JP4439865B2 - Fluid discharge method - Google Patents

Description

The present invention relates to an information-precision equipment, machine tools, areas such as FA, or a semiconductor, liquid crystal, display is relates to a fluid ejection how the minute flow rate required by the various production processes, such as surface mounting.

Liquid discharge devices (dispensers) have been used in various fields in the past, but with the recent needs for downsizing electronic components and increasing recording density, a small amount of fluid material can be supplied with high accuracy and stability. Control technology is required. For example, in the field of displays such as plasma displays, CRTs, and organic ELs, there is a great demand for patterning phosphors and electrode materials directly on the panel surface without a mask, instead of conventional methods such as screen printing and photolithography. To summarize the issues of dispensers for that,
(I) Finer application amount (ii) Higher accuracy of application amount (iii) Shortening application time.

  Now, machining accuracy in machining is entering the order of micron to submicron. Submicron machining is common in the semiconductor and electronic parts field, but the demand for ultra-precision machining is also increasing rapidly in the machining field, which is progressing with mechatronics. In recent years, with the introduction of ultra-precision machining technology, magnetostrictive elements represented by giant magnetostrictive elements and piezoelectric elements have been applied as microactuators. An ejecting apparatus that ejects a minute amount of liquid droplets at a high speed by using this electrostrictive element as a source of fluid pressure has been proposed.

  For example, JP-A 2000-167467 discloses a method of ejecting any one droplet using a giant magnetostrictive element. In FIG. 24, reference numeral 502 denotes a cylinder made of a nonmagnetic material such as a glass pipe or a stainless steel pipe. An injection nozzle 504 having a liquid storage portion 503 and a fine injection port is formed at the end of the cylinder 502. Inside the cylinder 502, an actuator 505 made of a giant magnetostrictive material having a rod shape is movably accommodated. A piston 506 is detachably provided at the end of the actuator 505 facing the injection nozzle 504.

  Between the other end of the actuator 505 and the stopper 507 at the end of the cylinder 502, a spring 508 is interposed to bias the actuator 505 forward by the spring 508. A coil 509 is wound around the cylinder 502 at a position close to the piston 506.

  In the injection apparatus having the above-described configuration, an instantaneous magnetic field is applied to the super magnetostrictive material by instantaneously passing a current through the coil 509 to generate an instantaneous transient displacement due to an elastic wave at the end of the super magnetostrictive material. Let As a result, the liquid filled in the cylinder 502 can be ejected from the nozzle 504 as a minute droplet.

  Conventionally, as a liquid ejecting apparatus, an air pulse type dispenser as shown in FIG. 25 has been widely used. For example, the technology is introduced in “Automation Technology No. 93.25 No. 7”.

  The dispenser according to this system applies a constant amount of air supplied from a constant pressure source to the interior 601 of the container 600 (cylinder) in a pulsed manner, and a certain amount of liquid corresponding to the increase in pressure in the cylinder 600 is supplied from the nozzle 602. It is what is discharged.

  For the purpose of high-speed intermittent application, a dispenser as shown in FIG. 26 (hereinafter referred to as “jet type” for convenience) has been put into practical use. 550 is a micrometer, 551 is a spring, 552 is a piston seal member, 553 is a piston chamber, 554 is a heater, 555 is a needle, 556 is a coating material that flows toward the sheet portion, 557 is a dot-like flying from a dispenser It is a coating material. 27A and 27B are model diagrams showing the vicinity of the discharge portion 558 of FIG. 26, FIG. 27A shows the suction process, and FIG. 27B shows the discharge process. Reference numeral 559 denotes a spherical convex portion formed at the discharge side end of the needle 555, 560 denotes a discharge tip portion, 561 denotes a spherical concave portion formed in the discharge tip portion 560, and 562 denotes a discharge nozzle. Reference numeral 563 denotes a pump chamber formed by a spherical convex portion 559 and a concave portion 561.

  In FIG. 27A, which is the suction process, when the supply air pulse of the piston chamber 553 is ON, the needle 555 rises against the spring 551. At this time, the suction portion 564 formed between the spherical convex portion 559 and the concave portion 561 is opened, and the coating material 556 is filled into the pump chamber 563 from the suction portion 564. In FIG. 27B, which is the discharge process, when the air pulse is OFF, that is, when no air pressure is applied to the piston chamber 553, the needle 555 is lowered by the force of the spring 551. At this time, the suction portion 564 is in a shielding state, and the fluid in the pump chamber 563 is compressed in a sealed space excluding the discharge nozzle 562, so that high pressure is generated and the fluid flies and flows out.

  Development has been made to apply an ink jet method, which has been widely used as a consumer printer, as an industrial coating apparatus. FIG. 28 shows a conventional example (Japanese Patent Laid-Open No. 11-10866) of a head portion in an ink jet recording apparatus. 651 is a base, 652 is a diaphragm, 653 is a laminated piezoelectric element, 654 is an ink chamber, 655 is A common ink chamber, 656 is an ink flow path (throttle portion), 657 is a nozzle plate, and 658 is a discharge nozzle. When a voltage is applied to the piezoelectric element 653 that is a pressure applying means, the piezoelectric element 653 deforms the diaphragm 652 in the thickness direction, and the volume of the ink chamber 654 decreases. As a result, the fluid is compressed and the pressure in the ink chamber 654 increases, so that a part of the fluid passes through the ink flow path 656 and flows back to the common ink chamber 655 side, but the remainder is discharged from the nozzle 658 to the atmosphere. Is done.

Japanese Patent Laid-Open No. 11-10866

  In recent years, in the field of circuit formation, which is becoming more highly accurate and ultra-fine, or in the field of manufacturing process of liquid crystal, optical disk, organic EL, etc., the formation of electrodes and ribs and phosphor screens of picture tubes such as PDP and CRT. Many of the fluid materials to be used are high viscosity powder fluids. In order to replace the conventional method with a direct patterning method using a dispenser, a minute amount of high-viscosity powder fluid containing fine particles having an average outer diameter of several microns to several tens of microns, such as phosphors, conductive capsules, solders, The biggest issue is how to apply electrode materials and the like finely onto the target substrate with high reliability at high speed and high accuracy without clogging the flow path.

  Hereinafter, taking the phosphor layer forming step of the plasma display panel as an example, the following problems of the prior art will be described.

[1] Problems of screen printing method and photolithography method [2] Problems in direct patterning of phosphor layer using conventional dispenser technology First, the above [1] will be described.

(1) Structure of Plasma Display Panel FIG. 29 shows an example of the structure of a plasma display panel (hereinafter referred to as PDP). The PDP is roughly composed of a front plate 800 and a back plate 801. A plurality of sets of linear transparent electrodes 803 are formed on a first substrate 802 which is a transparent substrate constituting the front plate 800. The second substrate 804 constituting the back plate 801 is provided with a plurality of sets of linear electrodes 805 orthogonal to the linear transparent electrodes. The two substrates are opposed to each other with a bias rib 806 on which a phosphor layer is formed, and a discharge gas is sealed in the bias rib 806. When a voltage equal to or higher than the threshold value is applied between the electrodes of both substrates, a discharge occurs at a position where the electrodes cross each other, and the discharge gas emits light, and the emitted light can be observed through the transparent first substrate 802. An image can be displayed on the first substrate side by controlling the discharge position (discharge point). In order to perform color display by PDP, a phosphor that develops a desired color by ultraviolet rays emitted during discharge at each discharge point is formed at a position corresponding to each discharge point (barrier rib partition). In order to perform full color display, RGB phosphors are formed.

  The configurations of the front plate 800 and the back plate 801 will be described in a little more detail.

  The front plate 800 is formed of two or more pairs of linear transparent electrodes 803 in parallel on the inner surface side of a first substrate 802 made of a transparent substrate such as a glass substrate, using ITO or the like. A bus electrode 807 for reducing the line resistance value is formed on the inner surface of the linear transparent electrode 803. A dielectric layer 808 that covers the transparent electrode 803 and the bus electrode 807 is formed on the entire inner surface of the front plate 800, and an MgO layer 809 that is a protective layer is formed on the entire surface of the dielectric layer 808. .

  On the other hand, on the inner surface side of the second substrate 804 of the back plate 801, a plurality of linear address electrodes 805 orthogonal to the linear transparent electrodes 803 of the front plate 800 are formed in parallel by a silver material or the like. A dielectric layer 810 that covers the address electrodes 805 is formed on the entire inner surface of the back plate 801. On the dielectric layer 810, each address electrode 805 is isolated, and a barrier rib (partition) 806 having a predetermined height is provided between the address electrodes 805 in order to keep the gap distance between the front plate 800 and the back plate 801 constant. It protrudes and is formed. With this barrier rib 806, cells 811 are formed along each address electrode 805, and phosphors 812 for each color of RGB are sequentially formed on the inner surface thereof. PDPs having a cell structure include those having discharge points one by one in an independent cell as shown in FIG. 29 and those having a cell structure (not shown) partitioned by a partition for each row. In recent years, the “independent cell system” has been attracting attention as a system that can improve the performance of the PDP. The reason is that by surrounding the cell with four barrier ribs in a waffle shape, light leakage between adjacent cells can be prevented and the area of the light emitter can be increased. As a result, the feature of the “independent cell method” is that the light emission efficiency and the light emission amount (brightness) can be improved and a high-contrast image can be realized. The phosphor layer formed on the cell wall is generally thickened to about 10 to 40 μm in order to improve the color development. In order to form the RGB phosphor layer, each cell is filled with a phosphor coating solution and dried to remove volatile components, thereby forming a thick phosphor on the inner surface of the cell and simultaneously discharging. A space filled with sex gas is created. In order to form such a thick phosphor pattern, a coating material containing a phosphor is a high-viscosity paste-like fluid (phosphor paste) of several thousand mPas to tens of thousands mPas with a reduced amount of solvent. And conventionally applied to the substrate by screen printing or photolithography.

(2) Problems of conventional screen printing method When the conventional screen printing method is adopted, if the screen is enlarged, the stretch of the screen plate due to the tension is large, and it is difficult to align the screen printing plate with high accuracy over the entire screen. became. Further, when the phosphor material is filled, the material is placed on the top of the partition wall, and in the case of the “independent cell method”, there is a problem that leads to crosstalk between the barrier ribs. Therefore, it is necessary to take a measure such as introducing a polishing process in order to remove the material attached to the top portion of the partition wall. Moreover, the filling amount of the phosphor material changes due to the difference in the squeegee pressure, and the pressure adjustment is very delicate and often depends on the skill level of the operator. Therefore, it is not easy to obtain a constant filling amount in all the independent cells over the entire area of the back plate.

(3) Problems with conventional photolithography Conventional photolithography has the following problems. In this method, after the photosensitive phosphor paste is press-fitted into the cells between the ribs, only the photosensitive composition press-fitted into the predetermined cells is left by the exposure and development processes. Thereafter, through a firing step, organic substances in the photosensitive composition are eliminated to form a phosphor layer pattern. In this construction method, since the paste used contains phosphor powder, the sensitivity to ultraviolet rays is low, and it is difficult to make the thickness of the phosphor layer 10 μm or more. Therefore, there is a problem that sufficient luminance cannot be obtained.

  In addition, when adopting photolithography, an exposure and development process is essential for each color, but since the phosphor is contained in a high concentration in the paste coating layer, the loss of the phosphor due to development removal is large, Since the effective utilization rate of the phosphor is limited to less than 30%, there is a significant problem in terms of cost.

[2] Problems in Direct Patterning of Phosphor Layer Using Conventional Dispenser Technology (1) Problems of Air Nozzle Dispenser Air Nozzle Dispenser (used widely in fields such as circuit mounting) Attempts have been made in the past to apply to video tubes using FIG. In the case of the air nozzle type, since it is difficult to continuously apply a high-viscosity fluid at a high speed, fine particles are applied after being diluted with a low-viscosity fluid. In the case of applying a phosphor for a picture tube such as PDP or CRT, the particle size of the fine particles is, for example, 3 to 9 μm, and the specific gravity is about 4 to 5. In this case, since the single particle is heavy, there is a problem that when the flow of the fluid is stopped, the fine particles are immediately deposited inside the flow path. Moreover, the air-type dispenser has a drawback of poor responsiveness. This drawback is due to the compressibility of the air confined in the cylinder and the nozzle resistance when the air passes through a narrow gap. That is, in the case of the air method, the time constant of the fluid circuit determined by the cylinder volume and the nozzle resistance is large, and after applying the input pulse, until the fluid starts to be discharged and transferred onto the substrate, 0.07 to 0.1 A time delay of about a second must be expected.

  In the case of the ejection device using the piezoelectric material or the giant magnetostrictive material described above with reference to FIG. 24, this is a proposal for applying a fluid that does not contain powder, and responds to the above-mentioned problems related to the powder fluid application process. Is expected to be difficult. In addition, when applying fluid using instantaneous transient displacement due to elastic waves, the reservoir 503 must always be filled with fluid with no gap in a constant volume state. There is no description on how to supply fluid to the liquid reservoir 503 in order to compensate for the fluid that is consumed every moment.

(2) Issues with jet dispensers
In the case of the dispenser shown in FIG. 26, in terms of application speed, it is sufficiently faster than the conventional dispensers such as the air type and the thread groove type, and it is also possible to deal with high viscosity fluids. In this method, the fluid can be intermittently applied by flying from the nozzle in a state where the distance between the nozzle and the facing surface is sufficiently separated. Thus, the application method for causing the fluid to fly from the nozzle is difficult in the air method and the screw groove method in which a steep pulsed generated pressure cannot be generated.

  In this method, as described above, the spherical convex portion formed on the end portion of the needle 555 and the spherical concave portion formed on the discharge side are meshed with each other, so that the sealed space 563 excluding the discharge nozzle 562 is obtained. Is created, and the sealed space is compressed to generate a high pressure to cause the fluid to fly out.

  In this case, in the compression process, there is no gap in the suction part 564 between the relatively moving members (projections and recesses), and the phosphor fine particles having an average particle diameter of 3 to 9 μm are mechanically squeezed and destroyed. . Application to powder fluid such as phosphor is often difficult due to various problems that occur as a result, that is, clogging of the flow passage, deterioration of the sealing performance of the suction portion 564 due to wear of members, and the like.

  Another problem of the above method is ensuring the accuracy of the absolute amount of application per dot when presumed to be used continuously for a long time. Assuming the case where the phosphor is intermittently applied in the “independent cell” of the PDP described above, several tens of heads are required in consideration of production tact during mass production. In the dispenser, the application amount per dot is determined by the volume of the sealed space, that is, the stroke of the needle 555 and the sealing performance of the suction part 564. However, it is expected to be extremely difficult in practice to keep the stroke and absolute position of each needle 555 of the tens of dispensers and the sealing performance of the suction part 564 with wear constant for a long time without variation. The

(3) Problems of Inkjet System In the inkjet system shown in FIG. 28, the viscosity of the fluid is limited to 10 to 50 mPa · s due to the driving method and structural limitations, and cannot cope with a high viscosity fluid. The particle size of the powder contained in the fluid is limited to about 0.1 μm from the viewpoint of clogging.

  In order to draw a fine pattern using an inkjet method, a low-viscosity nanopaste in which particles having an average particle size of about 5 nm are covered with a dispersant and independently dispersed has been developed. It is assumed that a phosphor layer is formed on the inner wall of the barrier rib (partition wall) of the “independent cell” of the above-described PDP using this nanopaste. However, in the process of filling the phosphor coating liquid in each cell and drying it, the coating material containing the phosphor is used to thicken the phosphor layer of about 10 to 40 μm as described above. Uses a paste-like fluid having a high viscosity and a reduced amount of solvent. A low-viscosity nanopaste that can only have a low phosphor content cannot form a phosphor layer having a predetermined thickness because the absolute amount of the phosphor is insufficient. Also, in order to obtain high brightness, it is usually best to use phosphor particles with a particle size on the order of several microns. However, the fact that the phosphor particle size cannot be easily changed at this stage is also a major problem with inkjet systems. It is.

  To summarize the above considerations, there is no method that can replace the screen printing method and photolithography method, for example, the direct patterning method that realizes the formation of an independent cell phosphor layer of PDP at this stage. .

  Hereinafter, the past proposal concerning the intermittent application dispenser of the present inventor will be described a little. In order to meet various recent demands related to the application of a minute flow rate, the present inventor, in Japanese Patent Application No. 2000-188899, gives a relative linear motion and a rotational motion between a piston and a cylinder. Proposed a coating method that controls the discharge rate by changing the relative gap between the fixed side and the rotating side using linear motion, and is pending as a "fluid supply device and fluid supply method" .

  In the above proposal, a thrust dynamic pressure seal is formed on the relative movement surface of the discharge-side end surface of the piston and its opposing surface, and the fluid is controlled to be shut off by the dynamic pressure sealing effect when the gap between the opposing surfaces is narrowed. .

  Japanese Patent Application No. 2000-208072 proposes a dispenser in which a positive displacement pump is configured by independently driving a piston and a cylinder storing the piston using two independent linear motion devices.

  Further, a theoretical analysis is performed on the dispenser structure disclosed in Japanese Patent Application No. 2000-188899, and an intermittent discharge method using a squeeze pressure generated by abruptly changing a gap between a piston end surface and a relative moving surface thereof, and A device is being proposed (Japanese Patent Application No. 2001-110945). This squeeze pressure is known as the hydrodynamic effect of the hydrodynamic bearing, but in order to use this squeeze pressure, the gap between the piston end face and its opposing face is narrow, for example, set to 20-30 μm or less. There is a need to.

  The present invention proposes a coating principle based on a new discovery that was not disclosed in the above-mentioned proposal. In other words, as a result of rigorous theoretical analysis assuming that the coating fluid is a viscous fluid, the piston end surface and its opposite surface are affected by the interaction between the pump characteristics of the fluid supply source and the flow rate fluctuation caused by abruptly changing the piston position. It has been found that even when the gap between them is sufficiently wide, a high generated pressure equivalent to or higher than the squeeze effect (that is, the secondary squeeze pressure) can be obtained.

Accordingly, the present invention proposes a fluid ejection how using the secondary squeeze pressure. If this discharge principle is used, the management of the gap between the piston end face and the opposed face may be simple, the structure is simple, and the total discharge amount per dot can be set by, for example, the rotational speed of the fluid supply pump. Therefore, practical handling ease, providing a high flow rate accuracy per dot, and a fluid ejection how the intermittent fluid delivery ultrafast and ultra small amount can be achieved with high reliability with respect to the liquid powders It is an object of the present invention.

  In order to achieve the above object, the present invention is configured as follows.

According to the first aspect of the present invention, the fluid is supplied from the fluid supply device to the gap while the two members are relatively moved in the gap direction of the gap formed between the two opposing surfaces of the two members. And a fluid discharge method for intermittently discharging the fluid by utilizing a pressure change caused by changing the gap and adjusting a discharge amount of the fluid per dot by a pressure and a flow rate characteristic of the fluid supply device. ,
In the coating method of intermittently applying the gap between the two opposing surfaces of the two members that move relative to each other with an amplitude h ₁ , the opposing surfaces of the two members with an amplitude h ₂ that is larger than the amplitude h ₁ . After the discharge is cut off by increasing the gap, the fluid is intermittently applied a plurality of times with the amplitude h ₁ so that the central value of the gap after the interruption gradually becomes equal to the central value of the gap immediately before the interruption. Discharge method is provided.

More fluid ejection how using the present invention, the following effects can be obtained.

      (1) Applicable to high viscosity fluid of several thousand to several tens of thousands mPa · s (cps).

      (2) No clogging occurs even with a discharge material containing a powder diameter of several μm or more.

      (3) Intermittent fluid discharge can be performed with a short cycle of msec order or less.

      (4) The fluid to be ejected can fly at a distance of 0.5 to 1.0 mm or more from the ejection nozzle.

      (5) The amount of fluid discharged per dot can be ensured with high accuracy.

(6) Multi-heading is easy and the structure is simple.
For example, if the present invention is used for phosphor coating of PDP, CRT display, electrode formation, surface mount dispenser, microlens molding, etc., the advantages can be fully exhibited, and the effect is tremendous.

  Before continuing the description of the present invention, the same parts are denoted by the same reference numerals or the same names in the accompanying drawings.

  Embodiments according to the present invention will be described below in detail with reference to the drawings.

  FIG. 1 is a model diagram showing a first embodiment of the present invention. Reference numeral 1 denotes a piston which is housed so as to be movable in the axial direction with respect to the housing 2. Reference numeral 3 denotes a sleeve that accommodates the outer peripheral portion of the piston 1, and is accommodated so as not to move in the axial direction with respect to the housing 2 on the fixed side but to be movable in the rotational direction.

  The piston 1 is driven by an axial drive device (arrow 4), and the sleeve 3 is driven by a rotation transmission device (arrow 5). Reference numeral 6 denotes a thread groove formed on the relative movement surface of the sleeve 3 and the housing 2 (a blackened portion in FIG. 1), and 7 denotes a fluid suction port. In this embodiment, a thread groove pump is used for the fluid supply device.

  8 is an end face of the piston 1, 9 is a fixed side facing surface thereof, 10 is a discharge nozzle formed at the center of the fixed side facing surface 9, and 11 is an opening portion of the discharge nozzle 10 formed on the fixed side facing surface 9. is there. The piston end surface 8 and the fixed-side facing surface 9 are two surfaces that relatively move in the gap direction.

  A coating fluid 12 is supplied between the sleeve 3 and the housing 2. Reference numeral 13 denotes a discharge chamber end (piston outer peripheral portion) formed between the lower end of the sleeve 3 and the housing 2, and the discharge chamber end 13 is constantly fluidized by a thread groove pump as an example of a fluid supply device. Is supplied.

  The axial drive device 4 (the specific structure is not shown) is provided between the piston 1 and the housing 2, and changes the axial relative position between the members 1 and 2. The axial direction drive device 4 is constituted by a piezoelectric actuator (100 in FIG. 9) and the like as described later in the first embodiment, for example. The axial driving device 4 can change the gap h between the piston end surface 8 and the opposed surface 9.

  In the present embodiment, the configuration conditions are different from the previous proposal (Japanese Patent Application No. 2001-110945) as follows.

(I) When the minimum value of the clearance h on the piston end surface is h = h _min , h _min is sufficiently large in the present embodiment, for example, h _min = 150 μm.

    (Ii) The thread groove pump is designed to be close to the metering pump, and its internal resistance Rs is sufficiently large.

  When the gap h is changed at a high frequency, a discharge chamber 14 (piston end surface portion) which is a gap portion between the piston end surface 8 and the opposed surface 9 is obtained by a secondary squeeze effect which will be described later newly found in the present proposal. Fluctuating pressure is generated.

  Further, a portion located at the center portion of the piston end surface 8 at 15 is called an upstream side of the discharge nozzle 10, and a gap portion formed by the screw groove and the housing 2 is called a screw groove chamber 16. A constant amount of fluid is continuously supplied to the discharge chamber 14 by a thread groove pump.

In the above application example of the present invention, the continuous flow (Analog) supplied from the pump is A / D converted into the intermittent flow (Digital) using the second-order squeezing effect, so that the piston end surface and the opposed surface are separated. This is based on the idea that the fluid can be intermittently applied at a high speed while keeping the gap h sufficiently large.
[1] Theoretical analysis (1) Derivation of basic equation In order to clarify the principle and effect of the present invention, a basic equation of a squeeze pump (tentative name) is derived.

  When a viscous fluid is present in a narrow gap between opposing planes and the gap distance changes with time, the fluid pressure is the Reynolds equation in the following polar coordinate with a squeeze action term Is obtained by solving

（１）式において、Ｐは圧力、μは流体の粘性係数、ｈは対向面間の隙間、ｒは半径方向位置、ｔは時間である。また、右辺が、隙間が変化するときに発生するスクイーズアクション効果をもたらす項である。図２に、スクイーズポンプ部の寸法関係を示す。なお、各記号における添字ｉは、図１における吐出ノズルの開口部１１の位置における値、添字０は吐出室端部１３（ピストン外周部）における値であることを示す。

In the equation (1), P is pressure, μ is a viscosity coefficient of fluid, h is a gap between opposing surfaces, r is a radial position, and t is time. The right side is a term that provides a squeeze action effect that occurs when the gap changes. FIG. 2 shows the dimensional relationship of the squeeze pump unit. The subscript i in each symbol indicates the value at the position of the opening 11 of the discharge nozzle in FIG. 1, and the subscript 0 indicates the value at the discharge chamber end 13 (piston outer peripheral portion).

として、（１）式の両辺を積分する。

Then, both sides of the equation (1) are integrated.

以下、未定定数ｃ_１，ｃ_２を求める。圧力勾配と流量の関係は

Hereinafter, undetermined constants c ₁ and c ₂ are obtained. The relationship between pressure gradient and flow rate is

ｒ＝ｒ_ｉ（図２参照）における流量Ｑ＝Ｑ_ｉとして、（２）式と（４）式からｃ_１を求めると

Assuming that flow rate Q = Q _{i at} r = r _i (see FIG. 2), c ₁ is obtained from equations (2) and (4).

さて、吐出室端部１３と、流体吸入口７間の流体抵抗Ｒｓが無視できない場合、吐出室端部１３（図２におけるｒ＝ｒ_０の位置）における圧力Ｐ＝Ｐ_０は

If the fluid resistance Rs between the discharge chamber end 13 and the fluid suction port 7 cannot be ignored, the pressure P = P ₀ at the discharge chamber end 13 (the position of r = r ₀ in FIG. 2) is

流体補給装置にねじ溝ポンプを用いた場合、流体抵抗Ｒｓはねじ溝ポンプの内部抵抗に等しい。また、上式においてＰ_Ｓ０は供給源圧力であり、ねじ溝ポンプの最大発生圧力Ｐｍａｘと、材料をねじ溝に供給するためのエアーによる補給圧力Ｐｓｕｐの和（Ｐ_Ｓ０＝Ｐｓｕｐ＋Ｐｍａｘ）に相当する。Ｑ_０はｒ＝ｒ_０における流量であり、（４）式から、

When a thread groove pump is used in the fluid supply device, the fluid resistance Rs is equal to the internal resistance of the thread groove pump. In the above equation, P _S0 is the supply source pressure, which corresponds to the sum of the maximum generated pressure Pmax of the thread groove pump and the replenishment pressure Psup by air for supplying the material to the thread groove (P _S0 = Psup + Pmax). Q ₀ is the flow rate at r = r ₀ , and from equation (4),

（３）式、（５）〜（７）式から未定定数ｃ_２が求まる。

(3), (5) - (7) is obtained undetermined constant _{c 2} from the equation.

ここで、任意の位置ｒにおける圧力Ｐを次のようにおく。

Here, the pressure P at an arbitrary position r is set as follows.

但し、

However,

吐出ノズルの開口部：ｒ＝ｒ_ｉ（図１の１１）において、Ｐ_ｉ＝Ａ＋ＢＱ_ｉとする。吐出ノズルの流体抵抗をＲ_ｎすれば、吐出ノズルを通過する流量はＱ_ｎ＝Ｐ_ｉ／Ｒ_ｎである。

At the opening of the discharge nozzle: r = r _i (11 in FIG. 1), P _i = A + BQ _i . If the fluid resistance of the discharge nozzle _{R n,} flow through the discharge nozzle is _{Q n = P i / R n} .

From the continuity of the flow, Q _i = Q _n , and the discharge nozzle upstream side pressure P _i (pressure at the point 15 in FIG. 1) is obtained as follows. A _i, the _{B i} is the value of A, B when the r = _{r i} in (10). Hereinafter referred discharge nozzle upstream pressure _{P i} and the discharge pressure _{P i.}

ここで、第１次スクイーズ圧力Ｐｓｑｕ１、第２次スクイーズ圧力Ｐｓｑｕ２を次のように定義する。

Here, the primary squeeze pressure Psqu1 and the secondary squeeze pressure Pscu2 are defined as follows.

  The primary squeeze pressure Psqu1 is due to a well-known squeeze effect generated between the piston end surface 8 and the opposing surface 9 by abruptly changing the gap between the piston end surface 8 and the relative moving surface 9. The narrower the pressure, the greater the pressure.

The present invention has found a method for generating the secondary squeeze pressure Pscu2 and a method for applying this action to, for example, ultra-high-speed intermittent coating, and the principle thereof is as follows. By rapidly changing the gap between the piston end surface and its relative movement surface, a flow rate fluctuation occurs between the end surface of the piston and the fluid supply source. This flow rate fluctuation corresponds to a change in the volume of the discharge chamber 14 (piston end surface portion) when the gap is changed. For example, when the volume is reduced, if the fluid resistance of the discharge nozzle is large, the fluid having no escape on the discharge side flows back to the thread groove pump side. As a result, a pressure Pscu2 proportional to the internal resistance Rs of the thread groove pump is generated.
From the equations (11) and (12), the pressure Pi upstream of the discharge nozzle is arranged.

吐出ノズルを通過する流量Ｑ_ｉは

The flow rate Q _i passing through the discharge nozzle is

吐出ノズルのノズル半径をｒ_ｎ、ノズル長さをｌ_ｎとおくと、吐出ノズル抵抗は

The nozzle radius of the discharge nozzle r _n, if the nozzle length is denoted by l _n, the discharge nozzle resistance

また、Ｒ_ｐは吐出ノズルの開口部（図１の１１）とピストン外周部（図１の吐出室端部１３）の間の流体抵抗である。

_Rp is the fluid resistance between the opening (11 in FIG. 1) of the discharge nozzle and the piston outer periphery (discharge chamber end 13 in FIG. 1).

Ｒ_ｓは、前述したように、ピストン外周部（図１の吐出室端部１３）と供給源側（吸入口７）の流路間の流体抵抗（ねじ溝ポンプの場合は内部抵抗）である。

As described above, R _s is the fluid resistance (internal resistance in the case of a thread groove pump) between the flow path on the piston outer peripheral portion (discharge chamber end portion 13 in FIG. 1) and the supply source side (suction port 7). .

(2) Equivalent circuit model Based on the above analysis results, the relationship between the pressure source and the load resistance is represented by an equivalent electric circuit model as shown in FIG.

(3) give the piston end face and the minimum gap hmin is sufficiently large. Table 1 Conditions and the piston input waveform of FIG. 4 between the opposing surfaces, the pressure P _i of the opening of the discharge nozzle, the expression (11) The results obtained using this are shown in FIG. A section of 0 ≦ t ≦ 2.0 msec corresponds to one cycle as the intermittent ejection device.

  The piston input waveform was evaluated by changing the stroke in three cases (hst = 10, 20, 30 μm) while the minimum gap between the piston end surface and the opposed surface remained constant (hmin = 150 μm).

5, in both cases of stroke, the pressure has a waveform that varies around the _P ic = 3.5MPa.

FIG. 6 shows the analysis result of the flow rate Q _i passing through the discharge nozzle. When the discharge nozzle resistance is Rn, the flow rate is Q _i = P _i / Rn. Flow rate: Qi, like the pressure waveform, has a waveform that fluctuates around Qic = 49 mm ³ / sec, although the amplitude varies depending on the stroke.

That is, it can be seen that the average flow rate does not depend on the size of the stroke of the piston, but is determined by the operating point (A in FIG. 8) determined by the thread groove pump characteristics and the discharge nozzle resistance.
The reason for this is as follows. In the equation (11), if h → ∞, the first squeeze pressure Psqu1 → 0 and Rp → 0, so the following equation is obtained. However, P _{s0 ≈P max} and Rs = P _max / Q _max .

（１７）式の第２項は、ピストン端面８とその対向面９で形成されるピストン端面部１４の幾何学的な容積変化分に相当する。変位ｈの時間微分ｄｈ／ｄｔは正と負を交互に持つ周期関数であり、１周期における時間積分値は０である。
すなわち、第２次スクイーズ圧力Ｐｓｑｕ２は、ねじ溝の連続流量（Analog）を間欠流量（Digital）に変えるＡ／Ｄコンバータとしての役割を担うのである。

The second term of the equation (17) corresponds to a geometric volume change of the piston end surface portion 14 formed by the piston end surface 8 and the opposed surface 9 thereof. The time derivative dh / dt of the displacement h is a periodic function having positive and negative alternately, and the time integral value in one period is zero.
That is, the secondary squeeze pressure Psqu2 plays a role as an A / D converter that changes the continuous flow rate (Analog) of the thread groove to the intermittent flow rate (Digital).

In FIG. 8, (I) shows the relationship between the pressure and the flow rate of the thread groove pump when the rotational speed N = 460 rpm (referred to as pressure / flow rate characteristics), and the maximum pressure: Pmax = 10 MPa (at Q = 0), the maximum The flow rate is Qmax = 77.35 mm ³ / sec (at P = 0). (III) is the fluid resistance of the discharge nozzle, and the intersection of (I) and (III) is the thread groove pump operating point A (Pic = 3.5 MPa, Qic = 49 mm ³ / sec).

  Table 2 shows an example of the specification of the thread groove that can obtain the characteristics of the pump.

The X-axis pressure in the graph of FIG. 8 is defined as a differential pressure (= P ₂ −P ₁ ) between the pressure P ₂ at the discharge chamber end 13 and the pressure P ₁ near the suction port 7. The thread groove pump can transport the largest flow rate when the differential pressure is minimum, that is, the piston 1 is raised, and the pressure at the lower end (discharge chamber end 13) of the thread groove 6 is P ₂ = −0. The time is 1 MPa (absolute vacuum). Therefore, in the graph of FIG. 8, the maximum transport amount of the pump is a flow rate when P = −0.1 MPa: Q≈80 mm ³ / sec, but since there is no large error, for convenience, P = 0 MPa (atmospheric pressure). Qmax = 77.35 mm ³ / sec is the maximum flow rate.

(4) Improvement of sharpness Now, when a fluid mass is repeatedly hit on the substrate while moving the discharge head and the substrate relative to each other, the waveform of the discharge pressure becomes a negative pressure just before the start of coating, and a sharp peak immediately after that. It is preferable that a positive pressure is generated and a negative pressure is obtained again. Due to the generation of negative pressure after discharge, the fluid at the tip of the discharge nozzle is again sucked into the nozzle and separated from the fluid on the substrate or the fluid in flight. That is, an intermittent application that is extremely sharp can be realized by a cycle of “negative pressure → steep positive pressure → negative pressure”.

None of the pressure waveforms in FIG. 5 satisfy P _i > 0, which does not satisfy the conditions for sharp and intermittent application. When the maximum value of the time derivative (piston speed) dh / dt of the displacement h is Vmax, the condition that the pressure waveform has a section where the negative pressure P _i <0 is obtained from the equation (17):

（１８）式を満足させるＱｍａｘを得るには、流体補給装置にたとえばねじ溝ポンプを用いる場合には回転数を変えればよい。Ｑｍａｘを小さくする程、スクイーズポンプ部の容積増加に供給量が追従できず、負圧が発生する時間は長くなる。

In order to obtain Qmax that satisfies the equation (18), for example, when a thread groove pump is used as the fluid supply device, the rotational speed may be changed. As Qmax is decreased, the supply amount cannot follow the increase in the volume of the squeeze pump unit, and the time during which negative pressure is generated becomes longer.

When the stroke is hst = 30 μm and the rotational speed of the thread groove is changed from N = 460 → 300 rpm and the maximum flow rate is reduced to Qmax = 77.35 → 50 mm ³ / sec under the conditions shown in Table 1, the discharge pressure is The waveform is shown in FIG. 7 in comparison with the case of N = 460 rpm. FIG. 8 shows the pressure / flow rate characteristics of the thread groove pump when the rotation speed N is 300 rpm. At this time, the operating point of the pump shifts from A to B. In FIG. 7, when N = 300 rpm (Qmax = 50 mm ³ / sec), Formula 18 is satisfied, and the waveform of the discharge pressure becomes a negative pressure immediately before the start of discharge, and again after the steep positive pressure is generated, the negative pressure again It has become. The reason why the negative pressure is generated is that, as described above, before and after the peak pressure is generated, the volume change of the piston end surface exceeds the maximum flow rate Qmax that can be supplied by the thread groove pump.

The minimum value of the discharge pressure is P _i = −1.4 MPa. This is because the analysis model assumes incompressibility, and when the atmospheric pressure is P _i = 0.0 MPa (gauge pressure). , A pressure of −0.1 MPa or less does not actually exist.

  The level of the negative pressure generation may be adjusted from the conditions of the process to be applied and the characteristics of the coating material, for example, spinnability (hardness of the coating line flowing out from the nozzle).

The application example of the embodiment of the present invention described above uses the secondary squeeze pressure by suppressing the generation of the primary squeeze pressure as much as possible by setting the gap between the piston end face and the opposing face sufficiently large. Thus, the continuous flow supplied from the fluid supply source is A / D converted into an intermittent flow and applied intermittently. In this case, the application amount per dot does not depend on the stroke of the piston, and is determined only by the pressure flow characteristics of the pump, which is an example of the fluid supply device, and the discharge nozzle fluid resistance.
Therefore,
(I) The discharge amount per dot is constant. (Ii) The period is constant. With respect to the coating process that requires the above (i) and (ii) at the same time, the present coating method and apparatus are extremely effective means (method and apparatus). )I will provide a.

For example, this is effective when R, G, and B phosphors are intermittently applied in independent cells (box-type ribs) on the back plate of a plasma display panel (PDP) that performs color display. In the case of a PDP, as will be described later with reference to the embodiment of FIG. 15, the independent cells are arranged on the panel in a grid pattern with high accuracy and geometrically symmetrical. In this case, the present dispenser capable of driving a constant amount of material into the independent cell at the same time interval at high speed can exhibit an unparalleled ability.
That is, the application example of the above embodiment of the present invention pays attention to the “geometric symmetry” of the object to be applied, and replaces this symmetry with “periodicity of time”, thereby applying 0.002 seconds. The following ultra-high-speed intermittent application is realized.

  Incidentally, when applying solder, an adhesive, etc. to a circuit board, for example, in circuit formation, the application time interval is usually random. Incidentally, in the case of a conventional air-type dispenser, the coating cycle is at most on the order of 0.05 to 0.1 seconds.

[2] Specific Embodiment FIG. 9 shows a first specific example of a dispenser structure to which the present invention is applied, and an axial drive device is provided to a central shaft (piston) penetrating a hollow outer peripheral shaft. The structure of the case is shown. Reference numeral 100 denotes a first actuator which is an example of an axial drive device, and uses a giant magnetostrictive element, a piezoelectric element, an electromagnetic solenoid, or the like. In the first embodiment, a laminated piezoelectric actuator having excellent responsiveness and capable of obtaining a high response and a large generated load is used.

  A piston 101 is driven in the axial direction by the piezoelectric actuator 100 that is a first actuator. By driving the piston 101, the squeeze pressure described above is generated on the discharge side end surface (discharge chamber) of the piston 101. The first actuator 100 is disposed inside the upper cylinder 102. Reference numeral 103 denotes a motor as a second actuator, which gives a relative rotational movement between a sleeve 104 that houses the piston 101 and an intermediate cylinder 105. Reference numeral 106 denotes a rotor of the motor 103, and reference numeral 107 denotes a stator.

  Reference numeral 108 denotes a thread groove that is formed on the outer surface of the sleeve 104 and is an example of a fluid supply device that pumps fluid to the discharge side. Between the sleeve 104 and the lower cylinder 109, a thread groove pump chamber 110 for obtaining a pumping action by relative rotation of the sleeve 104 and the lower cylinder 109 is formed.

  Further, a suction hole 111 communicating with the thread groove pump chamber 110 is formed in the lower cylinder 109. Reference numeral 112 denotes a discharge nozzle attached to the lower end portion of the lower cylinder 109, and a discharge hole 113 is formed in the center. Reference numeral 114 denotes a discharge side thrust end face of the sleeve 104. 115 and 116 are ball bearings for supporting the sleeve 104.

  In addition, 117 is a flange portion provided on the upper portion of the piston 101, 118 is a disk portion provided on the piezoelectric actuator 100, 119 is a displacement sensor for detecting the axial position of the piston 101, and 120 is a shaft for the flange portion 117. It is a hinge part formed in order to elastically deform in the direction. The dimensions of each member are determined so that an appropriate preload is applied to the piezoelectric actuator 100 by elastic deformation of the hinge portion 120.

  In the first embodiment, the piston 101 (central axis) is configured to penetrate the sleeve, and the piston 101 and the sleeve 104 are driven by separate actuators. That is, the piston 101 is driven only in the axial direction, and the sleeve 104 is driven only in the rotational direction.

  As already proposed in Japanese Patent Application No. 2000-188899, the present inventor uses a magnetostrictive element (or moving magnet) to give a linear motion to the shaft and a rotational motion to the shaft by a motor ( If the actuator has a two-degree-of-freedom actuator structure, the central shaft and the sleeve can be integrated into one shaft.

  FIG. 10 shows a second embodiment of the present invention, and shows a case where a thread groove pump which is an example of a fluid supply device and a piston are separated. A main shaft 51 is housed so as to be movable in the rotational direction with respect to the housing 52. The main shaft 51 is rotationally driven by a rotation transmission device (arrow 53) such as a motor. Reference numeral 54 denotes a thread groove formed in the relative movement surface of the main shaft 51 and the housing 52 (a blackened portion in FIG. 10), and 55 denotes a fluid inlet.

56 is an axial drive device for moving the piston 57 in the axial direction (arrow 58), 59 is an end surface of the piston 57, 60 is a fixed-side facing surface, and 61 is a discharge nozzle mounted on the housing 52. The piston end surface 59 and the fixed-side facing surface 60 form two surfaces (discharge chambers) that relatively move in the gap direction. 62 is a main shaft end, 63 is a piston outer peripheral portion, and 64 is a flow passage connecting the main shaft end 62 and the piston outer peripheral portion 63. A coating fluid 65 is always supplied to the piston outer peripheral portion 63 through a flow passage 64 by a thread groove pump 54 which is an example of a fluid supply device. Reference numeral 68 denotes a discharge chamber formed between the end surface of the piston 57 and the fixed-side facing surface 60. The axial drive device 56 (specific structure is not shown) changes the relative position in the axial direction between the piston 57 and the housing 52 on the fixed side. The axial direction driving device 56 changes the gap h between the piston end surface 59 and the facing surface 60 in the same manner as in the first embodiment of FIG. Similarly, the configuration conditions of the thread groove pump 54 and the piston 57 are as follows:
(I) When the minimum value of the gap h between the piston end surface and the opposed surface is h = h _min , h _min is sufficiently large, for example, h _min > 50 μm.

      (Ii) The thread groove is designed close to the metering pump, and its internal resistance Rs is sufficiently large.

  As shown in the second embodiment, if the application unit is configured by separating a pump unit 66 which is an example of a fluid replenishing device and a portion (piston driving unit 67) for driving a piston by an axial driving device, the above-described embodiment will be described. Depending on the target to be applied, the entire apparatus can be greatly simplified. For example, if the piston drive unit is configured using a piezoelectric element in the axial direction drive device, the piezoelectric element actuator can be made sufficiently compact.

  It supplements about the pressure generation principle of this invention. Even if the secondary squeeze pressure is not used, pressure can be generated if a “throttle” is formed in the flow path between the “piston having means (or device) for changing the gap” and the “fluid supply source”. . For example, in the case of the conventional inkjet method, reference numeral 656 in FIG. 28 corresponds to the stop. In the conventional ink jet compression / discharge process, this restriction contributes to pressure generation. However, in the suction process, this throttle serves as a fluid resistance when supplying fluid from the supply source to the piston portion (discharge chamber). Due to this fluid resistance, when a high-viscosity fluid having poor fluidity is intermittently applied at high speed, the fluid cannot be filled in the piston portion in a short time, which becomes the limit of the intermittent application cycle.

  In the second embodiment of the present invention, a thread groove pump is used, and the thread groove pump can transport the largest flow rate when the differential pressure is minimum, that is, during the suction process in which the piston is raised. The maximum flow rate Qmax of the thread groove can be freely selected regardless of the fluid viscosity, depending on the thread groove specifications, rotation speed, and the like. Therefore, in the dispenser of the above embodiment of the present invention, the restriction that the fluid filling time during the suction process gives to the intermittent period is avoided.

  The role of the thread groove pump in the present invention may be considered to be a “one-way diode” in which fluid is likely to flow in the forward direction (discharge side), but backflow is difficult.

[3] In the case of a multi-head (1) Problems of making a multi-head The above-described dispenser embodiment or example is a single head in which a pump unit and a piston drive unit, each of which is an example of a fluid replenishing device, are configured in pairs. Met.

  Hereinafter, a measure for further improving the production tact of the head of the present invention will be described.

As described above, a direct drawing method (direct patterning) using a dispenser is realized in order to solve the above-described problems related to the formation of the phosphor layer on the PDP, that is, the screen printing method and the photolithography method. There is a strong demand for it to be allowed. However, even when the phosphor layer is formed on the panel surface using a dispenser, a production tact equivalent to the screen printing method is desired.
When the present invention is applied to the process of intermittently applying a phosphor in an independent cell, the above-described conditions for the application process, (i) the discharge amount per dot is constant, (ii) the period is constant, (iii) ultra-high speed In addition to coating, the following conditions are required.

(Iv) Being a multi-head (v) Being able to correct the flow rate of each head The reason for the above (v) will be described below. As shown in the second embodiment, if a coating device is configured by separating a pump, which is an example of a fluid replenishing device, and an axial driving device for driving a piston, a plurality of piston drives from a set of pump units. An application head having a multi-nozzle can be realized by branching and replenishing fluid to the part.

  In the perspective view of FIG. 11, reference numeral 200 denotes a pump unit that is an example of a fluid supply device, and 201, 202, and 203 denote piston drive units A, B, and C each including a piezoelectric actuator and a piston. Reference numeral 204 denotes a housing in which a flow passage (corresponding to 64 in FIG. 10) connecting the pump unit 200 and the piston drive unit is formed.

  FIG. 12 shows an equivalent circuit model in the case of a multi-head.

  Pscu11, Pscu12, and Psqu13 are primary squeeze pressures of the piston drive units, Rp1, Rp2, and Rp3 are fluid resistances in the radial direction of the piston end surfaces, and Rn1, Rn2, and Rn3 are nozzle resistances. The magnitudes of Rp1 to Rp3 are inversely proportional to the cube of the gap h, as shown by equation (16). Rp1 to Rp3 are “variable resistors” that can adjust the flow rate without disassembling the coating apparatus.

  In the above-described embodiment, the gap h between the piston end surface and the opposing surface is set to be sufficiently large so as to suppress the generation of the primary squeeze pressure as much as possible. For example, the configuration is determined only by the number of revolutions). When fluid is branched and replenished from a set of pump units to a plurality of piston drive units, the flow rate supplied from the pump unit is as long as the dimensional accuracy and flow path resistance of each piston drive unit can be configured to be exactly the same. It is distributed equally. However, it is preferable that the flow rate accuracy can be finely adjusted for a coating target such as a display that requires accuracy of a coating amount of several percent.

(2) Flow rate adjustment method Here, we return to the basic equation (Equation 11) derived by this study again.

The graph of FIG. 13 shows the comparison of the discharge pressure characteristics using the equation (11) for the case where the minimum clearance of the piston h _min = 15 μm and h _min = 150 μm under the condition of the rotational speed N of the thread groove = 300 rpm. It is a thing. Contrary to the intuitive expectation, the comparison between the two results in a surprising result that the amplitude of the discharge pressure increases as the minimum piston gap h _min increases. The discharge amount per dot is larger when h _min = 150 μm.

When the minimum clearance h _{min of the} piston increases, the primary squeeze pressure Psqu1 → 0, but at the same time, the thrust fluid resistance Rp → 0 between the piston end surface and the opposed surface thereof, so the partial pressure ratio (= Rn / (Rs + Rp + Rn) (See equation 13)
In this analysis conditions, Psqu1 → 0 become than impact, because towards the effect of the partial pressure ratio is increased larger, the amplitude of the pressure Pi with increasing h _min is increasing.

The graph of FIG. 14B is obtained by calculating the discharge amount per dot with respect to the minimum piston gap h _min under the condition of N = 300 rpm in FIG. 14A. When it exceeds h _min = near 0.1 mm, the discharge amount Qs per dot converges to a constant value Qs → Qse without depending on h _min . As described above, the discharge amount convergence value Qse is determined by the pressure flow characteristics of a pump, which is an example of a fluid supply device, and the pump load (discharge nozzle fluid resistance Rn) regardless of the stroke of the piston, the minimum gap, and the like. Determined by.

  From the knowledge obtained from the above analysis, the flow rate adjustment of each head may be selected from the following.

(I) When the variation in the flow rate between the heads is large, the piston is within a region that is strongly influenced by the primary squeeze pressure, that is, in a range of 0 <h _min <h _x where the gradient of the discharge amount with respect to the gap is steep. The minimum gap h _min is set.

(Ii) When it is desired to secure the coating amount per dot with extremely high accuracy, the minimum gap h _{min of the} piston is set in the vicinity of h _min ≈h _x where the gradient of the discharge amount with respect to the gap is smooth.

The h _x is a value of the intersection of the envelope (I) of the Qs curve with respect to h _min in the region of 0 <h _min <h _x and the straight line (II) of Qs = Qse. In other words, h _x is the intersection of the tangent at h ₀ of the total discharge amount Qs per dot and the tangent at the point where the total discharge amount Q _s per dot is approximately constant.

The displacement of the piston can be arbitrarily controlled by providing a displacement sensor for detecting the absolute position of the piston and performing closed loop control. However, when an electrostrictive element such as a piezoelectric element or a giant magnetostrictive element is used, there is a stroke limit (0 to several tens of microns). Therefore, adjustment of the minimum gap h _min of the piston is performed by a mechanical method and electronic control. A combination of the methods described above may be used.

  For example, first, the position of the piston is mechanically coarsely positioned, and then the piston position of each head may be corrected again using electronic control based on the flow rate measurement data.

Regardless of which of the above (i) and (ii) is used for flow rate adjustment, if used together with the method for setting the output flow rate of the source pump, the flow rate adjustment is performed where the gap between the piston end surface and the opposed surface is sufficiently large. it can. For example, if the flow rate is too large and the minimum gap h _{min of the} piston must be set small, h _min can be set large by reducing the rotational speed of the thread groove pump. This point is advantageous when dealing with powdered fluid, as will be described later.

The above-mentioned measures used for correcting the flow difference between the heads of the multi-head can be applied to the case of a single head. In the case of a single head, the minimum clearance of the piston is set in the vicinity of h _min ≈ hx or in the range of 0 <h _min <hx, and instead of changing the motor speed of the pump, adjusting h _{min allows} high-speed flow rate control. Can do. The responsiveness of the rotational speed control of the motor is at most 0.01 to 0.05 second level, and there is a limit, but the control responsiveness of the piston driven by the electromagnetic strain element can be up to 0.001 second or less.

Instead of adjusting the flow rate with the minimum piston gap h _min , the flow rate may be adjusted with the average value or the center value of the piston input displacement waveform.

When the minimum gap of the piston is set in the vicinity of h _min ≈hx or in the range of 0 <h _min <hx, in order to improve the sharpness of intermittent application, the time differential of the gap h has the maximum value in Equation 13. When the primary squeeze pressure Psqu1 = Psqu10 and the secondary squeeze pressure Pscu2 = Pscu20, the motor rotation speed, piston stroke, intermittent frequency, etc. may be set such that Ps0 + Pscu10 + Pscu20 <0.

(2) Application device and application method As shown in the perspective view of FIG. 11, a plurality of piston drive units are provided for one set of pump units as an example of a fluid supply device. As a result, the entire apparatus can be greatly reduced in size. Normally, there is a limit to downsizing the pump unit, which is an example of a fluid replenishing device, but the piston drive unit can be applied with a small-diameter piezoelectric actuator. Can be small.

  Further, the multi-head shown in FIG. 11 may be a subunit, and a coating apparatus in which a plurality of subunits are combined may be used.

  Here, as shown in FIG. 15, a process is assumed in which the dispenser of the above-described embodiment or example of the present invention having a multi-nozzle moves a phosphor into an independent cell of a PDP while relatively moving on the substrate. To do. Reference numeral 850 denotes a second substrate constituting the back plate, and reference numeral 851 denotes an independent cell formed of barrier ribs. The independent cell 851 includes cells 851R, 851G, and 851B into which phosphors of RGB colors are implanted. Further, as the phosphor 852, an R color (red) phosphor 852R, a G color (green) phosphor 852G, and a B color (blue) phosphor 852B are used. In FIG. 15, only the nozzle part of the dispenser is shown, and the illustration of the dispenser body is omitted.

Here, attention is focused on only one nozzle 853. In this construction method in which the phosphor is ejected from the dispenser into the independent cell and driven, it is necessary to maintain the distance H between the tip of the discharge nozzle 853 and the barrier rib apex 854 as shown in the enlarged view of FIG. The reason is as follows. In the embodiment, for example, the volume of the PDP independent cell is V = 0.65 mm (vertical) × 0.25 mm (horizontal) × 0.12 mm (depth) ≈0.02 mm ^3. The paste needs to be filled. This is because, as described above, it is necessary to form a thick phosphor layer on the inner wall of the cell after the volatile matter is removed through the filling and drying process of the phosphor coating liquid.

  At the stage where the phosphor paste is driven into the cell, the high-viscosity paste is not quickly filled into the entire cell container due to its poor fluidity. The meniscus is filled with paste from the top while maintaining a shape that is raised above the barrier rib apex 854. Therefore, the meniscus is not flattened even at the stage where application to the target cell is completed. When the tip of the discharge nozzle 853 comes into contact with the raised phosphor meniscus in the middle of application, liquid adheres to the tip of the nozzle, so that the fluid flowing out of the nozzle is affected by the fluid soul at the tip of the nozzle, causing various problems. It becomes a factor. Therefore, it is necessary to maintain a sufficient distance H between the tip of the discharge nozzle 853 and the barrier rib apex 854.

  In order to prevent the liquid from adhering to the nozzle tip, H ≧ 0.5 mm was necessary in the examples. Furthermore, if H ≧ 1.0 mm, the liquid adhesion was sufficiently prevented, and a reliable intermittent application over a long time could be achieved.

  In a state where the gap H between the tip of the discharge nozzle 853 and the opposite surface thereof is kept sufficiently large and the flow path gap sufficiently larger than the powder particle diameter is maintained while the high-viscosity / powder fluid is allowed to fly. The method of aiming at high speed in the “independent cell” is made possible by the dispenser of the above embodiment or example of the present invention.

  In the case of the conventional type, the “jet type dispenser” (FIG. 26) and the “inkjet type” (FIG. 28) were able to fly the coating fluid.

However, as described above, in the case of the “jet type dispenser”, since there is a sliding portion with zero mechanical gap between the relatively moving members, it is difficult to use the powder fluid containing the phosphor fine particles for a long time. . In addition, it is difficult for the “inkjet type” to cope with a high-viscosity fluid of 100 mPa · s or more and a powder fluid with a particle diameter of several microns in view of its principle and structure. Therefore, if the characteristics of the coating apparatus using the present invention are summarized,
(1) Applicable to high viscosity fluids on the order of thousands to tens of thousands mPa · s (cps).
(2) Clogging does not occur even with a coating material containing a powder diameter of several μm or more.
(3) Intermittent application can be performed with a short period of msec order or less.
(4) The coating fluid can fly at a distance of 0.5 to 1.0 mm or more from the discharge nozzle.
(5) The coating amount per dot can be ensured with high accuracy.
(6) Multi-heading is easy and the structure is simple.
The above (1) to (6) are also necessary conditions for achieving the independent cell type phosphor layer by direct patterning using a dispenser instead of the conventional screen printing method and photolithography method. Hereinafter, the reason why the above (1) to (6) are necessary conditions and the reason why this dispenser has the above characteristics will be supplemented a little.

The reason why the above (1) is necessary for forming the phosphor layer is that, as described above, the phosphor layer of about 10 to 40 μm is thickened on the rib wall surface after coating and drying. This is because it is necessary to use a paste-like fluid having a high viscosity in which the amount of the solvent is reduced as the coating material containing the. In addition, one of the reasons why the present invention can cope with high-viscosity fluids in the order of thousands to tens of thousands mPa · s (cps), specifically 5000 to 100,000 mPa · s, This is because a thread groove pump is used in the fluid replenishing device, and a pumping pressure for pumping high-viscosity fluid to the piston side (discharge chamber) can be easily obtained with this thread groove pump. Further, when a high viscosity fluid is used, a large discharge pressure is generated because the squeeze pressure is proportional to the viscosity. When the generated pressure is P _i = 10 MPa, for example, from Table 1, the piston diameter D _o = 3 mm, the axial load applied to the piston f = 0.015 ² × π × 10 × 10 ⁶ ≈70 N. In the present embodiment, an electromagnetic strain actuator having a large load resistance that can withstand the load is used on the piston side.

The reason why the above (2) is required for forming the phosphor layer is that, as described above, in order for the display to obtain high luminance, phosphor fine particles having a particle size on the order of several microns are usually optimal. Because. The reason why the dispenser of the present invention is less likely to be clogged in the flow path is that the secondary squeeze pressure can be used, so that the minimum value h _min of the clearance between the piston that is most likely to clog and its opposing surface is This is because it can be set to be sufficiently larger than the particle diameter of the powder, for example, h _min = 50 to 150 μm or more.

The reason why the above (3) is required for achieving the independent cell type phosphor layer by direct patterning is as follows. For example, in the case of a 42 inch wide PDP, if the number of pixels is vertical 852 RGB × horizontal 480, the number of independent cells = 3 × 408960≈1.33 million. Assuming that the time allowed for the phosphor coating process T _P = 30 sec and 100 nozzles are mounted on the coating apparatus, the time per shot T _S = 30 × 100 / 1230,000 = 0.0024 sec. This value is 1/100 or less of the responsiveness of the conventional air type and thread groove type dispenser. Therefore, when considering mass productivity, there is a need for a quick response dispenser that far exceeds the conventional type.

One of the reasons why the dispenser of the present invention can realize the above (3) is that the gap h _min between the piston end surface and the opposed surface can be set large, for example, 50 to 150 μm or more, so that the fluid filling step ( This is because the fluid resistance of the flow path leading from the supply source pump to the discharge chamber (14 in FIG. 1 and 68 in FIG. 10) can be reduced as much as possible in the suction step with the piston raised. Since the fluid resistance of the radial flow path connected to the discharge nozzle is small, the filling time can be shortened even in the case of a highly viscous fluid having poor fluidity.

Further, in the present dispenser, an electromagnetic strain actuator using a piezoelectric element, a giant magnetostrictive element or the like having a high response of 0.1 msec or less can be used effectively. The stroke of the electromagnetic strain actuator is limited to about 30 to 50 μm at a practical level. However, since this dispenser uses the secondary squeeze pressure, a large pressure can be generated even when the gap h _min is large. As can be seen from Equation 12, the secondary squeeze pressure does not depend on the absolute value of the gap h, but only on the differential dh / dt (speed) of the gap. Therefore, by making use of the advantages of the magnetostrictive actuator capable of obtaining a high speed dh / dt, a discharge pressure having a high peak of 5 to 10 MPa or more can be easily obtained with a sharp and short period.

  In the case of the conventional “jet dispenser” (FIG. 26), it seems easy to replace the mechanism for driving the needle 555 with an electromagnetic strain actuator. However, in this case, in the suction process of FIG. 27A, the gap between the suction portion 564 formed between the spherical convex portion 559 and the concave portion 561 can take no more than several tens of μm at the stroke of the electromagnetic strain actuator. As a result, in the case of mainly high-viscosity fluid, it takes time to fill the pump chamber 553 with the fluid, so that the effect of the folding angle using the electromagnetic strain actuator having high-speed response cannot be utilized.

  The reason why the above (4) is necessary for forming the phosphor layer by direct patterning is that, as described above, the contact between the phosphor meniscus raised above the barrier rib apex and the tip of the discharge nozzle during the coating process This is because it is necessary to prevent this. In addition, as described above, the reason why the above (4) can be realized is that, as described above, the dispenser is sensitive to the high-speed response of the electromagnetic strain actuator, and can easily discharge pressure having a high peak of 5 to 10 MPa or more. It is because it can be obtained. Using a high peak pressure that overcomes the surface tension of the nozzle tip, it is possible to fly over a long distance even in the case of a high viscosity fluid.

  The reason (5) is required is that the accuracy of the phosphor filling amount in the independent cell is required, for example, about ± 5%. The reason why the above (5) can be realized is that the application amount per dot in the intermittent application of this dispenser basically does not depend on the stroke of the piston, the absolute position, and the viscosity of the application fluid. This is because it is determined only by the flow rate at the operating point of the characteristics and the discharge nozzle fluid resistance and the number of times of application per unit time. Specifically, when a thread groove pump is used as the supply source pump, a predetermined application amount per dot can be set only by changing the intermittent frequency and the rotation speed of the thread groove shaft.

In the case of a conventional dispenser, the piston stroke, the absolute position, and the coating fluid viscosity all have a great influence on the discharge amount, and therefore must be strictly managed. For example, in the case of an air dispenser, the discharge amount is inversely proportional to the fluid viscosity.
In the case of the jet type, the discharge amount is directly proportional to the stroke. In this dispenser,
The rotational speed of the thread groove shaft may be controlled so as to maintain a constant rotational speed using a DC servo motor, and there are few factors that impair the accuracy of intermittent application amount.

The reason why the above (6) is required is that in the case of direct patterning, it is necessary to mount at least several tens of heads on the coating apparatus. In order to be able to replace the conventional construction method, maintenance performance comparable to the screen printing method and the photolithography method is required.
The reason why the above (6) can be realized is that the application amount per dot in the intermittent application can be insensitive to the stroke and absolute position of the piston in the present application device, as in the above (5). This is because the configuration 67) can be simplified. That is, conventional dispensers are required for high-precision machining of the relatively moving members (57, 52 in FIG. 10) in the piston drive unit, accurate alignment between the members during assembly, and ensuring absolute accuracy of the piston stroke. These process controls are not so much required in this dispenser. Therefore, the entire multi-head that independently drives a plurality of pistons can be greatly simplified.

(3) Diaphragm type head structure FIGS. 17A to 17D show a third embodiment of the present invention. The discharge chamber (corresponding to 14 in FIG. 1 and 68 in FIG. 10) is formed by a diaphragm and its facing surface, and this diaphragm is directly driven by a piezoelectric actuator to vary the gap between the diaphragm and its facing surface. The thread groove pump, which is an example of the fluid supply device, and the piston for generating the squeeze pressure are configured separately as in the second embodiment.

  17A is a partial front sectional view, FIG. 17B is a side view, FIG. 17C is a top view, FIG. 17D is a view showing a flow path formed by an upper bottom plate and a lower bottom plate, and FIG. 17E is an enlarged partial cross section of the diaphragm portion. FIG.

  Reference numeral 301 denotes a main shaft which is housed so as to be movable in the rotational direction with respect to the housing 302. The main shaft 301 is rotationally driven by a motor 303 which is an example of a rotation transmission device. A bearing 324 supports the main shaft 301. 304 is a thread groove formed on the relative movement surface of the main shaft 301 and the housing 302, 305 is a fluid suction port, 306 is a syringe for storing the coating material 307, and 308 is an air pipe for supplying auxiliary air pressure. Reference numeral 309 denotes a joint for connecting the motor output shaft 310 and the main shaft 301, and 311 denotes a discharge port having a sufficiently large flow path diameter (about several millimeters) formed on the upper bottom plate on the thread groove pump side.

  312 is a piston, 313 is a piezoelectric actuator that is an example of an axial drive device that moves the piston 312 in the axial direction, 314 is a piezoelectric actuator housing that fixes the upper end of the piezoelectric actuator 313, and 315 is an end face of the piston 312. . 316 is an upper bottom plate, 317 is a lower bottom plate, 318 is an intermediate sheet, 319 is a flow passage formed between the upper bottom plate 316 and the lower bottom plate 317 using the thickness of the intermediate sheet 318. Reference numeral 320 denotes a diaphragm formed by reducing the thickness of the upper bottom plate 316, and reference numeral 321 denotes a discharge nozzle attached to the lower bottom plate 317. A discharge port 322 is formed in the lower bottom plate 317 and the discharge nozzle 321.

  Diaphragm 320 and its fixed-side facing surface 323 form two surfaces that move relative to each other in the gap direction. A piezoelectric actuator 313, which is an example of an axial driving device, changes the relative position in the axial direction between the diaphragm 320 and the fixed-side facing surface 323. This axial drive device changes the gap h [see FIG. 17E] of the relative movement surface in the same manner as the embodiment of FIGS.

  With the head structure of the present embodiment, the flow path from the outlet of the thread groove pump to the discharge port can be completely sealed, so that the piston portion need not be sealed.

  In addition, since the output end of the electromagnetic strain actuator can be driven in a state where it is pressed directly against the diaphragm, the mass of the mechanical operating portion can be reduced. That is, since the portion corresponding to the piston 57 in the structure of FIG. 10 can be reduced, the inertial load of the electromagnetic strain actuator can be reduced. As a result, intermittent application at a higher frequency became possible.

  FIG. 17A shows a simplified example of a control block diagram of the present coating apparatus. 325 is a command signal generator that gives a driving method of the piezoelectric actuator 313, 326 is a control controller, 327 is a driver that is a driving power source of the piezoelectric actuator 313, and 328 is position information from a linear scale provided on the stage. Control based on pre-determined command signals such as piston rise / fall waveforms, intermittent period, amplitude, minimum gap, and information 328 from the linear scale that detects the relative speed and relative position of the coating device and substrate The piezoelectric actuator 313 is driven by the driver 327 via the controller 326.

  In the embodiment, the piezoelectric actuator 313 is used as the axial direction driving device for the piston 312. However, a giant magnetostrictive actuator which is one of the electromagnetic strain actuators may be used.

(4) Another Method for Adjusting Flow Rate FIGS. 18A and 18B show a fourth embodiment of the present invention. The minimum number of pistons (312 in FIG. 17A) that generates a primary squeeze pressure and a secondary squeeze pressure is shown. A case where a flow rate correction function (apparatus) is separately provided in the middle of the flow path from the thread groove pump to each nozzle, instead of correcting the flow rate variation between the heads by setting the gap h _min is shown. FIG. 18A is a partial front sectional view, and FIG. 18B is a view showing a flow path connecting the thread groove pump and the diaphragm. 351 is a main shaft, 352 is a housing, 353 is a motor, 354 is a thread groove, 355 is a suction port, 356 is a syringe of a coating material 357, and 358 is an air pipe. 359 is a joint, 360 is a discharge port having a sufficiently large channel diameter (several millimeters) on the thread groove pump side, 361 is a main piston, 362 is a piezoelectric actuator as an example of an axial drive device, and 363 is for a piezoelectric actuator. A housing 364 is an upper bottom plate, 365 is a lower bottom plate, 366 is an intermediate sheet, and 367 is a flow passage formed between the upper bottom plate 364 and the lower bottom plate 365. 368 is a main piston diaphragm formed by reducing the thickness of the upper bottom plate 364, and 369 is a discharge nozzle. 370 is a flow rate correcting piezoelectric actuator, and 371 is a flow rate correcting diaphragm formed by reducing the thickness of the upper bottom plate 365. The point that the diaphragm 368 for main pistons and its fixed side opposing surface become two surfaces which move relatively in a clearance direction is the same as that of 3rd Example. In this case, however, the minimum clearance h _min of the main piston is sufficiently large, for example, h _min > 150 μm.

By varying the displacement of the output shaft (sub-piston) 372 of the flow rate correcting piezoelectric actuator 370, the gap h _S between the flow rate correcting diaphragm 371 and its opposing surface can be adjusted. Be determined gap h _S is, thereafter to keep the gap h _S, a state in which always applies a constant voltage to flow rate correction piezoelectric actuator 370.

To explain with the equivalent circuit model of the multi-head in FIG. 12, the magnitudes of Rp1 to Rp3 are inversely proportional to the cube of the gap h, as shown by equation (16). Rp1 to Rp3 for the main piston are Rp1 to Rp3 → 0 because h _min is sufficiently large. Instead, R′p1 to R′p3 (not shown) for correcting the flow rate are replaced. In the embodiment, h _S for correcting the flow rate is set to 50 μm or less. However, the flow rate from each nozzle is actually measured in a state where high-speed intermittent application is actually performed, and h _S for correcting the flow rate is experimentally determined. You may decide.

In the embodiment, a piezoelectric actuator is used as the flow rate correcting actuator 370, but a mechanical correcting means (or device) may be used. For example, a manual type in which the output shaft of the micrometer is a sub-piston may be used.

(5) About start / end control method Hereinafter, the start / end control method in the case of intermittently applying independent cells of PDP using the present invention will be described. Returning to FIG. 15, a process is assumed in which a dispenser having a multi-nozzle relatively moves on a substrate and implants a phosphor into an independent cell. Here, attention is focused on only one nozzle 853.

  Here, the panel surface has a “display region” 855 in which a phosphor layer is formed and a “non-display region” 856 in which no phosphor layer is formed on the outer periphery of the display region 854. An outer peripheral boundary portion of the “non-display area” 856 is indicated by a chain line 857.

  The nozzle 853 that has traveled at a high speed in the direction of the arrow 858 while intermittently applying the display area 855 on the panel surface enters the non-display area 856 at the same time as the dispensing of the dispenser is interrupted when the last intermittent application is completed. In this non-display area 856, after a U-turn as indicated by an arrow 859, after passing through the running section, the display area 855 is entered again, and the dispenser resumes intermittent discharge.

The graph of FIG. 19A is a displacement curve with respect to time of the piston, in which 950 is a piston and 951 is a discharge chamber (corresponding to 14 in FIG. 1). FIG. 19B shows the motor rotation speed N with respect to time t. After the nozzle 853 has struck the coating material into the cell at the end of the display area 855, the piston 950 moves up in a steady displacement pattern. At this stage, that is, at time t = T ₁ , the nozzle 853 starts to travel toward the non-display area 856, and at the same time, the piston 950 starts to rise again with a gentle inclination angle 952. When Q _P is the volume increase amount of the discharge chamber 951 per unit time due to the piston 950 rising and Q _max is the maximum flow rate of the thread groove pump, the discharge is cut off if Q _P > Q _max. Keep (see equation 18). At time t = T ₁ , the rotational speed N → 0 of the screw groove pump motor is simultaneously set. At this time, it is more preferable if the auxiliary air pressure (308 in FIG. 17A) is also cut off. The responsiveness of the motor control and the air pressure control is about two orders of magnitude lower than that of the electromagnetic strain element, and the rise / fall time is at most about T = 0.05 seconds. During this time T, the piston stroke and the piston diameter are set so that Q _P > Q _max is satisfied and at the same time the piston 950 continues to rise.

When the nozzle 853 travels in the U-turn section (non-display area 856) on the end face of the panel, the relative speed between the nozzle 853 and the panel is zero and extremely low before and after that. If the material continues to flow out of the nozzle in this section, the discharge from the plurality of nozzles overlaps, and the material is deposited on the substrate (however, the non-display area 856). As a result, troubles such as the deposited material sticking to the tip of the discharge nozzle occur. Therefore, it is preferable to keep the discharge blocked in the U-turn section. At time t = T ₃ resume discharge, this time taking into account the time Tm required for the rise of the motor, it is sufficient to start the rotation of the pre-motor. When the application amount immediately after the start of ejection is unstable, the nozzle is placed at the position of the non-display area 856, and after being discarded once or twice, application to the independent cell may be started.

  The method of moving from the application state to the cut-off state at high speed, for example, the method of raising the piston to cut off the discharge when the discharge nozzle is changed from the “display area” to the “non-display area” on the substrate is a continuous line application method. It can also be applied to cases. Moreover, the method of decreasing or zeroing the rotational speed of the motor at the same time as raising the piston can also be applied to continuous line coating.

For example, in the case of continuous line application, after drawing a continuous line in the “display area”, make a U-turn while maintaining the discharge cut-off state in the “non-display area” and apply the continuous line as soon as the discharge nozzle enters the “display area” again. Is started. Even in this case, since the secondary squeeze pressure found by the present invention can be used, the coating method and the dispenser structure described in the above [1] to [3] can be applied. For example, when the minimum value of the clearance h on the piston end surface is h = h _min , it can be set sufficiently large so that h _min > hx, for example, h _min = 150 μm. Therefore, even if the gap h fluctuates several microns due to the thermal expansion of the member, the influence on the fluctuation of the continuous coating flow rate is small. In addition, the method for obtaining hx, the method for correcting the flow rate of the multihead, and the like may be used by replacing the intermittent flow rate with the continuous flow rate, and the above-described contents can be used.

[4] Other supplementary explanation [4-1] Method for reducing the weight of the piston drive unit The perspective view of FIG. 20 shows a fifth embodiment of the present invention, and a pump unit and a piston drive unit, which are an example of a fluid supply device, are flexible. In this configuration, the pump part is arranged on the fixed side, and the piston driving part is arranged on the stage side that travels at high speed. In this case, since the piston drive unit may be light, it is advantageous for high-speed speed control / positioning control of the discharge nozzle tip with respect to the panel.

  Reference numeral 150 denotes a panel, and a pair of Y-axis direction conveying devices 151 and 152 are provided across both sides of the panel. Further, the X-axis direction transport device 153 is mounted on the Y-axis direction transport devices 151 and 152 so as to be movable in the YY ′ direction. Further, a Z-axis direction transport device 154 is mounted on the X-axis direction transport device 153 so as to be movable in the direction of arrow XX ′. The Z-axis direction transport device 154 is equipped with a piston drive unit 155 composed of a piezoelectric actuator and a piston.

A pump unit 156 is an example of a fluid replenishing device, and is disposed on the fixed side. Reference numeral 157 denotes a flexible pipe which is a flow path connecting the pump unit 156 (for example, corresponding to 66 in FIG. 10) and the piston drive unit 155 (for example, corresponding to 67 in FIG. 10). When compressibility due to the elasticity of the flexible pipe becomes a problem in high-speed intermittent application, this apparatus may be configured with the minimum gap h _{min of the} piston sufficiently small.

[4-2] Method for Providing Application Pause Section FIGS. 21A and 21B show an example in which an “application pause section” is provided for intermittent application. Specifically, this is an application method in which after n equal-quantity dots are hit at equal time intervals, application is paused for one, and the operation of hitting n equal-dots again at equal time intervals is repeated. For example, in the chip component bonding method in circuit formation, since bonding with only one kind of adhesive is necessary, this corresponds to the case where only this part needs to be stopped.

  21A is a displacement curve with respect to time of the piston, in which 750 is a piston, 751 is a discharge chamber (corresponding to 14 in FIG. 1), and 752 is a discharge nozzle. In FIG. 21B, reference numeral 753 denotes a substrate and 754 denotes dots applied on the substrate 753.

Starting from time t = T ₁ , the piston 750 performs intermittent application for n times on a straight line 755 that gently descends while repeating the rise and fall of the same amplitude. At time t = T ₂ , the piston 750 rises more than at steady state. The value of the start point of the straight line 756 at the start of intermittent application is equal to the value of the straight line 755 at t = T ₁ . Assuming that the period of the regular piston is ΔT, the time until it rises greatly and descends again is 2ΔT. After t = T _3, the piston 750 is on a straight line 756 that gradually drops, while repeating increase and decrease of the same magnitude to repeat the intermittent application. The end point value of the straight line 756 is equal to the value of the straight line 755 at t = T ₂ at the time point when n intermittent coatings are completed.

Between time t = T ₂ of time t = T _3, a section of the time width 2.DELTA.T, 2 times of coating total flow is charged into the discharge chamber 751 from the screw groove pump. However, in intermittent application at t = T ₃ , the piston drops only by the amplitude at the normal time, and therefore only the flow rate at the normal time is applied. The feature of the present invention that the discharge pressure does not depend on the absolute value of the minimum clearance h _min when the minimum clearance h _{min of the} piston is large is utilized here.

  The fluid accumulated once more in the discharge chamber 751 is then discharged while being evenly distributed by n intermittent applications. Therefore, by using the present method and apparatus, intermittent application with the same application amount per dot can be performed in the entire section having the application pause portion.

  The above method and apparatus are effective for a coating process in which the intermittent coating time interval is set to a constant value, for example, when the dispenser is fixed and the conveyor on which the substrate is mounted runs at a constant speed. is there.

[4-3] Method of changing intermittent application amount at a certain spot A case where the intermittent application amount is changed only at a certain spot will be described.

For example, when the application amount per dot for the nth time is doubled compared to the others after the start of application, the following is performed. The time interval between the (n−2) th time and the (n−1) th time is ΔT _n−1 , the time interval between the _n− 1th time and the nth time is ΔT _n , and the time interval between the nth time and the (n + 1) th time is ΔT _{n + 1} . Here, ΔT _n = 2 × ΔT _n−1 and ΔT _n−1 = ΔT _{n + 1} are set. The total flow rate that fills the discharge chamber from the thread groove pump at the (n-1) th time is Q _n = ΔT _n−1 × Q _max , and the total flow rate that fills the discharge chamber from the thread groove pump at the nth time is Q _n = ΔT _n * _Qmax = 2 * [Delta] Tn _-1 * _Qmax . Therefore, Q _n = 2 × Q _n−1 . Assuming that the total flow rate charged from the thread groove pump into the discharge chamber is proportional to the application amount per dot flowing out from the discharge nozzle, the n-th application amount is twice that of the other. However, in the section from the end of discharge to the start of discharge (ΔT _n ), the stroke of the piston is set to be sufficiently large so that the discharge chamber can maintain a sufficiently negative pressure state. In the above, the case where the nth application amount is applied twice as much as the others has been described, but conversely, when the nth application amount is reduced to ½ times the other application amounts, ΔT _n = What is necessary is just to set (DELTA) Tn _-1 / 2 and (DELTA) Tn _-1 = (DELTA) Tn _{+ 1} . Based on this idea, the dispenser can set an arbitrary coating amount at each spot.

  The conventional dispenser controls the application amount per dot by the mechanical displacement (stroke) of the piston, whereas in this dispenser, the application amount can be controlled by controlling the time interval.

[4-4] Method for obtaining inflection point hx of discharge amount Qs curve with respect to gap h _min As described above, the setting of the minimum value h _min of the gap between the piston end face and the opposed face is extremely difficult in the present invention. is important. If h _min > hx is set, stable intermittent application that does not depend on fluctuation (drift) of the piston stroke and the absolute position of the piston can be realized.

If h _min ≈hx is set, subtle flow rate correction between multiple heads can be performed. As a method of obtaining the inflection point hx,
(1) Experimental method The minimum value h _min of the gap between the piston end face and the opposed face is set, and the total discharge amount Qs per dot is obtained in a state where intermittent discharge is performed. The inflection point hx is obtained by plotting the measured value of Qs against h _min .
(2) Theoretical method (i) Strict method An input waveform h (t) of piston displacement is given, and the flow rate Q _i is obtained using equation (14). The total flow rate Qs per dot is obtained by integrating the flow rate Q _i in the discharge process section with time t. The inflection point hx is obtained by plotting the theoretical value of Qs against h _min . The graph of FIG. 14B is obtained by the above method.

(Ii) Simple method A method for obtaining the inflection point hx in a slightly simpler manner will be described below.

As described above, when the minimum gap hmin is sufficiently large, the flow rate: Q _i has a waveform that varies around the operating point Q _ic , although the amplitude varies depending on the size of the stroke hst.

  That is, the average flow rate does not depend on the size of the stroke of the piston, but is determined by the operating point (for example, A in FIG. 8) determined by the thread groove pump characteristics and the discharge nozzle resistance. That is, when comparing the magnitude of the total discharge amount Qs per dot under a constant period, the magnitude of the continuous flow rate when the stroke hst is zero may be compared.

In the equation (14), if hst = 0, Psqu1 → 0 and Pscu2 → 0. Since _PS0 does not depend on the gap h, the inflection point hx may be obtained by plotting the value of φ with respect to h using the function φ of the next gap h.

  When a thread groove pump is used for the fluid supply device, the internal resistance is Rs = Pmax / Qmax. In many cases, the maximum flow rate Qmax and the maximum pressure Pmax of the pump are theoretically obtained. If this is difficult, the pressure / flow rate characteristics corresponding to the graph of FIG. 8 can be obtained experimentally by the following method. good.

  For the maximum flow rate Qmax, continuous discharge is performed with the discharge nozzle detached, and the total flow rate per unit time is measured. The maximum pressure Pmax may be measured by mounting a jig equipped with a pressure sensor instead of the discharge nozzle and in a state where the flow rate is zero. When a pump other than the thread groove pump is used as the fluid supply device, if the pressure / flow rate characteristics are not linear, the internal resistance Rs may be obtained from the inclination angle by linearizing around the operating point.

  18A and 18B, as in another method of flow rate adjustment (fourth embodiment of the present invention), a flow rate correction function (in the middle of the flow path from the fluid supply device (for example, thread groove pump) to each nozzle ( When a device is provided, or when there is a restriction, the fluid resistance Rx of this portion may be added to the above Rs to obtain the internal resistance (Rs + Rx → Rs) of the apparent fluid supply device.

  The fluid resistances Rn and Rp are usually obtained from well-known theoretical equations (for example, equations 15 and 16). However, if the shape is complicated, numerical analysis may be used or may be obtained experimentally. In the case of an orifice whose length of the throttle portion is short with respect to the inner diameter, the linear resistance formula (for example, formula 15) does not hold, but in this case, the apparent fluid resistance may be obtained by linearizing around the operating point. .

Hereinafter, the features of the coating apparatus to which the present invention is applied will be added.
(I) The discharge amount Qs is hardly affected by the viscosity of the coating fluid.

  In Formula 14, fluid resistance Rn, Rp, Rs is proportional to viscosity μ. Further, if supply source pressure Ps0≈thread groove maximum pressure Pmax, Ps0 is proportional to viscosity μ.

Accordingly, the denominator / numerator viscosity μ in the equation (14) is canceled. For this reason, the discharge amount of the dispenser is unlikely to depend on the viscosity. Usually, the viscosity of a fluid varies greatly logarithmically with temperature. The insensitivity to the temperature change is a very advantageous feature in configuring the coating system.
(Ii) High application accuracy can be obtained and the structure is simple.

  When the dispenser of the present invention is applied to, for example, intermittent application of PDP, as described above, the application amount per dot in intermittent application is “the flow rate characteristic of the supply pump and the flow rate at the operating point of the discharge nozzle fluid resistance” ”And“ intermittent frequency ”. For example, when a thread groove pump is used as an example of a supply source pump, if a discharge nozzle is mounted on the apparatus, the application amount per dot is determined only by the rotation speed N of the thread groove pump and the frequency f of intermittent application. The

  Since the coating amount is insensitive to the stroke of the piston, the absolute position accuracy of the piston, and the viscosity of the coating fluid, the configuration of the piston driving unit (for example, 67 in FIG. 10) can be simplified.

  Hereinafter, in the case of a conventional dispenser, how the coating amount per dot is determined will be described.

  The air-type dispenser applies a constant amount of air supplied from a constant pressure source to a container (inside 600 in FIG. 25) in a pulsed manner, and discharges a certain amount of liquid corresponding to the increase in pressure in the container from the nozzle 602. It is something to be made. Therefore, (1) variation in discharge amount due to discharge pressure pulsation, (2) variation in discharge amount due to water head difference, (3) change in discharge amount due to change in liquid viscosity, and the like cause variations.

In the above (2), since the volume of the gap portion 600 in the cylinder differs depending on the remaining amount of liquid H, when a certain amount of high-pressure air is supplied, the degree of pressure change in the gap portion 600 greatly changes due to the H. The reason is that. If the remaining amount of liquid is reduced, there is a problem that the coating amount is reduced by, for example, about 50 to 60% compared to the maximum value. For this purpose, measures are taken such as detecting the remaining amount of liquid H for each discharge and adjusting the time width of the pulses so that the discharge amount becomes uniform.
The above (3) occurs, for example, when the viscosity of a material containing a large amount of solvent changes with time. As countermeasures for this, measures such as adjusting the pulse width so as to correct the influence of the viscosity change by programming the tendency of the viscosity change with respect to the time axis in advance in a computer have been taken.

  When intermittent application is performed with a conventional thread groove type dispenser (specific structure is not described), (1) an electromagnetic clutch is interposed between the motor and the thread groove, and this electromagnetic clutch is connected or released when discharging is turned ON / OFF. (2) The method of using a DC servomotor to start or stop rapid rotation was used. However, since the response is determined by the time constant of the mechanical system in any of the above, there is a restriction on high-speed intermittent operation. In addition, since there are many uncertain factors in the rotational characteristics at the time of transient response of the pump shaft (at the time of starting and stopping the rotation), it is difficult to strictly control the flow rate and there is a limit to the coating accuracy.

  In the case of the jet-type dispenser (FIG. 26), as described above, it is necessary to mesh the spherical convex portion formed at the end of the needle 555 and the spherical concave portion formed on the discharge side with high accuracy. .

  In the case of the ink jet method (FIG. 28), the diaphragm 652 is deformed in the thickness direction by the piezoelectric element 653 to reduce the volume of the ink chamber 654 → the fluid is discharged by increasing the pressure.

  In any of the above-described coating methods, the volume of the space directly connected to the discharge nozzle is changed by some means (or device), and the coating amount is controlled according to the scheme of “volume change in space = coating amount per dot”. Was the way. In the case of the dispenser of the present invention, as described above, the volume change of the space by the piston does not determine the application amount, but its role is to change the continuous flow rate (Analog) of the supply source pump to the intermittent flow rate (Digital). D converter. Therefore, with this dispenser, these process controls required for conventional dispensers, such as high-precision machining of the relatively moving member in the piston drive unit, accurate alignment between the members during assembly, and ensuring absolute accuracy of the piston / stroke, etc. It is greatly simplified.

Therefore, the entire multi-head that independently drives a plurality of pistons can be greatly simplified structurally.
(Iii) High reliability against clogging of powder fluid in the flow path If the present invention is applied, the opening area of the flow path from the suction port of the pump to the discharge nozzle can be made sufficiently large. Is expensive.

  In particular, since the gap h between the piston end surface, which is a flow passage leading to the discharge nozzle, and the opposing surface can be made sufficiently large, it is extremely advantageous for preventing clogging of powder (for example, in the case of a phosphor, the particle size is 7 to 9 μm). Become.

  For example, in a multi-head configuration, when finely adjusting the flow rate of each head, the slope of the discharge rate with respect to the gap is smooth by using together with the output flow rate setting method (adjusting the flow rate with the rotation speed) of the supply pump. A minimum gap (for example, hmin = 50 μm in FIG. 15) may be set in the vicinity of hmin≈hx. This numerical value of 50 μm is sufficiently larger than the commonly used powder diameter (several microns to several tens of microns). When fine adjustment of the flow rate is performed in the fourth embodiment (FIGS. 18A and 18B), or when the component accuracy of each component is good and the flow rate variation between the heads can be ignored, the minimum gap hmin is 150 to 200 μm, Or you may set more.

  The piston end face portion (for example, the discharge chamber 68 in FIG. 10) directly connected to the flow path of the discharge nozzle is a portion where the direction of the flow path changes greatly, and when handling a pulverized fluid, problems such as clogging are most likely to occur. It is a place. The point that a large gap in the flow path can be secured in this portion is the greatest feature of the present invention. When applying a powder fluid such as a phosphor or adhesive containing fine particles, the minimum gap δmin of the flow path may be set larger than the fine particle diameter φd.

δmin> φd (20)

As described above, in the embodiments and examples of the present invention, the thread groove type pump is used for the fluid supply device. In order to realize the present invention, the pump can be applied to other types of pumps than the thread groove type, but in the case of the thread groove type, various parameters (radial gap, thread groove angle, groove depth, groove and ridge) constituting the thread groove. It is advantageous that the maximum pressure Pmax, the maximum flow rate Qmax, and the internal resistance Rs (= Pmax / Qmax) can be freely selected by changing the ratio of the above.

  In addition, since the flow path can be configured in a completely non-contact manner, it is advantageous when handling a powder fluid.

  In addition, the form of the pump as the fluid supply device in the present invention is not limited to the thread groove type, and other types of pumps can be applied. For example, a Mono type, a gear type, a twin screw type, a syringe type pump or the like called a snake pump can be applied. Alternatively, a pump that simply pressurizes the fluid with high-pressure air may be used.

  FIG. 22 is a model diagram when the gear type is used for the fluid replenishing device according to the present invention, where 700 is a gear pump, 701 is a flow passage, 702a, 702b, and 702c are, for example, piezoelectric actuators or the like. , 703a, 703b, and 703c are pistons.

  The shape of the piston that constitutes the piston drive unit and its opposing surface may not be circular. The piston may be rectangular. In this case, the radius of a circle having an equivalent area is taken as the average radius.

  In the above-described embodiment and examples, each nozzle has a structure of one nozzle. As long as component accuracy can be ensured, one head may be equipped with n nozzles. In this case, for example, the above-described basic expression for obtaining the flow rate per dot may be calculated as n nozzles. For example, for nozzles of the same specification, calculation is performed as Rn → Rn / n. For example, when intermittently applying phosphor in an independent cell, if multiple nozzle holes are arranged in the longitudinal direction of the rectangular independent rib, it can be applied to the entire interior of the cell, which is effective in preventing the coating fluid from protruding from the rib. Is. In the case of the example, the shape of the PDP independent cell was 0.65 mm (vertical) × 0.25 mm (horizontal). In this case, for example, 0.65 mm may be divided into four, and nozzle holes may be formed at two places on the left and right (total of three places) including the central portion. Further, if a nozzle hole for applying the same color phosphor is formed in a direction perpendicular to the traveling direction of the stage and the phosphor is applied in a plurality of independent cells, the productivity is further improved.

  In the pumps of this embodiment and examples that handle minute flow rates, the stroke of the piston may be on the order of several tens of microns at most, and even if an electromagnetic strain element such as a giant magnetostrictive element or a piezoelectric element is used, the stroke limit is a problem. Not.

  In addition, when a high viscosity fluid is discharged, a large discharge pressure is expected to be generated due to the squeeze action. In this case, since an axial drive device that drives the piston requires a large thrust against high fluid pressure, an electromagnetic strain actuator that can easily generate a force of several hundred to several thousand N is preferably used. Since the magnetostrictive element has a frequency response of several MHz or more, the piston can be linearly moved with high response. Therefore, the discharge amount of the high-viscosity fluid can be controlled with high response and high accuracy.

If the responsiveness is sacrificed, a moving magnet type, moving coil type linear motor, electromagnetic solenoid, or the like may be used for the axial direction driving device for driving the piston. In this case, the stroke restriction is eliminated. (Not shown)
As can be seen from the equation (11) or the graphs of FIGS. 4 and 5, the generated pressure and flow rate due to the squeeze effect have a phase of Δθ = π / 2 with respect to the displacement input waveform between the piston end face and the opposed face gap. The waveform is advanced. That is, the fluid is discharged in a section where the piston is descending (dh / dt <0). For example, when intermittent application is performed while moving a substrate to be applied on a stage, in order to apply with high positional accuracy while aiming at the application location, the phase is Δθ = π with respect to the displacement input signal Sh of the piston gap. Considering that the coating is advanced by / 2, the timing of the stage and the displacement input signal Sh may be matched. For example, the stage may be moved while the piston is raised, and after the stop, the piston is lowered and applied to the target substrate. (Not shown)
FIG. 23A and FIG. 23B show application examples of the present invention in the case of using a bi-reflex type piezoelectric element used in a printer or the like. The bi-reflex type piezoelectric element is used to form two surfaces that move relative to each other, and a discharge chamber formed between the two surfaces communicates with a thread groove pump that is an example of a fluid supply device.

  Reference numeral 900 denotes a main shaft that is housed so as to be movable in the rotational direction with respect to the housing 901. The main shaft 900 is rotationally driven by a motor 902. Reference numeral 903 denotes a thread groove formed on the relative movement surface of the main shaft 900 and the housing 901. In this application example, a thread groove pump in which a groove 903 is formed on the surface of the very small diameter main shaft 900 or the inner surface of the housing 901 that houses the main shaft 900 is used as a supply source pump as a fluid supply device. This micro thread groove pump serves as a common fluid supply device for supplying fluid to a plurality of discharge chambers. 904 is a fluid suction port, 905 is a thin diaphragm, 906 is a bi-reflex type piezoelectric element (drive means in the gap direction) that deforms the diaphragm 905 in the thickness direction, and 907 is a discharge nozzle mounted on the housing 901. The discharge-side end face of the diaphragm 905 and its fixed-side facing face are two faces that move relative to each other in the gap direction, and the space formed by these two faces becomes the discharge chamber 908. Reference numeral 909 denotes a main shaft end portion, and reference numeral 910 denotes a flow passage connecting the main shaft end portion 909 and the discharge chamber 908. In the case of a piezoelectric actuator, there are several methods depending on the usage of deformation of the piezoelectric element. In the above application example, a diaphragm and a piezoelectric body are laminated, and the deflection of the diaphragm due to expansion and contraction in the surface direction of the piezoelectric body is used. Is used. In this case, a large number of multi-heads can be integrated in one coating unit by a high-density nozzle arrangement, so that productivity is greatly improved. Further, in this application example, the flow passage 910 connecting the fluid supply device and the discharge chamber 908 does not have a restriction (corresponding to 656 in FIG. 28) necessary in the case of the conventional ink jet method. Since there is no throttling, which is a factor causing a filling delay when sucking a high-viscosity fluid into the discharge chamber, a high-viscosity fluid can be used as compared with a conventional ink jet. For example, a high-viscosity fluid 10 times or more can be handled as compared with a conventional ink jet whose viscosity is about 100 mPa · s. In order to correct the variation in the flow rate between the heads, a flow rate correction function (device) may be provided in the middle of the flow path from the thread groove pump to each nozzle as shown in the fourth embodiment. However, even in this case, the fluid resistance of the throttle necessary for flow rate correction can be made sufficiently small so as not to hinder high-speed intermittent application.

  In the case of this application example, the generation principle of the discharge pressure is not only the primary and secondary squeeze pressures but also pressure due to elastic waves propagated in the liquid. However, even in this case, the high internal resistance of the thread groove pump prevents back flow and brings about the effect of efficiently flowing out the fluid from the discharge nozzle.

  As the piston or the diaphragm corresponding to the piston is driven at a higher frequency, the intermittent application becomes closer to continuous application. This intermittent application may be quasi-continuous to draw a continuous line.

  In this case, the flow rate as a continuous line can be adjusted by the same method as the adjustment of the coating amount per dot.

Moreover, if a configuration is adopted in which a small pipe having a small diameter is mounted as a time delay element on the discharge side and a discharge nozzle is provided at the tip thereof, pseudo-continuation can be achieved even at a lower frequency. (Not shown)
The present invention capable of intermittently discharging a minute amount of fluid at high speed and with high accuracy can be applied not only to a coating technique but also to various uses. For example, it can be applied as means (method or apparatus) for producing microlenses used in DVD optical pickups, cameras, printers, etc., instead of conventional glass molding.

Although the above description has described only intermittent application, the structure of the coating apparatus disclosed in [2] Specific Embodiments and Examples or [3] Multi-Head is applicable even in the case of continuous application. In this case, the flow rate may be adjusted by varying the gap between the piston end surface and the opposing surface. Alternatively, the start and end of the coating line may be controlled using generation of a squeeze pressure accompanying the rising and lowering of the piston. (Not shown)
It is to be noted that, by appropriately combining arbitrary embodiments of the various embodiments described above, the respective effects can be achieved.

It is a partial cross section figure of an example of the model of the example of application of this invention. It is a partial cross section figure which shows the dimensional relationship of each structural member. It is a figure of the equivalent electric circuit model of the example of application of this invention. It is a graph which shows an example of the displacement curve of a piston. It is a graph of the analysis result of the discharge pressure characteristic of this invention. It is a graph of the analysis result of the discharge flow rate characteristic of the present invention. It is a graph of the analysis result which compared the discharge pressure characteristic by changing rotation speed. It is a figure which shows the relationship between the flow volume and pressure of a thread groove pump. It is sectional drawing which shows 1st Example of this invention. It is a partial sectional view of a model showing the case where it is constituted by separating a thread groove pump and a piston in the second embodiment of the present invention. It is a perspective view which shows the multihead which is 3rd Example of this invention. It is a figure which shows the equivalent electric circuit model in the case of a multihead. It is a graph of the analysis result which changed the piston minimum clearance and compared the discharge pressure characteristic. It is a partial sectional view of a model around a piston. 14A is a graph showing the relationship between the total discharge amount per dot and the minimum piston gap in the present invention. It is a perspective view which shows the state which the dispenser of this invention implants a fluorescent substance in the independent cell of PDP. FIG. 16 is an enlarged perspective view of FIG. 15. It is a front fragmentary sectional view which shows 3rd Embodiment of this invention. It is a side view of the said 3rd Embodiment. It is a top view of the third embodiment. It is a figure which shows only the flow path formed with the upper baseplate and lower baseplate of the said 3rd Embodiment. FIG. 17B is an enlarged partial cross-sectional view of the diaphragm portion of FIG. 17A. It is a front fragmentary sectional view which shows 4th Embodiment of this invention. It is a figure of the model which shows the flow path which connects a thread groove pump and a diaphragm. It is a figure which shows the displacement curve h with respect to the time t of a piston. It is a figure which shows the rotation speed N of the motor with respect to time t. It is a perspective view which shows 5th Embodiment of this invention. It is a figure which shows the displacement waveform of a piston in the case of providing an "application | coating rest area" for intermittent application. It is a figure which shows the dot apply | coated on the board | substrate. It is a partial cross section figure of the model at the time of using a gear pump for the fluid replenishment apparatus of this invention. It is a top view which shows the example of application of this invention at the time of using a bimorph type piezoelectric element. It is a partial front sectional view of the application example. It is a partial cross section figure which shows the example of a conventional design of the injection apparatus using a giant magnetostrictive element. It is a partial cross section figure which shows the conventional air pulse system dispenser. It is a partial sectional view showing a conventional jet dispenser. It is a partial cross section figure of the model which shows the inhalation process of the conventional jet type dispenser. It is a partial cross section figure of the model which shows the discharge process of the conventional jet type dispenser. It is a partial sectional view showing a conventional ink jet. It is a perspective view which shows the structure of a plasma display panel (PDP).

Explanation of symbols

8, 9 Two surfaces that move relative to each other 6 Fluid supply device 7 Suction port 10 Ejection port

Claims (1)

Pressure change caused by changing the gap by supplying fluid from the fluid supply device to the gap while moving the two members relative to each other in the gap direction of the gap formed between the two opposing surfaces of the two members. A fluid discharge method for intermittently discharging the fluid by using the fluid and adjusting the discharge amount of the fluid per dot by the pressure and flow rate characteristics of the fluid supply device ,
In the coating method of intermittently applying the gap between the two opposing surfaces of the two members that move relative to each other with an amplitude h ₁ , the opposing surfaces of the two members with an amplitude h ₂ that is larger than the amplitude h ₁ . After the discharge is cut off by increasing the gap, the fluid is intermittently applied a plurality of times with the amplitude h ₁ so that the central value of the gap after the interruption gradually becomes equal to the central value of the gap immediately before the interruption. Discharge method.

Images