Journal of Advanced Mechanical Design,

Systems, and

Manufacturing

Vol. 3, No. 2, 2009
159

Static Stress Analysis of Link Plate of Roller
Chain using Finite Element Method and
Some Design Proposals for
Weight Saving*
Shoji NOGUCHI**, Kohta NAGASAKI***, Satoshi NAKAYAMA**,
Tohru KANADA****, Toshiaki NISHINO***** and Takashi OHTANI*****
**Department of Mechanical Engineering, Tokyo University of Science
2641 Yamasaki, Noda-city, Chiba 278-8510, Japan
E-mail:nog@rs.noda.tus.ac.jp
*** NSK Ltd., 1-5-50 Kugenuma-shimei, Fujisawa-city, Kanagawa 251-8501, Japan
E-mail: nagasaki-ko@nsk.com
**** Department of Mechanical Engineering, Kanto Gakuin University
1-50-1 Mutsuura-higashi, Kanazawa-ku, Yokohama-city, Kanagawa 236-8501, Japan
E-mail:kanada@kanto-gakuin.ac.jp
*****Enuma Chain Mfg. Co., Ltd.
300 Kami-kawasakicho, Kaga-city, Ishikawa 922-8678, Japan
E-mail:nisino@enuma.co.jp, tohtani@enuma.co.jp

Abstract
Roller chains have a long history as mechanical elements for transmission.
Although they have clear advantages over belts in terms of performance and
efficiency, their larger weight has always been a disadvantage. In this paper, the
authors propose some methods of weight saving for roller chains. These methods
are based on finite element method (FEM) analysis of the stress and deformation in
a link plate of a roller chain. The authors also suggest some approaches for
reducing stresses and weight saving in the link plate of the roller chain. Stresses are
3% higher in the proposed design, but the weight is reduced by 10%. Tensile tests
are performed on link plates made of resin, and the effectiveness of the proposed
model is confirmed.
Key words: Roller Chain, Link Plate, FEM Analysis, Stress Analysis, Weight
Saving, Design Proposal

1. Introduction
A roller chain is known and used as a mechanical element for drive train or parts
conveyance systems. Basically, its mechanism is a type of engaging between a chain and a
sprocket wheel. Compared with that in belts, the capacity of transmission in roller chain is
large and the efficiency is high owing to the absence of slip. Thus, the mechanism of roller
chains is applied to the rear-wheel drive of a bike, the hoist of a fork lift and the valve
switching mechanism of an internal-combustion engine.
However, when the transmission speed increases, the vibration of the mechanism
increases as is the case of gear engagement
(1)
. In a case of long-distance transmission using
a metal chain, a large driving force is required owing to the enlarged mass of the chain. To
solve such a problem, weight saving for the chain is desired. Regardless of the
above-mentioned background, research on the chain mechanism is rare. Moreover, design
*Received 25 Mar., 2009 (No. 09-0122)
[DOI: 10.1299/jamdsm.3.159]


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
160
research by stress analysis is practically not feasible.
Therefore, in this research, as the first step of weight saving for the chain, the authors
attempted to provide a design solution. That is, weight saving with the suppression of the
increase in stress is realized by analyzing the static stress of a link plate in a roller chain by
FEM (Finite Element Method).
First, the stress of a standard product in accordance with JIS was calculated by FEM for
loading static force. Second, the calculated stress was used as a benchmark. Third,
considering the weight saving (decrease of volume) with the suppression of the increase in
stress, the form of the link plate was explored by changing various design parameters. Then,
a prototype of link plate made of ABS resin was manufactured under the conditions used for
obtaining favorable results in simulated design proposal. Furthermore, the adequateness of
the analysis was verified by a tension test.

2. Three-dimensional CAD model and conditions of FEM analysis
Figure 1 shows the fundamental structure of a roller chain. The dimension of each part
is defined in JIS B 1801
(2)
or JCAS 1
(3)
. First, in this research, the stress on a commercially
available standard chain was analyzed. Then, when the form of the link plate was changed,
the analyzed stress was compared with that of the standard chain.
Fig.1 Schematic view of roller chain

The shape of a link plate is similar to a gourd with a constricted part. Principal
dimensions such as the pitch distance between pins, are shown in JIS. By considering the
symmetric properties, mesh segmentation was performed on a quarter model for FEM
analysis. The meshed model is shown in Fig.2. This model has 20000 meshed elements. To
simplify the analysis of link plate stress, the roller was not regarded because the roller did
not support the generated force when the chain was pulled independently.

Fig.2 Example of meshed model for FEM

The stress and deformation analyses were performed by applying a static tensile load in
the chain direction (X-direction). The load was not applied to the link plate directly but to
Inner link plate
Bush
Roller
Outer link plate


Pin


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
161
neighboring pins (or bushes). The deformations in the nodal points on the symmetry plane
were restrained in a plane perpendicular to the symmetry plane.
The analysis software used is MSC.Marc, which is a general-purpose FEM program.
Specifications of the personal computer used are as follows. The CPU (Central Processing
Unit) is Intel Xeon (3.06GHz) and the capacity of the main memory is 2GB. Although the
elapsed computation time depended on the mesh size, it was around 3 hours in this research.

3. Analytical results for standard chain
To determine the standard stress for weight saving design, the roller chain (size 520)
for a commercially available bike was analyzed as shown in Fig.3. Table 1 shows the
dimensions of the chain parts. Mesh segmentation was performed on this chain as shown in
Fig.2. Then, FEM analysis was applied for a static tensile load of 500N.













Fig.3 Photograph of roller chain for motorcycle #520

Table 1 Dimensions of chain parts
Material of plates SCM440E=206GPa, =0.3
Pin interval 15.875 mm
Pin Outer diameter 5.38 mm
Bush Inner diameter 5.5 mm
Height 14.9 mm
Thickness 2.0 mm Inner plate
Width 6.5 mm
Height 12.9 mm
Thickness 2.0mm Outer plate
Width 13.5 mm

The analytical results for tensile stress are shown in Fig.4 for the inner link plate and in
Fig.5 for the outer link plate. Both figures are displayed by observing from the outside of
the chain. On the color level scale, the upper half zone indicates tensile stress and the lower
half zone indicates compressive stress.
These results show that stress concentrates around a pin or bush. Moreover, the stress
at the constricted part of the link plate is large. Although studies on the analytical results for
stress occurring in the chain are few and a quantitative comparison with other research
studies is difficult owing to the different analytical conditions used, it can be verified that
the state of stress is similar to that in reference 4. In a recorded case of a damaged link plate,
cracks arise from the bottom point of contact hole of a pin or bush
(4)
. Therefore, the authors
believe that their analytical results are qualitatively accurate.
13.5mm
6.5mm
15.875mm
2mm


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
162

Fig.4 Results of stress analysis of inner plate


Fig.5 Results of stress analysis of outer plate




Fig.6 Deformation of chain caused by tension


Fig.7 Representative positions for stress analysis

B
B
A A
C
C
X
Z


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
163






































Fig.8 Stress distribution of inner link

Furthermore, the state of the stress distribution in the thickness direction varies
between the inner and outer link plates. In the inner link plate, the stress distribution is
slightly deviated by the outer side of the chain. In the contrast, in the outer link plate, the
stress distribution is slightly deviated by the inner side of the chain. It is considered that this
distinction arises from the deformations of the pin or bush, as shown in Fig.6. Owing to the
deformation of the pin or bush, the contact surface pressure concentrates on the corner. Thus,
it can be considered that the deformation of the link plate and the stress concentration arise
on the same side.
The stress distribution at a typical position (Fig.7) of link plates in Figs.4 and 5 is
shown in Figs.8 and 9. X-directional stresses on the A-A and B-B sections decrease toward
the outer side of the link plate from the bottom of the hole. At the point A of the outer link
-200
-150
-100
-50
0
50
100
150
0 1 2 3 4
Distance , mm
S
t
r
e
s
s

,

M
P
a
X-direction
Y-direction
Z-direction
-200
-150
-100
-50
0
50
100
150
0 1 2 3 4
Distance , mm
S
t
r
e
s
s

,

M
P
a
X-direction
Y-direction
Z-direction
-200
-150
-100
-50
0
50
100
150
0 1 2 3 4
Distance , mm
S
t
r
e
s
s

,

M
P
a
X-direction
Y-direction
Z-direction
A A
B B
C C


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
164







































Fig.9 Stress distribution of outer link

plate, the compressive stress concentrates inside the link plate. Therefore, the compressive
stress decreases toward the outer side of the outer link plate.
On the A-A section of the inner link plate, Y-directional stress is relatively large. The
area around the point A is tensile, but that around the point A is compressive. Thus, the state
of stress along the A-A section is complex and consists of multiple stress axes. The stress
along the C-C section is smaller than that of the bottom part of the hole and has a uniaxial
state of stress. The stress toward the center C from the outer constricted part decreases and
that of the point C is almost zero.

-200
-150
-100
-50
0
50
100
150
0 1 2 3 4
Distance , mm
S
t
r
e
s
s

,

M
P
a
X-direction
Y-direction
Z-direction
-200
-150
-100
-50
0
50
100
150
0 1 2 3 4
Distance , mm
S
t
r
e
s
s

,

M
P
a
X-direction
Y-direction
Z-direction
-200
-150
-100
-50
0
50
100
150
0 1 2 3 4
Distance , mm
S
t
r
e
s
s

,
M
P
a
X-direction
Y-direction
Z-direction
A A
B B
C C


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
165
4. Analytical results for weight saving model
4.1 Link plate with centrally located hole
A roller chain with link plates having centrally located holes is a commercial reality. It
has some effects, such as weight saving, heat cooling and mud ejection
(5)
. It is confirmed
that the stress in the central region (around the point C in the C-C section) of a link plate is
small and a centrally located hole in the link plate does not markedly affect the stress
change. Thus, a model with a centrally located hole in the inner link plate was established
and analyzed for the stress as shown in Fig.10.
The diameters of the centrally located hole are 3, 4, 5 and 6mm and other respects are
the same as the above-mentioned specifications in the previous chapter. The overall state of
stress is similar to that in Fig.4. Table 2 shows the changes in stress and weight saving ratio
compared with the weight without the centrally located hole. This result shows that the
stresses at the points B and A remain constant. At the point C outside the constricted part,
the tensile stress increases proportionally to the diameter of the centrally located hole.
However, this stress is 57-66% smaller than that of the point B which is the maximum.
Therefore, it is possible that a 10% maximum weight saving is realized without a change in
maximum stress.











Fig.10 Analysis models with center holes

Table 2 Stresses at representative positions and weight ratios
Diameter of
center hole,
mm
Maximum tensile
stress (point B), MPa
Maximum compressive
stress (point A), MPa
Tensile stress
(point C'),
MPa
Weight
ratio,
%
0 (Normal) 129.79 -181.10 73.88 100
3 132.32 -178.44 77.14 97.6
4 131.45 -182.97 81.03 95.8
5 132.84 -178.43 85.88 93.4
6 133.75 -178.56 88.54 90.5

4.2 Changing thickness of link plate
As shown in Fig.6, the direction of thickness is Z. If the thickness were changed, the
other dimensions would remain constant. Therefore, effective weight saving would be
realized. Thus, the stress was analyzed with a change in the thickness of the inner or outer
link plate for the standard roller chain indicated in chapter 3. Although the thickness of the
link plate of the standard chain was 2.00mm, four thicknesses, namely 1.75mm, 1.50mm,
1.25mm and 1.00mm, were set.
The model for the thickness of 1.00mm to be analyzed is shown in Fig.11. The stress
distribution in the link plate is similar to the state in Figs.4 and 5, and the position of stress
concentration is the same. The stresses at the typical positions of the inner and outer link
(a) Center hole of 3mm diameter
(b) Center hole of 6mm diameter


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
166










Fig.11 Analysis model for plate thickness of 1.00mm

Table 3 Stresses at representative positions and weight ratio in inner plate (X-direction)
Thickness of
plate, mm
Maximum
tensile stress
(point B), MPa
Maximum compressive
stress (point A), MPa
Tensile stress
(point C'),
MPa
Weight
ratio,
%
2.00 (Normal) 129.79 -181.10 73.88 100
1.75 148.16 -182.72 83.21 87.5
1.50 167.13 -186.87 94.90 75.0
1.25 171.98 -188.41 109.43 62.5
1.00 227.50 -182.85 133.58 50.0

Table 4 Stresses at representative positions and weight ratio in outer plate (X-direction)
Thickness of
plate, mm
Maximum
tensile stress
(point B), MPa
Maximum compressive
stress (point A), MPa
Tensile stress
(point C'),
MPa
Weight
ratio,
%
2.00 (Normal) 120.47 -116.40 58.21 100
1.75 131.81 -122.48 62.96 87.5
1.50 148.40 -132.42 69.92 75.0
1.25 172.85 -141.35 79.61 62.5
1.00 212.37 -137.93 95.93 50.0

plates are shown in Tables 3 and 4 respectively. The tensile stresses on the points B and C
increase when the thickness of the link plate decreases. Particularly, if the thickness
becomes smaller than 1.25mm, the stress increases. However, in the thickness range larger
than 1.50mm, weight saving ratio is similar to the rate of increase in tensile stress.
Consequently, weight saving with a decrease in the thickness of the link plate can be
realized using a higher strength material. Furthermore, the compressive stress at the point A
does not vary if the link plate becomes thinner.
4.3 Chamfering edges of link plate circumference
It is observed from the results in Figs.4 and 5 that the stress does not get centered at the
circumference of the link plate. Thus, a model with the chamfering of the edges of the link
plate circumference was focused. The chamfer dimension for the circumference is 1mm
(C1) and two models were established and analyzed. One model was chamfered at one edge
and the other model was chamfered at both edges. These models are shown in Figs.12 and
13. Moreover, a model with a centrally located hole of 3mm diameter, which was
chamfered at one edge, was also designed and analyzed, as shown in Fig.14.
The stresses of typical positions are shown in Table 5. In this table, the normal type
indicates a standard link plate. Type 1 indicates a model with the chamfering of one side,
type 2, a model with the chamfering of both edges, and type 3, a model with a centrally
located hole chamfered at one edge. It can be considered from the results in Table 5 that the
chamfering of one edge does not markedly affect the stress. Although the weight is reduced


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
167

Fig.12 Analysis model with chamfering on one side


Fig.13 Analysis model with chamfering on both sides


Fig.14 Analysis model with center hole and chamfering on one side

Table 5 Stresses at representative positions and weight ratio in outer plate (X-direction)
Analysis
model
type
Maximum tensile
stress (point B),
MPa
Maximum compressive
stress (point A), MPa
Tensile stress
(point C'), MPa
Weight
ratio, %
Normal 120.47 -116.40 58.21 100
Type 1 126.84 -117.54 58.11 93.3
Type 2 132.26 -108.96 55.26 86.6
Type 3 124.59 -118.58 60.23 90.6

by 14% by chamfering both edges, the tensile stress of the point B increases. This is
considered to be due to the circumference of the link plate becoming irritated. In the case of
type 3, the state of stress is similar to that of type 1 and the weight can be reduced by 3%.
The obtained results can be summarized as follows. To restrain the increase in stress
and to save weight, the form with a centrally located hole with the chamfering of one edge
is significant.



Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
168
5. Experimental proof obtained using prototype of link plate
Several link plates made of ABS resin were manufactured using a rapid prototyping
system. Then, tension tests were performed on the link plates and the above-mentioned
proposal was evaluated. The reason why a rapid prototyping system was used is as follows.
First, uniformity of specimens is required. Second, the thickness of the specimen plate
should be changed and analyzed sometime in the future. A rapid prototyping system is able
to fulfill that kind of demands inexpensively and simply.
The rapid prototyping system uses an extrusion technique. Namely, thermo plastics are
extruded from a fine nozzle and stacked layer by layer. The plastic material is in solution
when it is extruded from the nozzle. After the extrusion, it becomes solid with air cooling.
To maintain the layer thickness constant, the extruder capacity is controlled by a pump
depending on the scanning speed. Furthermore, the material for the supporting part is
stacked layer by layer from the other nozzle. This is to compensate for the strength of ABS
resin and to maintain the designed form. The material for the supporting part is dissolvable
in an alkaline solution. Thus, a complex form is realized by rapid prototyping.











Fig.15 Plate manufactured using rapid prototyping machine
(Right-hand side is showing a real plate.)

Figure 15 shows a link plate manufactured using a rapid prototyping system. The right
hand side of Fig.15 shows a real link plate. On the basis of the restriction in the dimensions
of the specimen mounted for the tension test machine, a double-size of link plate was
manufactured. In the tension test, two axes were reeved to both holes of the link plate. Then,
an axis was fixed through one hole, and the tension load and displacement of the other axis
through the other hole were measured. The tension speed was 0.1mm/s.
The relationship between the displacement of the axis and the tension load for the
manufactured inner link plate with a centrally located hole is shown in Fig.16. The curves in
Fig.16 are out of relation to the existence or diameter of the hole. This result coincides with
the pattern of the values analyzed by FEM, as shown in Table 2.
Figure 17 shows the relationship between the displacement of the axis and the tension
load for the manufactured outer link plate with the chamfering of the circumference edge.
As shown in Table 5, it can be observed that the tensile stress increases in some small
measures. Thus, the breaking force decreases in magnitude for the chamfered link plate.
However, the curves of types 1 and 3 are similar, and this result corresponds to the
analytical result.
A cracked link plate is shown in Fig.18. As shown in Figs.4 and 5, the crack is
generated in an area of large tensile stress. This phenomenon coincides with that indicated
in reference 4. Although the manufactured link plates are made of ABS resin and have a
larger plastic deformation than the metal material, the experimental results are good
indicators of the tendency of the analytical results.


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
169
















Fig.16 Relationship between tension load and displacement of inner ring with center hole














Fig.17 Relationship between tension load and displacement of outer ring with chamfering








Fig.18 Photograph of cracked plate

6. Conclusions
This research deals with a weight saving method for a link plate of a roller chain. The
stress analysis was performed by FEM, and some methods were considered with the
suppression of the increase in stress. The main conclusions are as follows.

(1) The design proposal for the use of a centrally located hole in a link plate has a
beneficial effect on weight saving and yields a negligible stress.
0
200
400
600
800
1000
0 0.2 0.4 0.6 0.8 1
Displacement , mm
T
e
n
s
i
o
n

l
o
a
d

,

N
Normal
d=3mm
d=4mm
d=5mm
d=6mm
0
200
400
600
800
1000
0 0.3 0.6 0.9 1.2 1.5
Displacement , mm
T
e
n
s
i
o
n

l
o
a
d

,

N
Normal
Type1
Type2
Type3
Crack


Journal of Advanced Mechanical Design,
Systems, and
Manufacturing
Vol. 3, No. 2, 2009
170
(2) Although chamfering the circumference of the link plate edge can be effective in
weight saving, it is clarified that the stress increases when the circumference
becomes irritated.
(3) In the link plate with a centrally located hole with the chamfering of the
circumference edge, the stress increases by almost 3%. However, 10% weight
saving can be realized and the design proposal is expected to be effective.
(4) A tension test was conducted for several manufactured link plates made of ABS
resin. Then, the appropriateness of weight saving design and stress analysis was
verified in this research.

In the future, the authors are going to investigate the aggregative state of stress and the
deformation of a chain including pins and bushes. On the basis of the obtained results, more
sufficient measures for weigh saving should be established. Furthermore, problems of
fatigue and safeness should be confirmed by dynamic load.

References
(1) e.c., Iwashimizu S., Fluctuation of Sprocket HEEL Teeth Load in Chain Transmission,
Bulletin of Laboratory in Ritsumeikan University, 61; 2002; 167 (written in Japanese).
(2) JIS B 1801; 1997, Short-Pitch Transmission Precision Roller Chains and Bush Chains
(written in Japanese).
(3) Standard of Japan Chain Association, Chains for Motor Bike, 1; 1981 (written in
Japanese).
(4) Nakagome M., Story of Chain, Japanese Standards Association,, 1997; 101 (written in
Japanese).
(5) Homepage of Enuma Chain Mfg. Co., Ltd.; http://www.enuma.co.jp/ (accessed on Dec.
10, 2008)
(6) Homepage of Marubeni Information Systems Co., Ltd.;
http://www.marubeni-sys.com/de/dimension/ (accessed on Dec. 10, 2008)