Jtam-A4.dvi


JOURNAL OF THEORETICAL

AND APPLIED MECHANICS

53, 4, pp. 799-809, Warsaw 2015
DOI: 10.15632/jtam-pl.53.4.799

EVALUATION OF SPECIFIC PROCESS PARAMETERS AND
ULTRASONICALLY ACTIVATED INJECTION AFFECTING THE QUALITY

OF FILLING IN THIN WALLED PLASTIC PARTS

Andrei Adam, Daniel Stan, Aurel Tulcan
Politehnica University Timisoara, Department of Materials and Manufacturing Engineering, Romania

e-mail: andreiadam83@yahoo.com

The paper presents a set of experiments focused on the study of rheological behavior of a
polymer flowing through a narrow section at the forming by injection of thin-walled plastic
parts. The paper addresses the use of ultrasonically activated injection for fabrication of po-
lymeric parts with thin wall features. In the experiment, a part with six different geometric
features has been created. The design-of-experiments approach is applied to correlate the
quality of the parts with the processing parameters. Four processing parameters are inve-
stigated using a screening factorial experimentation plan to determine their possible effect
on the filling quality of the moulded parts. The experiments have been conducted on a hot
runner mould with two nests in which the final (nest) nozzle has been modified to host,
as the central element, the ultrasonic horn of a sonic system. It has been found that the
ultrasonic activation applied on the active part of the mould does not play an important
role as a stand-alone factor but could amplify or strengthen the effect of classical setting
parameters (and influence factors) of the process: the melt temperature and injection pres-
sure. Because it is easier to stimulate and to control rheological properties of the melt by
setting the intensity of ultrasonic energy and, more important, the effect is forthwith, the
paper recommends the runner systems with ultrasonic activation as an alternative for the
hot runner with heating elements.

Keywords: ultrasonic activation, design-of-experiments, rheological behavior of the polymer,
thin-walled plastic parts

1. Introduction

Injectionmoulding is themost common formingmethod for themanufacturing of plastic parts.
With high productivity, it is based on accurate replication of the nest. In the industrial design
and manufacturing, it is always a challenge to make a proper compromise between the most
desirable shape of parts, tooling cost, their weight and as well as strength and rigidity (Adam
et al., 2013). The primary concern is to respect the quality requirements for themoulded parts.
Especially for thin-walled parts, difficulties in the process are due to poor rheological capabilities
of the melt flowing through thin section (negative of the thin-walled).
Based on the previous research results, for the ultrasonic activation of the extrusion (Stan,

1999; Stan et al., 2000), we supposed that the so called “thermo-pellicular effect” of the sonic
activation of the polymer melt under pressure could also be obtained in injection conditions to
improve the flow and replication capability of the melt in the above mentioned conditions.
The main approach to identify influential processing parameters in thin-wall injection mo-

ulding was by changing one parameter at a time while keeping the others constant, and then
observing the effects of that parameter (Wimberger-Friedl, 2000). This approach was inherited
fromconventional injectionmoulding. Itwas useful in drawingbasic conclusions about how each
parameter affects the filling quality of themoulded part. This approach, however, has twomain



800 A. Adam et al.

limitations (Eriksson et al., 2008): the first limitation is that it is relatively time consuming
whenmany parameters are being investigated, the second drawback is that it does not take into
consideration the effect of the interaction between two or more parameters, which is relevant
consideration in a complex process such as ultrasonically activated injection used on thin-walled
plastic parts.
In this paper, the design-of-experiments (DOE) approach is introduced into the research

domain as auseful alternative to conventionalmethods.Anumber of research groupshave useda
variety ofDOEexperimentation plans to investigate the relation between processing parameters
and part filling quality (Attia et al., 2009). The responses chosen for the experiments included
filling quality of micro-sized channels (Mönkkönen et al., 2002), part dimensions (Zhao et al.
2003; Aufiero, 2005; Baltes and Tierean, 2009; Pirskanen et al. 2005; Malloy, 1994) and flow
length (Jung et al., 2007).
Results presented in the literature show that different DOE designs yield different outputs.

For example, there is disagreement about the importance of holdingpressureand injection speed.
Furthermore, certain experiments have highlighted interactions between processing parameters
which have not been seen in other works. These differences in experimental results may be
due to different geometrical shapes as well as polymers and experimental set-ups used in each
experiment (Baltes and Tierean, 2009). It would, therefore, seem reasonable to claim that, at
present, significant processing parameters in thin wall injection are identified on case-by-case
basis and cannot be generalized for all situations.
This paper addresses the effects of ultrasonically activated injection and specific processing

parameters on the filling quality for thin walled plastic parts through the design-of-experiments
(DOE) approach.

2. Experimental part

2.1. Mould design

Mould flow simulations have been carried out using the finite element method (FEM)which
is one of the best methods to perform various computer and engineering simulations. This
method incorporates programs that have become essential parts of modern computer aided
design (Bariani et al., 2007).
Generally, using computer aided simulation in plastics technology, consists in assumption of

various scenarios with a certain number of injection points with different locations in relation
to the 3D configuration of the part to be injected with various injection point lengths and
diameters, and determining the correct size of the gate. For parts with thin walls this kind of
the engineering approach is very important because the location of the injection point should
be achieved as effectively as possible, so as to provide balanced mold filling (Chang, 2007).
For the experiment, a geometric part has been created that has flow channels with thickness

(gap, i) of 0.2mm, 0.3mm, 0.4mm, 0.5mm and 0.6mm (Fig. 1).
The experiments were done with the following injection parameters:

1. Tinj =230◦C, Pinj =200MPa (2000bar),

2. Tinj =280◦C, Pinj =200MPa (2000bar).

In the first case, the results show that in the flow channels with smaler gaps the injected
material (ABS)doesnot completely fill thewhole volume (Fig. 2).The significance of the colours
are: orange shows no problemswith the flow rate and red/grey that there will be problemswith
flow andmaybe no complete filling of the cavity.
In the second case, at a higher injection temperature, the results show that the flow channels

are completly filled, but for smaller gaps (i =0, 2 and 0, 3), there are still some issues (Fig. 3).



Evaluation of specific process parameters and ultrasonically activated... 801

Fig. 1. Proposed geometrical model of the injected part

Fig. 2. Flow simulations, first case

Fig. 3. Flow simulations, second case

2.2. Factorial experiment

For the experiment, we have chosen a patented invention (Fig. 4). The patent (Patent No.
118576 B/2003) relates to an inventive concept for the design of hot-runner injection molds in
which to improve the nests filling and quality of the parts, an ultrasonic converter (ultrasonic
transducer (1)+concentratorwave adapter (2)) is placed in thehotblock (plate) (3) centered on
each final nozzle (4), so that in the proximity of the injection point (gate) (5) a powerful thermo-
-pellicular effect is created in themolten plastic flowing in contact with thewave adapter and to
the top of this (Iclanzan et al., 2008). In order to estimate the effects of ultrasonic activation, the
second nestwas placed in themouldwith a classic (without ultrasonic activation) final injection
nozzle.



802 A. Adam et al.

Fig. 4. Ultrasonically activatedmould

The material used for the tests was acrylonitrile-butadiene-styrene (ABS) because of its
higher injection load resistance and a significant melt flow rate (MFI=35g/10min).

The tests were carried out on an injection mould (Fig. 5) that was mounted on the Krauss
Maffei KM 200-700 C2 injection machine.

Fig. 5. Cores of the injection mould

The injected parts that were obtained during the experiments (Fig. 6) were measured along
the length traveled by the plastic flow through the narrow cavities of the “ribs”, as the amount
of rheological capability of the melt.

The experimental planwas designed for four influence factors (independent variables), three
of them being the classical setting parameters for the control of rheological properties of the
melt: pressure, temperature and gap (opening of the free flowing section) plus the fourth factor,
ultrasonic activation of the injection (nest) nozzle. The response (dependent) variable considered
to be relevant for the assessment of rheological capabilities of the melt was the length of the
flow until solidification. The levels of the independent variables are presented in Table 1.

Experimental results obtained with and without ultrasonic activation are presented in
Table 2.



Evaluation of specific process parameters and ultrasonically activated... 803

Fig. 6. Parts obtained for different p-T combinations

Table 1. Levels of the independent variables

Pressure Temperature Gap Ultrasonic activation at
[bar] [◦C] [mm] maximum amplitude (US)

−1 +1 −1 +1 −1 +1 −1 +1

without
US

US of the injection nozzle:
1500 2000 230 280 0.2 0.5 Frequency: 35kHz

Power: 10% (Pnom =1kW)

Table 2.Experimental design results

Exp.
ctr.

Independent variables of the experimental design (coded) Dependent variable
Pressure Temperature Gap Ultrasoinc Lenght of the

P T i activation US rib L
[bar] [◦C] [mm] [−1=NO; +1=YES] [mm]

1 +1 −1 +1 +1 30
2 +1 +1 +1 +1 43
3 +1 −1 −1 −1 2
4 −1 +1 +1 −1 32
5 −1 −1 −1 −1 0
6 −1 −1 +1 +1 24
7 −1 +1 −1 −1 5
8 +1 −1 +1 −1 26
9 −1 +1 +1 +1 36
10 −1 +1 −1 +1 14
11 −1 −1 +1 −1 21
12 +1 −1 −1 +1 7
13 −1 −1 −1 +1 6
14 +1 +1 −1 −1 10
15 +1 +1 +1 −1 41
16 +1 +1 −1 +1 18

3. Experimental results

For the processing of the results, the StatgraphicsTM program has been used. The primary
influence factors that are taken into consideration are pressure, temperature and gap. Table 3
shows each of the estimated effects and interactions. Also shown is the standard error of each
effect, whichmeasures their sampling error, and that the largest variance inflation factor (V.I.F.)



804 A. Adam et al.

equals 1.0. For a perfectly orthogonal design, all factors would be equal to 1. Standard errors
are based on the total error with 5 d.f.

Table 3. Estimated effects for the travelled length

Effect Estimate Stnd. error V.I.F.

Average 19.6875 0.232177
A: Injection pressure 4.875 0.464354 1.0
B: Injection temperature 10.375 0.464354 1.0
C: Gap 23.875 0.464354 1.0
D: Ultrasonic activation 5.125 0.464354 1.0
AB 1.375 0.464354 1.0
AC 1.875 0.464354 1.0
AD −0.375 0.464354 1.0
BC 2.375 0.464354 1.0
BD 0.625 0.464354 1.0
CD −1.875 0.464354 1.0

In order to test the statistical significance of the effects, the analysis of variance, called
ANOVA has been used. The results obtained are shown in Table 3. The four independent va-
riables are encoded by A, B, C and D, and the interaction by AB, AC, AD, BC, BD and CD
which are products of the independent factors.

Table 4. ANOVA chart for the travelled length

Source Sum of squares D.f. Mean square F-ratio P-value

A: Injection pressure 95.0625 1 95.0625 110.22 0.0001
B: Injection temperature 430.563 1 430.563 499.20 0.0000
C: Gap 2280.06 1 2280.06 2643.55 0.0000
D: Ultrasoinc activation 105.063 1 105.063 121.81 0.0001
AB 7.5625 1 7.5625 8.77 0.0315
AC 14.0625 1 14.0625 16.30 0.0099
AD 0.5625 1 0.5625 0.65 0.4560
BC 22.5625 1 22.5625 26.16 0.0037
BD 1.5625 1 1.5625 1.81 0.2361
CD 14.0625 1 14.0625 16.30 0.0099
Total error 4.3125 5 0.8625
Total (corr.) 2975.44 15
R-squared=99.8551%, R-squared (adjusted for d.f.)=99.5652%
standard error of est.=0.928709, mean absolute error=0.445313
Durbin-Watson statistic=1.7971 (P =0.4020)
lag 1 residual autocorrelation=−0.0226449

The program determines, for each effect individually, the Fisher criteria which compare it
with the F-distribution table (critical values), and finally establishes the importance of that
factor which has a significant influence. In this case, the main factors and other three interac-
tions have P-values less than 0.05, indicating that they are significantly different from zero at
the 95.0% confidence level. The R-squared statistic indicates that the model as fitted expla-
ins 99.85% of the variability in the Travelled length. The adjusted R-squared statistic, which is
more suitable for comparingmodels with different numbers of independent variables, is 99.56%,
which shows that the mathematical modeling of the phenomenon is very good.



Evaluation of specific process parameters and ultrasonically activated... 805

The influence of the travelled lenght can be seen in the Pareto chart (Fig. 7), where the
vertical line represented for P =0.05 separates the significant factors from the others. The fact
that the release is random can also be seen in Fig. 8, where it can be observed that there is no
systematic arrangement between the experimentally obtained results and the residual dispersion
from the values estimated by themathematical model.

Fig. 7. Pareto chart for the traveled length

Fig. 8. Diagram of the residual

Based on the regression coefficients associated with the significant factors P , T , G, and
ultrasonic activation (US), Table 5, the length travelled by the plastic material a mathematical
model could be predictedMathematical model

Traveled length=19.6875+2.4375P +5.1875T +11.9375G+2.5625(US)

+0.6875PT +0.9375PG−0.1875P(US)+1.1875TG+0.3125T(US)
−0.9375G(US)

(3.1)

In Fig. 9, the main effects of the response function, i.e. the length traveled are shown. It is
noted that the four determinant factors and their interactions lead to an increase in the value
of the travelled lenght by increasing their minimum tomaximum.
In the next figures, variation of the estimated response surface (a) and its contours (b)

corresponding to the length travelled by the plastic material depending on combinations of the
four factors is shown.

4. Conclusions

The results of this paper show that for all the analyzed cases, with or without ultrasonic in-
jection, the gap i is the main influence factor. This situation is expected because, as in the
scientific literature, the lower limit usually considered for the gap (thinnest wall of the part) is



806 A. Adam et al.

Table 5. Estimated regression coefficients for the length travelled by the plastic material

Coefficient Estimate

Constant 19.6875
A: Presure 2.4375
B: Temperature 5.1875
C: Gap 11.9375
D: US activation 2.5625
AB 0.6875
AC 0.9375
AD −0.1875
BC 1.1875
BD 0.3125
CD −0.9375

Fig. 9. Main effects on the travelled length

Fig. 10. The influence of pressure and temperature

Fig. 11. The influence of pressure and gap



Evaluation of specific process parameters and ultrasonically activated... 807

Fig. 12. The influence of pressure and US activation

Fig. 13. The influence of temperature and gap

Fig. 14. The influence of temperature and US activation

Fig. 15. The influence of gap an US activation



808 A. Adam et al.

imin =0.5mm. Any increase of the flowing gap, even by 0.1mm, can decide on the capacity of
the melt to fill that area andmay dramatically affect the quality of the injected product.
Also the temperature T , shown in the Pareto chart (Fig. 7), is the second most important

factor influencing the value of the objective function L.
Note that theorderingof thegap and the temperature in thefirst twopositions is not affected

by including the ultrasonic activation in the study. This situation shows no major interactions
of the combined the i-US or T-US, excluding the possibility of spectacular effects from the
ultrasonic activation.
However, the sets of factorial experiments 24 in which the nozzle is ultrasonically activated

designate the US asthe third influencing factor after the contribution that it has to increase the
flow path.
Moreover, comparing the Pareto chart, one can conclude that the ultrasonic activation is

more important for the quality of the filling than the injection pressure (exceeding the influence
of the pressure in the Pareto charts) that conventionally has an outstanding influence on the
length travelled by themelt.
This situation is a solid argument for the implementation of the industrial ultrasonic activa-

tion in injection moldings.

Acknowledgment

This paper has been supported by the Sectoral Operational Programme Human Resources Deve-
lopment POSDRU/159/1.5/S/137516 financed from the European Social Fund and by the Romanian
Government.

References

1. Adam A., Cosma C., Dume A.I., Jadaneantu S., Pop A., 2013, Integration of CAD CAM
techniques in the development of an injection mould for automotive parts, Proceedings of AMS
2013 International Conference, Timisoara, Romania, 25-27

2. Attia U.M., Marson S., Alcock J.R., 2009Micro-injectionmoulding of polymermicrofluidic
devices,Microfluid Nanofluid, 7, 1-28

3. Aufiero R., 2005, The effect of process conditions on part quality in microinjection molding,
ANTEC: Proceedings of Annual Technical Conference, Boston,MA, 36-40

4. Baltes L., Tierean M., 2009, Influence of the part shape on the polymer flow inside mould
and overall strength, Annals of DAAM for 2009, Proceedings of the 20th International DAAM
Symposium, 20, 1, 1107-1108, ISSN: 1726-9679

5. Bariani P.F., Salvador M., Lucchetta G., 2007, Development of a test method for the
rheological characterization of polymers under the injectionmolding process conditions, Journal of
Materials Processing Technology, 191, 1/3, 119-122

6. Chang D.H., 2007,Rheology and Processing of Polymeric Materials, Vol. 2, Polymer Processing,
Oxford University Press Inc., NewYork, www.oup.com, ISBN: 978-0-19-518783-0

7. Eriksson L., Johansson N., Kettaneh-Wold N., Wikstrom C., Wold S., 2008,Design of
Experiments: Principles and Applications, 3rd ed., Umetrics, Ume̊a

8. IclanzanT., Serban I., Rotar D., Tut V., 2010, Ultrasonic activation, a possible alternative
application in the microinjectionmolding, International Journal of Material Forming, 3, suppl. 1,
547-550

9. IclanzanT., StanD., AbadieM., 2008,A possible new ultrasonic thermo-pellicular effect and
applications, International Journal of Material Forming, 743-746, doi: 10.1007/s12289-008-0282-y



Evaluation of specific process parameters and ultrasonically activated... 809

10. JungW.-C., HeoY.-M., ShinK.-H., YoonG.-S., Chang S.-H,, 2007,An experimatal study
onmicro injection paramaters,ANTEC: Proceedings of Annual Technical Conference, Cincinnati,
OH, 638-642

11. Malloy R., 1994,Plastic Part Design for Injection Molding, Carl Hanser Verlag, Munchen-New
York

12. MönkkönenK.,Hietala J., PääkkönenP.,PääkkönenE.J.,KaikurantaT., Pakkanen
T.T., Jääskeläinen T., 2002, Replication of sub-micron features using amorphous thermopla-
stics,Polymer Engineering and Science, 42, 1600-1608

13. Patent No. RO118576B, 2003,Ultrasonic activated mould (in Romanian)

14. Pirskanen J., Immonen J., Kalima V., Pietarinen J., Siitonen S., Kuittinen M.,
Mönkkönen K., Pakkanen T., Suvanto M., Pääkkönen E.J., 2005, Replication of sub-
micrometre features using microsystems technology,Plastics Rubber and Composites, 34, 222-226

15. StanD., 1999,Activationultrasoniquedes thermo-plastiques lorsde lamise enoeuvrepar injection
et extrusion, PhDThesis, Univ. Montpellier, France

16. ŜirbuA.N., 2009,Theoretical and experimental research on the ultrasonic activation effect on the
molten polymers rheology, PhDThesis, Politehnica University of Timisoara

17. ŜirbuN.A., IclănzanT., Oancă O., 2008, Conception, design and realization of the ultrasonic
assembly for flow activation devices of the polymeric media, Symposium of the Institute of Solid
Mechanics, SISOM, Symposium of Acoustics, Bucharest

18. Stan D., 1999, The ultrasonic activation of the thermo-plastics in forming processes by injection
andextrusion (inFrench),PhDThesis, 1999,USTLMontpellier 2,France,www.sudoc.fr/06769939,
accessed: 20 febr. 2015

19. StanD., IclanzanT.,AbadieM., 2000,The surfaceeffect in the caseof theultrasonicactivation
of the visco-ellastically pressurizedmedia (in French), Synthetic Polymer Journal,VII, 84-93

20. Wimberger-Friedl R., 2000, Injection molding of sub-µm grating optical elements, Journal of
Injection Molding Technology, 4, 78-83

21. Zhao J., Mayes R., Chen G., Xie H., Chan P., 2003, Effects of process parameters on the
micro molding process,Polymer Engineering and Science, 43, 1542-1554

Manuscript received July 27, 2014; accepted for print March 30, 2015