[0001] This invention relates to a printer head for an ink-on-demand type ink-jet printer
in which ink droplets are squirted out each time a driving pulse is applied to an
electromechanical transducer attached to one wall of an ink chamber.
[0002] Various types of ink-jet printers have been proposed as described in an article "Ink
Jet Printing" by Fred J. Kamphoefner published in the "IEEE Transactions on Electron
Devices, Vol. Ed-19, No. 4, April 1972, pp. 584-593. The ink-jet printer of ink-on-demand
type is described in detail, for example, in the United States Patent No. 3,946,398
entitled "Method and Apparatus for Recording with Writing Fluids and Drop Projection
Means Therefor" issued to E. L. Kyser et al, and the United States Patent No. 4,106,032
entitled "Apparatus for Applying Liquid Droplets to a Surface by using High Speed
Laminar Air Flow to Accelerate the same" issued to M. Miura et al. In GB-A-1 551 990
there is described a printer head for an ink-on-demand type ink jet printer for squirting
ink droplets on to a printing medium. This previously proposed printer head has a
nozzle through which ink droplets are to be squirted, a supply passage for supplying
ink from an ink tank, pressure exertion means for exerting pressure on the ink in
accordance with an electric signal to squirt the ink droplets and fluid control means
including two non-deformable valves which are disposed in the ink passage and can
be solenoid operated.
[0003] In contrast to this known embodiment, the present invention is characterised by the
features set out in claim 1.
[0004] In the conventional ink-on-demand type ink-jet printer, as shown in Fig. 1, a nozzle
104 and an ink feed port 105 communicate with a pressure chamber 103 which is filled
with ink and which genertes a pressure pulse by deforming a wall 102 with electromechanical
transducer means 101. The ink feed port 105 feeds the ink from an ink tank to the
pressure chamber 103. When a pressure is not applied, the meniscus of the ink is formed
at the end face of the nozzle 104 and balances with the pressure of the ink owing
to the surface tension. A piezoelectric element is mainly employed as the electromechanical
transducer means 101. The piezoelectric element is fastened to the wall 102. When
a driving voltage is applied across two electrodes 106 and 107 from a voltage source
108, an internal stress incurvates the wall 102 arises in the piezoelectric element
101.
[0005] The formation of ink droplet is carried out as follows. When the wall 102 is curved
inwardly of the pressure chamber 103 by applying the voltage to the piezoelectric
element 101, the internal volume of the pressure chamber decreases, and ink within
the pressure chamber is forced out of the pressure chamber. The ink pressure at this
time squirts the ink from the nozzle 104. The ink pressure from the pressure chamber
acts also on the ink feed port 105, and also an ink stream which returns from the
ink feed port to the ink tank arises. Subsequently, when the voltage applied to the
piezoelectric element 101 is returned to zero or a voltage of the opposite polarity
is applied so as to render the deformation of the wall 102 of the pressure chamber
null or to curve the wall outwardly of the pressure chamber, the internal volume of
the pressure chamber increases, and the ink is drawn back into the pressure chamber.
At this time, at the ink feed port 105, the ink is fed from the ink tank into the
pressure chamber, while at the nozzle 104, the meniscus is retracted from the nozzle
end into the nozzle. Subsequently, when the volume fluctuation in the pressure chamber
has stopped, the meniscus retracted into the nozzle reverts to the nozzle end again
under the action of the surface tension of the ink. With the transfer of the meniscus
succeeding to the increase of the internal volume of the pressure chamber, the ink
feed from the ink tank is continued at the ink feed port 105. At the time when meniscus
has reverted to the nozzle end and has come to a standstill, the supply for the ink
previously squirted from the nozzle 104 finishes off.
[0006] The droplet forming operation in the conventional ink-on-demand type ink-jet printer
has involved several problems as stated hereunder. In the first place, the ink pressure
generated by the deformation of the wall of the pressure chamber acts, not only on
the nozzle portion, but also on the ink feed port, so that the loss of energy dissipated
otherwise than the droplet formation is great. In order to squirt the ink droplets,
accordingly, the volume change of the pressure chamber needs to be made large by applying
high electric energy to the piezoelectric element. This has led to such problem as
degradation in the characteristics of the piezoelectric element and lowering in the
response rate of the droplet formation.
[0007] Secondly, there has been the problem that when the repetition period of the ink droplet
formation is shortened, the volumes and flight speeds of the squirted ink droplets
fluctuate. This is attributed to the fact that while the meniscus is being returned
into the nozzle by the change of the internal volume of the pressure chamber after
the ink droplet formation, the next droplet- formation operation starts, and that
the position and transfer speed of the meniscus at the starting of the operation change
depending upon the repetition period. In order to suppress such fluctuations of the
characteristics, the reverting speed of the meniscus retracted into the nozzle needs
to be increased. Since, however, the reversion of the meniscus is dependent upon the
surface tension of the ink and the value of this surface tension is inherent in the
ink material, it is subject to a limit to increase the reverting speed of the meniscus
by making the surface tension great.
[0008] In the next place, there has been the problem that when the repetition period of
the ink droplet formation is made still shorter, the volume of the ink droplet decreases,
the droplet formation stopping eventually. This problem has been known to be also
attributed to the limitation in the reverting speed of the meniscus retracted into
the nozzle.
[0009] In the prior art, accordingly, the number of ink droplets to be formed within one
second or the ink droplet frequency is at most 3KHz or so in a practical range, and
the highest frequency with the fluctuations of the characteristics neglected is approximately
10KHz. The prior art has therefore been unsuitable for high-speed high-density printing.
[0010] An ink-jet printer head in which an ink passage on the ink feed side is constructed
of a fluidic element in order to enhance the energy efficiency at the droplet formation,
is disclosed in the United States Patent No. 3,848,118. The fluidic element has the
effect that the flow resistance to an ink stream changes depending upon the direction
of the ink stream, and it intend to enhance the energy efficiency at the droplet formation.
Since, however, the fluidic element attains kind of rectification characteristic by
utilizing physical properties inherent in fluids, there are such problems that the
characteristic fluctuates depending upon the nature of the fluid used and that the
ratio of the flow resistance changes responsive to the directions of the ink stream
cannot he set large. Moreover, the fluidic element has been structurally complicated
and has been attended with much difficulty in disposing it within the ink-jet printer
heads.
[0011] An object of this invention is to provide a novel ink-jet printer head which solves
the various problems in the prior art.
[0012] According to this invention, there is provided a printer head for an ink-on-demand
type ink-jet printer for squirting ink droplets onto a printing medium, said printer
head comprising: a nozzle for squirting said ink droplets; a supply passage for supplying
ink in communication with an ink tank; pressure exertion means for exerting a pressure
on said ink in accordance with an electric signal to squirt said ink droplets; and
fluid control means having a valve which is deformed under the action of the ink pressure.
[0013] Other features and advantages of this invention will be apparent from the following
description of preferred embodiments of this invention taken in conjunction with the
accompanying drawings, wherein:
Fig. 1 is a cross-sectional view of a conventional ink-jet printer head;
Fig. 2 is a cross-sectional view of a first embodiment of this invention;
Figs. 3A, 3B, 4A, 4B, 4C, 5A, 5B, 5C, 6A, 6B, 6C, 7A, 7B, 9A, 9B, 9C and 10 show various
examples of a fluid control means used in the first embodiment;
Fig. 8 is a cross-sectional view of a second embodiment of this invention;
Figs. 11, 12A and 12B are diagrams useful for explaining the operation of the ink-jet
printer head;
Figs. 13A, 13B, 13C, 14A, 14B, 16A, 16B, 16C, 16D and 16E are diagrams showing examples
of the driving voltage to be applied to the ink-jet printer head and change in the
pressure and the change in the capacity of the pressure chamber; and
Figs. 15 and 17 are block diagrams of the driving voltage forming means.
[0014] Referring to Fig. 2, the first embodiment of this invention includes an ink-jet printer
head 10 which is constructed of a nozzle 14 for squirting ink, a supply passage 15
communicating with an ink tank (not shown) and for supplying the ink, a pressure chamber
13 filled with the ink, a piezoelectric element 11 fastened to a wall 12 of the pressure
chamber, first fluid control means 21 disposed between the nozzle 14 and the pressure
chamber 13, and second fluid control means 22 disposed between the pressure chamber
13 and the supply passage 15. When the ink passes through the fluid control means,
a pressure loss occurs. Herein, the ratio of the pressure loss to the flow rate of
the ink (hereinafter, referred to "flow resistance") changes depending upon the direction
of the passing ink stream. The fluid control means 21 and 22 operates so that the
flow resistance may become low under the action of the pressure of the ink in response
to the ink stream in the direction of from the supply passage side toward the nozzle
side, whereas the f!ow resistance may become high under the action of the ink pressure
in response to the ink stream in the direction of from the nozzle side toward the
supply passage side.
[0015] The formation of ink droplets in the ink-jet printer head 10 is carried out as follows:
When the wall 12 is curved inwardly of the pressure chamber 13 by applying a voltage
to the piezoelectric element 11 through electrodes 16 and 17 from a power source 18,
the pressure owing to the wall 12 acts on the ink within the pressure chamber 13.
As a result, the stream from the pressure chamber 13 toward the nozzle 14 acts on
the first fluid control means 21 so as to render the flow resistance low, whereas
the stream of the ink from the pressure chamber 13 toward the supply passage 15 acts
on the second fluid control means 22 so as to render the flow resistance high. Therefore,
the difference of the two flow resistances becomes large, the ink forced out of the
pressure chamber flows out principally toward the nozzle side, and the ink droplet
is squirted from the nozzle 14. Subsequently, when the voltage applied to the piezoelectric
element 11 is returned to zero or voltage of the opposite polarity is applied so as
to render the deformation of the wall 12 of the pressure chamber null or to curve
the wall outwardly of the pressure chamber, the interval volume of the pressure chamber
13 increases and the pressure inside the pressure chamber 13 decreases. As a result,
the stream from the nozzle 14 toward the pressure chamber 13 acts on the first fluid
control means 21 to render the flow resistance high, whereas the ink stream from the
supply passage 15 toward the pressure chamber 13 acts on the second fluid control
means 22 to render the flow resistance low. In this case, accordingly, the ink flows
principally from the supply passage 15 into the pressure chamber 13, and the extent
to which the meniscus in the nozzle portion is retracted into the nozzle lowers.
[0016] As apparent from the droplet forming operation described above, in the ink-jet printer
head 10, the deformation of the pressure chamber 13 acts for squirting the ink droplet
when the volume of the pressure chamber decreases, and it acts for supplying the ink
when the volume to the pressure chamber increases. Therefore, the dissipation of energy
onto the supply passage side at the ink droplet squirt as in the prior art lessens,
and the energy efficiency is enhanced. Further, since the retraction of the meniscus
into the nozzle at the ink supply lessens, the period of time required for the meniscus
to revert to the nozzle end part is shortened. In addition, since the change of the
internal volume of the pressure chamber by the piezoelectric element may be small
on the same order as the volume of the ink droplet, the period of time for forming
the droplet can be shortened in the extreme. Since excess electric energy need not
be applied to the piezoelectric element, the degradations of the characteristics of
the piezoelectric element are not incurred. Further, in the prior art, the ink supply
is relied on the surface tension of the meniscus at the nozzle portion, whereas the
ink is supplied from the supply passage owing to the increase of the internal volume
of the pressure chamber and forcibly performs the ink supply by the use of the external
energy such as electric energy, so that the ink supply for the high-speed droplet
formation is possible. Thus, the ink-jet printing apparatus according to this invention
has realized the droplet formation at very high speed.
[0017] The example employing the two fluid control means has been described above. However,
even when two fluid control means are not conjointly used, the use of the means either
between the pressure chamber and the nozzle or between the ink feed port and the pressure
chamber is effective: In case of employing the means between the ink feed port and
the pressure chamber, when the internal pressure of the pressure chamber is raised
to extrude the ink, the ink flows out toward the nozzle portion and the supply passage.
The fluid control means disposed on the supply passage side can be set so that the
flow resistance thereof may become greater than that in the nozzle portion under the
action of the ink pressure. As a result, most of the ink extruded from the pressure
chamber is squirted from the nozzle. Subsequently, when the internal pressure of the
pressure chamber is lowered to retract the ink into the pressure chamber, the ink
flows in the pressure chamber from the nozzle portion and the supply passage. At this
time, the fluid control means located on the supply passage side operates so that
the flow resistance may decrease. The flow resistance in the supply passage can be
set so as to become less than that in the nozzle portion. As a result, most of the
ink flows from the supply passage into the chamber. In this manner, the ratios between
the flow resistances in the nozzle portion and the supply passage in case of the extrusion
and retraction of the ink from and into the pressure chamber are made different, whereby
the same effect as in the first embodiment can be brought forth.
[0018] A similar effect is also attained in the case where the fluid control means is disposed
between the pressure chamber and the nozzle. In this case, the flow resistance of
the fluid control means is smaller under a high internal pressure than that under
a low internal pressure. In order to intensify the effect of the fluid control means,
it is desired to set the flow resistance of the fluid control means to be lower than
that of the ink supply passage when the internal pressure is high, and to be higher
than that of the ink supply passage when the internal pressure is low.
[0019] The stream of the ink in the on-demand type ink-jet printer head is always pulsatile,
and the quantity of the ink which is passed through a valve by one pulsatile stream
is as extremely small as approximately equal to at most the volume of the ink droplet.
On the other hand, in order to enhance the rectification effect based on the operation
of the valve, it is important that the volume of the space in which the valve moves
is made sufficiently smaller than the quantity of the ink to pass through the valve.
[0020] Referring to Figs. 3A and 3B, in the first example of the fluid control means, a
flat valve 301 made of an elastic member is arranged so as to cover an ink outflow
port 302. The valve 301 is fixed in close contact with a valve seat 304 by means of
a stationary portion 303. In the absence of the stream of the ink, also a movable
portion 305 lies in close contact with a valve seat 306. Now, when a pressure which
causes the ink flow upwardly from below the valve as shown by an arrow in Fig. 3B,
has acted on the valve, the valve 301 is pushed up and the ink flows out through the
opening 308 between the valve and the valve seat. Since the pressure loss at this
time occurs mostly in the opening 308, the dimension of the overlap parts 309 of the
valve and the valve seat needs to be made small so as to reduce the pressure loss.
As the material of the valve, there can be used thin films of metals such as gold,
nickel and stainless steel and various plastic films. For example, in case of employing
a polyethylene terephthalate film, the movable portion 305 and the outflow port 302
of the valve are respectively made square with each side being 200 um long and 180
11m long, and the length of the overlap part 309 is set to 10 11m. The valve is formed
by punching from the polyethylene terephthalate film of 20 11m thick. The valve is
fixed in such manner that the stationary portion 303 is pressed against the valve
seat 304 by a fixing member 307.
[0021] The wall 12 of the pressure chamber 13 is made of a cold-rolled stainless steel plate
of 0.4 mm thick. The piezoelectric transducer may be of NEPEC N-10 having dimensions
of 2 mmx26 mmxO.4 mm. The piezoelectric transducer is fastened to the wall 12 with
an epoxy type solventless thermosetting binder. As the nozzle 14, a hole which is
50 pm in diameter and 100 pm in length is formed by the electric discharge machining.
The droplet formation is carried out by supplying the piezoelectric transducer with
a pulse voltage which has a waveform corresponding to one wavelength of the cosine
wave. As a result, the formation of an ink droplet having a diameter of about 100
pm and an initial velocity of about 2.4 m/sec is observed at a pulse width of 55 psec
and a peak voltage of 80 V. When the recurrence frequencies of the pulses is changed,
the maximum value of the operating frequencies at which the fluctuations of the initial
velocity of the droplet fall within 10% is 18 KHz. In addition, in case where the
fluid control means is disposed on only either the nozzle side or the supply passage
side, the operating frequency is about 12 KHz under the operating conditions mentioned
above. On the other hand, in case where the ink-jet printer head of the same geometries
is used without providing the fluid control means, the operating frequency for fulfilling
the droplet velocity fluctuations of within 10% is at most only about 1.5 KHz. In
order to form an ink droplet having a diameter of 100 pm, a peak voltage of about
9 V is necessary, and the initial droplet velocity is about 1.8 m/sec. From these
results, it is apparent that the fluid control means according to this invention is
very greatly effective for raising the operating speed of the ink jet and that it
is effective for enhancing the energy efficiency in the droplet formation.
[0022] It is advantageous in the assemblage job to make the geometries of the valve as large
as possible. In this case, however, the geometries must be limited to the range in
which the volume of the moving space of the valve is smaller than the volume of the
droplet. In addition, as the valve enlarges, also its thickness needs to be increased.
For example, when the sides of square valve made of polyethylene terephthalate enlarged
to 300 pm and 400 pm, the thicknesses of the valves needed to be made 35 pm and 75
pm, respectively. It has been confirmed that when the dimensions of the valve are
further enlarged, the effect of the valve lowers abruptly.
[0023] The fluid control means in this invention has its one feature in exploiting the displacement
of the valve owing to the ink pressure. In order to hold the reliability of the valve
high, the displacement of the valve needs to be made within the elastic limit thereof.
In case where the valve is displaced beyond the elastic limit, it is deformed and
cannot return to its original closed state. With the cantilever valve as in the first
example shown in Fig. 3, a pressure range in which the elastic limit is not exceeded
is narrow. Therefore, it has sometimes been the case that an excess ink pressure acts
to deform the valve at, for example, the initial charging of the ink-jet printer head
with the ink.
[0024] Referring to Figs. 4A and 4B, a doughnut-shaped disc valve 401 made of an elastic
member is fixed in close contact with a valve seat 404 by a stationary portion 403,
and in the absence of the stream of the ink, also a movable portion 405 lies in close
contact with a valve seat 406 so as to blockade an ink outflow port 402. Such valve
is constructed of a components as shown in Fig. 4C. More specifically, the valve seats
404 and 406 are unitarily formed in a manner to have the annular outflow port 402
therebetween. The disc valve 401 formed with a hole in its central part is stacked
on the valve seats, and a ring-shaped fixing member 407 is further stacked on the
valve, to fix the valve.
[0025] When a pressure causing the ink to flow upwardly acts on the valve, the valve 401
is pushed up similarly to the cantilever valve in Fig. 3, the ink flows out through
the opening 408 between the valve and the valve seats as shown in Fig. 4B. The deformation
of the disc valve around the central hole thereof as in Fig. 4B includes an elongation
in the circumferential direction of the central hole in addition to the same simple
bending in the radial direction as in the cantilever valve of Fig. 3. Accordingly,
the disc valve is more difficult of deformation than the cantilever valve and has
its durability sharply enhanced against the action of intense pressure.
[0026] An example of the geometries of the fluid control means constructed by the use of
a polyethylene terephthalate film will be given below. The doughnut-shaped disc valve
401 is formed by punching from the polyethylene terephthalate film 20 11m thick. The
diameter of the central hole is set at 300 um, and the outside diameter of the movable
portion 405 of the valve 401 is set at 500 µm. The outside diameter of the valve seat
406 is set at 320 µm so that the length of the overlap parts 409 of the movable portion
405 and the valve seat 406 may become 10 µm. The fluid control means thus constructed
is applied to the ink-jet printer head 10 shown in Fig. 2. As a result, the same effects
as in the fluid control means shown in Fig. 3 has been confirmed.
[0027] In case of employing a gold foil as the material of the disc valve 401, the thickness
of the valve needed to be as small as 5 µm when the diameter of the central hole and
the outside diameter of the movable portion of the disc valve are set equal to those
in the case of the polyethylene terephthalate film. In fabricating such thin disc
valve, various microscopic machining techniques have been known. For example, according
to the machining technique called "electroforming", an electrode in the flat shape
of the disc valve is plated with gold up to a predetermined thickness in the vertical
direction, whereby the disc valve of the gold foil can be formed.
[0028] The thickness of the valve must be made smaller as the elastic modulus of the material
used is greater. For example, in case where the disc valve having the same geometries
as those of the disc valve of the gold foil is made of stainless steel, the pressure
for the droplet formation needs to be made approximately double that in the case of
the gold foil. In order to perform the droplet formation under the same conditions,
accordingly, the thickness of the stainless steel valve needs to be made smaller than
5 um. The thin valve, however, has such problems that the handling is difficult and
that thin foils are difficult to obtain in case of some materials.
[0029] Referring to Fig. 5A and 5B, in the third example of the fluid control means, a valve
501 arranged in close contact with a valve seat 506 so as to cover an outflow port
502 is supported by fine supporting arms 510, and is fixed to the valve seat 506 by
a stationary portion 503. As understood from Fig. 5C, the third example can be constructed
by successively stacking the valve seat 506 having the outflow port 502 in its central
part, a valve member 511 which is centrally located and in which the valve 501 to
cover the outflow port is unitary with the peripheral stationary ring 503 through
the fine supporting arm 510, and a fixing member 507. In response to the upward stream
of the ink, the valve 501 is pushed up as shown in Fig. 5B and the ink flows out through
the opening 508 between the valve and the valve seat 506. In the third example, the
operation of the valve involves the flexures and elongation of the supporters 510,
and the valve has its durability sharply enhanced against the action of intense pressure
in comparison with the cantilever valve of Fig. 3. Moreover, since the displacement
of the valve is based on the deformation of the supporter portion, the quantity of
the displacement of the valve can be set larger than in the disc valve of Fig. 4,
resulting in the advantage that the versatility of the selection of the valve material
and the versatility of the design are sharply enhanced.
[0030] An example of the geometries in the case where a stainless steel sheet of 10 µm thick
is used will be given below. In case of the configuration in which the disc valve
501 has four supporting arms 510 similarly to the valve member 511 shown in Fig. 5C,
the outside diameter of the disc valve 501 is set at 200 µm, the width and length
of the supporting arm are respectively set at 50 µm and 400 pm, the diameter of the
ink outflow port 502 is set at 180 µm, and the length of the overlap parts of the
disc valve 501 and the valve seat 506 is set at 10 pm. The fluid control means thus
constructed may be applied to the ink-jet printer head shown in Fig. 2. As a result,
the same effects as in the fluid control means shown in Figs. 3 and 4, has been confirmed.
This example is highly practical because of the additional advantages that the stainless
steel readily available can be used and that the outside dimensions can be set large.
[0031] While, in any of the fluid control means thus far described, the movable portion
of the valve is arranged so as to lie in close contact with the valve seat at a standstill,
such close-contact arrangement is not always necessary.
[0032] Referring to Figs. 6A and 68, in the fourth example of the fluid control means, a
disc valve 601 is fixed in close contact with a valve seat 604 by a stationary portion
603, and in the absence of the stream of the ink, a movable portion 605 lies in a
position separate from a valve seat 606 and the parts of the ink before and behind
the valve communicate. However, when a pressure acts to cause the ink to flow downwardly,
as shown in Fig. 6B, the valve 601 flexes and comes into contact with the valve seat
606, and it acts to prevent the stream of the ink.
[0033] On the other hand, under the action of a pressure causing the ink to flow upwardly
as shown in Fig. 6C, the valve 601 flexes upwards and the ink flows out upwards through
the opening 608 between the valve and the valve seat. In this manner, to the end of
attaining the rectification effect of permitting the ink to flow in one direction,
it is not always necessary that the valve and the valve seat lie in close contact
at a standstill.
[0034] In case of employing a doughnut-shaped disc valve which has been actually made of
a polyethylene terephthalate film of 20 um thick and in which the diameter of the
central hole is 300 µm and the outside diameter of the movable portion is 500 um,
the quantity of flexure of the valve under the ordinary droplet-forming conditions
is readily calculated within the scope of the fundamental knowledge of the material
strength and is found to be approximately 3 µm. A stationary rectification effect
is accordingly achieved when the spacing between the valve and the valve seat at the
standstill is up to 3 11m or so. Needless to say, the fact that the valve and the
valve seat need not always lie in close contact at the standstill is applicable, not
only to the cases of employing the doughnut-shaped disc valves, but also to the fluid
control means shown in Figs. 3 and 5.
[0035] When, in the fourth example, the spacing between the valve and the valve seat at
the standstill is increased beyond the quantity of flexure of the valve, the ink comes
to flow also in the opposite direction, and the rectification effect weakens gradually.
However, in a range in which the spacing between the valve and the valve seat is not
very great as compared with the quantity of flexure of the valve, the flow resistance
can be greatly changed depending upon the direction of the ink stream, and hence,
a satisfactory function can be exercised as the fluid control means. There are considered
as the flow resistance an inertial resistance and a viscous resistance which are based
on the ink stream, and a loss term in the part in which the sectional shape of the
flow passage changes. The viscous resistance becomes the greatest with respect to
the steady flow after the valve has flexed a fixed amount. As a basic characteristic
concerning viscous fluids, it has been known that the viscous resistance is proportional
to d-
3 with respect to the spacing d between the valve and the valve seat. Accordingly,
in case where, for example, the spacing between the valve and the valve seat at the
standstill is set at 8 um for a valve flexure quantity of 3 pm, the flow resistance
changes about 10 times in dependence on the direction of the ink, and a satisfactory
function can be achieved as the fluid control means. The fluid control means as thus
far described in which the ink stream in the opposite direction is not perfectly cut
off but the flow resistance changes greatly in dependence on the direction of the
ink stream, is readily performed in such a way that the spacing between the valve
and the valve seat in each of the examples shown in Figs. 3, 4 and 5 is much greater
than the quantity of flexure of the valve.
[0036] The above-mentioned structures in which the valve and the valve seat are separated
at the standstill have an important advantage from the viewpoint of practical use.
At the standstill, the ink meniscus is formed at the end face of the nozzle of the
ink-jet printer head, and a liquid component in the ink in the nozzle portion is continually
vaporizing. Now, in case where the fluid control means is disposed between the pressure
chamber and the supply passage and where the ink passages before and behind the fluid
control means are separated from each other by the valve, the ink in the nozzle decreases
and the meniscus is retracted into the nozzle. When the retraction of the meniscus
arrives at the pressure chamber, the air is introduced into the pressure chamber and
a stable droplet formation can no longer be executed. In contrast, when the ink passage
parts before and behind the fluid control means are communicating even at the standstill,
the ink is supplied from the supply passage side to the extent that the ink in the
nozzle portion has decreased due to the vaporization of the liquid component, and
hence, the ink meniscus is always kept at the nozzle end face.
[0037] As the fluid control means whose flow resistance changes depending upon the direction
of the ink stream, there will now be described one having a valve structure somewhat
different from those of the examples described above. The fifth example of the fluid
control means shown in Figs. 7A and 7B is constructed of a wall member 35 defining
a flow passage of the ink, a plate member 31 provided with an aperture, a spacer 32,
a film 33 which can be deformed by the pressure of the ink, and a frame member 34
which has a frame for securing the film and whose outer side is a penetration portion.
The bore of the plate member 31 is made smaller than the diameter of the film. The
film 33 may be made of sheets of metals such as gold and stainless steel, films of
plastics, etc. The ink flows from the passage 36 to the passage 40 through the aperture
37, the opening 38 between the plate member 31 and the film 33, and the penetration
portion 39 of the frame member 34. When the ink flows from the passage 40 to the passage
36, it passes in the reverse order.
[0038] The operation will now be described. When the pressure has become higher on the side
of the passage 36 than on the side of the passage 40, the ink begins to move from
the passage 36 toward the passage 40. At this time, letting P" P
2, P
3 and P
4 denote respective pressure in or on the passage 36, the surface of the film 33 facing
the aperture 37, the penetration portion 39 and the passage 40, the relationship among
the pressures is represented by

Therefore, a pressure difference of (P
2-P
4) arises between both the surfaces of the film 33, and the film curves downwards.
Therefore, the gap width of the opening enlarges, and the flow resistance of this
part decreases. For this reason, when the ink flows from the passage 36 toward the
passage 40, the overall flow resistance of this flow passage decreases.
[0039] In contrast, when the pressure of the passage 40 is higher and the ink flows in the
opposite direction, the relationship of the pressure of the respective portions becomes

Therefore, the film 33 curves upwards conversely to the foregoing, the gap width of
the opening 38 decreases and the flow resistance increases. For this reason, the overall
flow resistance of the flow passage increases. Thus, the flow passage whose flow resistance
changes depending upon the direction of the ink stream can be provided, and a high-speed
ink-jet printer head may be provided by employing it as the fluid control means.
[0040] In order to intensify the effects by making the change of the flow resistance dependent
upon the direction of the ink stream great, it is desirable that the ratio k=d
2/d
1 between the gap width d
1 of the opening 38 and the width d
2 of the deformation of film 33 is close to 1 (unity). The magnitude of the ratio between
the flow rates dependent upon the direction of the stream becomes

under an identical pressure. For example, the material of the film is polyethylene
terephthalate of 10 um thick, the diameter of the film is 400 pm, and the gap width
of the opening 38 is 10 pm. Then, supposing the pressure difference between the front
and rear of the film is 0.5 atmosphere, k=0.5 is obtained and the ratio of the flow
rates becomes about 60.
[0041] Referring to Fig. 8, in the second embodiment of this invention, the fluid control
means is arranged between the ink feed port and the pressure chamber in the conventional
ink-jet printer head shown in Fig. 1. As the fluid control means, a passage 42 having
a fixed gap width is disposed between an ink feed port 41 and a pressure chamber 103,
the wall of the passage on one side is provided with a hole communicating with the
pressure chamber, and a film 43 is fastened to a part of the passage 42. As the film,
the sheet of a metal, plastic or the like can be employed as mentioned above.
[0042] In operation, when a voltage is applied to a piezoelectric element 101 to curve the
wall 102 of the pressure chamber 103 inwards, internal pressure of the pressure chamber
rises. Then, the film 43 curves to reduce the gap width of the passage 42. Therefore,
the flow resistance of the passage 42 increases, and most of the ink extruded from
the pressure chamber is squirted from the nozzle 104. Subsequently, when the voltage
is returned to its original value, incurving wall 102 undergoes a force which restores
it to its original position, and hence, the pressure of the pressure chamber becomes
lower than the external pressure. Therefore, the film 43 returns onto the pressure
chamber side conversely to the foregoing, and the flow resistance of the passage 42
decreases. For this reason, most of the ink to flow into the pressure chamber is applied
through the passage 42. Accordingly, the quantity of the retracted ink is smaller
than the quantity of the squirted ink in the nozzle portion, the period of time in
which the retracted meniscus return to the nozzle end is shortened, and it is permitted
to shorten the period of the ink droplet formation.
[0043] In each of the various examples of the fluid control means as stated above in which
the flow resistance changes depending upon the direction of the ink stream, the change
of the flow resistance is the sum of the fixed value independent of the direction
of the ink stream and a value varying under the action of the valve. Accordingly,
the flow resistance can also be changed in dependence on the direction of the ink
stream in such a way that an auxiliary ink passage having a fixed flow resistance
is disposed jointly with the fluid control means which has previously been illustrated
in Figs. 3, 4 and 5 and which has the complete rectification action. For example,
Fig. 9 shows another example of the fluid control means in which auxiliary ink passages
penetrating at all times are respectively provided in the valve seats in the three
kinds of fluid control means shown in Figs. 3,4 and 5. Fig. 9 shows the case where
the valve is open under the action of the ink pressure and where the ink is flowing
upwards. In order for such fluid control means to exercise a satisfactory function
in the ink
-jet head, the flow resistance to the ink passing through opening 908 between a valve
901 and a valve seat 906 needs to become sufficiently lower than the flow resistance
of the auxiliary ink passage 912.
[0044] A still another example of the fluid control means is shown in Fig. 10, in which
the parts of the ink passage covered by the valve and the valve seat, as shown in
Figs. 3, 4 and 5, are communicated by an auxiliary passage which is provided so as
to bypass the valve and the valve seat. In this case, the flow resistance to the ink
passing through the opening 908 between the valve 901 and the valve seat 906 needs
to be sufficiently lower than the flow resistance of the auxiliary ink passage 912.
[0045] The driving voltage to be applied to the piezoelectric transducer 11 or 101 will
be described. When the driving voltage is applied to the piezoelectric transducer,
a pressure is generated in accordance with its voltage value φ or with a quantity
of deflection, and the magnitude of the pressure can be approximated as follows:

Here, A and B are constants that are determined by the dimension and material of the
piezoelectric oscillation plate, P is the pressure, φ is the voltage, φ
o is the initial voltage, V is the capacity of the pressure chamber and V
o is the capacity of the pressure chamber when φ=φ
2 and P=0.
[0046] As can be understood from the equation (1), the pressure occurring on the piezoelectric
oscillation plate increases in proportion to the amplitude (φ-φ
o) of the impressed voltage. When the voltage is applied to the piezoelectric oscillation
plate, the pressure occurs in accordance with the equation (1) and the ink is caused
to flow out from the pressure chamber. As the capacity of this pressure chamber decreases,
the pressure also decreases progressively and when a predetermined voltage I)
a is kept being applied, the pressure approaches zero. In this instance, the capacity
of the pressure chamber progressively approaches the saturation value Va(P=
0). The behaviour of the capacity change in this case is shown in Figure 11 a. In the
drawing, symbols b and c represent the behaviours when other voltages q)
b and φ
c are applied, respectively. Incidentally, the relation between these voltages is φ
a<φ
b<φ
c. It can be understood from this drawing that the deflection quantity of the capacity
becomes greater with an increasing amplitude of the voltage. It can also be understood
that the time required for deflection can be shortened by increasing the amplitude
of the voltage if a predetermined capacity change is to be effected by this oscillation
plate. It will be assumed the case in which the capacity is to be changed from V.
to V
a. As shown in Figure 11, when φ
a is impressed, the time required for deflection theoretically becomes infinite, and
when the voltage increases to
4)
b and then to φ
c, the time required for deflection becomes shorter to T
b and T
e, respectively.
[0047] Next, deflection will be considered when a pulse voltage is applied to this piezoelectric
oscillation plate. The width of this driving pulse is related with an ink droplet
formation characteristic. In the ink jet head such as shown in Figure 2, behaviour
of the droplet formation will be considered by changing the width of the driving pulse.
In this case, when the pulse width becomes greater and exceeds a predetermined width,
the volume and initial speed of the ink droplet become constant irrespective of the
pulse width. This is because the deflection of the pressure chamber reaches the saturation
value in Figure 11. However, in this pulse width, a large number of sub-droplets,
that are called "satellite", of ink are formed in addition to the main ink droplet
and cause recording problems. Furthermore, the time required for the deflection to
attain its maximum becomes about 0.2 msec in ordinarily available ink jet recording
heads in this pulse width. In other words, if such a pulse width is employed, it is
not possible to obtain a high ink droplet-forming frequency. Accordingly, as a practical
driving pulse is used one that shortens the application time and finishes the pulse
application before the deflection attains its maximum. By use of such a pulse, the
formation of the satellites can be eliminated, and the repetition frequency of the
driving pulse can be enhanced since the pulse width is narrow.
[0048] Figure 12B shows the behaviour of the change in the capacity of the pressure chamber
when the driving pulse having a pulse width, at which the capacity of the pressure
chamber does not attain the maximum deflection quantity, such as shown in Figure 12A,
is employed. In this drawing, V
o represents the initial capacity and V
1 does the saturation value when φ
1 is impressed. When the impressed voltage reaches φ
1 from φ
o, the capacity of the pressure chamber so changes as to asymptotically approach the
saturation value V
1 and becomes V
2 at the time t
1 of finish of the φ
1 application. The deflection quantity of the pressure chamber capacity from V
o is smaller than the deflection quantity up to the saturation value V
i.
[0049] Next, when the impressed voltage is returned to φ
o, the capacity of the pressure chamber so changes from V
2 as to asymptotically approach the saturation value V
o when φ
o is impressed. Here, the time during which the pulse is impressed will be compared
with the time required for the capacity of the pressure chamber to return to V
o after the finish of the pulse application. The relation between the time required
for predetermined deflection and the impressed voltage is already described with reference
to Figure 11. Since the saturation is not yet attained, the deflection when the voltage
is impressed from φ
o to φ
1 corresponds to the case when the capacity of the pressure chamber changes from V
o to V
a as φ
b or φ
c is impressed, as shown in Figure 11. On the other hand, when the voltage is returned
from φ
o to φ
1, the behaviour corresponds to that of the case when φ
a is impressed in Figure 11 because the capacity of the pressure chamber asymptotically
approaches V
o at the voltage φ
o from V
2.
[0050] Accordingly, it can be understood that the voltage changes from φ
o to φ
1 and the time required for the pressure chamber capacity to return from V
2 to Vo when the voltage is returned from φ
1 to φ
o becomes greater than the time required for the one to change from V
o to V
1. If the ink droplet is formed by use of such a driving waveform as shown in Figure
12A, therefore, its repetition frequency becomes high, and if the application time
of the driving pulse and the repetition frequency approach each other, a succeeding
voltage is applied before the deflection of the pressure chamber is not capable of
returning its initial value V
o.
[0051] If there is any initial deflection as given by the aforementioned equation (1) when
the succeeding voltage pulse is applied, the pressure generated by the pressure chamber
becomes smaller even if the same voltage is impressed. For this reason, the velocity
of the jetted ink droplet and the volume of the droplet decrease. The decreases in
the ink droplet velocity and the ink droplet volume may presumably be attributed to
the slow action of the pressure chamber capacity in returning to its initial value.
On the basis of this assumption, the present invention contemplates to enhance the
ink droplet forming a frequency by making it possible for the capacity of the pressure
chamber to return to its initial value within a shorter period of time.
[0052] Figure 13A shows an example of the driving pulse in accordance with the present invention.
The waveform is so arranged that it first changes from φ
1 to φ
1 at the time of jetting of the ink and when the deflection of the pressure chamber
is to be returned to its initial state, it is set to φ
2 having an amplitude in the opposite direction to φ
1 with respect to φ
o before the voltage is returned to φ
o and is then allowed to return to φ
o. The mode of change in the capacity of the pressure chamber when this driving waveform
is applied is shown in Figure 13B. As the impressed voltage changes from φ
o to φ
1 and attains the stage in which the ink droplet is about to be jetted, the capacity
of the pressure chamber exhibits the same change as in Figure 12B. Next, when the
impressed voltage changes from φ
1 to φ
2, a pressure is generated, which pressure is greater than one when the impressed voltage
changes from φ
1 to φ
o. Accordingly, the capacity of the pressure chamber so changes as to asymptotically
approach from V
2 to V
3 which is greater than the initial value V
o. This is the same mode of change when the pressure chamber capacity changes with
a varying application voltage in Figure 11. Hence, the capacity of the pressure chamber
reaches the initial value V
o more quickly than when the voltage is simply returned to φ
o. The impressed voltage is returned from φ
2 to φ
o when the capacity of the pressure chamber reaches V
a. Though the application time of φ
2 in this description is the time required for the pressure chamber capacity to return
to V
o, some increases or decreases of the application time may be effected in practice
in consideration of the force of inertia of the ink or the like. This application
time of φ
2 varies in accordance with the magnitude of | φ
1-φ
2 | and when this absolute value is great, the application time becomes short, and
it becomes greater as the value | φ
1-φ
2 | is small and asymptotically approaches | φ
1-φ
o | , as can be appreciated clearly from the explanation with reference to Figure 11.
[0053] The driving waveform shown in Figure 13A consists of two pulses continuously combined
with each other, the former being for forming the ink droplet and the latter, for
shortening the returning time of the capacity of the pressure chamber. However, other
driving waveforms may be employed, as well, such as one consisting of two spaced-apart
pulses as shown in Figure 13C.
[0054] Though the foregoing embodiment uses a rectangular driving waveform for the ease
of description, the present invention is not limited in particular to such a waveform.
Namely, it is possible to use a waveform in which a predetermined time constant is
applied to the rise and fall of the pulse, a triangular wave, a sine wave, a trapezoidal
wave, and so forth. Namely, after the voltage for jetting the ink droplet is applied,
the voltage is not merely returned to its initial value φ
o, but a voltage of an amplitude in the opposite direction to the voltage pulse for
jetting the droplet is applied to the initial potential φ
o so that the capacity of the pressure chamber is capable of more rapidly returning
to its initial value V
o. Hence, the waveform for this purpose is not limitative, in particular.
[0055] It is possible to use, for example, one period of a sine wave as the driving waveform
such as shown in Figure 14A. In this waveform, the phase of the sine wave is so adjusted
that when one period of the driving waveform finishes, the capacity of the pressure
chamber returns to its initial value V
o, as shown in Figure 14B, in this waveform, too, the capacity of the pressure chamber
returns to its initial state more rapidly than a waveform devoid of such a voltage
that is lower than φ
o and is applied to the above-mentioned waveform.
[0056] The driving waveform in accordance with the present invention can be synthesized
by generally known methods. Figure 16 shows a representative of circuits for forming
the driving waveform such as shown in Figure 13A. In the drawing, when a trigger pulse
for forming the ink droplet is applied to a mono-multiple vibrator 211, there is produced
a rectangular pulse 212. This pulse width determines the <1>1 application time of
the driving pulse. The pulse 212 is applied to a second mono-multiple vibrator, which
is triggered by the rear end of the pulse and generates a rectangular pulse 214. The
pulse width of this pulse determines the <1>2 application time of the driving pulse.
The rectangular pulses 212 and 214 are subjected to the amplitude adjustment by an
amplitude control circuit 215 and are then applied to positive and negative input
terminals of a differential amplifier 216, thereby yielding a driving signal 217.
[0057] Another driving waveform in accordance with the present invention will further be
described. Figure 16A shows an example of the driving waveform and Figure 16B shows
an example of a compensating waveform. Driving waveforms of the present invention
synthesized from these waveforms are shown in Figures 16C, 16D and 16E. Figure 17
shows an example of the circuit for forming these driving waveforms. In Figure 17,
when a trigger 210 signal is applied to a driving waveform forming circuit 218, a
waveform such as shown in Figure 16A is produced as the output. This trigger signal
10 is also applied to a delay circuit 219 and is applied to a compensating waveform
forming circuit 219 after the passage of a predetermined time, thereby yielding the
waveform such as shown in Figure 16B. These outputs are applied to an amplitude control
circuit 221 and are then synthesized by a synthesizing circuit such as a differential
amplifier 222. In accordance with the delay time determined by the delay circuit 219,
there are obtained a waveform composed of two mutually overlapping waveforms such
as shown in Figure 16C, two continuous waveforms such as shown in Figure 16D and two
spaced-apart waveforms such as shown in Figure 16E. When the ink jet printer is actuated
by any of these waveforms, the force of returning the deflected piezoelectric oscillation
plate to its original state becomes great by applying, after the formation of the
ink droplet, a voltage having an amplitude in the opposite direction with respect
to the voltage applied at the time of jetting the ink droplet and consequently, the
returning time to the original state becomes shorter. The delay time is selected in
accordance with the signal response characteristic of the piezoelectric oscillation
plate of the ink jetter printer to be employed and with the amplitude of the signal
so that the time required for the capacity of the pressure chamber to return to its
original state is shortened.
[0058] As described in the foregoing, in accordance with the present invention, a voltage
having an amplitude in the opposite direction to the driving voltage for jetting the
ink with respect to the initial voltage is applied to the piezoelectric oscillation
plate in order to allow the capacity of the pressure chamber to immediately return
to its initial state after jetting of the ink. According to this arrangement, the
velocity and size of the ink droplet do not change even if the repetition frequency
for forming the ink droplet is increased, and there can thus be obtained an ink jet
printer having a high printing speed and high printing quality.