Optimization of Box Type Girder of Overhead Crane

Muhammad Abid, Muhammad Hammad Akmal, Shahid Parvez

GIK Institute of Engineering Sciences and Technology, Topi, Pakistan

Abstract
Double girder box girder over head cranes are used for heavy duty applications in
the industry. In this paper a detailed parametric design optimization of the main
girder of box type is performed for a 150Ton capacity and 32m long span crane,
after its basic design using available design rules. Design optimization is
performed using detailed 3D finite element analysis by changing the number, shape
and location of horizontal stiffeners along the length of the girder and number and
location of stiffeners along the vertical direction to control any possible buckling,
light weight and for safe stress and deflection. During optimization, primarily
calculated thickness of the box girder plates is not changed.

Keywords: Box, girder, optimization, overhead, crane, FEA

1. Introduction
Overhead cranes are used for the handling and transfer of heavy loads from one
position to another, thus they are used in many areas of industry such as in
automobile plants and shipyards [1,2] etc. Their design features vary widely
according to their major operational specifications such as: the type of motion of
crane structure, weight and type of the load, location of the crane, geometric
features and environmental conditions. Since the crane design procedure is highly
standardized with these components, main effort and time is spent mostly for
interpretation and implementation of available design standards [3]. There are
many of the published studies on their structural and component stresses, safety
under static loading and dynamic behavior [5-16]. Solid modeling of bridge
structures and finite element analysis (FEA) to find the displacements and stress
values has been investigated by Demirsoy [17]. Solid modeling techniques applied
for the road bridge structures, and these structures analysed with finite element
method has given by [18-20]. DIN-Tashenbuch and F.E.M (Federation Européan
de la Manutention) rules offer design methods and empirical approaches and
equation that are based on previous design experiences and widely accepted design
procedures. DIN-Tashenbuch 44 and 185 are collection of standards related to the
crane design. DIN norms generally state standard values of design parameters.
F.E.M rules are mainly an accepted collection of rules to guide the crane designers.
It includes criteria to decide on the external loads to select crane components. In
this paper a detailed parametric design optimization of the main girder of box type
610 M. Abid, M.H. Akmal and S. Parvez

is performed for a 150Ton capacity and 32m long span crane, after its basic design
using available DIN and F.E.M design rules. Design optimization is performed
using detailed 3D FEA, by changing the number, shape and location of horizontal
stiffeners along the length of the girder and number and location of stiffeners along
the vertical direction to control light weight and for safe stress and deflection.
During optimization, primarily calculated thickness of the box girder plates is not
changed. Three case studies are carried out for optimization using;
x horizontal stiffeners only (study-1)
x vertical stiffeners only (study-2)
x Both horizontal and vertical stiffeners (study-3)

Figure 1. Initial geometry of the overhead crane girder

2. Modeling, Material Properties and Meshing

A complete box girder is modeled in ANSYS software and is shown in Figure 1
with all its dimensions. Thickness of side plates = 16mm, top and bottom plates =
22mm, vertical stiffeners = 10mm, width of top and bottom plates = 960mm and
height of side plates in the center = 2600mm. However during FEA due to its
symmetry, only half of the model is used and is optimized with different
geometries under applied loading conditions. Initially box with rail at the top is
analyzed without any stiffener. Then different horizontal and vertical stiffeners at
different stages were modeled and glued to the outer box keeping in view the
manufacturing process and symmetry in front. Linear elastic material model is used
for steel Rst-37.2 with Young’s modulus of 207GPa, Poisson’s ratio of 0.3,
allowable stress 157 MPa and density of 7.86*10-6 kg/m3. 3-D, 10 nodded higher
order quadrilateral SOLID187 elements having three degrees of freedom at each
node are used. Free Mesh option is used to mesh the entire geometry and is shown
in Figure 2.
 Optimization of Box Type Girder of Overhead Crane 611

2.1 Boundary Conditions

Considering crane standing at one position and lifting the load, as mostly is the
recommendation for crane operation, hence during design calculation and finite
element analysis, no horizontal force is considered to be acting on the main girder.
Main girder is fully fixed at the ends where it is joined to the end carriages. A three
point bending loading strategy is applied considering the distance between two
wheels of the trolley to be very small. Load is applied along the rail width equally
distributed on all the 6 nodes. For different case studies load applied is considered
with the self weight of the main girder and is discussed in related sections below.
Due to the symmetry of the geometry, symmetry boundary conditions are applied
on the plates as shown in Figure 2(a).

Symmetric
Load on rail
Boundary
conditions

Fully fixed

Figure 2(a). FE model with applied boundary conditions

Figure 2(b). FE model using SOLID WORKBENCH

3. Results and Discussion

Maximum bending stress with and without stress concentration points are shown in
Figure 3 and 4.
612 M. Abid, M.H. Akmal and S. Parvez

Figure 3. Bending stress in girder with maximum at rail due to stress concentration where
load is applied

Figure 4. Bending stress in girder by removing the volumes to avoid stress concentration,
hence redistributing the stresses.

3.1 Study-1: Optimization Using Horizontal Stiffeners

In this case optimization is performed by changing the number, position and shape
of horizontal stiffeners only. The details of all the cases are summarised in Table 1.
It is noted that there is no considerable decrease in the maximum deflection by
using the L-shape stiffeners, however better results are achieved using the C-shape
horizontal stiffeners. Using two C-shape horizontal stiffeners at 400 and 1700mm
 Optimization of Box Type Girder of Overhead Crane 613

from the top plate, the best optimized results (maximum deflection=37.32mm and
maximum bending stress=176MPa, mass of girder=16999kg) are achieved.
Analysis is also performed by modeling the girder in Ansys Workbench. Using
built in solid elements and free meshing and removing the stress concentration
points, maximum deflection = 36.24mm and bending stress = 165 MPa is observed.
Although maximum bending stress is more than the allowable but can be neglected
due to the stress concentrations in all the cases.

3.2 Study-2: Optimization Using Vertical Stiffeners

In this case optimization is performed by changing the number and position of

plate stiffeners along the length of the girder. Results are summarized in the Table
2. It is observed that by increasing the number of vertical stiffeners from one to
two and so on, a decrease in the maximum deflection from 37.74mm to 34.79mm
is observed. By increasing the number of vertical stiffeners, corresponding
decrease observed is small. Hence using seven vertical stiffeners@2000 mm from
each other deflection reduced to 34.79mm, but an increase in mass (1042kg) of the
girder is observed as we increase the stiffeners from 1 to 7. A maximum bending
stress of 160MPa is observed which is very close to the allowable stress of the
flange material.
Using Workbench and neglecting the stress concentration, maximum deflection is
reduced to 29.52mm and maximum bending stress is reduced to 135 MPa
respectively, and is within the allowable limits.

3.3 Study-3: Optimization Using Both the Horizontal

and Vertical Stiffeners

In this case, analyses are performed by changing the number and location of the
vertical stiffeners along the length of the girder in addition to the two C-shape
horizontal stiffeners positioned equally along the height of the girder. Results are
summarized in the Table. 3. Two C-shape horizontal stiffeners are used as most
optimized results were concluded using these in study-1. In these cases, the number
of vertical stiffeners is increased, the value of maximum deflection decreases from
34.23 to 34.06mm and the value of maximum bending stress decreases from 166 to
160MPa. It is interesting to note that using vertical stiffeners from 3 to 7,
maximum deflection and bending stress remains the same. But using more vertical
stiffeners, mass of girder is increased. Vertical plates are used here in order to
avoid lateral buckling. Using Workbench model and neglecting stress
concentrations, maximum deflection and stress is reduced to 29.32mm and
131MPa and is within the allowable limits.
After that, box girder is modeled by using the dimensions such that two C-shape
horizontal stiffeners are placed @625mm and 1250mm from the top plate and
twenty one vertical stiffeners are used in the half model of girder. First four
vertical stiffeners are located along the support and varying cross section and the
remaining 17 vertical stiffeners are located along the length of the girder where the
height of the girder is uniform. For optimization 21 and 31 vertical stiffeners are
614 M. Abid, M.H. Akmal and S. Parvez

also used and analysis is performed. In addition position and orientation of the
horizontal stiffeners is also changed such as using inverted C-shape stiffeners and
so on and results are summarized in Table 4. Using 17 vertical stiffeners in
addition to two C-shape stiffeners, a maximum deflection = 32.45mm and
maximum bending stress = 218MPa is observed. Using Worbench model with 17
vertical stiffeners and removing stress concentrations, maximum deflection and
stress is reduced to 28.62mm and 132 MPa. By using L-shape horizontal stiffeners
in addition to vertical stiffeners, results are also found in good agreement to that
using 2 C-shape stiffeners but with a slight increase in the weight of the girder.
Using inverted 2 C-shape stiffeners, no difference in results is observed but from
manufacturing point of view, this is not appreciated.

(a) (b) (c)

Figure 5. Different Orientations of horizontal stiffeners
 Optimization of Box Type Girder of Overhead Crane 615

Table 1. Results comparison by changing the shape, number and location of horizontal
stiffeners

Max Mass of
# and type Max bending
Location deflection girder
of stiffeners stress (MPa)
(mm) (kg)

No stiffener ----- 47.49 210 15865

Touching top plate 41.91 206 16488
1 C-Shape @400mm from top plate 42.98 205 16457
horizontal @650mm from top plate 42.97 325 16466
stiffener @890mm from top
180x70x8 plate(aligned with lower 41.94 208 16438
plate)
Equally divided
throughout the height 39.77 187 16972

1st@710mm, 2nd@1655
2 C-Shape 40.63 353 16999
mm from top plate
horizontal
1st@400mm,
stiffener
2nd@1700mm from top 37.32 176 16999
180x70x8
plate
1st@400mm,
2nd@1700mm from top 36.24 165 16999
plate(WORKBENCH)
Equally divided
3 C-Shape 40.97 318 17539
throughout the height
horizontal
1st@710mm, 2nd@1340
stiffener
mm, 3rd@1970mm from 39.36 350 17537
180x70x8
top plate
Touching upper plate
1 L-Shape 46.54 208 16466
horizontal
@400mm from top plate
stiffener 44.90 213 16466
156x156x8
@878mm from top plate 42.75 206 16438
2 L-Shape Equally divided
throughout the height 42.00 200 16971
horizontal
stiffener
1st@722mm, 2nd@1661
156x156x8 43.71 201 17001
mm from top plate
Equally divided
3 L-Shape throughout the height 45.00 204 17540
horizontal st nd
stiffener 1 @722mm, 2 @1348
156x156x8 mm, 3rd@1974mm from 43.71 203 17538
top plate
616 M. Abid, M.H. Akmal and S. Parvez

Table 2. Results comparison by changing the number and location of vertical stiffeners
Maximum Maximum
Location and number of Mass of girder
deflection bending stress
vertical stiffeners (kg)
(mm) (MPa)
1@6500mm from center 37.74 179 16039
2@12000mm from each other 35.39 167 16213
3@6000mm from each other 35.03 166 16386
4@4000mm from each other 34.86 165 16560
5@3000mm from each other 34.79 165 16734
6@2400mm from each other 34.78 165 16907
7@2000mm from each other 34.79 160 17081
7@2000mm from each other
29.52 135 17081
(WORKBENCH)

Table 3. Results comparison by changing the location of vertical stiffeners in addition to

two C-shape stiffeners of study-1
Location and # of Maximum Maximum bending Mass of girder
vertical stiffeners deflection (mm) stress (MPa) (kg)
3@6000mm 34.24 166 17472
5@3000mm 34.07 164 17806
7@2000mm 34.06 165 18140
7@2000mm
29.32 131 18140
(WORKBENCH)

Table 4. Results comparison by changing the number and location of vertical stiffeners in
addition to two different types of horizontal stiffeners
Number and type of using stiffeners Max Max bending Mass of
deflection stress (MPa) girder
(mm) (kg)
ORIENTATION OF HORIZONTAL STIFFENER AS PR FIG. 1.5a
17@750mm along uniform height 32.45 218 20104
17@750mm along uniform height
(WORKBENCH) 28.62 132 20104
21@600mm along uniform height 32.45 221 20779
31@400mm along uniform height 32.70 224 22451
ORIENTATION OF HORIZONTAL STIFFENER AS PR FIG. 1.5b
2 C-Shape horizontal stiffeners 32.35 220 20114
ORIENTATION OF HORIZONTAL STIFFENER AS PR FIG. 1.5c
2 C-Shape horizontal stiffeners 32.40 220 20123
CHANGING POSITION OF HORIZONTAL STIFFENERS
2 C-Shape stiffeners @866 and
1733mm from top plate 32.05 218 20123
2 C-Shape stiffeners @866 and
1733mm from top plate 28.18 129 20123
(WORKBENCH)
CHANGING THE SHAPE OF HORIZONTAL STIFFENERS
2 L-Shape horizontal stiffeners 32.31 220 20169
 Optimization of Box Type Girder of Overhead Crane 617

4. Conclusions
From detailed optimization studies following results are concluded;

1. The most optimized case concluded is with 2 C-Shape horizontal

stiffeners equally distributed along the height with 17 vertical stiffeners
along the uniform height and 4 along the support point and varying
section. Here, the maximum deflection and stress is reduced to 28.18mm
and 129 MPa according to the Workbench model.
2. The results achieved from the model of ANSYS Workbench are 10 %
more accurate than ANSYS model and are concluded due to less
discretisation error.
3. Orientation of the horizontal stiffeners does not make visible difference in
the results.
4. The minimum deflection is achieved by equally dividing the horizontal
stiffeners along the height.
5. To control longitudinal and lateral buckling, use of horizontal and vertical
stiffeners is strongly recommended. In addition, inclusion of stiffeners
increases the strength of the girder.
6. In order to further reduce the weight of the girder in future, variation of
plate thicknesses and use of other sections is recommended.

5. References
[1] Oguamanam, D.C.D., Hansen, J.S., Heppler, G.R., (1998) Dynamic Responce of an
Overhead Crane System,Journal of Sound and Vibration, 213 (5), 889 – 906.
[2] Otani, A., – Nagashima, K. –Suzuki, J.: Vertical Seismic Responce of Overhead Crane,
Nuclear Eng. And Design, 212, 1996, p. 211 – 220.
[3] Erden, A. (2002) Computer Automated Access to the “F.E.M. rules” for Crane Design,
Anadolu University Journal of Science and Technology, 3 (1), 115-130.
[4] Anon, A. (1998) New Thinking in Mobile Crane Design, Cargo Systems, 5 (6), 81.
[5] Baker J.: (1971) Cranes in Need of Change, Engineering, 211 (3), 298.
[6] Buffington K.E. (1985) Application and Maintenance of Radio Controlled Overhead
Travelling Cranes, Iron and Steel Engineer,62 (12), 36.
[7] Demokritov V.N. (1974) Selection Of Optimal System Criteria For Crane Girders,
Russian Engineering Journal, 54 (4), 7.
[8] Erofeev M.J. (1987) Expert Systems Applied To Mechanical Engineering Design
Experience with Bearing Selection and Application Program, Computer Aided
Design, 55 (6), 31.
[9] Lemeur M., Ritcher C., Hesser L. (1977) Newest Methods Applied to Crane Wheel
Calculations in Europe, Iron and Steel Engineer, 51 (9), 66.
[10] McCaffery F.P. (1985) Designing Overhead Cranes for Nonflat Runways, Iron and
Steel Engineer,62 (12), 32.
[11] Reemsyder H.S., Demo D.A. (1978) Fatigue Cracking in Welded Crane Runway
Girders, Causes and Repair Procedures, Iron and Steel Engineer, 55 (4), 52.
[12] Rowswell J.C., Packer J.A. (1989) Crane Girder Tie-Back Connections, Iron and Steel
Engineer, 66 (1), 58.
618 M. Abid, M.H. Akmal and S. Parvez

[13] Moustafa, K.A., Abou-El-yazid, T.G. (1996) Load Sway Control of Overhead Cranes
with Load Hoisting via Stability Analysis, JSME Int. Journal, Series C, 39 (1), 34–40.
[14] Oguamanam, D.C.D., Hansen, J.S., Heppler, G.R. (2001) Dynamic of a Three-
dimensional Overhead Crane System, Journal of Sound and Vibration, 242 (3), 411–
426.
[15] Auering, J.W., Troger, H. (1987) Time Optimal Control of Overhead Cranes with
Hoisting of the Load, Automatica, 23 (4), 437–447.
[16] Huilgol, R.R., Christie, J.R., Panizza, M.P. (1995) The Motion of a Mass Hanging
From an Overhead Crane, Chaos, Solutions & Fractals, 5 (9), 1619–1631.
[17] Demirsoy, M. (1994) Examination of the Motion Resistance of Bridge Cranes, PhD.
Thesis, Dokuz Eylul University, Izmir, Turkey.
[18] Ketill, P., Willberg, N.E. Application of 3D Solid Modeling and Simulation Programs
to a Bridg Structure, PhD. Thesis, Chalmers University of Technology, Sweden.
[19] Celiktas, M. (1998) Calculation of Rotation Angles at the Wheels Produced by
Deflection Using Finite Element Method and the Determination of Motion Resistance
in Bridge Cranes, J. Of Mechanical Design, 120.
[20] Alkin, C. (2004) Solid Modeling of Overhead Crane’s Bridges and Analyse with Finite
Element Method, M.Sc. Thesis, Istanbul Technical University, Turkey.
[21] Scheffer, M., Feyrer, K., Matthias, K. (1998) Fördermaschinen Hebezeuge, Aufzüge,
Flurförderzeuge, Vieweg & Sohn, Wiesbaden.
[22] Kogan, J.: Crane Design. (1976) Theory and Calculations of Reliability, John Wiley &
Sons, New York.
[23] Errichello, R. (1983) Gear Bending Stress Analysis, ASME Journal of Mechanical
Design 105, 283–284.
[24] Moaveni, S. (1999) Finite Element Analysis : Theory and Application with ANSYS,
Prentice-Hall, New Jersey.
[25] Verschoof, J. (2000) Cranes Design, Practice and Maintenance, Professional
Engineering Pub. London.