Improvement of
Drive Energy Efficiency
in a Shear Mode
Piezo Inkjet Head
Yoshio Takeuchi,
Hiroshi Takeuchi, Katsuaki Komatsu, Shinichi Nishi
Summary
Recent
improvements of inkjet printer technology
enabling the high
print quality and high-speed
printing have
been remarkable, however, for higher-
speed printing,
the development of multi-channel
inkjet head whose
energy efficiency is higher, is
necessary. We
have analyzed an ink flow in a
shear mode piezo
inkjet head and an ink droplet
forming process
through computational simulation.
By using the
results of the simulation, we designed
optimum shapes of
an actuator, an ink channel and
a nozzle, and
made a prototype of inkjet head
employing a
funnel type nozzle. Then we
experimentally
confirmed an improvement of the
drive efficiency.
Abstract
The recent
acceleration in high print quality and
high-speed inkjet
printers commands the
development of an
energy efficient multi-channel
print head to accommodate
these ever-advancing
printers. In
response, we have computationally
simulated a shear
mode inkjet head in order to
analyze its fluid
flow dynamics and jet forming
process. As a
result, we have been able to
optimize the
shape of the actuator, channel, and
nozzle of the
inkjet head. In particular, a funnel
type nozzle has
proven to provide good energy
efficiency in a
prototype print head based on the
results of our
simulation and analysis.
1 Introduction
For increasing
the printing speed of the inkjet
printer, the
improvement of the ink-ejecting rate of
the print head is
a matter of course, and the number
of channels needs
to be increased.
For increasing
the number of channels, the problem
in the structure
including the manufacturing is
important, and it
is also important to improve the
drive efficiency
to minimize energy necessary for
droplet ejection,
for controlling deterioration of print
quality and
droplet ejection stability caused by
temperature rise
of the print head during printing.
In order to
overcome this problem, we analyzed the
drive efficiency
characteristics of a shear mode
piezo inkjet head
through computer simulation.
The shear mode
piezo inkjet head is a head driven
by shearing
stress generated by applying an
electrical field
in the direction perpendicular to the
polarization
direction of the piezoelectric material.
The
characteristic of an actuator composed of such
piezoelectric
material was analyzed by using a finite
element method
simulation software that can make
a structure and
an electrical field to be coupled, and
for the ink flow
within the print head and for the
process of
droplet ejection from the nozzle, a finite
difference method
simulation software which can
analyze free
surface flow, was used.
The electric
energy applied to the actuator, the
elastic energy
applied to the ink in channels, and
the kinetic
energy of the droplet were estimated
through the
simulation, and the relationship
between the
factors (such as shapes of actuator
and channel,
nozzle shape, piezoelectric material,
adhesive layer,
ink characteristics, drive voltage
waveform) and
drive efficiency, was analyzed.
Herein, the
relationship between the shape of inkjet
head and drive
efficiency will be mainly discussed.
2 Structure of
the inkjet head and driving energy
Fig. 1 Structure
of shear mode piezo inkjet head
Fig. 1 is a
structural view of a shear mode piezo
inkjet head. When
grooves are mechanically
formed in a PZT
(lead zirconate titanate) substrate,
channels and
actuators which are walls of channels
are formed. A
cover plate is bonded on the upper
surface of the
walls, and a nozzle plate is bonded
on the front
surface of the substrate, and the ink is
fed into the
channels.
Fig. 2 PZT
actuator (cross section)
A sectional view
of the actuator that is cut at a right
angle to the flow
direction of channel is shown in
Fig. 2. When an
electrical field is applied in the
direction
orthogonal to the polarization direction of
the PZT,
actuators are deformed, and the ink in the
channel is
pressurized. When the pressure wave
generated in the
channel is reflected between
nozzles and the common
ink chamber, and
resonated, the
pressure applied to the nozzle
change in time,
and an ink droplet is ejected.
Fig. 3 Droplet
ejection process
A result of the
simulated droplet ejection process is
shown in Fig. 3.
For low voltage drive, the drive
waveform shown on
the upper part in Fig. 3 is used.
When the voltage
is changed, pressure generates
within the
channel, after that, it oscillates at a
resonance
frequency and gradually attenuates.
(middle in Fig.
3). At the time of the rise of voltage
applied in the
direction in which the volume of
channel is
increased, negative pressure is
generated. When
the negative pressure reaches
the peak of the
positive pressure after a half period,
the voltage is
applied in the direction in which the
channel volume is
decreased, that is, in the reverse
polarity to the
first rise of the voltage. Then the
positive pressure
for droplet ejection is reinforced.
The result of the
simulation of the time change of
the pressure in
the channel and the process of
forming droplet
from the nozzle, is shown in the
lower part of
Fig. 3(1).
Fig. 4 Resonance
frequency vs. droplet ejection
For the higher
speed and higher print quality, it is
necessary to
enhance the pressure resonance
frequency in the
channel. The reason for this is
that the droplet
volume is inversely proportional to
the resonance
frequency, as shown by the following
expression.
Vd = pr2 × v/(2 ×
f)
Vd: Volume of droplet r: Radius of the nozzle
v: Velocity of droplet f: Resonance frequency
The computation
result of the relationship between
the resonance
frequency of the pressure applied to
the nozzle and
values of the necessary pressure for
constant velocity
of a ejected droplet is shown in
Fig. 4. When the
frequency is increased, the
necessary
ejection pressure increases rapidly, that
is, the drive
voltage is increased.
Further, when the
number of channels or the drive
frequency is
increased to improve printing speed
and print
quality, the generated heat (including the
heat generated in
the drive circuit) also increase
rapidly.
Wa = (1/2)× C ×
V2 × A × fd × N
Wa: Total generated heat,
C: Electrostatic capacity of the actuator,
V: Drive voltage
fd: Drive frequency
A: Waveform coefficient
N: Number of channels
Some part of this
generated heat, namely the heat
generated by the
dielectric loss of the piezoelectric
material forming
the actuator, and by the resistance
of electrode
transfer to the ink in the channel, and
causes the ink
temperature rise. Because of the
short distance
between the actuator and the ink, ink
temperature rises
in a very short time, and changes
in ink
characteristics cause fluctuations of the
droplet velocity
and droplet volume, resulting in a
decline in print
quality. Further, when the
temperature rise
is remarkable, there is a risk that
stable ejection
can not be achieved.
3 Shape of inkjet
head and drive efficiency
3.1 Actuator and
ink channel
A calculated
example of the actuator displacement
by voltage
application is shown on the left side in
Fig. 5. The
compliance (displacement /force) of
the actuator is
calculated as a counter pressure
displacement by
the internal pressure rise on the
right side in
Fig. 5. The ratio of the compliance of
an actuator to
the compliance of the ink in the
channel is called
the compliance ratio (kcr). The
compliance ratio
shows the ratio of the volume
change of the
actuator by pressure difference
between the
channels to the volume change of
pressurized ink
in channel.
Fig. 5 Actuator
deformation analysis
Pressure P
generated in the channel by voltage
application can
be calculated using the following
expression.
Herein, . is a constant determined by
the channel drive
pattern. The generated pressure
is decreased
because the actuator is forced back
by the rise of
internal pressure in the channel.
P = 2 × (.x/W) ×
B × V/(1 + . × kcr)
.x: an averaged displacement of the actuator
by unit voltage application
V: Drive voltage
W: Channel width
B: Bulk modulus of the ink
The speed of the
pressure wave propagating in the
channel is also
changed depending on the value of
the compliance
ratio. The reason is that the
change of volume
of the ink by the channel internal
pressure is
practically increased by the deformation
of the actuator,
that is, bulk modulus of the ink is
apparently
decreased. Therefore, the change of
shape also
influences on the resonance frequency,
then the
attention must be paid.
Propagation speed
of pressure wave
C0 = (B/.)1/2/(1
+ . × kcr)1/2
.: Density of ink
Resonance
frequency of pressure wave
f = C0(1 + a)/4L
a: Shape factor
L: Channel length
Since the
generated pressure is proportional to the
displacement of
the actuator, it is essential to
design so that
the displacement per unit of applied
voltage is
increased. The relationship between the
displacement and
the elastic energy applied to the
ink is determined
by the following expression.
E = (1/2) × B ×
(x/W)2 × L × H × W
E: Elastic energy of to the ink
x: Average displacement of the actuator
L: Channel length
W: Channel width
H: Channel depth
Further, the
relationship between pressure P
generated in
channel and the energy is expressed
by the following
expression:
E = (1/2) × P2 ×
L × H × W/B
Fig. 6 Voltage
sensitivity vs. channel width
Fig. 6 shows an
example to calculate how the ratio
of voltage
sensitivity and compliance ratio are
changed when the
channel width is changed under
the constant
channel pitch (channel width +
actuator
thickness). When the channel is shallow,
even when the
channel width increases, the
compliance ratio
(dashed line) is only slightly
increased, then
the voltage sensitivity (solid line)
does not drop.
When the channel depth is
increased, the
compliance ratio increases rapidly,
then the voltage
sensitivity drops sharply as the
channel width
increases. Fig. 7 shows the
relationship
between channel width and elastic
energy in which the
channel depth is a parameter.
When channel
depth is decreased, even when
voltage
sensitivity is high, the elastic energy is
lowered because
the section area of the channel is
decreased.
Fig. 7 Ink
elastic energy vs. channel width
It is important
to design the shapes of the actuator
and channel so
that efficiency of conversion from
input electrical
energy to ink elastic energy
increases.
However, it must be realized that the
electrostatic
capacity of the actuator changes
depending on its
shape. The electrostatic capacity
is proportional
to the channel length and to its
depth, and is
inversely proportional to the thickness
of the actuator.
Further, the optimum cross
sectional shape
changes depending on the
characteristic of
the piezoelectric material
(piezoelectric
constant, relative dielectric constant,
elastic
constant), the characteristic of the ink (bulk
modulus) or an
adhesive layer (1). The resonance
frequency of the
channel is also influenced by the
cross sectional
shape, however, it is almost
inversely
proportional to channel length. Fig. 8
shows how the
resonance frequency is changed for
the channel
length.
Fig. 8 Channel
length vs. resonance frequency
3.2 Nozzle
When a nozzle
diameter is decreased, the droplet
volume decreases,
however, the viscous resistance
in the nozzle is
greatly increased, and the energy
loss grows
greater. The Fig. 9 shows the
relationship
between the nozzle diameter and the
droplet velocity,
in which the ink viscosity is a
parameter. In the
case where ink viscosity is high, if
the nozzle
diameter is decreased, a lowering of ink
velocity is
remarkable. The reason is that velocity
down effect of
the viscosity is greater than the
velocity up
effect of the accelerated flow rate by the
ratio of cross
sectional area of the channel to that of
the nozzle.
Fig. 9 Nozzle
diameter vs. droplet velocity
Particularly, in
the case of high viscosity ink, when
resistance of the
nozzle is decreased, the increase
of ink droplet
velocity is large. Fig. 10 shows a
change of droplet
velocity, in the case of changing
the taper angle
of nozzle.
Fig. 10 Taper
angle vs. droplet velocity
When a nozzle diameter
is small, the influence of
taper angle is
great. In order to reduce nozzle
resistance, it is
also effective that the length of
nozzle is
reduced, whereby however, the stiffness
of the nozzle
plate is also reduced, and the
pressure in the channel
is lowered by the increase
of compliance,
and the fluctuation of the jet
trajectory
increases.
Further, the
taper angle can not also be increased,
because it
affects a jet trajectory accuracy. In
addition, viscous
resistance of the nozzle largely
influences on the
ink replenishment time after
droplet ejection
as well as on attenuation of the
pressure wave,
and therefore, attention must be
paid to the
design of nozzles (1).
Fig. 11 SEM
photograph of nozzle cross section
In order to
reduce energy loss and to stabilize the
jet trajectory, a
funnel type nozzle shown on the left
side in Fig. 11
shows good characteristics. Fig. 12
is an example in
which the droplet velocity was
calculated when
the nozzle diameter was changed
on a taper type
nozzle and a funnel type nozzle.
Compared to the
conventional taper type nozzle
having a small
taper angle, a greater increase of
droplet velocity
can be expected.
Fig. 12 Nozzle
shape vs. droplet velocity
4 Characteristics
of the prototype print head
In accordance
with the result of the simulation, the
shapes of the
actuator and the channel were
designed to
achieve high drive efficiency, and a
print head whose
nozzle shape was changed from
the conventional
taper type to the funnel type was
made on a trial
basis. Specifications of the
prototype print
head are shown in Table 1. As
drive method of
the print head, a so-called 3-cycle
firing by which
the ink is ejected every three
channels at a
time and the drive of all channels is
completed by 3
cycles was used, because of the
actuators that
are shared for the adjoining
channels.
Further, an oil-based ink with a
relatively high
viscosity was used. The driving
energy necessary
for ejection of one droplet in this
print head was
0.45 µJ.
Table 1
Specifications of prototype head
Nozzle
Funnel
Type
Ink
droplet volume
15 pl
Viscosity
10 mPa·sec
drive frequency
13 kHz
surface tension
28 mN/m
number of
channels
512 ch.
density
0.89 g/cm3
channel array
density
180 dpi
5 Consideration
Driving energy
for the shear mode piezo inkjet head
is imparted to
the actuator as electrical energy, and
the greater part
of the energy is consumed in the
drive circuit,
and a part of the rest is converted to
elastic energy in
the ink in the channels by a
displacement of
the actuator. This elastic energy
propagates in the
channel as a pressure wave to
form a standing
wave. Then, it pressurizes the ink
in the nozzle to
eject a droplet.
The required
energy to eject the ink droplet includes
the energy to
form the droplet surface and the
kinetic energy of
the droplet, and in addition, a
considerable
energy is consumed for the flow of the
ink in the
nozzle. Further, even after droplet
ejection, more
energy is consumed until the
residual
oscillation of the ink is terminated.
When the driving
energy of the prototype inkjet
head is roughly
calculated, the elastic energy of the
ink in each
channel is 6 nJ, which is nearly two-digit
smaller than the
electrical input energy of 0.45 µJ,
and the droplet
surface forming energy is 0.08 nJ,
and the droplet
kinetic energy is about 0.22 nJ.
The shapes of the
actuator, channel, and nozzle
were optimized
based on the computational
simulation
analyses, the results were that a
prototype print
head proved that its drive efficiency
was twice or more
than the conventional one.
6 Conclusion
On the basis of
driving efficiency analyses of the
shear mode piezo
inkjet head through
computational
simulation, the channel shape or
nozzle shape have
been optimized, and an inkjet
head of better
drive efficiency was made on a trial
basis. If the
electrostatic capacity of wiring section
is reduced, the
efficiency can further be raised
several times.
Since the
fabrication of the shear mode piezo inkjet
head of a
multi-channel type is comparatively easy
and a high
efficiency drive is possible (3), the shear
mode piezo inkjet
head is promising as a head for a
higher speed and
a higher print quality printers in
near future.
• References
1) Yoshio
Takeuchi, Konica Tech. Rep., Vol 15, 31
(2002).
2) Iwaishi,
Miyaki, Kawamura, Kato, Mikami, Japan
Hardcopy 2000
Papers (2000).
3) Alfred
Zollner, Peter Moestl, SPIE 2949, 434
(1997).