Materials Science and Technology

Comparison of thermo-mechanical stresses produced in work rolls during hot and cold
rolling of Cartridge Brass 1101
--Manuscript Draft-Manuscript Number:

MST11289R2

Full Title:

Comparison of thermo-mechanical stresses produced in work rolls during hot and cold
rolling of Cartridge Brass 1101

Article Type:

Research Paper

Keywords:

Hot Rolling; Numerical Simulation; Thermo-Mechanical Hoop Stress; Shear Stress;

Cold Rolling; Compression Testing

Corresponding Author:

Faisal Qayyum, M.S.

University of Engineering and Technology, Taxila
Jauharabad, Punjab PAKISTAN

Corresponding Author Secondary

Information:
Corresponding Author's Institution:

University of Engineering and Technology, Taxila

Corresponding Author's Secondary

Institution:
First Author:

Faisal Qayyum, M.S.

First Author Secondary Information:

Order of Authors:

Faisal Qayyum, M.S.

Masood Shah, PhD
S. Manzoor, PhD
Mohsin Abbas, M.S.

Order of Authors Secondary Information:

Abstract:

Thermo-mechanical stresses play an important role in defining the life of the work roll
used in hot rolling process. In this research temperature dependent mechanical
properties of cartridge brass are determined experimentally using high temperature
compression tests at different temperatures and strain rates. Real life measurements
are made from a brass rolling mill as input data for the simulation boundary conditions.
Hot rolls are made of AISI H11 hot work tool steel. Temperature dependent mechanical
properties of AISI H11 steel are used. Thermal and mechanical stresses produced in
the work rolls during hot rolling process are predicted using a thermoplastic finite
element approach in the ABAQUS StandardTM software. Hot rolling is compared with
cold rolling to determine the effects produced on the work rolls. A criterion is introduced
to compare the severity of stresses produced on the rolling surfaces in case of hot
rolling and cold rolling based on the yield stress of the roller material for different
temperatures. A method for separating thermal and mechanical stresses in the
simulation is also described.

Powered by Editorial Manager and ProduXion Manager from Aries Systems Corporation

Manuscript (NOT PDF FILE AT REVISE STAGE)

Click here to download Manuscript (NOT PDF FILE AT REVISE STAGE): Simulation of Hot Rolling of Brass V17.docx

Comparison of thermo-mechanical stresses produced in work rolls during

hot and cold rolling of Cartridge Brass 1101
Faisal Qayyum*; Masood Shah*; S. Manzoor*; Mohsin Abbas*
*Mechanical Engineering Department, University of Engineering and Technology, Taxila,
Pakistan
Abstract: Thermo-mechanical stresses play an important role in defining the life of the
work roll used in hot rolling process. In this research temperature dependent
mechanical properties of cartridge brass are determined experimentally using high
temperature compression tests at different temperatures and strain rates. Real life
measurements are made from a brass rolling mill as input data for the simulation
boundary conditions. Hot rolls are made of AISI H11 hot work tool steel. Temperature
dependent mechanical properties of AISI H11 steel are used. Thermal and mechanical
stresses produced in the work rolls during hot rolling process are predicted using a
thermoplastic finite element approach in the ABAQUS Standard TM software. Hot
rolling is compared with cold rolling to determine the effects produced on the work
rolls. A criterion is introduced to compare the severity of stresses produced on the
rolling surfaces in case of hot rolling and cold rolling based on the yield stress of the
roller material for different temperatures. A method for separating thermal and
mechanical stresses in the simulation is also described.
Keywords: Hot Rolling; Numerical Simulation; Thermo-Mechanical Hoop Stress; Shear
Stress; Cold Rolling; Compression Testing.
NOTATIONS
cr
cs
kr
ks
q
r
T
t
x
y

r
s
Th-Mec
Mec
Th

Work-roll specific heat

Strip specific heat
Work-roll thermal conductivity
Strip thermal conductivity
Rate of heat of deformation
Radial direction
Temperature
Time
Rolling direction
Thickness direction
Peripheral direction
Work-roll density
Strip density
Thermo-mechanical hoop stresses
Pure Mechanical hoop stresses
Pure thermal hoop stresses
Percent ratio of hoop stress to yield stress in work rolls
Hoop stress
Yield stress
Yield strength of H11 Tool Steel at 25o centigrade
1

mH11

Slope indicating reduction in yield strength of H11 with an increase in temperature.

I. INTRODUCTION
Materials used in the work roll for hot rolling process undergo severe thermo-mechanical
loading during the process cycle. In the past most of the work has been done to determine the
formability of the work piece, rather than the effect of thermo-mechanical loading on the
rolls. A brief summary of the work already done on the rolling process is discussed. ZoneChing and Chang-Cheng1 developed a three-dimensional, control-volume based finitedifference model for predicting the temperature distribution and the thermal expansion in a
rotating cylindrical work roll. They took into account all boundary conditions that might be
encountered during the actual hot rolling process i.e. heat convection to surrounding air and
cooling water, boiling phenomenon of the cooling water and radiation losses, etc. Results
obtained from their mathematical model were found close to experimental results. They
found out that greatest amount of thermal expansion of the work roll along the radial
direction in the middle steady stage of rolling is located at a central position closer to the exit
in the contact region. CG Sun, et.al.2 presented an integrated finite elementbased model for
the prediction of the steady-state thermo-mechanical behavior of the roll-strip system and of
roll life in hot strip rolling. The model comprised of basic finite-element models, which are
incorporated into an iterative-solution procedure to deal with the interdependence between
the thermo-mechanical behavior of the strip and that of the work roll, which arises from rollstrip contact, as well as with the interdependence between the thermal and mechanical
behavior. Results showed that the model was effective for the prediction of the detailed
aspects of the effect of the process parameters and, consequently, for the successful
exploration of the process conditions that can enhance the roll performance. L. M. Galantucci
and L. Tricarico3 used commercial FEM software ANSYS 5.0 to simulate the hot rolling
process to estimate temperature distribution and stress strain fields in roller and plate during
hot rolling process. Main variables and characteristics of rolling process were expressed in
parametric form. The results of simulation were verified using already published data. Martha
P. Guerrero, et.al.4 developed three models (two using the finite-difference method and one
integrating the heat flow to the roll) to calculate the transient heat flow into the work rolls,
and a fourth model was developed to compute the temperature distribution in the work roll at
steady-state condition. It was concluded that each model has its own value, and that all of
them should be used when a comprehensive analysis of the rolling conditions is required.
Kazutake Komori5 developed a numerical model to analyze material deformation and
temperature distribution in rolling of round bars. They investigated equivalent strain and
temperature distribution in three-roll and two-roll rolling processes, and difference between
them was clarified. M. Raudensky, et.al.6 developed a full scale model to carry out
experiments and also numerically simulated the problem to get deeper insight of the problem.
They used the system to give general recommendation for cooling of work-rolls during hot
rolling. They concluded that cooling increases with increasing number of nozzles, and that
distant sprays help cool rolls more efficiently. A. Prez, et.al.7 predicted the thermal response
of a work roll in a continuous hot strip mill, under typical working conditions. Their
mathematical model considered length of the strip, temperature variation along the length of
the strip, and temperature variation from the beginning to the end of the process. They
superimposed the three different levels of this phenomenon (i) the independent cycle of the
2

roll, (ii) rolling of a strip-rest, and (iii) a whole campaign. They concluded that the surface of
rollers reaches steady state condition after 10 cycles, and that whole roller reaches steady
state condition when effective time of laminating is greater than 1300s. Cheng Gang Sun8
gave a finite element (FE)-based analysis of the interfacial thermo-mechanical behavior of the
roll and strip in tandem mill during hot rolling. They investigated the validity of the proposed
model by comparing with experimental measurements. Then, using that model, they studied
the effect of various process parameters on interfacial thermo-mechanical behavior of the roll
and strip. C. Fedorciuc, et.al.9 worked on thermal fatigue crack propagation in work-rolls
during hot rolling. They developed 2D implicit FEM models considering complex thermomechanical interactions and cooling cycles. From that they investigated how stress state
inside the roll contributes in a different manner to energy release at the crack tip, depending
on initial crack length. A stress intensity factor (SIF) approach was used to derive the crack
growth rate. They found out that with increasing crack length there is increase in mechanical
stresses and decrease in thermal stresses at crack tip. L. Khalili10 developed a mathematical
model to assess thermo-mechanical behavior of work rolls during hot rolling processes. He
worked on AA2024. Thermal and mechanical responses at steady-state conditions were
investigated during the study using roll pressure and temperature field as governing boundary
conditions. They concluded that both thermal and mechanical aspects are important in
thermo-mechanical stresses developed within the work rolls, and that cooling design of rolls
can significantly alter both the magnitude and the distribution of thermo-mechanical stresses.
Ignoring heat conduction along roll axis i.e. z-direction, the basic heat transfer equation in
cylindrical coordinates for a work roll can be written as Eq. 111
(

Eq. 1

It should be noted that there is a strong thermal relationship between work-roll and strip,
therefore it is necessary to calculate the temperature distribution of the strip at the same
time12. The governing heat conduction equation can be employed as Eq. 2.
(

Eq. 2

Materials used in the work roll for hot rolling process undergo severe thermo-mechanical
loading during the process cycle. This imposes restrictions on the amount of deformation that
can be achieved in each pass.
In this work coupled thermo-mechanical FE model is used. Right boundary conditions,
process parameters and temperature dependent material data is needed to accurately simulate
this problem. Temperature and strain dependent mechanical data for brass was measured in
this study. The boundary conditions and other parameters were determined from the actual
process of 2 high brass rolling mill. Analysis has been carried out for the first pass and the
final pass. It is believed that the thermal stresses are highest in the first pass when the billet is
hottest and the bite angle is maximum. In the last pass it is believed that the thermal stresses
would be lower due to cooling of billet, however mechanical stresses might show higher
values due to hardening of billet material. For comparison purposes the analysis is also
carried out for cold rolling in the first pass as well as for the last pass, where highest rolling
stresses are expected. The model has been used to determine the hoop and shear stresses
produced in the work- rolls during hot and cold rolling. The sequence of these stresses is
3

determined and it is found that the peak hoop, shear stress and peak temperature are not in
phase and occur at different points during the rolling process. All the parameters of hot
rolling are compared with the cold rolling process. A criterion to compare the severity of the
stresses produced in the hot rolling and the cold rolling process is also introduced, it is based
on the comparison of hoop stresses produced with the yield strength of the roll material.
II. EXPERIMENTAL
II.1 Material
Hot rolling of Cartridge brass 110113 is done in a 2 high rolling mill in which rolls are made
of AISI H11 tool steel14. Chemical composition of cartridge brass 1101 and AISIH11 tool
steel are shown in Table I and
Table II.
Table I Chemical composition of Cartridge Brass 11013
Elements

Cu

Pb

Fe

Zn

Others

Weight %

70.0

0.05

0.04

29.84

0.15 max

Table II Chemical composition of H11 tool steel14

Elements

Weight % 0.330.43

Mn

Si

Cr

Ni

Mo

Cu

0.200.50

0.801.20

4.755.50

< 0.3

1.101.60

0.30.6

<
0.25

<
0.03

<
0.03

Brass compression tests were performed on MTS 810 servo hydraulic, universal testing
machine using the standard ASTM E915. Specimens of 25mm (height) 20mm (width)
12mm (thickness) were used. Three tests for each testing condition are carried out. The
testing parameters are shown in Table III. Thus a total of 60 experiments are carried out.
Range of strain rate for brass compression testing was obtained from the numerical
simulation. It was noted that the maximum strain rate during hot rolling occurs at the start of
first pass where strain rate is 40 strain/sec. The lowest speed where there are significant
stresses in the roll material is 5 strain/sec. The strain rates taken here cover most of the
deformation speeds during rolling, especially in the zones of high rolling stresses. At the end
of the bite angle the strain rate approaches zero but the stress in this zone is comparatively
lower.
In industrial practice a 120mm thick brass billet is hot rolled to reduce the thickness down to
23mm in 5 passes. Before rolling, the billet is heated up to 800 C and during the process it
cools down to almost 570 C. Maximum reduction ratio of 30% is attained in first pass at a
bite angle of 18o. In the last pass the reduction ratio is 23% with a bite angle of 8o. The testing
parameters cover the range of temperatures expected in the rolling process.

Table III Testing parameters for compression testing of brass

Temperature

Strain Rates (strain / sec)

800C

16

32

48

700C

16

32

48

600C

16

32

48

500C

16

32

48

25C

16

32

48

III. FINITE ELEMENT ANALYSIS

To carry out the FE analysis of the process, the problem is simplified and only one roller and
half of billet is simulated (symmetric). Detailed finite element analysis is performed on the
assembly of roll and billet as shown in Fig. 5 (first pass shown only). The roll is designed as a
2-d deformable shell of 730mm diameter and the billet is defined as a 2-d deformable shell of
1000mm (length) by 60mm (height) rectangular block. Instances are assembled together by
using displacement command. The contact bite angle is maintained to be 18o for the first pass.
All the parameters are adjusted in the same manner as above for the last pass as well.
Contact was defined to be rough surface contact, with maximum conductivity at contact and
zero conductivity when distance between roller and billet 1mm. The billet was given initial
temperature of 800 C and 1m/sec initial velocity. The roller is pinned from the center, given
30 C initial temperature and a constant angular velocity of 2.7397 rad/sec.
III.1 Material properties
Roller was assigned density of 7847 kg/m3, 0.3 Poissons ratio and temperature dependent
data of youngs modulus, conductivity, expansion and specific heat was specified as shown in
Table IV14, 16, 17. Temperature dependent stress strain data of AISI H11 tool steel was used
[unpublished data from Institut Clement Ader, Campus Jarlard, Alle Des Sciences, Albi
81000, France] as shown in Fig. 1.

Temperature
(0C)

Conductivity
( W/(mK)

Young's
modulus (GPa)

Thermal expansion
coefficient ( 110-6 K-1)

Specific heat
( KJ/(KgK)

20

29.8

213

13.5

0.482

200

31.8

184

13.6

0.516

300

32.1

178

13.7

0.589

400

31.8

169

13.9

0.621

500

30.9

153

14

0.669

600

29.4

126

14.2

0.745

Table IV Temperature dependent physical properties of H11 tool steel

1400

1200

Stress (MPa)

1000

800

600

20C
300C
500C
600C

400

200C
400C
550C

200
0

0.002

0.004
0.006
Strain (mm/mm)

0.008

0.01

Fig. 1 Temperature dependent plastic strain data of H-11 tool steel at 10-3 strain/sec
Brass was assigned temperature dependent conductivity data as shown in Table I. Plastic
strain curve for brass at room temperature under compression test is shown in Fig. 2. This
stress strain diagram is for 16 strain/sec. Tensile properties are found to be strain independent
in the range of deformation speeds as described in Table III. At room temperature the yield
strength is 470MPa with a very high strain hardening. Density, Poissons ratio, modulus of
elasticity, thermal expansion coefficient and specific heat for brass are shown in Table VI.

Table V Temperature dependent conductivity of Cartridge Brass 1101

Temperature
(0C)

Conductivity
( W/(mK)

25

115

100

136

300

148

700

157

1200

True compressive stress (MPa)

1000

800

600
16 strain/sec at room temperature
400

200

0
0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

True strain (mm/mm)

Fig. 2 Plastic strain data of brass at room temperature at 16 strain/sec
Table VI Mechanical properties of Cartridge Brass 110118
Density

Youngs
Modulus

Poissons ratio

Thermal
Coefficient

8530 kg/m3

10 GPa

0.3

1.8710-5 K-1

Expansion Specific Heat

0.382 KJ/K

Tests to determine the compressive material properties of brass at elevated temperatures were
conducted as discussed earlier in experimentation section. Results of these compression tests
were used as input data for numerical simulation. Compressive stress-strain curves have been
plotted for different temperatures in Fig. 3 at a strain rate of 16 strain/sec for all the curves.
The plot depicts that strength of brass is highly temperature dependent. It can clearly be seen
that at 800oC the ultimate tensile strength (UTS) of brass is only 90 MPa, while at 500oC UTS
increases up to 260MPa. Also the strain hardening reduces to a great extent with an increase
7

in temperature. It can be concluded that with the rise in temperature the brass billet becomes
softer, hence will exert less mechanical loadings on the work roll during rolling.
300
800 C
700 C

Compressive stress (MPa)

250

600 C
500 C

200

150

100

50

0
0

0.05

0.1

0.15

0.2

0.25

Strain (mm/mm)
Fig. 3 Stress-strain curves of cartridge brass 1101 for 16 strain/sec at different
temperatures
It can be observed from the Fig. 4 that stress/strain curves at varying strain rates do not show
much scatter. The maximum deviation in this curve was found to be 5% between strain rates
of 4 and 48 strain/sec at 15% strain. Variation in strain rates, within the range studied here,
does not cause considerable change in the stress strain values at 600oC. Hence it can be
concluded that varying deformation speed does not affect the strength of brass even at
elevated temperatures.

Compressive stress (MPa)

250

200

150
4 strain/sec
100

16 strain/sec
32 strain/sec
48 strain/sec

50

0
0

0.05

0.1

0.15

0.2

0.25

Strain (mm/mm)
Fig. 4 Compressive stress-strain curve of cartridge brass 1101 for different stain rates
at 600oC
III.2 Meshing
The roller model was sectioned and then seeded to create a homogenous controlled mesh of
aspect ratio close to 1 using explicit CPS4RT elements (4-node plane stress thermally
coupled quadrilateral, bilinear displacement and temperature, reduced integration, hourglass
control) with total 24120 elements and 24480 nodes. The billet was seeded with approximate
global size of 0.007 and then meshed using explicit, 4-node plane stress, same as the roller.
There are a total of 1287 elements and 1440 nodes in the billet. Meshed roller and billet are
shown in Fig. 5 for the first pass only. It must be added that the model was checked for mesh
dependency and the results were verified.

Fig. 5 Geometry, assembly and mesh of roll and billet

Fig. 6 is presented as an example of the stress profiles produced in the rolls according
to different conditions.

Fig. 6 Contour plot of hoop stresses produced at bite angle during first pass of hot
rolling process

10

IV. RESULTS AND DISCUSSION

The numerical simulation gives us an insight into the stresses produced thermally,
mechanically and coupled thermo-mechanically in the roll. Some researchers have worked on
determining the hoop stresses which are responsible for radial cracks19-21. However, it is
known that many rolls fail due to fretting / scaling of the roll surface. Fretting may be
produced due to shear stresses. Numerical simulation has been used to determine the shear
stress and depth of shear stress produced in addition to the hoop stresses.
1100

Cold Rolling (First Pass)

Hot Rolling (First Pass)

900

Cold Rolling (Last Pass)

Compressive hoop stress (MPa)

Hot Rolling (Last Pass)

700

500

300

100

-100

-10

-5

10

15

20

25

30

Bite angle (o)

Fig. 7 Hoop stresses in hot and cold rolling first pass and last pass
Fig. 7 shows the plot of compressive hoop stresses produced on the surface of work-roll at
bite angle, during hot and cold rolling for the first pass and the last pass. In cold rolling, bite
angle starts at 0o and ends at 18o, contact initiates at 0o and the compressive hoop stress at the
surface rapidly increases. At 14o or 78% of the bite angle compressive stresses reach at the
peak value of 1020 MPa after that the stress values rapidly reduce till 18o bite angle, where
the contact ends. While in hot rolling, pre-contact region -10o to 0o shows the increase in
hoop stress, it is due to conduction of heat to the roller. The stresses rise rapidly till contact
angle of 4 and remains close to 380 MPa till 17o and decreases rapidly once the contact is
terminated. As compared above, one can observe that much lesser hoop stresses are produced
in hot rolling than in cold rolling. However, the trend of the rise and fall of stresses is similar
in both the cases, with a characteristic plateau and a short peak just before the drop in stress.
The severity of this peak is much pronounced in cold rolling as compared to hot rolling.
Where as in the last pass the contact angle is reduced to only 8o. The stresses in hot rolling are
slightly higher (376 MPa) due to cooling and subsequent hardening of the material as
11

compared to the first pass. However for the last pass, as expected, the peak stress (775MPa) is
much lower due to the reduced bite angle. Still this peak stress of cold rolling is too high as
compared to the strength of the material.
400
300

Shear stress (MPa)

200

100
0
-100
-200

Shear stress in cold rolling (First Pass)

Shear stress in cold rolling (Last Pass)
Shear stress in hot rolling (First Pass)
Shear stress in hot rolling (Last Pass)

-300
-400
-10

-5

10
Bite angle (o)

15

20

25

30

Fig. 8 Comparison of maximum shear stress produced in cold rolling first pass (39mm deep),
last pass (23mm deep) and hot rolling during first pass (26mm deep), last pass (13mm deep)
The depth of maximum shear stress is of importance because fretting fatigue and resulting
spalling is the direct result of this shear stress. The rolls supplied by the manufacturer are case
hardened to 50mm depth. Redressing of the roll surface causes thinning of this case. The
depth of the shear stress is an important consideration for the life of the rolls, as the maximum
shear stress depth should not go beyond this hardened case.
Fig. 8 shows the plots of shear stress produced in work-roll during cold rolling process in the
first pass and last pass. It can be seen that shear stress profile is positive as well as negative.
Shear stresses show fully reverse behaviour with peak to peak value of 716 MPa (first pass).
These fully reversed cyclic stresses cause more damage as compared to the single
compressive hoop stress produced in the roller material. An interesting feature of this stress
distribution is the span along the roller. The hoop and radial stresses are mostly confined to
the part of the roller that is in contact with the billet as shown in Fig. 7. However, in the case
of shear stress, the contact starts with maximum value of shear, that changes magnitude and
sign to the lowest point just after the end of contact. The total span of action of shear stress
starts well before contact (almost 10 degrees) and remains after contact as well (almost 15
degrees). Similar behaviour is seen for last pass as well. Reducing the bite angle is also seen
to have an effect on the depth of maximum shear stress. It is found to be 39mm in the case of
first pass which redues to 23 mm depth for the last pass with a magnitude of 490MPa.
12

The maximum shear stresses produced in work-rolls during hot rolling for first pass and the
last pass are shown in Fig. 8. Also, the maximum shear stress depth is reduced in the case of
hot rolling in both first pass and last pass to 185 MPa at 26mm depth and 210 MPa at 13mm
respectively. The magnitude of shear stress in the last pass for hot rolling is found to be
higher than that of the first pass. That is again, due to the cooling and subsequent hardening
of the billet material.

Maximum (at 26mm depth)

Middle (at 13mm depth)
On Surface

100

Shear stress (MPa)

50

-50

-100

-10

-5

10
Bite angle (o)

15

20

25

30

Fig. 9 Shear stress profile for hot rolling first pass

13

120

Shear Stress (MPa)

70

20

-30
On Surface
Maximum (at 13mm depth)
Deep (at 26mm depth)

-80

-130
-10

-5

10

15

20

25

30

Bite Angle (o)

Fig. 10 Shear stress profile for hot rolling last pass
Comparison of shear stresses in the first and last pass at different depths is provided for hot
rolling only, as shown in Fig. 9 and Fig. 10 for reference. The same results are seen for cold
rolling. This graph shows that the shear stress magnitude increases from surface to a
maximum value at a certain depth. This depth is important, as stated above, to determined the
required case hardening depth.

14

400
Thermo-Mechanical
Pure Mechanical
Pure Thermal

Compressive stress (MPa)

350
300
250
200
150
100
50
0
-50
-10

-5

10

15

20

25

30

Bite angle (o)

Fig. 11 Pure thermal, pure mechanical and thermo-mechanical hoop stresses on the surface of
rolls during hot rolling in first pass

15

400
Thermo-Mechanical
Pure Mechanical
Pure Thermal

Compressive stress (MPa)

350
300
250
200
150
100
50
0
-50
-10

-5

5
10
Bite angle (o)

15

20

25

30

Fig. 12 Pure thermal, pure mechanical and thermo-mechanical hoop stresses on the surface of
rolls during hot rolling in last pass
A method for separating thermal and mechanical stresses has been developed. To find pure
thermal stresses produced in the work roll, brass billet was assigned very low yield strength
and UTS, so that the billet may not exert any mechanical load on the roll. The pure thermal
stress curve is shown in Fig. 11 and Fig. 12. Pure thermal stresses were subtracted from
thermo-mechanical stresses to find pure mechanical stresses produced during hot rolling. The
plots indicate that the thermal stresses are almost 35% of the total stresses produced in the hot
rolling process, rest 65% are due to mechanical effects for the first pass. In the last pass due
to cooling of the billet the mechanical component of stress increases to almost 75% and the
thermal stress component is reduced to 25%. The coupled thermo-mechanical stress should
follow the Eq. 3
Th-Mec = Th + Mec Eq. 3
This separation of stresses is useful in optimization of the rolling process, wherein the total
stress component (objective function) can be reduced by adjusting the two independent
parameters i.e., billet temperature (cause of Th) and bite angle (cause of Mec).

16

120

Cold Rolling (First Pass)

Hot Rolling (First Pass)
Cold Rolling (Last Pass)
Hot Rolling (Last Pass)

Percentage hoop stress (%)

100
80
60
40
20
0
-20
-10

-5

10

15

20

25

30

Bite angle (o)

Fig. 13 Comparison of percentage hoop stresses in hot and cold rolling for first pass and last
pass
In Fig. 13 a criteria is presented to compare the severity of stresses produced in hot rolling
and cold rolling process. It is based on the fact that during hot rolling although, lesser stresses
are produced, the yield strength of the material evolves continuously due to heating. It is thus
deemed advantageous that a proportional criteria be used when determining rolling
parameters like bite angle and billet temperature. A model is developed for this purpose given
in Eq. 4 and Eq. 5. Eq. 4 is used to calculate the percentage of hoop stress to yield stress of
the roll material. Eq. 5 is a linear interpolation from the H11 material property data as shown
in Fig. 1. The criteria is to present the hoop stress as a percentage of the yield stress of the roll
material at the instantaneous temperature.
Eq. 4
[

Eq. 5

Where
(913 MPa for H11) is the yield strength of the roll material at room temperature
and mH11 (-0.89 for H11) is the slope indicating reduction in yield strength with an increase in
temperature.
It is found that during hot rolling even though the temperature of the roll increases, the hoop
stress only reaches almost 45% of the yield stress for both the first and last pass. Whereas in
cold rolling the percentage ratio of hoop stress generated to the material yield strength at
17

room temperature is 110% and 85% for first pass and last pass respectively. These values are
too high and will severely damage, or even crack the rolls. The brass billet is thus hot rolled
when high reduction ratios are required.
This model (Eq. 4 and Eq. 5) is especially useful where even higher billet temperatures are
used, as in steel rolling or in applications of continuous strip rolling. In these cases the
temperature of the rolls rises to a point that it causes substantial thermal softening of the roll
surface. This model would help establish a quick criterion of the effective load on the roll
even on the shop floor by simply measuring roll surface temperature using a pyrometer. The
value of
would ideally be provided by the roll manufacturer.
V. CONCLUSION
A coupled thermo-mechanical analysis of the brass strip rolling process is presented. The
boundary conditions used are based on the real process parameters of the brass rolling mill.
Material models used are the temperature dependent data for the H11 (roll material) and
Cartridge Brass. Material properties for brass are determined through compression tests at
different temperatures and strain rates. Analysis is carried out for the first and last pass only,
in a 5 pass rolling process.
Hoop stress profiles are determined through numerical simulation. It is seen that the stresses
for cold rolling exceed the yield stress of the roll material by a large extent. The reduction in
bite angle reduces the maximum hoop stress from 1020MPa in the first pass down to 775MPa
in the last pass. However, for hot rolling, stresses produced are thermal as well as mechanical
in nature. In the first pass of hot rolling, the maximum thermal stress is about 35% of the
maximum hoop stress produced. In the last pass the thermal contribution to the hoop stress is
reduced to 25% of the maximum hoop stress. Maximum coupled thermo-mechanical hoop
stress reaches almost 380 MPa in both the first and the last pass.
Shear stress is also produced in the roll at a depth below the work surface of the roll. In cold
rolling the magnitude of the shear stress reaches 716 MPa at a depth of 39mm which is
reduced to 490 MPa at a depth of 23mm in the last pass. In the case of hot rolling the
maximum shear stress of 185 MPa is produced 26mm below the roll surface in the first pass
and 210 MPa at a depth of 13 mm in the last pass. The increase in the shear stress for last pass
is due to cooling and subsequent hardening of the billet. This depth is of importance because
the fretting cracks will be produced at this depth tangent to the roll surface. It is also
interesting to note that the manufacturer recommends high hardening up to 50mm depth of
roll. Redressing of the roll surface with time will reduce the thickness of the hardened surface
and the roll will be susceptible to early damage due to thermo-mechanical cyclic loads.
The analysis presented here successfully decouples the thermal and mechanical effects on the
roll during the rolling process. The FE model and method described here can be used to
readjust the process parameters of the rolling mill for optimization of the rolling process and
increasing roll life. Change in bite angle will have a direct effect on the mechanical stress and
the change in billet temperature will have an impact on the thermal stresses.
A criteria is presented to compare the severity of stresses produced in hot rolling and cold
rolling process. It is based on the fact that during hot rolling although, lesser stresses are
produced in the roll, the yield stress of the roll material evolves continuously due to heating.
18

The criteria is to present the hoop stress as a percentage of the yield stress of the roll material
at the instantaneous temperature. It is found that during hot rolling even though the
temperature of the roll increases, the hoop stress only reaches almost 45% of the yield stress
for both the first and last pass. This tool can be used to control the effective load on the rolls
during the process by simply measuring the roll surface temperature.
ACKNOWLEDGEMENT
This study was funded and supported by Department of Mechanical Engineering, University
of Engineering and Technology Taxila, Pakistan, which is gratefully acknowledged.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.

16.
17.
18.
19.
20.
21.

Z.-C. Lin and C.-C. Chen, Journal of materials processing technology, 1995, 49(1),
125-147.
C. Sun, S. Hwang, C. Yun, and J. Chung, Metallurgical and Materials Transactions
A, 1998, 29(9), 2407-2424.
L. M. Galantucci and L. Tricarico, Journal of Materials Processing Technology,
1999, 9293(0), 494-501.
M. P. Guerrero, C. R. Flores, A. Prez, and R. Cols, Journal of Materials Processing
Technology, 1999, 94(1), 52-59.
K. Komori, Journal of materials processing technology, 1999, 92, 450-457.
M. Raudensky, J. Horsky, and M. Pohanka, Journal of Materials Processing
Technology, 2002, 125126(0), 700-705.
A. Prez, R. L. Corral, R. Fuentes, and R. Cols, Journal of Materials Processing
Technology, 2004, 153154(0), 894-899.
C. Gang Sun, Tribology international, 2005, 38(4), 413-422.
C. Fedorciuc Onisa and D. C. J. Farrugia, International Journal of Material
Forming, 2008, 1(1), 363-366.
L. Khalili, S. Serajzadeh, and B. Koohbor, Metall and Materi Trans B, 2012, 43(6),
1638-1648.
H. Carslaw, Clarendon, Oxford, 1959.
S. Serajzadeh, A. K. Taheri, and F. Mucciardi, Modelling and Simulation in Materials
Science and Engineering, 2002, 10(2), 185.
R. K. Bell: 'Methods for the chemical analysis of NBS copper-base spectrochemical
standards'; 1965, For sale by the Superintendent of Documents, US Govt. Print. Off.
M. Handbook, Properties and Selection of Metals, American Society for Metals, 1961,
1206.
A. E9, 89a (Reapproved 2000), Standard Test Method of Compression Testing of
Metallic Materials at Room Temperature, Annual Book of ASTM Standards. 2000,
ASTM International PA.
P. D. Harvey: 'Engineering properties of steel'; 1982, American Society for Metals
Metals Park, Ohio.
J. Davis, ASM International, Metals Park, OH, 1996.
E. G. West: 'Copper and its Alloys'; 1982, Ellis Horwood New York.
M. Dnckelmeyer, C. Krempaszky, E. Werner, G. Hein, and K. Schrkhuber, Steel
Research International, 2010, 81(9), 86-89.
D.-F. Chang, Journal of Materials Processing Technology, 1999, 94(1), 45-51.
Z.-C. Lin and V.-H. Lin, Journal of Materials Processing Technology, 1997, 70(13),
62-76.
19

20

Copyright Form

Materials, Science and Technology (the Journal)

ASSIGNMENT OF COPYRIGHT
In order for W.S Maney & Son Ltd, trading as Maney Publishing (the Publisher), to ensure the widest
possible dissemination and protection of articles published in the Journal, we request authors to assign
worldwide copyright in print, digital and other media in their article, including abstracts, to Institute of
Materials, Minerals and Mining. This enables the Publisher to ensure maximum international copyright
protection against infringement, and to disseminate your article, and the Journal, to the widest possible
worldwide readership. Should the article be rejected from the Journal for any reason, all rights assigned
within this document revert back to the author(s) of the article.
Please read the explanatory notes of this agreement on the following page.
1.

In consideration of the undertaking set out in paragraph 2, the author(s) as beneficial owner(s)
hereby assign(s) to Institute of Materials, Minerals and Mining all rights, title and interest in the
copyright throughout the world in the article named below (which shall be taken to include any or
all supplementary material associated with the article*) for the full legal term of copyright.
Article title: Comparison of thermo-mechanical stresses produced in work rolls during hot and cold
rolling of Cartridge Brass 1101
Name(s) of Author(s): Faisal Qayyum; Masood Shah; S. Manzoor; Mohsin Abbas
Manuscript no./ref. (if known): MST11289

If this paper is sponsored for Open Access by your grant funders, please follow the MORE
OpenChoice route by signing and returning this copyright assignment to the address below, and find
details of the policy and how to make payment at www.maneyonline.com/openaccess.
So that there should be no doubt, it is understood and agreed that this assignment includes the right to
publish or adapt (subject to paragraphs 3 and 4) the material in the article for use in conjunction with
computer systems, including networks. The article may be published in printed, CD-ROM, microfiche, online,
or other machine-readable form, or in any other format or medium (electronic or otherwise) approved by
Institute of Materials, Minerals and Mining; and subject to data mining.
2.

Maney Publishing, as the Journals publisher, hereby undertakes to prepare for publication and
publish in the Journal, the Article named in paragraph 1.

3.

The editor of the Journal, and the Publisher, are empowered to make such editorial changes as may
be necessary to make the Article suitable for publication. Every effort will be made to consult the
Author(s) if substantive changes are required.

4.

The author(s) hereby assert(s) their moral rights under the terms of the Copyright Designs and
Patents Act 1988 to be identified as the Author(s) of the Article.

5.

The author(s) warrant(s) that the article is the author(s) original work, has not been published
before, either in part or whole, and is not currently under consideration for publication elsewhere.
The author(s) also warrant(s) that (a) the article contains no libellous or unlawful statements, (b) it in
no way infringes the rights of others (including the holders of copyright in material used within the
article), (c) that permission has been obtained to reproduce any material or illustrations for which
they do not hold copyright, and that (d) the author(s), as the owners of the copyright, is (are) entitled
to make the assignment set out in this agreement. The author(s) hereby indemnifies/indemnify the
Publisher against any claims for breach of the warranties given above.

STM October 2013

6.

In cases where the article is a translation of a previously published article, the author(s) warrant(s)
that full permission has been obtained to carry out and publish the translation of the article from its
original language, and that it in no way infringes the rights of others (including the holders of
copyright in material used within the original Article).

7.

The author(s) warrant(s) that the article contains no plagiarised material and agree(s) that the article
may be checked for plagiarism and other ethical questions (including checks using automated
software packages). The Publisher reserves the right to withdraw or retract any article for which, in
its reasonable opinion, valid concerns exist over the originality of the article, or its compliance with
the Journal's ethical policy. Maney's plagiarism and ethical policy may be read in full at
www.maneyonline.com/page/authors/publishingethics-general.

*If the author(s) cannot assign copyright for any or all the supplementary material, the clause in parentheses
should be struck out and a licence to publish provided for this material. If copyright is held by a third party,
permission to reproduce this material must be obtained from the copyright holder as laid out in paragraph 5
of this form and submitted to the publisher.
Copyright Assignment
I confirm that all the co-authors, named above in 1, know that the article has been submitted to the Journal.
I hereby sign this Assignment of Copyright with the full knowledge and agreement of all authors.

Date: 4th Feburary 2014

Signed:
Printed name: Faisal Qayyum

Please note: it is essential for publication that Maney Publishing receive this signed Assignment of
Copyright form. Without it, we will be unable to proceed to publication.
Please complete all specified fields and return this signed form to the Production Editor of the Journal.
Production Editor, MST, for Maney Publishing [email protected];
fax: +44 (0) 20 7451 7307
1 Carlton House Terrace, London SW1Y 5AF, UK

STM October 2013

ASSIGNMENT OF COPYRIGHT: EXPLANATORY NOTES

1.

The Journals policy is to acquire copyright for all contributions for the following reasons:
a.
ownership of copyright by a central organisation helps to ensure maximum international protection
against infringement;
b.
requests for permission to reproduce articles in books, course packs or for library loan can be handled
efficiently and with sensitivity to changes in international copyright law and to the general desirability of
encouraging the widest dissemination of knowledge. At the same time, this relieves editors and
societies of a time-consuming and costly administrative burden;
c.
the demand for research literature to be delivered in machine-readable form, on-line or on CD-ROM or
downloaded on a file server, can be met efficiently, with proper safeguards for authors, editors and
journal owners.

2.

As the information market changes, there are opportunities to reach both individual readers and institutions (e.g.
companies, schools and public libraries) that are unlikely to subscribe to the Journal. This involves working with
other organisations to publish online services or CD-ROM, or to deliver copies of individual articles. It also
involves registering the Journal with the Copyright Licensing Agency in the UK and the Copyright Clearance
Center in the USA, which offer centralised licensing arrangements for photocopying. Income received from all
of these sources will be used to further the interests of the Journal.

3.

As a consequence of the above, your article will be published in electronic, machine-readable form, and
normally printed, in the Journal, and may be stored electronically either by Maney or by a duly licensed third
party for delivery as an individual article copy or as part of a larger collection of articles, from a range of journals
to meet the specific requirements of a particular market. Assignment of copyright signifies agreement to the
Journal making such arrangements.

4.

It may be that the author is not able to make the assignment solely by him- or herself:
a.
If it is appropriate, the authors employer may sign this agreement. The employer may reserve the
right to use the article for internal or promotion purposes (by indicating on this agreement), and reserve
all rights other than copyright.
b.
If the author is a UK Government employee, TSO will grant a non-exclusive licence to publish the
article in the Journal in any medium or form provided that Crown Copyright and user rights (including
patent rights) are reserved.
c.
If the author is a US Government employee and the work was done in that capacity, the assignment
applies only to the extent allowed by US law.

5.

Under the UKs Copyright Design and Patents Act 1988, the author has the moral right to be identified as the
author whenever and wherever the article is published. He or she also has the right to object to derogatory
treatment or distortion of the work. The Journal encourages assertion of this right, as it represents best
publishing practice and provides an important safeguard for all authors. Clause 4 ensures that moral rights are
asserted, as required by the Act.

6.

Authors who wish to reproduce material from previously published sources or where the copyright is owned by
a third party, such as sections of text, tables or images, must obtain written permission from the copyright
holder and the author(s)/artist(s) of the original material. In the case of translations, the author(s) must obtain
written permission from the copyright holder of their original article upon which the translation is based.
Copyright is required for use in all formats (including digital), in perpetuity and in all geographical regions worldwide (and in colour for illustrations where relevant). For more information please see
www.maneyonline.com/page/authors/copyrightandpermissions.

7.

The Journal will not withhold permission from the author to use the content of his or her own article elsewhere
after publication in the Journal, provided acknowledgement is given to the Journal as the original source of
publication. Further permission is required for republication to be made in a commercial product and authors

STM October 2013

should contact [email protected] with their request. Permission is never unreasonably

withheld.
8.

The corresponding author shall receive a PDF file (Eprint) of the final, published, version of the article, which
may be forwarded and shared with all co-authors, and other research associates, but it cannot be archived or
put on a personal or institutional website, or in a subject based open access repository. No commercial use
shall be made of the article, or the PDF file, without Maneys prior, written permission.

9.

Authors may:
a.
put a copy of their un-refereed, unedited article (i.e. before peer-review - sometimes called a pre-print),
on their personal or institutional website, or in an institutional, or subject based repository; and
b.
put a copy of their final accepted version of the article (i.e. after peer-review but without Maney editing
and typesetting - sometimes called a post-print) on a non-commercial institutional repository subject to
an embargo period of twelve months from the date of publication, as long as (i) full acknowledgement
is made of the Journal issue in which the article is published, when this information is available; (ii)
there is a link to the published online version of the article on Maneys chosen online hosting platform,
to the Journal, and the Publishers website at www.maneyonline.com, and any other information
specified in the copyright transfer statement; and (iii) no commercial use is made of the article. A preprint of an article should be replaced by a post-print once the article has been accepted for publication.
(In the case of translations of original articles, this is subject to the permissions given by the copyright
holder of the original article). For more information please visit
www.maneyonline.com/page/authors/copyrightandpermissions or email
[email protected]

10.

Exceptions may be made to this policy in the following circumstances:

Sponsored open access papers (published as MORE OpenChoice papers) can be distributed under the terms
of one of three different Creative Commons licences:
1. CC-BY: Creative Commons Attribution Licence 3.0
This licence permits others to re-use, reproduce, re-distribute and/or build upon your work, including for
commercial purposes, providing they acknowledge you as author, and attribute original publication to the
journal.
Articles using this licence will carry the following copyright line:
Institute of Materials, Minerals and Mining 2014
MORE OpenChoice articles are open access and distributed under the terms of the Creative Commons
Attribution Licence 3.0
2. CC-BY-NC: Creative Commons Attribution Non-Commercial Licence 3.0
This licence permits others to re-use, reproduce, re-distribute or build upon your work providing they
acknowledge you as author, and attribute original publication to the journal, and that reuse is restricted to
noncommercial purposes, i.e. research or educational use.
For this licence, the copyright line will reads as follows:
Institute of Materials, Minerals and Mining 2014
MORE OpenChoice articles are open access and distributed under the terms of the Creative Commons
Attribution Non-Commercial Licence 3.0
3.

CC-BY-NC-ND: Creative Commons Attribution Non-Commercial No Derivs Licence 3.0 This licence permits
others to re-distribute your work providing they acknowledge you as author, and attribute original
publication to the journal. Your work cannot be altered, transformed or built upon, and use is restricted to
non-commercial purposes, i.e. research or educational use.

STM October 2013

Articles using this licence will carry the copyright line:

Institute of Materials, Minerals and Mining 2014
MORE OpenChoice articles are open access and distributed under the terms of the Creative Commons
Attribution Non-Commercial No Derivs Licence 3.0
Unless otherwise specified, the Creative Commons Attribution Non-Commercial Licence 3.0 (CC-BY-NC) will be
used.

STM October 2013

Response to reviewer comments

Comments from the Editors and Reviewers

The paper has a large number of figures
restrictions imposed on the journal.

which

conflict

with

the

space

Number of figures have been reduced from 24 to 13.

1 is trivial and should be deleted - the size of the specimens is given in
the text.
Deleted
Figures 4-7 and 9 could be more compactly presented as a simple table.
Fig. 4-7 have been presented
presented as Table V Page 7

in

Table

IV

Page

6,

and

Fig.

has

been

You should also look critically at the other figures to see whether some can
be omitted or combined (e.g. using different symbols or colours to present
results form the first and last pass on the same figure. This will mean renumbering the remaining figures in the text and in their captions.
-

Fig. 2 Page 6 has been deleted as it has been explained in text (Last
paragraph, Page 4)
Fig. 3 Page 7 has been deleted and text has been referred to Fig. 5
Page 10 due to same geometry
Fig. 15 Page 16 and Fig. 16 Page 17 have been combined as Fig. 7 Page
11
Fig. 17 Page 18 and Fig. 18 Page 19 have been combined as Fig. 8 page
12
Fig. 23 and Fig. 24 Page 23 have been combined as Fig. 13 Page 17.

All the figures

accordingly.

have

been

re-numbered,

re-captioned

and

cited

in

text

An alternative would be to present representative results in the paper and

make fuller details available as supplementary material that will be
available online, linked to the paper.
N/A
Tables 1 and 2: should read weight% (not weighted).
Corrected
Please check the manuscript carefully throughout since there are a number of
further English and grammatical errors that require attention.
Manuscript has been re-read and corrected, but not much mistakes and errors
were found.