JPH07276649A - Manufacture of ink jet head - Google Patents

Abstract

PURPOSE:To make it possible to shorten a manhour, that is, to improve a throughput while maintaining the injection quality of a discharge orifice at a high level by dividing a cutting step for the determination of a flow passage terminal into plural stages in the thickness direction of an article to be cut, and setting the cutting feed speed of an upper stage part to be higher than the cutting feed speed of an area including the discharge orifice. CONSTITUTION:An area in the depth direction of cutting an article to be cut, for example, is divided into three sections, and the cutting is completed in the number of three times. An area B is a cutting area including a discharge orifice 17. The cutting of the area B requires high precision and high quality so that the face of the discharge orifice is formed and the length of a flow passage is determined. In order to ensure these requirements, the feed speed v2 when cutting the area B is equivalent to the speed at which the constituent material of a head is cut at high quality. It is presumed that the feed speeds v1, v2 when cutting the areas A and B be v1>3v2. It is possible to prevent the cut surface at the upper part from becoming affected adversely and thereby enhance the yield by cutting the area C at a feed speed v3 which is lower than the feed speed v2 at which the area B is cut.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for manufacturing an ink jet head, and more particularly to a cutting process for precisely defining an ink flow path length and forming a high quality nozzle opening. Is.

[0002]

2. Description of the Related Art As a method for manufacturing an ink jet head, various methods have been proposed. As a method useful in mass production, cost reduction, and quality stabilization, a large area substrate is used to form a large number of heads. A method of cutting and separating is considered. In particular, the manufacturing method of obtaining a large number of uniform inkjet heads by sticking two Si wafers and dicing is a very useful method in mass production, cost reduction, and quality stabilization. Such a manufacturing method is disclosed, for example, in JP-A-61-230954.
It is described in JP-A-1-166965 and the like.

Dicing (cutting) is used as a technique common to such a method of manufacturing an ink jet head, and the ink jet head is cut out.
At this time, in many inkjet head structures, the dicing is performed to form / form a nozzle surface and a nozzle opening that greatly affect the ink droplet ejection directionality, and at the same time, the ink flow path length that greatly affects the ejection characteristics, particularly the ink droplet volume. Will be decided. Therefore, various proposals have been made for the dicing method.

For example, in the technique disclosed in Japanese Patent Laid-Open No. 60-196354, the cutting process is carried out in two or more stages according to the material of the head. Further, Japanese Patent Application Laid-Open No. 2-184451 discloses a method of performing dicing with a peripheral speed determined by a diameter of a blade used for dicing and a rotation speed being a certain value or more. Japanese Unexamined Patent Publication (Kokai) No. 5-57897 describes a method of grooving a portion that will be a nozzle opening before joining substrates, then joining the substrates and separating the grooves. In Japanese Patent Laid-Open No. 4-234666, after the resin layer portion of the nozzle opening portion is previously pattern-etched,
A method for joining and cutting substrates is described. In Japanese Patent Application Laid-Open No. 4-234667, after the resin layer portion of the nozzle opening portion is pattern-etched in advance, the portion other than the resin layer portion that serves as the nozzle opening is grooved, and then the substrate is joined and separated at the groove portion. It describes how to do. These techniques are intended to economically manufacture a head having good dimensional accuracy without missing a nozzle opening.

Since the dicing process of the ink jet head affects the ink flow path length and the shape of the nozzle opening as described above, a blade that can be processed with high precision is used.
As a blade to be used, a resin blade coated with a resin (resin) of ultra-fine diamond is generally used. However, this resin blade is soft and gradually deforms during cutting to cause an error in the cutting position.

In the technique described in the above-mentioned Japanese Patent Laid-Open No. 60-196354, since the cutting is simply performed in a plurality of times, a soft resin blade is used to obtain a high quality ejection port. Therefore, it is impossible to maintain the cutting position with high accuracy over the cutting process of many lines. Also,
The technique described in Japanese Patent Laid-Open No. 5-57897 can achieve high-quality and high-precision cutting, but unless the two substrates are accurately adhered, the upper or lower part of the ejection port is displaced, and the ink This will affect the discharge direction of. Especially in the high density head such as 600 DPI,
This effect poses a problem, and thus a problem remains in the bonding technology.

After pattern etching the resin layer portion of the nozzle opening portion disclosed in JP-A-4-234666 and JP-A-4-234667, the portion other than the resin layer portion around the nozzle opening is removed. In the method of dicing, since there is no cutting of the resin layer portion, there is no clogging of the blade, and there is no burr due to cutting in the resin layer portion,
A good discharge port can be obtained. However, a gap corresponding to the dicing position accuracy occurs between the edge of the resin layer that is pattern-etched at the discharge port and the edge of the substrate that is formed by cutting. There is a difference in level. This step difference is a problem particularly in a high density head such as 600 DPI, and a problem remains in the dicing position accuracy.

In the technique disclosed in Japanese Unexamined Patent Publication No. 2-184451, a high-quality ejection port can be obtained in the initial stage by defining the peripheral speed of the blade. However, when a large number of heads are obtained by dicing, there is a problem that the resin blade gradually deforms as described above, and the quality and accuracy in the latter half deteriorate. Further, the conditions for achieving the quality are largely related to the cutting depth, the feed rate, and the supply of the grinding water, and it is difficult to specify only the peripheral speed of the blade.

With respect to processing techniques other than the ink jet head, Japanese Patent Laid-Open Nos. 3-281170 and 4-257 are known.
As described in Japanese Patent No. 405, etc., it is proposed that a dressing material is fixed in the vicinity of a work to be cut, and a cutting blade is dressed (dressed) while cutting to achieve high cutting quality at all times. ing. However, with these processing techniques, the dress material must be fixed at the same time as the workpiece, which is problematic in terms of workability and cost.

[0010]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned circumstances, and it is possible to easily and accurately perform cutting processing and obtain a high quality ink jet head. It is an object of the present invention to provide a method for manufacturing an inkjet head.

[0011]

The present invention provides a method of manufacturing an inkjet head according to claim 1, wherein the inkjet head is provided with an ink flow path and an ejection port for ejecting an ink droplet connected to the ink flow path. A method of forming the discharge port and / or its peripheral surface by a cutting process using a rotary cutting blade, and / or defining an ink flow path length, wherein the cutting process is performed in a thickness direction of a workpiece. A region including the discharge port is divided into multiple stages, and a region above the region B not including the discharge port is defined as A, and relative feed speeds of the rotary cutting blade with respect to the workpiece in each region are V _B and V. _{If A} , then V _A > V _B. The condition of this relative feed speed can be V _A ≧ 3 × V _B. Area A
It is also possible to divide into more stages and cut.

In the method of manufacturing an ink jet head according to a fourth aspect, diamond particles are hardened with resin in an ink jet head having an ink flow path and an ejection port for ejecting an ink drop connected to the ink flow path. A method of cutting while rotating a resin-based cutting blade, the number of rotations: P (rotation / sec), the cutting edge radius: R
(Mm), cutting edge width: W (mm), feed rate: V (mm
/ Sec), depth of cut: H (mm), cutting volume per unit area of cutting blade: U = V ×
H × W / (2 × π × R × P × W) is 0.9 × 10 ⁻⁵ (m
m ³ · sec / mm ² · sec) or more, and at least one or more cutting steps for mainly cutting silicon are included.

According to a fifth aspect of the present invention, there is provided an ink jet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path, wherein the resin system is prepared by solidifying diamond particles with a resin. The cutting condition of the area mainly made of silicon including the discharge port is to cut while rotating the cutting blade of the cutting blade, and the cutting volume per unit area and unit time of the cutting blade: U is 1.4 ×
It is characterized by cutting under the condition of less than 10 ⁻⁵ (mm ³ · sec / mm ² · sec).

According to a sixth aspect of the present invention, there is provided the method for producing the ink jet head according to the first aspect, wherein when the object to be cut is mainly made of a silicon material, the region B is defined in the fifth aspect. The cutting is performed under the conditions shown, and the region A is cut under the conditions described in claim 4.

According to a seventh aspect of the present invention, there is provided an ink jet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path, wherein the resin system comprises diamond particles hardened with a resin. The cutting volume per unit time and unit time of the cutting blade: U is 1.9.
At least one cutting step for cutting glass mainly under the condition of × 10 ^-6 (mm ³ · sec / mm ² · sec) or more
It is characterized by including more than one step.

In the method of manufacturing an ink jet head according to claim 8, in an ink jet head having an ink flow path and an ejection port for ejecting an ink droplet connected to the ink flow path, a resin system in which diamond particles are hardened with a resin is used. The cutting condition of the region mainly made of glass material including the discharge port is set to be 1.5 while the cutting volume per unit area of the cutting blade is 1.5.
It is characterized by cutting under the condition of less than × 10 ⁻⁶ (mm ³ · sec / mm ² · sec).

According to a ninth aspect of the present invention, there is provided the method of producing the ink jet head according to the first aspect, wherein when the object to be cut is mainly made of a glass material, the region B is defined as the area of the eighth aspect. The cutting is performed under the conditions shown, and the region A is cut under the conditions shown in claim 7.

According to a tenth aspect of the present invention, there is provided the method of producing the ink jet head according to the first, sixth or ninth aspect, wherein the workpiece is divided into three or more stages in the thickness direction. Cutting area of cutting edge of cutting process, cutting volume per unit time: U
Where B is a region including the ejection port, A is an upper region of the region B that does not include the ejection port, and C is a lower region C of the region B that does not include the ejection port, U _C ≦ U _B <U It is characterized by setting to _A.

In the method of manufacturing an ink jet head according to the eleventh aspect, an ink jet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path is ground by rotating a rotary cutting blade. In the method of manufacturing a head, the grinding water does not directly contact the rotary cutting blade, and the grinding water is supplied to a grinding region of the rotary cutting blade via a grinding groove formed by the rotary cutting blade. It is characterized by being done.

According to a twelfth aspect of the present invention, there is provided the method of producing the ink jet head according to the eleventh aspect, wherein the supply angle of the grinding water is 0 to 45 with respect to the feed direction of the rotary cutting blade. It is characterized in that it is °.

According to a thirteenth aspect of the present invention, in the method for producing an ink jet head, in the invention of the first aspect, the sixth aspect, the ninth aspect, or the tenth aspect, when the grinding step divided in the thickness direction of the object to be ground is carried out, When the region including the ejection port is B, the upper region of the region B not including the ejection port is A, and the lower region of the region B not including the ejection port is C, at least the region B and the region C are defined. In the grinding step, the grinding is performed under the conditions for supplying the grinding water according to claim 11 or 12.

[0022]

According to the invention described in claims 1 to 3, in the cutting process for determining the flow path end, the process is divided into multiple steps in the thickness direction of the object to be cut, and the cutting feed rate of the upper part is discharged. By setting the cutting feed rate in the region including the outlet to be equal to or higher than the cutting feed rate, preferably 3 times or higher, the process time can be shortened, that is, the throughput can be improved while maintaining the quality of the ejection port. At this time, by further subdividing the cutting of the upper part into n pieces and cutting, the feeding speed can be increased by n times or more, and the effect is further enhanced.

In order to stably and accurately process a large number of heads with high quality and high accuracy, in the invention as set forth in claim 4, the cutting conditions for sharpening and correcting the shape of the resin blade are mainly silicon materials. In the case of cutting, the unit area of the blade, the cutting volume per unit time: U is 0.9 × 10
^-5 (mm ³ · sec / mm ² · sec) or more. Further, in the invention described in claim 7, in the case of cutting glass material mainly, the unit area of the blade, the cutting volume per unit time: U is 1.9 × 10 ⁻⁶ (mm ³ · sec / m
m ² · sec) or more. Under these conditions, cutting is performed in a region that does not include the discharge port or in another place, so that the blade can be sharpened and the shape can be corrected during the process.

In the invention as set forth in claim 5, the cutting conditions of the region including the discharge port are as follows: when cutting mainly a silicon material, the unit area of the blade and the cutting volume per unit time:
U is 1.4 × 10 ^-5 (mm ³ · sec / mm ² · se
Less than c). Further, in the invention described in claim 8, in the case of mainly cutting a glass material, the unit area of the blade, the cutting volume per unit time: U is 1.5 × 10 ⁻⁶
It is less than (mm ³ · sec / mm ² · sec). By cutting the region including the discharge port under these conditions, it is possible to obtain a highly accurate and high quality end face of the discharge port.

In the inventions according to claims 6 and 9, the discharge port surface is cut in multiple steps, and the upper region of the region including the discharge port is cut under the conditions described in claim 4 or 7. By cutting the region including the discharge port under the conditions described in claim 5 or 8, after the blade is sharpened and the shape is corrected when the upper region of the region including the discharge port is cut, the region including the discharge port is cut. Can be cut with high precision, and a high-quality discharge port can be obtained.

Further, in the invention described in claim 10,
In multi-step cutting with three or more steps, the cutting conditions in the lower step region are made smaller than the cutting conditions in the upper step region so that the unit area of the blade and the cutting volume per unit time: U become smaller, so that the upper step due to the runout of the blade Secondary damage to the cutting surface can be prevented and the yield can be improved.

In the invention as set forth in claim 11, the grinding water (cooling water) does not directly contact the rotary cutting blade, and the rotary cutting blade does not come into direct contact with the rotary cutting blade through the grinding groove formed by the rotary cutting blade. By supplying the water to the grinding area at the tip, the runout of the blade is suppressed and the grinding water is efficiently supplied to the grinding area, so that highly accurate grinding can be performed and the nozzle quality is good. In addition, it is difficult for the blade tip to have a non-uniform shape change. Further, as in the invention described in claim 12, a more preferable result is obtained by setting the supply angle of the grinding water to 0 to 45 ° with respect to the feed direction of the rotary cutting blade (the bottom surface of the grinding groove).

In the thirteenth aspect of the present invention, when the grinding step divided in the thickness direction of the object to be ground is performed, at least the region including the discharge port and the lower step region thereof are the same as in the eleventh and twelfth aspects. High-precision grinding can be performed by grinding under the indicated grinding water supply conditions.
A high quality discharge port can be obtained.

[0029]

FIG. 1 is a process drawing showing an embodiment of a method for manufacturing an ink jet head of the present invention, and FIG. 2 is an enlarged sectional view showing an example of a wafer during a head cutting and cutting / separating process. In the figure, 1 is a heater wafer, 2 is a channel wafer, 3a, 3b are alignment marks, 4, 4a,
4b is a head chip, 5 is a photosensitive resin layer, 6 is an adhesive,
7a, 7b and 7c are cutting positions, 8 is a cleaning liquid, 11 is a heater, 12 is a pit, 13 is an ink reservoir, 14 is an ink flow path, and 15 is a bonding pad.

In the process of FIG. 1A, the heater wafer 1 is manufactured. The heater wafer 1 is composed of a Si wafer, and as shown in FIG. 2, heaters 11, individual electrodes and common electrodes (not shown), bonding pads 15, not shown, are formed on the Si wafer by the number of head chips 4a on the heater wafer side.
A protective layer and the like (not shown) are formed by an LSI process. Further, in the step of FIG. 1B, after the photosensitive polyimide resin is spin-coated, the desired pattern is exposed and developed to form the photosensitive resin layer 5, and baking (Cure) is performed to complete the heater wafer 1. . As the photosensitive polyimide resin, for example, Ciba-Geigy Probim
IDE (registered trademark), Pyral manufactured by Du-Pont
in PI2722, etc. can be used. As shown in FIG. 2, the photosensitive resin layer 5 is not formed on the heater 11 and the bonding pad 15. The concave portion on the heater 11 becomes a pit 12 for controlling the shape of bubbles generated in the heater 11. The alignment mark 3a is also formed on the heater wafer 1 during the above process.

In the process of FIG. 1C, the channel wafer 2
To make. The channel wafer 2 is also made of a Si wafer, and as shown in FIG. 2, the ink flow path 14, the ink reservoir 13 and the like corresponding to the heater on the heater wafer 1 are formed by anisotropic etching. Also, the bonding pad 1
A recess is also formed in the portion corresponding to 5. Further, the alignment mark 3b is also formed. As a result, the channel wafer 2 is completed.

After that, in the step of FIG. 1D, the adhesive 6 spin coated on a PET film, for example, is thinly and uniformly applied by transfer onto the adhesive surface side of the channel wafer 2.

Next, in the step of FIG.
A heater wafer 1 manufactured in steps (A) and (B);
The channel wafer 2 manufactured in the steps of FIGS. 1C and 1D is aligned and bonded. The alignment is performed using the alignment marks 3a and 3b that are patterned in advance on the bonding surface side of each wafer, and is achieved with high accuracy by the alignment device while observing using an infrared microscope, for example. The two wafers are aligned and bonded, and then baked.

The process of FIG. 1F is a process of cutting and cutting and separating each head chip 4. In this step, the bonding pad 15 provided on the heater wafer 1 is exposed, so that the channel wafer 2 is cut at the cutting position 7a, and the ejection port surfaces of the two wafers bonded in FIG. 1E. In order to define the flow path length and to define the flow path length, a dicing step of cutting at the cutting position 7b and a cutting step of cutting at the cutting position 7c and cutting and separating into each head chip. In FIG. 2, the portion to be cut is indicated by a one-dot chain line. Here, the cutting is performed so as to form the groove in the dicing process, but the cutting may be performed so as to also serve as the cutting process. By this step, the individual head chips 4 are cut and separated.

The process of FIG. 1G is a process of cleaning each head chip 4 cut and separated in the process of FIG. 1F with a cleaning liquid 8. If chips and dust generated by the cutting shown in FIG. 1 (F) remain in the head chip 4, defective ejection of ink or the like will occur and the image quality will deteriorate. Therefore, in this step, the chips and dust are washed away, and the head chip 4 is brought into a clean state.

FIG. 3 is an explanatory view of an example of the ink jet head. In the figure, the same parts as those in FIGS. 1 and 2 are designated by the same reference numerals and the description thereof will be omitted. Reference numeral 16 is a bonding pad, 17 is a discharge port, 18 is a fixing substrate, and 19 is a bonding wire. The head chip 4 manufactured by the process as shown in FIG. 1 is die-bonded on the fixing substrate 18, and the bonding pad 16 on the printed wiring board (not shown) on the fixing substrate 18 and the bonding pad 15 on the heater wafer 1 are bonded. And the bonding wire 1
An electrical connection is made by 9. After that, the ink jet head is completed by connecting sealing, joints, etc., if necessary.

FIG. 4 is a front view showing an example of cutting the discharge port during the dicing process. In the figure, 21 is a blade.
In the manufacturing process of the inkjet head shown in FIG. 1, the dicing process for forming the discharge port surface and defining the flow path length is conventionally formed by one process. However, in this embodiment, for example, as shown in FIG.
Divide into 3 parts and cut. Area B is a cutting area including the discharge port. The upper area of the area B is referred to as an area A, and the lower area of the area B is referred to as an area C.

The cutting of the region B requires high precision and high quality because of the formation of the discharge port surface and the determination of the flow path length. Therefore, the feed speed v2 at the time of cutting the region B is the speed for cutting the constituent material of the head with high quality. The feed rate v1 when cutting the area A is, for example, the area B
The feed rate v2 at the time of cutting is set to 3 times.
The feed speed v3 when cutting the area C is the same as the feed speed v2 when cutting the area B. The feed speed v3 at the time of cutting of the region C does not have a discharge port, so that the cutting can be performed at a higher feed speed. However, due to the high feed rate, the blade swings, causing area A and area B
May affect the cutting surface of. Therefore, area C
Feed rate v3 during cutting of
By performing cutting at a speed of 2 or less, it is possible to prevent the upper cut surface from being affected and improve the yield.

More specifically, the constituent material of the head is mainly Si.
In the area B, the feed rate v2 for cutting this with high quality is, for example, 2.0 mm / sec.
And it is sufficient. The feed speed v1 is 6.
It may be 0 mm / sec. The feed rate v3 may be 2.0 mm / sec, which is the same as the feed rate v2. Also,
Cutting depth h1, h2, h3 of each subdivided cutting process
Are 350, 350 and 400 μm, respectively. The blade 21 used is a diamond ultrafine diameter (grain size 3
No. 000) is coated with resin (resin) and is a resin blade with a diameter of 52 mm.
It is set to rpm.

Here, the number of rotations of the blade is P (rotation / s
ec), cutting edge radius is R (mm), cutting edge width is W (m)
m), the feed rate is V (mm / sec), and the cutting depth is H (mm), the unit area of the cutting blade and the cutting volume per unit time U = V × H × W / (2 × π × R x P x
Consider W) (mm ³ · sec / mm ² · sec). In the above specific example, the cutting volume U ₁ of the region A = 2.57 × 1
0 ^-5 , the cutting volume U ₂ of the region B = 8.58 × 10 ^-6 , and the cutting volume U ₃ of the region C = 1.03 × 10 ^-5 .

Generally, a high-quality cutting is carried out by using a blade 21 such as a resin blade which is heavily worn. For example, in the case of Si, the cutting volume U <1.0 × 10 ^-5 (m).
If the cutting is continued at (m ³ · sec / mm ² · sec), the cutting edge becomes non-uniform due to wear. FIG. 5 is an explanatory diagram of the tip cross-sectional shape of the blade. FIG. 5 (A) shows an initial blade tip cross-sectional shape, and FIG.
(B) and FIG. 5 (C). The wear as shown in FIG. 5B is largely caused by the load and heat at the blade tip, and becomes more remarkable when the grinding depth is deep. When the tip of the blade has a cross-sectional shape as shown in FIG. 5 (C), the force that the blade receives from the workpiece is uneven, and the blade is bent during cutting.

FIG. 6 is an explanatory view of the shape of the tip of the blade and the bending of the cut surface. As shown in FIG. 6 (A), in the initial blade tip cross-sectional shape, since the force received from the object to be cut is uniform, the blade can be cut straight. However, with the cross-sectional shape of the tip of the blade as shown in FIG. 6 (B), the blade bends during cutting and the cutting is performed obliquely. This blade bending is 50 μm
The flow path length (TCL) accuracy varies to some extent.

Generally, in order to suppress the above-described blade bending, it is sufficient to reduce the amount of blade protruding from the flange (blade holding jig) and increase the blade width. However, the blade protrusion amount cannot be made equal to or less than the cutting depth, and in the above-described specific example, since two general Si wafers are bonded together, it is difficult to set the blade protrusion amount to 1 mm or less. In addition, increasing the blade width results in a wider cutting width, which reduces the head removal height when cutting a large number of head chips from the wafer, leading to an increase in head chip cost. It is also possible to carry out a step (truing) of adjusting the shape of the blade every time a certain cutting is performed. However, in this method, it is necessary to perform truing several times within one wafer, and alignment with the object to be cut is required each time, so that productivity is significantly deteriorated.

FIG. 7 is an explanatory diagram of changes in the feed speed of the blade and the shape of the tip of the blade. As described with reference to FIG. 5, at the blade feed rate that achieves high cutting quality, wear as shown in FIG. 7A was exhibited. However, when the blade feed rate is increased by about 3 times or more, FIG. As in (B), the amount of wear increases, but the shape becomes uniform. This is because the excessive cutting resistance accelerates the loss of diamond particles and the abrasion of the resin portion. Even with a blade that has been unevenly worn once, by performing cutting by high-speed feed, the convex portions of uneven wear are evened out, and a certain shape correction is possible. At this time, new diamond particles and new pockets (hollows) are formed at the same time, and a sharpening effect is obtained. At this time, if the blade feed speed is a certain speed or higher, the load on the blade becomes excessively large, which causes large shake and breakage of the blade. This is greatly affected by the cutting depth and the like, but the limit is about 10 times under the same conditions.

By applying the selected feed rate in consideration of such characteristics, the entire wafer cutting process can be performed.
Short cutting time, high precision and high quality have been achieved. That is, when the discharge port surface is conventionally cut once, the cutting depth is deep. Therefore, in order to perform cutting with high accuracy and high quality, the cutting speed is about 1/3 of the feed speed in the region B. Need to cut. However, as in the above-described embodiment, since a part of the region having no discharge port is cut by high-speed feed, the cutting time can be shortened as a whole, and the precision and quality of the region having the discharge port can be maintained. be able to.

FIG. 8 is a front view showing another example of cutting the discharge port during the dicing process. In this example, the upper area of the area B including the ejection port 17 is further divided into three. This 3
The divided upper regions A1, A2, A3 are cut at a feed rate of 20 mm / sec, for example. By doing so, the entire cutting time is shortened, and since the cutting is performed at a high speed with a shallow cutting depth, the blade shape correction is better and more stable than in the case shown in FIG. Here, for example, a blade diameter of 50 mm and a rotation speed of 4
If the cutting depth is 100 rpm and the cutting depth is 100 μm, the cutting volume per unit area of the cutting blade U ₁ = 1.9.
It is 1 × 10 ⁻⁵ (mm ³ · sec / mm ² · sec).

FIG. 9 is an enlarged sectional view showing an example of a dicing processed portion of the wafer. In the figure, the same parts as those in FIG. Reference numeral 31 is a heater protection film, and 32 is a common electrode. When forming the ejection surface and determining the flow path length by dicing, as shown in FIG. 9, the shape of the channel wafer 2 may differ in the blade width direction on the cutting line. Cutting the portion having such a shape promotes uneven wear of the blade. But,
When the cutting method as shown in FIG. 8 is used, the shape of the blade is corrected to a good state before cutting the region including the discharge port, and therefore, stable even when cutting the region including the discharge port. Can be processed as desired. Therefore, no extra space such as lengthening the flow path for cutting is required, which leads to improvement of the head removal height.

In the example of FIG. 8, the area B and the area C are used.
The cutting depths were set to 450 and 350 μm, respectively. At this time, the cutting area per unit area and the cutting volume per unit time in the regions B and C are U ₂ = 8.60 × 10 5, respectively.
^-6 , U ₃ = 6.69 × 10 ^-6 (mm ³ · sec / mm ²
・ Sec). Thereby, the load on the blade in the cutting process of the region C is reduced, and interference with the surface formed by the upper cutting due to the deflection of the blade is prevented,
It is possible to reduce defects near the discharge port. As a method of making the load on the blade smaller in the cutting process in the region C than that of the upper portion, a method of changing the rotation speed or feed speed of the blade is also possible, but changing / stabilizing time is necessary to change the rotation speed. Since the head manufacturing process time is lengthened, the method of changing the cutting depth or the feed rate as described above is preferable.

As described above, the wear as shown in FIG. 5B is largely due to the load and heat at the blade tip. The cutting by the cutting blade as described above is a process also called grinding, and heat is generated in the grinding region.
Therefore, grinding water (cooling water) is supplied to the grinding area to prevent abrasion due to heat. However, the wear at the central portion in the blade width direction as shown in FIG. 5B is largely due to insufficient supply of grinding water.

FIG. 10 is an explanatory view of a conventional grinding water supply method, and FIG. 10 (A) is a schematic view of the supply method.
10B is a sectional view of the ground portion, and FIG. 10C is a plan view of the ground portion. In the figure, 21 is a blade, 22 is grinding water,
Reference numeral 23 is a grinding water nozzle, 24 is an object to be ground, and 25 is a grinding groove. The feed direction of the blade 21 is from right to left in the figure. In the conventional method of supplying the grinding water 22, FIG.
As shown in (A), the grinding water nozzle 23 and the blade 21
A connecting the centers of the grinding water nozzle 23 and the blade 21
Between the outer circumference of the blade 21 and the line b tangent thereto
It supplies so that 2 may hit directly. This is because the grinding water 22 is well supplied to the side surface of the blade 21. When a blade using a diamond having an extremely small diameter as described above, that is, a blade having small irregularities on the side surface is used and cutting is performed under the condition that the blade runout is small, the grinding water 22 is provided between the object to be ground and the blade 21. There is not enough space to feed. Therefore, in the method shown in FIG. 10 (A), sufficient grinding water 22 is not supplied to the grinding action portion of the tip portion of the blade 21, and as described above, in the blade width direction as shown in FIG. 5 (B). Wear was likely to occur in the central part of the. This tendency becomes more remarkable as the grinding groove becomes deeper. Further, as shown in FIGS. 10B and 10C, in the conventional method of supplying the grinding water 22, the grinding water 22 is directly applied to the blade 21, so that most of the grinding water 22 is on the workpiece 24. And was not sufficiently supplied to the ground part. It may be possible to improve the shortage of the grinding water 22 by increasing the flow rate, but since the grinding water 22 is directly applied to the blade 21, an excessive load is applied to the blade 21 to cause a runout of the blade and As a result, the grinding quality is deteriorated.

FIG. 11 is an explanatory view of an example of the method for supplying the grinding water according to the present invention, and FIG. 11 (A) is a schematic view of the method for supplying the grinding water.
11B is a sectional view of the ground portion, and FIG. 11C is a plan view of the ground portion. Reference numerals in the figure are the same as those in FIG. The feeding direction of the blade 21 is from the right to the left in the figure, as in FIG. In the method for supplying the grinding water 22 according to the present invention, the grinding water 22 is not directly supplied to the blade 21,
The grinding water 22 is supplied to the grinding portion via the grinding groove 25 formed by grinding. By supplying the grinding water 22 in such a manner, the amount of the supplied grinding water 22 scattered on the surface of the workpiece 24 is reduced, and the grinding groove 2
The grinding water 22 is sent to the grinding portion along with the rotation of the blade 21 along 5. Therefore, the grinding water 22 can be sufficiently supplied to the grinding portion, wear as shown in FIG. 5B is unlikely to occur, and good grinding quality can be obtained.

Further, in the method for supplying the grinding water 22 according to the present invention, the grinding water 2 with respect to the feeding direction of the blade 21 is used.
It is desirable that the supply angle 2 of 2, that is, the supply angle θ of the grinding water 22 with respect to the bottom surface of the grinding groove 25 be 0 ° to 45 °, and more preferably 0 ° to 30 °. Good grinding quality can be obtained.

FIG. 12 is an explanatory diagram of a comparison result of the cutting time, the flow path length accuracy, and the yield of the head manufactured by the manufacturing method of the present invention and the conventional manufacturing method. As a conventional head manufacturing method, a method of forming a discharge port surface by one-time cutting and a conventional method of simply cutting in multiple stages (three stages) were used. At this time, the grinding water supply method used was the conventional supply method shown in FIG. As a manufacturing method of the present invention, a case of cutting into three parts shown in FIG. 4 in “Invention example 1” is shown, and shown in FIG. 8 in “Invention example 2” and “Invention example 3”. It shows a method of cutting the uppermost stage further into three. The “invention example 2” and the “invention example 3” differ in the method of supplying the grinding water. In “Invention Example 1” and “Invention Example 2”, the method for supplying grinding water of the present invention as shown in FIG. 11 was used. The supply angle θ of the grinding water is 25 °. Further, in "Example 3 of the present invention", the conventional grinding water supply method as shown in FIG. 10 was used. Here, in each example, the flow rate of the grinding water was 1.5.
1 / min.

FIG. 12 shows the cutting time, the accuracy of the flow path length, and the yield from the quality of the discharge port when cutting is performed by each method. The flow path length accuracy of this evaluation result is 2
Alignment accuracy within about μm is included. Figure 1
As can be seen with reference to 2, both three methods of the present invention
It is clear that the accuracy of the flow path length and the yield are greatly improved and the cutting time is shortened as compared with the conventional case. That is, comparing the “conventional one-step example” and the “conventional multi-step example” with the “third example of the present invention”, by using the cutting method of the present invention, the flow path length accuracy is greatly improved and the cutting time is also increased. It has been shortened and the yield has been improved. Further, from the comparison between “Invention example 2” and “Invention example 3”, the yield can be greatly improved by changing the method of supplying the grinding water. As described above, even if only one of the cutting method and the grinding water supply method is used, it is greatly improved as compared with the conventional method. However, as in “Invention Example 1” and “Invention Example 2”, By using both the cutting method of the present invention and the method of supplying grinding water, better cutting can be performed.

FIG. 13 is a block diagram showing another head structure. 13A is a front view and FIG. 13B is a sectional view. In the figure, parts similar to those in FIGS. 1 to 4 are designated by the same reference numerals, and description thereof will be omitted. 41 is a photosensitive resin layer,
42 is a partition.

The heater wafer 1 is composed of a Si substrate, and the heater 13 and electrodes (not shown), signal circuits, etc. are well known LSIs.
It is formed by technology. A photosensitive resin layer 5 made of polyimide is provided thereon, and pits 12, bonding pad portions (not shown) and the like are patterned. The channel wafer 2 functions as a top plate in this example, and a photosensitive glass plate is used. An ink reservoir 13 is formed on the channel wafer 2, a photosensitive resin layer 41 is provided, and a photosensitive resin is patterned thereon to form partition walls 42. Next, the heater wafer 1 and the channel wafer 2 are bonded to each other to form the ink flow path 14, which is cut and separated to manufacture an inkjet head. The discharge port 17 appears on the cut surface.

FIG. 14 is a front view showing an example of cutting the ejection port at the dicing step in another head structure. In this example, the cutting surface is divided into a region B including the photosensitive resin layer 5, a region above the region B, and a region C below the region B. Further, the area above the area B is divided into two areas, and areas A1 and A2 are cut for each area. The number of divisions of this area is not limited to this. In the head structure of this example, since the photosensitive glass is used as the channel wafer 2, the region B is cut under conditions suitable for the photosensitive glass. Area A1, A2
The cutting is performed at a feed rate v1 that is faster than the feed rate v2 in the cutting of the area B. The cutting depth is made shallow by dividing the region, and the feed speed v1 of the regions A1 and A2 can be set to about 10 times the feed speed v2 of the region B. By cutting at such a feed rate, the blade can be sharpened. Since the heater wafer 1 is made of a Si substrate, the cutting of the region C is performed at a feed rate suitable for the cutting of Si. As a result, it is possible to reduce the influence on the already cut surface due to the blurring of the blade or the like.

Specifically, for example, the cutting depths of the regions A1 and A2 are 200 μm, the region B is 450 μm, the region C is 550 μm, and a blade having a diamond grain size of 600 and a diameter of 52 mm called a metal resin is used.
Cutting is performed under the condition of a rotation speed of 40,000 rpm. At this time, the feed rate v1 when cutting the areas A1 and A2 is 1.
0 mm / sec (U ₁ = 1.84 × 10 ⁻⁶ ), and the feed rate v2 when cutting the region B is 0.1 mm / sec (U ₂ = 4.13 × 1) which is a condition suitable for the photosensitive glass.
0 ^-7 ), the region C is a feed rate v3 as a condition suitable for Si.
= 1.0 mm / sec (U ₃ = 5.05 × 10 ⁻⁶ ).

As shown in the above two examples, not only the object to be cut (Si, glass, etc.) but also the resin type of the blade to be used, the ultrafine diameter of diamond and the cutting depth, the rotational speed of the blade, the cooling water ( Depending on how the grinding water is supplied,
The feed rate for cutting with high quality is different. The feed rate suitable for the object to be cut in the present invention means an upper limit rate in consideration of a margin within the range of the feed rate that can be selected from the quality around the discharge port when the above-mentioned conditions are constant. pointing. The feed rate selected in this way is that the crushed (or rounded corners) diamond ultra-fine diameter of the blade is appropriately missing, and a new diamond ultra-fine diameter (diamond grains with sharp edges) is exposed. High quality dicing quality can be obtained within a certain range.

Further, the condition of blade correction / dressing according to the present invention is to perform the correction of the blade shape by forcibly causing the diamond extra-fine diameter to be lost in response to the above-mentioned condition of the appropriate diamond extra-fine diameter loss. Is.
For example, in a resin blade, these are substantially determined by the material to be cut and the blade unit area / cutting volume per unit time: U, which is determined by the blade diameter, the number of revolutions, the feed rate and the cutting depth.

FIGS. 15 to 17 are explanatory views of the cutting volume per unit area / unit time, the quality of the discharge port, and the worn shape of the blade. FIGS. 15 and 16 show the case where the object to be cut is Si, FIG. 17 shows the case where the object to be cut is glass. These tables show the results of evaluation of the quality of the discharge port and the change in the shape of the blade when the cutting volume per unit area / unit time: U was changed. Here, “good ejection port quality” means that the average ejection port has no defects or defects of 2 μm or less. Further, as a good blade wear shape, the deformation amount after a cutting length of 200 cm (d in FIG. 5). ) Is 50 μm or less is a criterion. For example, with a blade having a thickness of 200 μm, if the amount of deformation of the wear shape of the blade exceeds 50 μm, the probability that the cross-sectional shape of the blade becomes the shape shown in FIG. There is.

When the material to be cut is Si, a blade having a diamond grain size of 3000 and a blade width of 200 μm is used, the protrusion amount from the flange that is the blade pressing is 1.5 mm, and the grinding water is efficiently applied to the cutting site. An angle, for example, the supply angle θ shown in FIG. 11 (B) is an angle of 25 °, and an amount of 1.5 l / min is supplied. The diamond grain size of the blade has a great influence on the quality of the outlet. When the material to be cut is Si, about 1000 to 5000 are selected in order to improve the quality of the discharge port. Here, the results shown in FIG. 15 and FIG. 16 are obtained by sliding the average quality required for chipping of the ejection port, and the same result is obtained in the above-mentioned range of the diamond grain size even if the grain size is different.

As shown in FIGS. 15 and 16, when the object to be cut is Si, the condition for obtaining good discharge port quality is U =
1.4 × 10 ⁻⁵ (mm ³ · sec / mm ² · sec) or less, preferably U = 2 to 7 × 10 ⁻⁶ (mm ³ · s)
ec / mm ² · sec) is good. The feed accuracy with which the worn shape of the blade is good is 0.9 × 10 ^-5 (mm ³ · se
c / mm ² · sec), preferably 1.5 to 3.
0 × 10 ⁻⁵ (mm ³ · sec / mm ² · sec) is preferable. × in the column of outlet quality and blade wear shape in the table
× indicates the occurrence of blade damage due to excessive cutting resistance, ○
X indicates good quality in the initial state due to blunting of the blade, but deteriorates with time.

From FIGS. 15 and 16, U = 0.9 to 1.8.
× 10 ^-5 (mm ³ · sec / mm ² · sec) and 4.3
~ 6.4 × 10 ^-5 (mm ³ · sec / mm ² · sec)
In the vicinity area, the evaluation of the discharge port quality and the blade wear shape depending on the size of U may be reversed. This is because the value of U depends on the number of rotations of the blade, the cutting depth, the feed rate, and the blade diameter, and there are many combinations of these values for one value of U. In the above-mentioned region, the quality of the discharge port can be ensured depending on the combination, and the blade wear shape can be improved by another combination. Among these values, the feed rate and the cutting depth are more influential than other factors, and particularly the feed rate greatly affects the wear shape of the blade. Therefore, even if the value of U is the same, it is effective to correct the shape of the blade by making the cutting depth shallow and increasing the feed rate.

Even when the material to be cut shown in FIG. 17 is a glass material, there is a slight difference, but the tendency is almost the same as when the material to be cut is Si, and the value of U is about 1/10. Has become. The blade used is a diamond grain size of 60.
The blade is No. 0 and has a width of 300 μm, and other conditions such as the amount of protrusion from the flange and the cooling water are the same as when the workpiece is Si. The evaluation shown in FIG. 17 is an evaluation when the photosensitive glass with a groove is cut, and is different from the actual head shape. In addition, the quality of the ejection port is judged to be good by not having a chip of 10 μm or more in the photosensitive glass groove portion of the sample. In the case of an actual head, this is
This is because the head configuration as shown in FIG. 3 is premised and there is no glass in the vicinity of the ejection port, so that the quality required from the ejection directionality becomes gradual.

From FIG. 17, the cutting of the region including the discharge port is U
= 1.5 × 10 ^-6 (mm ³ · sec / mm ² · sec)
Hereinafter, the blade correction condition is U = 1.9 × 10 ⁻⁶ (mm ³
・ Sec / mm ²・ sec) or more is selected.

Cutting volume per unit area / unit time: U
The blade thickness is not ideally related to the value of. However, in reality, when the thickness of the blade becomes extremely thick, the supply of the cooling water to the cutting action portion becomes non-uniform, causing defective cutting. In addition, when the blade is extremely thin, vibration of the blade occurs and quality deterioration occurs. Therefore, a blade having a width of about 0.05 to 1.0 mm is usually selected.

The present invention is not limited to the above-mentioned examples, and the blade feeding direction and the like can be freely selected.
In the case of the triangular discharge port shape as shown in FIG. 4 and FIG. 8, by making the blade rotation direction from the bottom to the top with respect to the blade feeding direction particularly in the cutting region B including the discharge port. It is possible to improve the defect due to chipping on the slope of the ejection port. Further, the head structure is not limited to the above-mentioned structure, and can be applied to inkjet heads of various structures, for example, a structure in which only the flow path length is defined by cutting and an orifice plate is separately attached. Applicable.

Further, in the above description, the example in which the cutting for correcting the shape of the blade is applied to a part of the multi-step cutting on the discharge port surface has been shown, but the present invention is not limited to this example. For example, the discharge port surface is cut by conventional one-step cutting,
The cutting 7a for exposing the wire bond portion may be cut under the cutting conditions for correcting the shape of the blade, and after the shape of the blade is corrected, the discharge port surface may be cut.

[0070]

As is apparent from the above description, according to the present invention, it is possible to stably manufacture a large number of heads with high precision and high quality without requiring an additional step or the like. In addition, the production time can be shortened.

[Brief description of drawings]

FIG. 1 is a process drawing showing an example of a method for manufacturing an inkjet head of the present invention.

FIG. 2 is an enlarged cross-sectional view showing an example of a wafer during a head cutting and cutting / separating step.

FIG. 3 is an explanatory diagram of an example of an inkjet head.

FIG. 4 is a front view showing an example of cutting a discharge port during a dicing process.

FIG. 5 is an explanatory diagram of a sectional shape of a tip of a blade.

FIG. 6 is an explanatory diagram of the tip shape of the blade and the bending of the cut surface.

FIG. 7 is an explanatory diagram of changes in blade feed speed and blade tip shapes.

FIG. 8 is a front view showing another example of cutting discharge ports during a dicing process.

FIG. 9 is an enlarged cross-sectional view showing an example of a dicing processed portion of a wafer.

FIG. 10 is an explanatory diagram of a conventional grinding water supply method.

FIG. 11 is an explanatory diagram of an example of a method for supplying grinding water according to the present invention.

FIG. 12 is an explanatory diagram of a comparison result of the cutting time, the flow path length accuracy, and the yield of the head manufactured by the manufacturing method of the present invention and the conventional manufacturing method.

FIG. 13 is a configuration diagram showing another head structure.

FIG. 14 is a front view showing an example of cutting ejection ports during a dicing process in another head structure.

FIG. 15 is a unit area when the cutting object is Si.
It is explanatory drawing of the cutting volume per unit time, the quality of a discharge port, and the wear shape of a blade.

FIG. 16 is a unit area when the cutting object is Si.
It is explanatory drawing of the cutting volume per unit time, the quality of a discharge port, and the wear shape of a blade.

FIG. 17 is an explanatory diagram of the cutting volume per unit area / unit time, the quality of the discharge port, and the worn shape of the blade when the object to be cut is glass.

[Explanation of symbols]

1 ... Heater wafer, 2 ... Channel wafer, 3a, 3b ...
Alignment marks, 4, 4a, 4b ... Head chip,
5 ... Photosensitive resin layer, 6 ... Adhesive agent, 7a, 7b, 7c ... Cutting position, 8 ... Cleaning liquid, 11 ... Heater, 12 ... Pit, 1
3 ... Ink reservoir, 14 ... Ink flow path, 15, 16 ... Bonding pad, 17 ... Discharge port, 18 ... Fixing substrate, 1
9 ... Bonding wire, 21 ... Blade, 22 ... Grinding water, 23 ... Grinding water nozzle, 24 ... Grinding object, 25 ... Grinding groove, 31 ... Heater protective film, 32 ... Common electrode, 41 ... Photosensitive resin layer, 42 ... partition.

Claims (13)

[Claims] 1. An ink jet head having an ink flow path and a discharge port for ejecting ink droplets connected to the ink flow path, wherein the discharge port and / or its peripheral surface are formed by a cutting process using a rotary cutting blade, And / or a method of defining the ink flow path length, wherein the cutting step is divided into multiple stages in the thickness direction of the object to be cut, the region including the ejection port is B, and the region not including the ejection port. When the upper region of B is A and the relative feed rates of the rotary cutting blade to the workpiece in each region are V _B and V _A ,
_A method of manufacturing an inkjet head, wherein V _A > V _B. 2. The invention according to claim 1, wherein V _A ≧
3. A method for manufacturing an inkjet head, which is characterized in that it is 3 × V _B. 3. The method for manufacturing an ink jet head according to claim 1 or 2, wherein the region A is further divided into a plurality of stages and cut. 4. A method of cutting while rotating a resin-based cutting blade, in which diamond particles are hardened with a resin, in an inkjet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path. Rotation speed: P (rotation / sec), cutting edge radius: R (mm), cutting edge width: W (mm), feed rate: V (mm / sec),
When the cutting depth is H (mm), the cutting area per unit area of the cutting blade and the cutting volume per unit time: U = V × H × W / (2
× π × R × P × W) is 0.9 × 10 ⁻⁵ (mm ³ · sec
/ Mm ² · sec) or more and at least one or more cutting steps for mainly cutting silicon under the conditions. 5. An ink jet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path, which is a method of cutting while rotating a resin-based cutting blade in which diamond particles are hardened with a resin. The cutting conditions of the area mainly made of silicon including the discharge port are set as follows: U of 1.4 × 10 ^-5
A method for producing an ink jet head, which comprises cutting under a condition of less than (mm ³ · sec / mm ² · sec). 6. The invention according to claim 1, wherein when an object to be cut is mainly made of a silicon material, a region B is cut under the condition shown in the above claim 5, and a region A is shown in the above claim 4. A method for manufacturing an inkjet head, which comprises cutting under different conditions. 7. An ink jet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path, which is a method of cutting while rotating a resin-based cutting blade in which diamond particles are hardened with a resin, Cutting area per unit area of cutting blade: Cutting volume: U is 1.9 × 10
^-6 (mm ³ · sec / mm ² · sec) A method for producing an inkjet head, which comprises at least one or more cutting steps for mainly cutting glass under the condition of not less than ⁶ mm ² . 8. An ink jet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path, by a method of cutting while rotating a resin-based cutting blade in which diamond particles are hardened with a resin, The cutting conditions of the area mainly made of glass material including the discharge port are set as follows: Cutting volume per unit area of cutting blade: U = 1.5 × 10
^-6 (mm ³ · sec / mm ² · sec) A method for producing an inkjet head, characterized by cutting under the condition. 9. The invention according to claim 1, wherein when the object to be cut is mainly made of a glass material, region B is cut under the conditions shown in claim 8 and region A is shown in claim 7. A method for manufacturing an inkjet head, which comprises cutting under different conditions. 10. The invention according to claim 1, 6 or 9, wherein when the object to be cut is divided into three or more stages in the thickness direction, the cutting blade of the cutting process has a unit area and a cutting volume per unit time: When U is a region including the ejection port B, an upper region of the region B not including the ejection port A, and a lower region C of the region B not including the ejection port, U _C
A method of manufacturing an inkjet head, wherein ≦ U _B <U _A is set. 11. A method for manufacturing an inkjet head, which grinds an inkjet head having an ink flow path and an ejection port for ejecting ink droplets connected to the ink flow path while rotating the rotary cutting blade, wherein The inkjet head is characterized in that the grinding water does not come into direct contact with the grinding water and is supplied to the grinding region of the rotary cutting blade via the grinding groove formed by the rotary cutting blade. Of manufacturing. 12. The method of manufacturing an ink jet head according to claim 11, wherein the supply angle of the grinding water is 0 to 45 ° with respect to the feed direction of the rotary cutting blade. 13. The method according to claim 1 or 6 or 9 or 10.
In the invention described in [1], when the grinding step divided in the thickness direction of the object to be ground is performed, the region including the discharge port is B,
When the upper region of the region B not including the ejection port is A and the lower region of the region B not including the ejection port is C, at least in the grinding process of the region B and the region C, the method of claim 11 Alternatively, the method for producing an inkjet head is characterized in that grinding is performed under the condition for supplying the grinding water described in 12.