MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.

1051/matecconf/201815701012
MMS 2017

Computer simulation of the load test of an

extrusion blow moulded product
Michał Modławski*,1, Jacek Nabiałek1, Tomasz Jaruga1
1
Department of Polymer Processing, Czestochowa University of Technology, Armii Krajowej 19c,
42-201 Częstochowa, Poland

Abstract. The results of compression "top load" test of an extrusion blow moulded
container, performed in ANSYS Mechanical software, were presented. The non-
uniform wall thickness of the blown product was considered because the thickness
distribution was the result of the first step of the simulation in ANSYS Polyflow - the
simulation of the product manufacturing by extrusion blow moulding. The hydrostatic
pressure factor was taken into account during the "top load" simulation, because such
a test is usually done for filled products. It was found that the biggest stress values
occur around the inlet as well as in the side walls of the container.
Keywords: extrusion blow moulding simulation, top load test

1 Introduction
Extrusion blow moulding technology is used for many years, but because of the growing
demand on plastics packagings in the last years this process is developing fastly, which is
remarkable not only by increasing number of machines working in the industry but also by
increasing the technological innovations [1].
Although the simulation of injection moulding process is widespread today, the
simulation of blow moulding is still not used so much. The results of this simulation could
contribute to the improvement in new products development thanks to the possibility of
design modification at the product design stage. The computer simulation gives also an
useful information about the processing parameters efficiency which is important while
designing a product and a mould, that means before taking the costs of building a tool.
The research works regarding the computer simulation of extrusion blow moulding
done so far were mainly about wall thickness distribution optimization [2-10], by
controlling the extruded parison wall thickness distribution, for example. It is not only
possible to optimize the wall thickness distribution, but also to reduce the mass of the
product, using a computer simulation as a tool [7]. The processing conditions like blowing
pressure, parison temperature or its wall thickness have a big impact on products’
properties and on their quality. However, not all quality factors can be simulated nowadays
– for example the surface roughness of a product and the degree of cavity mapping. These
factors are an important quality indicators and they highly depend on processing parameters
[11].

*
Corresponding author: [email protected]
Reviewers: Kateryna Kravchenko, Ján Vavro

© The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons
Attribution License 4.0 (http://creativecommons.org/licenses/by/4.0/).
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

The modeling of the product strength [8-10] with the reference to the results of blow
moulding computer simulation is especially important because it allows modifying product
design and processing conditions to obtain the required strength. The wall thickness
distribution result should be used to simulate the strain and stress of the loaded container or
to simulate the stress in the case of free fall of the container which occurs during the typical
“drop test” of such products [10].
A canister-type container is one of typical extrusion blow moulded products. This
product is specific due to its cubicoidal shape which makes the risk to thin the wall
significantly in the corners during the moulding process as well as due to the handle used
for carrying, but also being problematic during moulding [12-13].

2 Simulation research
The aim of this article was to determine the stress and strain distribution in a canister-type
product, during the compression test, using a computer simulation. This product should be
manufactured from high-density polyethylene, in an extrusion blow moulding process. The
canister was loaded with the concentrated force of 100N, coming from the palletization of
the filled products during their storage and transport, and loaded with the hydrostatic
pressure. The computer simulation of the extrusion blow moulding process was done using
ANSYS Polyflow software. The stress and strain distribution was simulated using ANSYS
Mechanical software which was convenient to use in this case because of the possibility to
transfer the data of manufacturing simulation from ANSYS Polyflow.

2.1 The project of the canister

The canister was designed in Siemens NX 8.5 CAD software. The following assumptions
regarding the design and dimensions of this container were made:
 the canister volume is no less than 5 l. The level of the free surface of the filling liquid
should be below the neck,
 easy demoulding of the product is possible because the surfaces paralell to the mould
opening direction are rounded, except of the upper surface of the handle, which is flat,
and except of the bottom surface of this container,
 the side walls of the canister are flat; in these places usually labels are placed,
 there are no geometrical elements making canister more stiff (increasing its strength)
except of the bottom,
 the bottom of the canister is of the shape which enables the canister to be put on a flat
surface and increasing the strength of this product,
 the inlet was extended and evened with the top surface of the handle (as the cap height).
The thread was neglected to simplify the shape used in the computer simulation.
The designed shape of the canister is presented in Fig. 1 with its main dimensions. The
shape of this container is symmetrical so only a half of the canister was assumed in the
blow moulding process and mechanical strength simulation.

2
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

Fig. 1. The shape and main dimensions of the canister

2.2 Blow moulding computer simulation

During the blow moulding process simulation 1 second time of this process was considered.
In this time the mould is closed with the velocity of 0.2 m/s (from 0 to 0.4 s) and after that
the parison is blown by the gas overpressure in the time interval from 0.4 to 1.0 s. The
assumed value of the gas overpressure increases between 0.4 and 0.5 s time and reaches the
value of 0.4 MPa (Fig. 2).

Fig. 2. The function of mould closing speed and blowing gas overpressure versus time
This type of container is manufactured very often from high-density polyethylene (PE-
HD). In this simulation PE-HD Tipelin 6000B, produced by MOL Petrochemicals was
selected. The rheological viscosity curve was determined for this material in the
temperature of 160°C and in the shear rate range of 0.005 ÷ 0.5 s-1 using Rheolyst AR
1000N rotational rheometer, manufactured by TA Instruments. The average value of
viscosity in this shear rate range is 56.4·103 Pa·s. The density of this material in the liquid
state is 0.770 g/cm3 in the temperature of 160°C. The density was determined using pvT-
100 device.
The diameter of the extruded parison was 130 mm. Its initial thickness was computed
by Parison Programming module of ANSYS Polyflow software. Some initial simulations
were conducted to obtain this thickness. It was assumed that the minimal thickness of the

3
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

canister should be no less than 0.6 mm. The thickness distribution of the parison along its
height is shown in Fig. 3.

Fig. 3. Parison dimensions – wall thickness distribution

The boundary conditions were imposed on the horizontal and vertical edges of the
parison. The condition on vertical edges involves plane symmetry with a normal vector
directed along Y axis. The direction of Y axis corresponds to the direction of the mould
closing and opening. The boundary condition on the horizontal edges is defined by zero
value of: the normal velocity and the force tangent to the plane containing the horizontal
parison edge. Between the finite elements of the parison and mould mesh a contact
condition was imposed. The contact is applied when the distance between the nodes of
these meshes is smaller than 0.4 mm.
The size of the parison mesh finite elements is 5 mm, which means the distance
between the nodes. The mesh modifier Mapped Face Meshing was used to obtain even
distribution of nodes, without mesh deformation. Adaptive Meshing function was also
applied to achieve good mesh quality in the handle area. This function is used to increase
the number of finite elements (mesh density) when:
 the distance between a node of the parison mesh and a node of the mould mesh –
expression 1   dB  from the equation (1) - is smaller than the defined value,
2

 the radius of the parison mesh curvature – expression A  r  C is of a small value.

The maximum number of mesh divisions equal 2 was assumed. The mesh elements are
divided when:

4
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

   2

new _ size  max size _ min,  A  r  C   1   d 

(1)

   B   


where:
 size_min – minimum size of an element in the parison –1.6 mm was assumed for
computations,
 A – fraction of the radius of curvature – 0.2 was assumed,
 B – typical distance when mesh is begun to refine – value of 5 mm,
 C – coefficient of proportionality – the assumed value was 0,
 d – distance between node E of a mesh parison and the nearest node N of a mesh of the
mould,
 r – local radius of curvature at node N [14].
The simulation results of blow moulding – the distribution of the canister wall thickness is
shown in Fig. 4. The thickness distribution of the product - canister is homogeneous, except
of some small thin areas in the handle and some thicker wall below the neck by the parting
plane. This proves that distribution of the initial parison thickness was well determined.

Fig. 4. Blow moulding simulation result

2.3 Computer simulation of the loaded product

The model was supplemented with two plates to simulate the load. The plate on the bottom
of the canister is imposed with fixed support condition and the plate on the top is loaded
with 100 N concentrated force, directed down along the axis Z-. The value of 100 N acts on
one canister when being transported in the bottom layer, below two other layers of these
products filled with water. This is an usual condition of the transport of such products. The
numerical calculations were conducted for two cases. In the first case a load caused by
hydrostatic pressure was not taken into account, but in the second case it was.

5
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

The contact condition was defined between the bottom surface of the canister and the
upper surface of the bottom plate as well as between the upper surface of the handle
together with the upper edge of an inlet and a bottom surface of the upper plate.

Fig. 5. Preparing load simulation a) mesh used for the load simulation, b) canister wall thickness
distribution imported from ANSYS Polyflow solver
For the numerical computations of the load test the surface model with the additional
thickness parameter was involved. The thickness values of the container were imported
from a text file, generated during the parison blow moulding simulation in Polyflow solver.
The thickness distribution after the blow moulding process computer simulation is
presented in Fig. 5a. The thickness value was generated inside the canister so the outer
surface of the container was unchanged. After mapping of the thickness 3D finite element
mesh can be generated. The average size of the container mesh edges was equal 3 mm, after
adding a mesh modifier (Body Sizing) to the canister model. The mesh is shown in Fig. 5b.

2.4 Results of computer simulation of the load

The numerical simulation results of the load test were: stress distribution and ready product
deformation, computed for two cases:
 canister loaded by a vertical force equal 100 N,
 canister loaded by a vertical force equal 100 N and the internal walls loaded by
hydrostatic pressure resulting from filling the container by a liquid.
Loading the designed canister with the vertical force equal 100 N causes that the stress is
created in the product. The stress distribution is shown in Fig. 6. The maximum value
equals 9.13 MPa and occurs in the area below the inlet. A big stress values are also in the
zone of the vertical walls of the canister handle.

6
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

Fig. 6. Distribution of the stress without hydrostatic pressure (Equivalent Stress)

The canister deformation caused by the vertical force of 100 N is presented in Fig. 7.
This force causes the biggest displacement of the rear wall in a place where the vertical
surface changes into the handle wall at 11° angle. A significant deformation occurs also in
the analogical zone on a front wall.

7
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

Fig. 7. Strain distribution – real scale (Total Deformation)

After adding a load resulting from hydrostatic pressure the maximum value of the stress
increased by 19%. In this case the side walls of the canister are much more loaded – see
Fig. 8 compared with Fig. 6. The maximum stress value calculated occurs in the bottom
zone of the side walls.
The maximum value of stress equal 10.9 MPa does not exceed the maximum tensile
strength at yield, declared by the manufacturer of this plastic: 32 MPa [15]. In order to
predict whether the canister wall can be thinner than this obtained from the current
simulation some numerical simulation of dynamical loads should be done.

8
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

Fig. 8. Stress calculated considering hydrostatic pressure (Equivalent Stress)

Filling the canister with a liquid causes a large deformation of this product in the side
wall zone and also in a zone of front and back vertical walls (Fig. 9). The displacement
values caused by the vertical force are relatively small in comparison to the displacement
values caused by the hydrostatic pressure.

9
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

Fig. 9. Total deformation – in three axes directions

Conclusions
A study of a combined computer simulation was presented in this article. The mechanical
strength simulation of a canister-type plastic container loaded with a compression force was
made, but based on the results of the simulation of manufacturing this product by blow
moulding – on the base of the wall thickness distribution result, since the wall thickness
impacts the local product strength.
Results of numerical calculations of strength made using the data coming from he
simulation of manufacturing process allow obtaining more accurate data, important
regarding the predicted quality of a new product. Using the results of polymer processing
simulation for strength calculations improves the quality of forecasting the properties of
plastics products.
On the base on the computer simulation study the following conclusions can be drawn:
 simulation of the extrusion blow moulding process allows predicting many properties of
a designed product which is going to be manufactured. A key result of this simulation is
the distribution of the product wall thickness,
 using results of the polymer processing simulation further in mechanical strength
computer simulation enables stress and strain analysis of a designed product without
making real tests, and eliminating the problems with incorrect product design,
 contemporary computer simulation programs allow making virtual analyses concerning
each aspect of the plastic product lifetime - starting from a shape design, through usage
properties prediction up to the product utilization.

10
MATEC Web of Conferences 157, 01012 (2018) https://doi.org/10.1051/matecconf/201815701012
MMS 2017

The authors wish to thank Department of Polymer Engineering and Technology from Wrocław
University of Science and Technology for the rheological measurements and Department of Polymer
Processing from Lublin University of Technology for p-v-T measurements.

References
1. H. Meetwall, Virtual prototyping for the blow molding industry. (ANSYS U.S.A.
2012)
2. S. Tanoue, T. Kajiwara, K. Funatsu, K. Terada, M. Yamabe, Numerical Simulation of
Blow Molding – Prediction of Parison Diameter and Thickness Distributions in the
Parison Formation Process. Polym. Eng. Sci. 36, 2008-2017 (1996)
3. S. I. Tanifuji, Overall Numerical Simulation of Extrusion Blow Molding Process.
Polym. Eng. Sci. 40, 1878-1893 (2000)
4. K. Pepliński, M. Bieliński, Polyflow software use to optimize the parison thickness in
blowing extrusion. Journal of POLISH CIMAC. R. 4 (3), (2009)
5. C. Gauvin, F. Thibauld, D. Laroche, Optimization of Blow Molded Part Performance
through Process Simulation. Polym. Eng. Sci. 43, 1407-1414 (2003)
6. Y. Fukuzawa, S. Tanoue, Y. Iemoto, R. Kawachi, H. Tomiyama, Three-dimensional
simulation on multi-layer parison shape at pinch-off stage in extrusion blow molding.
Polym. Eng. Sci. 50, 1476-1484 (2010)
7. J. C. Yu, X. X. Chen, T. R. Hung, F. Thibauld, Optimization of extrusion blow molding
processes using soft computing and Taguchi's method. J. Intell. Manuf. 15, 625-634
(2004)
8. M. Modławski, J. Nabiałek, T. Jaruga, Symulacja odkształcenia wyrobu
rozdmuchowego o stałej i zmiennej grubości ścianki. Tworzywa Sztuczne w Przemyśle
34, 94-97 (2016)
9. D. Kwiatkowski, M. Modławski, T. Jaruga, Symulacje komputerowe grubości ścianki
butelki uzyskiwanej w procesie wytłaczania z rozdmuchiwaniem. Przetwórstwo
Tworzyw 21, 256-261 (2015)
10. P. Klein, F. Fradet, H. Meetway, T. Marchal, Virtual Prototyping Applied to a Blow-
Molded Container. (Proceedings of SPE ANTEC, Orlando, Florida, U.S.A., 2012)
11. K. Pepliński, M. Bieliński, Wpływ warunków przetwórstwa na wybrane wskaźniki
użytkowe wytworów wytłaczanych z rozdmuchiwaniem. (Postęp w przetwórstwie
materiałów polimerowych, Ed. J. Koszkul, E. Bociąga, 176-183, Częstochowa 2006)
12. D. V. Rosato, A. V. Rosato, D. P. DiMattia, Blow Molding Handbook. (Hanser
Publishers, Munich, 2004)
13. K. Pepliński, Problem rączki. Jak wyeliminować problemy w procesie wytłaczania z
rozdmuchiwaniem podczas formowania rączek pojemników?. Plastics Review 71-72,
46-47 (2007)
14. ANSYS Polyflow help documents (2017)
15. TIPELIN 6000B Technical Data Sheet - MOL Petrochemicals:
https://mol.hu/en/polymers/products/hdpe-medium-and-high-density-polyethylene,
(Access: 07.2017)

11