Journal of Materials Processing Technology 186 (2007) 138–146

Modeling of temperature distribution in ultrasonic welding

of thermoplastics for various joint designs
K.S. Suresh, M. Roopa Rani, K. Prakasan ∗ , R. Rudramoorthy 1
Department of Production Engineering, PSG College of Technology, Peelamedu, Coimbatore 641004, India
Received 30 May 2006; received in revised form 10 September 2006; accepted 14 December 2006

Abstract
Use of engineering plastics in structural and non-structural applications is rapidly increasing. As the demand for plastics increases so does
the requirements for joining. Of the many techniques that are available for joining of thermoplastics, ultrasonic welding is one of the preferred
processes. Thermoplastic polymers are categorized according to their molecular structure as amorphous and semi-crystalline. Ultrasonic welding
of these two types of thermoplastics is expected to be quite different. As heating is confined to the interface area, quality of weld mainly depends
on temperature at the interface. So study on temperature distribution during welding is very important to predict the quality of weld. Viscoelastic
heating is most critical to ultrasonic welding of thermoplastics because it is the main mechanism by which heat is developed at the interface. Heat
developed due to viscoelastic heating depends on applied frequency, square of amplitude and loss modulus. In this study, modeling of temperature
distribution for various joint designs of thermoplastics as practised by industry is attempted and simulation is done in ANSYS. Model is validated
by measurement of temperature during welding.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Ultrasonic welding; Joint designs; Mathematical modeling; Viscoelastic heating

1. Introduction where the horn is located at a distance which is more than 6 mm

from the joint.
A typical ultrasonic welding process starts when a power
supply changes 50 Hz of electrical energy into high ultrasonic
2. Literature review
frequency of 20–40 kHz. To transform this electrical energy
into mechanical energy, a converter—a lead zirconate titanate A number of papers on joint design by practising industry
electrostrictive element expands and contracts at the reso- were reviewed to understand the joints practised by them. The
nant frequency. The converter is coupled to the mechanical authors also studied the components welded using USW from
impedance transformer called a horn. The horn transmits the various industries. The paper by Devine [1] discussed a num-
energy to the joint area where frictional heat is produced to melt ber of joints practised by the industry. These include butt joint,
the plastic momentarily, causing it to fuse together. A pneumat- step joint, tongue and groove joints. It has been suggested that
ically activated press applies pressure to the part to be welded. the energy director is suitable for most of the amorphous plas-
The energy director is the protruding part on one side of the joint tics. Semi-crystalline material may receive incomplete fusion
interface. It directs and focuses the ultrasonic energy to smaller when an energy director is used because material displaced from
contact area and has the effect of increasing welding speed and energy director may solidify before it flows across the joint to
quality. These energy directors are in the form of triangular, rect- form a seal. For this reason author suggested the use of a shear
angular, or semicircular shapes. Ultrasonic welding is generally joint for welding semi-crystalline polymers.
divided into (a) near field welding in which the horn is located Frantz [2] suggested three basic guidelines to get quality
within a distance of 6 mm from the joint and (b) far field welding joints in plastics by ultrasonic welding. These are (a) the use
of a small initial contact area between the mating surfaces to
∗ Corresponding author. Tel.: +91 422 2572177.
ensure efficient energy concentration (b) the alignment of the
E-mail address: prakasan [email protected] (K. Prakasan). mating surfaces and (c) uniformity of mating surfaces for con-
1 Professor & The Principal PSG College of Technology. sistent energy transfer and controllable melting and welding.

0924-0136/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmatprotec.2006.12.028
 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146 139

Kenney [3] on engineering thermoplastics with crystalline struc- The quality of weld depends on the heat affected zone
tures suggested a shear joint because it is self sealing, reduces (HAZ) and temperature distribution for a given set of param-
heat loss and prevents oxidation in susceptible resins. The author eters. The parameters include amplitude, frequency, material
also suggested an alternative to shear joint. This method involves characteristics, weld time and joint configuration. Amorphous
molded-in-texturing of the welding surface opposite to an energy thermoplastics are energy efficient and usually require low
director. The textured surface consists of peaks and valleys of ultrasonic energy levels. They are characterized by a random
3–6 mm in size. During welding the textured surface produces molecular arrangement and broad softening range. This allows
more compact melt and better heat transfer across the joint. A the material to flow easily without premature solidification.
study by Chuah et al. [4] discussed the effects of the shape of Semi-crystalline resins need higher ultrasonic levels because
the energy directors on far field ultrasonic welding. It was found they have an orderly molecular structure. In the solid state the
that the energy directors with semicircular shape yielded high- molecules absorb mechanical vibrations and very high amount
est welding efficiency followed by the rectangular and triangular of energy is necessary to breakdown the crystalline structure so
shapes. A semicircular shaped energy director had a greater con- that material melts and flows.
tact area than the other two energy directors and hence greater
welding efficiency. 3. Objective
A study was made by Benatar et al. [5] on high frequency
ultrasonic wave propagation and attenuation. Measurements From the literature review it is concluded that there is enough
were made in order to estimate the dynamic mechanical mod- scope to carry out research in mathematical modeling and eval-
uli of the polymers. The estimated moduli were entered into a uation of temperature distribution for different joint designs in
lumped parameter model to predict heating rates and energy ultrasonic welding. Thus this work is taken up. The objective
dissipation. Benatar and Cheng [6] developed a model for of this work is to model the temperature distribution during
wave propagation in viscoelastic materials to predict vibration ultrasonic welding for an amorphous (ABS) and semi-crystalline
amplitude experienced at the joint interface. This study indi- (HDPE) thermoplastics for several possible joint designs using
cates that increasing the length of the samples to a half wave software ANSYS and validate the model by measurement of
length can improve the far field welding of semi-crystalline temperature. This will help understand the spread of HAZ in
polymers by maximizing the amplitude of vibration at the ultrasonic welding and quality of weld.
joint interface. Ultrasonic welding of oriented poly propylene
(OPP) using tie-layer materials has been examined by Tateishi 4. Mathematical analysis
et al. [7]. The thermal cycle at the joint interface was eval-
uated using high speed data acquisition system. The authors The governing differential equation to be solved is
found that the total time required for completion of the weld- ∂2 T ∂2 T ∂T
ing process decreases when applied pressure and amplitude are k + k 2 + Q − ρc =0
∂x 2 ∂y ∂t
increased.
Aliosio et al. [8] investigated the ultrasonic welding of poly- where ρ is density (kg/m3 ), c specific heat capacity (J/kg K), Q
carbonate, ABS and Noryl—using rectangular energy directors. volumetric heat generation rate (W/m3 ) and k is isotropic thermal
They modeled the viscoelastic heating of the energy directors conductivity (W/m K).
assuming adiabatic heating, and used elastic analysis to estimate The standard sample shown in Fig. 1 (developed by Nedu-
the strain amplitude within parts. landse Philips) [5] was used for analysis and experimental work.

Fig. 1. Model used for analysis and experimental work.

140 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146

Applying suitable initial and boundary conditions this equa- 6 mm from the interface. So the intensity of energy at the inter-
tion is solved using ANSYS software. face is reduced because of attenuation (result of true absorption
and scattering) of waves.
• Initial condition
6. ANSYS analysis
T (x, y, 0) = T0 = room temperature
A standardized sample which was developed by Nedulandse
Philips was used for analysis. Standard specimen with different
• Boundary conditions
joint designs (triangular, semicircular, and shear) is modeled.
Q(0, y, t) = Qconvection , 0 ≤ y ≤ 30, 0 ≤ t ≤ tweld time , For modeling, axy-symmetry concept is used because the speci-
men is symmetric about the Z axis. For this full transient analysis
Q(29, y, t) = Qconvection , 0 ≤ y ≤ 30, 0 ≤ t ≤ tweld time the properties considered are coefficient of thermal conductiv-
ity (k) specific heat (c) and density (ρ). The properties of these
where Qconvection is the heat lost due to convection to the sur-
plastics are listed in Table 1. The type of element used for analy-
roundings from lateral surfaces of area A with overall heat
sis is PLANE 55 (solid-Quad 4 node). In meshing, a finer mesh
transfer coefficient (h) of 0.1 J/m2 ◦ C.
(smart size of 2) size than default value was used. The governing
equation was solved by applying the initial and boundary con-
Q(x, 0) = Qvisco , 0 ≤ x ≤ 29, where Qvisco is the viscoelastic ditions as mentioned earlier. The preprocessor for semicircular
heating (heat flux) in W/m2 . joint design (ABS-near field) is shown in Fig. 2 as an example.
The following assumptions were made for calculating vis- Using various options available in postprocessor the required
coelastic energy dissipated. outputs are obtained.

(1) The triangular energy director and semicircular energy 7. Experimental procedure
director can be approximated as a rectangular energy direc-
tor of equal cross-sectional area. All the welding was carried out using a conventional ultrasonic plastic
(2) Amplitude of vibration is taken by the top part and the energy machine (1500 W, 20 kHz) manufactured by M/s National Indosonic. Horn made
of aluminum–titanium alloy was used for this study. The horn used for welding
director. had diameters of 28 and 30 mm for near field and far field welding, respectively.
(3) The loss modulus is not a function of temperature. The materials evaluated are

5. Viscoelastic heating • ABS (amorphous);

• HDPE(semi-crystalline).

A viscoelastic material that is subjected to a sinusoidal strain A standardized sample (which was developed by Nedulandse Philips) was
dissipates some energy through intermolecular friction. The stor- used in all the experiments in near field and far field welding. The specimen
age modulus for a viscoelastic material is the in-phase modulus with different energy directors were made by injection molding. A ZWICK
and it is a measure of the ability to store energy. The loss modulus 1484 tensile tester was used to measure the strengths of the welded joints. Test
procedures according to ASTM standard D638-97 (Standard Test Method for
is out of phase modulus and it is a measure of energy dissipated.
Tensile properties of plastics) were used. The deformation rate used in the testing
The average energy dissipated per unit time is given by was 50 mm/min.

ωε20 E
Q= J/m3 s (1) 8. Data acquisition system
2
where Q is the average power dissipated, ω = 2πf, f the applied The temperatures at different points of the specimen were
frequency, ε0 the maximum strain and E is the loss modulus. monitored in real time using a high speed data acquisition
The peak strain ε0 during welding was calculated based on an system. The data acquisition system includes sensors (ther-
elastic analysis. mocouple), terminal block, DAQ card and analyzing software.
Viscoelastic heating is most critical to ultrasonic welding of The block diagram of data acquisition system is shown in
thermoplastics because without sufficient heating the interface Fig. 3(a). All the temperatures were measured using SWG 36
cannot melt and fuse to form the joint. In the case of near field Alumal–Cromal (type K) thermocouples positioned at different
welding horn is very close to the interface. But in the case of far points on the specimen as shown in figure no. 1. The thermo-
field welding horn is located at a distance which is more than couple signal was fed into a terminal block (TBX 68 T) and

Table 1
Properties of the thermoplastics used
Material Coefficient of thermal Density (ρ) Specific heat (c) Melting Loss modulus Maximum tensile
conductivity (k) (W/m ◦ C) (kg/m3 ) (J/kg ◦ C) point (◦ C) (E ) (GPa) strength (MPa)

ABS 0.15 1181 2000 90 0.42 60

HDPE 0.43 997 1500 135 0.52 29
 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146 141

Fig. 2. Preprocessor for semicircular energy director (ABS-near field).

Fig. 3. (a) Process data acquisition system and (b) experimental setup for temperature measurement.
142 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146

then to DAQ card (NI 4351). The experimental setup is shown

in Fig. 3(b). A programme made for measurement of tempera-
ture (multi channel) in Lab View, was used to get the real time
temperature variation during welding.

9. Results and discussion

9.1. Effect of joint design and field

As discussed in the earlier section, experi-

ments and simulation studies were carried out
and the results are presented in this section.
Figs. 4(a), 5(a), 6(a), 7(a), 8(a), 9(a), 10(a), 11(a) and 12(a)
show the simulated results and Figs. 4(b), 5(b), 6(b), 7(b),
8(b), 9(b), 10(b), 11(b) and 12(b) show the experimental results
for the same conditions. In all these representations, predicted
and experimental values are found to be in agreement.
Figs. 4(a), 5(a), 6(a), 7(a), 8(a) and 9(a) and 4(b), 5(b), 6(b),
7(b), 8(b) and 9(b) represent the results for ABS
specimen for various welding conditions. Out of this
Figs. 4(a), 5(a) and 6(a) and 4(b), 5(b) and 6(b) are for near field
welding and remaining are for far field welding configurations.
Fig. 4(a) represents simulated results with semicircular
energy director. At the end of weld time (0.3 s) the interface Fig. 5. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
temperature developed is 52.362 ◦ C. Under the same weld- temperature distribution (experimental results). Material, ABS; type of joint,
ing conditions experimentally observed value is 52 ◦ C which triangular energy director; field, near field; frequency, 20 kHz; amplitude, 60 ␮m.
is shown in Fig. 4(b) (the percentage error of 0.7). Fig. 5(a) Welding conditions: weld pressure, 3 bar; weld time, 0.3 s; hold time, 1.0 s.
and (b) represents the variation of temperature with triangular
energy director. The interface temperature obtained in simula-

Fig. 4. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
temperature distribution (experimental results). Material, ABS; type of joint, Fig. 6. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b) tem-
semicircular energy director; field, near field; frequency, 20 kHz; amplitude, perature distribution (experimental results). Material, ABS; type of joint, shear
60 ␮m. Welding conditions: weld pressure, 3 bar; weld time, 0.3 s; hold time, joint; field, near field; frequency, 20 kHz; amplitude, 60 ␮m. Welding conditions:
1.0 s. weld pressure, 3 bar; weld time, 0.3 s; hold time, 1.0 s.
 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146 143

Fig. 9. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b) tem-
Fig. 7. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
perature distribution (experimental results). Material, ABS; type of joint, shear
temperature distribution (experimental results). Material, ABS; type of joint,
joint; field, far field; frequency, 20 kHz; amplitude, 60 ␮m. Welding conditions:
semicircular energy director; field, far field; frequency, 20 kHz; amplitude,
weld pressure, 3 bar; weld time, 0.8 s; hold time, 1.0 s.
60 ␮m. Welding conditions: weld pressure, 3 bar; weld time, 0.8 s; hold time,
1.0 s.
tion is 60.942 ◦ C and in experiment it is 56 ◦ C. The percentage
error in this case is 8.75. The temperature at the interface for
triangular energy director is more than with semicircular energy
director. The triangular energy director concentrates more vis-
coelastic heat developed towards the interface. So initializing of
melting will start quickly for small weld time.
The temperature variations for shear joints are shown in
Fig. 6(a) and (b). Maximum interface temperature in simulation
and experimental analysis are 68.695 and 68 ◦ C, respectively.
Shear joint is a closed joint so that heat loss due to convection
is less. This may be the reason for higher interface temperature
than that in energy director joints.
The temperature distributions for ABS in far field configura-
tion are shown in Figs. 7(a), 8(a) and 9(a) and 7(b), 8(b) and 9(b).
In far field welding, the intensity of ultrasonic waves would
decrease inversely with the distance from the energy source. This
results from two basic causes-scattering and true absorption,
which are combined in the term attenuation. The change in inten-
sity of ultrasonic wave with distance as a result of attenuation
can be represented as shown in Eq. (2):

I0
2αz = 20 log dB (2)
I
where I0 and I are intensity at the beginning and the end, respec-
Fig. 8. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
tively, for a section of length z with attenuation coefficient α.
temperature distribution (experimental results). Material, ABS; type of joint,
triangular energy director; field, far field; frequency, 20 kHz; amplitude, 60 ␮m. The attenuation coefficient α (in dB/m) for plastics at 20 kHz is
Welding conditions: weld pressure, 3 bar; weld time, 0.8 s; hold time, 1.0 s. in the range of 10–100 [9]. For this study α is taken as 50 dB/m.
144 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146

Fig. 10. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
temperature distribution (experimental results). Material, HDPE; type of joint,
semicircular energy director; field, near field; frequency, 20 kHz; amplitude,
60 ␮m. Welding conditions: weld pressure, 3 bar; weld time, 0.8 s; hold time, Fig. 11. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
1.0 s. temperature distribution (experimental results). Material, HDPE; type of joint,
triangular energy director; field, near field; frequency, 20 kHz; amplitude, 60 ␮m.
Welding conditions: weld pressure, 3 bar; weld time, 0.8 s; hold time, 1.0 s.

Fig. 7(a) and (b) represents variation of temperature with

semicircular energy director in far field configuration. The inter- ture of 104.856 ◦ C as shown by simulation results and 78 ◦ C by
face temperature observed at the end of 0.8 s (weld time) is actual measurements. These results are shown in Fig. 12(a) and
72.446 ◦ C for simulation studies and 70 ◦ C is observed from (b). The energy director for ABS specimen absorbed a larger
experiments (with percentage error of 3.4). Fig. 8(a) and (b) portion of the welding energy than that of HDPE. The energy
are representing temperature distribution with triangular energy absorption ratio is defined as the ratio of the energy absorbed
director. A maximum interface temperature of 89.379 ◦ C is by the plastic material at the energy director to that of energy
observed in simulation and 80 ◦ C in experimental studies (with dissipated by the ultrasonic welder. For this analysis absorp-
error of 11.75%). In far field welding also, value of interface tion ratio for ABS is taken as 50% and for HDPE it is 30%
temperature is more for triangular energy director than with [4].
semicircular energy director. In the case of shear joint max- Temperature distribution obtained from the analysis can be
imum interface temperature developed in ANSYS analysis is used to find out the nature of heat affected zone (HAZ). It was
105.26 ◦ C and in experimental studies it is 74 ◦ C as shown in found that only a small portion of the welding energy was trans-
Fig. 9(a) and (b). ferred by heat conduction in the welded part because plastics are
Figs. 10(a), 11(a) and 12(a) and 10(b), 11(b) and 12(b) rep- very bad conductors of heat.
resent the results for HDPE specimen in near field configuration It may be noted that experimental values are less than that
with a weld time of 0.8 s. Fig. 10(a) and (b) are representing obtained from simulation studies (Tables 2–4) probably because
temperature distribution for HDPE with semicircular energy the viscoelastic dissipation diminishes as the material heats
director obtained from ANSYS analysis and experimental stud- above its Tg (glass transition temperature). Also the surface tem-
ies, respectively. Maximum interface temperature in simulation perature is increasingly affected by heat loss from the surface
is 70.758 ◦ C and in experimental studies it is 70 ◦ C (with error by convection. The assumption that maximum strain ampli-
of 1%). For triangular energy director, an interface temperature tude is equal to horn displacement during the calculation of
of 82.357 ◦ C in simulation and 72.5 ◦ C in experimental studies maximum strain is an ideal one. Position of maximum ampli-
are obtained as in Fig. 11(a) and (b) (percentage error is about tude of vibration depends on wave length and thickness of
13.6). Shear joint experiences a maximum interface tempera- specimen.
 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146 145

Table 2 perature of 56 ◦ C for ABS in the near field configurations in

Comparison of simulated and experimental results (interface temperature) a short weld time of 0.3 s as shown in Fig. 13(a). Concentra-
(ABS-near field)
tion of energy is more in triangular energy director because
Type of joint Simulated Experimental of its small cross-sectional area. So initializing of melting will
results (◦ C) results (◦ C) start quickly for small weld times. In far field welding, shear
Triangular energy director 60.942 56 joint exhibits maximum UTS and interface temperature value
Semicircular energy director 52.362 52 followed by a joint with triangular energy director. In shear
Shear joint 68.695 68 joint, heat loss due to convection is less because it is a closed
one.
Table 3
Fig. 13(b) represents the variation of UTS with various
Comparison of simulated and experimental results (interface temperature) joints for HDPE in near field welding. For HDPE triangular
(ABS-far field) energy director joint was found to have high UTS but shear
Type of joint Simulated Experimental
joint was found to be with higher interface temperature than
results (◦ C) results (◦ C) other joint designs. In the far field region only shear joint
could be welded with standard specimen which had very low
Triangular energy director 89.379 80
Semicircular energy director 72.446 70
UTS.
Shear joint 105.26 74 Weld strength increases with interface temperature only up
to certain temperature and then decreases. This is the temper-
ature at which the properties will change sufficiently so that
the plastic material ceases to function in the actual use. This is
9.2. Effect of material called the maximum use temperature. A rule of thumb for prac-
tical applications is that plastic material should not be welded
ABS—an amorphous material can be easily welded using at temperature above 75% of its glass transition temperature for
ultrasonic welding in far field and near field configurations. amorphous polymers and 75% of its melting point for semi-
But PE—a semi-crystalline thermoplastic is very difficult to crystalline polymers [10].
weld in far field. Amorphous plastics are better conductors of
ultrasonic waves and therefore ultrasonic energy can be trans-
mitted easily. The semi-crystalline PE is a viscous material
and it tends to vibrate at reduced levels when the ultrasonic
wave energy is dissipated from the horn. A phase difference
between the vibration of part to be welded and the horn may
result in energy dissipation at the interface of the welded part
and the horn. The thermocouple placed at the interface of the
horn shows higher temperature in the case of PE than ABS.
Visual observation revealed the formation of a thin melted film
of the welded part on the horn after the welding process. So
it can be concluded that less energy may be transmitted to the
energy director for HDPE part. This may be hindering the trans-
mission of energy leading to the observed difference between
temperatures obtained from simulation studies and temperature
measurements.

9.3. Temperature distribution and ultimate tensile strength

(UTS)

Triangular energy director was found to have high UTS

values compared to other joints with optimum interface tem-

Table 4
Comparison of simulated and experimental results (interface temperature)
(HDPE-near field)
Type of joint Simulated Experimental
results (◦ C) results (◦ C)

Triangular energy director 82.357 72.5 Fig. 12. (a) Temperature distribution (in ◦ C) (from ANSYS analysis) and (b)
Semicircular energy director 70.758 70 temperature distribution (experimental results). Material, HDFE; type of joint,
Shear joint 104.856 78 shear joint; field, near field; frequency, 20 kHz; amplitude, 60 ␮m. Welding
conditions: weld pressure, 3 bar; weld time, 0.8 s; hold time, 1.0 s.
146 K.S. Suresh et al. / Journal of Materials Processing Technology 186 (2007) 138–146

compared to other energy director joints in the near field

configuration for ABS. For shear joint, interface tempera-
ture is more than that in the triangular energy director joint
with reduced UTS values. In far field configuration for ABS,
shear joint having maximum interface temperature and UTS
is followed very closely by triangular energy director. In
HDPE triangular energy director joint was found to have
high UTS in near field configuration.
(5) UTS increases with interface temperature up to maximum
use temperature and then decreases for amorphous and semi-
crystalline thermoplastics. This is the temperature at which
the properties will change sufficiently so that the plastic
material ceases to function in the actual use.
(6) The semi-crystalline thermoplastics are viscous and ultra-
sonic waves get damped. Therefore, for semi-crystalline
thermoplastics, a phase difference between the vibration of
part to be welded and horn may result in energy dissipation
at the part–horn interface.

Acknowledgements

Authors express sincere gratitude to the Department of Sci-

Fig. 13.(a) Variation of UTS (N/mm2 ) values of ABS with different joint designs ence and Technology, New Delhi, for funding this project (no.
and (b) variation of UTS (N/mm2 ) values of HDPE with different joint designs. SR/WOS-A/ET-20/2003). Thanks are due to the Management
tr, triangular energy director joint; sc, semicircular energy director joint; sh, and the Principal, PSG College of Technology, Coimbatore
shear joint. for providing the necessary infrastructure to setup the labora-
tory with Department of Production Engineering. The support
10. Conclusions extended by the Department of Civil Engineering, PSG College
of Technology for the tensile testing of specimens is gratefully
From the studies carried out as presented in the previous acknowledged.
sections, the following points can be highlighted.
References
(1) Both amorphous and semi-crystalline polymers were suc-
cessfully welded in the near field configuration for all [1] J. Devine, Ultrasonic plastic welding basics, Weld. J. (AWS) (2001)
the joint designs. Amorphous polymer exhibited stronger 29s–33s.
bonds. In far field, only amorphous polymer got welded for [2] J. Frantz, Joining plastics in sound way, Mach. Design 69 (7) (1997) 61–65.
[3] W.E. Kenney, Designing plastic parts for ultrasonic assembly, Mach.
all the joint designs. Only shear joint was possible in the
Design 64 (10) (1992) 65–68.
case of HDPE which is a semi-crystalline polymer. [4] Y.K. Chuah, L.-H. Chien, B.C. Chang, S.-J. Liu, Effects of the shape of the
(2) Viscoelastic heating is most critical to ultrasonic welding energy director on far field ultrasonic welding of thermoplastics, Polym.
of thermoplastics, which depends on applied frequency, Eng. Sci. 40 (1) (2000) 157–167.
square of the amplitude and loss modulus of material. This [5] A. Benatar, R.V. Eswaran, S.K. Nayar, Ultrasonic welding of thermoplas-
tics in the near field, Polym. Eng. Sci. 29 (23) (1989) 1689–1698.
is observed from the expression for viscoelastic heating and
[6] A. Benatar, Z. Cheng, Ultrasonic welding of thermoplastics in the far field,
found to be in agreement with observed results. Polym. Eng. Sci. 29 (23) (1989) 1699–1704.
(3) Results obtained from ANSYS analysis are very close to [7] N. Tateishi, T.H. North, R.T. Woodhama, Ultrasonic welding using tie-layer
the measured values of temperature. Experimental values materials. Part 1. Analysis of process operation, Polym. Eng. Sci. 32 (9)
of interface temperature for various joint designs are less (1992) 600–610.
[8] C.J. Aliosio, D.G. Wahl, E.E. Whetsel, A simplified thermoviscoelastic
than simulated values probably because the viscoelastic dis-
analysis of the ultrasonic bonding, SPE Technical Papers, Brookfield, CT,
sipation diminishes as the material heats above its glass 1972.
transition temperature and the formation of film of polymer [9] J. Krautkrammer, H. Krautkrammer, Ultrasonic Testing of Materials, 4th
at the tip of the horn after repeated welds. fully revised ed., Narosa Publishing House, New Delhi, 1993, pp. 108–113.
(4) Triangular energy director was found to contribute towards [10] A. Brent Strong, Plastics-Materials and Processing, Prentice Hall, Inc.,
Upper Saddle River, NJ, 1996, pp. 82–85.
achieving higher interface temperatures and UTS values