JP2006194598A - Strain sensor and its manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a strain sensor and its manufacturing method capable of forming an insulating layer and wiring on the same plane with suppressed irregularities, suppressing to the minimum an influence of wiring steps or the like on printability of resistor paste, and reducing to the utmost characteristic dispersion of a strain-sensitive resistor. <P>SOLUTION: This strain sensor is characterized by having a substrate 21, and the insulating layer 22 with a plurality of electrodes 12 buried therein on the substrate 21, and having a protection layer 25 equipped with the strain-sensitive resistor 23 on the electrodes 12, for covering the strain-sensitive resistor 23 and at least a part of the electrodes. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a strain sensor using a metal elastic body and a manufacturing method thereof.

  Conventionally, a strain sensor has been proposed in which a strain sensitive resistor is formed on an elastic metal body via an insulator. Such strain sensors are widely used in applications such as smart airbags for automobiles (adjusting the opening of the airbag according to the weight of the passenger in the passenger seat), and further improvements in performance and cost are desired in the future.

  A conventional strain sensor is shown in FIG. FIG. 17 is a cross-sectional view showing a configuration of a conventional strain sensor. As shown in FIG. 17, a strain sensitive resistor 3 and a plurality of convex electrodes 4 connected to both ends thereof are formed on a substrate 1 with an insulating layer 2 interposed therebetween. The protective layer 5 is formed so as to cover the strain sensitive resistor 3 and the convex electrode 4.

  FIG. 18 is a diagram showing a state in which a plurality of strain sensitive resistor elements are connected in a predetermined shape to form a bridge circuit. In FIG. 18, the plurality of strain sensitive resistor elements 6 form a predetermined bridge circuit, so that the sensitivity of the sensor can be increased. 18 corresponds to the strain-sensitive resistor 3 in FIG. 17 and the convex electrode 4 connected to both ends thereof.

  However, in the case of such a structure, the characteristic as a bridge circuit cannot be obtained unless the resistance value variation of the individual strain sensitive resistance elements 6 shown in FIG. In this case, when the resistance values of the plurality of strain sensitive resistor elements 6 forming the bridge circuit vary from one another, calibration may not be possible with a semiconductor chip (not shown in FIG. 18) connected to the bridge circuit. Because.

  There have been several approaches to this problem. FIG. 19 shows an example of a conventional measure, which is proposed in Patent Document 1 and the like. In FIG. 19, when the resistor 10 is formed between the first conductive pattern 7 and the second conductive pattern 8, the sag 11 is easily formed. In this sag 11, the first conductive pattern 7 and the second conductive pattern 8 are raised in a convex shape on the substrate to a thickness of about 10 to 30 μm, so that the resistor paste printed thereon flows during printing. It is caused by or blurring. Since the sag 11 causes the variation of the resistance value, the variation of the resistance value is attempted to be reduced by forming the convex connection portion 9 in FIG.

  However, even if these methods are used, since the conductive pattern is raised, the sag 11 is inevitably generated.

As prior art document information relating to the invention of this application, for example, Patent Document 1 is known.
JP 2004-55946 A

  However, in the conventional configuration, as shown in FIG. 19, the resistor and the first conductive pattern 7, the second conductive pattern 8, or the convex electrode 4 are essentially raised on the insulating layer 2 in a convex shape. Therefore, when the resistor paste is inevitably printed on these electrodes, sagging 11 occurs due to the influence of the thickness of the electrodes, and the resistance value variation of the plurality of strain sensitive resistance elements 6 is likely to occur. As a result, the yield and characteristics of the strain sensor may be affected.

  In the present invention, the insulating layer and the wiring can be formed on the same plane while suppressing unevenness, minimizing the influence on the printability of the resistor paste due to the level difference of the wiring, and minimizing the characteristic variation of the strain sensitive resistor. An object of the present invention is to provide a strain sensor that can be used and a manufacturing method thereof.

  In order to achieve the above object, the present invention includes a substrate and an insulating layer in which a plurality of electrodes are embedded on the substrate, the strain sensitive resistor is provided on the electrode, and the strain sensitive resistor and the electrode A strain sensor having a protective layer covering at least a part of the resistance of the strain sensitive resistor by forming the strain sensitive resistor between a plurality of electrode patterns embedded in an insulating layer. Since the value can be stabilized and the variation in resistance value can be suppressed, the effect of stabilizing the characteristics of the strain sensor and increasing the yield can be obtained.

  A strain sensor in which a strain sensitive resistor is formed between a plurality of electrode patterns embedded in an insulating layer formed on a metal elastic body, and the strain sensitive resistor and a part of the electrode are covered with a protective layer By forming the surfaces of the insulating layer and the wiring on the same plane so that the steps are minimized, the printability of the resistor paste is improved, and a plurality of strain sensitive resistors are printed and formed at a time. Even in this case, the variation in the characteristics of the strain sensitive resistor can be suppressed, and the effect that the characteristics of the strain sensor can be stabilized can be obtained.

(Embodiment 1)
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a cross-sectional view showing a configuration of a strain sensor according to Embodiment 1 of the present invention. In FIG. 1, reference numeral 12 denotes an embedded electrode, which is embedded in an insulating layer 22 formed on a substrate 21 made of a metal elastic body so that only the surface thereof is exposed. A strain sensitive resistor 23 is formed between the plurality of embedded electrodes 12, and the strain sensitive resistor 23 and the embedded electrode 12 are protected by a protective layer 25. As shown in FIG. 1, since the wiring is embedded in the insulating layer 22 as the embedded electrode 12, no step is generated between the insulating layer 22 and the embedded electrode 12. In addition, even when the strain sensitive resistor 23 is formed by printing and firing the resistor paste, sagging or the like does not occur, and thus the variation in resistance values of the plurality of strain sensitive resistors forming the bridge circuit can be reduced. Therefore, the output of the bridge circuit can be stabilized.

  Next, the manufacturing method of the strain sensor of the present invention will be described with reference to FIGS.

  2A and 2B are cross-sectional views showing a process of forming a predetermined electrode pattern on the unsintered insulating layer. In FIG. 2A, reference numeral 13 denotes an unfired insulating layer made of, for example, unfired glass, which is dried after a predetermined glass paste is printed in a predetermined pattern on a substrate 21 made of a metal elastic body. It is. In FIG. 2B, reference numeral 14 denotes a cured wiring, which is obtained by printing a predetermined electrode paste in a predetermined pattern on the unfired insulating layer 13 and then curing the electrode paste.

  FIG. 3 is a cross-sectional view showing a process of embedding the cured wiring in the unfired insulating layer. In FIG. 3, the cured wiring 14 is formed on the unfired insulating layer formed on the substrate 21, and the press device 16 set at a predetermined temperature by the heater 17 is moved in the direction of the arrow 18. The cured wiring 14 is pushed into the unfired insulating layer 13 through the movement and the antifouling film 15.

  Thus, by using the antifouling film 15 at the time of pressing, it is possible to prevent the members peeled off from the unfired insulating layer 13 and the cured wiring 14 from adhering to the surface of the pressing device 16, Yield can be increased. Further, by using a long film material as the antifouling film 15, the antifouling film 15 can be wound up little by little at every press so that a new surface always appears. By taking up little by little in this way, even if dust or detached members adhere to the surface of the antifouling film 15 by the press, this does not affect the next press.

  FIG. 4 is a cross-sectional view showing a step of pushing the cured wiring into the unfired insulating layer. In FIG. 4, when the pressing device 16 applies pressure in the direction of the arrow 18, the cured wiring 14 is embedded in the unfired insulating layer 13.

  FIG. 5 is a cross-sectional view after pressing. In FIG. 5, the press device 16 is pulled up in the direction of the arrow 18 and the antifouling film 15 is also peeled off. Then, the hardened wiring 14 is embedded in the unfired insulating layer 13 of the substrate 21.

  FIG. 6 is a cross-sectional view showing a step of simultaneously firing the unfired insulating layer and the cured electrode. In FIG. 6A, an unfired insulating layer 13 made of unfired glass and a hardened wiring 14 embedded therein are formed on a substrate 21 made of a metal elastic body. FIG. 6B shows the result after firing. In FIG. 6B, an insulating layer 22 formed by baking the unfired insulating layer 13 and an embedded electrode 12 formed by baking the cured wiring 14 are formed on the substrate 21. Further, the unfired insulating layer 13 and the cured wiring 14 are simultaneously fired, so that they are formed as a single body with sufficient adhesion strength without a gap between them, and unevenness due to the embedded electrode 12 is generated on the surface. Can be prevented. FIG. 6C shows a case in which the strain sensitive resistor 23 is formed on the embedded electrode 12, and this configuration can prevent the occurrence of sagging or the like when the strain sensitive resistor 23 is formed.

  This will be described in more detail. First, as the unsintered insulating layer 13, a thermoplastic resin such as ethyl cellulose or butyral resin is used, and by adding a plasticizer or the like, it is soft and easily squeezed when pressed (or one whose volume is likely to change, It is desirable to use a highly compressible material. Further, it is desirable to use a hardened and hard-to-deform (low compressibility) wire after being hardened by using an electrode paste using a curable resin as the hardened wire 14.

  Thus, it becomes easy to embed in the non-fired insulating layer 13 without deforming the hardened wiring 14 by using the difference in hardness or compressibility between the non-fired insulating layer 13 and the hardened wiring 14. Further, it is possible to prevent the occurrence of a bulge (a slight rise in the pushed-in periphery) due to the cured wiring 14 embedded in the unfired insulating layer 13.

  Next, a specific description will be given. First, as the unsintered insulating layer 13, a glass paste was prepared by dispersing a predetermined glass powder in a resin solution. Here, as the resin solution, polyvinyl butyral resin was dissolved in a mixed solvent of α-terpineol and BCA (butyl carbitol acetate), and a phthalic acid-based high boiling point solvent was added as a plasticizer.

  And the said glass powder was added in this resin solution, the viscosity was adjusted, after making it disperse | distribute uniformly using the ceramic 3 roll mill. Finally, the paste was filtered using a commercially available mesh to obtain a glass paste. This glass paste is printed on the substrate 21 in a predetermined pattern using a screen printing method, then leveled for about 10 minutes, and finally dried at about 200 ° C. for about 10 to 30 minutes. An unfired insulating layer 13 as shown in A) was formed.

  In order to prevent pinholes and bubbles during screen printing from affecting the yield of the product, the unfired insulating layer 13 was repeatedly formed by printing, drying, printing, and drying to form a plurality of layers. In addition, as for the average particle diameter of glass powder, 0.1-10 micrometers is desirable. Since glass powder of less than 0.1 μm is expensive, it may affect the product price. When the thickness exceeds 10 μm, a large glass powder may be mixed with 20 μm or 30 μm, and in some cases 40 μm and large glass powder, which may be a problem when printing or baking the glass paste.

  On the other hand, as a material for the curable wiring 14, an electrode paste was prepared by dispersing commercially available silver palladium powder as a conductive powder in a curable resin solution. Here, as the curable resin solution, a curable resin such as a commercially available epoxy resin or xylene resin was used, and a solvent such as α-terpineol or BCA (butyl carbitol acetate) was added for viscosity adjustment as needed. Then, the silver palladium powder is added to the curable resin solution, and a three roll mill is used so that the silver palladium powder is not deformed (between the rolls, the silver palladium powder is flaked). When crushed, it may affect the sinterability). And after adjusting the viscosity, finally, this paste was filtered using a commercially available mesh (net) to complete an electrode paste.

  Then, this electrode paste was printed in a predetermined shape on the unfired insulating layer 13 as shown in FIG. The electrode paste was cured at 200 ° C. for 10 to 30 minutes. Here, as the curable resin, in addition to the thermosetting type, a UV (ultraviolet) curable type, an EB (electron beam) curable type, a two-component reaction curable type, or the like can be used. The curable resin is preferably mainly composed of C, H, N, and O, and preferably does not contain elements such as S, Cl, P, Na, and K. In addition to these elements, metal atoms may affect the characteristics after firing in some cases.

  The thus prepared sample was embedded with the hardened wiring 14 inside the unfired insulating layer 13 using a press as shown in FIGS. 3, 4, and 5. Then, as shown in FIG. 6, with the cured wiring 14 embedded in the unfired insulating layer 13, these were baked together to form the insulating layer 22 with the embedded electrode 12 embedded in the surface.

  When a predetermined strain-sensitive resistor paste was printed on the embedded electrode 12 in a predetermined shape, it could be formed with high accuracy without causing a resistance value variation such as bleeding or sagging. As a result, the resistance value variation of the plurality of strain sensitive resistor elements constituting the bridge circuit can be suppressed within a range that can be adjusted by the semiconductor chip with a high yield of 99.8% or more.

  For comparison, a conventional example having the shape shown in FIG. 17 was produced. First, using the same members as described above, samples were prototyped in the same order as in FIGS. 2A and 2B, and then the pressing steps of FIGS. 13), the unfired insulating layer 13 and the cured wiring 14 were fired at the same time. Then, when a predetermined strain sensitive resistor paste is printed on this, sagging or blurring occurs, the resistance value variation of the plurality of strain sensitive resistor elements constituting the bridge circuit increases, and the range that cannot be adjusted by the semiconductor chip (Something that does not fit the specs) occurred from several percent to several tens of percent.

  Therefore, as shown in FIG. 19, the pattern shape of the wiring connected to the strain sensitive resistor has been devised, but there is no change in sagging or blurring, and the resistance value variation does not fall within the specifications. A certain amount of sample was generated.

  For comparison, as the cured wiring 14, instead of the thermosetting resin solution, a resin solution (polyvinyl butyral resin dissolved in an organic solvent or a plasticizer) used for producing a glass paste is used. Used, silver palladium powder is added here, and kneaded with a three-roll mill to prepare a predetermined paste, which is formed on the unfired insulating layer 13 as shown in FIGS. 2 (A) and 2 (B). Printed (hereinafter referred to as conventional wiring).

  When this was pressed as shown in FIGS. 3, 4, and 5, the conventional wiring was crushed so that the pattern spread on the surface without biting into the unfired insulating layer 13. The crushed conventional wiring was fired simultaneously with the unfired insulating layer 13. Thereafter, a strain sensitive resistor 23 was formed in the same manner. However, when the resistance value variation of a plurality of strain sensitive resistors constituting the bridge circuit was measured in the samples prepared in this way, some of them were more than several tens of percent, and some of them were short-circuited between the wires, and the product yield was It was low.

  Thus, in the case of the present invention, when the electrode pattern is embedded in the unsintered insulating layer by pressing, the unsintered insulating layer is desirably soft and excellent in compressibility (or highly air permeable). As one electrode pattern, it is desirable that the electrode pattern is hard and hard to be compressed (not easily deformed by pressing force).

  In FIG. 6B, the thickness of the insulating layer 22 is preferably 10 to 500 μm. When the thickness is less than 10 μm, there is a possibility that an insulation failure occurs between the substrate 21 and the embedded electrode 12. In addition, when the thickness exceeds 500 μm, the material cost of the insulating layer 22 may increase. Here, the insulating layer 22 has a thickness sandwiched between the substrate 21 and the embedded electrode 12. It is desirable to form the insulating layer 22 with one or more layers.

  For example, by forming the insulating layer 22 with a plurality of layers, even if dust adheres to the insulating layer 22 or bubbles are generated, it is difficult to cause an insulation failure. In addition, in FIG. 6C, the thickness of the insulating layer 22 made of glass paste is preferably 10 to 500 μm. When the thickness of the insulating layer 22 made of glass paste covering the embedded electrode 12 is less than 10 μm, it may be easily affected by dust and pinholes. Moreover, when thickness exceeds 500 micrometers, since the usage-amount of a glass paste increases, there exists a possibility that cost may become high.

The thickness of the substrate 21 is desirably 1 to 100 mm. When the thickness of the substrate 21 is less than 1 mm, the proof stress required as a strain sensor may not be obtained. Further, when the thickness of the substrate 21 exceeds 100 mm, it becomes difficult to process the substrate 21 (for example, punching with a mold or laser processing). The area of the substrate 21 is preferably 0.1 to 1000 cm ² . When the area of the substrate 21 is 0.1 cm ² or less, the plurality of strain sensitive resistors 23 can be as small as 0.1 cm ² even when the step between the surface of the insulating layer 22 and the embedded electrode 12 is within ± 5 μm. For example, the bridge circuit requires four strain sensitive resistors 23 and is difficult to form by printing. If the area of the substrate 21 exceeds 1000 cm ² , the material cost of the substrate 21 may increase and affect the cost of the strain sensor.

  Moreover, as for the thickness of the hardened | cured wiring 14, 50-200 micrometers is desirable. When the thickness is less than 5 μm, the resistance value of the embedded electrode 12 is increased, which may affect the characteristics of the strain sensor. On the other hand, when the thickness is greater than 200 μm, the amount of wiring members increases, resulting in an increase in cost. The use of an AgPd-based electrode material for the embedded electrode 12 can improve the solder erosion resistance, and it is easy to directly mount a chip component, a semiconductor chip or the like by soldering.

  Further, it is desirable that the unsintered insulating layer 13 be one or more layers. For example, by making the unsintered insulating layer 13 into two or three layers, it is possible to reduce the effects of adhesion of glass paste dust, pinholes, and the like.

  In the case where the unsintered insulating layer 13 is multilayered, it can be formed by further applying a glass paste after applying and drying the glass paste. The insulating layer 22 in contact with the substrate 21 is crystallized glass, and the crystallization rate is desirably 40% or more. When the insulating layer 22 in contact with the substrate 21 is amorphous glass, it may be softened during the firing of the strain sensitive resistor 23 to reduce the adhesive strength at the interface with the substrate 21.

  Therefore, when the insulating layer 22 in contact with the substrate 21 is made of crystallized glass, the glass does not soften as the insulating layer 22 when the strain sensitive resistor 23 is fired, so that the adhesive force at the interface with the substrate 21 does not decrease. When crystallized glass is used, the crystallization rate is preferably 40% or more. When the crystallization rate is less than 40% (that is, 50% or more of the insulating layer is amorphous glass), the amorphous glass component is softened during the firing of the strain sensitive resistor 23 and affects the adhesive strength with the substrate 21. May give.

  The crystallization rate of the insulating layer 22 can be evaluated using equipment such as X-ray diffraction. The upper limit of the crystallization rate is sufficient up to about 80%. This is because, in order to increase the crystallization rate of the crystallized glass to 80% or more, the firing conditions and the glass composition for increasing the crystallization rate are optimized, which may affect the process cost. Therefore, when used in a strain sensor, it is desirable that the crystallization rate is 40 to 100% (substantially 80% or less is sufficient).

  The thickness of at least the embedded electrode 12 is preferably 5 to 100 μm. It is desirable to form a space (for example, a solder mounting portion called a land or the like) in which a chip component can be mounted on a part of the embedded electrode 12 as necessary. Thus, by forming the strain sensitive resistor 23 in a part of the embedded electrode 12 and mounting a chip component such as a semiconductor or a square chip resistor in the other part, the strain sensor can be reduced in size and cost. I can plan.

  In the first embodiment of the present invention, the embedded electrode 12 of the mounting portion such as soldering can be embedded in the insulating layer 22 as necessary. When the thickness of the embedded electrode 12 is less than 5 μm, the wiring resistance increases, which may affect the characteristics as a strain sensor. Further, when the thickness of the embedded electrode 12 exceeds 100 μm, the material cost of the member for forming the embedded electrode 12 increases, which may increase the product cost.

In addition, as for the thickness of the strain sensitive resistor 23, 5-50 micrometers is desirable. When the thickness of the strain sensitive resistor 23 is 5 μm or less, there is a possibility that the strain sensitive resistor 23 may be affected by a minute step even when the step between the surface of the insulating layer 22 and the embedded electrode 12 is within ± 5 μm. In addition, the influence of the insulating layer 22 serving as a base of the strain sensitive resistor 23 (for example, mutual diffusion of glass material or the like) is easily received. Further, when the thickness of the strain sensitive resistor 23 is larger than 50 μm, the material cost of the strain sensitive resistor 23 is increased, which may increase the product cost. The area of the strain sensitive resistor 23 is preferably 0.1 to 100 mm ² . When the area of the strain sensitive resistor 23 is less than 0.1 mm ² , even when the step difference between the surface of the insulating layer 22 and the embedded electrode 12 is within ± 5 μm as in the strain sensor of the present invention, the strain sensitive resistor 23. The pattern becomes smaller and may be affected by this minute step.

Further, when the area of the strain sensitive resistor 23 exceeds 100 mm ² , the material cost of the strain sensitive resistor 23 increases, which may increase the product cost. As the resistance value of the strain sensitive resistor 23, a commercially available resistor paste using ruthenium oxide having a high GF (Gauge Factor, gauge factor, rate of change in resistance value against strain) can be selected and used.

  Further, at least the entire surface of the strain sensitive resistor 23 and the embedded electrode 12 or a part of the strain sensitive resistor 23 and the embedded electrode 12 may be covered with a protective layer (not shown) having a thickness of 10 to 500 μm with resin or glass. it can. Thus, by covering at least a part of the strain sensitive resistor 23 and the embedded electrode 12 with the protective layer, it is possible to protect from the external environment of the strain sensitive resistor 23 and the embedded electrode 12 and to improve reliability. Can do.

  Further, by exposing a part of the embedded electrode 12, various chip parts and the like can be mounted using the embedded electrode 12, and the strain sensor can be reduced in size and cost. The thickness of the protective layer covering at least the strain sensitive resistor 23 is preferably 10 to 500 μm. When the thickness of the protective layer is less than 10 μm, pinholes may be generated in the protective layer. If the thickness of the protective layer exceeds 500 μm, the material cost of the protective layer forming member may affect the cost.

  Here, the protective layer is preferably resin or glass. When a resin is used for the protective layer, the temperature for forming the protective layer can be lowered, and the characteristics of the strain sensitive resistor 23 are not affected by the heat during the formation of the protective layer. When glass is used for the protective layer, the temperature at the time of forming the protective layer is lowered by 100 ° C. or more from the firing temperature of the strain sensitive resistor 23 (for example, when the firing temperature of the strain sensitive resistor is 850 ° C., The firing temperature of the glass is preferably less than 700 ° C.). This is to suppress the influence on the characteristics of the strain sensitive resistor by the heat treatment at the time of firing the glass material to be the protective layer.

  Further, as shown in FIG. 6B, the substrate 21 has an insulating layer 22 made of glass paste and a buried electrode 12 embedded in the surface of the insulating layer 22 fired at the same time, so that the insulating layer shown in FIG. 22 and the embedded electrode 12 are formed, and the firing temperature is preferably 500 to 950 ° C. here. When the embedded electrode 12 and the glass paste are fired at a low temperature of less than 500 ° C., the embedded electrode 12 is not sufficiently sintered and the wiring resistance does not decrease or the sintering strength of the glass paste cannot be obtained sufficiently. There is a case. Further, when the firing temperature is higher than 950 ° C., the substrate 21 may be oxidized and discolored, and the proof stress may be reduced.

  It is also effective to add a glass paste in the range of 0.5 to 20 wt% to the paste for forming the embedded electrode 12 in advance. By previously adding a glass component to the interior of the embedded electrode 12 in this way, even if the embedded electrode 12 is simultaneously fired in the state of being embedded in the glass paste, it is possible to prevent the occurrence of peeling or cracking at the mutual interface. Can do. In addition, when the addition amount of a glass paste is less than 0.5 wt%, the addition effect may not be acquired. On the other hand, if the added amount exceeds 30 wt%, the resistance value of the buried electrode 12 increases, which may affect the characteristics of the strain sensor.

  In addition, as for the press apparatus 16 shown in FIG.3, FIG.4, FIG.5 etc., it is desirable to heat to fixed temperature as needed. For example, a temperature of 50 to 200 ° C., particularly 70 to 150 ° C. is desirable. If the temperature is too high, the unsintered insulating layer 13 becomes too soft and easily deforms, or adheres to the surface of the antifouling film 15. Therefore, when the press device 16 is heated, a temperature of room temperature to 250 ° C. is desirable.

  The pressing time is preferably 0.1 second to 10 minutes. If the pressing time is less than 0.1 seconds, a sufficient embedding effect may not be obtained. Further, when the pressing time exceeds 10 minutes, productivity may be affected.

The press pressure is preferably 1-1000 kg / cm ² . When the pressure is less than 1 kg / cm ² , it may be difficult to embed the cured wiring 14 in the unfired insulating layer 13. If the press pressure exceeds 1000 kg / cm ² , the rigidity is high and an expensive press device having a large total pressure is required, which increases the manufacturing cost and may deform the unfired insulating layer 13.

  In the strain sensor of the present invention, thermal adhesiveness can be obtained by using a thermoplastic resin for the glass paste of the insulating layer 22. As this thermoplastic resin, PVB (polyvinyl butyral) resin or acrylic resin can be used. Such resins are widely used in the production of ceramic green sheets, and a resin having high adhesion (or thermal transferability) to the substrate 21 may be selected from these resins.

  The insulating layer 22 made of glass paste can be a glass paste in which a predetermined glass powder is dispersed at 40 to 90 wt% in a resin solution made of a resin and an organic solvent. This is because when the proportion of glass powder in the glass paste is less than 40 wt%, pinholes are likely to occur in the insulating layer 22 formed after drying. When the glass powder ratio exceeds 90 wt%, the resin solution ratio is less than 10 wt%, the fluidity of the glass paste is lowered, and pinholes are easily generated in the insulating layer 22 when the embedded electrode 12 is covered. .

  The embedded electrode 12 has an electrode paste in which conductive powder is dispersed at 40 to 90 wt% in a resin solution composed of a resin and a solvent, printed in a predetermined shape on a resin film, and dried at a temperature of 50 to 200 ° C. It is desirable. When the conductive powder contained in the electrode paste is less than 40 wt%, the thickness of the embedded electrode 12 is reduced after firing, the wiring resistance is increased, and the characteristics of the strain sensor are affected. When the conductive powder is more than 90 wt%, the resin solution ratio is less than 10 wt%, so that the fluidity of the electrode paste is lowered and the printing of the predetermined pattern is affected. Moreover, when the drying temperature of electrode paste is less than 50 degreeC, drying time may become long and production cost may be raised. When the drying temperature exceeds 200 ° C., the resin film may be thermally deformed.

  The antifouling film 15 does not need to be peeled off after being pressed. For example, when an anti-smudge film 15 that is flammable and thin is used, the anti-smudge film 15 is left on the unfired insulating layer 13 as it is, and simultaneously fired together with the unfired insulating layer 13 and the cured wiring 14. May be.

  As described above, after the non-fired insulating layer in which the electrode pattern is embedded on the substrate made of the metal elastic body, the non-fired insulating layer and the electrode pattern are simultaneously fired to form the electrode embedded insulating layer, A strain sensor can be manufactured at low cost by forming a strain sensitive resistor on the electrode and further covering at least a part of the strain sensitive resistor and the electrode-embedded insulating layer with a protective layer.

  Similarly, after forming an unfired insulating layer on a substrate made of a metal elastic body, an electrode pattern is embedded on the unfired insulating layer, and the unfired insulating layer and the electrode pattern are simultaneously fired to embed the electrode. After forming the insulating layer, a strain-sensitive resistor is formed on the electrode, and at least a part of the strain-sensitive resistor and the electrode-embedded insulating layer is covered with a protective layer to manufacture a strain sensor. Also good. Further, after forming an unsintered insulating layer on the metal elastic body, forming an electrode pattern on the unsintered insulating layer, and embedding the electrode pattern in the unsintered insulating layer using a press device, After firing the unsintered insulating layer and the electrode simultaneously to form an electrode buried insulating layer, a strain sensitive resistor is formed on the electrode, and at least one of the strain sensitive resistor and the electrode buried insulating layer is formed. The present strain sensor can also be manufactured by covering the portion with a protective layer.

  The unsintered insulating layer is preferably dried in the range of 50 to 400 ° C. after a glass paste in which glass powder is dispersed in a resin solution is printed in a predetermined shape on a metal elastic body. If the glass paste is less than 50 ° C., the drying time becomes too long. When the temperature is higher than 400 ° C., the glass paste is hardened or the plasticizer in the glass paste is evaporated, which may make it difficult to embed the hardened wiring. Further, by adjusting the particle size distribution of the glass powder, the resin amount, or the drying conditions of the glass paste, the finished unsintered insulating layer can be brought into a state having a high porosity. Even if the electrode pattern is pressed and embedded in the unfired insulating layer produced in this way, it is possible to prevent the occurrence of a bulge (a slight increase in the pressed periphery).

  Moreover, as for the thickness of a non-baking insulating layer, 10-1000 micrometers is desirable. When the thickness is less than 10 μm, predetermined insulation may not be obtained. If the thickness exceeds 1000 μm, material costs may affect the product price.

  The unsintered insulating layer is preferably formed of glass powder, resin, and some plasticizer. As an average particle diameter of glass powder, 0.5-10 micrometers is desirable. When the average particle diameter of the glass powder is less than 0.5 μm, the cost associated with the pulverization and classification of the glass powder increases. Moreover, when the average particle diameter of glass powder exceeds 10 micrometers, when the hardened wiring 14 is embedded, the pattern of the hardened wiring 14 may be damaged. As the plasticizer, commercially available phthalic high-boiling solvents (for example, various products such as DOP, DBB, BBP are commercially available) can be used.

  Moreover, it is desirable to add such a plasticizer in the range of 5 to 500 wt% with respect to the resin. When the plasticizer is less than 5 wt%, the effect of adding the plasticizer (that is, the effect of softening the resin or the effect of lowering the glass transition temperature of the resin) may not be obtained. On the other hand, when the added amount of the plasticizer exceeds 500 wt%, the plasticizer may come out of the unfired insulating layer (or the plasticizer may ooze out on the surface).

  Some resins may be soft without adding a plasticizer. Therefore, when the glass transition temperature (also referred to as Tg, which can be measured by thermal analysis) of the resin used is 20 ° C. or higher, it is desirable to add a plasticizer. Further, when Tg is less than 0 ° C., it can be used without adding a plasticizer. However, such a material having a low Tg is rubbery even at room temperature and may be difficult to handle. Therefore, the resin used in the present invention is desirably a resin having a Tg of 0 to 100 ° C. and a plasticizer added in the range of 5 to 500 wt%.

  By using such a combination of resin (or resin + plasticizer), the unfired insulating layer is hard at room temperature (at least without being sticky to rubber) and is heated to a certain temperature (for example, 100 ° C. When set to the front and back), the unsintered insulating layer can be softened (the cured wiring 14 is easily recessed) during pressing. After the press is completed (the state shown in FIGS. 5 and 13), after the press is released, the unfired insulating layer 13 in which the cured wiring 14 is embedded is easily cooled (if necessary, a cooling device or a cooling plate). The press cycle can be accelerated.

  In particular, when the base film 20 is peeled, the peelability of the cured wiring 14 from the base film 20 can be made smooth by cooling the unfired insulating layer 13 and the cured wiring 14.

  As described above, when the cured wiring 14 is formed on the unfired insulating layer 13 by printing an electrode paste in which conductive powder is dispersed in a curable resin solution in a predetermined shape and then curing the electrode paste ( 3 and 7), even when formed on the base film (in the case of FIGS. 10 and 11), the surface of the unfired insulating layer 13 is not damaged without damaging the pattern shape. Can be bitten. In addition, as a curable resin solution, what melt | dissolved curable resin in the predetermined | prescribed organic solvent can be used.

  The thickness of the electrode pattern is preferably 5 to 200 μm. When the thickness of the electrode pattern is less than 5 μm, the solder mounting strength of each chip resistor, semiconductor component, or connector may be affected. Moreover, when thickness exceeds 200 micrometers, the influence on material cost will increase. Moreover, it is desirable that this electrode pattern is composed of conductive powder and a curable resin when pressed (or embedded) at least on the unfired insulating layer. During this pressing (or embedding), if a certain amount or more of a solvent component (or plasticizer component, etc.) remains in the electrode pattern, the electrode pattern is deformed in an undesired direction (for example, the pattern is crushed). Etc.). Further, as the conductive powder, conductive metal powder (or alloy powder) such as silver, palladium, platinum, and gold can be used.

  The curing temperature of the electrode paste is preferably 60 to 400 ° C. If it is less than 60 ° C., sufficient curing cannot be obtained (or a resin material that cures at a low temperature of less than 60 ° C. is likely to be cured even at room temperature, which may shorten its pot life). When the curing temperature exceeds 400 ° C., the resin may be decomposed by heat. The heat treatment time is preferably 0.1 second to 30 minutes. If thermosetting is attempted in less than 0.1 seconds, it is necessary to use a special and expensive resin. When the heat treatment time exceeds 30 minutes, productivity may be affected, and the curing itself needs to be performed by batch processing, which makes process management complicated. Therefore, by setting the heat treatment time to 0.1 seconds to 30 minutes (desirably 1 to 10 minutes), a highly productive method such as roll-to-roll can be selected.

  The unsintered insulating layer and the electrode pattern embedded therein are preferably fired at 600 to 1000 ° C. at the same time. When the firing temperature is less than 600 ° C., the wiring and the insulating layer may be insufficiently sintered. When the firing temperature exceeds 1000 ° C., the firing furnace is expensive, and it is necessary to use an expensive special member for the substrate 21.

The press pressure is preferably 1 to 1000 kg / cm ² . When the pressing pressure is less than 1 kg / cm ² , the cured wiring 14 may not be sufficiently embedded in the unfired insulating layer 13. On the other hand, when the pressing pressure exceeds 1000 kg / cm ² , the unfired insulating layer 13 itself may be abnormally deformed or cracked.

  The pressing time is preferably 0.1 second to 10 minutes. When the pressing time is less than 0.1 seconds, the unfired insulating layer 13 is elastically deformed, and after the press is removed, the cured wiring 14 to be embedded is not embedded, but rises to the surface (unfired insulation). In some cases, the elasticity of the layer 13 returns to the original state). Further, when the pressing time exceeds 10 minutes, productivity may be affected. Therefore, the pressing time (the time during which the pressing pressure is actually generated on the surface to be pressed) is desirably 1 second or more.

  The pressing temperature is preferably 20 to 300 ° C. When the press is less than 20 ° C., the cured wiring 14 may not sufficiently penetrate the unfired insulating layer 13. When the press temperature is 300 ° C. or higher, the unfired insulating layer 13 may be abnormally deformed by heat. Therefore, desirably, by setting the pressing temperature in the range of 80 ° C. or more and 150 ° C., the hardened wiring 14 can be easily embedded in the unfired insulating layer 13, and thermal deformation of the unfired insulating layer 13 itself (for example, the external dimensions are In some cases, it may change, crack, or crush interlayer connection portions such as via holes formed in the unfired insulating layer 13).

  As described above, by pressing each member (the electrode pattern side to be embedded is hard and the embedded unsintered insulating layer is soft and fluffy) under the appropriate pressing conditions described above, unsintered insulation of the part where the electrode pattern is not embedded is performed. The layer can also be densified during this pressing, reducing the porosity in the green insulating layer. As a result, it is needless to say that voids (small voids and holes) are not easily generated inside the insulator formed by firing the unfired insulating layer, and the insulation resistance, the insulation yield, and various reliability can be improved.

In the case of using the amorphous glass insulating layer, SiO ₂ is 40 to 80 wt%, CaO is 5 to 15 wt%, PbO is 3~15wt%, Al ₂ O ₃ is 110 wt.%, ZrO ₂ is 2~10wt % Amorphous glass can be used. Since such amorphous glass has high matching properties with the strain sensitive resistor, it can be used as a base glass for the strain sensitive resistor.

Here, when the proportion of SiO ₂ is less than 40 wt% or exceeds 80 wt%, the matching with the strain sensitive resistor may change (for example, the resistance value or TCR of the resistor may change). TCR means temperature dependency of resistance value). Also, when the proportion of CaO or PbO is small (CaO is less than 5 wt%, PbO is less than 3 wt%) or too much (CaO is 15 wt% or more, PbO is 15 wt% or more), it affects the matching with the strain sensitive resistor. May give. This is thought to be because these CaO and PbO are part of the glass element that constitutes the strain sensitive resistor.

(Embodiment 2)
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  7 to 9 are cross-sectional views illustrating a method of embedding only a predetermined portion of the wiring by providing protrusions on the surface of the pressing device.

  In FIG. 7, protrusions are formed on the surface of the press device 16. Then, as shown in FIG. 8, by pressing the protruding portion of the pressing device 16 against the surface of the unfired insulating layer 13, only the necessary portion of the cured wiring 14 can be embedded in the unfired insulating layer 13. Further, at this time, the press device 16 is subjected to processing such as applying R (smoothed in a curved shape) or trapezoid (smoothing so that the projecting portion is not sharp) to perform pressing. In this case, only necessary portions can be embedded in the unfired insulating layer 13 while suppressing damage to the cured wiring 14. 7 to 9, the antifouling film 15 and the like are omitted.

  FIG. 10 is a cross-sectional view showing a state in which a product in which only a part of the predetermined wiring is embedded is manufactured. In FIG. 10A, an unfired insulating layer 13 is formed on a substrate 21, and a hardened wiring 14 partially embedded is formed on the unfired insulating layer 13. Next, the unsintered insulating layer 13 and the cured wiring 14 are fired simultaneously, so that the insulating layer 22 and part of the insulating layer 22 are embedded on the substrate 21 as shown in FIG. The embedded electrode 12 and the convex electrode 24 are formed.

  In FIG. 10, the difference between the embedded electrode 12 and the convex electrode 24 is only whether or not the embedded electrode 12 is embedded in the insulating layer 22. This is because the cured wiring 14 is not sintered in FIGS. 7 to 10A. The only difference is whether it was embedded in the layer 13 with a press. Thus, by simultaneously firing the hardened wiring 14 and the unfired insulating layer 13, it is easy to match the sintering shrinkage characteristics of each other, and it is possible to prevent the occurrence of cracks (breaking or peeling), insufficient adhesion, and the like. Next, a strain sensitive resistor 23 is formed on the embedded electrode 12 as shown in FIG. Finally, as shown in FIG. 10D, a protective layer 25 is formed so as to cover the strain sensitive resistor 23, the embedded electrode 12, and the convex electrode 24.

  Here, in FIG. 10D, reference numeral 19 denotes a window, which is a portion without the protective layer 25. Predetermined convex electrodes 24 and embedded electrodes 12 are exposed inside the window 19 formed in the protective layer 25, and various chip components, semiconductor elements, and the like are mounted by using the window 19. be able to. Moreover, since the influence of the water | moisture content to the convex electrode 24 or the embedded electrode 12 can be suppressed by using a glass-type material as the protective layer 25, even when a silver-type material is used for these electrode members, it is a migration process. Occurrence can be prevented.

  Thus, by devising the shape of the press device 16, only necessary portions can be embedded in the insulating layer 22 without damaging the cured wiring 14, so that various product forms can be dealt with.

(Embodiment 3)
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  In Embodiment 3, a process of embedding a wiring member formed on a film in an insulating layer will be described with reference to FIGS.

  FIG. 11 is a cross-sectional view showing a stage before the wiring member formed on the film is pressed against the insulating layer. In FIG. 11, reference numeral 20 denotes a base film, and a hardened wiring 14 is formed on the base film 20 in a predetermined pattern. Then, as shown in FIG. 11, the press device 16 adjusted to a predetermined temperature by the heater 17 is moved in the direction of the arrow 18.

  FIG. 12 is a cross-sectional view showing a process of pressing the wiring member formed on the film against the insulating layer. In FIG. 12, the cured wiring 14 formed on the base film 20 is pressed against the surface of the unfired insulating layer 13 by the pressing device 16.

  FIG. 13 is a cross-sectional view showing a state where a wiring member formed on a film is transferred to an insulating layer. In FIG. 13, the press device 16 is pulled up in the direction of the arrow 18, and the base film 20 is further peeled off from the unfired insulating layer 13, so that the cured wiring 14 formed on the surface of the base film 20 is unfired. It can be embedded in the surface of the insulating layer 13.

  This will be described in more detail. First, as the base film, a resin film such as a commercially available PET film can be used. And it becomes easy to peel off the base film 20 as shown in FIG. 13 from the unbaked insulating layer 13 by using the resin file which surface-treated beforehand as the base film 20. When the surface treatment is not performed here, when the base film 20 is peeled off from the non-fired insulating layer 13, the forming member of the non-fired insulating layer 13 may stick to the base film 20 and be peeled off. As such surface treatment, various treatment methods such as silicone treatment, fluorine treatment, and resin coating have been proposed by the film manufacturer, and an optimum one (or an appropriate product number) may be selected from these methods.

  Here, when the surface treatment is too strong (the water and oil repellent treatment is strong), it is difficult to form the cured wiring 14 on the base film 20 (the electrode ink may be repelled and printing may not be performed properly). Further, the cured wiring 14 may be easily peeled off from the base film 20 (it may be peeled off even if it is touched with a fingertip) and may be difficult to use.

  As the base film 20, it is desirable to use a resin film having a thickness of 10 to 200 μm. In the case of a film having a thickness of less than 10 μm, there is a case where it is difficult to handle because there is no elasticity of the film. Further, when the film thickness is 200 μm or more, the film cost increases. Moreover, as a film, films, such as a polyester film, a polypropylene film, PET (polyethylene terephthalate) film, PEN, a polyimide, can be used.

  In addition, electrode printing on such a film can be continuously performed at high speed by using a screen printing method, a rotary screen printing method, or the like. Further, the alignment (positioning) of the electrode pattern formed on the film and the unfired insulating layer 13 is not limited to image recognition (for example, holes and markings formed in the substrate 21 can be read), but also to external alignment, sprocket It is also possible to perform alignment using pins such as holes or holes.

(Embodiment 4)
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  In Embodiment 4, the case where an internal electrode is formed inside an insulating layer will be described with reference to FIGS. FIG. 14 is a cross-sectional view showing a state in which an internal electrode is formed inside an insulating layer in which a buried electrode is formed. In FIG. 14, 26 is an internal electrode. By forming the internal electrode 26 inside the insulating layer 22 in this way, the noise resistance of the strain sensor (the sensor output is affected by external noise or the like) can be improved. Furthermore, since a part of the embedded electrode 12 and the convex electrode 24 (not shown in FIG. 14) can be connected to the internal electrode 26 through a through hole or the like as required, the wiring can be multilayered. It is easy, the substrate 21 can be downsized, and the product cost can be reduced.

  The thickness of the internal electrode 26 is desirably 1 to 50 μm. When the thickness is less than 1 μm, the resistance value may increase or the pattern may be easily cut. If it is thicker than 50 μm, the material cost of the internal electrode 26 may affect the product cost.

  Moreover, as a member which comprises the internal electrode 26, what mainly has silver is desirable. Then, by adding palladium, platinum or the like as necessary, it becomes easy to match the sintering profile of the internal electrode 26 with that of the glass, and as a result, the electrode materials constituting the internal electrode 26 together with the unfired insulating layer 13 are collectively collected. Even when baked, the occurrence of delamination (delamination) or the like can be prevented. Moreover, the adhesive strength of the insulating layer 22 and the internal electrode 26 can be raised by adding a glass member to the internal electrode 26 in the range of 0.5-20 wt% as needed. In this case, when the addition amount of the glass member is less than 0.5 wt%, the addition effect may not be obtained. On the other hand, if the addition amount exceeds 20 wt%, the resistance value of the internal electrode 26 increases, which may affect the characteristics of the completed strain sensor.

(Embodiment 5)
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  In Embodiment 5, the case where the insulating layer is formed using a plurality of layers of different members will be described with reference to FIGS. FIG. 15 is a cross-sectional view showing a state in which the insulating layer on which the embedded electrode is formed is composed of a plurality of layers of different members. In FIG. 15, 27 is a first insulating layer, and 28 is a second insulating layer. As described above, the embedded electrode 12 is formed in the second insulating layer 28, and the first insulating layer 27 is formed between the substrate 21 and the second insulating layer 28. The insulation yield with respect to the substrate 21 can be increased. In addition, by using a plurality of insulating layers, a material having high adhesive strength with the substrate 21 is selected as the first insulating layer 27 in contact with the substrate 21, and the second insulating layer 28 in contact with the strain sensitive resistor 23 is sensitive. A thing with high matching property with the distortion resistor 23 can be selected.

  In particular, in the strain sensitive resistor 23, the resistance value creeps (a phenomenon in which the resistance value changes with time) or the GF linearity is affected (in other words, depending on the degree of matching with the underlying insulating layer). The rate of change in resistance value with respect to strain may not be constant), or the TCR of the resistor (temperature change in resistance value or change in resistance value depending on temperature) may increase.

  When the TCR becomes large, the output of the strain sensor is easily affected by the temperature of the strain sensor. When this strain sensor is used for a smart airbag in a passenger seat of a passenger car, the output is affected by the room temperature of the passenger car. There is a possibility. As described above, the matching between the strain sensitive resistor 23 and the second insulating layer 28 formed immediately below the strain sensitive resistor 23 is not only a material design but also a thermal contraction or heat of each other at the time of firing. Also susceptible to differences in expansion coefficient. Furthermore, the strain sensitive resistor 23 itself is limited to a very few product numbers that can withstand practical use, considering GF, its reliability, TCR, and the like. Because of this background, even if the material of the substrate 21 changes, the strain sensitive resistor 23 and the second insulating layer 28 often have to be used as they are.

  In such a case, it is only necessary to develop a new material for the first insulating layer 27 in accordance with the material of the new substrate 21 by using the fifth embodiment, and the second insulating layer 28 and the strain sensitive resistor 23 are conventionally used. The same thing can be diverted. Thus, the development speed can be increased, and a stable quality strain sensor can be provided to the user at low cost.

  This will be described in more detail. For example, it is desirable to use crystalline glass as the first insulating layer in contact with the substrate 21. Crystalline glass has a crystallizing temperature exceeding the crystallization temperature, and the remelting temperature after the crystallization is 1000 ° C. or higher. Therefore, it is known that crystallized glass has high strength and is difficult to break, and silver or the like is difficult to diffuse compared to general amorphous glass. Therefore, the crystallized glass used in the present invention preferably has a crystallization temperature in the range of 550 to 750 ° C.

When the crystallization temperature of the crystallized glass is within this range, the remelting temperature after crystallization is 1200 ° C. or higher. Therefore, even if the strain sensitive resistor 23 is fired at a high temperature of 800 to 900 ° C., Since the crystallized glass remains solidified on the substrate 21 and does not re-dissolve, delamination (delamination) and voids (small holes) do not occur when the strain sensitive resistor 23 is fired. As such crystallized glass, it is desirable to use a crystallized glass composition mainly composed of MgO, B ₂ O ₃ and SiO ₂ . Since these members have a relatively large coefficient of thermal expansion (for example, 90 × 10 ⁻⁷ to 130 × 10 ⁻⁷ / ° C.), it is possible to use a SUS material (or a SUS steel material) having excellent heat resistance for the substrate 21. it can. Moreover, since such SUS steel has high yield strength, the reliability of the strain sensor thus completed can be improved.

  Next, the case where an internal electrode is formed inside an insulating layer made of a plurality of different members will be described with reference to FIG. FIG. 16 is a cross-sectional view showing a state in which the insulating layer in which the embedded electrode is formed is composed of a plurality of layers of different members, and an internal electrode is formed in the insulating layer. In FIG. 16, a plurality of insulating layers such as a first insulating layer 27 and a second insulating layer 28 are formed on a substrate 21, and an internal electrode 26 is built in a part of the insulating layer. Has been. Thus, by forming the internal electrode 26 inside the first insulating layer 27, the internal electrode 26 can be used as a shield layer or a GND layer (ground layer), and the strain sensor can be used as a housing, various facilities, Alternatively, when mounted on a passenger car or the like, the influence of noise from such a case can be suppressed.

  This will be described in more detail. First, a first glass paste prepared by dispersing crystallized glass powder in a resin solution using three rolls so as to be an insulating layer 27 after firing was printed on the substrate 21 in a predetermined shape. Further thereon, an internal electrode paste mainly composed of silver was printed so as to become the internal electrode 26 after firing. Further, the first glass paste was printed in a predetermined shape on the electrode paste so as to become the insulating layer 27 after firing. And on this, the amorphous glass powder was disperse | distributed to the resin solution, and the 2nd glass paste was printed on this by the predetermined shape.

  Then, as shown in FIGS. 11 to 13, the hardened wiring 14 was embedded in the surface of the second glass paste. Then, the first glass paste, internal electrode paste, second glass paste, cured wiring, and the like were fired at the same time. Thereafter, a strain resistor paste was printed thereon so that a plurality of strain resistor pastes were formed in a predetermined pattern so as to form a bridge circuit, and baked at 850 ° C. to form a strain sensitive resistor 23. Then, an overcoat glass paste was printed in a predetermined pattern so as to cover at least the strain-sensitive resistor 23 and fired at 600 ° C.

  The firing temperature of the glass paste for overcoat is preferably 500 to 750 ° C. When baked at 750 ° C. or higher, glass components such as PbO contained in the strain sensitive resistor are redissolved and easily diffused into the overcoat material. As a result, the resistance value shifts or the resistance value varies. May have an effect. When the firing temperature is less than 500 ° C., pinholes and bubbles may occur in the completed overcoat (corresponding to the protective layer 25 in FIG. 16). As described above, as the glass member for forming the protective layer 25, the protective layer 25 can be formed in a temperature range in which the resistance value of the strain sensitive resistor is hardly affected by setting the firing temperature to 500 to 750 ° C. The generation of bubbles and pinholes in the protective layer 25 can be suppressed (the higher the firing temperature, the less likely the bubbles and pinholes are generated in the protective layer 25. However, if the firing temperature exceeds 750, the resistance of the strain sensitive resistor 23 Value may be affected).

(Embodiment 6)
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  In Embodiment 6, members used for the insulating layer of the present invention will be described in detail. First, as the insulating layer, glass or a composite material made of glass and ceramic can be used. The thickness is preferably 5 to 500 μm. When the thickness is less than 5 μm, sufficient insulating properties may not be obtained between the electrode pattern and the metal elastic body. On the other hand, when the thickness exceeds 500 μm, the material cost of the insulating layer may affect the product cost. As the glass material, amorphous glass or crystalline glass can be combined with a single layer or a plurality of layers.

When using a crystallized glass, the composition of which is MgO 30~55wt%, B ₂ O ₃ is 5-30 wt%, SiO ₂ is 10 to 25 wt%, BaO is 5-30 wt%, the Al ₂ O ₃ 1-30 wt %, CaO is 6 wt% or less, and ZrO ₂ is preferably 5 wt% or less.

When amorphous glass is used, the composition is 40 to 80 wt% for SiO _{2, 5} to 15 wt% for CaO, 3 to 15 wt% for PbO, 1 to 10 wt% for Al ₂ O ₃ , and ₂ to 10 wt% for ZrO _2. Is desirable. If SiO ₂ is too small, less than 40 wt%, or exceeds 80 wt%, sufficient sinterability may not be obtained.

  As the insulating layer, a composite material obtained by adding a ceramic member to these glass members and firing them simultaneously can also be used.

As the ceramic member used here, ceramic powder such as SiO ₂ , ZrO ₂ , Al ₂ O ₃ , MgO, calcium oxide and titanium oxide can be used alone or in combination. In this case, the average particle size of these ceramic powders is preferably less than 0.01 to 10 μm. When the average particle size is less than 0.01 μm, the ceramic powder becomes expensive and may affect the product cost. If the average particle size exceeds 10 μm, it is difficult to disperse the ceramic powder, and as a result, the sinterability of the insulating layer and the sintering uniformity may be affected. Thus, by adding ceramic powder as a kind of filler in the insulating layer and co-firing with the glass member, the thermal expansion coefficient of the completed insulating layer can be finely adjusted and the cost can be reduced.

Moreover, when these glass is made into multiple layers, the thermal expansion coefficient can be optimized, without affecting mutual sinterability by making a part of each composition differ. For example, when a plurality of glass layers are used as insulating layers, the glass components contained therein (for example, glass components such as SiO ₂ and MgO) can be optimized for each layer.

  When amorphous glass and crystalline glass are used as a multilayer, it is desirable to use crystalline glass on the base (that is, the side close to the metal elastic body). After crystallized glass has been crystallized by firing on a metal elastic body, the crystallized glass is very stable without being re-dissolved by heat treatment in a subsequent process. Moreover, the effect which suppresses the spreading | diffusion can also be anticipated with respect to the member which spread | diffuses easily, such as silver. As a result, as shown in FIG. 14, even when the internal electrode 26 is built in the insulating layer 22, a member mainly composed of silver can be selected as the member of the internal electrode 26, and the influence of the diffusion is affected. It can be suppressed.

  When an electrode mainly composed of silver is used for the internal electrode, the thickness is preferably 1 to 50 μm. If it is less than 1 μm, necessary characteristics may not be obtained. If the thickness exceeds 50 μm, the product cost may be affected. Further, if necessary, the sintering shrinkage can be adjusted by adding an appropriate amount of Pd or Pt and further a glass member to the internal electrode, so that the internal electrode 26 and the insulating layer 22 adjacent thereto are fired simultaneously. (Even if fired at the same time, delamination and the like can be prevented).

  Moreover, as a strain sensitive resistor, GF (for example, GF = 10-20) higher than a commercially available metal foil gauge (GF = 2) can be obtained by using a ruthenium compound. In addition to ruthenium oxide, various ruthenium compounds can be used as such a ruthenium compound, and a commercially available resistor paste can also be used as such a material.

  In this case, the thickness of the strain sensitive resistor is preferably 1 to 100 μm. When the thickness of the strain sensitive resistor is less than 1 μm, the resistance value tends to vary, and it is necessary to use a fine resistor member, and the GF value may decrease. Further, when the thickness is 100 μm or more, the product cost may be affected. Moreover, by adding lead oxide in the range of 1 to 70 wt% in the strain sensitive resistor, the sinterability of the strain sensitive resistor can be enhanced, and the adhesion strength with the base can also be enhanced. In addition, when the addition amount of lead oxide is less than 1 wt%, the addition effect may not be obtained. On the other hand, if it exceeds 70 wt%, the resistance value may become too high, or the dependency of the resistance value on the firing temperature may become high, and the resistance value may easily fluctuate.

(Embodiment 7)
DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings.

In the seventh embodiment, a strain sensitive resistor used in the strain sensor proposed in the present invention will be described. As the strain sensitive resistor, RuO ₂ or Ru compound is 10 to 40 wt%, PbO is 40 to 60 wt%, SiO ₂ is 5 to ₂₀ wt%, B ₂ O ₃ is ₂ to 10 wt%, Al ₂ O ₃ is 1 to 10 wt%. It is desirable to use the ones that are included.

If the RuO ₂ or Ru compound is less than 10 wt%, the resistance value may be too high to be used as a strain sensor. When the ratio of RuO ₂ or Ru compound exceeds 40 wt%, Ru may be a noble metal and may affect the product price. Moreover, when PbO is less than 40 wt%, GF may become low. Moreover, when PbO is 60 wt% or more, it may be influenced by the glass material used for overcoat. Further, when the proportion of SiO ₂ is less than 5 wt% or exceeds 20 wt%, the characteristics of the resistor may be easily affected by the firing temperature. Further, when the proportion of B ₂ O ₃ is less than 2 wt% or exceeds 10 wt%, the characteristics of the resistor may become unstable even when the proportion of Al ₂ O ₃ is less than 1 wt% or exceeds 10 wt%.

  Of the ruthenium compounds constituting the strain sensor, 50% or more is preferably a ruthenium compound having a size of 0.1 to 3 μm. When the ruthenium compound has a small diameter of less than 0.1 μm and occupies 50% or more of the ruthenium compound, the GF value may be small. To measure the size of the ruthenium compound, the paste is dried in the paste state, and in the case of a completed strain sensor, the cross section of the product is used by SEM (scanning electron microscope) or XMA (X-ray microanalysis). It can be confirmed by analyzing.

  When analyzing with XMA, Ru (ruthenium) and Pb (lead) fluorescent X-rays are close in wavelength, and depending on the equipment, Ru and Pb elements cannot be separated cleanly, and the shape and size of the ruthenium compound cannot be distinguished. is there. In this case, the size of the Ru compound can be measured more accurately by using a wavelength dispersion type XMA apparatus instead of the energy dispersion type. The size distribution of the Ru compound is determined by observing an arbitrary cross section of the resistor with XMA, sampling about 10 to 100 Ru compounds observed here, and measuring each dimension from the mapping data of XMA. (In this case, it is not the true size of each Ru compound, but the size of the Ru compound in cross section, but this measurement method is convenient and it is desirable to use this method). In addition, the size distribution of the Ru compound may be obtained.

According to the experiments by the inventors, even in the paste state (the Ru compound is not fired), the Ru compound is confined in the glass and its cross section (the Ru compound is also measured in its cross section). Even when measured by, there was no particular difference. The Ru compound need not be limited to the Ru oxide. If necessary, K ₂ O, CaO, or CuO is added in the range of 0.1 to 3 wt%, so that the strain sensitive characteristics of the resistor can be improved or stabilized. In this case, when the addition amount of these trace elements is less than 0.1 wt%, it is difficult to control the addition amount and the addition effect may not be obtained. If the amount added exceeds 3 wt%, the characteristics may become unstable. Further, by adding B ₂ O ₃ in the range of 1 to 10 wt%, the dependency of the strain sensitive resistor on the firing temperature can be suppressed. In this case, when the addition amount of B ₂ O ₃ is less than 1 wt%, the addition effect may not be obtained. If it exceeds 10 wt%, other characteristics of the resistor may be affected.

  The strain sensor and the manufacturing method thereof according to the present invention can stably form the strain sensor characteristics and increase the productivity because the strain sensitive resistor that determines the characteristics of the strain sensor can be formed with high accuracy and stability. Therefore, the product quality can be stabilized and the cost can be reduced, and a highly accurate strain sensor can be supplied to the market at a low cost.

Sectional drawing which shows the structure of the sensor in Embodiment 1 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 1 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 1 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 1 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 1 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 1 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 2 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 2 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 2 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 2 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 3 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 3 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 3 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 4 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 5 of this invention. Sectional drawing which shows the manufacturing process of the sensor in Embodiment 5 of this invention. Sectional view showing the configuration of a conventional sensor The figure which shows a mode that several strain sensitive resistor elements are connected in the predetermined shape in the conventional sensor, and form a bridge circuit. Diagram showing an example of conventional sensor countermeasures

Explanation of symbols

DESCRIPTION OF SYMBOLS 12 Embedded electrode 13 Unbaked insulating layer 14 Cured wiring 15 Antifouling film 16 Press apparatus 17 Heater 18 Arrow 19 Window 20 Base film 21 Substrate 22 Insulating layer 23 Strain sensitive resistor 24 Convex electrode 25 Protective layer 26 Internal electrode 27 First insulating layer 28 Second insulating layer

Claims (28)

A substrate and an insulating layer in which a plurality of electrodes are embedded on the substrate; a strain sensitive resistor provided on the electrode; and a protective layer covering at least a part of the strain sensitive resistor and the electrode. Characteristic strain sensor. The strain sensor according to claim 1, wherein the substrate is made of a metal elastic body. 2. The strain sensor according to claim 1, wherein an internal electrode is formed inside the insulating layer, and the internal electrode is made mainly of silver and glass is added, and has a thickness of 1 to 50 μm. The strain sensor according to claim 1, wherein the insulating layer is formed by laminating crystallized glass and amorphous glass with a thickness of 3 to 300 μm. The strain sensor according to claim 1, wherein the insulating layer is made of glass or a composite material made of glass and ceramic, and has a thickness of 5 to 500 μm. 2. The strain sensor according to claim 1, wherein the electrode is made of silver or silver and palladium as a main component and glass is added, and the thickness thereof is 1 to 100 [mu] m. The strain sensor according to claim 1, wherein the protective layer is a composite material made of glass or glass and ceramic, and has a thickness of 5 to 500 μm. Insulating layer, at least MgO is 30~55wt%, B ₂ O ₃ is 5-30 wt%, SiO ₂ is 10 to 25 wt%, BaO is 5~25wt%, AL ₂ O ₃ is 1-30 wt%, CaO is 6wt The strain sensor according to claim 1, 5 or 7, which is a crystallized glass having a crystallization rate of 40% to 80%. Insulating layer is at least SiO ₂ is 40 to 80 wt%, CaO is 5 to 15 wt%, PbO is 3~15wt%, AL ₂ O ₃ is 110 wt.%, ZrO ₂ is an amorphous glass of 2 to 10 wt% The strain sensor according to claim 1, 5, or 7. The strain sensor according to claim 1, wherein the strain sensitive resistor is a ruthenium compound added with glass and has a thickness of 1 to 100 µm. The strain sensitive resistor contains at least RuO ₂ 10-40 wt%, PbO 40-60 wt%, SiO ₂ 5-20 wt%, B ₂ O ₃ 2-10 wt%, Al ₂ O ₃ 1-10 wt%. The strain sensor according to claim 1. 2. The strain sensor according to claim 1, wherein 50% or more of the ruthenium compound constituting the strain sensitive resistor is a ruthenium compound having a size of 0.1 to 3 [mu] m. A non-sintered insulating layer in which a plurality of electrodes are embedded in advance is laminated on a substrate, and then the non-sintered insulating layer is fired simultaneously with the electrodes to form an electrode embedded layer, and then a strain sensitive resistor is formed on the electrodes. Forming a body, and then covering at least a part of the strain-sensitive resistor and the electrode-embedded insulating layer with a protective layer. Forming an unsintered insulating layer on the substrate, then forming an electrode pattern on the unsintered insulating layer, and then embedding the electrode pattern in the unsintered insulating layer using a pressing device; The insulating layer and the electrode are fired at the same time, and then an electrode buried insulating layer is formed, then a strain sensitive resistor is formed on the electrode, and at least a part of the strain sensitive resistor and the electrode buried insulating layer A method for manufacturing a strain sensor, characterized by covering the substrate with a protective layer. The press device is provided with a protrusion on a surface to be pressed, and the method for manufacturing a strain sensor. A predetermined electrode pattern is formed on the film, this film is set on a substrate on which a non-fired insulating layer is formed, and then the non-fired insulation of the electrode pattern is formed from the film side using a press device. Press to embed on the layer, then peel off only the film, then fire the unfired insulating layer and the electrode pattern at the same time, then form the electrode embedded insulating layer, and then strain sensitive on the electrode A method for manufacturing a strain sensor, comprising: forming a resistor, and further covering at least part of the strain-sensitive resistor and the electrode-embedded insulating layer with a protective layer. The method for manufacturing a strain sensor according to claim 13, wherein the substrate is made of a metal elastic body. The unsintered insulating layer is obtained by drying a glass paste, in which glass powder is dispersed in a resin solution, in a predetermined shape on a metal elastic body and then drying in a range of 50 to 400 ° C. A method for manufacturing the strain sensor according to claim 16. The method for manufacturing a strain sensor according to any one of claims 13 to 16, wherein the unfired insulating layer has a thickness of 10 to 1000 µm. The method for manufacturing a strain sensor according to claim 13, wherein the unsintered insulating layer comprises glass powder, a resin, and a plasticizer. The method for producing a strain sensor according to claim 13, wherein the electrode pattern is obtained by printing an electrode paste in which conductive powder is dispersed in a curable resin solution in a predetermined shape and then curing the electrode paste. The method for manufacturing a strain sensor according to claim 13, wherein the electrode pattern has a thickness of 5 to 200 μm. The method for manufacturing a strain sensor according to claim 13, wherein the electrode pattern is made of a conductive powder and a curable resin. 17. The method for producing a strain sensor according to claim 13, wherein the electrode pattern is obtained by curing the predetermined electrode paste at a temperature of 60 to 400 [deg.] C. for 0.1 seconds to 30 minutes or eliminating the need for a solvent. The method for producing a strain sensor according to claim 13, wherein the unfired insulating layer and the electrode pattern are fired at 600 to 1000 ° C. at the same time. The method for producing a strain sensor according to claim 14, wherein the pressing pressure is 1-1000 Kg / cm ² . The method for producing a strain sensor according to claim 14, wherein the pressing time is 0.1 second to 10 minutes. The method for producing a strain sensor according to claims 14 to 16, wherein the pressing temperature is 20 to 300 ° C.

Images