0.06
CuO.33
BO.OO4
1.59(a)
2.69
3.86
9.48
4.8O(b)
40(a)
68
98
241
122(b)
9.86
4.63
5.42
4.61
8.23
3.32
4.70
1.28
1.62
1.70
1.83
250
118
138
117
209
84
119
33
41
43
46
(a) Does not include the effect ofV. (b) Does not include the effect ofV and B
Table 24-5
Grossman Numbers and Film Coefficients for Selected Quenchants

Quenmant
tem.l!erature
0C
OF
32
90
55
130
Past oil
60
140
25% polyvinyl pyrrolidone
43
110
65
150
27
150
300
80
Quencbant
Water
Conventional oil
Martempering oil
Air
Qnenchant
veioci!1
mls
ftlmin
0.00
0.25
0.51
0.76
0.00
0.25
0.51
0.76
0.00
0.25
0.51
0.76
0.00
0.25
0.51
0.76
0.51
0.51
0.00
2.54
5.08
0
50
100
150
0
50
100
150
0
50
100
150
0
50
100
150
100
100
0
500
1000
Grossman
number
(H-h/2k)
1.1
2.1
2.7
2.8
0.2
0.6
1.5
2.4
0.5
1.0
1.1
1.5
0.8
1.3
1.5
1.8
0.7
1.2
0.05
0.06
0.08
Effective film coefficient

W/m2.K
BtultV . h . OF
5000
9000
12000
12000
1000
2500
6500
10500
2000
4500
5000
6500
3500
6000
6500
7500
3000
5000
200
250
350
880
1600
2100
2100
180
440
1100
1850
350
790
880
1200
620
1100
1200
1300
530
880
35
44
62
Source: Ref 12
<;of
r"
il~
,"'\f
r"'
ri
I ~
I,~1"
Is
i'"

.:.
The diameter of a cylinder that will quench to produce 50%

martensite at the center (D) in commercial quenching operations is
always less than the ideal critical diameter (Dr). The relation be
tween D and Dl for quenches of various H values is illustrated in
Figure 24-48. For example, if a steel has Dl =4 in. (102 mm) and is
quenChed in water where H 1.5, a 3.5 in. (89 mm) diameter bar
can be expected to quench to 50% martensite at the center. If the
same steel is quenched in oil with H = 0.4, a cylinder with a diameter
of2.25 in. (57 mm) can be expected to harden to produce 50%
martensite at the center.
The relationship between ideal critical diameter, quench severity,
and the thickness of plates that can be hardened is illustrated in

Figure 24-49. For example, assume that a plate of 4340 steel with a
DI of 6.42 in. (163 mm) is quenched in still water (H=1.1 from Table
24-5). The corresponding critical plate thickness is 3.0 in. (76 mm).
This is the thickness of a plate which can be hardened to 50%
martensite. A 6 in. (152 mm) diameter rod or cylinder will harden to
50% martensite in the center under the same conditions. The differ
ence in hardened depth between plates and cylinders arises from the
slower cooling rate in plates because they have lower surface area
to mass ratios. The bar actually has twice the surface to volume ratio
(when the bar diameter equals the plate thickness) and consequently
can cool faster during quenching.
Cooling Rate Equivalence. Different parts of a section cool
at different rates when being quenched. The rates of cooling are
24-24
Part 5B: Processing Properties
Table 24-6
End-Quench Distance vs. Bar Diameter (30)
When the end-quench hardness curve of a steel has been found, this table enables one to estimate the hardnesses that would be
obtained at the centers of quenched round bars of different diametersl when that same steel is quenched with various severities of
q!Jench. For each successive 1/16 in. position, the hardness obtained In the end-quench test would be found at the center of the bar
size shown.
Equivalent bar diameter when guenched
Distance from end in

end-quench test inches
Ilt6
21t6
31t6
Vg
4ft6
5/16
1/4
61\6
3fs
'l'16
1Yt6
1/2
91t6
10/16
lllt6
1~6
l31i6
I4ft6
15/16
16/16
171i6
%
7/g
181\6
191i6
1Ys
20/16
11/4
21/16
22;16
23/16
241t6
25/16
26/16
27/16
28/16
1%
1V2
1%
1%
291t6
30/16
31/16
32/16
1%
2
Still
oil
Circulated
oil
Still
water
Circulated
water
Still
brine
Infmiteor
idealized
quench
B"'O.25
B=O,45
B=1.0
B"'1.S
B-2.0
B=oo
0.1
0.2
0.35
0.5
0.6
0.8
1.0
1.1
1.3
1.4
1.6
0.15
0.3
0.55
0.80
0.95
1.2
1.4
1.5
1.7
1.9
2.1
2.2
2.4
2.5
2.7
2.8
3.0
3.1
3.3
3.4
3.5
3.6
3.7
3.8
4.0
4.0
4.1
4.2
4.3
4.4
4.5
4.5
0.3
0.5
0.85
1.15
1.4
1.6
1.8
2.1
2.3
2.5
2.8
3.0
3.2
3.3
3.5
3.7
3.9
4.0
4.1
4.3
4.4
4.5
4.7
4.8
4.9
5.0
5.1
5.2
5.3
5.4
5.5
5.5
0.35
0.65
1.0
1.3
1.6
1.8
2.0
2.3
2.5
2.7
3.0
3.2
3.4
3.5
3.7
3.9
4.1
4.2
4.4
4.5
4.7
4.8
5.0
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.8
0.4
0.75
1.25
1.5
1.75
2.0
2.3
2.5
2.7
2.9
3.2
3.4
3.5
3.7
3.9
4.1
4.2
4.4
4.5
4.7
4.8
4.9
5.1
5.2
5.3
5.4
5.5
5.6
5.6
5.7
5.8
5.9
0.7
1.15
1.6
1.9
2.2
2.4
2.7
2.9
3.2
3.4
3.6
3.8
4.0
4.2
4.4
4.6
4.7
4.9
5.0
5.1
5.3
5.4
5.5
5.6
5.8
5.9
6.0
6.1
6.2
6.2
6.3
6.4
1.7
1.9
2.0
2.1
2.3
2.4
2.5
2.6
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.6
3.7
3.8
affected by the severity ofquench and section thickness as indicated

in Figures 24-48 and 49 (37). The cooling rates along the length of
Jominy end-quench bars also varies depending on the distance from
the quenched end. Consequently, there should be a cooling rate
somewhere along an end-quench bar which corresponds to the
cooling rate at a point of interest in a part being quenched.
To determine this point on the end-quench bar, the cooling rates
of interest in the actual part are measured over some temperature
intervaltypically from 1350 OF (730C) down to 600 or 700 OP (315
or 370C). This is the range where the greatest tendency of trans
formation to pearlite or bainite occurs.
Figure 24-50 provides the end-quench equivalent at the surface,
center, and two intermediate positions in cylindrical sections
quenched in solutions providing different quench severities (38).
These curves allow one to determine the hardness at a particular
distance from the end of a Jominy end quench bar and predict the
hardness at the surface, half radius, 3/4 radius. and center of bars
quenched in oil, water, and salt. Sitnilar data are presented in tabular
form in Table 24-6 (39).
The curves in Figure 24-50 and data in Table 24-6 can be used for
shapes having other cross sections such as square or hexagonal bars
by determining equivalent diameters. The equivalent diameter is
considered to be the diameter of a circle that can be inscribed in the

section of interest.
Data on end-quench equivalence of plates are not numerous.
Table 24-7 provides the equivalent end-quench distance for the
center of various plate thicknesses under different quench condi
tions. Table 24-8 presents end-quench equivalent data for different
parts of plate sections quenched in agitated water. Whether a graph
such as Figure 24-50 or a table such as Table 24-8 is used, the result
is equivalent distance from the quenChed end for points in the cross
section of the quenched piece having equivalent hardness.
Specifying Hardenability. Hardenability is specified in sev

eral different ways. The Society of Automotive Engineers Recom
mended Practice J-435a covers three grades of 0.3% carbon steel.
The hardenability bands for the three grades are illustrated in Figure
24-51. The choice of analysis to fit the chosen grade is usually made
by the manufacturer.
When the hardenability band has been defined for the steel, it can
be accepted by both producer and customer. The end-quench curve
for each heat made to this specification is required to fall within the
band. The specification may call for a range of hardness at one or
more distances from the quenched end, i.e. Rockwell C 38 to 53 at

DISTANCE FROM
END - mm
QUENCHED
'f -,~
.1
~ :'.~
6.
1 ~
-:'
ffi
I-
w
::!:
OIL
SALT
2456789
"
1.1
150 E
E
125
loom
l1.,
75
(t:
5O~
<{
(t:
<{
m
l1.,
L...
Fpm
I
2
3
4
5
6
7
8
9
WATER
OIL
OIL
OIL
OIL
SALT
SALT
SALT
SALT
200
750
200
50
0
135
50
35
5
3'4 RADIUS
WATER
Fpm
I
2
3
4
5
6
7
8
9
SALT
2~,~ ~ /6/7,9
OIL
WATER
OIL
OIL
OIL
OIL
OIL
SALT
SALT
SALT
IIrl
I"
CENTER
I
E
125
WATER
lOOm
l1.,
75
0
(t:
50
25
<{
is
w
~
::!:
::!
CI
8 10
20
40
DISTANCE FROM QUENCHED ENDSIXTEENTHS OF AN INCH
Table 24-7
10
20
40
DISTANCE FROM QUENCHED END

SIXTEENTHS OF AN INCH
Equivalent End-Quench Distance Plates-Hardness at Center of Section

Plate
thickness
~1
(t:
<{
~.
is
'0
150
200
750
700
200
50
0
135
35
5
(t:
w
I
w
::!:
Fig. 24-50 Correlation curves for

identical cooling times in
end-quench hardenabiJity
specimens and round bars
quenched in hot salt, oil, and
water. Water was at 75 of (24
0C); mineral oil [Saybolt universal
viscosity at 100 of, (38C), 79
sec], at 120 of (49 C); molten
salt, at 400 of (204C) (7).
::!:
25
6 r
(t:
<{
oL...
~"
WATER
OIL
2
RADIUS
<{
'1'1
>j
-,
,j
1/2
l1.,
:1
60
40
20
SURFACE
,i
10
15
=
5
(t:
<{
I :~
24-25
Still oil
in.
mm
0.5
1.0
1.5
2.0
2.5
3.0
3.5
12.7
25.4
38.1
50.8
63.5
76.2
88.9
1;16 in.
mm
4.5
10.0
15.5
21.5
29.5
7.1
15.8
24.6
34.1
46.8
6116 in. (9.5 JIl1) from the quenched end. The distance is generally
chosen as the ~quivalent to a critical region iii the part being heat
treated. Other specifications may set maximum and minimum dis
tances from the quenched end for a given hardness, i.e. Rockwell C
40 at 3/16 to 14/16 in. (4.7-22.2 mm) from the quenched end.
Still other hardenability specifications may set lower and upper
limits on the ideal critical diameter. This range is not calculated from
the extreme composition limits of the steel grade, but from the
distribution of Dr values determined from several production heats.
Hardenability Bands. Hardenability bands for cast steels

have been determined by the Steel Founders' Society of America.
They include carbon steels and many of the frequently used low
alloy steels. Several bands are illustrated in Figures 24-27 through
24-45. Some of the figures show the effect of carbon in alloys where
the rest of the analysis is essentially constant.
The hardenability bands are useful in selecting alloys for particu
lar applications. The hardenability bands for wrought steels are
similar to comparable cast steels if both steels have similar compo
sitions and grain sizes.
Equivalent end-quench distance

guenchedin
Stillwater
1/16 in.
mm
2.0
6.0
8.5
12.0
15.0
19.5
26.0
3.2
9.5
13.5
19.1
23.8
30.9
41.3
Ag!tated water
mm
%6 in.
2.0
6.0
8.5
12.0
14.0
17.5
24.0
3.2
9.5
13.5
19.1
22.2
27.8
38.1
Figures 24-43, 44, and 45 illustrate the effect of 0.001 to 0.006%

boron on the end-quench hardness band. Excessive additions of
boron are not recommended because of the possibility of boride or
carboride compound formation and the detrimental effects of these
compounds on machinability and mechanical properties.
Part of the width of the band for a given alloy is due to variations
in analysis, including residual elements, from heat to heat. Some
elements, chromium and molybdenum, for example, have an appre
ciable effect on hardenability even when present in small quantities.
As a consequence, residual alloy element concentrations must be
monitored and controlled as closely as specified elemental concen
trations if hardness values within the hardenability band are to be
maintained.
Quenching To Produce Martensite

The term quenching is sometimes used to refer to the rapid heat
extraction required when parts are being austempered or'martem
pered in hot oil or salt baths, but its use in this section is restricted
to the process of rapidly removing heat for the purpose of forming
'.
24-26
"
Table 24-8 Equivalent End-Quench Distance for Plates-Quenched in Agitated Water

Equivalent end-quench distance in 1/16 in. and mm
{at three distances from the surface}
I;.. in.
(6.4mm)
(1.6mm)
Plate
thickness
in.
(mm)
Ilt6 in.
0.5
1.0
1.5
2.0
2.5
3.0
(12.7)
(25.4)
(38.1)
(50.8)
(63.5)
(76.2)
1.3
2.0
2.5
2.5
3.0
3.0
DISTANCE FROM QUENCHED
(2.1)
(3.2)
(3.9)
(3.9)
(4.7)
(4.7)
(3.2)
(6.4)
(9.5)
(12.7)
(15.9)
(19.8)
2.0
4.0
6.0
8.0
10.0
12.5
Center
2.0
6.0
8.5
12.0
14.0
17.5
(3.2)
(9.5)
(13.5)
(19.1)
(22.2)
(27.8)
1600~1----------------------------------~
END - mm
Cooling curves
u.
o
h'
}St,:e
~
(I)
}s~e
Cooling rate curves

0'
12
16
20
24
28
DISTANCE FROM QUENCHED ENO

SIXTEENTHS cr: AN INCH
20
40
60 80 100 120 140 180 180 200

TIme (sec) or cool rate (OF/sec)
Fig. 2452 Cooling curves and COOling rate curves at center of a

1.5 in. diameter probe quenched in unagitated hot water
Fig. 24-51 Hardenability bands specified by SAE J-435a
martensite. Just how rapidly the steel must be cooled to miss the
nose of the transformation curve depends on the hardenability of the
particular alloy. More hardenable materials do not require fast cool
ing rates, but achieving martensitic microsttuctures in most con
sttuction grades of steel requires faster cooling rates than can be
obtained in air. It is usually necessary for steel to be sprayed or
immersed in a liquid quenchant to increase the rate of heat extrac
tion. but in a few cases it is possible to use gas quenching, especially
agitated gas under pressure.
There are three stages of heat extraction associated with quench
ing. Duplicate temperature-time curves in a part being quenched and
corresponding cooling rate curves are illustrated in Figure 24-52.
The various stages involved during cooling are also labeled in this
figure. Slow cooling occurs in stage A as a result of a vapor barrier
formation around the part Heat transfer during this period occurs
primarily by radiation through the vapor blanket. Stage B quenching
is relatively short in duration but represents the interval where the
rate of heat removal is highest. During this stage the vapor blanket
collapses and high heat extraction rates associated with nucleate
boiling on the part surface are achieved. The final stage of quench
ing (C) is associated with the end of boiling and relatively slow heat
removal by conduction and convection from the steel into the liquid
quenchant (8,36).
In general, agitating the quenchant accelerates heat removal by
limiting the duration of stage A and accelerating heat removal in
stages B and C. An increase in the quenchant temperature usually
slows the rate of heat removal if other conditions are held constant.
The cooling rates and temperature ranges of the three stages vary
with the type ofquencpant and the mass ofthe parts being quenched.
The fastest overall cooling rates are achieved with brine solutions,
generally followed by water. polymer solutions in water, oils. inert
gases, and air.
Transformation diagrams provide data on various steels and the
cooling rates necessary to form martensite and avoid ferrite or
pearlite formation. The required cooling rate is determined by the
hardenability of the steel. The actual rate that can be achieved is a
function of the quenchant and facility. and the thickness and geome
try of the part, and the heat capacity and thermal conductivity of the
alloy. The quenching capacity of a given facility depends on the
quenchant used, its temperature, and the extent of agitation.
Quenchants
Many materials and solutions have been used as quenchants for
steels. The fastest quench rates are usually obtained using water or
brine, i.e. water-salt solutions. Slower. less drastic quenchants in
clude the water-based polymers. oils, and salts. Very slow rates can
be obtained by cooling in air or inert gas atmospheres. This review
is limited to commonly used materials including water, brine, water
based polymers, oils, with a limited review of salts.
Water. Water has several advantages over other quenchants. Water

is readily available, cheap. not hazardous, and can be used to
produce very high quench rates. However. its ability to extract heat
can cause cracking in more hardenable steels. except in simple
shapes. Water quenching is prone to cause cracking and distortion
in parts with a difference in cross section area of more than 1:4 and
parts with holes and grooves.
The bath temperature during water quenching is normally in the
range of55 to 75 OF and should be less than 110 oF. Agitation ofthe
bath is needed if soft spots are to be avoided. Soft spots can
"
24-27
Table 24-9 Effects of Water Temperature on Cooling Rates and Predicted Hardness in 1.5 inch Diameter
Steel Bars Quenched from 1550 of
Alloy 4130
Film
Quenchant
j:a
f'
Water
Water
Water
Water
Water
Water
Water
Water
Water
Bath
temp
Velocity
(OF)
(FPM)
80
90
100
120
140
160
180
200
212
Cooling rate at
1300 of
650F
(F/sec)
400F
58.6
58.1
55.8
43.7
16.8
10.1
8.5
6.3
5.4
26.2
26.3
25.8
25.9
25.7
24.6
24.2
23.7
22.6
13.2
13.0
12.5
12.2
50
50
50
50
50
50
50
50
50
sometimes be detected by etching a part in hot HCI:H20, which

causes the soft spots to appear as light areas. Contamination of water
baths with sludge, oils, and scale reduces the cooling rate, and
contamination with salt increases the cooling rate (7,10,27).
Changes in the quenching medium can seriously affect the quench
severity. An increased temperature reduces the cooling effect of
water, but has little effect on that of oils. The water temperature is
not important up to 75 OF (25C) but between 75 and 120 OF (25
and 50C), the cooling power decreases materially. Above 120 OF
(50C) the cooling power of water is quite low. Dissolved salts
increase the cooling power of water, while emulsions reduce the
cooling rates that can be obtained. Additional data on water quench
ing and some of the control measures has recently been published
(38).
The effect of water bath temperature on cooling curves and
cooling rate curves produced when 1.5 in. diameter bars were
quenched into water at 80, 90, 140, and 160 Of flowing at 50 fpm is
illustrated in Figure 24-53. Some characteristics of these conditions
are presented in Table 24-9. Water at 140 OF produced a cooling rate
at 1300 OF of 16.8 OF per second. With a lower water bath tempera
ture of 80 OF, the cooling rate at 1300 OF (which is commonly used
as an indicator of the ability of quenchants to extract heat) was 58.6
OF per second, about 3.5 times higher than water at 140 OF.
Brine. Brine can be prepared by mixing sodium chloride, calcium

chloride, or sodium hydroxide with water. A 10% mixture of salt in
water provides a material that can rapidly remove heat from cast
ings. The salt elevates the boiling temperature and hastens the
collapse of the vapor blanket in Stage A cooling. Agitation and bath
temperature are not as critical in brine solutions, nor are soft spots
as much of a problem as with water. The uniformity of heat removal
is enhanced by the presence of salt. Brines are not as sensitive as
water to contamination but the salt concentration must be controlled.
Equipment and parts exposed to brine must be protected, however,
from its corrosive action.
,~
I
~;.;:
"J;,
F*:,;~
coefficient
Oils. Oil quenchants are readily available and widely used. Suc
cessful performance demands that the oil produce the desired cool
ing rate and resist thermal degradation during prolonged use. Deg
radation of quenching oil can result from contamination, oxidation,
or thermal cracking at elevated temperatures (27,40).
There are four general classifications of quenching oils: conven
tional, fast, hot, and martempering. Other types such as those used
in vacuums exist, but these are generally restricted to very special
ized operations.
11.3
10.6
9.7
9.1
9.9
(BTU/)
(ft2 . oF)
1652.5
1589.4
1346.5
617.1
132.8
73.5
60.1.
42.5
35.8
1oo0~;;::==------------------~
1400 L
1200
u.
e...1000
2!
~ 800
Oi
If
600
I-
400
(\)
200
00
10
20
30 40 50 00 70 80 90 100
Time (sec) or cool rate (OF/sec)
Fig. 24-53 COOling curves and cooling rate curves produced by (a)
80 OF, (b) 90 OF, (c) 140 OF, (d) 160 OF water flowing at 50 fpm past
a 1.5 in. diameter bar
Conventional oils have additives for improving thermal stability.

The viscosity is usually about 100 Saybolt universal seconds (SUS)
at 100 OF, with a range of 70-300 SUS. Conventional oils are
normally used at a bath temperature of about 120-150 OF to quench
harden steels (27).
Fast oils usually have a lower viscosity than conventional oils,
and additives are present to speed quenching and improve thermal
stability. The additives present include inhibitors to slow oxidation,
wetting agents to increase the quenching rate, and detergents and
dispersants to control sludge and contamination. The viscosity of
fast oils ranges from about 70 to 125 SUS at 100 OF. Fast oils are
used at 120-150 OF for less hardenable alloy steels (27).
Hot oils are designed for use in baths operated at temperatures of
200-300 OF and sometimes up to 400 OF. The higher bath tempera
tures require that hot oils have a higher viscosity and improved
oxidation resistance. The viscosity of hot oil is typically in the range
of 250 to 3000 SUS at 100F. Hot oils are used to harden alloys or
part shapes that are more prone to distortion and cracking.
Martempering oils are used at temperatures greater than 300 OF
and usually at 450 OF. Because of the high operating temperature,
martempering oils need the highest viscosity and thermal stability.
The viscosity of martempering oil is usually around 2000 SUS at
.
24-28
1600
1400
__ 1200n
u..
~ 1000
l!:!
~ 800
//15%/5%
Q)
~ 600
Q)
O!~--~--~~~~---L--~--~~--~--~~
15 30
45 60 75 90 105 120 135 150

Fig. 24-54 Cooling curves and cooling rate curves in a 1 in.

diameter stainless probe quenched in 5, 15. and 25% at 110F
and flowing at 50 fpm
E
!!!
15
30
45
60
75
90
105 120 135 150

Fig. 24-55 Cooling curves and coaling rate curves in a 1 in.
diameter stainless probe quenched in 10% PAG at 80,100, and
120 OF and flowing at 50 fpm
u..
100
~
0
E
Q)
I
o.
o
15
30
45
60
II!
75
90 105 120 135 150

Fig. 24-56 Cooling curves and cooling rate curves in a 1 in.
diameter stainless probe quenched in 20% PAG 110 OF and
flowing at 0,50, and 100 fpm
100 OF, and the oils are used with martempering treatments applied
to very crack-sensitive parts (41).
The duration of each quench stage is considerably different from
those provided by water. The duration of all stages (A, B, and C) is
longer than produced by water, resulting in slower cooling rates
through the martensite formation range.
An ideal quenchant would provide fast A and B stage cooling to
prevent the formation of pearlite or bainite, and produce a slow C
stage to minimize distortion and cracking during martensite forma
tion. A fast oil has a faster A and B stage than conventional oil, but
a similarly slow C stage. Therefore, a fast oil will harden a part more
than a conventional oil without necessarily contributing to the dis
tortion.
Used conventional oils usually produce faster cooling rates than
new oils because thermal cracking and the increased soluble resin
and acid contents increase the ability of the oil to wet parts. This
increase in quenching speed is usually undesirable, and the faster
rates indicate oil degradation and instability (10.42,43).
Water contamination of oil quench baths can occur from conden
sation or leaks in the oil cooling system. Water contamination can
sometimes be heard as a crackling or frying sound during quench
ing. Contamination is undesirable since it can lead to staining,
foaming, nonuniform hardness, and cracks. Water contamination
can even lead to steam explosions.
Staining indicates oil breakdown and is associated with varnish or
lacquer precipitation on parts during quenching. Carbon steels are
very sensitive to staining. Oil oxidation and as little as 0.03% water
contribute to staining. The tendency toward staining can be qualita
tively assessed by keeping a thin coat of oil on a test panel at 500 OF
in air for 24 hours (27,41).
The use of oil quenchants requires that proper consideration be
given to the fue hazard, ventilation of smoke, and protection from
water contamination. Oil usually ignites for one of two reasons-a
partially submerged load leading to overheating of the bath surface
and ignition or operation of the oil too close to the flash point. Oil
quenching systems should be designed to fully immerse the tray or
basket if any part of the system fails. Low oil viscosity helps
minimize fue danger by aiding in removing the heated oil from the
part surface more quickly.
In addition to these considerations, oil tanks must have an ade
quate drain in case of fue; fue extinguishers must be available;
safety covers, and other safety equipment must be considered during
design and construction. Flame and smoke are both minimized if the
oil boiling point exceeds 630 oF. Smoke formation during quench
ing necessitates good ventilation over the quench tank, but the
ventilation should automatically cease in case of fire (40).
Polymer Solutions. The technology ofpolymer quenching has

seen dramatic growth in the past 20 years. The principal quenchant
has historically been water, but with the discovery of petroleum,
hydrocarbon products became a major quenching medium. More
recently, it has been established that certain water-soluble organic
polymers are useful in modifying the cooling characteristics of
water to achieve properties between those of water and oil. Three
types of polymers are in prominent commercial use in the United
States. These are: polyalkylene glycol (PAG), polyvinyl pyrrolidone
(PVP), and polysodium acrylate. Polyvinyl alcohol (PVA) has been
used but has only limited use in the United States at the present time
(40).
In general, it is the objective of polymer quenchant suppliers to
provide products that can be used to provide quench severities
ranging from those equivalent to oil to those that may be more
severe than water.
Polyalkylene glycol. Polyalkylene glycol, or polyalkylene glycol
esters, were fust introduced as commercial products in the early
1940s. These materials are formulated by the copolymerization of
ethylene and propylene oxides although higher alkylene oxides
andlor aryl oxides may also be used.

By varying the molecular weights and the ratio of alkylene oxides
used to manufacture the product, polymers having broad range of
properties may be produced. The proper selection of a PAG polymer
provides a product that is soluble in water at room temperature.
However, PAG polymers typically exhibit inverse solubility-that
is, they are soluble at room temperature but insoluble at elevated
temperatures. This phenomenon provides a mechanism for control
ling the heat transfer from hot metal by producing a polymer-rich
layer around the part Then, as the metal part cools and approaches
the temperature of the quenchant, the PAG redissolves in the quen
chant bath.
Polyalkylene glycol quenchants have been particularly useful
because of the processing latitude they provide. Quench severities
from slow oil to brine may be possible by varying the bath tempera
ture, agitation rate, and polymer concentration. Relatively diluted
solutions will provide a quench severity similar to water but reduce
cracking and distortion because of the slow cooling provided at low
temperatures.
PAG quenchants can also be selected to provide improved wetting
so that the cooling rates achieved are faster than water and approach
those achieved by brines. Thus, "brine quenching" is possible with
out the hazards and corrosiveness attendant with the use of salts or
caustic solutions.
Three principal parameters are recognized as controlling the rate
of cooling during polymer quenching:
Quenchant concentration
Quenchant temperature
Quenchant agitation
The influence of PAG concentration on cooling rates in a I in.
diameter probe is illustrated in Figure 24..-54 (40). The slower cool
ing rates achieved at the higher concentrations correspond to an
increase in the thickness of the polymer layer that surrounds the part
during quenching. Polyalkylene glycol quenchants are not sensitive
to minor changes in polymer concentration. This strong concentra
tion dependency is a recognized deficiency ofpolyvinyl alcohol and
some other "film-forming" polymers used as quenchants.
PAG solutions decrease in their cooling capability as the bath
temperature is elevated. The curves shown in Figure 24-55 are
illustrative of the general trends that occur with changes in bath
temperature. More detailed data would require specific identifica
tion of the particular PAG quenchant employed.
The use of polyglycol quenchants without agitation is not recom
mended. In general, low to moderate agitation is essential to ensure
that adequate replenishment of polymer occurs at the hot metal
surface and to provide uniform heat transfer from the hot part to the
surrounding qu~nchant Vigorous agitation may be essential to
achieve a rapid 'jllte of cooling (for example, with a low hardenabil
ity steel) to avoid undesirable transformation products. As agitation
is increased, the; cooling rates rapidly increase, as clearly illustrated
in Figure 24-56. A contour plot illustrating the variation in Gross
man H factor, with varying polymer concentration, bath tempera
ture, and circulation rates of an illustrative PAG quenchant is shown
in Figure 24-57 (40).
The refractive index of oxyalkylene glycol polymer solutions (in
the range employed for quenching) is almost linear with concentra
tion. Thus, the refractive index of a PAG quenchant solution can
serve as a measure of product concentration. Refractive index vs.
quenchant concentration calibration plots are usually available from
quenchant manufacturers. However, since this relationship is almost
linear, a calibration plot can be easily constructed by the heat treater
by obtaining refractive index values at 0, 10, and 20% concentra
tions. A best fit line can be calculated by a linear regression analysis
of the data. The resulting equation can then be used to calculate the
quenchant concentration from a refractometer reading.
"
140
Q)
Cire rate = 50
Cire rate 75
Clre rate = 100
... ,
130
"
...... , ...
:s
~
Q)
"".~,
." ,
135
,,
'
....."
"
125
"
"
Q.
E
.l9
:6
nl
III
120
115
........
'
""
...
24-29
"
"" .... 0.4

'
' ......... 0.6 """
"
"
"""',
0.4
"'. 0.6
.....
...
0.8 ......... ,
...
"
... ... ,
...
L"" . . "
'.
'It",
... ...
'
.. .
"
'.
...
'"
'
'.
' ...............
110
...
105
100 " ,
15 16
17
....,
18
19
20
, '"
21 22
,".,
23 24
,....
25
Polymer concentration
Fig.24-57 Grossman hardenability factor of UCON quenchant E
Industrial optical refractometers that employ an arbitrary scale

may also be calibrated. Whereas such instruments prove valuable
for day-to-day monitoring of the quenchant concentration, the re
fractometer also registers other water-soluble components such as
salts in the quenchant. When the refractometer reading begins to
provide erroneous numbers, another analytical test is required to
define the "effective" quenchant concentration.
Kinematic viscosity measurements, which correlate with concen
tration are also useful with PAG quenchants, Plots of concentration
of kinematic viscosity are available from quenchant manufacturers
and are similar to those for refractive index with one exception:
viscosity is usually a nonlinear function of polymer concentration.
Therefore, higher order regression analyses must be used to relate
concentration to viscosity. Additional tests for pH, inhibitor level,
and conductance also are useful adjuncts to a monitoring program
(40).
Polyvinyl pyrroUdone. Polyvinyl pyrrolidone is derived from the
polymerization of N-vinyl-2-pyrrolidone. PVP is a water-soluble
polymer characterized by its complexing and colloidal properties
and by its physiological inertness. PVP is available as a white,
free-flowing powder, manufactured in four molecular-weight
grades.
Solutions of PVP in water were fIrst introduced as quenchants in
1975. The quantity ofpolymer recommended for a solution concen
trate is generally about 10% polymer solids and solution must
contain a rust inhibitor and a bactericidal preservative.
As with other polymer-type quenchants, concentration, bath tem
perature, and agitation play a role in establishing the cooling char
acteristics, The quenching rates tend to be faster with PVP quen
chants during the stable film and nucleate and boiling stages, but
slower during the convection stage. Since PVP is soluble up to the
boiling point of water, a broader working range of temperatures for
quenching can be employed. Optical refractometer readings will
provide initial control of concentration, but backup with viscosity
measurements is strongly recommended (40).
Polyacrylates. Sodium polyacrylates (PA) represent a third class
ofpolymers used to formulate aqueous polymer quenchants ..The PA
polymer can be produced by the homo-polymerization of sodium
acrylate, or the alkaline hydrolysis of a polyacrylate ester.
,)
"'
24-30
1000
1000
1800
27 C (80 oF)
-1800
45 C (110F)
- 1600
1600
Cooling curves of
polyacrylate solutions
concentration, wt%
--2.5%
- - - 5.0%
10.0%
- - 20.0%
;j
~
Q)
Cl.
1400
- 1200
600
400
- 1000
- 800
600
,
~.~
,'.... ," .......
200
u.
0
Q)
Cl.
u
.,
600
400
---
--
!!!
Q)
01
0
80
20
40
lime, s
60
--2.5%
- - - 5.0%
10.0%
- - 20.0%
Table 24-10
u.,
1000
i
~
oi 600
-1400
20
40
lime, s
60
Q)
Cl.
E
Q)
l-
~
Q)
OF
Water
90
90
90
90
130
130
130
130
140
140
140
140
110
110
110
110
150
300
80
80
80
25% water soluble

polymer
Conventional oil
Marternper oil
Air
- 1400
2.5%
5.0%
10.0%
20.0%
- 1200
.,u.
1000
i
1!
800
400
600
Q)
Cl.
E
Q)
I
400
200
'-'
200
01
0
80
:::I
j200
20
40
lime, S
60
80
Grossman Numbers and Film Coefficients for Selected Quenchants
Quendumt
Fast oil
-
--
'5
~
Q)
Temperature
Water
- 1600
u.
800
-1
01
0
Cl.
E
Q)
11800
COOling curves of
concentration, wt%
1200
600
\ \
!!!
Q)
80
1000 I
1400
400
2001
i
.a
200
-1
1600
2i
75 C (170 OF)
Cooling curves of
concentration. wt%
u.
400
..............~
~.
1800
~ ~~
l'~
- 800
200
60"C (140 OF)
!!!
1400
- 600
200
1000
2.50/0
5.0%
10.0%
20.0%
Q)
400
40
Time,s
20
1200
i
.a
0
0
Cooling curves of
concentration, wt%
800
Fig, 24-58 Cooling

rate of a polyacrylate
quenchant as a
function of
concentration and
temperature
Veloci!l:
C
32
32
32
32
55
55
55
55
60
60
60
60
43
43
43
43
65
150
27
27
27
m/sec
ft/min
0.00
0.25
0.51
0.76
0.00
0.25
0.51
0.76
0.00
0.25
0.51
0.76
0.00
0.25
0.51
0.76
0.51
0.51
0.00
2.54
5.08
0
50
100
150
0
50
100
150
0
50
150
150
0
50
100
150
100
100
0
500
1000
GroSSll1an
number
H=h/2k
1.07
2.07
2.70
2.76
0.20
0.64
1.49
2.44
0.49
1.04
1.13
1.51
0.81
1.33
1.50
1.77
0.74
1.21
0.05
0.06
0.08
Effective film coefficient

Btujhft2F
W/cm2sec
0.46
0.89
1.17
1.19
0.08
0.27
0.65
1.05
0.21
0.45
0.49
0.65
0.35
0.58
0.65
0.77
0.32
0.52
0.021
0.026
0.034
300
621
810
828
60
192
447
732
147
312
339
453
243
399
450
531
222
363
15
18
24
"

1.0 I
11111111111 " " 1l1111l111117171l7T/??'~
24-31
--::::::::...-----
0.9
0.8
0.7
0.6
Q
~
Q
0.5
0.4
0.3
0.2
0.1
011111111111 11111111111[1111[[111[1111111111111111111111111111111111111111111111111111111
0.1
0.2
0.3
0.4 0.5
0.8 1.0
2
3
DorHD
5 6 789 10
I
15
I
20
Fig. 24-59 Curves of DulD vs. D or HD for estimating severity of quench (H) for quenching baths.
I
~.
Like PVP, aqueous sodium poly acrylate solutions are soluble up

to the boiling point of water. Their ability to reduce the rate of heat
removal is proportional to the molecular weight of the polymer and
its effect on solution viscosity. By varying the polymer molecular
weights, a whole family of quenchants can be designed to cover
applications ranging from water to oils.
The effect of polymer concentration and temperature for one of
the commercially available sodium polyacrylate quenchants is
shown in Figure 24-58. The cooling curves obtained with polyacry
late solutions can be almost linear with time, which is a result of the
extended A -phase cooling and reduced cooling rates during nucleate
boiling. This property of the polyacrylate quenchants allows their
use in hardening crack-prone parts made of high hardenability
steels. Cooling curves obtained with polyacrylate quenchants are
illustrated in Figure 24-58. Both the concentration of the polyacry
late in water and the water temperature can be varied to affect the
cooling rates in steel parts.
With increasing PA concentration and bath temperature, the cool
ing rate can be teduced to such an extent that many ferrous metals
do not transfonrl to martensite at all, but form bainite or fine pearlite.
The "nonmartensitic" quenching can be utilized for achieving lower
hardness values.
As with all polymer quenchants, sodium polyacrylate solutions
require agitation, the degree of which depends on the specific
application. As a general rule, a high degree of agitation is recom
mended for hardening operations.
The concentration control of the sodium polyacrylate quenchants
is based on kinematic viscosity. They also should be periodically
checked with other laboratory methods including cooling curve
analysis to monitor the influence of contaminations. This service is
usually provided by the quenchant supplier (40).
Salts. Salt baths are usually made of mixtures ofKN03, NaN03,

and NaN02 with small amounts of dissolved water. Water levels are
typically in the range of 0.25-2.0% and the quench severity in
creases with the water content. Salts can be used instead of oil, and
can operate at temperatures of 300-750 OF in austempering or
marquenching operations. Salt baths are controlled by periodically

measuring electrical conductivity, thermal conductivity, melting
point, weight loss on drying, and using certain titration techniques.
Quenching Variables. Quench facility operation involves the

careful analysis oftwo intimately interrelated factors. The first is the
type of quenchant and the operating conditions in use, and the
second is the hardenability of the particular alloy(s) being heat
treated. The hardenability is a function of the chemical composition
and grain size ofthe specific heat, as previously described.
Water is an excellent quenchant, but its heat removal charac
teristics are variable depending on the bath temperature and the
relative velocity between the water and the parts being quenched.
Water also can be a rather severe quenchant, so oils or polymer
water solutions often are used to moderate the heat removal rates.
Data on the effect of quenchant type, temperature, and quenchant
velocity on the Grossman number and film coefficient produced by
water, a fast oil, a conventional oil, a martemper oil, a 25% water
soluble polymer solution and air are presented in Table 24-10 (36).
Water at 80 to 90 OF can produce film coefficients of over 1000
Btulh ft2 OF depending on the extent of agitation. The film coeffi
cient decreases rather rapidly however, as the water temperature
rises or as the water velocity past the part decreases.
The heat removal characteristics of water at 130 OF are substan
tially reduced, especially at lower velocities. compared to 80 OF
water. The film coefficient of 130 OF water without agitation was
only 60 Btulh ft2 OF, but increased to approximately 730 Btulh ft2 OF
with a velocity of 150 fpm.
Thus, without agitation, water at 90 OF can remove heat approxi
mately 5 times faster than water at 130 OF without agitation. At 50
fpm, 130 OF water has less than 1/3 the heat removal capacity of9O
OF water at the same velocity. At a velocity of 100 fpm, the heat
removal capacity of 130 OF water is approximately 112 that of9O OF
water.
These data are perhaps sufficient to illustrate the variaeility of
water and water bond quenchants as a heat removal media. When
quenching large castings, the localized temperature and relative
24~32

2
2.0
50
H-YAWE {SEVERITY OF QUENCH l
I.B
5.00 STRONG BRINE QUENCH,

1.6
0.35
c
I
1.4
a::
1.2
ll..
40 E
E
VIOLENT AGITATION
2.00 POOR BRINE QUENCH,
a:::
30m
NO AGITATION
I .00
ll..
1.0
a:::
ffi O.B
I-
20~
~ 0
.6
::E
Cl
Cl
0.4
10
"'-HOT SALT AT 450 F (232" C)
POOR WATER QUENCH,

NO
0.70
AGITATION
STRONG OIL QUENCH,

VIOLENT
AGITATION
0.50 VERY GOOD OIL QUENCH,

GOOD
AGITATION
0.35 GOOD OIL QUENCH,

MODERATE AGITATION
0.2
0
0.20 POOR OIL QUENCH.

0
II
NO
AGITATION
12
DISTANCE FROM QUENCHED END

OF THE END - QUENCH HARDENAB1LlTY SPECIMEN,
CORRESPONDING TO CENTER OF ROUND BAR, SIXTEENTHS OF AN INCH
DATA ARE FOR

CENTER HARDENING
Fig.24-60 Grossman chart relating bar diameter. hardenability of steel, and severity of quench (H-value). See text for discussion of method
for using this chart to evaluate severity of quench (7).
velocity of water past parts have significant effects on heat removal

rates. The hardenability of the specific alloy and the heat removal
rate significantly affect the hardness that can be expected in
quenched parts.
Measuring Quench Severity. Methods for determining the

ability of a quenching solution or facility to extract heat can be
divided into two broad categories. The mechanical procedures for
evaluating quench severity usually employ hardness measurements
or etching procedures applied to quenched parts to determine the
depth ofmartensite formation in a particular alloy during quenching
under standardized conditions. Thermal techniques involve measur
ing the time required to cool between specified temperatures, deter
mining the rate of cooling in a standard specimen, or defining by
some other thermal means, the ability of a quenchant to extract heat.
Grossman Technique. The Grossman technique has been used for
many years to evaluate the quench severity "H" provided by a
quenchant. The use ofthis procedure is based on the observation that
with increasing bar diameter, the unhardened core diameter in
creases more rapidly with the use ofless severe quenchants. Deter
mining the Grossman "H" value involves quenching a series of bars
of different diameters (D) in a particular quenchant or quench
facility. The bars are then sectioned and the cross section of the
unhardened diameter (Du) determined by either hardness measure
ments or by chemical etching.
A graph on the appropriate scale is made of DuID for the several
quenched bar sizes (8). This DulD curve can be superimposed in
Figure 24-59 and matched with the best fitting curve. The HD factor
on the abscissae is the product of the Grossman H factor times the
bar diameter D. This value is read from the scale and divided by the
actual bar diameter to obtain the quench severity factor H. One
limitation on this procedure is that it may not be possible to match a
curve if there is a large difference between the bar diameter and the
depth of hardening (8,44,45).
The Grossman chart illustrated in Figure 24-60 can be used to
obtain H values for quench severity under operating conditions (38).
As an example of the use of the chart, the hardness at the center of

a 1.2 in. (30.5 mm) diameter bar quenched in still water (H = 1.0)
will be the same as the hardness 4116 in. (6.4 mm) from the
quenched end of the end-quench bar of the same alloy.
To determine the H value of a commercial quench, an end-quench
bar and cylinders of two different diameters are needed. The cylin
der diameters are chosen so that the center hardnesses fall on the
sloping part of the end-quench curve. The samples and the end
quench bar can be made from a wrought steel bar. The analysis is
selected to give a usable end-quench curve.
The two cylinders are quenched with a load of castings. In the
example in Figure 24-60 for still water quench, the hardness at the
center of the 1.5 in. (38.1 mm) diameter bar equalled that at 5116 in.
(8 mm) from the end of the end-quench bar. The hardness at the
center of the 1 in. (25.4 mm) diameter bar was equivalent to 3116 in.
(4.8 mm) from the quenched end of the end-quench bar. These two
points and a line joining them are plotted in Figure 24-60. The H
value of this quench is slightly over 1.0. The curves for oil and hot
salt were determined in the same way.
Rushman Technique. The Rushman method is a variation of the
bar quenching technique. Two bars having the same composition,
intermediate hardenability (typically 1141), and different diameters
are quenched in the facility of interest. The bars are then cut in two
and the hardness at the bar center measured. Using the bar diameter
and the equivalent Jominy distance based on hardness achieved
when the two bars were quenched, the data points are placed on a
Lamont diagram illustrated in Figure 24-61 (39,46). The use of this
procedure involves:
1. Quenching 2 bars (typically 1141) of different diameters (typi
cally 112 and 1 in.) in the quenchant or quench facility of interest
2. Determining the Jominy curve for the steel used in the bars
3. Comparing center hardnesses in the bar with the Jominy curve
and assigning a Jominy distance to the cooling rate at the bar
center
Fig. 24-61 (a) Location on

end-quenched Jominy har
denability specimen corre
sponding to the center of
round bars (30) (b) Location
on end-quenched Jominy
hardenability specimen cor
responding to the half-radius
position in round bars (30)
7.0
175
...~
5.0 (0.2)
2.0 (0.08)
1.5
1.0 (0.04)
0.70
150
5.0
125
lii
.~
0.50 (0.02)
.S
.5
4.0
::t:""
CD
E
0.20 (0.008)
75
I .
0.35
lii
100
(j)
-0
24-33
Iii
III
'0
50
2.0
25
1.0
(J)
0.25
0.50
0.75
1.25
1.50
1.75
Distance from water-cooled end, in.

(a)
250
1-
10.0
rlR = 0.5
9.0 1
8.0
200 I-
E
E
150
..:
.=
6.0 -
Qj
Q)
0.50 (0.02)
Q)
Q)
E 5.0
t1l
...
'6
t1l
'6
t1l
OJ
5.0 (0.2)
2.0 (0.08)
1.5
1.0 (0.04)
0.70
'=r.'LL
7.01
100
...
t1l
OJ
0.35
....I
g
I
.=
:i
4.0 -
0.20 (0.008) c0Q)

::J
3.01
//'// /
/'
cr
'5
~
50
"n 1
///////
.~
./
>
Q)
(/)
Ol~~------~--------~--------~--------~
0.5
1.5
.Distance from water-cooled end, in.
(b)
4. By comparison with Figure 24-61 or Table 24-6 an "H" value for

the quenchant can be assigned
Cooling Curves. Informative data can also be obtained by

measuring the actual cooling rates in a test specimen, and the
resulting data can be used to characterize the quenchant behavior
under a variety of conditions. 'I)'pical cooling curves were illus

trated in Figures 24-53 and 54. Generally, a specimen is machined
from a non-scaling non-transfonning material such as silver, nickel,
or an austenitic stainless steel. The test specimen is usually cylindri
cal with a diameter ranging between 112 to 2 inches or longer and a
length-to-diameter ratio greater than 4. Two positions in the speci-
il
24-34
'"
Table 24-11
Hardenability Qata on 4130

Low
specification
JII:
Chemical com~sition
High
Actual
composition
specification
Carbon(%)
Silicon(%)
Manganese (%)
Nicke1(%)
Chromiwn(%)
Molybdenum (%)
Copper(%)
Vanadium (%)
Boron(%)
0.250
0.150
00400
0.000
0.800
0.150
0.000
0.000
0.000
0.315
0.379
0.576
0.103
1.140
0.244
0.125
0.014
0.000
0.350
0.400
0.700
0.103
1.500
0.250
0.125
0.014
0.000
Alloy factors
Carbon
Silicon
Manganese
Nickel
Chromium
Mo1ybdenwn
Copper
Vanadiwn
0.l35
1.105
2.333
1.000
2.728
1.450
1.000
1.000
0.170
1.265
2.920
1.035
3.462
1.732
1.046
1.024
0.189
1.280
3.333
1.035
4.240
1.750
1.046
1.024
1.38
4.18
6.63
Ideal diameters (in.)

Martensite hardness (Rc)
M(99)
47.3
M(95)
43.9
M(90)
41.5
M(80)
38.2
M(50)
34.7
Transformation temperatures (oF)
A(3)
1507.9
A(l)
l350.7
B(s)
1216.5
M(s)
756.0
M(90)
570.6
51.2
47.9
45.5
42.0
38.3
1509.1
l363.3
1092.6
691.9
506.5
53.3
49.9
4704
43.9
40.0
1500.0
l370.8
1009.3
653.7
468.3
men, one near the surface and one in the center, are instrumented
with thermocouples to provide information on the cooling rates
actually achieved. The thermocouple and recording system used
must be capable ofmeasuring rapid temperature changes, especially
if a thermocouple is placed near the surface of a probe. This type of
equipment has been used to study quenchants, spray quenching,
agitation, bath temperature, and effects of surface condition on heat
extraction from metal.
Quench Factors. Jominy curves have been used to represent

hardenability of various grades of steel for many years, and the
Grossman "H" value has been used to provide information about the
ability of a quenchant to extract heat from parts. A more recently
introduced method for quantifying quenchant and steel interactions
is that of quench factors (40,47,48).
Quench factor analysis provides a single number that interrelates
the cooling rate as produced by the quenchant, the section thickness
of the part, and the transformation rate of the alloy as reflected by
the time-temperature-property (TIP) curve. The TIP curve is the
mathematical representation of the start of transformation curve and
reflects the amount of transformation required to reduce the as
quenched hardness or strength from the maximum value by a fixed
percentage. The alloy composition and quenCh conditions must be
controlled so the maximum quenCh factor corresponding to the
minimum acceptable as-quenched hardness is not exceeded.
Lower quench factor values are associated with higher as
quenched hardness values. The critical value of the quench factor
(Q) is the maximum value that will result in the desired hardness
value and can be defined in terms of the desired martensite content,

i.e. 99%, 95%, 90%, etc.
Quench Factor Calculation. Quench factors can be calculated
from time-temperature (cooling curve) data and the equation de
scribing the transformation kinetics of the alloy of interest (CT
function).
The CT function is usually described with an equation ofthe form:
CT = - Kl * Kz
*exp [ R*T(K4-1i
(K3*K~) 1* exp (KS)J
j R*T
(Eq5)
where CT = critical time required to form a constant amount of a new

phase or reduce the hardness by a specified amount (the locus of the
critical time values as a function of temperature forms the TfP curve),
Kl =constant which equals the natural logarithm ofthe fraction untrans
formed during quenching, i.e. the fraction defined by the TfP curve, K2
= constant related to the reciprocal ofthe number ofnucleation sites, K3
= constant related to the energy required to form a nucleus, K4 =
constant related to the solvus temperature, K5 =constant related to the
activation energy for diffusion, R = 8.3143 J/(OK mol), and T tempera
ture oK.
The constants Kl, K2, K3, K4, and K5 define the shape on the TIP
curve.
An incremental quench factor, q, for each time step in the cooling
curve is calculated using Eq 6:
~t
(Eq6)
q= CT
where q incremental quench factor, and t time step used in cooling

curve data acquisition.
The incremental quench factor, q, represents the ratio of the
amount of time a steel was at a particular temperature divided by the
time required for a specified amount of transformation, typically
1%, at that temperature.
The incremental quench values are summed over the entire trans
formation range between Ar3 and Ms to produce the cumulative
quench factor, Q, according to Eq 7:
T=Ms
~t
Q=l:q=LCT
(Eq7)
T=NJ
The cumulative quench factor reflects the heat extraction charac

teristics of the quenchant as a function of the quenchant type,
velocity, and temperature over the transformation range of the steel
being quenched. This factor includes section thickness effects on
cooling rate and the transformation rate of the alloy as a function of
temperature.
The calculated quench factor can be used to predict the as
quenched hardness in steel using the following equation:
Hp= Hmin + (Hmax - Hmin) exp (KIQ)
(Eq8)
where Hp = predicted hardness, Hmin = minimum hardness for alloy,

Hmax = maximum hardness for alloy, exp = base of the naturalloga
rithm, Kl = In (0.995) =-D.OO501, and Q = quench factor.
Hardenability of Cast 4130 Steel

Some calculated hardenability data on cast 4130 steel is presented
in Table 24-11 (47). The first column in Table 24-11 provides
composition and hardenability data on a cast 4130 steel assuming all
elements to be at the low end of the composition range. While it is
unrealistic for all of the alloy elements in a particular grade of steel
to lie on the lower specification limit, nevertheless the calculation is

made to illustrate the potential extremes that can be encountered.
The carbon range has been expanded from the normal AISI range of
0.28 to 0.33% to cover the range normally accepted in castings, i.e.
0.25 to 0.35% carbon. The second column provides data on a cast
steel composition actually tested, and the third column provides data
assuming all elements to be at the high end ofthe composition range.
The composition data is followed by calculated alloy factors and
ideal diameters for the low, actual, and high end compositions.
The next data group in Table 24-11 contains calculated martensite
hardness data. At 99% martensite, a 4130 steel containing 0.25%
carbon can be expected to have a maximum hardness of Rockwell
C 47.3, and steel containing 0.31% carbon is expected to have
maximum martensite hardness of Rc 51. As the martensite content
decreases, the as-quenChed hardness decreases to a value ofRc 38.3
at 50% martensite. The Ms temperature for the steel was calculated
to be 692 OF, and the temperature at which the steel had transformed
to 90% martensite was 506 OF using equations published by Atkins
(3).
Calculated and experimental Jominy hardness values for the low
specification composition, the actual composition, and the high
specification composition 4130 are presented in Table 24-12. The
predicted Jominy hardness for the low end composition, the actual
composition, and the high composition are illustrated in Figure
24-62 (47).
The solid lines represent the calculated (predicted) Jominy hard
ness distribution using the methods, procedures, and coefficients
described in Caterpillar Tractor Company Specification 1E38. The
stars on Figure 24-62 illustrate the experimental Jominy hardness
measurements on the steel after austenitizing and quenching from
1575 oF. There was reasonably good agreement between the pre
dicted Jominy hardness and the measured hardness in this alloy.
Table 24-12
MTL#25
J
position
Calculated Jominy Hardness on 4130
specification
(Rc)
47
1
2
3
4
44
6
7
8
9
38
28
23
21
20
19
18
10
17
12
14
16
18
20
24
28
32
16
15
15
14
13
12
12
11
Actual
composition
(Rc)
Wgh
specification
(Rc)
51
51
51
51
50
49
45
53
53
53
53
53
53
53
53
53
53
53
51
50
50
49
44
42
41
38
37
34
33
32
30
29
27
46
46
44
E~rimental hardness data
TTP Curves for Cast 4130 Steel
position
Hardness
(Rc)
1
2
3
4
52
52
51
50
50
49
47
45
42
42
39
38
36
35
34
5
6
Most published TTT curves were developed for specific steel

compositions and do not take into account the possible variations in
composition and shifts in TTT curve position that can occur within
the allowable composition range. Using quenCh factor analysis it is
possible to calculate the start of the transformation curve for a steel
where a modest amount of data exists.
The approximate TI'P curves have been developed for the upper
and lower limit compositions of cast 4130 steel given in Table
24-11. The shift in "C curve" with composition is clearly evident in
Figure 24-63 (47). These "C curves" are plotted for times to 1000
8
9
10
12
14
16
18
20
24
28
32
70
65
60
33
31
30
1600
1400
~ 55
~ 50
LL
1200
III
1000
45
~ 40
~a.
"0
24-35
35
::r: 30
t!
25
20
400
200
ol~__~~__~~--~--~--~
15
10
0
6 8 10 1214 16 18 2022 242628 30 32

Jominy distance, 16th's
Fig. 24-62 Calculated and experimental Jominy data on cast 4130

steel
10-1
100
Critical time (sec)

Fig. 24-63 TIP curves for cast 4130 steel. (a) Low specification
composition, (b) Actual composition, and (c) High specification
composition
.,'
"
24-36
Part 58: processing Properties
Table 2413 Measured and Predicted Hardness at

17 Positions in Water Quenched 4130 Steel
60
54
Cooling
rate
Casting
at1300F
(OF/sec)
position
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Quencb
fador
205.4
121,7
72,1
52.9
38.8
31,3
23.8
19.9
17,0
12.7
9.8
8.1
6.8
5.9
4.6
3.9
3.3
Predicted
bardness
(Rc)
Measured
48
hardness
(Rc)
42
5L8
50.9
49.9
48.8
47.7
52
51
50
50
49
47
45
42
42
39
38
36
35
34
33
31
30
7,2
12.4
18.5
25.6
33.1
41.5
50.1
58.7
67.3
86.8
103.1
119.1
134.8
150.1
175.3
197.6
215.6
46.4
45.2
44.1
43.0
40.7
38.9
37.3
35.8
34.5
325
31.0
29.9
36
<>
a:
30r
o Measured hardness
Predicted hardness
24
18
12
1 Cast 4130 MTl25J 20-Aug-90 13:46:36
0
10-1
102
10 1
100
1Q3
Quench factor
Fig. 24-64 Correlation between predicted hardness and measured
hardness at 18 locations in a quenched part (Mt1 25)
(a)
(b)
100X
250X
Fig.24-65 (a) Ferrite-pearlite structu re (light and dark constituents, respectively) of nonnalized 0.2% C steel. (b) Martensite structure of quenched
and tempered Ni-Cr-Mo (8635) steel
(a)
(b)
100X
100X
Fig. 24-66 (a) Ferrite-pearlite structure of 1 in. (25.4 mm) thick plate casting of 0.20% C steel obtained by normalizing. (b) Same as in (a), but
finer, obtained by quenching and tempering
Representative Properties:
UTS
A
B
YS
BHN
ksi
MFa
ksi
MFa
El
%
149
163
77
531
558
51
57
352
393
31
27
81
RA
%
OF
60
-45
-65
61
RT
NDTT
ft Ib
-43
-54
80
100
108
136
Cv - Im~act Enel1O'
OF(-18C)
ftlh
J
65
78
88
106
-(i() OF ( -51C)
ftlb
28
29
J
38
39
"
24-37
Surface
%T
Center
100x
500x
~ "',>"J.
Fig. 24-67 The ferrite-pearlite structure of a quenched and tempered 4 in. (102 mm) thick, A-216-WCC type, carbon steel plate casting
seconds (16.7 J;Irinutes) rather than the usual 1,000,000 seconds (20
days) used in many published diagrams. Transformations that occur
over a 20 day period are of little interest in heat-treating operations.
At 1100 OF, the transformation begins in about 0.15 second in the
low specification composition, in one second in the actual compo
sition, and after about 3 seconds in the high specification composi
tion.
The "e curves" illustrate the shift in the start of transformation
with alloy content, and the mathematical expression describing the
curves allows hardness predictions to be made under a wide variety
of quenching conditions. Some correlations between quench factor
and predicted hardness and measured hardness in cast 4130 are
presented in Table 24-13 and graphically illustrated in Figure 24-64.
The solid line in Figure 24-64 represents the predicted hardness as
a function of quench factor, and the data points are measured
hardness values at locations in a quenched part where cooling curves
were available.
Microstructure and Properties

The type of metallurgical structure which produces a given heat
treatment depends to a large extent on alloy content, quench prac
tice, section size being heat treated, and the location within a given
section.
A ferrite-pearlite structure will form during normalizing most
carbon steels and many low alloy steels (Figure 24-65A); martensite
forms during quenching provided the cooling rate is high enough for
the particular alloy composition (Figure 24-658).
Even though quenching may not produce martensite, the resulting
structure is usually finer and has superior toughness compared to air
cooled parts. These differences are illustrated in Figure 24-66A and
8 for low hardenability carbon steeL
The location within a given component, or section, may: have an
effect on the structure and properties depending on hardenabiJity of
the particular alloy. For carbon steel (Figure 24-67) minor differ
"
,-;)'
24-38 Part 58: Processing Properties
(a) 250X
(b) 250X
Fig.24--68 (a) The martensitic microstructure at the surface of a quenched and tempered Ni-Cr-Mo (cast 8635) 17 in. (432 mm) thick gear
blank. (b) The acicular, ferrite-pearlite structure of the casting in A-at the center of the 17 in. (432 mm) thick section
Representative Properties:
illS
A
YS
EI
RA
ksi
MPa
ksi
MPa
BRN
160
110
1103
146
1007
14
2
38
2
345
280
758
HOT UPSET
(DOLING NON UNIFORMLY
..
(
Distortion and cracking during quenching often limits the severity

of the quenchant and equipment that may be used. A more severe
quench produces martensite to a greater depth (with a steel of given
hardenability), but it also increases the likelihood of cracking.
Distortion is a result of three general phenomena:
UNIFORMLY COLD
LONG - COMPRESSION
41
Residual Stress Distortion and Cracking
SHRINKING
SHORT - TENSION
30
4
tively. The toughness, in tenns of Nil Ductility Transition Tempera

ture, NDTI, was -40, -30, and -20 of (-40, -34, and -29C),
respectively (49).
Quite pronounced microstructural and mechanical property gradi
ents can be expected when parts are quenched and do not through
harden because of insufficient hardenability. An example involving
a 17 in. (432 mm) thick gear blank of cast 8635 steel that was
quenched and tempered is illustrated in Figure 24-68. Near-surface
areas were martensitic while a structure of acicular ferrite, the
pearlite was observed at locations 8.5 in. (216 mm) from the surface.
The gradient had relatively small effects on strength, but major
changes in ductility and toughness as shown by the data in Figure
24-68.
UNIFORMLY HOT
C. - Iml!act Energy at RT
ftlb
J
Fig.24-69 Warpage caused by nonuniform quenching of a 30%

nickel steel block (10)
ences are found in the ferrite-pearlite structure of a4 in. (101.6 mm)
thick A216-WCC type carbon steel plate that was quenched and
tempered as illustrated in Figure 24-67. The hardness at the surface,
114 T. and' center locations was BHN 163, 142, and l37, respec
1. Warping during heating or during holding at the austenitizing

temperature
2. Thennally induced deformation caused by high thermal gradi
ents between thick and thin sections and from surface to center
of a part
3. Transformation induced stresses caused by the volumetric
changes, especially when austenite transforms to martensite
Warping is generally considered to be the result of nonuniform

heating or deformation caused by excessive weight on the part.
Warping can sometimes be reduced by lowering the rate of heating
to minimize thermal gradients during heating. Warping can also be
~,
AI
I"
UlltFOIMLT NOT
3.000" (COLD)
~o
L
DOUN'
ann InllG MOT UP'SrT
B
"
~I
"7:
.. ~
."l
c
Ie
SMIIHS .DOGS"
.j
1--.-' .
I
.J 1" (0)
tIII"--....
r ....----------------- ... ,
~,
I
~
I;;
;;:
JA'
,I
CDlIllllS mlS"
. . . . . . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ '"
-i-,
' __ '4'
24-39
,ti
~~I
i;1
~.'1
High Tempera/lin!
Transformation .... ~V
,,
1::
~O.
t::
ct
~f/A~
~
~0.6
~<f./
tfr
,~~
,,'
~
~O.
:S
O.
'/ /
. .X'Ms /
#'
-0.1
'"
\V
A
.~~.
~'
... ... ~~
MA(HINIIIG TH[ SHEU Off IlMOVlS ltlSIU

STiUS '10M COl( AlLOWlIIG n TO SHOIUIl
Fig.24-70 Effect of quenching on 30% nickel steel cylinder-water
quenched from 1425 of (774 0c) (10)
~/J
d'
~
c,9/
~~,
I'
!
.
!
.
.~
16
i~'
!'
~
~.
1t
f-
reduced by changing the racking procedure to reduce the amount of

stress on a part and by supporting parts properly on the tray or rack.
Thermally induced distortion is a result of nonuniform thermal
contraction during cooling. Warping during nonuniform cooling of
a part is schematically illustrated in Figure 24-69(A). Assume that
the bar in Figure 24-69(A) was initially at a uniformly high tempera
ture. If the bar is quenched nonuniformly principally from one side,
as illustrated in Figure 24-69(B), the more rapidly cooled side will
contract quicker and at a higher rate than the more slowly cooled
side. Since the rapidly cooled side becomes shorter and stronger as
it cools, it causes plastic deformation to occur in the hot side, which
has a lower flow stress. The deformation is followed by cooling and
contraction on the more slowly cooled side.
When the part finally cools to a uniform temperature, the part is
warped in the opposite direction to the initial warpage, as illustrated
in Figure 24-69(C), with the slowly cooled side being shorter and
the surface being concave on the more slowly cooled side. The
plastic deformation on the hot side results in compressive stresses
on the rapidly cooled side and tensile stresses on the more slowly
cooled side (SO).
Deformation resulting from nonuniform thermal contraction is
also illustrated in Figure 24-70. If a part is initially uniformly hot, as
illustrated in Figure 24-70A, and is rapidly quenched, the outer
surface shrinks while the center is still relatively hot This process
puts the outer surface into tension and the inside into compression,
causing plastic flow in the softer core (Figure 24-70B). As the center
of the part cools and the temperature reaches a uniformly low value.
thermal contraction in the center of the bar occurs, which reverses
the stress state and places the center in tension, as illustrated in
Figure 24-70C. Any machining on the part will partially relieve the
stress resulting in distortion (Figure 24-70D).
Fig. 24-71 Dilatometric studies show that when cooling a steel

from a high temperature it experiences reversals of its dimensional
contractions which vary, depending on the quenching rate (12)
0.014
0.012
E
0.010
E 0.008
c: 0.006
.~
i'i5
0.004
0.002
0.6
0.8
Carbon, %
1.4
Fig. 2472 Linear expansion in steel after quenching to produce a

fully martensitic structure
Deformation also occurs frequently in steel parts being hardened
partly as a result of the volumetric changes associated with marten
site formation. The expansion that occurs on heating 4340 steel from
room temperature to 1600 OF is illustrated in Figure 24-71 (40).
When the steel is rapidly cooled, the austenite transformll;tion tem
perature is suppressed to a relatively low value where martensite
begins to form. As martensite forms, there is a volumetric expansion
"
iii"
24-40
Table 24-14 Specific Volume of Phases Present in Carbon Tool Steels

Range or
carbon
(percent)
Phase or phase mixture
Austenite
Martensite
Ferrite
Cementite
Epsilon carbide
Graphite
Ferrite plus cementite
Low-carl:lon martensite plus epsilon carbide
Ferrite plus epsilon carbide
Om2
Om2
OmQOO
~~
~~
100
Om2
Om2
Om2
Calculated specific volume

at 68 of,cm 3/g
0.1212 +0.0033 (%C)
0.1271 +0.0025 (%C)
0.1271
0.1300.001
0.140 0.002
0.451
0.1271 +0.0005 (%C)
0.1277 + 0.0015 (%C-0.25)
0.1271 + 0.0015 (%C)
Size Changes on Hardening Carbon Tool Steels

Reaction
Spheroidite-austenite
Austenite-martensite
Spheroidite-martensite
Austenite-lower bainite(a)
Spheroidite-Iower bainite(a)
Austenite-aggregate of ferrite and cementite(b)
Spheroidite-aggregate of fenite and cementite(b)
Volume change
Dimensional change
(percent)
in.fm.
-4.64+2.21 (%C)
4.64-0.53 (%C)
1.68(%C)
4.64 - 1.43 (%C)
0.78(%C)
4.64-2.21 (%C)
...{).0155 + 0.0074 (%C)

0.0155-0.0018 (%C)
0.0056 (%C)
0.0155 -0.0047 (%C)
0.0026(%C)
0.0155 - 0.0074 (%C)
(a) Lower bainite is assumed to be a mixture offerrite and epsilon carbide. (b) Upper bainite and pearlite are assumed to be mixtures of fenite and cementite.
IJ"
ANNEALED
\.
D.
.1
3.000"
WATER QUENCHED 142soF.
(a)
HARD-]]O/o~
TOUGH.23%
J,..
(b)
.300"
3.000"
.. I
ROCKWELL (68.
WET GROUND TO .300"D.
I:
~-----~--~---------~--~~
'"( TOUGH CORE-ROCKWEll (-46
.300"
,
.~
.. --------------- - - - -- I
(c)
I..
CONTRACTS .008"
.. I
Fig. 24-73 Strains produced by hardening a cylinder of 1.10%

carbon tool steel (10)
in the metal that results in the ftnished part dimensions being larger
than the initial dimensions. The volumetric expansion associated
with martensite funnation as a function of carbon concentration in
steel is illustrated in Figure 24-72 (36).
When rapid cooling a through hardening steel, the outer surface is
quenched to martensite and the subsequent contraction of the core
produces surface compression and center tension, as illustrated in
Figure 24-73 (50). This ftgure schematically illustrates a bar that
was initially 3 in. long and 5/8 in. diameter. Water quenching from
1425 OF caused expansion in the center of the bar, resulting in a

0.0015 in. increase in diameter. If the hardened surface of the bar
were ground to a fmal diameter of 0.3 in. to relieve the internal stress
the bar would be 0.008 in. shorter than the as-quenched length. A
wide variety of similar experiments have been conducted to demon
strate that the stresses developed during quenching and martensite
formation can result in signiftcant defonnation of steel parts.
The fonnation of other phases can also contribute to distortion
because of volumetric change associated with speciftc transfonna
tions. The speciftc volume of several phases present in carbon tool
steels is presented in Table 24-14 (51).
In general, the amount of distortion occurring during quenching
depends on the size and shape of the part. the part composition, and
the characteristics of the quenchant employed. Parts with section
size ratios greater than 1 to 4; large parts with relatively thin cross
sections; and parts containing slots, keyholes, drilled holes, or
grooves cause problems because of the difficulty of achieving uni
form heating and cooling rates. Some of these difficulties are illus
trated in Figure 24-74 (SO).
Distortion occurring in a track shoe, because of the presence of a
lightening groove, is illustrated in Figure 24-74a. Flanges and
bosses also create conditions making it difficult to achieve uniform
cooling rates, as illustrated in Figure 24-74b. The thin, large diame
ter flange is difficult to cool at a rate consistent with the bar, and
almost invariably produces quench cracks.
Keyholes and grooves often cause a shaft to distort during heat
treating, and keyways with sharp comers often initiate cracks. The.
keyway cracking problem can be minimized to some extent with
generous radii, as illustrated in Figure 24-74<:.
The steel composition can also have a signiftcant effect on warp
ing and distortion. Higher carbon alloys expand more during
martensite formation and those with higher hardenabilities are prone
to more distortion than lower hardenability alloys. Similarly, high
austenitizing temperatures and low martensite start (Ms) tempera
tures aggravate distortion.
The characteristics of the quench procedure, particularly condi
tions promoting high rates of cooling through the transformation
range, aggravate distortion and cracking. Rapid quenching to a
'I.l'
,.
(;':
ti
c
:i2
(J
I!!
(J
..c:
(J
l00~~
IcKJ
80/
60
cQ)
:::l
eT
'0
Distortion due 10 lightening groove
c
0
(a)
j,1
,;~
';l
,I
0.75~7.621
40 ~ C-M" _I,
o Ni, Cr, Mo
20
steels
I!!
u..
~16.00in~300d
24-41
lam
(a)
Ii
0-:Cl
100
:i2
80
..c:
(J
60
C-Mn steels
o Ni, Cr, Mo
steels
(J
cQ)
(b)
:::l
eT
'0
c
0
15
40
Not fully
martensitic
20
(!!
u..
0
0.4
.L_ _
0.5
0.6
0.7
(b)
Fig.24-75 (a) Relation between the fraction of the test specimens

that quench cracked and Ms temperature. (b) Relation between
fraction of the test specimens that quench cracked and Ceq (60)
The tendency for cracking tends to decrease as the martensite start

temperature increases. The martensite start temperature can be ap
proximated by using the following equation (51):
(c)
Ms=S21-3S3C-22S Si-24.3Mn-17.3 Ni-17.7 Cr-2S.8Mo
Fig. 24-74 (a) Ttlis track shoe of SAE 1037, water quenched.
distorted up to 0;010 in. due to lightening of groove. Redesign to
remove grooves (which improved uniformity of section) reduced
distortion to 0.063 in. max (1). (b) In this design, the flange will
invariably crack during quenching. (c) Grooves will cause a shaft to
warp in heat treating. A keyway with sharp corners often initiates
cracks in quenching. The problem is avoided with a radius, as
shown.
I~
\:c
~,: ,
\",
4;;
temperature just above the Ms, followed by slow cooling through

the martensite formation range, where volumetric expansion occurs,
minimizes distortion. Uniform rates of heat removal per unit of
surface area also tend to minimize distortion (50,51,52,53).
Cracking occurs for the same reasons as distortion, but cracks
form when the localized strain exceeds the failure strain in the
material. Usually the cracks follow austenite grain boundaries, but
cracking does not appear to be related to the austenite grain size.
(Eq9)
The correlation between the occurrence of quench cracks and the
Ms temperature calculated using Eq 9 in two steels is illustrated in

Figure 24-75a (7).
Cracking has been also related to composition using the equation:
Ce<j '" C + MnlS + Mo/S + Cr/lO + NilSO
where ~ is the carbon equivalent. Quench cracks were prevalent at

carbon equivalent values above 0.525, as illustrated in Figure 24-75b.
Quench cracking also becomes more prevalent as the ideal diame
ter increases (53,54,55). The danger of cracking a part is at a
maximum when the martensite skin thickness is about 50-60% of
the cross sectional area (53).
Cracking is commonly associated with nonuniform quenching.
Uniform surface compressi ve stresses produced by a uniform rate of
heat removal help suppress cracks. However, as the quench severity
increases, maintaining a uniform rate of heat extraction over the
entire part surface becomes more dufficult (26,27,56,57).
S"
24-42'
""l
';"",.'
REFERENCES
31. S.E. Feldman, ''The Minitech Computerized Alloy Steel Informa
tion System,".Peoc. of Metallurgical Soc. of AIME Heat Treatment
1. Troiano and Greninger, Metal Progress, Vol 50, 1946, P 303
CommitteelASM Activity on Phase Transformation Symposium,
2. Grange and Stewart, ''The Temperature Range of Martensite For
Chicago, IL, Oct 24-26, 1977, P 546-567
mation," Trans. AlME 167, New York, 1946, p 467
32. 1M. Hodge and MA Orehoski, "Relationship Between Har
3. M. Atkins, Atlas ofContinuous Coolin8 Transformation Diagrams

denability and Percentage of Martensite in Some Low-Alloy
for Engineering Steels, ASM International, Metals Park, Ohio,

Steels," Transactions of the American Institute of Mining and
1980
Metallurgical Engineers, Vol 167, 1946, p 627
4. Atlas of Isothermal Transformation and Cooling Transformation
33. MA Grossman and E.C. Bain, Principles ofHeat Treatment, 5th

Diagrams, ASM International, Metals Park, Ohio, 1979
ed., American Society for Metals

5. U.S. Steel, Atlas ofIsothermal Transformation Diagrams, 3ed ed.
34. "Hardenability Calculations and Data," Steel Casting Design. En
Pittsburgh, PA
gineering Data File, Section 12, Steel Founders' Society of Amer
6. G. Krauss, Principles ofHeat Treatment ofSteels, ASM, 1980
ica, Rocky River, OH

7. Sourr:e Book on Heat Treating, Vol I, Materials and Processes,
35. "Calculating Hardenability Curves from Chemical Composition,"
ASM,1975
Metal Progress, June 1977, p 114
'
8. Siebert, Doane, and Breen, The Hardenability ofSteels, ASM, 1977
36. C.E. Bates, "Predicting Properties and MinicnizilIg Residual Stress

9. K. Sachs, "Presentation of Transformation Data," Heat Treatment
in Quenched Steel Parts," Journal ofHeat Trealing, Vo16 (No. I).

p2745
ofMetals, Voll, 1974, p 43-57
37. MA Grossman, Elements of Hardenability, American Society for
10. K.E. TheIning, "ccr Continuous Cooling Transformation Dia
Metals, OH, 1952
grams with Natural Cooling," Scandinavian Journal ofMetallurgy,
38. "Quenching of Steel," Metals Handbook, Vol 2, 8th ed, American
1978, p 252-263
Society for Metals, 1964, p 15
11. Climax Molybdenum Company, Austenite Transformation Kinet

39. J.L. Lamont, "How to Estimate Hardening Depth in Bars," Iron
ics ofFerrous Alloys,Voll and 2, Greenwich, CT

Age, Vol 152, Oct 14, 1943, P 64-70
12. The Making, Shilping, and Treating ofSteel, U.S. Steel Corporation,
40. C.E. Bates, G.E. Totten, and R.L. Brennan, "Quenching of Steel,"
7thed.,1957
ASM Metals Handbook, 10th ed., Vol 4, ASM International, Aug
13. Nickel Alloy Steels Data Book, 3rd ed, International Nickel Com
1991,p67-120
pany, Inc., Bulletin A, Section 2
41. S.B. Lasday, "Metal Quenching with Oils and Synthetic Medi.a,"
14. R.A Granger and N.M. Steward, ''The Temperature Range of
Industrial Heating, Vol 43, 1976, p 8-19
Martensite Formation," Trans., Vol 167, AIME, 1946, P 467
42. N.F. Squire, ''Modem Quenching Oils and Techniques for Foundry
15. R.C. Voigt, M. Blair, and 1. Rassizadehghani, "High Strength Low

Heat Treating," Modem Casting, p 30-34
Alloy Cast Steels," Proc. 1990 ASME Pressure Vessels and Piping
43. "Heat Treatment Problems Solved by New Medium," Metal Mbrk
Conference, Nashville, TN, 17-21 June 199O,p 147-154
ing Production. Vol 111 (No.4), 1976, P 64-65
16. R.C. Voigt, "AnalYSis of Intercritical Heat Treatment of Cast

44. MA Grossman, Principles ofHeat Treatment, ASM, 1935
Steels," J. Heat Treat, (No.7), 1989,p95-105

45. B.D. Wakefield, "Agitation Quenching: A Good Way to Avoid High
17. R. Molnar, "A Literature Survey on Certain Aspects of Austenitic

Alloying Costs," Iron Age, Oct 1, 1970, P 54-55
Manganese Steels," Steel Castings Institute of Canada, June 1974

46.
W.F.
Rushman, "How to Determine the Quench Severity in Oil and
18. G. Coletta, et al., "Contributions a I'Etude des Transformations des
Salt Baths," Metal Progress, Dec 1963, P 91
Austenites a 12 pour cent Mn," Revue de Metallurgia, Vol 54 (No.
47. CE. Bates, "Effects ofComposition and Quenching Practice on the
6), June 1957, p 433-486
As-Quenched Hardness of Cast 4130 Steel," Proceedings of 44th
19. K.J. Irvine, DJ. Crow, and F.B. Pickering, ''The Physical Metal
Steel Founders' Society of America Technical and Operating Con
lurgy of 12% Chromium Steels," Journal of the Iron and Steel
ference, Chicago, IL, Nov 8-10, 1990, P 1-38
Institute, Aug 1960, p 386
48. C.E. Bates, T. Landig, and G. Seitanakis, "Quench Factor Analysis:
20.1. Beech, "Metallurgy of 13% Cr Steel Castings," Foundry Trade
A Powerful Tool Comes ofAge," Heat Treating, Vol 85, Dec 1985,
Journal, Nov 3, 1960, P 569
p 13-17
21. P. Shahinian and J.R. Lane, "Modifying 12% Cr Steel for Better
49. Steel Castings Handbook, 5th ed., Steel Founders' Society of Amer
Properties," Iron Age, Oct 1956, P 106
ica, 1980, Chapt 21, P 15
22. EE. Thurn, Ed., The Book of Stainless Steels, 2nd ed., American
50. Palmer. et al., Tool Steel Simplified, 4th ed, Chilton, 1978
Society for Metals, 1935
51. R. W. Edmonson, Dimensional Changes in Steel During Heat Treat
23. RD. Solomon and T.M. Devine, Jr., "Duplex Stainless Steels--A
ment," Metal Treating, Vol 20 (No.6), 1969-1970, P 3-19
Tale of Two Phases," ASM MetalslMaterials Technology Series
52. D.M. Nakhimov and D.I. Bron, "Deformation of Steel During
8201-089
Quenching." Metal Science and Heat Treatment ofMetals, 1966, p
24. E. Hombogen and R.C. Glenn, "A Metallographic Study ofPeecipi
540-545
tation of Copper from Alpha Iron," Trans. AlME, Vol 218, Dec
53. N.I. Kobasko. "Intensification of the Quenching Process in Steel,"
1%0, p 1064-1069
Metal Science and Heat Treatment ofMetals, 1970, p 153-155
25. E Hombogen, "Precipitation from Binary Substitutional Solid
54. T. Kunitake and S. Sugisawa, "Quench Cracking Susceptibility of
Solutions of Alpha Iron," Proc. Metallurgical Society AlME, Vol
Steel," Sumimoto Searr:h, 1971, p 16-25
28, Cleveland, OH, Oct 1965, P 1-69
55. M. Herenyi, Handbook of Experimental Stress Analysis, John
26. R.F. Kerr, ''Thinking Through to Successful Heat Treatment," Met
Wiley, 1950
alEngineering Quarterly, 1971,p 1-4
56. Bogatyrev. et al., "Influence of Cooling Rate on the Formation of
Cracks During Quenching ofSteel," Metal Science and Heat Treat

27. Sourr:e Book on Heat Treating, Vol II, Production and Engineering
ment ofMetals, 1%7, p 255-261
Practices, ASM, 1975
57. Golovin, et al., "Cooling Conditions and the Formation of Quench
28. ASTM A-255, Jominy Test

29. "Hardenability Slide Rule," United States Steel Corp., 1970
Cracks," Metal Science and Heat Treatment of Metals, 1%9, p
30. Stecal computer program from ASM International, OH

912-913

Hardenability and Heat Treatment

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Hardenability and Heat Treatment

Hochgeladen von

Copyright:

Verfügbare Formate

.-.

-+t~wRiO;$! t I till' "i19

?H"ij'td n 'e: :alii it'll