Metal Forming Lecture 2
Metal Forming Lecture 2
Metal Forming Lecture 2
g.
ng
k
E
Spring 23-24
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
3D Stress strain analyses and yield locus
.S
la
Dr
ca
Dr. Shamik Basak
ni
e
ha
ke
Assistant Professor
ec
or
fM
Ro
Department of Mechanical and Industrial Engineering
.o
IIT
pt
Metal Forming
1
Stress state at a point
g.
Metal forming process involves complex stress and strain states.
ng
• State of stress at a point comprises the stress vectors or components of stress vectors acting on
E
sa
three mutually perpendicular planes passing through that point.
al
Ba
ri
st
ik
du
m
In
ha
nd
.S The three components for
ABCD face can be written as,
la
Dr
σxx, σxy and σxz .
ca
ni
e
ha
ke
First subscript: orientation of the
ec
or
normal to the plane on which the
fM
Ro
stress is acting
.o
Metal Forming
2
Stress state at a point
g.
ng
Stress state at a point can be represented by these
E
manner as shown below
sa
al
𝜎𝑥𝑥 𝜎𝑥𝑦 𝜎𝑥𝑧
Ba
ri
st
𝜎𝑖𝑗 = 𝜎𝑦𝑥 𝜎𝑦𝑦 𝜎𝑦𝑧
ik
dz
du
𝜎𝑧𝑥 𝜎𝑧𝑦 𝜎𝑧𝑧
In
Here i = the plane of which stress is acting and j =
ha
nd
.S direction on which the stress is acting
dy
la
Dr
Diagonal element of the matrix is NORMAL STRESS COMPONENTS
ca
dx
ni
All other element of the matric is SHEAR STRESS COMPONENTS
e
ha
ke
ec
or
So there are NINE unknown terms as of now.
fM
Ro
Now, doing couple balance
.o
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
ABC plane normal is along the 𝒏 is the unit vector
fM
Ro
𝒏 = 𝒏𝟏 𝒆𝟏 + 𝒏𝟐 𝒆𝟐 + 𝒏𝟑 𝒆𝟑
.o
IIT
pt
Metal Forming
4
Stresses on an inclined plane
g.
S1, S2 and S3 be the magnitudes of the three components. Let ‘As’ be the area of the triangular face
ng
ABC, and A1, A2 and A3 be areas of the triangular faces BPC, CPA and APB, respectively.
E
sa
𝑺 = 𝑺𝟏 𝒆𝟏 + 𝑺𝟐 𝒆𝟐 + 𝑺𝟑 𝒆𝟑
al
Ba
ri
Therefore, balancing force along the three axes
st
ik
du
m
𝑆1 . 𝐴𝑠 = 𝜎11 . 𝐴1 + 𝜎21 . 𝐴2 + 𝜎31 . 𝐴3
In
ha
𝑆2 . 𝐴𝑠 = 𝜎12 . 𝐴1 + 𝜎22 . 𝐴2 + 𝜎32 . 𝐴3
nd
.S 𝑆3 . 𝐴𝑠 = 𝜎13 . 𝐴1 + 𝜎23 . 𝐴2 + 𝜎33 . 𝐴3
la
Dr
The area components can be written in form of unit normal vectors as follows
ca
𝑛1 = 𝐴1 Τ𝐴𝑠 , 𝑛2 = 𝐴2 Τ𝐴𝑠 , 𝑛3 = 𝐴3 Τ𝐴𝑠
ni
e
ha
ke
Hence substituting the above equations, it can be written as
ec
or
𝑆1 = 𝜎11 . 𝑛1 + 𝜎21 . 𝑛2 + 𝜎31 . 𝑛3
fM
Ro
𝑆2 = 𝜎12 . 𝑛1 + 𝜎22 . 𝑛2 + 𝜎32 . 𝑛3
.o
IIT
𝑆3 = 𝜎13 . 𝑛1 + 𝜎23 . 𝑛2 + 𝜎33 . 𝑛3
pt
De
g.
ng
The magnitude of the resultant stress vector can be denoted as
E
sa
al
𝑺= 𝑺 = 𝑺𝟐𝟏 + 𝑺𝟐𝟐 + 𝑺𝟐𝟑
Ba
ri
st
ik
The magnitude of normal stress 𝜎𝑛 acting on the inclined
du
m
plane along 𝒏
ෝ
In
ha
nd
.S 𝑺. 𝒏 = 𝑺𝟏 𝒆𝟏 + 𝑺𝟐 𝒆𝟐 + 𝑺𝟑 𝒆𝟑 . 𝒏𝟏 𝒆𝟏 + 𝒏𝟐 𝒆𝟐 + 𝒏𝟑 𝒆𝟑
la
Dr
𝑺. 𝒏 = 𝑺𝟏 𝒏𝟏 + 𝑺𝟐 𝒏𝟐 + 𝑺𝟑 𝒏𝟑 = 𝑺𝒊 𝒏𝒊
ca
Hence,
ni
𝝈𝒏 = 𝑺𝒊 𝒏𝒊
e
ha
ke
Hence, if we know stress at a
ec
or
𝝈𝒏 = 𝝈𝒊𝒋 𝒏𝒋 𝒏𝒊
point we can calculate shear and
fM
Ro
normal stress at any arbitrary Hence the magnitude of the shear stress
.o
plane.
IIT
pt
𝝈𝒔 = 𝑺𝟐 − 𝝈𝟐𝒏
De
Metal Forming
6
Principal Stresses
g.
ng
• When the total stress vector on a plane is directed along the normal to the plane, the
E
sa
stress vector is called as principal stress and the plane is principal plane.
al
Ba
ri
• THERE IS NO SHEAR STRESS COMPONENTS ON A PRINCIPAL PLANE.
st
ik
du
• For a given stress state there is only ONE set of three principal planes and only ONE set
In
of three principal stresses acting on them.
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
De
Metal Forming
7
Principal Stresses
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
We can write as
la
Dr
𝜎11 − 𝜎 . 𝑛1 + 𝜎21 . 𝑛2 + 𝜎31 . 𝑛3 = 0
ca
𝜎12 . 𝑛1 + 𝜎22 − 𝜎 . 𝑛2 + 𝜎32 . 𝑛3 = 0
ni
e
ha
𝜎13 . 𝑛1 + 𝜎23 . 𝑛2 + 𝜎33 − 𝜎 . 𝑛3 = 0
ke
ec
We can write the equations in a tonsorial form as
or
fM
Ro
𝝈𝒊𝒋 − 𝜎𝜹𝒊𝒋 𝒏𝒊 = 0
.o
IIT
Here, 𝜹𝒊𝒋 is the kronecker delta function which can be written as
pt
De
Metal Forming
8
Principal Stresses
g.
ng
Above equations are the homogeneous equations in n1, n2 and n3. For a non-trivial solution
E
sa
of these equations, the determinant of coefficients of the variables in the above equations must
al
be zero
Ba
ri
st
ik
𝑑𝑒𝑡 𝝈𝒊𝒋 − 𝜎𝜹𝒊𝒋 = 0
du
m
Or
In
ha
nd
.S
la
Dr
𝜎11 − 𝜎 𝜎21 𝜎31
ca
𝜎12 𝜎22 − 𝜎 𝜎32 = 0
ni
𝜎13 𝜎23 𝜎33 − 𝜎
e
ha
ke
ec
or
This is a cubic equation and the three roots of this equation are the three principal stresses. The
fM
Ro
above equation, after expanding can be rewritten as
.o
IIT
𝝈𝟑 − 𝑰𝟏 𝝈𝟐 + 𝑰𝟐 𝝈 − 𝑰𝟑 = 0
pt
De
Metal Forming
9
Principal Stresses
g.
Where, I1, I2 and I3 are the coefficients of this equation and have values as given below
ng
k
E
𝐼1 = 𝜎11 + 𝜎22 + 𝜎33
sa
al
Ba
2 2 2
ri
𝐼2 = −𝜎12 − 𝜎23 − 𝜎31 + 𝜎11 𝜎22 + 𝜎22 𝜎33 + 𝜎33 𝜎11
st
ik
2 2 2
𝐼3 = 𝜎11 𝜎22 𝜎33 + 2𝜎12 𝜎23 𝜎31 − 𝜎12 𝜎33 − 𝜎23 𝜎11 − 𝜎31 𝜎22
du
m
In
ha
nd
Also, the I3 can be represented as the following determinant
.S
la
Dr
𝜎11 𝜎21 𝜎31
ca
𝐼3 = 𝜎12 𝜎22 𝜎32
ni
e
𝜎13 𝜎23 𝜎33
ha
ke
ec
or
With the change of co-ordinate system, the different stress components change,
fM
Ro
however, the coefficients I1, I2 and I3 do not change.
.o
IIT
Hence, these coefficients are called as INVARIANTS of the stress tensor.
pt
De
Metal Forming
10
Principal Stresses
g.
ng
k
E
sa
al
Let us denote the principal stresses by 𝜎1 , 𝜎2 , 𝜎3 . The
Ba
ri
st
forces are acting as shown in Fig. In this case the stress
ik
du
tensors are reduced to as
In
ha
𝐼1 = 𝜎1 + 𝜎2 + 𝜎3
nd
.S 𝐼2 = 𝜎1 𝜎2 + 𝜎2 𝜎3 + 𝜎3 𝜎1
la
Dr
ca
𝐼3 = 𝜎1 𝜎2 𝜎3
ni
e
ha
ke
ec
or
fM
Ro
.o
Metal Forming
11
Stresses acting on OCTAHEDRAL plane
g.
The octahedral plane is a plane equally inclined to directions of the three principal stresses.
ng
k
1 1 1
E
Direction cosines of normal to this plane with respect to 𝜎1 , 𝜎2 , 𝜎3 are , , .
sa
3 3 3
al
Ba
ri
1 𝐼1
st
𝜎𝑛,𝑜𝑐𝑡 = 𝜎1 𝑛12 + 𝜎2 𝑛22 + 𝜎3 𝑛32 = 𝜎1 + 𝜎2 + 𝜎3 =
ik
3 3
du
m
In
ha
nd
.S
𝜎𝑠,𝑜𝑐𝑡 = 𝑆 2 − 𝜎𝑛2 = 𝑆12 + 𝑆22 + 𝑆32 − 𝜎𝑛,𝑜𝑐𝑡
2
la
Dr
ca
2
1
ni
= 𝜎1 𝑛1 2 + 𝜎2 𝑛2 2 + 𝜎3 𝑛3 2 − 𝜎 + 𝜎2 + 𝜎3
3 1
e
ha
ke
ec
𝟏
or
𝟐 𝟐 𝟐 𝟎.𝟓
𝝈𝒔,𝒐𝒄𝒕 = 𝝈𝟏 − 𝝈𝟐 + 𝝈𝟐 − 𝝈𝟑 + 𝝈𝟑 − 𝝈𝟏
fM
Ro
𝟑
.o
1
De
2 0.5
𝜎𝑠,𝑜𝑐𝑡 = 2 𝜎1 + 𝜎2 + 𝜎3 − 6 𝜎1 𝜎2 + 𝜎2 𝜎3 + 𝜎3 𝜎1
3
Metal Forming
12
Stresses acting on OCTAHEDRAL plane
g.
ng
In terms of stress invariants, it can be represented
E
sa
as
al
Ba
ri
1 2 2 2
st
𝜎𝑠,𝑜𝑐𝑡 = 2𝐼 − 6𝐼2 0.5 = 𝐼 − 3𝐼2 0.5
ik
du
3 1 3 1
In
ha
nd
.S In terms of generalized stress state, it can be
la
Dr
written as
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
𝟏
IIT
pt
𝟎.𝟓
𝝈𝒔,𝒐𝒄𝒕 = 𝝈𝟏𝟏 − 𝝈𝟐𝟐 𝟐
+ 𝝈𝟐𝟐 − 𝝈𝟑𝟑 𝟐
+ 𝝈𝟑𝟑 − 𝝈𝟏𝟏 𝟐
+ 𝟔 𝝈𝟐𝟏𝟐 + 𝝈𝟐𝟐𝟑 + 𝝈𝟐𝟑𝟏
De
Metal Forming
13
Deviatoric stress tensor : J2 invariant
g.
ng
The hydrostatic pressure does not make a metal yield. It only produces elastic volume
E
change.
sa
al
Ba
ri
1 1 1
st
𝜎𝑝 = 𝜎𝑖𝑖 = 𝜎11 + 𝜎22 + 𝜎33 = 𝜎1 + 𝜎2 + 𝜎3
3 3 3
ik
du
m
In
ha
𝜎11 𝜎12 𝜎13
nd
.S
𝜎𝑖𝑗 = 𝜎21 𝜎22 𝜎23
la
Dr
𝜎31 𝜎32 𝜎33
ca
ni
𝜎11 − 𝜎𝑝 𝜎12 𝜎13 𝜎𝑝 0 0
e
𝜎𝑖𝑗 = 𝜎21
ha
𝜎22 − 𝜎𝑝 𝜎23 + 0 𝜎𝑝 0
ke
ec
𝜎31 𝜎32 𝜎33 − 𝜎𝑝 0 0 𝜎𝑝
or
fM
Ro
.o
Deviatoric component
IIT hydrostatic component
pt
1
𝝈′𝒊𝒋 = 𝝈𝒊𝒋 − 3 𝜹𝒊𝒋 𝝈𝒌𝒌
Here, 𝜹𝒊𝒋 is the kronecker delta function which can be written as Metal Forming
𝜹𝒊𝒋 = 𝟏 𝑖𝑓 𝑖 = 𝑗 𝑎𝑛𝑑 𝜹𝒊𝒋 = 𝟎 𝑖𝑓 𝑖 ≠ 𝑗 14
Deviatoric stress tensor : J2 invariant
g.
′ ′ ′
= 𝜎11 − 𝜎𝑝 , 𝜎22 = 𝜎22 − 𝜎𝑝 , 𝜎33
ng
𝜎11 = 𝜎33 − 𝜎𝑛
E
sa
al
′ ′ ′ ′ ′ ′
𝜎12 = 𝜎12 = 𝜎21 = 𝜎21 , 𝜎23 = 𝜎23 = 𝜎32 = 𝜎32 , 𝜎31 = 𝜎31 = 𝜎13 = 𝜎13
Ba
ri
st
ik
Hence the stress state at a point can be re written as
du
m
′ ′ ′ 𝜎𝑝 0 0
In
𝜎11 𝜎12 𝜎13
ha
′ ′ ′
+ 0 𝜎𝑝 0
nd
.S 𝜎𝑖𝑗 = 𝜎21 𝜎22 𝜎23
′ ′ ′ 0 0 𝜎𝑝
𝜎31 𝜎32 𝜎33
la
Dr
ca
For deviatoric stress tensor we need to calculate the stress invariants
ni
e
ha
𝑑𝑒𝑡 𝝈′𝒊𝒋 − 𝜎𝜹𝒊𝒋 = 0
ke
ec
or
fM
Ro
′ ′ ′
𝜎11 −𝜎 𝜎12 𝜎13
.o
′ ′ ′
𝜎21 𝜎22 −𝜎 𝜎23 = 0
IIT
pt
′ ′ ′
𝜎31 𝜎32 𝜎33 −𝜎
De
Metal Forming
15
Deviatoric stress tensor: J2 invariant
g.
ng
k
E
This is a cubic equation and the three roots of this equation are the three principal deviatoric
sa
al
stresses. The characteristic equation for deviatoric stress tensor is
Ba
ri
st
𝟐
𝝈′𝟑 − 𝑱𝟏 𝝈′ − 𝑱𝟐 𝝈′ − 𝑱𝟑 = 0
ik
du
m
In
ha
nd
.S
Where, J1, J2 and J3 are the coefficients of this equation and have values as given below
la
Dr
𝐽1 = 𝜎′11 + 𝜎′22 + 𝜎′33
ca
ni
e
ha
ke
′2 ′2 ′2
𝐽2 = 𝜎12 + 𝜎23 + 𝜎31 − 𝜎 ′11 𝜎 ′ 22 − 𝜎 ′ 22 𝜎 ′ 33 − 𝜎′33 𝜎′11
ec
or
fM
Ro
2
𝜎′33 − 𝜎′223 𝜎′11 − 𝜎31
′2
.o
Metal Forming
16
Deviatoric stress tensor: J2 invariant
g.
ng
The value of the first deviatoric invariant is ZERO as shown
E
′ ′ ′
𝐽1 = 𝜎11 + 𝜎22 + 𝜎33 = 𝜎11 + 𝜎22 + 𝜎33 − 3𝜎𝑝 = 0
sa
al
Ba
ri
Further, the J2 can be written as
st
ik
du
′ 2′ ′ 2 ′ ′2 ′ ′ ′ ′
𝐽2 = 𝜎12 + 𝜎23 + 𝜎31 − 𝜎11 𝜎22 − 𝜎22 𝜎33 − 𝜎33 𝜎11
In
ha
nd
Using
.S
la
Dr
′ ′ ′
𝜎11 + 𝜎22 + 𝜎33 =0
ca
ni
Rearranging
e
ha
ke
ec
1
or
′ ′ 2 ′ ′ 2 ′ ′ 2 ′2 ′2 ′ 2
𝐽2 = 𝜎11 − 𝜎22 + 𝜎22 − 𝜎33 + 𝜎33 − 𝜎11 + 6𝜎12 + 6𝜎23 + 6𝜎31
fM
Ro
6
.o
Also,
IIT
pt
′ ′
𝜎11 − 𝜎22 = 𝜎11 − 𝜎22
De
′ ′
𝜎22 − 𝜎33 = 𝜎22 − 𝜎33
′ ′
𝜎33 − 𝜎11 = 𝜎33 − 𝜎11
′ ′ ′
𝜎12 = 𝜎12 , 𝜎23 = 𝜎23 , 𝜎31 = 𝜎31 Metal Forming
17
Deviatoric stress tensor: J2 invariant
g.
ng
Replacing all,
E
sa
al
1
Ba
ri
2 2 2 2 2 2
𝐽2 = 𝜎 − 𝜎22 + 𝜎22 − 𝜎33 + 𝜎33 − 𝜎11 + 6𝜎12 + 6𝜎23 + 6𝜎31
st
6 11
ik
du
m
In
ha
𝟏 𝟐 1
nd
𝑱𝟐 = 𝑰𝟏 + 𝑰𝟐 = − 𝜎1′ 𝜎2′ + 𝜎2′ 𝜎3′ + 𝜎3′ 𝜎1′ = 𝜎1′2 + 𝜎2′2 + 𝜎3′2
.S
𝟑 2
la
Dr
ca
ni
𝟐 𝟑 𝟏 1 ′3
𝑰𝟏 − 𝑰𝟏 𝑰𝟐 + 𝑰𝟑 = 𝜎1 𝜎2 𝜎3 = 𝜎1 + 𝜎2′3 + 𝜎3′3
′ ′ ′
e
𝑱𝟑 =
ha
ke
𝟐𝟕 𝟑 3
ec
or
Also
fM
Ro
.o
𝟏 IIT 𝟐
pt
𝟎.𝟓
𝝈𝒔,𝒐𝒄𝒕 = 𝟐𝑰𝟐𝟏 − 𝟔𝑰𝟐 = 𝑱
𝟑 𝟐
De
g.
ng
For deviatoric stress tensor can be written as
E
sa
al
𝝈′𝒊𝒋 = 𝝈𝒊𝒋 − 𝜎𝑚 𝜹𝒊𝒋
Ba
ri
Or
st
ik
1
du
𝒔𝒊𝒋 = 𝝈𝒊𝒋 − 𝜎𝑚 𝜹𝒊𝒋 𝜎𝑚 = 𝜎 + 𝜎22 + 𝜎33
m
3 11
In
ha
Or
nd
.S
𝑠 = 𝜎 − 𝜎𝑚 𝐼
la
Dr
ca
Characteristic Eqn for Cauchy deviatoric stress tensor reduces to
ni
𝒔𝟑 − 𝑱𝟐 𝒔 − 𝑱𝟑 = 0
e
ha
ke
ec
or
𝑱𝟐 𝑱𝟐 2𝜋 𝑱𝟐 4𝜋
Its solution, 𝒔𝟏 = 𝟐 cos 𝜽, 𝒔𝟐 = 𝟐 cos − 𝜃 , 𝒔𝟑 = 𝟐 cos −𝜃
fM
Ro
𝟑 𝟑 3 𝟑 3
.o
IIT
pt
Metal Forming
19
J2 plasticity
g.
Effective/Equivalent stress
ng
k
E
Equivalent or effective stress term is often used in plasticity. They are also called as representative
sa
al
stress. The effective stress can be mathematically written as
Ba
ri
st
𝟑
ഥ=
𝝈 𝟑𝑱𝟐 = 𝝈
ik
𝟐 𝒔,𝒐𝒄𝒕
du
m
Dimensionless parameters
In
ha
𝐼1 𝜎𝑚
nd
.S Stress triaxiality, η = = ഥ
3 3𝐽2 𝜎
la
Dr
ca
Normalized third deviatoric stress invariant (ξ)
ni
e
ha 𝐽3 3
3 Τ2
27 𝐽3 −1 ≤ ξ ≤ +1
ke
ξ = cos 3𝜃 = =
ec
2 𝐽2 2 𝜎ത 3
or
−1 ≤ 𝜃ҧ ≤ +1
fM
Ro
6𝜃
Lode angle parameter = 𝜃ҧ = 1 −
.o
𝜋 𝜋
IIT 0 ≤ 𝜃ҧ ≤
pt
𝟐 𝟐 2𝜋 𝟐 4𝜋
𝒔𝟏 = 𝟏 + 𝟑η cos 𝜽 𝝈𝒎 , 𝒔𝟐 = 𝟏 + 𝟑η cos −𝜃 𝝈𝒎 , 𝒔𝟑 = 𝟏 + 𝟑η cos −𝜃 𝝈𝒎
3 3
Metal Forming
For detailed derivations follow class note 20
J2 plasticity
g.
ng
Relation among them (For plane stress condition)
E
sa
𝜋 27 1
al
ξ = cos 3𝜃 = cos 1 − 𝜃ҧ =− η η2 −
Ba
ri
2 2 3
st
For detailed derivations follow class note
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
De
Metal Forming
21
Planes of maximum shear stresses
g.
ng
If we change the orientations of the inclined plane, we shall come across a set of planes on which
E
the shear stresses are the maximum but normal stresses are not zero. These are called
sa
al
planes of maximum shear stress. We know already
Ba
ri
st
𝜎𝑛 = 𝜎1 𝑛12 + 𝜎2 𝑛22 + 𝜎3 𝑛32 and 𝜎𝑠2 = 𝑆 2 − 𝜎𝑛2
ik
du
m
In
ha
Since 𝜎𝑠 is a function of 𝑛1 , 𝑛2 , 𝑛3 ., its maxima or minima can be obtained by solving the
nd
following equations.
.S
la
Dr
𝜕𝜎𝑠 𝜕𝜎𝑠 𝜕𝜎𝑠
ca
= = =0
𝜕𝑛1 𝜕𝑛2 𝜕𝑛3
ni
e
ha
ke
Also we know,
ec
or
𝑛12 + 𝑛22 + 𝑛32 = 1, 𝑠𝑜 𝑛32 = 1 − 𝑛12 − 𝑛22
fM
Ro
Now
.o
IIT
𝟐
𝝈𝟐𝟏 − 𝝈𝟐𝟑 𝒏𝟐𝟏 𝝈𝟐𝟐 𝝈𝟐𝟑 𝒏𝟐𝟐 + 𝝈𝟐𝟑 𝝈𝟑 𝒏𝟐𝟏 𝝈𝟑 𝒏𝟐𝟐
pt
𝝈𝟐𝒔 = + − − 𝝈𝟏 − + 𝝈𝟐 − + 𝝈𝟑
De
Metal Forming
22
Planes of maximum shear stresses
g.
Differentiating w.r.t 𝑛1 and Rearranging
ng
2𝑛1 𝜎1 − 𝜎3 𝜎1 − 𝜎3 − 2𝑛12 𝜎1 − 𝜎3 − 2𝑛22 𝜎2 − 𝜎3
k
=0
E
sa
al
Differentiating w.r.t 𝑛2 and Rearranging
Ba
ri
2𝑛2 𝜎2 − 𝜎3 𝜎2 − 𝜎3 − 2𝑛12 𝜎1 − 𝜎3 − 2𝑛22 𝜎2 − 𝜎3 =0
st
ik
du
m
Taking as general case, 𝜎1 ≠ 𝜎2 ≠ 𝜎3
In
ha
nd
.S
Case I: We can get, 𝑛1 = 0 and, 𝑛2 = 0 which gives, 𝑛3 = ±1
la
Dr
ca
Case II: Again we can think that 𝑛1 = 0 but 𝑛2 ≠ 0 so, 1 − 𝑛22 = 0
ni
e
Which gives, 𝑛2 = ±
1
ha
so we can get, 𝑛3 = ±
1
ke
ec
2 2
or
1 1
Case III: Similarly, we can take 𝑛2 = 0, and 𝑛1 = ± , 𝑛 = ±
fM
Ro
2 2
1 1
.o
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
1 1 1 1
Dr
𝑛1 = 0 𝑛2 = ± , 𝑛3 = ± 𝑛2 = 0, 𝑛1 = ± , 𝑛3 = ±
2 2
ca
2 2
ni
e
ha
ke
ec
or
fM
Ro
.o
1 1
IIT 𝑛3 = 0, 𝑛1 = ± , 𝑛2 = ±
pt
2 2
De
Metal Forming
24
Equilibrium Equation
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
•
IIT
A deformable body subjected to external forces, sets up internal resistance stresses due to movement of its atoms.
pt
De
• The equilibrium at any point of the body is established by Newton’s second law of motion
• For equilibrium, the resultant forces along each of the three axes must be zero.
Metal Forming
• Let, F1, F2, F3 be the components of the body force acting along x1, x2 and x3 respectively. 25
Equilibrium Equation
g.
ng
For equilibrium, the resultant forces along each of the three axes must be zero.
E
Let, F1, F2, F3 be the components of the body force acting along x1, x2 and x3 respectively. Hence, by
sa
al
equilibrium in the X direction we can get
Ba
ri
st
ik
du
𝜕𝜎11 𝜕𝜎21 𝜕𝜎31 𝑑𝑉1
m
𝑑𝑥1 𝑑𝑥2 𝑑𝑥3 + 𝑑𝑥2 𝑑𝑥1 𝑑𝑥3 + 𝑑𝑥3 𝑑𝑥1 𝑑𝑥2 + 𝐹1 𝑑𝑥1 𝑑𝑥2 𝑑𝑥3 = 𝜌 𝑑𝑥1 𝑑𝑥2 𝑑𝑥3
In
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝑑𝑡
ha
nd
.S
Here, ρ is the density of the material and v1 is the velocity component along X1 direction.
la
Dr
ca
𝜕𝜎11 𝜕𝜎21 𝜕𝜎31 𝑑𝑉1
+ + + 𝐹1 = 𝜌
ni
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝑑𝑡
e
ha
ke
Similarly, for X2 and X3 directions, the equations can be written as
ec
or
𝜕𝜎12 𝜕𝜎22 𝜕𝜎32 𝑑𝑉2
fM
Ro
+ + + 𝐹2 = 𝜌
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝑑𝑡
.o
IIT
pt
Metal Forming
26
Equilibrium Equation
g.
Neglecting body forces the equations can be written as
ng
k
𝜕𝜎11 𝜕𝜎21 𝜕𝜎31 𝑑𝑉1
E
sa
+ + =𝜌
al
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝑑𝑡
Ba
ri
st
𝜕𝜎12 𝜕𝜎22 𝜕𝜎32 𝑑𝑉2
ik
+ + =𝜌 Equations of Motion
du
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝑑𝑡
In
ha
𝜕𝜎13 𝜕𝜎23 𝜕𝜎33 𝑑𝑉3
nd
.S 𝜕𝑥1
+
𝜕𝑥2
+
𝜕𝑥3
=𝜌
𝑑𝑡
la
Dr
In steady state conditions the accelerations are zero, hence, forces due to accelerations are also
ca
equal to zero. The equations can be written as
ni
e
ha
𝜕𝜎11 𝜕𝜎21 𝜕𝜎31
ke
+ + =0
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3
ec
or
fM
Equations of
Ro
𝜕𝜎12 𝜕𝜎22 𝜕𝜎32
+ + =0 Equilibrium
.o
+ + =0
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
Please find the class
la
Dr
ca
notes for details
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
De
Metal Forming
28
Yield conditions
g.
ng
During uniaxial tensile test we may write the yield condition as given below. The material is in plastic
k
state if and only if
E
sa
al
𝝈𝟏𝟏 − 𝝈𝟎 ≥ 𝟎
Ba
ri
Where, 𝜎0 is the yield strength of the material in tension or compression.
st
ik
du
m
However, in generalized cases, the yield condition in multi-axial stress state can be written in some
In
ha
function of stresses as given below
nd
.S
𝑭 𝝈𝒊𝒋 = 𝟎
la
Dr
ca
ni
Assumption:
e
ha
ke
• The material is homogeneous (composition is uniform) and isotropic.
ec
or
• The material is rigid perfectly plastic
fM
Ro
• There is no hysteresis.
.o
• The Bauschinger effect is negligible (i.e. yield point in tension and compression is same)
IIT
pt
Metal Forming
29
Yield conditions
g.
ng
We know
E
𝐹 𝐽2 , 𝐽3 = 0
sa
al
Ba
ri
𝐹 𝐽2 , 𝐽3 = 𝐴 + 𝐵𝐽2 + 𝐶𝐽3 + 𝐷𝐽2 𝐽3 + 𝐸𝐽2 2 + 𝐹𝐽3 2 + ⋯ = 0
st
ik
du
m
In
Neglecting the higher order terms
ha
nd
.S
la
𝐴 + 𝐵𝐽2 + 𝐶𝐽3 = 0
Dr
ca
Now if we neglect Bauschinger effect and von Mises yield model, then 𝐽3 can be neglected. So the
ni
e
above equation can be reduced to ha
ke
ec
𝐴 + 𝐵𝐽2 = 0
or
fM
i.e.
Ro
𝑱𝟐 = 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕
.o
IIT
pt
Metal Forming
30
von Mises yield criterion
g.
As discussed that von Mises yield model can be represented as
ng
k
𝑱𝟐 = 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕
E
sa
al
1 2 2 2
𝜎11 − 𝜎22 2 + 𝜎22 − 𝜎33 2 + 𝜎33 − 𝜎11 2 + 6𝜎12
Ba
+ 6𝜎23 + 6𝜎31 =𝐶
ri
6
st
ik
du
2 2 2
𝜎11 − 𝜎22 2 + 𝜎22 − 𝜎33 2 + 𝜎33 − 𝜎11 2 + 6𝜎12 + 6𝜎23 + 6𝜎31 = 6𝐶 = 𝐶1
In
ha
nd
.S
So in terms of principal stresses the von Mises yield model can be written as
la
Dr
𝟐 𝟐 𝟐
ca
𝝈𝟏 − 𝝈𝟐 + 𝝈𝟐 − 𝝈𝟑 + 𝝈 𝟑 − 𝝈𝟏 = 𝟔𝑪 = 𝑪𝟏
ni
e
ha
ke
To evaluate the constant value, we need to perform the uniaxial tensile test and torsion tests.
ec
or
fM
Ro
.o
IIT
For uniaxial tensile tests, 𝜎1 = 𝑌 , 𝜎2 = 𝜎3 = 0 Clearly, 𝑪𝟏 = 𝟐𝒀𝟐
pt
De
g.
ng
So FINAL von Mises YIELD model equation
E
sa
al
Ba
ri
𝝈𝟏𝟏 − 𝝈𝟐𝟐 𝟐
+ 𝝈𝟐𝟐 − 𝝈𝟑𝟑 𝟐
+ 𝝈𝟑𝟑 − 𝝈𝟏𝟏 𝟐
+ 𝟔𝝈𝟐𝟏𝟐 + 𝟔𝝈𝟐𝟐𝟑 + 𝟔𝝈𝟐𝟑𝟏 = 𝟐𝒀𝟐 = 𝟔𝑲𝟐
st
ik
du
m
In
ha
In the principal stress it can be written as
nd
.S 𝟐 𝟐 𝟐
𝝈𝟏 − 𝝈𝟐 + 𝝈𝟐 − 𝝈𝟑 + 𝝈𝟑 − 𝝈𝟏 = 𝟐𝒀𝟐 = 𝟔𝑲𝟐
la
Dr
ca
ni
𝑲 = 𝟎. 𝟓𝟕𝟕𝒀
e
ha
ke
ec
Hence, under von Mises criteria
or
fM
Ro
yield strength of material under SHEAR = 0.577 x yield strength of material under TENSION
.o
IIT
pt
De
Metal Forming
32
Tresca yield criterion
g.
Maximum of 𝝈𝟏 − 𝝈𝟐 , 𝝈𝟐 − 𝝈𝟑 , 𝝈𝟑 − 𝝈𝟏 = 𝑪𝟐
ng
For uniaxial tensile tests, 𝜎1 = 𝑌 , 𝜎2 = 𝜎3 = 0 Clearly, 𝑪𝟏 = 𝒀
E
sa
al
Ba
ri
For torsion tests (pure shear), 𝜎1 = 𝐾 , 𝜎2 = −𝐾, 𝜎3 = 0 Clearly, 𝑪𝟏 = 𝟐𝑲
st
ik
du
m
Maximum of 𝝈𝟏 − 𝝈𝟐 , 𝝈𝟐 − 𝝈𝟑 , 𝝈𝟑 − 𝝈𝟏 = 𝒀 = 𝟐𝑲
In
ha
nd
.S 𝑲 = 𝟎. 𝟓𝒀
la
Dr
Hence, under Tresca criteria
ca
ni
yield strength of material under SHEAR = 0.5 x yield strength of material under TENSION
e
ha
ke
ec
or
fM
Ro
.o
IIT
𝟒𝑱𝟑𝟐 − 𝟐𝟕𝑱𝟐𝟑 − 𝟑𝟔𝑲𝟐 𝑱𝟐𝟐 + 𝟗𝟔𝑲𝟒 𝑱𝟐 = 𝑪 = 𝟔𝟒𝑲𝟔
pt
De
Upon simplification
2
𝑠1 − 𝑠2 − 4𝐾 2 𝑠2 − 𝑠3 2
− 4𝐾 2 𝑠3 − 𝑠1 2
− 4𝐾 2 = 0
Metal Forming
2
𝜎1 − 𝜎2 − 4𝐾 2 𝜎2 − 𝜎3 2
− 4𝐾 2 𝜎3 − 𝜎1 2
− 4𝐾 2 =0
33
Graphical representation of yield criterion
g.
2D yield locus
ng
k
E
sa
Revisiting the von-Mises yield equation,
al
Ba
ri
𝜎1 − 𝜎2 2 + 𝜎2 − 𝜎3 2 + 𝜎3 − 𝜎1 2 = 2𝑌 2
st
ik
du
m
Now for plane stress condition, 𝜎3 = 0
In
ha
nd
.S
At that condition the equation reduces to
la
Dr
ca
2𝜎12 + 2𝜎12 − 2𝜎1 𝜎2 = 2𝑌 2
ni
e
ha
ke
𝜎1 2 𝜎2 2 𝜎1 𝜎2
ec
or
+ − =1
𝑌 𝑌 𝑌 𝑌
fM
Ro
.o
Metal Forming
34
Graphical representation of yield criterion
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
The von Mises hypothesis is shown by an elongated ellipse while Tresca hypothesis is
De
Metal Forming
35
Graphical representation of yield criterion
g.
ng
• Stress state comparison
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
De
Metal Forming
36
Graphical representation of yield criterion
g.
3D representation
ng
k
E
sa
al
Ba
X
ri
st
ik
du
P
In
ha
nd
.S
la
Dr
N
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
This 3D space is also known as Haigh-Westergaard Stress Space.
pt
De
Metal Forming
37
Graphical representation of yield criterion
g.
OX: Equally inclined to all of the stress direction. Here, 𝜎1 = 𝜎2 = 𝜎3
ng
k
E
sa
P is the point of the yield locus. Clearly,
al
Ba
ri
st
𝑶𝑷 = 𝝈𝟐𝟏 + 𝝈𝟐𝟐 + 𝝈𝟐𝟑
ik
du
m
In
𝝈𝟏 𝝈𝟐 𝝈𝟑 𝟑𝝈𝒎
ha
𝑶𝑵 = + + = = 𝟑𝝈𝒎 = 𝟑ηഥ
𝝈
nd
.S 𝟑 𝟑 𝟑 𝟑
la
Dr
ON is the hydrostatic component of the material
ca
ni
𝑃𝑁 = 𝑂𝑃2 − 𝑂𝑁 2 = 𝜎12 + 𝜎22 + 𝜎32 − 3𝜎𝑚
2
e
ha
ke
ec
Or it can be re-written in terms of deviatoric component as
or
fM
Ro
𝟐 𝟐 𝟐 𝟐 𝟐
𝑷𝑵 = 𝒔𝟏 + 𝒔𝟐 + 𝒔𝟑 = 𝟐𝑱𝟐 = 𝒀
𝟑
.o
IIT
Hence, it can be confirmed that the PN component is the deviatoric component of stresses.
pt
De
Metal Forming
38
Graphical representation of yield criterion
g.
ng
k
E
sa
al
Ba
ri
st
• If the point P lies inside von Mises cylinder, the
ik
P
du
m
material is elastic or a rigid body.
In
ha
P’
nd
• It becomes plastic only when the point P lies on
.S N
la
Dr
the surface of the cylinder.
ca
N’
ni
• Points outside of the cylinder are not defined.
e
ha O
ke
ec
• A similar discussion to Tresca prism as well.
or
fM
Ro
.o
IIT
pt
De
Metal Forming
39
Introduction to π-plane
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
The π-plane is passing through origin and equally inclined to all of the plane.
.o
IIT
pt
Metal Forming
40
Graphical representation of deviatoric plane
g.
ng
k
E
sa
al
Ba
ri
st
ik
du
m
In
ha
nd
.S
la
Dr
ca
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
De
Metal Forming
41
Other isotropic yield criteria
g.
Drucker yield criterion can be written as
ng
k
E
sa
𝑱𝟑𝟐 − 𝒄𝑱𝟐𝟑 = 𝑲𝟔
al
Ba
ri
st
ik
This is a homogenous function in stress of the sixth degree
du
m
27 9
In
for which c must lie in the range − ≤𝑐≤4
ha
8
nd
.S Drucker–Prager YC
la
Drucker–Prager yield criterion
Dr
ca
𝑱𝟐 = 𝐀 + 𝐁𝑰𝟏
ni
The constants A, B are determined from experiments.
e
ha
ke
ec
or
fM
Ro
𝑱𝟐 + 𝜶𝑰𝟏 = 𝒄𝟏
.o
IIT
pt
De
Metal Forming
Mohr-Coulomb YC
42
Yield criterion for anisotropic materials
g.
Hill48 anisotropic yield model
ng
k
E
sa
In 1948, Dr. R. Hill suggested first ever yield model for anisotropic materials. This is inspired
al
Ba
by von Mises yield model. This anisotropy yield model is popularly known as Hill48 anisotropic
ri
st
yield model.
ik
du
m
In
ha
𝑭 𝝈𝟐𝟐 − 𝝈𝟑𝟑 𝟐
+ 𝑮 𝝈𝟑𝟑 − 𝝈𝟏𝟏 𝟐
+ 𝑯 𝝈𝟏𝟏 − 𝝈𝟐𝟐 𝟐
+ 𝟐𝑳𝝈𝟐𝟐𝟑 + 𝟐𝑴𝝈𝟐𝟑𝟏 + 𝟐𝑵𝝈𝟐𝟏𝟐 = 𝟏
nd
.S
la
Dr
ca
Here, F, G, H, L, M and N are the anisotropic constants and these are the material parameters. If, X,
ni
Y and Z are the yield strength of the material w.r.t the principal direction of the anisotropy then the
e
following equations can be derived. ha
ke
ec
𝟏
or
𝑮 + 𝑯 = 𝑿𝟐 𝑓𝑜𝑟 𝜎11 = 𝑋, 𝑎𝑙𝑙 𝑜𝑡ℎ𝑒𝑟 = 0
fM
Ro
𝟏
.o
𝟏
De
Metal Forming
43
Yield criterion for anisotropic materials
g.
Hill48 anisotropic yield model
ng
k
From these set of equation F, G, H values can be found out. Such as
E
sa
al
𝟏 𝟏 𝟏
Ba
ri
𝟐𝑭 = 𝟐 + 𝟐 − 𝟐
st
𝒀 𝒁 𝑿
ik
𝟏 𝟏 𝟏
du
m
𝟐𝑮 = 𝟐 + 𝟐 − 𝟐
𝑿 𝒁 𝒀
In
ha
𝟏 𝟏 𝟏
nd
.S 𝟐𝑯 = 𝟐 + 𝟐 − 𝟐
𝑿 𝒀 𝒁
la
Dr
ca
ni
Similarly, if we assume that R, S, T are the yield strength of the material in pure shear condition w.r.t.
e
ha
the principal anisotropy direction then we can get
ke
ec
𝟏
or
𝟐𝑳 = 𝑹𝟐 𝑓𝑜𝑟 𝜎23 = 𝑅, 𝑎𝑙𝑙 𝑜𝑡ℎ𝑒𝑟 = 0
fM
Ro
𝟏
.o
𝟏
𝟐𝑵 = 𝟐 𝑓𝑜𝑟 𝜎12 = 𝑇, 𝑎𝑙𝑙 𝑜𝑡ℎ𝑒𝑟 = 0
De
Metal Forming
44
Yield criterion for anisotropic materials
g.
Hill48 anisotropic yield model
ng
k
E
sa
al
If anisotropy effect is completely neglected, then the material becomes isotropic and the condition
Ba
ri
for isotropy will become as
st
ik
du
𝟏 𝟑
𝟑𝑭 = 𝟑𝑮 = 𝟑𝑯 = 𝑳 = 𝑴 = 𝑵 = 𝟐𝑲𝟐 = 𝟐𝒀𝟐
In
ha
nd
.S
In that condition, the Hill48 anisotropic yield model reduces to von Mises isotropic yield model
la
Dr
ca
𝝈𝟏𝟏 − 𝝈𝟐𝟐 𝟐
+ 𝝈𝟐𝟐 − 𝝈𝟑𝟑 𝟐
+ 𝝈𝟑𝟑 − 𝝈𝟏𝟏 𝟐
+ 𝟔𝝈𝟐𝟏𝟐 + 𝟔𝝈𝟐𝟐𝟑 + 𝟔𝝈𝟐𝟑𝟏 = 𝟐𝒀𝟐 = 𝟔𝑲𝟐
ni
e
ha
ke
ec
or
fM
Ro
.o
IIT
pt
De
Metal Forming
45
Strain analyses
g.
ng
• A body is under deformation if there is a relative motion between the particles
E
sa
• Strain is the measurement of the deformation
al
Ba
ri
Lagrangian measurement of strain Initial configuration,
st
ik
du
m
R’
In
𝑃𝑄 ∥ 𝑥2 and 𝑃𝑄 ∥ 𝑑𝑥2 𝑃𝑅 ∥ 𝑥3 and 𝑃𝑅 ∥ 𝑑𝑥3
ha
nd
X3 R .S 𝜽 𝑃𝑅 ⊥ 𝑃𝑄 and ∡𝑅𝑃𝑄 = 90°
la
Q’
Dr
P’
ca
For deformation both length of the line segments
ni
P Q Deformed
and angle between the line segments are changed
e
ha
ke
Undeformed
ec
or
X2
fM
Ro
Longitudinal Strain: It is defined as change in 𝑃′ 𝑄′ − 𝑃𝑄 𝑃′ 𝑅′ − 𝑃𝑅
.o
𝐸22 = 𝐸33 =
length per unit length
IIT 𝑃𝑄 𝑃𝑅
pt
Shear Strain:
De
g.
R’
ng
𝑃 ≡ 𝑥1 , 𝑥2 , 𝑥3 , Q ≡ 𝑥1 , 𝑥2 + 𝑑𝑥2 , 𝑥3
E
sa
Q’
al
X3 𝜽 Let the displacement of the particle from P to P’ is U and displacement
Ba
P’
ri
R T of the particle from Q to Q’ is U+dU
st
ik
du
dx3
m
ഥ and 𝑄𝑇 + 𝑇𝑄′ = QQ′ so 𝑇𝑄′ = dU
𝑃′𝑄′ = 𝑃𝑄 + dU ഥ
In
ha
nd
P dx2 Q .S Let dU has three components dU1, dU2 and dU3 along X1, X2 and X3
directions, respectively.
la
X2
Dr
ca
𝜕𝑈1 𝜕𝑈1 𝜕𝑈 𝜕𝑈1
𝑑𝑈1 = 𝑑𝑥1 + 𝑑𝑥2 + 1 𝑑𝑥3 So 𝑑𝑈1 = 𝑑𝑥
ni
𝜕𝑥1 𝜕𝑥2 𝜕𝑥3 𝜕𝑥2 2
e
𝜕𝑈2 ha 𝜕𝑈3
ke
Similarly 𝑑𝑈2 = 𝑑𝑥 and 𝑑𝑈3 = 𝑑𝑥
ec
𝜕𝑥2 2 𝜕𝑥2 2
or
fM
Ro
Hence, the displacement vectors can be rewritten as ഥ = 𝑑𝑈1 𝑒1ҧ + 𝑑𝑈2 𝑒2ҧ +𝑑𝑈3 𝑒3ҧ
𝑑𝑈
.o
𝑃𝑄 = 𝑑𝑥2 . 𝑒2ҧ ഥ
IIT
𝑃′𝑄′ = 𝑃𝑄 + 𝑑𝑈
pt
Metal Forming
47
Strain-displacement relationship
g.
ng
k
E
sa
𝑃′ 𝑄′ − 𝑃𝑄
al
𝐸22 =
Ba
ri
𝑃𝑄
st
ik
du
m
2 2 2
1 𝜕𝑈2 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
In
ha
For infinitesimally small displacement 𝐸22 = 1 + . 2.
2
+ .
𝜕𝑥2 2 𝜕𝑥2
+
𝜕𝑥2
+
𝜕𝑥2
−1
nd
.S
la
Dr
ca
2 2 2
𝜕𝑈1 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
ni
𝐸11 = + . + +
e
𝜕𝑥1 2
ha 𝜕𝑥1 𝜕𝑥1 𝜕𝑥1
ke
ec
or
2 2 2
𝜕𝑈2 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
fM
𝐸22 = + . + +
Ro
𝜕𝑥2 2 𝜕𝑥2 𝜕𝑥2 𝜕𝑥2
.o
IIT
2 2 2
𝜕𝑈3 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
pt
𝐸33 = + . + +
𝜕𝑥3 2
De
Metal Forming
48
Strain-displacement relationship
g.
Shear strain calculation For infinitesimally small angle
ng
k
E
𝜋 𝜋 𝑃′ 𝑄′. 𝑃′ 𝑅′
sa
𝛾23 = − 𝜃 ≅ sin − 𝜃 = cos 𝜃 =
al
2 2 𝑃′ 𝑄′ . 𝑃′ 𝑅′
Ba
ri
st
𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
ik
du
𝑃′ 𝑄′ = 𝑑𝑥2 𝑒1ҧ + 𝑑𝑥2 + 𝑑𝑥2 𝑒2ҧ + 𝑑𝑥 𝑒ҧ
𝜕𝑥2 𝜕𝑥2 2 3
m
𝜕𝑥2
In
ha
nd
𝜕𝑈1
.S 𝜕𝑈2 𝜕𝑈3
𝑃′ 𝑅′ = 𝑑𝑥3 𝑒1ҧ + 𝑑𝑥3 𝑒2ҧ + 𝑑𝑥3 + 𝑑𝑥 𝑒ҧ
𝜕𝑥3 𝜕𝑥3 𝜕𝑥3 3 3
la
Dr
ca
ni
𝑃′ 𝑄′ = 𝑑𝑥2 1 + 𝐸22 Similarly 𝑃′ 𝑅′ = 𝑑𝑥3 1 + 𝐸33
e
ha
ke
ec
or
𝑃′ 𝑄′. 𝑃′ 𝑅′
fM
Ro
𝛾23 = cos 𝜃 =
𝑃′ 𝑄′ . 𝑃′ 𝑅′
.o
IIT
pt
De
Metal Forming
49
Strain-displacement relationship
g.
𝜕𝑈2 𝜕𝑈3 𝜕𝑈1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈2 𝜕𝑈3 𝜕𝑈3
ng
𝛾23 = + + . + . + .
𝜕𝑥3 𝜕𝑥2 𝜕𝑥2 𝜕𝑥3 𝜕𝑥3 𝜕𝑥2 𝜕𝑥2 𝜕𝑥3
E
sa
al
Ba
𝜕𝑈1 𝜕𝑈3 𝜕𝑈1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈2 𝜕𝑈3 𝜕𝑈3
ri
𝛾13 = + + . + . + .
st
𝜕𝑥3 𝜕𝑥1 𝜕𝑥1 𝜕𝑥3 𝜕𝑥1 𝜕𝑥3 𝜕𝑥1 𝜕𝑥3
ik
du
m
𝜕𝑈1 𝜕𝑈2 𝜕𝑈1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈2 𝜕𝑈3 𝜕𝑈3
In
ha
𝛾12 = + + . + . + .
𝜕𝑥2 𝜕𝑥1 𝜕𝑥1 𝜕𝑥2 𝜕𝑥1 𝜕𝑥2 𝜕𝑥1 𝜕𝑥3
nd
.S
la
Dr
ca
2 2 2
𝜕𝑈1 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
ni
𝐸11 = + . + +
𝜕𝑥1 2 𝜕𝑥1 𝜕𝑥1 𝜕𝑥1
e
ha
ke
ec
2 2 2
or
𝜕𝑈2 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
fM
𝐸22 = + . + +
Ro
𝜕𝑥2 2 𝜕𝑥2 𝜕𝑥2 𝜕𝑥2
.o
IIT
2 2 2
𝜕𝑈3 1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈3
pt
𝐸33 = + . + +
𝜕𝑥3 2 𝜕𝑥3 𝜕𝑥3 𝜕𝑥3
De
Metal Forming
50
Strain-displacement relationship
g.
ng
k
E
Hence strain tensor can be written as
sa
al
Ba
𝐸11 𝐸12 𝐸13
ri
1
st
𝐸𝑖𝑗 = 𝐸21 𝐸22 𝐸23 Where, 𝐸12 = 𝐸21 = γ12
ik
2
du
𝐸31 𝐸32 𝐸33
In
ha
1 𝜕𝑈𝑖 𝜕𝑈𝑗 1 𝜕𝑈𝑚 𝜕𝑈𝑚
In generalized terms
nd
.S 𝐸𝑖𝑗 = +
2 𝜕𝑋𝑗 𝜕𝑋𝑖
+ .
2 𝜕𝑋𝑖 𝜕𝑋𝑗
la
Dr
Linear terms Non-linear terms
ca
ni
For infinitesimally small strain, we can neglect the non-linear terms.
e
ha
ke
ec
or
fM
Ro
𝟏 𝝏𝑼𝒊 𝝏𝑼𝒋
𝜺𝒊𝒋 = +
.o
𝟐 𝝏𝑿𝒋 𝝏𝑿𝒊
IIT
pt
De
Metal Forming
51
Strain-displacement relationship
g.
ng
Hence strain tensor can be written as
E
ε11 ε12 ε13
sa
al
ε𝑖𝑗 = ε21 ε22 ε23
Ba
ri
ε31 ε32 ε33
st
ik
du
m
In
The components of strain tensor are
ha
nd
𝜕𝑈1 .S 1 𝜕𝑈1 𝜕𝑈2 1 𝜕𝑈1 𝜕𝑈3
ε11 = ε12 = + ε13 = +
la
𝜕𝑋1 2 𝜕𝑋2 𝜕𝑋1 2 𝜕𝑋3 𝜕𝑋1
Dr
ca
ni
1 𝜕𝑈1 𝜕𝑈2 𝜕𝑈2 1 𝜕𝑈2 𝜕𝑈3
ε21 = + ε22 = ε23 = +
e
2 𝜕𝑋2 𝜕𝑋1 ha𝜕𝑋2 2 𝜕𝑋3 𝜕𝑋2
ke
ec
or
fM
Ro
1 𝜕𝑈1 𝜕𝑈3 1 𝜕𝑈2 𝜕𝑈3 𝜕𝑈3
.o
Metal Forming
Principal strains
g.
ng
For pure deformation case 𝑑𝑈𝑖 = ε𝑖𝑗 𝑑𝑥𝑗
E
sa
al
Ba
ri
𝑑𝑈
Let principal strain = ε =
st
𝑑𝑅
ik
du
m
The dU has component in dU1, dU2 and dU3 in X1, X2 and X3 direction, respectively.
In
ha
nd
.S ε=
𝑑𝑈 𝑑𝑈1 𝑑𝑈2 𝑑𝑈3
= = =
la
𝑑𝑅 𝑑𝑋1 𝑑𝑋2 𝑑𝑋3
Dr
ca
ni
𝑑𝑋1 𝑑𝑋2 𝑑𝑋3
And direction cosine 𝑛2 = 𝑛3 =
e
𝑛1 =
ha 𝑑𝑅 𝑑𝑅
ke
𝑑𝑅
ec
or
fM
Ro
From previous relationship
.o
IIT
𝑑𝑈𝑖 = ε 𝑑𝑥𝑖 = ε𝑖𝑗 𝑑𝑥𝑗
pt
De
g.
ng
These equation can be written as 𝜀11 − ε ε21 ε31
E
sa
𝜀12 𝜀22 − 𝜀 𝜀32 = 0
al
𝑑𝑒𝑡 ε𝒊𝒋 − ε𝜹𝒊𝒋 = 0 𝜀13 𝜀23 𝜀33 − 𝜀
Ba
ri
st
ik
du
The above equation, after expanding can be rewritten as
In
ha
𝜺𝟑 − 𝑰′𝟏 𝜺𝟐 + 𝑰′𝟐 𝜺 − 𝑰′𝟑 = 0
nd
.S
Where, I’1, I’2 and I’3 are the coefficients of this equation and have values as given below
la
Dr
ca
𝐼′1 = 𝜀11 + 𝜀22 + 𝜀33 = tr(𝜀𝑖𝑗 )
ni
e
ha 1 2
ke
2 2 2 2
𝐼′2 = −𝜀12 − 𝜀23 − 𝜀31 + 𝜀11 𝜀22 + 𝜀22 𝜀33 + 𝜀33 𝜀11 = tr(𝜀𝑖𝑗 ) − tr(𝜀𝑖𝑗 )
ec
2
or
fM
2 2 2
𝐼′3 = 𝜀11 𝜀22 𝜀33 + 2𝜀12 𝜀23 𝜀31 − 𝜀12 𝜀33 − 𝜀23 𝜀11 − 𝜀31 𝜀22 = det(𝜀𝑖𝑗 )
Ro
.o
IIT
pt
Metal Forming
Volumetric Strains
g.
The dilatation (the relative variation of the volume) is the first strain invariant or trace of the tensor
ng
k
E
∆𝑉
sa
𝛿= = 𝐼′1 = 𝜀11 + 𝜀22 + 𝜀33
al
𝑉0
Ba
ri
st
ik
du
If we consider a cube with an edge length a, it is a quasi-cube after the deformation
In
ha
nd
.S
la
∆𝑉 𝑎 1 + 𝜀11 . 𝑎 1 + 𝜀22 . 𝑎 1 + 𝜀33 − 𝑎3
Dr
= = 𝜀11 + 𝜀22 + 𝜀33 + 𝜀11 . 𝜀22 + 𝜀22 . 𝜀33 + 𝜀33 . 𝜀11 + 𝜀11 . 𝜀22 . 𝜀33
𝑉0 𝑎3
ca
ni
e
as we consider small deformations, ha
ke
ec
∆𝑉
or
= 𝜀11 + 𝜀22 + 𝜀33 = 3𝜀ℎ𝑦𝑑
fM
𝑉0
Ro
.o
IIT
pt
De
Metal Forming
55
Strain deviator tensor
g.
The infinitesimal strain tensor 𝜀𝑖𝑗 , similarly to the Cauchy stress tensor, can be expressed as the sum
ng
of two other tensors:
E
sa
al
𝜺𝒊𝒋 = 𝜺′𝒊𝒋 + 𝜺𝒉𝒚𝒅 𝜹𝒊𝒋
Ba
ri
st
where
ik
du
m
𝜀11 + 𝜀22 + 𝜀33
In
𝜺𝑴 = 𝜺𝒉𝒚𝒅 =
ha
3
nd
and, .S
la
Dr
′ ′ ′
𝜀11 𝜀12 𝜀13 𝜀11 𝜀12 𝜀13 𝜀𝑀 0 0 𝜀11 − 𝜀𝑀 𝜀12 𝜀13
ca
′
𝜀21 ′
𝜀22 ′
𝜀23 = 𝜀21 𝜀22 𝜀23 − 0 𝜀𝑀 0 = 𝜀21 𝜀22 − 𝜀𝑀 𝜀23
ni
′
𝜀31 ′
𝜀32 ′
𝜀33 𝜀31 𝜀32 𝜀33 0 0 𝜀𝑀 𝜀31 𝜀32 𝜀33 − 𝜀𝑀
e
ha
ke
ec
or
fM
Ro
Similar to stress tensor, deviatoric strain tensor has the invariants of J’1, J’2 and J’3.
.o
IIT
pt
De
Metal Forming
56
Octahedral strains
g.
ng
Let (n1,n2,n3) be the directions of the three principal strains. The shear strain on an octahedral plane
E
sa
is called the octahedral shear strain and is given by
al
Ba
ri
1
st
2 2 2 2 2 2 0.5
𝜀𝑠,𝑜𝑐𝑡 = 𝜀 − 𝜀22 + 𝜀22 − 𝜀33 + 𝜀33 − 𝜀11 + 6 𝜀12 + 𝜀23 + 𝜀31
3 11
ik
du
The normal strain on an octahedral plane is given by
In
ha
𝜀11 + 𝜀22 + 𝜀33
nd
.S 𝜀𝑛,𝑜𝑐𝑡 =
3
la
Dr
ca
The effective strain can be mathematically written as
ni
𝜀 ҧ = 2𝜀𝑠,𝑜𝑐𝑡
e
ha
ke
ec
or
fM
Ro
2 2 2 0.5
𝜀ҧ = 𝜀1 − 𝜀2 + 𝜀2 − 𝜀3 + 𝜀3 − 𝜀1
3
.o
IIT
pt
De
Ro In
or du
ke st
e ri al
E ng
g.
Metal Forming
58