Precise Finite-Element Model for Pulleys Based

on the Hamiltonian Form of Elasticity

Xiangjun Qiu1; Chang Chai2; and Ryan Lemmon3

Abstract: Conveyor pulleys are typical axisymmetric structures subjected to nonaxisymmetric loading. Taking full advantage of the character-
istics of the pulley structure, this paper presents a precise finite-element formulation for pulley stress analysis based on the Hamiltonian form of
elasticity. In the model, the solution is expanded into a set of Fourier series; a paired set of state variables are selected from the Fourier coef-
ficients; and the governing equations are reorganized in Hamiltonian form with the use of the paired state variables. The general solutions to
the Hamiltonian system can be obtained numerically and formulated into a finite-element model, from which the final stress solution for a pulley
can be found. Numerical examples show that this method is much more efficient than the conventional FEM with comparable accuracy.
DOI: 10.1061/(ASCE)EM.1943-7889.0000746. © 2014 American Society of Civil Engineers.
Author keywords: Pulley; Fourier series; Hamiltonian system; Precise finite-element method (FEM); Timoshenko beam; Shell and plate.

Introduction of a conveyor pulley is shortened from a few days in the 1990s to less
than 1 h today.
A conveyor pulley consists of rim, disk, and shaft components as- Still, designers never stop seeking more efficient methods for
sembled through weld and locking devices (Fig. 1). In a belt con- pulley stress analysis for the purpose of efficient design optimization
veyor system, pulleys play important roles as drive pulley, tail among various options. Qiu and Sethi (1993) developed a very effi-
pulley, take-up pulley, and bend pulley. In a typical, long-distance cient and effective analytical model for pulley stress distribution
conveyor system, where the belt tension can be as high as several named modified transfer matrix method (MTMM). It uses the paired
million Newtons, a pulley is subjected to severe bending and tor- displacement-stress variables to reorganize the governing equations,
sional load. Failure of a pulley in a conveyor system may be cat- expresses the general solution in the form of transfer matrix, and
astrophic. Therefore, accurate and efficient modeling of stress further modifies the transfer matrix form into a finite-element for-
distribution in a pulley is of significant scientific and engineering mulation. But this method relies on the availability of the analytical
value. The effort of modeling stress distribution in a pulley structure solution of the governing equation for each pulley component. In case
can be traced to the 1960s, when Lange (1963) tried to use an the governing equations are too complicated to obtain an analytical
approximate analytical method to estimate the conveyor pulley solution, an approximate solution has to be used as a substitute. In Qiu
stress. But modern FEM results have proven that Lange’s work was and Sethi’s (1993) work, the major approximate solution employed
very inaccurate (Qiu and Sethi 1993). was Timoshenko and Woinowsky-Krieger’s (1959) approximate
Currently, according to Patel et al. (2011), the most popular solution to the cylindrical shell, in which the terms of circumferential
method used for pulley stress analysis is the FEM. Ravikumar and displacement in strain-displacement relations are neglected for
Chattopadhyay (1999) developed a two-dimensional (2D) finite- mathematical convenience. The consequence of this approximation is
element model based on Fourier series for pulley stress analysis. that four of six modes of rigid body motions are no longer preserved in
Martins and Kovesdy (2012) summarized the use of three- the solution and errors are unavoidable in certain cases.
dimensional (3D) solid elements to solve the conveyor pulley This paper modifies the Qiu and Sethi (1993) model by (1)
stress problem. Many people reported the use of commercial FEM presenting the governing equations for each pulley component in
software to compute the pulley stress distributions (Sethi and a form of Hamiltonian system without any approximation, (2) de-
Nordell 1993; Pathan et al. 2011). As computer hardware technology riving the general solution to the Hamiltonian system using a precise
advances, the computational time for completion of a FEM analysis numerical integration method, (3) reorganizing the general solution
in finite-element form, and (4) numerically solving the finite-element
equations for final stress solution. This modification leads to a new
method named precise FEM (PFEM). It must be pointed out that the
1
Chief Scientist, High Fidelity Simulation, Metso Minerals, 4820 Cen- fundamental ideas of PFEM are similar to those used in Qiu’s (2006,
tennial Blvd., Colorado Springs, CO 80919. E-mail: Xiangjun.Qiu@metso 2009) boundary element models for rolling resistance. The in-
.com terrelation of the model presented in this paper to the rolling re-
2
Associate Professor, School of Machinery and Automobile Engineer- sistance models (Qiu 2006, 2009) is addressed subsequently.
ing, Hefei Univ. of Technology, Hefei, Anhui 230009, China (correspond-
ing author). E-mail: [email protected] Governing Equations, Hamiltonian Systems,
3
President, Advanced Conveyor Technologies Inc., 3911 E. 132 N,
and PFEM
Rigby, ID 83442. E-mail: [email protected]
Note. This manuscript was submitted on February 14, 2013; approved on
October 30, 2013; published online on November 4, 2013. Discussion period Fig. 1 shows a pulley structure, which, from the mechanics point of
open until July 11, 2014; separate discussions must be submitted for view, consists of cylindrical shell, plate, and beam components. The
individual papers. This paper is part of the Journal of Engineering theories of shell, plate, and beam are used to develop a precise finite
Mechanics, © ASCE, ISSN 0733-9399/04014057(12)/$25.00. element for each component.

© ASCE 04014057-1 J. Eng. Mech.

 Fig. 1. Profile of a pulley structure

Modeling Rim as Thin-Walled Cylindrical Shell

Fig. 2 shows the coordinate system for a cylindrical shell with radius
R and the stresses and stress couples acting on a differential element.
According to Ventsel and Krauthammer (2001), the equilibrium
equations can be written as
∂N1 1 ∂N21
þ þX ¼0 (1)
∂x R ∂u

∂N12 1 ∂N2 Q2
þ þ þY ¼0 (2)
∂x R ∂u R

∂Q1 1 ∂Q2 N2
þ 2 þZ ¼0 (3)
∂x R ∂u R

∂M1 1 ∂M21
Q1 ¼ þ (4)
∂x R ∂u

∂M12 1 ∂M2
Q2 ¼ þ (5)
∂x R ∂u

M21
N12 2 N21 2 ¼0 (6)
R

where N1 , N2 , N21 , and N12 5 membrane stresses; Q1 and Q2 Fig. 2. Stresses on rim cylindrical differential element
5 transverse stresses; and M1 , M2 , M12 , and M21 5 stress couples
(see Fig. 2 for the positive senses of these stresses and stress couples);
X, Y, and Z 5 components of external surface load in x-, q-, and z-
directions, respectively; the constitutive equations can be written as  
M12 ¼ M21 ¼ 2Dð1 2 mÞ 1 ∂ w 2 1 ∂v
2

 (12)
Et ∂u 1 ∂v w R ∂x∂u R ∂x
N1 ¼ þ m þ m (7)
1 2 m2 ∂x R ∂u R

 where u, v, and w 5 displacements in x-, u-, and z-directions, re-
N2 ¼ Et 1 ∂v þ w þ m ∂u (8) spectively; E and m 5 Young’s modulus and Poisson’s ratio of the
1 2 m2 R ∂u R ∂x pulley material; and D 5 Et 3 =12ð1 2 m2 Þ. The boundary conditions

 on the edge x 5 xp can be formulated in terms of the following four
N21 ¼ Et 1 ∂u þ ∂v (9) pairs of displacement-stress variables:
2ð1 þ mÞ R ∂u ∂x
  ðu, v, w, w, N1 , F1 , V1 , M1 Þ (13)
M1 ¼ 2D ∂ w 1 ∂2 w 2 m 1 ∂v
2
þ m (10)
∂x2 R2 ∂u2 R2 ∂u where
 
M2 ¼ 2D 12 ∂ w 1 ∂v þ m ∂2 w
2
2 (11) M12
R ∂u2 R2 ∂u ∂x2 F1 ¼ N21 þ 2 (14)
R

© ASCE 04014057-2 J. Eng. Mech.

 ∂M1 2 ∂M12 P
‘
V1 ¼ þ (15) wðx, uÞ ¼ w0 ðxÞ þ ½wn ðxÞcos nu þ w2n ðxÞsin nu (31)
∂x R ∂u n51

∂w
w¼2 (16) For simplicity, drop the (x) in all the preceding Fourier coefficients
∂x [e.g., N1,n stands for N1,n ðxÞ].
Substituting Eqs. (20)–(31) into Eqs. (7)–(19)
Eliminating Q1 , Q2 , M21 , and N12 , from Eqs. (1)–(6) and Eqs. (13)–
(15), the equilibrium equations can be reduced to the following: N_ 1,n þ n N21 þ Xn ¼ 0 (32)
R

∂N1 1 ∂N21
þ þX ¼0 (17) 2 n N2,n þ F_ 1,n 2 n2 M2,n þ Yn ¼ 0 (33)
∂x R ∂u R R

2
∂F1 1 ∂N2 1 ∂M2 2 n 2 M2,n þ V_ 1,n 2 1 N2,n þ Zn ¼ 0
þ þ 2 þY ¼0 (18) (34)
∂x R ∂u R ∂u R R

∂V1 1 ∂2 M2 N2 _ 1,n þ 2 n M12,n

V1,n ¼ M (35)
þ 2 2 þZ ¼0 (19) R
∂x R ∂u2 R

F1,n ¼ N21,n þ 2 M12,n (36)

Because of axisymmetry, the solution can be expressed by the fol- R
lowing Fourier series:
w_ n ¼ 2wn (37)
P
‘

uðx, uÞ ¼ u0 ðxÞ þ ½un ðxÞcos nu þ u2n ðxÞsin nu (20) Et mn m
N1,n ¼ _
u n þ v n þ wn (38)
n51 1 2 m2 R R

P
‘

vðx, uÞ ¼ v0 ðxÞ þ ½vn ðxÞsin nu þ v2n ðxÞcos nu (21) N2,n ¼ Et n v þ 1 w þ mu_ (39)
n n n
n51 1 2 m2 R R


P
‘ Et n
wðx, uÞ ¼ w0 ðxÞ þ ½wn ðxÞcos nu þ w2n ðxÞsin nu (22) N21,n ¼ 2 un þ v_ n (40)
n51
2ð1 þ mÞ R
 
‘ 
P  mn2 mn
N1 ðx, uÞ ¼ N1,0 ðxÞ þ N1,n ðxÞcos nu þ N1,2n ðxÞsin nu (23) M1,n ¼ D w_ n þ 2 wn þ 2 vn (41)
R R
n51
 
‘ 
P  n2 n
N2 ðx, uÞ ¼ N2,0 ðxÞ þ N2,n ðxÞcos nu þ N2,2n ðxÞsin nu (24) M2,n ¼ D wn þ 2 vn þ mw_ n (42)
R2 R
n51


‘ 
P  n 1
M12,n ¼ 2Dð1 2 mÞ 2 wn 2 v_ n (43)
N21 ðx, uÞ ¼ N21,0 ðxÞ þ N21,n ðxÞsin nu þ N21,2n ðxÞcos nu R R
n51
(25) where the updot symbol denotes the derivative with respect to
x (i.e., u_ n 5 dun =dx); Xn , Yn , and Zn 5 Fourier coefficients of X, Y,
‘ 
P  and Z, respectively.
M1 ðx, uÞ ¼ M1,0 ðxÞ þ M1,n ðxÞcos nu þ M1,2n ðxÞsin nu (26) Let
n51

‘   qn ¼ ðun , vn , wn , wn ÞT and
P
M2 ðx, uÞ ¼ M2,0 ðxÞ þ M2,n ðxÞcos nu þ M2,2n ðxÞsin nu (27)  T
n51 pn ¼ 2pRN1,n , 2pRF1,n , 2pRV1,n , 2pRM1,n (44)

‘ 
P  From the preceding equations, one can derive the following ordinary
M12 ðx, uÞ ¼ M12,0 ðxÞ þ M12,n ðxÞsin nu þ M12,2n ðxÞcos nu differential equation (ODE) for the nth-order Fourier coefficients of
n51
q and p of Eq. (44)
(28)

 


q_ n An Bn qn 0
‘ 
P  5 þ (45)
V1 ðx, uÞ ¼ V1,0 ðxÞ þ V1,n ðxÞcos nu þ V1,2n ðxÞsin nu (29) p_ n Cn Dn pn fn
n51
where
‘ 
P 
F1 ðx, uÞ ¼ F1,0 ðxÞ þ F1,n ðxÞsin nu þ F1,2n ðxÞcos nu (30)
n51 fn 5ð22pRXn ,22pRYn ,22pRZn , 0ÞT (46)

© ASCE 04014057-3 J. Eng. Mech.

 2 3 2 3
mn m 1 2 m2
0 2 2 0
6 R R 7 6 0 0 0 7
6  7 6 2pREt 7
6 7 6 7
6 EtnR 4 1 2 m nD 7
2 6 ð1 þ mÞR 7
6 0 0 7 6 7
6 2 EtR2 þ 4ð1 2 m2 ÞD7 6 0 0 0 7
An ¼ 6EtR þ 4ð1 2 m ÞD
2
7 Bn ¼ 6 pEtR2 þ 4pð1 2 m2 ÞD 7 (48)
6 7 6 7
6 21 7 6 7
6 0 0 0 7 6 0 0 0 0 7
6 7 6 7
4 mn mn2 5 4 1 5
0 2 2 2 0 0 0 0
R R2 2pRD
(47)

2 3
4pð1 2 mÞn2 EtD 4pð1 2 mÞn2 EtD
6 0 0 2 2 7
6 EtR3 þ 4ð1 2 m2 ÞRD EtR þ 4ð1 2 m2 ÞD 7
6   7
6 7
6 2pn2 Et 2p 1 2 m n D 2pnEt 2p 1 2 m n D 7
2 2 2 3
6 0 þ þ 0 7
6 R R3 R R3 7
Cn ¼ 6
6   7
7
6 2pnEt þ 2p 1 2 m n D 2pEt þ 2p 1 2 m n D 7
2 3 2 4
6 0 0 7
6 R R3 R R3 7
6  2 27
6 7
4 4pð1 2 mÞn2 EtD 4pn ð1 2 mÞD 16pð1 þ mÞ 1 2 m n D 5
2 2
2 2 0 0 2
EtR þ 4ð1 2 m2 ÞD R EtR3 þ 4ð1 2 m2 ÞRD
(49)

2 3 where
0 2 2 EtnR 2 0 0 " #21 " # ( )
6 EtR þ 4ð1 2 m ÞD 7
6 7 2Tn,qp 0 2Tn,qq I qn ð0Þ
6 mn mn 7 Kn ¼ ; Un ¼
6 7 ;
6 0 0 7 2Tn,pp 2I 2Tn,pq 0 qn ðxÞ
6 R R2 7
6 7 ( ) " #21 ( )
Dn ¼ 6 7 (50) 2pn ð0Þ 2Tn,qp 0
6 m mn2 7 Fn,q
6 0 0 7 Fn,int ¼ ; and Fn,ext ¼
6 R R2 7 pn ðxÞ 2Tn,pp 2I Fn,p
6  7
6 7
4 4 1 2 m2 nD 5
0 2 2 1 0
EtR þ 4ð1 2 m2 ÞD The following comments apply to these preceding equations:
1. The mathematical properties described by Eq. (51) guar-
It can be seen that An , Bn , Cn , and Dn have the following properties: antee that the ODEs of Eq. (45) are a Hamiltonian system
(Vinogradov and Kupershmidt 1981; Zhong 1995). The corre-
An ¼ 2DTn ; Bn ¼ BTn ; Cn ¼ CnT (51) sponding Hamiltonian function can be expressed as Hðpn , qn Þ
5 1=2pTn Bn pn 1 pTn An qn 2 1=2qTn Cn qn . With the use of Ham-
iltonian function, Eq. (45) also can be written as q_ n 5 ∂H=∂pn
The general solution to Eq. (45) can be written as and p_ n 5 2ð∂H=∂qn Þ
Note that the mathematical properties of Eq. (51) only can be


 ðx

qn ðxÞ qn ð0Þ 0 preserved when the displacement-stress pairs of variables,
¼ Tn ðxÞ þ Tn ðx 2 tÞ dt (52) called state variables, for boundary conditions [i.e., Eq. (13)]
pn ðxÞ pn ð0Þ fn ðtÞ
0 are selected correctly.
2. It is difficult to obtain the transfer matrix Tn ðxÞ of Eq. (52)
where Tn ðxÞ 5 transfer matrix. Let analytically. But Tn ðxÞ can be calculated precisely by using the
so-called precise numerical integration method suggested by
 ðx

 Zhong et al. (1996).
Tn,qq Tn,qp 0 Fn,q 3. Because Eq. (54) takes the form of a finite-element model, it
Tn ðxÞ ¼ and Tn ðx 2 tÞ dt ¼
Tn,pq Tn,pp fn ðtÞ Fn,p can be assembled with other finite elements discussed sub-
0 sequently to form a global FEM equation to represent the
(53) mechanical behavior of a pulley structure. Because there is no
interpolation approximation used in the development of
Eq. (52) can be reorganized in a form of finite-element model Eq. (54) [i.e., the formulation of Eq. (54) is mathematically
exact], the authors call the precise finite-element model de-
Kn Un ¼ Fn,int þ Fn,ext (54) veloped in this paper the PFEM.

© ASCE 04014057-4 J. Eng. Mech.

 positive senses of these stresses and stress couples); the constitutive
equations can be written as


Et ∂w 1 ∂v w
Nr ¼ þ m þ m (61)
1 2 m2 ∂r r ∂u r


Et 1 ∂v w ∂w
Nu ¼ þ þ m (62)
1 2 m2 r ∂u r ∂r


Et 1 ∂w ∂v v
Nru ¼ þ 2 (63)
2ð1 þ mÞ r ∂u ∂r r
 
Mr ¼ 2D ∂ u2 þ m 12 ∂ u2 þ m 1 ∂u
2 2
(64)
∂r r ∂u r ∂r
 
1 ∂2 u 1 ∂u ∂2 u
Mu ¼ 2D þ þ m (65)
r2 ∂u2 r ∂r ∂r 2

Fig. 3. Stresses on disk plate differential element  

Mru ¼ Dð1 2 mÞ 1 ∂ u 2 12 ∂u
2
(66)
r ∂r∂u r ∂u

4. It is noted that in Qiu’s (2006, 2009) boundary element models where u, v, and w 5 displacements in x-, u-, and r-directions, re-
for rolling resistance, a similar form of governing ODE to that spectively; and the boundary conditions on the edge r 5 rp can be
of Eq. (45) is derived, which can be proved to satisfy the formulated in terms of the following four pairs of displacement-
properties of Eq. (51) in the domain of complex numbers. stress variables:
5. In Qiu and Sethi (1993), the terms associated with the cir-
cumferential displacement v in Eqs. (10)–(12) are omitted for ðu, v, w, w, Vr , Nru , Nr , 2Mr Þ (67)
the purpose of convenience to obtain analytical solution of
transfer matrix (Timoshenko and Woinowsky-Krieger 1959). where
The consequence of the omission is that four modes of rigid
body motions are no longer preserved. w ¼ ∂u (68)
∂r
Modeling Disk as Thin-Walled Circular Plate
1 ∂Mru
V r ¼ Qr 2 (69)
Fig. 3 shows the coordinate system for a circular plate and the r ∂u
stresses and stress couples acting on a differential element. The
thickness, denoted by t, of the circular plate varies as the following
function of radius: Eliminating Qr and Qu from Eqs. (57)–(59) and Eq. (68)

t ¼ cr b (55) ∂Vr Vr 2 ∂Mru 1 ∂2 Mu

þ 2 2 þ 2 ¼0 (70)
∂r r r ∂u r ∂u2
According to Ventsel and Krauthammer (2001), the equilibrium
equations can be written as ∂Mr Mr 2 Mu 2 ∂Mru
Vr ¼ þ 2 (71)
∂r r r ∂u
∂Nr Nr 2 Nu 1 ∂Nru
þ þ ¼0 (56)
∂r r r ∂u Because of axisymmetry, the solution can be expressed by the fol-
lowing Fourier series:
∂Nru ∂N ∂N
þ 2 ru þ 1 u ¼ 0 (57)
∂r r r ∂u P
‘
uðr, uÞ ¼ u0 ðrÞ þ ½un ðrÞcos nu þ u2n ðrÞsin nu (72)
n51
∂Qr Qr 1 ∂Qu
þ 2 ¼0 (58)
∂r r r ∂u
P
‘
vðr, uÞ ¼ v0 ðrÞ þ ½vn ðrÞsin nu þ v2n ðrÞcos nu (73)
∂Mr Mr 2 Mu 1 ∂Mru n51
Qr ¼ þ 2 (59)
∂r r r ∂u
P
‘
∂Mru M ∂M wðr, uÞ ¼ w0 ðrÞ þ ½wn ðrÞcos nu þ w2n ðrÞsin nu (74)
Qu ¼ þ 2 ru 2 1 u (60) n51
∂r r r ∂u
‘ 
P 
where Nr , Nu , and Nru 5 membrane stresses; Qr and Qu 5 transverse Nr ðr, uÞ ¼ Nr,0 ðrÞ þ Nr,n ðrÞcos nu þ Nr,2n ðrÞsin nu (75)
stresses; and Mr , Mu , and Mru 5 stress couples (see Fig. 3 for the n51

© ASCE 04014057-5 J. Eng. Mech.

 ‘  

P Nu,n ¼ Et 1 w þ n v þ mw_
Nu ðr, uÞ ¼ Nu,0 ðrÞ þ Nu,n ðrÞcos nu þ Nu,2n ðrÞsin nu (76) 12m r
2 n
r
n n (89)
n51


‘ 
P  Nru,n ¼ Et v_ n 2 n wn 2 1 vn (90)
Nru ðr, uÞ ¼ Nru,0 ðrÞ þ Nru,n ðrÞsin nu þ Nru,2n ðrÞcos nu 2þm r r
n51
(77)  
m mn2
Mr,n ¼ 2D w_ n þ u_ n 2 2 un (91)
‘   r r
P
Mr ðr, uÞ ¼ Mr,0 ðrÞ þ Mr,n ðrÞcos nu þ Mr,2n ðrÞsin nu (78)
n51  
1_ n2
Mu,n ¼ 2D un 2 2 un þ mw_ n (92)
‘ 
r r
P 
Mu ðr, uÞ ¼ Mu,0 ðrÞ þ Mu,n ðrÞcos nu þ Mu,2n ðrÞsin nu (79)
n51

n n
Mru,n ¼ Dð1 2 mÞ 2 wn þ 2 un (93)
r r
‘ 
P 
Mru ðr, uÞ ¼ Mru,0 ðrÞ þ Mru,n ðrÞsin nu þ Mru,2n ðrÞcos nu Let
n51
(80) qn ¼ ðun , vn , wn , wn ÞT and
 T
‘ 
P  pn ¼ 2prVr,n , 2prNru,n , 2prNr,n ,22prMr,n (94)
Vr ðr, uÞ ¼ Vr,0 ðrÞ þ Vr,n ðrÞcos nu þ Vr,2n ðrÞsin nu (81)
n51
From the preceding equations, one can derive the following ODE
P
‘ for the nth-order Fourier coefficients of q and p of Eq. (94)
wðr, uÞ ¼ w0 ðrÞ þ ½wn ðrÞcos nu þ w2n ðrÞsin nu (82)
n51
 

q_ n An Bn qn
5 (95)
For simplicity, drop the (r) in all the preceding Fourier coefficients. p_ n Cn Dn pn
Substituting Eqs. (72)–(82) into Eqs. (56) and (57), Eqs. (61)–(68),
and Eqs. (70) and (71), and denoting the derivative with respect to r where
by the up dot symbol (i.e., u_ n 5 dun =drÞ
2 3
rN_ r,n þ Nr,n 2 Nu,n þ nNru,n ¼ 0 (83) 0 0 0 1
6 0 0 7
6 1=r n=r 7
An 56 7 (96)
r N_ ru,n þ 2Nru,n 2 nNu,n ¼ 0 (84) 4 0 2mn=r 2m=r 0 5
mn2 =r2 0 0 2m=r
2n n2
rV_ r,n þ Vr,n 2 Mru,n 2 Mu,n ¼ 0 (85) 2 3
r r 0 0 0 0
6 7
6 1þm 7
rVr,n _ r,n þ Mr,n 2 Mu,n 2 2nMru,n
¼ rM (86) 60 0 0 7
6 prEt 7
6 7
Bn 56
6 12m2
7
7 (97)
u_ n ¼ wn (87) 60 0 0 7
6 2prEt 7
6 7

 4 5
Et m mn 1
Nr,n ¼ w_ n þ wn þ vn (88) 0 0 0
12m 2 r r 2prD

2 3
2pð2 2 2m þ n2 2 m2 n2 Þn2 D 2pðm2 þ 2m 2 3Þn2 D
6 0 0 7
6 r3 r2 7
6 7
6 2pn2 Et 7
6 2pnEt 7
6 0 0 7
6 r r 7
Cn 56 7 (98)
6 2pnEt 2pEt 7
6 0 0 7
6 7
6 r r 7
6 7
4 2pðm þ 2m 2 3Þn D
2 2
2pð1 2 m2 þ 2n2 2 2n2 mÞD 5
0 0
r2 r

© ASCE 04014057-6 J. Eng. Mech.

 2 3
0 0 0 2mn2 =r2 qn ¼ Sq ðrÞqpn and pn ¼ Sp ðrÞppn (103)
6 0 21=r 7
6 nm=r 0 7
Dn 56 7 (99) where
4 0 2n=r m=r 0 5
21 0 0 m=r qpn ¼ ðun , vn , wn , rwn ÞT
 T
It can be seen that An , Bn , Cn , and Dn satisfy the properties of a linear ppn ¼ 2pr 323b Vr,n , 2pr 12b Nru,n , 2pr12b Nr,n , 22pr223b Mr,n
Hamiltonian system described by Eq. (51). The general solution to
Eq. (95) can be written as Sq ðrÞ ¼ diag½ 1 1 1 1=r 


  
qn ðrÞ qn ðr0 Þ Sp ðrÞ ¼ diag r 22þ3b rb rb r 21þ3b
5 Tn ðr, r0 Þ (100)
pn ðrÞ pn ðr0 Þ
where diag½. . . 5 diagonal matrix; and b 5 parameter defined in
Eq. (55). Substituting Eq. (103) into Eq. (95), one obtains a new set
where Tn ðr, r0 Þ 5 transfer matrix. Let of ODEs with constant coefficients

p " #

Tn,qq Tn,qp d qn ~n B
A ~n qpn
Tn ðr, r0 Þ5 (101) ¼ (104)
Tn,pq Tn,pp d ln r ppn ~n D
C ~n ppn

Following the same procedure of development of Eq. (54), one can where
rewrite Eq. (100) in a form of finite-element model
2 3
0 0 0 1
Kn Un ¼ Fn (102) 6 0 0 7
~n ¼ 6
A 6
1 n 7
7 (105)
4 0 2mn 2m 0 5
where mn2 0 0 12m
" #21 " # ( ) 2 3
2Tn,qp 0 2Tn,qq I qn ðr0 Þ 0 0 0 0
Kn ¼ ; Un ¼ ; 6 7
2Tn,pp 2I 2Tn,pq qn ðrÞ 6 7
60 1 þ m
0
0 0 7
( ) 6 pEc 7
2pn ðr0 Þ 6 7
and Fn ¼ ~n ¼ 66 7
B 12m 2 7 (106)
pn ðrÞ 60 0 0 7
6 2pEc 7
6 7
6  7
4 6 12m 5 2
To obtain accurate numerical solution of Tn ðr, r0 Þ, make the 0 0 0
following transformation pEc3

2  3
2p 2 2 2m þ n2 2 m2 n2 n2 Ec3 2p m2 þ 2m 2 3 n2 Ec3
6 0 0 7
6 12ð1 2 m2 Þ 7
12ð1 2 m2 Þ
6 7
6 7
6 0 2pn2 Ec 2pnEc 0 7
~n ¼ 6
C 7 (107)
6 7
6 0 2pnEc 2pEc 0 7
6  2  7
6 7
4 2p m þ 2m 2 3 n2 Ec3 2p 1 2 m þ 2n 2 2n m Ec 5
2 2 2 3
0 0
12ð1 2 m2 Þ 12ð1 2 m2 Þ

2 3   21
2 2 3b 0 0 2mn2 Sq ðrÞ 0 Sq ðr0 Þ 0
6 0 7 Tn ðr, r0 Þ ¼ Tnp ðln r 2 ln r0 Þ
21 2b 0 Sp ðrÞ 0 Sp ðr0 Þ
~n ¼ 6
D 6
nm 0 7
7 (108)
4 0 2n m2b 0 5 (109)
21 0 0 1 þ m 2 3b
Modeling Shaft as Beam and Bar
The pulley shaft is treated as a beam for bending and a bar for tension
where c is defined in Eq. (55). For Eq. (104), one can obtain accurate and torsion (Fig. 4). As proved by Qiu and Sethi (1993), the fol-
numerical solution Tnp ðln r 2 ln r0 Þ and then use Eq. (103) to find the lowing mathematically exact beam element can be obtained by use
solution of Timoshenko’s beam theory:

© ASCE 04014057-7 J. Eng. Mech.

 Fig. 4. Beam element for shaft

Kbm Ubm ¼ Fbm (110)

Fig. 5. Finite-element divisions for a pulley
where

Ubm ¼ ðws0 , g0 , wsx , gx ÞT

Fbm 5 ðV0 , M0 , Vx , Mx ÞT same as that in conventional FEM. Special attention, however, must
2 3 be paid to the connection between the shaft element and the disk
12EIs 6EIs 12EIs 6EIs
2 element. In beam and bar theories, a major assumption is that plane
6 L3 ð1 þ aÞ L2 ð1 þ aÞ L3 ð1 þ aÞ L2 ð1 þ aÞ 7
6 7 sections remain plane. According to this assumption, the shaft beam
6 7 element only can contribute to the Fourier series term of n 5 1
6 ð4 þ aÞEIs 12EIs 7
6 6EIs
2
6EIs 7 and 21; and the shaft bar element only can contribute to the Fourier
6 L2 ð1 þ aÞ Lð1 þ aÞ L2 ð1 þ aÞ L3 ð1 þ aÞ 7
6 7 series term of n 5 0. Therefore, at the point of connection between
Kbm 5 6
6
7
7
6 12EIs 6EIs 12EIs 6EIs 7 the shaft and disk, the following conditions are imposed to connect
62 3 2 2 2 7
6 L ð1 þ aÞ L ð1 þ aÞ L3 ð1 þ aÞ L2 ð1 þ aÞ7 the shaft with the disk:
6 7
6 7
4 6EIs ð22aÞEIs 6EIs ð4 þ aÞEIs 5 wn ¼ 2nvn n ¼ 21, 1 (114)
2 2
L2 ð1 þ aÞ Lð1 þ aÞ L ð1 þ aÞ Lð1 þ aÞ
(111) un ¼ Rshaft wn n ¼ 21, 1 (115)

where ws0 5 ws ðx0 Þ, wsx 5 ws ðxÞ, g0 5 gðx0 Þ, gx 5 gðxÞ; ws ðxÞ and ws ¼ wn n ¼ 21, 1 (116)
gðxÞ 5 transverse displacement and rotation angle of the beam axis,
respectively; VðxÞ and MðxÞ 5 transverse force and moment acting
at the cross section of the beam; Is 5 area moment of inertia; g ¼ wn n ¼ 21, 1 (117)
L 5 x 2 x0 ; and a 5 12EIs =ðL2 kaGÞ where k is a shape factor equal
to 0.75 for circular cross section, a is the area of the cross section, us ¼ un n¼0 (118)
and G is the shear modulus.
For tension and torsion loading, the following mathematically b ¼ vn =Rshaft n¼0 (119)
exact element applies:
Kbar Ubar ¼ Fbar (112) un ¼ vn ¼ wn ¼ 0, wn ¼ 0 jnj . 1 (120)
where
where Rshaft 5 radius of the shaft at the connection point.
T The belt tension applies a pressure and shear stress on the surface
Ubar ¼ ðus0 , b0 , usx , bx Þ
of the pulley rim. Fig. 6. shows a typical distribution of such
Fbar 5 ðN0 , Mt0 , Nx , Mtx ÞT a loading. The pressure and shear stress, as a periodic function of u,
2 3 can be expanded into Fourier series. The coefficients of the pressure
EA 0 2EA 0
6 L L 7 and shear stress are the terms of Zn and Yn in Eq. (46). Due to the
6 7 nature of loading, Xn 5 0 in Eq. (46). The integration term of Eq. (52)
6 7
6 GJs GJs 7
6 0 0 2 7 (113) can be evaluated by using any of the numerical integration methods,
6 L L 7
6
Kbar 5 6 7 of which 5 points Gauss quadrature is recommended.
7
6 2EA EA
0 7
6 0 7
6 L L 7
6 7 Examples
4 GJs GJs 5
0 2 0
L L Two examples are given to demonstrate the validity of PFEM for
pulley stress analysis. One is a bend pulley and the other is a drive
where us0 5 us ðx0 Þ, usx 5 us ðxÞ, b0 5 bðx0 Þ, bx 5 bðxÞ; us ðxÞ and
pulley. The major parameters for both pulleys are listed in Table 1. In
bðxÞ 5 axial displacement and rotational angle about axis of the shaft,
the two examples, the pulleys (Fig. 1) are analyzed by using both
respectively; NðxÞ and Mt ðxÞ 5 tension and torsional moment acting
PFEM and ANSYS 10 FEM. Owing to symmetry, only half of the
at the cross section, respectively; and Js 5 polar moment of inertia.
pulley is modeled.

Assembly of Global Finite Elements

FEM Model
Fig. 5 shows a division of finite elements for a pulley. The assembly In the ANSYS FEM model for pulley, two types of axisymmetric-
method of the precise finite elements discussed previously is the harmonic structural solid elements [Plan 25 (four nodes per element)

© ASCE 04014057-8 J. Eng. Mech.

and Plan 83 (eight nodes per element)] and two types of meshes 1. The deviations of the stress solutions of Plan 25 model with
[coarse mesh (1,088 elements) and fine mesh (4,690 elements)] are coarse and fine meshes with respect to the solution of Plan 83
tested for convergence. There are 71 (236 , n , 36) Fourier terms model with fine mesh are around 17 and 12%, respectively, in
used in the FEM model. Fig. 7 shows the distribution of stress terms of RMS error defined as
component sx . From Fig. 7, it is observed that there is a tendency for sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
solutions to converge to the solution of the Plan 83 model with fine PN  Test 2
mesh as the mesh gets finer and/or the order of element’s polynomial i51 si 2 sP83Fine
i
N
(interpolation) function gets higher. Other stress components not RMS_ERðsÞ ¼ (121)
shown in Fig. 7 have similar trends to that of sx . Further calculations ksk‘
indicate the following:
2. The deviation of the stress solutions of Plan 83 model with
coarse mesh with respect to that of Plan 83 model with fine
meshes is around 1% in terms of RMS error.
Using the preceding deviation numbers and FEM h-p conver-
gence criterion, the authors concluded that the ANSYS Plan 83 el-
ement model with fine mesh yields a sufficient accurate solution for
pulley stresses. Therefore, ANSYS Plan 83 element model with fine
meshes (3,662 and 4,690 elements for the bend and drive pulleys,
respectively) was chosen to calculate stresses for the bend and drive
pulleys specified in Table 1, respectively, for the purpose of vali-
dation and verification of PFEM.

Validation and Verification of PFEM

In a PFEM model, only 20 elements are employed; there are 71
(236 , n , 36) Fourier terms used in a PFEM model.
Figs. 8–15 demonstrate the results of stress distributions at
locations A, B, C, and D obtained by the PFEM and ANSYS. To
quantitatively estimate the numerical deviation between the results

Fig. 6. Loading on pulley surface

Table 1. Model Parameters

Case 1: Case 2:
Parameters bend pulley drive pulley
Material properties
E (GPa) 206.9 206.9
m 0.3 0.3
Rim geometry
Length (mm) 1,550 2,200
Diameter (mm) 610 1,362 Fig. 7. Observation of convergence of stress sx calculated by ANSYS
Thickness (mm) 20 38 for Case 2, at location B inside of rim
Belt width (mm) 1,300 2,000
Disk geometry
Locking device width (mm) 88 141
Hub outer diameter (mm) 380 800
Hub inner diameter (mm) 250 515
Hub width (mm) 117 155
Disk parameter, c 381 4,189
Disk parameter, b 20.433 20.760
Shaft geometry
Diameter (mm) 190 420
Length (mm) 2,240 3,360
Bearing-bearing distance (mm) 1,950 2,900
Disk-disk distance (mm) 1,280 2,000
Loading parameters
u1 (degrees) 210 0
u2 (degrees) 270 180
T1 (kN) 197 1,018 Fig. 8. Case 1, at location A inside of rim, REðsx Þ 5 9:3%, REðsu Þ
T2 (kN) 196 633 5 8:2%

© ASCE 04014057-9 J. Eng. Mech.

of PFEM and ANSYS, the following RMS error function, denoted as Evident in Figs. 8–15 and Tables 2 and 3, at locations A and D the
REðsÞ, was used first to measure the relative deviation: up bond of RE values are within 9–10% and the up bond of errors of
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi s1max and s2min are within 10%; while at locations B and C the up
PN  PFEM 2
i51 si 2 sANSYSi bond of RE values are within 12–17% and the up bond of errors of
N s1max and s2min are within 16–19%. The larger discrepancies of
REðsÞ 5 (122) stresses of PFEM from those of ANSYS at locations B and C may be
ksk‘
owing to the fact that locations B and C are much closer to the joint
Figs. 8–15 contain the REðsÞ numbers for each stress component. region between the rim and the disk, where the 3D stress state cannot
Secondly, Tables 2 and 3 give a direct comparison of principal be accurately modeled by the 2D shell and plate theories. However,
stresses calculated by PFEM to those calculated by ANSYS at the inaccuracies of stress distributions in locations B and C are within
locations where extreme values (denoted s1max and s2min ) occurred. 16–19% and, therefore, are within the engineering allowable margin.
The PFEM computation for each pulley only takes less than
a second to finish on a laptop computer with a 1.8-GHz AMD
processor. ANSYS computations take approximately 14 and 34 min,
respectively, for the bend and drive pulleys on a desktop computer
with a 3.07-GHz Inter Core i7–950 processor. Also, it is worth
noting that one of the strengths of PFEM is the much easier set-up of
the model and analysis by using only the information from Table 1.
On the other hand, the full FEM model of the pulley, which can
handle precisely the pulley geometry and yield accurate solution at
locations of stress concentrations, requires a lot of manual prepa-
ration work, such as mesh adjustment and test of solution conver-
gence. Thus, the authors suggest a way to combine the strengths of
both PFEM and FEM in the pulley design process: use of PFEM for
preliminary optimization (i.e., selection of good candidate designs
Fig. 9. Case 1, at location B inside of rim, REðsx Þ 5 11:9%, REðsu Þ among hundreds of candidates) and use of FEM for the final op-
5 5:8%, REðt xu Þ 5 3:0% timization (i.e., selection of the best design among a few good
candidates).

Fig. 10. Case 1, at location C inside of plate, REðsr Þ 5 8:1%, REðsu Þ Fig. 12. Case 2, at location A inside of rim, REðsx Þ 5 3:9%, REðsu Þ
5 5:7%, REðt ru Þ 5 6:8% 5 5:3%

Fig. 11. Case 1, at location D inside of plate, REðsr Þ 5 8:9%, REðsu Þ Fig. 13. Case 2, at location B inside of rim, REðsx Þ 5 17:1%, REðsu Þ
5 4:1%, REðt ru Þ 5 4:0% 5 8:2%, REðtxu Þ 5 4:8%

© ASCE 04014057-10 J. Eng. Mech.

 Table 3. Case 2 Extreme Stresses (MPa) and Bearing Reaction Force (kN)
by PFEM
Stress u Error
Location type (degrees) ANSYS PFEM (%)
Cross section A inner s1max 90 15.5 14.6 5.2
surface s2min 20 232.0 230.9 3.4
Cross section B inner s1max 355 41.0 42.3 3.3
surface s2min 310 215.4 214.7 5.1
Cross section C inner s1max 150 42.0 39.4 6.0
surface s2min 315 2100.2 283.8 16.4
Cross section D inner s1max 275 64.5 63.8 1.1
surface s2min 335 250.9 245.8 9.9
Fig. 14. Case 2, at location C inside of plate, REðsr Þ 5 9:6%, REðsu Þ Bearing center E Reaction F N/A 825.4 823.5 0.2
5 8:4%, REðtru Þ 5 7:9%

Table 4. Case 1 Extreme Stresses (MPa) and Bearing Reaction Force (kN)
by MTMM
Stress u Error
Location type (degrees) ANSYS MTMM (%)
Cross section A inner s1max 310 21.5 23.5 29.5
surface s2min 250 228.7 226.6 7.4
Cross section B inner s1max 345 32.0 34.2 6.6
surface s2min 0 211.0 28.5 22.5
Cross section C inner s1max 355 40.1 41.6 3.6
surface s2min 310 243.0 234.3 20.1
Cross section D inner s1max 340 51.4 52.9 2.9
surface s2min 130 246.6 242.1 9.8
Bearing center E Reaction F N/A 126.3 120.2 4.8

Fig. 15. Case 2, at location D inside of plate, REðsr Þ 5 8:7%, REðsu Þ

5 7:1%, REðt ru Þ 5 5:6%
Conclusions

A rigorous and efficient precise finite-element model for conveyor

Table 2. Case 1 Extreme Stresses (MPa) and Bearing Reaction Force (kN) pulley has been developed. Two examples of pulleys were chosen
by PFEM to validate PFEM models by comparison of the numerical solutions
to ANSYS FEM solution. The numerical examples demonstrated
Stress u Error
the capabilities and the degree of accuracies of the model. The
Location type (degrees) ANSYS PFEM (%)
methodology may serve as a useful tool for designers of industrial
Cross section A inner s1max 310 21.5 23.5 9.6 pulleys.
surface s2min 250 228.7 226.4 8.0
Cross section B inner s1max 345 32.0 33.9 5.8
References
surface s2min 0 211.0 29.9 9.8
Cross section C inner s1max 355 40.1 41.8 4.2
ANSYS 10 [Computer software]. Canonsburg, PA, ANSYS.
surface s2min 310 243.0 234.7 19.2
Lange, H. (1963). “Investigations on stress in belt conveyor pulleys.” Ph.D.
Cross section D inner s1max 340 51.4 53.4 4.0
thesis, Technical Univ. Hannover, Hannover, Germany.
surface s2min 130 246.6 242.9 7.9 Martins, J. A., and Kövesdy, I. (2012). “Overview in the application of FEM
Bearing center E Reaction F N/A 126.3 125.8 0.4 in mining and the study of case: Stress analysis in pulleys of stacker-
reclaimers: FEM vs. analytical.” Chapter 13, Finite element analysis:
Applications in mechanical engineering, F. Ebrahimi, ed., InTech,
Rijeka, Croatia, 277–296.
Patel, R. R., Joshi, S. P., and Agrawal, P. M. (2011). “Studies on some
Comparison of PFEM to MTMM
aspects of conveyor drive pulley design.” National Conf. on Recent
Table 4 shows the s1max and s2min results of the bend pulley obtained Trends in Engineering & Technology, BVM Engineering College,
by using Qiu and Sethi’s (1993) MTMM. Comparing the results in Anand, Gujarat, India.
Table 4 with those in Table 2, the authors conclude that PFEM Pathan, N. W., Bhope, D. V., and Khamankar, S. D. (2011). “Investigation
outperforms MTMM marginally. However, it is noted that at the of stresses in flat belt pulley by FEM and photoelasticity.” Int. J. Eng.
Sci. Technol., 3(10), 7444–7451.
cross section E (bearing location), the resultant reaction force cal-
Qiu, X. (2006). “Full two-dimensional model for rolling resistance: Hard
culated by MTMM is 5% lower than that of ANSYS while the re- cylinder on viscoelastic foundation of finite thickness.” J. Eng. Mech.,
sultant reaction force calculated by PFEM yields almost the same 10.1061/(ASCE)0733-9399(2006)132:11(1241), 1241–1251.
value as that of ANSYS. This is owing to the fact that PFEM preserve Qiu, X. (2009). “Full two-dimensional model for rolling resistance. II:
all modes of rigid body motions while MTMM only preserves two Viscoelastic cylinders on rigid ground.” J. Eng. Mech., 10.1061/(ASCE)
of six modes of rigid body motions. 0733-9399(2009)135:1(20), 20–30.

© ASCE 04014057-11 J. Eng. Mech.

Qiu, X., and Sethi, V. (1993). “A new pulley stress analysis method based on Ventsel, E., and Krauthammer, T. (2001). Thin plates and shells theory,
modified transfer matrix method.” J. Bulk Solids Handling, 13(4), 713–724. analysis, and applications, Marcel Dekker, New York.
Ravikumar, M., and Chattopadhyay, A. (1999). “Integral analysis of conveyor Vinogradov, A. M., and Kupershmidt, B. A. (1981). “The structures of
pulley using finite element method.” Comp. Struct., 71(3), 303–332. Hamiltonian mechanics.” Integrable systems, Cambridge University
Sethi, V., and Nordell, L. K. (1993). “Modern pulley design techniques Press, London, 173–240.
and failure analysis methods.” Proc., SME Annual Meeting & Ex- Zhong, W. (1995). A new systematic methodology for theory of elasticity,
hibition, Society for Mining, Metallurgy, and Exploration, Engle- Dalian University of Technology Press, Dalian, China.
wood, CO. Zhong, W., Zhu, J., and Zhong, X. (1996). “On a new time integration
Timoshenko, S., and Woinowsky-Krieger, S. (1959). Theory of plates and method for solving time dependent partial differential equations.”
shells, 2nd Ed., McGraw Hill, New York. Comput. Methods Appl. Mech. Eng., 130(1–2), 163–178.

© ASCE 04014057-12 J. Eng. Mech.

Copyright of Journal of Engineering Mechanics is the property of American Society of Civil
Engineers and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.