Assignment 1 Final-Ahmad Firouzian Nejad
Assignment 1 Final-Ahmad Firouzian Nejad
Submitted to
Professor Allan T. Dolovich
Submitted by
Ahmad Firouzian Nejad
NSID: tfj859
1|Page
In course material, we already derived the differential equation for the linear elastic small deflection of
the pure-bending beam as follows:
2 2
d v d v M
M =EI 2 → 2 = (1)
dx dx EI
Where E and I are Yong’ module and second moments of area respectively,
We can solve the given problem by integrating it as follow:
2
∫ ddxv2 dx =∫ EI
M dv M
dx → = x +C 1
dx EI
(2)
Then
dv M M
∫ dx dx=∫ ( EI x+C 1)dx → v ( x )= 2 EI x 2 +C 1 x +C 2 (3)
Where C1 and C2 are the integral coefficients and can be determined by meeting the boundary conditions
for a cantilever beam at x=0 as follow:
v ( 0 )=0 →C 2=0
dv (4)
( 0 )=0 →C 1=0
dx
So, the vertical deflection of the neutral surface of the beam will be:
2|Page
M 2
v ( x )= x (5)
2 EI
The vertical deflection of the neutral surface at end point C can be calculated as follows:
M 2
v c =v ( L )= L (6)
2 EI
It should be noted that the second moment of area for a circle cross section in respect to z-axis (see figure
1) can be calculated as following:
Where r and d are radius and diameter of circular cross section, respectively.
3|Page
i) Geometric approaches:
As we know, the deformed shape of a beam under constant pure bending has a constant curvature radius,
EI
¿ , similar to a circle (see figure 1). The general equation of a circle is as following:
M
2 2 2
(x−x 0 ) +( y− y 0) =ρ (1)
4|Page
Figure 2. Cartesian coordinate system with a circle of radius ρ centered at (0,ρ)
2
d y
For 2
>0 (i.e. 0≤θ≤π/2 or 3π/2 ≤θ≤2π) → v ( x )= y ( x )=ρ−√ ρ2−x 2
dx
(3)
2
d y
For 2
<0 (i.e. π/2≤θ≤3π/2) → v ( x )= y ( x )=ρ+ √ ρ2−x 2
dx
It should be noted that x and y in the equation (3) are spatial coordinates system, so to calculate the
deflection of an arbitrary point X in neutral line of the beam, we need to find x in terms of X. From
geometry (see figure 2), we know that:
X EI MX (4)
x=ρ sin θ=ρ sin = sin ( )
ρ M EI
X MX
Where θ is the angle with respect to -y axis in Radians and can be calculated by θ= = in which X
ρ EI
is a location of arbitrary point along neutral-line of beam (in material coordinate system). So now, we can
rewrite the equations (3) as following:
√ X
ρ
X
ρ
X
v ( x )=ρ−√ ρ2−x 2=ρ− ρ2−ρ2 sin2 =ρ−ρ 1−sin2 =ρ(1−cos )
ρ √
√ X
(*) 1−sin 2 =cos
ρ
X
ρ
√
v ( x )=ρ+ √ ρ 2−x 2=ρ+ ρ2−ρ2 sin2
X
ρ √X
= ρ+ ρ 1−sin2 = ρ(1−cos )
ρ
X
ρ
(**)
√ X
1−sin 2 =−cos
ρ
X because cos
ρ
X
ρ
is minus in this angle range
5|Page
So the deflection of an arbitrary point X in neutral line of the beam can be calculated by:
(
v ( θ )= y (θ )=ρ ( 1−cos θ ) → v ( X )= y ( X )= ρ 1−cos
X
ρ) EI
= (1−cos
M
MX
EI
) (6)
ML
If θ L = is too large (i.e. ≥ 2π), then the deformed shape of beam will be a circle. It should be noted
EI
that vertical deflection could also be derived directly from geometric relationships as well as the
relationship between Cartesian and polar coordinate systems (see figure 2).
6|Page
II. Integrating the general nonlinear curvature equation:
| |
2
d v
2
1 dx
κ= = 2 3
ρ dv 2
[1+( )]
dx (1)
2 2
d v −d v
2 2 2 2
d v dx d v dx
a) if 2
≥ 0 κ= 2 3
and b) if 2
≤ 0 κ= 2 3
dx dv dx dv
[1+( ) ] 2 [1+( ) ] 2
dx dx
2
d v
2 2
d v dx
a ¿ if 2
≥ 0 κ= 2 3
dx dv
[1+( ) ] 2
dx
7|Page
2 2
u 2 2 u +1−1 2 2 1 2 2 2 2 1
2
=κ x → 2
=κ x → 1− 2
=κ x → 1−κ x = 2
1+ u 1+u 1+u 1+u
1 1 κ x κx
2 2 (6)
2 2
1+u = →u = −1= →u=±
2 2
1−κ x
2 2
1−κ x
2 2
1−κ x √1−κ2 x 2
By considering (*), the only acceptable answer for u will be:
κx
u=
√ 1−κ2 x2
(7)
dv κx κ x dx
= → dv=
dx √ 1−κ2 x 2 √1−κ2 x 2
If we integral form both sides of above equation, we will have:
κ x dx −1
∫ dv=+∫ → v (x )= √1−κ 2 x 2+ C2
√1−κ 2
x 2 κ
(8)
v ( x )=− ρ 1−
√x 2
ρ ()
+C2 =−√ ρ2−x 2+ C2
To meet the boundary condition at x=0 , the deflection of beam should be zero so v (0)=0
2
−d v
2 2
d v dx
b ¿ if 2
≤ 0 κ= 2 3
dx dv 2
[1+( ) ]
dx
The problem-solving method is exactly the same as above but we only need to pay attention to the sign.
So we can summarize it.
dv (10)
u=
dx
So we have:
8|Page
du
−
1 dx −du (11)
κ= = → =κ dx
ρ 3 3
[ 1+ u ] 2 2
[1+u ]
2 2
L
To meet the boundary condition, i.e. @ x=0 → v (0)=2 ρ (It happens when ≥ π)
ρ
v ( 0 )=ρ+C 2=2 ρ→ C 2=+ ρ
(17)
v ( x )=ρ+ √ ρ 2−x 2
For summary:
9|Page
2
d v
If if 2
≥0 → v ( x )=ρ−√ ρ2−x 2
dx
2 (18)
d v
If if 2
≤0 → v ( x )=ρ+ √ ρ −x
2 2
dx
As we see, the answer is exactly the same as obtained answer in section (i)
10 | P a g e
Table 1. Comparison of linear and nonlinear theories for pure bending beam deflection
Small deformation in beam theory refers to the assumption that the deformations of the beam under load
are small enough that the geometry of the beam does not change significantly. This assumption is also
known as the linear elasticity assumption and valid for small deflections of the materials that behave
linearly under stress. In this condition, the deflection predicted by linear theory is typically within a few
percent of the actual deflection (see the results in table 1 for M=10 & 20 in.lb). Linear theory may not
accurately predict the behavior of the beam under large deflections (see results in table 1 for M=30 & 40
in.lb). As we can see for all the moments, the linear theory over predicts the deflection of the end beam
when compared to the nonlinear theory and the percent difference grows by increasing the moment. The
linear theory for the pure bending beam has several limitations, which can affect its accuracy in certain
cases. Some of these limitations include:
- Small deflection assumption: The linear theory assumes that x ≈ X so u c=0 and the deflection of the
beam is small compared to the length of the beam. This assumption may not hold true for beams that
undergo large deflections, where the nonlinear behavior of the beam cannot be ignored.
d2v
dv
≪1 1 dx 2 1 d2 v
so κ= = → κ= ≃ (1)
dx ρ dv 2 32 ρ dx2
[1+( ) ]
dx
- Homogeneity assumption: The linear theory assumes that the material of the beam is homogeneous
which means that the material properties do not vary with position. In reality, some materials like
11 | P a g e
functionally graded materials (FGMs) have some degree of variation in their properties, which can affect
the accuracy of the analysis.
M =∫ Y ( EYρ ) dA= Eρ ∫ Y dA
2
(2)
- Elasticity assumption: The linear theory assumes that the material of the beam behaves elastically,
meaning that the deformation is reversible and directly proportional to the applied load. This assumption
may not hold true for materials that exhibit plastic deformation or other nonlinear behavior.
−EY
σ =Eε= (3)
ρ
Here, we also assumed that the cross-section of the beam is uniform along its length (i.e. I≠I(x) & I=cte).
2
∫ ddxv2 dx =∫ EI
M 1
dx= ∫ Mdx
EI (4)
In reality, many beams have non-uniform cross-sections, which can affect the distribution of stresses and
deflections along the length of the beam.
12 | P a g e
Figure 1. Deformed configuration of the beam under M=10 in.lb
13 | P a g e
As it can be seen in figure 1 and 2, the linear theory over predicts the deflection of the beam
when compared to the nonlinear theory.
For a cantilevered beam under a pure moment, the deflection predicted by linear theory would be
a parabolic curve, while the deflection predicted by nonlinear theory would be a part of the
circle. In the linear theory, we assume that the deflection and slope of the beam are so small such
that:
2
d v
2 2
dv 1 dx 1 d v
≪1 so κ= = → κ= ≃ (1)
dx ρ dv
2 3
ρ dx2
[1+( ) ] 2
dx
In other words, each point on the neutral line in linear theory has only y-displacement (v) and
zero x- displacement u=0 (i.e. x ≈ X), whereas for nonlinear theory, each point can have both x
and y-displacements.
It should be noted that when the load, and therefore the deflection is large, the Euler–Bernoulli
beam theory's starting assumption "which is neutral plane does not deform" is violated for linear
method. In fact, if we calculate the length of the neutral line using the arc length formula (see
equation 2), we notice that it is stretched, not constant.
Arclength=∫ 1+(
√ dy 2
dx
) dx (2)
In addition, the linear theory is not capable of predicting the deformed shape of a cantilevered
beam when it is subjected to very large moments. The linear theory predicts a deflection as a
parabolic function while experimental evidence and general theory show that the deformed shape
of the beam become even circular, so these large deformations cannot be accurately predicted
using linear theory (see figure 2).
14 | P a g e
According to the course material the longitudinal stress σ can be calculated for both linear and general
methods by the following formula:
−MY (1)
σ=
I
−Md d
σ min= =−3259.493 psi@ Y = =0.25∈¿
2I 2 (2)
Md −d
σ max= =3259.493 psi@ Y = =−0.25∈¿
2I 2
15 | P a g e
List of key points
LIST ALL SELECTED KEYPOINTS. DSYS= 0
NO. X,Y,Z LOCATION KESIZE NODE ELEM MAT REAL TYP ESYS
1 0.00 0.00 0.00 0.00 1 0 0 0 0 0
2 15.0 0.00 0.00 0.00 2 0 0 0 0 0
List of nodes
NODE X Y Z
1 0.00000000000 0.00000000000 0.00000000000
2 15.0000000000 0.00000000000 0.00000000000
3 1.00000000000 0.00000000000 0.00000000000
4 2.00000000000 0.00000000000 0.00000000000
5 3.00000000000 0.00000000000 0.00000000000
6 4.00000000000 0.00000000000 0.00000000000
7 5.00000000000 0.00000000000 0.00000000000
8 6.00000000000 0.00000000000 0.00000000000
9 7.00000000000 0.00000000000 0.00000000000
10 8.00000000000 0.00000000000 0.00000000000
11 9.00000000000 0.00000000000 0.00000000000
12 10.0000000000 0.00000000000 0.00000000000
13 11.0000000000 0.00000000000 0.00000000000
14 12.0000000000 0.00000000000 0.00000000000
15 13.0000000000 0.00000000000 0.00000000000
16 14.0000000000 0.00000000000 0.00000000000
16 | P a g e
Figure 1. The beam model and node number
List of elements
1 1 1 1 0 1 1 3 0
2 1 1 1 0 1 3 4 0
3 1 1 1 0 1 4 5 0
4 1 1 1 0 1 5 6 0
5 1 1 1 0 1 6 7 0
6 1 1 1 0 1 7 8 0
7 1 1 1 0 1 8 9 0
8 1 1 1 0 1 9 10 0
9 1 1 1 0 1 10 11 0
10 1 1 1 0 1 11 12 0
11 1 1 1 0 1 12 13 0
12 1 1 1 0 1 13 14 0
13 1 1 1 0 1 14 15 0
14 1 1 1 0 1 15 16 0
15 1 1 1 0 1 16 2 0
17 | P a g e
Figure 2. The beam model and element number
List of loads
LIST NODAL FORCES FOR SELECTED NODES 1 TO 16 BY 1
CURRENTLY SELECTED NODAL LOAD SET= FX FY FZ MX MY MZ
NODE LABEL REAL IMAG
2 MZ 40.0000000 0.00000000
18 | P a g e
Linear theory
SOLUTION OPTIONS
PROBLEM DIMENSIONALITY. . . . . . . . . . . . .3-D
DEGREES OF FREEDOM. . . . . . UX UY UZ ROTX ROTY ROTZ
ANALYSIS TYPE . . . . . . . . . . . . . . . . .STATIC (STEADY-STATE)
GLOBALLY ASSEMBLED MATRIX . . . . . . . . . . .SYMMETRIC
LOAD STEP OPTIONS
LOAD STEP NUMBER. . . . . . . . . . . . . . . . 1
TIME AT END OF THE LOAD STEP. . . . . . . . . . 1.0000
NUMBER OF SUBSTEPS. . . . . . . . . . . . . . . 1
STEP CHANGE BOUNDARY CONDITIONS . . . . . . . .DEFAULT
PRINT OUTPUT CONTROLS . . . . . . . . . . . . .NO PRINTOUT
DATABASE OUTPUT CONTROLS. . . . . . . . . . . .ALL DATA WRITTEN
FOR THE LAST SUBSTEP
NODE UX
1 0.0000
2 0.0000
3 0.0000
4 0.0000
5 0.0000
6 0.0000
7 0.0000
8 0.0000
9 0.0000
10 0.0000
11 0.0000
12 0.0000
13 0.0000
14 0.0000
15 0.0000
16 0.0000
MAXIMUM ABSOLUTE VALUES
NODE 0
VALUE 0.0000
19 | P a g e
PRINT U NODAL SOLUTION PER NODE
***** POST1 NODAL DEGREE OF FREEDOM LISTING *****
LOAD STEP= 1 SUBSTEP= 1
TIME= 1.0000 LOAD CASE= 0
THE FOLLOWING DEGREE OF FREEDOM RESULTS ARE IN THE GLOBAL COORDINATE SYSTEM
NODE UY
1 0.0000
2 14.698
3 0.65325E-001
4 0.26130
5 0.58792
6 1.0452
7 1.6331
8 2.3517
9 3.2009
10 4.1808
11 5.2913
12 6.5325
13 7.9043
14 9.4068
15 11.040
16 12.804
MAXIMUM ABSOLUTE VALUES
NODE 2
VALUE 14.698
20 | P a g e
PRINT ELEMENT TABLE ITEMS PER ELEMENT
***** POST1 ELEMENT TABLE LISTING *****
STAT CURRENT CURRENT CURRENT CURRENT
ELEM SMIS32 SMIS33 SMIS34 SMIS35
1 -3266.2 3266.2 -0.86750E-013 0.86750E-013
2 -3266.2 3266.2 -0.86750E-013 0.86750E-013
3 -3266.2 3266.2 -0.86750E-013 0.86750E-013
4 -3266.2 3266.2 -0.86750E-013 0.86750E-013
5 -3266.2 3266.2 -0.86750E-013 0.86750E-013
6 -3266.2 3266.2 -0.86750E-013 0.86750E-013
7 -3266.2 3266.2 -0.86750E-013 0.86750E-013
8 -3266.2 3266.2 -0.86750E-013 0.86750E-013
9 -3266.2 3266.2 -0.86750E-013 0.86750E-013
10 -3266.2 3266.2 -0.86750E-013 0.86750E-013
11 -3266.2 3266.2 -0.86750E-013 0.86750E-013
12 -3266.2 3266.2 -0.86750E-013 0.86750E-013
13 -3266.2 3266.2 -0.86750E-013 0.86750E-013
14 -3266.2 3266.2 -0.86750E-013 0.86750E-013
15 -3266.2 3266.2 -0.86750E-013 0.86750E-013
MINIMUM VALUES
ELEM 15 1 1 1
VALUE -3266.2 3266.2 -0.86750E-013 0.86750E-013
MAXIMUM VALUES
ELEM 1 15 1 1
VALUE -3266.2 3266.2 -0.86750E-013 0.86750E-013
Figure 5. Deformed shape of the beam and longitudinal bending stress contour plot
21 | P a g e
Figure 6. SMIS32 contour plot (bending stress on the top of the beam)
Figure 7. SMIS33 contour plot (bending stress on the bottom of the beam)
22 | P a g e
Nonlinear theory
SOLUTION OPTIONS
PROBLEM DIMENSIONALITY. . . . . . . . . . . . .3-D
DEGREES OF FREEDOM. . . . . . UX UY UZ ROTX ROTY ROTZ
ANALYSIS TYPE . . . . . . . . . . . . . . . . .STATIC (STEADY-STATE)
NONLINEAR GEOMETRIC EFFECTS . . . . . . . . . .ON
NEWTON-RAPHSON OPTION . . . . . . . . . . . . .PROGRAM CHOSEN
GLOBALLY ASSEMBLED MATRIX . . . . . . . . . . .SYMMETRIC
LOAD STEP OPTIONS
LOAD STEP NUMBER. . . . . . . . . . . . . . . . 1
TIME AT END OF THE LOAD STEP. . . . . . . . . . 1.0000
NUMBER OF SUBSTEPS. . . . . . . . . . . . . . . 10
MAXIMUM NUMBER OF EQUILIBRIUM ITERATIONS. . . . 15
STEP CHANGE BOUNDARY CONDITIONS . . . . . . . .DEFAULT
TERMINATE ANALYSIS IF NOT CONVERGED . . . . . .YES (EXIT)
CONVERGENCE CONTROLS. . . . . . . . . . . . . .USE DEFAULTS
PRINT OUTPUT CONTROLS . . . . . . . . . . . . .NO PRINTOUT
DATABASE OUTPUT CONTROLS. . . . . . . . . . . .ALL DATA WRITTEN
FOR THE LAST SUBSTEP
23 | P a g e
Figure 8. Deformed shape of the beam and x displacement contour plot
24 | P a g e
Figure 9. Deformed shape of the beam and y displacement contour plot
Stress Analysis
PRINT ELEMENT TABLE ITEMS PER ELEMENT
***** POST1 ELEMENT TABLE LISTING *****
STAT CURRENT CURRENT CURRENT CURRENT
ELEM SMIS32 SMIS33 SMIS34 SMIS35
1 -3266.2 3266.2 -0.25822E-007 0.25822E-007
2 -3266.2 3266.2 -0.23874E-007 0.23874E-007
3 -3266.2 3266.2 -0.21926E-007 0.21926E-007
4 -3266.2 3266.2 -0.19980E-007 0.19980E-007
5 -3266.2 3266.2 -0.18034E-007 0.18034E-007
6 -3266.2 3266.2 -0.16091E-007 0.16091E-007
7 -3266.2 3266.2 -0.14151E-007 0.14151E-007
8 -3266.2 3266.2 -0.12215E-007 0.12215E-007
9 -3266.2 3266.2 -0.10285E-007 0.10285E-007
10 -3266.2 3266.2 -0.83620E-008 0.83620E-008
11 -3266.2 3266.2 -0.64475E-008 0.64475E-008
12 -3266.2 3266.2 -0.45431E-008 0.45431E-008
13 -3266.2 3266.2 -0.26486E-008 0.26486E-008
14 -3266.2 3266.2 -0.76234E-009 0.76234E-009
15 -3266.2 3266.2 0.11191E-008-0.11191E-008
MINIMUM VALUES
ELEM 14 1 1 15
VALUE -3266.2 3266.2 -0.25822E-007-0.11191E-008
MAXIMUM VALUES
ELEM 1 14 15 1
VALUE -3266.2 3266.2 0.11191E-008 0.25822E-007
25 | P a g e
Figure 10. Deformed shape of the beam and longitudinal bending stress contour plot
Figure 11. SMIS32 contour plot (bending stress on the top of the beam)
26 | P a g e
Figure 12. SMIS33 contour plot (bending stress on the bottom of the beam)
27 | P a g e
Table 1. Ansys results for displacements
Ansys Ansys
X Linear method Nonlinear method
(in)
u (in) v (in) u (in) v (in)
0 0 0 0 0
1 0 6.53E-02 -2.13E-03 6.53E-02
2 0 0.2613 -2.13E-02 0.26
3 0 0.58792 -7.41E-02 0.58085
4 0 1.0452 -0.17688 1.0224
5 0 1.6331 -0.34479 1.577
6 0 2.3517 -0.59204 2.2353
7 0 3.2009 -0.93148 2.9861
8 0 4.1808 -1.3743 3.8165
9 0 5.2913 -1.9302 4.7124
10 0 6.5325 -2.6065 5.6586
11 0 7.9043 -3.4088 6.6389
12 0 9.4068 -4.3405 7.6365
13 0 11.04 -5.4028 8.6346
14 0 12.804 -6.5946 9.616
15 0 14.698 -7.9126 10.564
28 | P a g e
Table 2. Comparison between theory and Ansys results
Linear Nonlinear
Parameter
Theory Ansys Error* Theory Ansys Error*
uc (in) 0 0 0 -7.8913 -7.9126 -0.26919
vc (in) 14.6677 14.698 -0.002061 10.5497 10.564 -0.13537
σmax (psi) 3259.493 3266.24 -0.002066 3259.493 3266.24 -0.20657
(Theory −Ansys )
* Error= ×100
Ansys
Figure 13. Comparison of theoretical and Ansys results for deformed beam configuration
As we can see in figure 13, both the linear and nonlinear theoretical methods accord well with
the corresponding results (displacement and stress) in Ansys. The relative error is less than 1%.
Each point on the neutral line in linear method (both theory and Ansys) has only y-displacement
29 | P a g e
(v) and zero x- displacement u=0 (i.e. x ≈ X), whereas for nonlinear method (both theory and
Ansys), each point can have both x and y-displacements. In general, the linear method (both
theory and Ansys) overpredicts the y-displacement for the same node when compared to the
nonlinear method.
Moreover, for M=40 in.lb, the linear method (both theory and Ansys) is incapable of accurately
predicting the deformed shape of a cantilevered beam. The linear method predicts a deflection as
a parabolic function, but nonlinear method (both theory and Ansys) predicts a circular shape for
the deformed beam.
30 | P a g e