JP6012833B2 - Semiconductor device, manufacturing method thereof, flow rate sensor and humidity sensor - Google Patents

Description

  The present invention relates to a flow sensor and a manufacturing technique thereof, and more particularly, to a resin-sealed flow sensor and a technique effective when applied to the manufacturing technique.

  Japanese Patent Laid-Open No. 2011-122984 (Patent Document 1) discloses a mold having an elastic film as a manufacturing method of a flow rate sensor in which a flow rate detection unit for detecting a flow of gas (air) is partially exposed from a sealing body. Describes a technique for pouring a resin while clamping a substrate on which a semiconductor chip is mounted.

  Japanese Patent Application Laid-Open No. 2004-74713 (Patent Document 2) describes a technique for clamping a component with a mold provided with a release film sheet and pouring a resin as a method for manufacturing a semiconductor package.

JP 2011-122984 A JP 2004-74713 A

  For example, currently, an internal combustion engine such as an automobile is provided with an electronically controlled fuel injection device. This electronically controlled fuel injection device has the role of operating the internal combustion engine efficiently by appropriately adjusting the amount of gas (air) and fuel flowing into the internal combustion engine. For this reason, in the electronically controlled fuel injection device, it is necessary to accurately grasp the gas (air) flowing into the internal combustion engine. For this reason, the electronic control fuel injection device is provided with a flow rate sensor (air flow sensor) for measuring the flow rate of gas (air).

  Among flow sensors, a flow sensor manufactured by a semiconductor micromachining technique is particularly attracting attention because it can reduce cost and can be driven with low power. Such a flow sensor has, for example, a diaphragm (thin plate portion) formed by anisotropic etching on the back surface of a semiconductor substrate made of silicon, and a heating resistor and a temperature measuring device on the surface of the semiconductor substrate opposite to the diaphragm. The flow rate detection part which consists of a resistor is formed.

  The actual flow sensor includes, for example, a second semiconductor chip formed with a control circuit unit for controlling the flow rate detection unit in addition to the first semiconductor chip formed with a diaphragm and a flow rate detection unit. The first semiconductor chip and the second semiconductor chip described above are mounted on a substrate, for example, and are electrically connected to wiring (terminals) formed on the substrate. Specifically, for example, the first semiconductor chip is connected to a wiring formed on the substrate by a wire made of a gold wire, and the second semiconductor chip uses a bump electrode formed on the second semiconductor chip. , Connected to the wiring formed on the substrate.

  In this way, the first semiconductor chip and the second semiconductor chip mounted on the substrate are electrically connected via the wiring formed on the substrate. As a result, the flow rate detection unit formed in the first semiconductor chip can be controlled by the control circuit unit formed in the second semiconductor chip, and a flow rate sensor is configured.

  At this time, the gold wire (wire) connecting the first semiconductor chip and the substrate is usually fixed by potting resin in order to prevent contact due to deformation. That is, the gold wire (wire) is covered and fixed by the potting resin, and the gold wire (wire) is protected by the potting resin. On the other hand, the first semiconductor chip and the second semiconductor chip constituting the flow sensor are usually not sealed with potting resin. In other words, a normal flow sensor has a structure in which only a gold wire (wire) is covered with a potting resin.

  Here, the fixing of the gold wire (wire) with the potting resin is not performed in a state in which the first semiconductor chip is fixed with a mold or the like. Therefore, the contraction of the potting resin causes the first semiconductor chip to deviate from the mounting position. There's a problem. Furthermore, since the potting resin is formed by dropping, there is a problem that the dimensional accuracy of the potting resin is low. As a result, the mounting position of the first semiconductor chip on which the flow rate detection unit is formed varies for each individual flow sensor, and the formation position of the potting resin is slightly different. Variations will occur.

  For this reason, in order to suppress the performance variation of each flow sensor, it is necessary to correct detection performance for every flow sensor, and the necessity of adding the performance correction process in the manufacturing process of a flow sensor arises. In particular, when the performance correction process is lengthened, there is a problem that the throughput in the manufacturing process of the flow sensor is reduced and the cost of the flow sensor is increased.

  Furthermore, since the potting resin does not promote curing by heating, the time until the potting resin is cured becomes long, and the throughput in the manufacturing process of the flow sensor is reduced.

  An object of the present invention is to provide a technique capable of improving performance by suppressing performance variation for each flow sensor (including a case where performance improvement is achieved by improving reliability).

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  For example, a flow sensor in a typical embodiment includes a first chip mounting portion and a first semiconductor chip bonded to the first chip mounting portion via an adhesive, and the above-described adhesive is Of the first chip mounting portion, the first chip mounting portion is also formed on a region overlapping with the diaphragm formed on the semiconductor chip in plan view.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  It is possible to improve performance by suppressing performance variation for each flow sensor.

FIG. 3 is a circuit block diagram showing a circuit configuration of the flow sensor in the first embodiment. FIG. 3 is a plan view showing a layout configuration of a semiconductor chip that constitutes a part of the flow sensor in the first embodiment. It is a figure which shows the mounting structure of the flow sensor in related technology, and is a figure which shows the structure before sealing with resin. (A) is a top view which shows the mounting structure of the flow sensor in related technology, (b) is sectional drawing cut | disconnected by the AA line of (a), (c) is the back surface of a semiconductor chip. FIG. It is a figure which shows the mounting structure of the flow sensor in related technology, and is a figure which shows the structure after sealing with resin. (A) is a top view which shows the mounting structure of the flow sensor in related technology, (b) is sectional drawing cut | disconnected by the AA line of (a), (c) is B of (a). It is sectional drawing cut | disconnected by the -B line. It is sectional drawing which shows the process of resin-sealing the flow sensor in related technology. It is a figure which shows the mounting structure of the flow sensor in Embodiment 1, and is a figure which shows the structure before sealing with resin. (A) is a top view which shows the mounting structure of the flow sensor in Embodiment 1, (b) is sectional drawing cut | disconnected by the AA line of (a), (c) is a semiconductor chip. It is a top view which shows the back surface. It is a figure which shows the mounting structure of the flow sensor in Embodiment 1, and is a figure which shows the structure after sealing with resin. (A) is a top view which shows the mounting structure of the flow sensor in Embodiment 1, (b) is sectional drawing cut | disconnected by the AA of (a), (c) is (a) It is sectional drawing cut | disconnected by the BB line of FIG. FIG. 6 is a cross-sectional view showing a manufacturing process of the flow sensor in the first embodiment. It is sectional drawing which shows the manufacturing process of the flow sensor following FIG. FIG. 10 is a cross-sectional view showing a flow sensor manufacturing process following FIG. 9. FIG. 11 is a cross-sectional view showing a manufacturing process of the flow sensor following FIG. 10. FIG. 6 is a cross-sectional view showing a manufacturing process of the flow sensor in the first embodiment. FIG. 13 is a cross-sectional view showing a manufacturing process for the flow sensor following FIG. 12. It is sectional drawing which shows the manufacturing process of the flow sensor following FIG. FIG. 15 is a cross-sectional view showing a manufacturing process of the flow sensor following FIG. 14. FIG. 16 is a cross-sectional view showing the manufacturing process of the flow sensor following FIG. 15. It is sectional drawing which shows the resin sealing process of the flow sensor in the modification 1. It is sectional drawing which shows the resin sealing process of the flow sensor in the modification 1. In the modification 2, it is a figure which shows the structure of the flow sensor after resin sealing. (A) is a top view which shows the structure of the flow sensor after resin sealing, (b) is sectional drawing cut | disconnected by the AA line of (a), (c) is (a). It is sectional drawing cut | disconnected by the BB line | wire. It is a figure which shows the mounting structure of the flow sensor in Embodiment 2, and is a figure which shows the structure after sealing with resin. (A) is a top view which shows the mounting structure of the flow sensor in Embodiment 2, (b) is sectional drawing cut | disconnected by the AA of (a), (c) is (a) It is sectional drawing cut | disconnected by the BB line of FIG.

  In the following embodiments, when it is necessary for the sake of convenience, the description will be divided into a plurality of sections or embodiments. However, unless otherwise specified, they are not irrelevant to each other. There are some or all of the modifications, details, supplementary explanations, and the like.

  Further, in the following embodiments, when referring to the number of elements (including the number, numerical value, quantity, range, etc.), especially when clearly indicated and when clearly limited to a specific number in principle, etc. Except, it is not limited to the specific number, and may be more or less than the specific number.

  Further, in the following embodiments, the constituent elements (including element steps and the like) are not necessarily indispensable unless otherwise specified and apparently essential in principle. Needless to say.

  Similarly, in the following embodiments, when referring to the shape, positional relationship, etc., of components, etc., unless otherwise specified, and in principle, it is considered that this is not clearly the case, it is substantially the same. Including those that are approximate or similar to the shape. The same applies to the above numerical values and ranges.

  In all the drawings for explaining the embodiments, the same members are denoted by the same reference symbols in principle, and the repeated explanation thereof is omitted. In order to make the drawings easy to understand, even a plan view may be hatched.

(Embodiment 1)
<Circuit configuration of flow sensor>
First, the circuit configuration of the flow sensor will be described. FIG. 1 is a circuit block diagram showing a circuit configuration of the flow sensor according to the first embodiment. In FIG. 1, the flow sensor in the first embodiment has a CPU (Central Processing Unit) 1 for controlling the flow sensor, and an input circuit 2 for inputting an input signal to the CPU 1. And it has the output circuit 3 for outputting the output signal from CPU1. The flow rate sensor is provided with a memory 4 for storing data, and the CPU 1 can access the memory 4 and refer to the data stored in the memory 4.

  Next, the CPU 1 is connected to the base electrode of the transistor Tr via the output circuit 3. The collector electrode of the transistor Tr is connected to the power source PS, and the emitter electrode of the transistor Tr is connected to the ground (GND) via the heating resistor HR. Therefore, the transistor Tr is controlled by the CPU 1. That is, since the base electrode of the transistor Tr is connected to the CPU 1 via the output circuit 3, an output signal from the CPU 1 is input to the base electrode of the transistor Tr.

  As a result, the current flowing through the transistor Tr is controlled by an output signal (control signal) from the CPU 1. When the current flowing through the transistor Tr is increased by the output signal from the CPU 1, the current supplied from the power source PS to the heating resistor HR is increased, and the heating amount of the heating resistor HR is increased.

  On the other hand, when the current flowing through the transistor Tr decreases due to the output signal from the CPU 1, the current supplied to the heating resistor HR decreases, and the heating amount of the heating resistor HR decreases.

  As described above, the flow rate sensor according to the first embodiment is configured such that the amount of current flowing through the heating resistor HR is controlled by the CPU 1 and the amount of heat generated from the heating resistor HR is thereby controlled by the CPU 1. I understand that.

  Subsequently, in the flow rate sensor according to the first embodiment, a heater control bridge HCB is provided in order for the CPU 1 to control the current flowing through the heating resistor HR. The heater control bridge HCB is configured to detect the amount of heat released from the heating resistor HR and output the detection result to the input circuit 2. As a result, the CPU 1 can input the detection result from the heater control bridge HCB, and controls the current flowing through the transistor Tr based on this.

  Specifically, as shown in FIG. 1, the heater control bridge HCB includes resistors R1 to R4 that form a bridge between the reference voltage Vref1 and the ground (GND). In the heater control bridge HCB configured as described above, when the gas heated by the heating resistor HR is higher than the intake air temperature by a certain temperature (ΔT, for example, 100 ° C.), the potential of the node A and the node B The resistance values of the resistors R1 to R4 are set so that the potential difference between the potentials of the resistors R1 to R4 is 0V. That is, the resistors R1 to R4 constituting the heater control bridge HCB are referred to as a component in which the resistor R1 and the resistor R3 are connected in series and a component in which the resistor R2 and the resistor R4 are connected in series. The bridge is configured so as to be connected in parallel between the voltage Vref1 and the ground (GND). A connection point between the resistor R1 and the resistor R3 is a node A, and a connection point between the resistor R2 and the resistor R4 is a node B.

  At this time, the gas heated by the heating resistor HR comes into contact with the resistor R1 constituting the heater control bridge HCB. Therefore, the resistance value of the resistor R1 constituting the heater control bridge HCB mainly changes depending on the amount of heat generated from the heating resistor HR. When the resistance value of the resistor R1 changes in this way, the potential difference between the node A and the node B changes. Since the potential difference between the node A and the node B is input to the CPU 1 via the input circuit 2, the CPU 1 controls the current flowing through the transistor Tr based on the potential difference between the node A and the node B.

  Specifically, the CPU 1 controls the amount of heat generated from the heating resistor HR by controlling the current flowing through the transistor Tr so that the potential difference between the node A and the node B becomes 0V. That is, in the flow rate sensor according to the first embodiment, the CPU 1 causes the gas heated by the heating resistor HR to be only a certain temperature (ΔT, for example, 100 ° C.) higher than the intake air temperature based on the output of the heater control bridge HCB. It can be seen that the feedback control is performed so as to maintain a high constant value.

  Subsequently, the flow sensor in the first embodiment has a temperature sensor bridge TSB for detecting the gas flow rate. The temperature sensor bridge TSB is composed of four temperature measuring resistors that form a bridge between the reference voltage Vref2 and the ground (GND). The four resistance temperature detectors are composed of two upstream resistance temperature detectors UR1 and UR2, and two downstream resistance temperature detectors BR1 and BR2.

  That is, the direction of the arrow in FIG. 1 indicates the direction in which the gas flows. The upstream resistance thermometers UR1 and UR2 are provided on the upstream side of the gas flow direction, and the downstream resistance thermometers BR1 and BR2 is provided. The upstream resistance thermometers UR1 and UR2 and the downstream resistance thermometers BR1 and BR2 are arranged so that the distance to the heating resistor HR is the same.

  In the temperature sensor bridge TSB, an upstream resistance temperature detector UR1 and a downstream resistance temperature detector BR1 are connected in series between the reference voltage Vref2 and the ground (GND), and the upstream resistance temperature detector UR1 and the downstream resistance temperature detector. The connection point of BR1 is node C.

  On the other hand, an upstream resistance temperature detector UR2 and a downstream resistance temperature detector BR2 are connected in series between the ground (GND) and the reference voltage Vref2, and a connection point between the upstream resistance temperature detector UR2 and the downstream resistance temperature detector BR2. Is node D. Then, the potential of the node C and the potential of the node D are configured to be input to the CPU 1 via the input circuit 2. The upstream resistance thermometers UR1 and UR2 and the downstream temperature sensor are set so that the potential difference between the potential of the node C and the potential of the node D becomes 0V when the flow rate of the gas flowing in the arrow direction is zero. Each resistance value of the resistors BR1 and BR2 is set.

  Specifically, the upstream resistance thermometers UR1 and UR2 and the downstream resistance thermometers BR1 and BR2 are configured to have the same distance from the heating resistor HR and the same resistance value. For this reason, it can be seen that the temperature sensor bridge TSB is configured such that the potential difference between the node C and the node D is 0 V in the absence of wind regardless of the amount of heat generated by the heating resistor HR.

<Flow sensor operation>
The flow sensor according to the first embodiment is configured as described above, and the operation thereof will be described below with reference to FIG. First, the CPU 1 outputs an output signal (control signal) to the base electrode of the transistor Tr via the output circuit 3, thereby causing a current to flow through the transistor Tr. Then, a current flows from the power supply PS connected to the collector electrode of the transistor Tr to the heating resistor HR connected to the emitter electrode of the transistor Tr. For this reason, the heating resistor HR generates heat. The gas heated by the heat generated from the heat generating resistor HR heats the resistor R1 constituting the heater control bridge HCB.

  At this time, when the gas heated by the heating resistor HR is increased by a certain temperature (for example, 100 ° C.), the resistor is set so that the potential difference between the node A and the node B of the heater control bridge HCB becomes 0V. Each resistance value of R1 to R4 is set. For this reason, for example, when the gas heated by the heating resistor HR is increased by a certain temperature (for example, 100 ° C.), the potential difference between the node A and the node B of the heater control bridge HCB becomes 0V, This difference potential (0 V) is input to the CPU 1 via the input circuit 2. Then, the CPU 1 recognizing that the difference potential from the heater control bridge HCB is 0 V outputs an output signal (control signal) for maintaining the current amount of current to the base electrode of the transistor Tr via the output circuit 3. Output.

  On the other hand, when the gas heated by the heating resistor HR deviates from a certain temperature (for example, 100 ° C.), a non-zero potential difference is generated between the node A and the node B of the heater control bridge HCB. The difference potential is input to the CPU 1 via the input circuit 2. Then, the CPU 1 recognizing that the difference potential from the heater control bridge HCB is generated, outputs an output signal (control signal) to the base electrode of the transistor Tr via the output circuit 3 so that the difference potential becomes 0V. Output.

  For example, when a potential difference in a direction in which the gas heated by the heating resistor HR becomes higher than a certain temperature (for example, 100 ° C.) is generated, the CPU 1 controls the control signal so that the current flowing through the transistor Tr decreases. (Output signal) is output to the base electrode of the transistor Tr. On the other hand, when a potential difference in a direction in which the gas heated by the heating resistor HR becomes lower than a certain temperature (for example, 100 ° C.) is generated, the CPU 1 increases the current flowing through the transistor Tr. A control signal (output signal) is output to the base electrode of the transistor Tr.

  As described above, the CPU 1 performs feedback control based on the output signal from the heater control bridge HCB so that the potential difference between the node A and the node B of the heater control bridge HCB is 0 V (equilibrium state). To do. From this, it can be seen that in the flow rate sensor according to the first embodiment, the gas heated by the heating resistor HR is controlled to have a constant temperature.

  Next, an operation for measuring the gas flow rate in the flow sensor according to the first embodiment will be described. First, the case of a windless state will be described. When the flow rate of the gas flowing in the direction of the arrow is zero, the upstream resistance temperature detectors UR1 and UR2 are set so that the potential difference between the node C potential and the node D potential of the temperature sensor bridge TSB becomes 0V. Each resistance value of the downstream resistance thermometers BR1 and BR2 is set.

  Specifically, the upstream resistance thermometers UR1 and UR2 and the downstream resistance thermometers BR1 and BR2 are configured to have the same distance from the heating resistor HR and the same resistance value. Therefore, in the temperature sensor bridge TSB, regardless of the amount of heat generated by the heating resistor HR, if there is no wind, the difference potential between the node C and the node D becomes 0V, and this difference potential (0V) is passed through the input circuit 2. Are input to the CPU 1. Then, the CPU 1 recognizing that the potential difference from the temperature sensor bridge TSB is 0 V recognizes that the flow rate of the gas flowing in the direction of the arrow is zero, and the gas flow rate Q is zero via the output circuit 3. Is output from the flow sensor in the first embodiment.

  Next, consider the case where gas is flowing in the direction of the arrow in FIG. In this case, as shown in FIG. 1, the upstream resistance temperature detectors UR1 and UR2 arranged on the upstream side in the gas flow direction are cooled by the gas flowing in the arrow direction. For this reason, the temperature of the upstream resistance thermometers UR1 and UR2 decreases. On the other hand, the downstream resistance thermometers BR1 and BR2 arranged on the downstream side in the gas flow direction have a temperature because the gas heated by the heating resistor HR flows to the downstream resistance thermometers BR1 and BR2. Rises. As a result, the balance of the temperature sensor bridge TSB is lost, and a non-zero differential potential is generated between the node C and the node D of the temperature sensor bridge TSB.

  This difference potential is input to the CPU 1 via the input circuit 2. Then, the CPU 1 recognizing that the potential difference from the temperature sensor bridge TSB is not zero recognizes that the flow rate of the gas flowing in the arrow direction is not zero. Thereafter, the CPU 1 accesses the memory 4. Since the memory 4 stores a comparison table (table) in which the difference potential and the gas flow rate are associated with each other, the CPU 1 accessing the memory 4 calculates the gas flow rate Q from the comparison table stored in the memory 4. . In this way, the gas flow rate Q calculated by the CPU 1 is output from the flow rate sensor in the first embodiment via the output circuit 3. As described above, according to the flow rate sensor of the first embodiment, it can be seen that the flow rate of gas can be obtained.

<Flow sensor layout configuration>
Next, the layout configuration of the flow sensor according to the first embodiment will be described. For example, the flow sensor in the first embodiment shown in FIG. 1 is formed on two semiconductor chips. Specifically, the heating resistor HR, the heater control bridge HCB, and the temperature sensor bridge TSB are formed on one semiconductor chip, and the CPU 1, the input circuit 2, the output circuit 3, the memory 4, and the like are formed on another semiconductor chip. . Hereinafter, a layout configuration of a semiconductor chip on which the heating resistor HR, the heater control bridge HCB, and the temperature sensor bridge TSB are formed will be described.

  FIG. 2 is a plan view showing a layout configuration of the semiconductor chip CHP1 constituting a part of the flow sensor according to the first embodiment. First, as shown in FIG. 2, the semiconductor chip CHP1 has a rectangular shape, and gas flows from the left side to the right side (arrow direction) of the semiconductor chip CHP1. As shown in FIG. 2, a rectangular diaphragm DF is formed on the back surface side of the rectangular semiconductor chip CHP1. The diaphragm DF indicates a thin plate region where the thickness of the semiconductor chip CHP1 is reduced. That is, the thickness of the region where the diaphragm DF is formed is thinner than the thickness of the other semiconductor chip CHP1.

  As shown in FIG. 2, a flow rate detection unit FDU is formed in the surface region of the semiconductor chip CHP1 opposite to the back surface region where the diaphragm DF is formed. Specifically, a heating resistor HR is formed at the center of the flow rate detection unit FDU, and a resistor R1 that forms a heater control bridge is formed around the heating resistor HR. And the resistor R2-R4 which comprises a heater control bridge is formed in the outer side of the flow volume detection part FDU. A heater control bridge is configured by the resistors R1 to R4 formed in this way.

  In particular, since the resistor R1 constituting the heater control bridge is formed in the vicinity of the heating resistor HR, the temperature of the gas heated by the heat generated from the heating resistor HR is accurately reflected in the resistor R1. Can do.

  On the other hand, since the resistors R2 to R4 constituting the heater control bridge are arranged apart from the heating resistor HR, they can be made less susceptible to the heat generated by the heating resistor HR.

  Therefore, the resistor R1 can be configured to react sensitively to the temperature of the gas heated by the heating resistor HR, and the resistors R2 to R4 are not easily affected by the heating resistor HR and have a constant resistance value. The value can be easily maintained. For this reason, the detection accuracy of the heater control bridge can be increased.

  Further, upstream temperature measuring resistors UR1 and UR2 and downstream temperature measuring resistors BR1 and BR2 are arranged so as to sandwich the heating resistor HR formed in the flow rate detection unit FDU. Specifically, upstream resistance thermometers UR1 and UR2 are formed on the upstream side in the arrow direction in which gas flows, and downstream resistance thermometers BR1 and BR2 are formed in the downstream in the arrow direction in which gas flows.

  With this configuration, when the gas flows in the direction of the arrow, the temperature of the upstream resistance thermometers UR1 and UR2 can be lowered and the temperature of the downstream resistance thermometers BR1 and BR2 can be increased. . Thus, the temperature sensor bridge is formed by the upstream resistance thermometers UR1 and UR2 and the downstream resistance thermometers BR1 and BR2 arranged in the flow rate detection unit FDU.

  The heating resistor HR, the upstream resistance thermometers UR1 and UR2, and the downstream resistance thermometers BR1 and BR2 are formed by sputtering a metal film such as platinum or a semiconductor thin film such as polysilicon (polycrystalline silicon), for example. It can be formed by patterning by a method such as ion etching after forming by a method such as the CVD method or the CVD (Chemical Vapor Deposition) method.

  The heating resistor HR configured as described above, the resistors R1 to R4 constituting the heater control bridge, and the upstream temperature sensing resistors UR1 and UR2 and the downstream temperature sensing resistors BR1 and BR2 constituting the temperature sensor bridge are: These are connected to the wiring WL1 and drawn out to the pads PD1 arranged along the lower side of the semiconductor chip CHP1.

  As described above, the semiconductor chip CHP1 constituting a part of the flow sensor according to the first embodiment is laid out. The actual flow rate sensor includes one semiconductor chip on which the heating resistor HR, the heater control bridge HCB and the temperature sensor bridge TSB are formed, and another one on which the CPU 1, the input circuit 2, the output circuit 3, the memory 4, and the like are formed. The semiconductor chip has a structure in which these semiconductor chips are mounted on a substrate.

  Below, the related technique regarding the mounting structure of a flow sensor is demonstrated first, Then, the necessity for the improvement which this related technique has is demonstrated. Next, a description will be given of the mounting configuration of the flow sensor according to the first embodiment, which is devised to overcome the room for improvement of the related art.

<Description of related technologies>
FIG. 3 is a diagram showing a mounting configuration of the flow sensor FSP in the related art, and is a diagram showing a configuration before sealing with resin. In particular, FIG. 3A is a plan view showing a mounting configuration of the flow sensor FSP in the related art. 3B is a cross-sectional view taken along line AA in FIG. 3A, and FIG. 3C is a plan view showing the back surface of the semiconductor chip CHP1.

  First, as shown to Fig.3 (a), the flow sensor FSP in related technology has the lead frame LF which consists of copper materials, for example. The lead frame LF includes a chip mounting portion TAB1 and a chip mounting portion TAB2 inside the dam bar DM constituting the outer frame body. The semiconductor chip CHP1 is mounted on the chip mounting portion TAB1, and the semiconductor chip CHP2 is mounted on the chip mounting portion TAB2.

  The semiconductor chip CHP1 has a rectangular shape, and a flow rate detection unit FDU is formed substantially at the center. A wiring WL1 connected to the flow rate detection unit FDU is formed on the semiconductor chip CHP1, and the wiring WL1 is connected to a plurality of pads PD1 formed along one side of the semiconductor chip CHP1. That is, the flow rate detection unit FDU and the plurality of pads PD1 are connected by the wiring WL1. These pads PD1 are connected to a lead LD1 formed on the lead frame LF via a wire W1 made of, for example, a gold wire. The lead LD1 formed on the lead frame LF is further connected to a pad PD2 formed on the semiconductor chip CHP2 via a wire W2 made of, for example, a gold wire.

  On the semiconductor chip CHP2, an integrated circuit made of semiconductor elements such as MISFET (Metal Insulator Semiconductor Field Effect Transistor) and wirings is formed. Specifically, an integrated circuit constituting the CPU 1, the input circuit 2, the output circuit 3 or the memory 4 shown in FIG. 1 is formed. These integrated circuits are connected to the pads PD2 and PD3 that function as external connection terminals. The pad PD3 formed on the semiconductor chip CHP2 is connected to the lead LD2 formed on the lead frame LF via a wire W3 made of, for example, a gold wire. In this manner, the semiconductor chip CHP1 in which the flow rate detection unit FDU is formed and the semiconductor chip CHP2 in which the control circuit is formed are connected via the leads LD1 formed in the lead frame LF. Recognize.

  Subsequently, as shown in FIG. 3B, a chip mounting portion TAB1 is formed on the lead frame LF, and the semiconductor chip CHP1 is mounted on the chip mounting portion TAB1. The semiconductor chip CHP1 is bonded to the chip mounting portion TAB1 with an adhesive ADH. A diaphragm DF (thin plate portion) is formed on the back surface of the semiconductor chip CHP1, and a flow rate detection unit FDU is formed on the surface of the semiconductor chip CHP1 facing the diaphragm DF. On the other hand, an opening OP1 is formed at the bottom of the chip mounting portion TAB1 existing below the diaphragm DF.

  Further, as shown in FIG. 3B, on the surface (upper surface) of the semiconductor chip CHP1, in addition to the flow rate detection unit FDU, a pad PD1 connected to the flow rate detection unit FDU is formed, and this pad PD1 Is connected to a lead LD1 formed on the lead frame LF via a wire W1. In addition to the semiconductor chip CHP1, the semiconductor chip CHP2 is also mounted on the lead frame LF, and the semiconductor chip CHP2 is bonded to the chip mounting portion TAB2 with an adhesive ADH. Further, the pad PD2 formed on the semiconductor chip CHP2 and the lead LD1 formed on the lead frame LF are connected via a wire W2. Further, the pad PD3 formed on the semiconductor chip CHP2 and the lead LD2 formed on the lead frame LF are electrically connected through a wire W3.

  FIG. 3C is a plan view showing the back surface of the semiconductor chip CHP1. As shown in FIG. 3C, a diaphragm DF is formed on the back surface of the semiconductor chip CHP1, and an adhesive ADH is applied so as to surround the diaphragm DF. That is, in the related art, it can be seen that the adhesive is formed so as not to overlap with the diaphragm DF in plan view.

  In the related art flow sensor FSP, the mounting configuration of the flow sensor FSP before sealing with resin is as described above, and the mounting configuration of the flow sensor FSP after sealing with resin will be described below.

  FIG. 4 is a diagram illustrating a mounting configuration of the flow sensor FSP in the related art, and is a diagram illustrating a configuration after sealing with resin. In particular, FIG. 4A is a plan view showing a mounting configuration of the flow sensor FSP in the related art. 4B is a cross-sectional view taken along the line AA in FIG. 4A, and FIG. 4C is a cross-sectional view taken along the line BB in FIG. 4A.

  As shown in FIGS. 4A to 4C, in the flow rate sensor FSP in the related art, the semiconductor chip CHP2 is sealed with a sealing body including the resin MR, and a part of the semiconductor chip CHP1 and the chip mounting are provided. Part of the part TAB1 is sealed with a sealing body including the resin MR. Specifically, in the flow sensor FSP in the related art, the resin MR is formed so as to cover a part of the side surface and the upper surface of the semiconductor chip CHP1 while exposing the flow rate detection unit FDU formed on the upper surface of the semiconductor chip CHP1. ing.

  The flow rate sensor FSP in the related art configured as described above is resin-sealed by, for example, a manufacturing process shown in FIG. FIG. 5 is a cross-sectional view showing a process of resin-sealing the flow rate sensor FSP in the related art.

  As shown in FIG. 5, the semiconductor chip CHP1 is fixed on the chip mounting portion TAB1 formed on the lead frame LF with an adhesive ADH. Then, the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM through the second space. Thereafter, the resin MR is poured into the second space under heating, whereby a part of the semiconductor chip CHP1 is sealed with the resin MR.

  At this time, as shown in FIG. 5, the inner space of the diaphragm DF is separated from the second space by the adhesive ADH. Therefore, when the second space is filled with the resin MR, the diaphragm DF It is possible to prevent the resin MR from entering the internal space.

  Further, the upper mold UM is formed with a recess so as to secure a first space SP1 (sealed space) surrounding the flow rate detection unit FDU formed on the upper surface SUR (CHP) of the semiconductor chip CHP1. From this, when the upper mold UM is pressed onto the semiconductor chip CHP1, the recesses formed in the upper mold UM surround the flow rate detection unit FDU formed in the semiconductor chip CHP1 and the vicinity thereof. For example, the side surface and part of the upper surface of the semiconductor chip CHP1 can be sealed while the one space SP1 (sealed space) is secured. That is, according to the related art, a part of the semiconductor chip CHP1 can be sealed while exposing the flow rate detection unit FDU formed in the semiconductor chip CHP1 and the vicinity thereof.

  Specifically, in FIG. 5, as a method of manufacturing the flow sensor, the lower mold BM and the upper mold UM on which the elastic film LAF is installed are mounted on the chip mounting portion TAB1 formed on the lead frame LF. A state in which components such as the manufactured semiconductor chip CHP1 are clamped is shown. FIG. 5 shows a process of injecting the resin MR into the second space formed between the upper mold UM and the lower mold BM. In particular, FIG. 5 shows a cross-sectional view of the flow sensor in the flow direction of air (gas). As shown in FIG. 5, the end portion of the semiconductor chip CHP1 is pressed by the upper mold UM via the elastic film LAF, whereby the semiconductor chip CHP1 is fixed by the upper mold UM.

  In the related art described above, the semiconductor chip CHP1 in which the flow rate detection unit FDU is formed can be sealed in a state where the semiconductor chip CHP1 is fixed with a mold. A part of the chip CHP1 can be sealed with the resin MR. This means that a part of the semiconductor chip CHP1 can be sealed with the resin MR while suppressing the positional deviation of each flow sensor according to the manufacturing method of the flow sensor FSP in the related art, and is formed on the semiconductor chip CHP1. This means that it is possible to suppress variations in the position of the flow rate detection unit FDU.

  As a result, according to the related art, the position of the flow rate detection unit FDU that detects the flow rate of gas can be matched by each flow rate sensor, so that the performance variation of detecting the gas flow rate in each flow rate sensor can be suppressed. That is, according to the related technology that seals a part of the semiconductor chip CHP1 while being fixed by the mold, it is possible to suppress the performance variation for each flow sensor FSP compared to the technology using the potting resin.

  In the manufacturing process of the flow sensor FSP in the related technology that employs the resin sealing process, it is necessary to smoothly release the sealing body including the resin MR from the lower mold BM. Therefore, as shown in FIG. 5, a protrusion pin (ejector pin) EJPN capable of moving up and down is inserted into the lower mold BM. By using this protrusion pin EJPN, sealing after resin sealing is performed. The stationary body can be easily released from the lower mold BM.

  Here, for example, as shown in FIG. 3, there is a possibility that the dimension of the mounting height H from the lower end portion of the lead frame LF before resin sealing to the upper end portion of the semiconductor chip CHP1 may vary. This is because there is a certain variation in the thickness dimension of each component (lead frame LF, adhesive material ADH, semiconductor chip CHP1). That is, the mounting height H may vary due to variations in the thickness of each component (lead frame LF, adhesive member ADH, semiconductor chip CHP1).

  If the dimensional variation of the mounting height H occurs, the semiconductor chip CHP1 may be broken or the resin leakage onto the semiconductor chip CHP1 may occur when components are clamped by the upper mold UM and the lower mold BM. Will increase.

  Therefore, in the related art, in order to prevent breakage of the semiconductor chip CHP1 or occurrence of resin leakage onto the semiconductor chip CHP1, for example, as shown in FIG. 5, between the semiconductor chip CHP1 and the upper mold UM. An elastic film LAF is interposed between the two.

  Thereby, for example, when the thickness of the semiconductor chip CHP1 is thinner than the average thickness, a gap is generated when the lead frame LF mounting the semiconductor chip CHP1 is sandwiched between the upper mold UM and the lower mold BM. Since the gap can be filled with the elastic film LAF, resin leakage onto the semiconductor chip CHP1 can be prevented.

  On the other hand, when the thickness of the semiconductor chip CHP1 is larger than the average thickness, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the elastic film LAF is used for the semiconductor chip CHP1. Therefore, the dimension in the thickness direction of the elastic film LAF changes so as to absorb the thickness of the semiconductor chip CHP1. As a result, even if the thickness of the semiconductor chip CHP1 is larger than the average thickness, it is possible to prevent the semiconductor chip CHP1 from being subjected to an excessive force, and as a result, to prevent the semiconductor chip CHP1 from being broken. Can do.

  That is, according to the manufacturing method of the flow sensor FSP in the related art, the semiconductor chip CHP1 is pressed by the upper mold UM through the elastic film LAF. For this reason, the mounting variation of the components due to the thickness variation of the semiconductor chip CHP1, the adhesive ADH, or the lead frame LF can be absorbed by the thickness change of the elastic film LAF.

  Thus, according to the related art, the clamping force applied to the semiconductor chip CHP1 can be relaxed. As a result, it is possible to prevent breakage such as cracks, chips or cracks of the semiconductor chip CHP1. That is, according to the manufacturing method of the flow sensor FSP in the related art, the semiconductor chip CHP1 is removed from damage represented by cracks, chips, cracks, etc. of the semiconductor chip CHP1 due to an increase in clamping force due to component mounting variation. Can be protected.

<Room for improvement in related technologies>
In the related art described above, further improvement can be achieved from the viewpoint of protecting the semiconductor chip CHP1 from breakage such as cracking, chipping, or cracking of the semiconductor chip CHP1 due to an increase in clamping force due to component mounting variation. There is room. Below, the room for the improvement which this related technology exists is demonstrated.

  For example, when the dimensional variation of the mounting height H shown in FIG. 3 is large, the dimensional variation of the mounting height H may not be absorbed only by the dimensional change of the elastic film LAF. That is, even if the chip mounting portion TAB1 to which the semiconductor chip CHP1 is bonded is sandwiched between the upper mold UM and the lower mold BM via the elastic film LAF, the mounting height H is so large that it cannot be absorbed by the thickness change of the elastic film LAF. Is high, a clamping force more than necessary is applied to the semiconductor chip CHP1, and the possibility that the semiconductor chip CHP1 is broken becomes obvious.

  On the other hand, when the mounting height H is so low that the mounting height H cannot be absorbed by the thickness change of the elastic film LAF, a gap is generated between the elastic film LAF and the semiconductor chip CHP1, and the resin flows into this gap. The risk of resin leakage on the semiconductor chip CHP1 becomes obvious.

  In this regard, it is conceivable to increase the thickness of the elastic film LAF to cope with the dimensional variation in the mounting height H. However, it is conceivable that the material cost increases when the thickness of the elastic film LAF is increased.

  Furthermore, since the elastic film LAF having a large thickness has poor followability to the upper mold UM, wrinkles are easily generated in the elastic film LAF. The wrinkles of the elastic film LAF are likely to occur particularly in a place where the size of the upper mold UM changes. For example, the wrinkle of the upper mold UM for securing the first space SP1 (sealed space) shown in FIG. It is easy to occur in the part. The wrinkles of the elastic film LAF generated in this way are transferred to the sealing body of the flow sensor FSP, which is a molded product (product), in the resin sealing step shown in FIG. As a result, an appearance defect of the flow sensor FSP occurs, resulting in a decrease in yield.

  Therefore, in the technical idea of the first embodiment, in order to improve the above-described matters, a contrivance is given focusing on the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1. Below, the technical idea in this Embodiment 1 which gave this device is demonstrated.

<Mounting configuration of flow sensor in embodiment 1>
FIG. 6 is a diagram showing a mounting configuration of the flow sensor FS1 in the first embodiment, and is a diagram showing a configuration before sealing with resin. In particular, FIG. 6A is a plan view showing the mounting configuration of the flow sensor FS1 in the first embodiment, and FIG. 6B is a cross-sectional view taken along the line AA in FIG. 6A. It is. FIG. 6C is a plan view showing the back surface of the semiconductor chip.

  First, as shown in FIG. 6A, the flow sensor FS1 in the first embodiment has a lead frame LF made of, for example, a copper material. The lead frame LF includes a chip mounting portion TAB1 and a chip mounting portion TAB2 inside the dam bar DM constituting the outer frame body. The semiconductor chip CHP1 is mounted on the chip mounting portion TAB1, and the semiconductor chip CHP2 is mounted on the chip mounting portion TAB2.

  The semiconductor chip CHP1 has a rectangular shape, and a flow rate detection unit FDU is formed substantially at the center. A wiring WL1 connected to the flow rate detection unit FDU is formed on the semiconductor chip CHP1, and the wiring WL1 is connected to a plurality of pads PD1 formed along one side of the semiconductor chip CHP1. That is, the flow rate detection unit FDU and the plurality of pads PD1 are connected by the wiring WL1.

  These pads PD1 are connected to a lead LD1 formed on the lead frame LF via a wire W1 made of, for example, a gold wire. The lead LD1 formed on the lead frame LF is further connected to a pad PD2 formed on the semiconductor chip CHP2 via a wire W2 made of, for example, a gold wire.

  In the semiconductor chip CHP2, an integrated circuit including semiconductor elements such as MISFETs and wirings is formed. Specifically, an integrated circuit constituting the CPU 1, the input circuit 2, the output circuit 3 or the memory 4 shown in FIG. 1 is formed. These integrated circuits are connected to the pads PD2 and PD3 that function as external connection terminals. The pad PD3 formed on the semiconductor chip CHP2 is connected to the lead LD2 formed on the lead frame LF via a wire W3 made of, for example, a gold wire. In this manner, the semiconductor chip CHP1 in which the flow rate detection unit FDU is formed and the semiconductor chip CHP2 in which the control circuit is formed are connected via the leads LD1 formed in the lead frame LF. Recognize.

  Although not shown here, a polyimide film may be formed on the outermost surface of the semiconductor chip CHP1 for the purpose of stress buffering with the resin to be bonded, surface protection, insulation, and the like.

  Note that the dam bar DM shown in FIG. 6A has a function of preventing resin leakage in a resin sealing process described later, and the dam bar DM is cut and removed after the resin sealing process.

  Subsequently, as shown in FIG. 6B, a chip mounting portion TAB1 is formed on the lead frame LF, and the semiconductor chip CHP1 is mounted on the chip mounting portion TAB1. The semiconductor chip CHP1 is bonded to the chip mounting portion TAB1 with an adhesive ADH1. A diaphragm DF (thin plate portion) is formed on the back surface of the semiconductor chip CHP1, and a flow rate detection unit FDU is formed on the surface of the semiconductor chip CHP1 facing the diaphragm DF.

  On the other hand, an opening OP1 is formed at the bottom of the chip mounting portion TAB1 existing below the diaphragm DF. Here, an example is shown in which the opening OP1 is formed at the bottom of the chip mounting portion TAB1 existing below the diaphragm DF, but the technical idea in the first embodiment is not limited to this. A lead frame LF in which the opening OP1 is not formed can also be used.

  Further, as shown in FIG. 6B, on the surface (upper surface) of the semiconductor chip CHP1, in addition to the flow rate detection unit FDU, a pad PD1 connected to the flow rate detection unit FDU is formed, and this pad PD1 Is connected to a lead LD1 formed on the lead frame LF via a wire W1. In addition to the semiconductor chip CHP1, the semiconductor chip CHP2 is also mounted on the lead frame LF, and the semiconductor chip CHP2 is bonded to the chip mounting portion TAB2 with an adhesive ADH1.

  Further, the pad PD2 formed on the semiconductor chip CHP2 and the lead LD1 formed on the lead frame LF are connected via a wire W2. Further, the pad PD3 formed on the semiconductor chip CHP2 and the lead LD2 formed on the lead frame LF are electrically connected through a wire W3.

  The pad PD1 of the first semiconductor chip CHP1 and the pad PD2 of the second semiconductor chip CHP2 can be directly connected by a gold wire.

  The adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 and the adhesive ADH1 that bonds the semiconductor chip CHP2 and the chip mounting portion TAB2 are, for example, thermosetting such as epoxy resin or polyurethane resin. An adhesive having a resin as a component and an adhesive having a thermoplastic resin such as a polyimide resin, an acrylic resin, or a fluororesin as a component can be used.

  In addition, the adhesive ADH1 contains a thermosetting resin or a thermoplastic resin as a main component, a metal material such as gold, silver, copper, or tin, or an inorganic material containing silica, glass, carbon, mica, talc, or the like as a component. By doing so, it is assumed that conductivity can be given and the linear expansion coefficient can be controlled.

  Here, the semiconductor chip CHP1 and the chip mounting portion TAB1 can be bonded by applying the adhesive ADH1 or attaching a sheet-like adhesive. For example, as shown in FIG. 6C, a sheet-like adhesive ADH1 is attached to the back surface of the semiconductor chip CHP1.

  At this time, in the first embodiment, for example, as illustrated in FIG. 6B, the adhesive ADH1 is also formed on the chip mounting portion TAB1 over a region overlapping with the diaphragm DF in plan view. Yes. In other words, it can also be said that the adhesive ADH1 is also formed on the projection surface of the diaphragm DF onto the chip mounting portion TAB1.

  In this case, at least one opening (through hole) OPA is formed in the adhesive ADH1, and the opening OPA formed in the adhesive ADH1 is formed in the chip mounting portion TAB1. It communicates with the opening OP1. As a result, the internal space formed immediately below the diaphragm DF is connected to the external space via the opening OPA and the opening OP1.

  FIG. 6C is a plan view showing the back surface of the semiconductor chip CHP1. As shown in FIG. 6C, a diaphragm DF is formed on the back surface of the semiconductor chip CHP1, and an adhesive ADH1 is formed on the back surface of the semiconductor chip CHP1 on which the diaphragm DF is formed. The adhesive material ADH1 is also formed in a region overlapping the diaphragm DF in a plan view, and an opening OPA is formed in the adhesive material ADH1 formed in the region overlapping the diaphragm DF. That is, the opening OPA is formed so as to overlap the diaphragm DF in a plane.

  Next, the mounting configuration of the flow sensor FS1 in the first embodiment after sealing with resin will be described. FIG. 7 is a diagram showing a mounting configuration of the flow sensor FS1 in the first embodiment, and is a diagram showing a configuration after sealing with resin. In particular, FIG. 7A is a plan view showing a mounting configuration of the flow sensor FS1 in the first embodiment. 7B is a cross-sectional view taken along the line AA in FIG. 7A, and FIG. 7C is a cross-sectional view taken along the line BB in FIG. 7A.

  In the flow rate sensor FS1 according to the first embodiment, as illustrated in FIG. 7A, a part of the semiconductor chip CHP1 and the semiconductor chip CHP2 are exposed while the flow rate detection unit FDU formed in the semiconductor chip CHP1 is exposed. The entire structure is covered with the resin MR. That is, in the first embodiment, the pad forming region of the semiconductor chip CHP1 and the entire region of the semiconductor chip CHP2 are collectively sealed with the resin MR while exposing the region where the flow rate detection unit FDU is formed. .

  Here, in the first embodiment, for example, a convex portion made of resin MR may be provided so as to cover the wire W1 electrically connected to the pad PD1 formed on the semiconductor chip CHP1. That is, in order to reliably seal a component such as a gold wire (wire) having a high loop height, a convex portion can be formed on the resin MR (sealing body). However, as shown in FIGS. 7A and 7B, in the first embodiment, the convex portion is not an essential component. That is, if the gold wire (wire) that electrically connects the pad PD1 and the lead LD1 formed on the semiconductor chip CHP1 can be sealed with the resin MR without providing the convex portion, the resin MR (sealing) It is not necessary to provide a convex part on the (stopping body).

  As the above-described resin MR, for example, a thermosetting resin such as an epoxy resin or a phenol resin, or a thermoplastic resin such as polycarbonate, polyethylene terephthalate, polyphenylene sulfide, or boricylene terephthalate can be used. Furthermore, the resin MR can be made conductive or linearly expanded by mixing a metal material such as gold, silver, copper, or tin, or an inorganic material containing silica, glass, carbon, mica, talc, or the like as a component. The coefficient may be controlled.

  According to the first embodiment, the sealing with the resin MR can be performed, for example, in a state where the semiconductor chip CHP1 in which the flow rate detection unit FDU is formed is fixed with a mold. As a result, according to the first embodiment, a part of the semiconductor chip CHP1 and the semiconductor chip CHP2 can be sealed with the resin MR while suppressing the displacement of the semiconductor chip CHP1. This means that according to the flow sensor FS1 in the first embodiment, a part of the semiconductor chip CHP1 and the entire region of the semiconductor chip CHP2 can be sealed with the resin MR while suppressing the displacement of each flow sensor FS1. This means that variation in the position of the flow rate detection unit FDU formed in the semiconductor chip CHP1 can be suppressed.

  From this, according to this Embodiment 1, since the position of the flow volume detection part FDU which detects the flow volume of gas can be made to correspond in each flow sensor FS1, the performance variation which detects gas flow volume in each flow sensor FS1. The remarkable effect which can suppress can be acquired.

  Here, in the first embodiment, the adhesive ADH1 is formed on the back surface of the semiconductor chip CHP1 on which the diaphragm DF is formed, and this adhesive ADH1 is also formed in a region that overlaps the diaphragm DF in plan view. Yes. An opening OPA is formed in the adhesive ADH1 formed in the region overlapping with the diaphragm DF. Further, as shown in FIGS. 7B and 7C, an opening OP1 is formed at the bottom of the chip mounting portion TAB1 below the diaphragm DF formed on the back surface of the semiconductor chip CHP1, and the chip mounting is performed. An opening OP2 is provided in the resin MR covering the back surface of the part TAB1.

  Thereby, according to the flow sensor FS1 according to the first embodiment, the inner space of the diaphragm DF includes the opening OPA formed in the adhesive ADH1, the opening OP1 formed in the bottom of the chip mounting portion TAB1, and It communicates with the external space of the flow sensor FS1 through the opening OP2 formed in the resin MR. As a result, the pressure in the inner space of the diaphragm DF and the pressure in the outer space of the flow rate sensor FS1 can be made equal, and it is possible to suppress the stress from being applied to the diaphragm DF.

  Further, in the flow rate sensor FS1 in the first embodiment, as shown in FIG. 7B, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is the diaphragm DF. The thickness of the adhesive material ADH1 formed in the inner space is smaller than the maximum value HDF. Further, in the flow rate sensor FS1 in the first embodiment, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is between the semiconductor chip CHP2 and the chip mounting portion TAB2. It is smaller than the thickness HCT2 of the adhesive ADH1 sandwiched between the two.

  In the first embodiment, the upper surface SUR (MR) of the resin MR is higher than the upper surface SUR (CHP) of the semiconductor chip CHP1, and the resin MR partially covers the upper portion of the semiconductor chip CHP1. It is shown. However, the configuration of the flow sensor FS1 in the first embodiment is not limited to such a shape. For example, the upper surface SUR (MR) of the resin MR is more than the upper surface SUR (CHP) of the semiconductor chip CHP1. May be present, and the resin MR may not partially cover the upper portion of the semiconductor chip CHP1 at this location.

<Method for Manufacturing Flow Sensor in First Embodiment (Feature of First Embodiment)>
The flow sensor FS1 in the first embodiment is configured as described above. Hereinafter, a method for manufacturing the flow sensor FS1 in the first embodiment will be described with reference to the drawings. First, using the cross-sectional view taken along the line BB in FIG. 7A, the characteristic process in the first embodiment will be described from the viewpoint of clarification (FIGS. 8 to 11). Thereafter, the flow rate sensor FS1 in the first embodiment will be described from the viewpoint of clarifying that it has a two-chip structure including the semiconductor chip CHP1 and the semiconductor chip CHP2 (FIGS. 12 to 16).

  First, as shown in FIG. 8, for example, a lead frame LF made of a copper material is prepared. The lead frame LF is formed with a chip mounting portion TAB1, and an opening OP1 is formed at the bottom of the chip mounting portion TAB1.

  Next, as shown in FIG. 9, an adhesive ADH1 is formed on the chip mounting portion TAB1. As this adhesive ADH1, for example, an adhesive made of a thermosetting resin such as an epoxy resin or a polyurethane resin, or an adhesive made of a thermoplastic resin such as a polyimide resin, an acrylic resin or a fluororesin may be used. it can. In addition, the adhesive ADH1 contains a thermosetting resin or a thermoplastic resin as a main component, a metal material such as gold, silver, copper, or tin, or an inorganic material containing silica, glass, carbon, mica, talc, or the like as a component. By doing so, conductivity may be given or the linear expansion coefficient may be controlled.

  Then, an opening OPA is formed in the adhesive ADH1. The opening OPA is formed so as to communicate with the opening OP1 formed in the chip mounting portion TAB1.

  Subsequently, as shown in FIG. 10, the semiconductor chip CHP1 is mounted on the chip mounting portion TAB1. Specifically, the semiconductor chip CHP1 is connected to the chip mounting portion TAB1 formed on the lead frame LF with an adhesive ADH1.

  At this time, in the first embodiment, the adhesive ADH1 is also formed in a region overlapping the diaphragm DF in a plan view, and the opening OPA is formed in the adhesive ADH1 formed in the region overlapping the diaphragm DF. Has been. The semiconductor chip CHP1 communicates with the opening OPA formed in the adhesive ADH1 and the opening OP1 formed in the bottom of the chip mounting portion TAB1 so that the diaphragm DF formed in the semiconductor chip CHP1 communicates. It is mounted on the chip mounting portion TAB1.

  In the semiconductor chip CHP1, a flow rate detection unit FDU, wiring (not shown), and pads (not shown) are formed by a normal semiconductor manufacturing process. Then, for example, the diaphragm DF is formed at the position of the back surface opposite to the flow rate detection unit FDU formed on the surface of the semiconductor chip CHP1 by anisotropic etching.

  Thereafter, although not shown in the drawing, the pads formed on the semiconductor chip CHP1 and the leads formed on the lead frame LF are connected by wires (wire bonding). This wire is formed of, for example, a gold wire.

  Next, as shown in FIG. 11, a part of the semiconductor chip CHP1 is sealed with a resin MR (molding process). That is, a part of the semiconductor chip CHP1 is sealed with the resin MR (sealing body) while exposing the flow rate detection unit FDU formed in the semiconductor chip CHP1.

  Specifically, first, an upper mold UM and a lower mold BM with a protruding pin EJPN inserted are prepared.

  Subsequently, a part of the upper mold UM is brought into close contact with the upper surface of the semiconductor chip CHP1, and a first space SP1 surrounding the flow rate detection unit FDU is formed between the upper mold UM and the semiconductor chip CHP1, The lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the mold UM and the lower mold BM via the second space.

  Thereafter, the resin MR is poured into the second space by the plunger under heating. At this time, as shown in FIG. 11, the protrusion pin EJPN is inserted into the lower mold BM. The protruding pin EJPN is configured to move up and down by a protruding pin plate (not shown), and the protruding pin plate (not shown) can be moved up and down by a pressurizing mechanism and a spring of the molding apparatus. It is configured.

  Next, when the resin MR is cured, the lead frame LF on which the semiconductor chip CHP1 is mounted is removed from the upper mold UM and the lower mold BM. Specifically, first, after removing the upper mold UM from the sealing body including the resin MR, the ejection pin EJPN also moves upward. As a result, the sealing body in which a part of the semiconductor chip CHP1 is sealed is released from the lower mold BM by the protrusion by the protrusion pin EJPN moved upward. As described above, the flow sensor FS1 in the first embodiment can be manufactured.

  Here, the first embodiment is characterized in that, in the resin sealing step shown in FIG. 11, the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is used as the upper mold UM and the lower mold BM. And the resin A is sealed in a state where the adhesive ADH1 is compressed. As a result, it is possible to absorb the dimensional variation in the mounting height caused by the variation in the thickness dimension of each component (lead frame LF, semiconductor chip CHP1).

  That is, in the first embodiment, even when the thickness dimension of each component (lead frame LF, semiconductor chip CHP1) varies, the adhesive ADH1 interposed between the semiconductor chip CHP1 and the chip mounting portion TAB1. However, by compressing and deforming the upper mold UM against the semiconductor chip CHP1, it is possible to absorb variations in the thickness of each component.

  At this time, the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed, so that a part of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is flat. In view, it is pushed out to the area overlapping the diaphragm DF and the outer area of the semiconductor chip CHP1.

  As a result, according to the first embodiment, in the flow rate sensor FS1 after resin sealing, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is equal to the diaphragm DF. It becomes smaller than the maximum value HDF of the thickness of the adhesive ADH1 formed in the inner space.

  In particular, in the flow rate sensor according to the first embodiment, the elastic modulus of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is smaller than the elastic modulus of the semiconductor chip CHP1. Specifically, for example, considering the case where the adhesive ADH1 is made of an epoxy resin, the elastic modulus of the epoxy resin is about 1 GPp to 20 GPa at room temperature. On the other hand, the elastic modulus of silicon (Si) constituting the semiconductor chip CHP1 is 193 GPa at room temperature.

  From this, it can be seen that according to the flow sensor FS1 in the first embodiment, the elastic modulus of the adhesive ADH1 is smaller than the elastic modulus of the semiconductor chip CHP1. This indicates that the adhesive ADH1 is softer than the semiconductor chip CHP1 and easily deforms. The elastic modulus is a proportional constant when the Hooke's law that stress and strain in an elastic body are proportional to each other is expressed in the form of “stress is proportional to strain”.

  As described above, in the first embodiment, a material having a lower elastic modulus than that of the semiconductor chip CHP1 is used for the adhesive ADH1. Thus, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the adhesive ADH1 is so absorbed that the semiconductor chip CHP1 absorbs the variation in the thickness dimension of the component before the semiconductor chip CHP1 is broken. The thickness changes. As a result, even when the mounting height of the component is higher than the average value, it is possible to prevent the semiconductor chip CHP1 from being applied more than necessary, and as a result, it is possible to prevent the semiconductor chip CHP1 from being broken. it can.

  On the other hand, for example, even when the component mounting height is lower than the average value, the upper mold UM and the lower mold are so compressed that the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed. By adjusting the dimension between the BM and the BM, a compressive load is applied to the end portion of the semiconductor chip CHP1, so that resin leakage onto the semiconductor chip CHP1 can be prevented.

  For example, as described in the related art, there is a possibility that the mounting height H may vary due to the variation in the thickness of each component (lead frame LF, semiconductor chip CHP1). If such a variation in the mounting height occurs, there is a risk that the semiconductor chip CHP1 breaks or the resin leaks onto the semiconductor chip CHP1 when the component is clamped by the upper mold UM and the lower mold BM. Rise.

  In this regard, as described in the related art described above, in order to prevent the semiconductor chip CHP1 from being broken, for example, as shown in FIG. 5, the elastic film LAF is interposed between the semiconductor chip CHP1 and the upper mold UM. The structure which interposes can be considered.

  When this configuration is employed, if the thickness of the semiconductor chip CHP1 is larger than the average thickness, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the semiconductor chip CHP1 The dimension of the elastic film LAF in the thickness direction changes so as to absorb the thickness. As a result, even if the thickness of the semiconductor chip CHP1 is larger than the average thickness, it is possible to prevent the semiconductor chip CHP1 from being subjected to an excessive force, and as a result, to prevent the semiconductor chip CHP1 from being broken. It is thought that you can.

  However, when the dimensional variation of the mounting height is large, there may be a case where the dimensional variation of the mounting height cannot be absorbed only by the dimensional change of the elastic film LAF. That is, even if the chip mounting portion TAB1 to which the semiconductor chip CHP1 is bonded is sandwiched between the upper mold UM and the lower mold BM via the elastic film LAF, the mounting height is so large that the thickness change of the elastic film LAF cannot be absorbed. If it is high, a clamping force more than necessary is applied to the semiconductor chip CHP1, and the possibility that the semiconductor chip CHP1 breaks becomes obvious.

  In this regard, it is conceivable to increase the thickness of the elastic film LAF to cope with the dimensional variation of the mounting height. However, increasing the thickness of the elastic film LAF increases the material cost. It is done.

  Furthermore, since the elastic film LAF having a large thickness has poor followability to the upper mold UM, wrinkles are easily generated in the elastic film LAF. The wrinkles of the elastic film LAF are likely to occur particularly in a place where the size of the upper mold UM changes. For example, the wrinkle of the upper mold UM for securing the first space SP1 (sealed space) shown in FIG. It is easy to occur in the part. The wrinkles of the elastic film LAF generated in this way are transferred to the sealing body of the flow sensor FSP, which is a molded product (product), in the resin sealing step shown in FIG. As a result, an appearance defect of the flow sensor FSP occurs, resulting in a decrease in yield.

  On the other hand, in the first embodiment, when there is a dimensional variation of a large mounting height, it is conceivable to increase the thickness of the adhesive ADH1 in order to absorb the dimensional variation of the large mounting height. In this case, unlike the elastic film LAF that needs to follow the shape of the upper mold UM, even if the thickness of the adhesive ADH1 is increased, no wrinkle is generated in the adhesive ADH1, and the wrinkle is There is no problem of being transferred to the appearance.

  Furthermore, since the installation area of the adhesive ADH1 is smaller than the installation area of the elastic film LAF installed in the entire upper mold UM, even if the thickness of the adhesive ADH1 is increased, the thickness of the elastic film LAF is increased. The material cost can be reduced as compared with the case.

  That is, according to the first embodiment, wrinkles generated in the elastic film LAF are transferred to the finished product as in the related art described above even when there is a large variation in the height of the mounting height. And no side effects represented by a significant increase in material costs. Therefore, according to the first embodiment, the flow sensor FS1 represented by the breakage of the semiconductor chip is generated without generating new concerns even when there is a dimensional variation in the large mounting height. Factors that decrease reliability can be effectively suppressed, and thereby the reliability of the flow sensor FS1 can be improved.

  As described above, in the first embodiment, the dimensional variation in the mounting height of the component is absorbed by the dimensional change in the thickness direction of the adhesive ADH1. Therefore, it is desirable that the elastic modulus of the adhesive ADH1 is smaller than the elastic modulus of the semiconductor chip CHP1 from the viewpoint of effectively absorbing the dimensional variation in the component mounting height and suppressing the breakage of the semiconductor chip CHP1. Recognize. Further, when the thickness of the adhesive ADH1 is small, the amount of dimensional change in the thickness direction is reduced, and there is a possibility that the dimensional variation of the mounting height cannot be absorbed sufficiently. It is necessary to make the thickness appropriate for the dimensional variation of the mounting height to be achieved.

  According to the first embodiment, for example, as shown in FIG. 11, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is the internal space of the diaphragm DF. It becomes smaller than the maximum value HDF of the thickness of the adhesive ADH1 formed in the above. That is, since the thickness HCT of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed and thinned, the thickness of the adhesive ADH1 near the end of the diaphragm DF is partially increased. Become. As a result, according to the first embodiment, the contact area between the end portion of the diaphragm DF and the adhesive ADH1 is increased, and separation between the semiconductor chip CHP1 and the chip mounting portion TAB1 can be effectively prevented.

  That is, according to the first embodiment, the adhesive ADH1 adheres not only to the region sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1, but also to a part of the diaphragm DF. For this reason, according to the first embodiment, the contact area between the semiconductor chip CHP1 and the adhesive ADH1 can be increased. Furthermore, according to the first embodiment, the elastic modulus of the adhesive ADH1 is smaller than the elastic modulus of the semiconductor chip CHP1.

  Therefore, according to the first embodiment, the adhesive ADH1 serves as a cushioning material that protects the semiconductor chip CHP1 from external impacts in addition to the originally provided function of bonding the semiconductor chip CHP and the chip mounting portion TAB1. Will also work. That is, according to the first embodiment, the elastic modulus of the adhesive ADH1 is smaller than the elastic modulus of the semiconductor chip CHP1, and the contact area with the semiconductor chip CHP1 is increased by contacting a part of the diaphragm DF. Due to the effect, the impact from the outside can be absorbed by the adhesive ADH1, so that the damage of the semiconductor chip CHP1 due to the impact from the outside can be suppressed.

  That is, according to the first embodiment, the semiconductor chip CHP1 can be prevented from being damaged during the manufacturing process of the flow sensor FS1, and the remarkable effect can be obtained that the semiconductor chip CHP1 can be prevented from being damaged in the completed flow sensor FS1. be able to.

  Specifically, as described above, according to the first embodiment, even when the dimensional variation of the mounting height of the component occurs, the dimensional variation is absorbed by the change in the thickness direction of the adhesive ADH1. can do. For this reason, in the resin sealing step (molding step), it is possible to suppress an unnecessary clamping force from the upper mold UM from being applied to the semiconductor chip CHP1, thereby suppressing the semiconductor chip CHP1 from being broken. it can.

  Furthermore, according to the first embodiment, even in the completed flow sensor FS1, the elastic modulus of the adhesive ADH1 is smaller than the elastic modulus of the semiconductor chip CHP1, and the adhesive is in contact with a part of the diaphragm DF. The contact area between the material ADH1 and the semiconductor chip CHP1 is increased. For this reason, according to the flow sensor FS1 in the first embodiment, it is possible to effectively absorb the external impact by the adhesive ADH1, and as a result, it is possible to suppress damage to the semiconductor chip CHP1 due to the external impact. It is.

  In the resin sealing process (molding process) in the first embodiment, since the upper mold UM and the lower mold BM having a high temperature of 80 ° C. or higher are used, the heated upper mold UM and the lower mold BM Heat is transferred from the mold BM to the resin MR injected into the second space in a short time. As a result, according to the manufacturing method of the flow sensor FS1 in the first embodiment, the heating / curing time of the resin MR can be shortened.

  For example, as described in the section of the problem to be solved by the invention, when only fixing a gold wire (wire) with a potting resin, the potting resin does not promote curing by heating, so the potting resin The time until curing becomes long, and the problem that the throughput in the manufacturing process of the flow sensor is lowered becomes obvious.

  On the other hand, in the resin sealing step in the first embodiment, as described above, since the heated upper mold UM and the lower mold BM are used, the heated upper mold UM and the lower mold BM are used. Heat conduction from the mold BM to the resin MR can be performed in a short time, and the heating / curing time of the resin MR can be shortened. As a result, according to the first embodiment, the throughput in the manufacturing process of the flow sensor FS1 can be improved. As described above, the flow sensor FS1 in the first embodiment can be manufactured.

  Subsequently, from the viewpoint of clarifying that the flow sensor FS1 in the first embodiment has a two-chip structure including the semiconductor chip CHP1 and the semiconductor chip CHP2, a method for manufacturing the flow sensor FS1 in the first embodiment will be described. (FIGS. 12 to 16).

  First, as shown in FIG. 12, for example, a lead frame LF made of a copper material is prepared. In the lead frame LF, a chip mounting portion TAB1, a chip mounting portion TAB2, a lead LD1, and a lead LD2 are integrally formed, and an opening OP1 is formed at the bottom of the chip mounting portion TAB1.

  Next, as shown in FIG. 13, an adhesive ADH1 is formed on the chip mounting portion TAB1 and the chip mounting portion TAB2. Then, an opening OPA is formed in the adhesive ADH1. The opening OPA is formed so as to communicate with the opening OP1 formed in the chip mounting portion TAB1.

  Subsequently, as shown in FIG. 14, the semiconductor chip CHP1 is mounted on the chip mounting portion TAB1, and the semiconductor chip CHP2 is mounted on the chip mounting portion TAB2. Specifically, the semiconductor chip CHP1 is connected to the chip mounting portion TAB1 formed on the lead frame LF with an adhesive ADH1. At this time, the semiconductor chip CHP1 is formed such that the diaphragm DF formed in the semiconductor chip CHP1 communicates with the opening OPA formed in the adhesive ADH1 and the opening OP1 formed in the bottom of the chip mounting portion TAB1. Is mounted on the chip mounting portion TAB1.

  The semiconductor chip CHP1 is formed with a flow rate detection unit FDU, wiring (not shown), and a pad PD1 by a normal semiconductor manufacturing process. And the diaphragm DF is formed in the position of the back surface facing the flow volume detection part FDU formed in the surface of the semiconductor chip CHP1 by anisotropic etching, for example. Further, the semiconductor chip CHP2 is also mounted on the chip mounting portion TAB2 formed on the lead frame LF by the adhesive ADH1. In the semiconductor chip CHP2, semiconductor elements (not shown) such as MISFETs, wirings (not shown), pads PD2, and pads PD3 are formed in advance by a normal semiconductor manufacturing process.

  At this stage, the thickness of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 and the thickness of the adhesive ADH1 that bonds the semiconductor chip CHP2 and the chip mounting portion TAB2 are the same. It has become.

  Next, as shown in FIG. 15, the pad PD1 formed on the semiconductor chip CHP1 and the lead LD1 formed on the lead frame LF are connected by a wire W1 (wire bonding). Similarly, the pad PD2 formed on the semiconductor chip CHP2 is connected to the lead LD1 and the wire W2, and the pad PD3 formed on the semiconductor chip CHP2 is connected to the lead LD2 and the wire W3. The wires W1 to W3 are made of, for example, a gold wire.

  Thereafter, as shown in FIG. 16, the surface of the semiconductor chip CHP1 excluding the flow rate detection unit FDU and its vicinity, the wire W1, the lead LD1, the wire W2, the entire main surface of the semiconductor chip CHP2, the wire W3 and a part of the lead LD2 Sealing with resin MR (molding process).

  Specifically, a part of the upper mold UM is brought into close contact with the upper surface of the semiconductor chip CHP1, and the first space SP1 surrounding the flow rate detection unit FDU is formed between the upper mold UM and the semiconductor chip CHP1, A lead frame LF on which the semiconductor chip CHP1 and the semiconductor chip CHP2 are mounted is sandwiched between the upper mold UM and the lower mold BM via the second space. That is, as shown in FIG. 16, the lead frame LF on which the semiconductor chip CHP1 and the semiconductor chip CHP2 are mounted is sandwiched between the upper mold UM and the lower mold BM through the second space (sealed space). Thereafter, the resin MR is poured into the second space (sealed space) under heating, so that the surface of the semiconductor chip CHP1 excluding the flow rate detection unit FDU and its vicinity, the wires W1, the leads LD1, the wires W2, and the semiconductor chips CHP2 The entire main surface, part of the wire W3 and the lead LD2 are sealed with resin MR.

  At this time, as shown in FIG. 16, the inner space of the diaphragm DF is separated from the second space by the adhesive ADH1, and therefore, when the second space is filled with the resin MR, the diaphragm DF It is possible to prevent the resin MR from entering the internal space.

  Further, in the first embodiment, since the semiconductor chip CHP1 in which the flow rate detection unit FDU is formed can be fixed with a mold, the position of the semiconductor chip CHP1 is suppressed while suppressing the displacement of the semiconductor chip CHP1. A part and the semiconductor chip CHP2 can be sealed with the resin MR. This is because, according to the flow sensor manufacturing method of the first embodiment, a part of the semiconductor chip CHP1 and the entire region of the semiconductor chip CHP2 can be sealed with the resin MR while suppressing the displacement of each flow sensor. This means that variation in the position of the flow rate detection unit FDU formed in the semiconductor chip CHP1 can be suppressed.

  As a result, according to the first embodiment, since the position of the flow rate detection unit FDU that detects the gas flow rate can be matched by each flow rate sensor, it is possible to suppress the performance variation of detecting the gas flow rate in each flow rate sensor. A remarkable effect can be obtained.

  Here, in the method of manufacturing the flow sensor according to the first embodiment, the lower mold is formed so that the flow rate detection unit FDU formed in the semiconductor chip CHP1 is surrounded by the first space SP1 isolated from the second space. The lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the BM and the upper mold UM. Thereby, according to this Embodiment 1, the surface area of other semiconductor chip CHP1 can be sealed, exposing the flow volume detection part FDU currently formed in semiconductor chip CHP1, and its vicinity area.

  Subsequently, as shown in FIG. 16, in the first embodiment, the resin MR also flows into the back surface side of the lead frame LF. Therefore, in the first embodiment, since the opening OP1 is formed at the bottom of the chip mounting portion TAB1, there is a concern that the resin MR flows into the inner space of the diaphragm DF from the opening OP1.

  Therefore, in the first embodiment, the shape of the lower mold BM that sandwiches the lead frame LF is devised. Specifically, as shown in FIG. 16, a protruding insertion piece IP1 is formed in the lower mold BM, and is formed in the lower mold BM when the lead frame LF is sandwiched between the upper mold UM and the lower mold BM. The projecting insert piece IP1 is inserted into an opening OP1 formed at the bottom of the chip mounting portion TAB1. As a result, the insertion piece IP1 is inserted into the opening OPA and the opening OP1 without a gap, so that the resin MR can be prevented from entering the inner space of the diaphragm DF from the opening OP1.

  That is, in the first embodiment, a protruding insertion piece IP1 is formed in the lower mold BM, and this insertion piece IP1 is inserted into the opening OP1 formed at the bottom of the chip mounting portion TAB1 during resin sealing. doing. Note that the insert piece IP1 may be configured to penetrate not only the opening OP1, but also the opening OPA formed in the adhesive ADH1.

  For example, in the first embodiment, in the resin sealing process (molding process), a part of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 overlaps the diaphragm DF in plan view. It is pushed out to the region and the outer region of the semiconductor chip CHP1. In this case, it is conceivable that a part of the adhesive ADH1 extruded into the region overlapping with the diaphragm DF blocks the opening OPA formed in the adhesive ADH1. Even in this case, by configuring the insert piece IP1 so as to penetrate not only the opening OP1 but also the opening OPA formed in the adhesive ADH1, the opening OPA formed in the adhesive ADH1 is blocked. It can be prevented from being peeled off.

  Furthermore, in the first embodiment, the shape of the insert piece IP1 is devised. Specifically, in the first embodiment, the insert piece IP1 includes an insertion portion that is inserted into the opening OP1 and a pedestal portion that supports the insertion portion. The cross-sectional area is large. Thus, the insert piece IP1 has a structure in which a step portion is provided between the insertion portion and the pedestal portion, and the step portion is in close contact with the bottom surface of the chip mounting portion TAB1.

  By configuring the insert piece IP1 in this way, the following effects can be obtained. For example, in the case where the shape of the insert piece IP1 is configured only from the insertion portion described above, the insertion portion is inserted into the opening portion OP1, so the diameter of the insertion portion of the insertion piece IP1 is slightly smaller than the diameter of the opening portion OP1. It has become. Therefore, when the insertion piece IP1 is configured only from the insertion portion, even if the insertion portion of the insertion piece IP1 is inserted into the opening OP1, there is a slight gap between the inserted insertion portion and the opening OP1. Conceivable. In this case, the resin MR may enter the inner space of the diaphragm DF from the gap.

  Therefore, in the first embodiment, the insertion piece IP1 is configured to be formed on the pedestal portion having a larger cross-sectional area than the insertion portion. In this case, as shown in FIG. 16, the insertion portion of the insertion piece IP1 is inserted into the opening OP1, and the pedestal portion of the insertion piece IP1 comes into close contact with the bottom surface of the chip mounting portion TAB1. As a result, even if a slight gap is generated between the insertion part of the insertion piece IP1 and the opening OP1, the pedestal part is firmly pressed against the back surface of the chip mounting part TAB1, so that the resin MR enters the opening OP1. It can be prevented. That is, in the first embodiment, the insertion piece IP1 is configured to provide the insertion portion on the pedestal portion having a larger cross-sectional area than the insertion portion, and therefore the resin MR reaches the opening OP1 by the pedestal portion. The combination of the fact that the step portion formed between the pedestal portion and the insertion portion is pressed against the chip mounting portion TAB1 makes the resin MR the internal space of the diaphragm DF through the opening OP1. It is possible to effectively prevent intrusion.

  Thereafter, when the resin MR is cured, the lead frame LF on which the semiconductor chip CHP1 and the semiconductor chip CHP2 are mounted is removed from the upper mold UM and the lower mold BM. Thereby, the flow sensor FS1 in the first embodiment can be manufactured.

  In the flow sensor FS1 manufactured at this time, as a result of using the lower mold BM in which the insertion piece IP1 is formed in the resin sealing process, as shown in FIG. 7B, for example, on the bottom surface of the chip mounting portion TAB1. An opening OP1 is formed, and an opening OP2 communicating with the opening OP1 is formed in the resin MR. The opening OP2 is produced as a result of forming a pedestal on the insert piece IP1, and the cross-sectional area of the opening OP2 is larger than the cross-sectional area of the opening OP1. As a result, according to the flow sensor FS1 according to the first embodiment, the internal space of the diaphragm DF flows through the opening OP1 formed in the bottom of the chip mounting portion TAB1 and the opening OP2 formed in the resin MR. It communicates with the external space of the sensor FS1. As a result, the pressure in the inner space of the diaphragm DF and the pressure in the outer space of the flow rate sensor FS1 can be made equal, and it is possible to suppress the stress from being applied to the diaphragm DF.

  Here, the first embodiment is characterized in that, in the resin sealing step shown in FIG. 16, the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is used as the upper mold UM and the lower mold BM. And the resin A is sealed with the adhesive ADH1 compressed. As a result, it is possible to absorb the dimensional variation in the mounting height caused by the variation in the thickness dimension of each component (lead frame LF, semiconductor chip CHP1).

  That is, in the first embodiment, even when the thickness dimension of each component (lead frame LF, semiconductor chip CHP1) varies, the adhesive ADH1 interposed between the semiconductor chip CHP1 and the chip mounting portion TAB1. However, by compressing and deforming the upper mold UM against the semiconductor chip CHP1, it is possible to absorb variations in the thickness of each component.

  At this time, the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed, so that a part of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is flat. In view, it is pushed out to the area overlapping the diaphragm DF and the outer area of the semiconductor chip CHP1.

  As a result, according to the first embodiment, in the flow rate sensor FS1 after resin sealing, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is equal to the diaphragm DF. It becomes smaller than the maximum value HDF of the thickness of the adhesive ADH1 formed in the inner space.

  Further, in the first embodiment, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is sandwiched between the semiconductor chip CHP2 and the chip mounting portion TAB2. It can also be said that the thickness of the adhesive ADH1 is thinner than the thickness HCT2. That is, as shown in FIG. 16, in the semiconductor chip CHP1, since the semiconductor chip CHP1 is pressed (clamped) by the upper mold UM, the adhesion sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1. As a result of the compression of the material ADH1, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is reduced. On the other hand, in the semiconductor chip CHP2, since the entire semiconductor chip CHP2 is sealed with the resin MR, the semiconductor chip CHP2 itself is not suppressed (clamped) by the upper mold UM. For this reason, a large compressive force is not applied to the adhesive ADH1 sandwiched between the semiconductor chip CHP2 and the chip mounting portion TAB2. As a result, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is equal to the thickness of the adhesive ADH1 sandwiched between the semiconductor chip CHP2 and the chip mounting portion TAB2. It is thinner than HCT2.

  As described above, the first embodiment is characterized in that the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is between the semiconductor chip CHP2 and the chip mounting portion TAB2. It can be seen that the adhesive material ADH1 sandwiched becomes thinner than the thickness HCT2 of the adhesive material ADH1.

<Typical effects in the first embodiment>
According to the flow sensor FS1 in the first embodiment, the following effects can be obtained.

  (1) According to the first embodiment, the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 uses a material having a lower elastic modulus than the semiconductor chip CHP1. For this reason, for example, as shown in FIG. 11, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the adhesive ADH1 absorbs variations in the mounting dimensions of the components. Furthermore, the dimension in the thickness direction of the adhesive ADH1 changes.

  As a result, even when the mounting height is higher than the average value, it is possible to prevent the semiconductor chip CHP1 from being excessively applied with force, and as a result, it is possible to prevent the semiconductor chip CHP1 from being broken. On the other hand, even when the mounting height is low, the dimension between the upper mold UM and the lower mold BM is adjusted so that the adhesive ADH1 bonding the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed. By doing so, a compressive load can be applied to the end of the semiconductor chip CHP1, thereby preventing resin leakage onto the semiconductor chip CHP1.

  That is, the dimensional variation in the mounting height of the components due to the variation in the thickness of the semiconductor chip CHP1 and the lead frame LF is absorbed by the change in the thickness of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1. Can do. Thus, according to the first embodiment, the clamping force applied to the semiconductor chip CHP1 can be relaxed. As a result, it is possible to prevent breakage such as cracks, chips or cracks of the semiconductor chip CHP1.

  The feature of the first embodiment described above is that, for example, as shown in FIG. 7B, the flow sensor FS1 after resin sealing is sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1. It can be said that the thickness HCT of the rare adhesive material ADH1 is manifested as being smaller than the maximum value HDF of the thickness of the adhesive material ADH1 formed in the internal space of the diaphragm DF.

  In particular, when the flow sensor FS1 has a two-chip structure, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is equal to the semiconductor chip CHP2 and the chip mounting portion TAB2. It can be said that the feature of the first embodiment is also manifested in that the thickness is smaller than the thickness HCT2 of the adhesive ADH1 sandwiched between the two.

  (2) According to the first embodiment, for example, as shown in FIG. 7B, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is the diaphragm. It becomes smaller than the maximum value HDF of the thickness of the adhesive ADH1 formed in the internal space of the DF. That is, since the thickness HCT of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed and thinned, the thickness of the adhesive ADH1 near the end of the diaphragm DF is partially increased. Become. As a result, according to the first embodiment, the contact area between the end portion of the diaphragm DF and the adhesive ADH1 is increased, and separation between the semiconductor chip CHP1 and the chip mounting portion TAB1 can be effectively prevented.

  (3) Further, according to the first embodiment, the adhesive ADH1 protects the semiconductor chip CHP1 from external impacts in addition to the originally provided function of bonding the semiconductor chip CHP1 and the chip mounting portion TAB1. It will also function as a cushioning material. That is, according to the first embodiment, the elastic modulus of the adhesive ADH1 is smaller than the elastic modulus of the semiconductor chip CHP1, and the contact area with the semiconductor chip CHP1 is increased by contacting a part of the diaphragm DF. Due to the effect, the impact from the outside can be absorbed by the adhesive ADH1, so that the damage of the semiconductor chip CHP1 due to the impact from the outside can be suppressed.

  (4) In the first embodiment, for example, as shown in FIGS. 11 and 16, the semiconductor chip CHP1 in which the flow rate detection unit FDU is formed can be performed in a state of being fixed by a mold. A part of the semiconductor chip CHP1 and the semiconductor chip CHP2 can be sealed with the resin MR while suppressing the displacement of the CHP1.

  This is because, according to the flow sensor manufacturing method of the first embodiment, a part of the semiconductor chip CHP1 and the entire region of the semiconductor chip CHP2 can be sealed with the resin MR while suppressing the displacement of each flow sensor. This means that variation in the position of the flow rate detection unit FDU formed in the semiconductor chip CHP1 can be suppressed.

  As a result, according to the first embodiment, since the position of the flow rate detection unit FDU that detects the gas flow rate can be matched by each flow rate sensor, it is possible to suppress the performance variation of detecting the gas flow rate in each flow rate sensor. A remarkable effect can be obtained.

<Modification 1>
In the first embodiment, the example in which the dimensional variation in the mounting height of the component is absorbed by the change in the thickness of the adhesive ADH1 has been described. On the other hand, in the first modification, the variation in the mounting height of the component is absorbed by the change in the thickness of the adhesive ADH1, and also due to the change in the thickness of the elastic film LAF attached to the upper mold UM. An example of absorption will be described.

  FIG. 17 is a cross-sectional view showing the resin sealing step of the flow rate sensor according to the first modification, corresponding to the cut surface taken along the line BB in FIG. FIG. 18 is a cross-sectional view showing the resin sealing process of the flow rate sensor according to the first modification, corresponding to the cut surface taken along the line AA in FIG.

  The resin sealing step in the first modification shown in FIGS. 17 and 18 has, for example, the same configuration as the resin sealing step in the first embodiment shown in FIGS. 11 and 16, and the differences are as follows. An elastic film LAF having an elastic modulus smaller than that of the semiconductor chip CHP1 is attached to the upper mold UM.

  In the resin sealing process in Modification 1 thus configured, as in the first embodiment, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, The dimension in the thickness direction of the adhesive ADH1 changes so as to absorb the variation in the mounting dimensions of the components.

  As a result, even when the mounting height is higher than the average value, it is possible to prevent the semiconductor chip CHP1 from being excessively applied with force, and as a result, it is possible to prevent the semiconductor chip CHP1 from being broken. On the other hand, even when the mounting height is low, the dimension between the upper mold UM and the lower mold BM is adjusted so that the adhesive ADH1 bonding the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed. By doing so, a compressive load can be applied to the end of the semiconductor chip CHP1, thereby preventing resin leakage onto the semiconductor chip CHP1.

  Further, in the resin sealing step in the first modification, for example, when the thickness of the semiconductor chip CHP1 is thinner than the average thickness, the lead frame LF on which the semiconductor chip CHP1 is mounted is replaced with the upper mold UM and the lower mold BM. It is conceivable that a gap is formed when sandwiched between the two, but since this gap can be filled with the elastic film LAF, resin leakage onto the semiconductor chip CHP1 can be prevented.

  On the other hand, when the thickness of the semiconductor chip CHP1 is larger than the average thickness, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the elastic film LAF is used for the semiconductor chip CHP1. Therefore, the dimension in the thickness direction of the elastic film LAF changes so as to absorb the dimensional variation in the mounting height of the component. As a result, even if the thickness of the semiconductor chip CHP1 is larger than the average thickness, it is possible to prevent the semiconductor chip CHP1 from being subjected to an excessive force, and as a result, to prevent the semiconductor chip CHP1 from being broken. Can do.

  As described above, according to the resin sealing process in the first modification, the thickness of the adhesive ADH1 interposed between the semiconductor chip CHP1 and the chip mounting portion TAB1 is affixed to the upper mold UM. The variation in the mounting height of the component can be absorbed by both the change in the thickness of the elastic film LAF. As a result, also in the first modification, it is possible to prevent the semiconductor chip CHP1 from being broken.

  In the first modification, for example, even when the dimensional variation in the mounting height of the component becomes large, the component is caused by both the change in thickness due to the adhesive ADH1 and the change in thickness due to the elastic film LAF. The dimensional variation of the mounting height can be effectively absorbed. Therefore, for example, even when there is a dimensional variation in the mounting height of a component that cannot be absorbed only by the change in thickness due to the adhesive ADH1, by utilizing the change in thickness due to the elastic film LAF, It is possible to effectively absorb the dimensional variation in the mounting height of the component.

  Further, in the first modification, even when the dimensional variation in the mounting height of the component is large, the change in the thickness of the adhesive ADH1 is also used, so that only the change in the thickness of the elastic film LAF is used. It is not necessary to increase the thickness of the elastic film LAF as compared to the case where the dimensional variation in the mounting height of the component is absorbed. As a result, according to the first modification, even when the elastic film LAF is used, it is possible to suppress the generation of wrinkles in the elastic film LAF, thereby reducing the appearance defect of the flow sensor due to the wrinkles. Can be suppressed. Moreover, since it is not necessary to increase the thickness of the elastic film LAF, an increase in manufacturing cost can be suppressed.

<Modification 2>
Next, the flow sensor FS1 in the second modification will be described. In the first embodiment, for example, as shown in FIG. 7B and FIG. 7C, the example in which the semiconductor chip CHP1 is disposed on the chip mounting portion TAB1 via the adhesive ADH1 has been described. In the second modification, for example, as illustrated in FIG. 19, an example in which a plate-like structure PLT is inserted between the semiconductor chip CHP1 and the chip mounting portion TAB1 will be described.

  FIG. 19 is a diagram illustrating the structure of the flow sensor FS1 after resin sealing in the second modification. FIG. 19A is a plan view showing the structure of the flow sensor FS1 after resin sealing. FIG. 19B is a cross-sectional view taken along the line AA in FIG. 19A, and FIG. 19C is a cross-sectional view taken along the line BB in FIG. is there.

  As shown in FIGS. 19B and 19C, in the flow rate sensor FS1 in the second modification, a plate-like structure PLT is formed over the lower layer of the semiconductor chip CHP1 and the lower layer of the semiconductor chip CHP2. Recognize. The plate-like structure PLT has, for example, a rectangular shape, and has an outer dimension that includes the semiconductor chip CHP1 and the semiconductor chip CHP2 in plan view. That is, in the second modification, the plate-like structure PLT is mounted on the chip-mounting portion TAB1 and the chip-mounting portion TAB2 via the adhesive ADH2, and the semiconductor chip CHP1 and the chip AHP1 are mounted on the plate-like structure PLT by the adhesive ADH1. The semiconductor chip CHP2 is mounted.

  Further, a groove DPLT is formed in the plate-like structure PLT, and the groove DPLT connects the internal space of the diaphragm DF formed in the semiconductor chip CHP1 and the opening OP3 formed in the resin MR. Yes. As a result, the pressure in the inner space of the diaphragm DF and the pressure in the outer space of the flow rate sensor FS1 can be made equal, and it is possible to suppress the stress from being applied to the diaphragm DF.

  Here, in the second modification, the opening OP3 formed in the resin MR is formed in a region that does not overlap the diaphragm DF in a planar manner. This advantage will be described. For example, in the first embodiment shown in FIG. 7B, the opening OP2 is formed in a region directly below the diaphragm DF in plan view. Also in this case, the inner space of the diaphragm DF and the outer space of the flow sensor FS1 communicate with each other via the opening OP1 and the opening OP2, and the pressure in the inner space of the diaphragm DF and the outer space of the flow sensor FS1 The pressure can be made equal.

  However, in the case of such a configuration, the opening OP2 is arranged at a position where the gas flows. That is, the pressure of the external space becomes unstable due to the gas flowing in the external space near the opening OP2. That is, when the internal space of the diaphragm DF and the external space are communicated with each other by the opening OP2 formed in the region immediately below the diaphragm DF, the pressure in the internal space of the diaphragm DF is caused by the flow of gas flowing through the external space. May become unstable.

  Therefore, in the second modification, for example, as shown in FIG. 19B, the opening OP3 formed in the resin MR is formed in a region that does not overlap with the diaphragm DF in a plan view so as to be away from the position where the gas flows. It is arranged. Thereby, according to this modification 2, the pressure of the internal space of diaphragm DF can be stabilized, without being influenced by the flow of gas. That is, according to the second modification, the opening OP3 is provided in a place that is not easily affected by the gas flow, so that the pressure in the inner space of the diaphragm DF is equal to the pressure in the outer space of the flow sensor FS1, and the diaphragm The pressure in the internal space of the DF can be stabilized.

  Next, in the second modification, for example, as shown in FIG. 19B, the plate-like structure PLT is arranged on the chip mounting portion TAB1 and the chip mounting portion TAB2. The plate-like structure PLT is bonded to the chip mounting portion TAB1 and the chip mounting portion TAB2 using, for example, an adhesive ADH2, but can also be bonded using a paste material.

  On the plate-like structure PLT, the semiconductor chip CHP1 is mounted via the adhesive ADH1, and the semiconductor chip CHP2 is mounted via the adhesive ADH1. At this time, when the plate-like structure PLT is made of a metal material, it can be connected to the semiconductor chip CHP1 with a wire, and can also be connected to the semiconductor chip CHP2 with a wire.

  The plate-like structure PLT described above mainly functions as a cushioning material against an improvement in rigidity of the flow sensor FS1 and an external impact. Further, FIG. 19B shows an example in which the pad PD1 of the semiconductor chip CHP1 and the pad PD2 of the semiconductor chip CHP2 are directly connected by a gold wire. However, when the plate-like structure PLT is made of a conductive material, it is electrically connected to the semiconductor chip CHP1 (pad PD1) or the semiconductor chip CHP2 (pad PD2) and used to supply a ground potential (reference potential). It is also possible to stabilize the ground potential. For example, when the plate-like structure PLT uses a highly rigid material such as a metal material, the rigidity of the flow sensor FS1 can be improved. On the other hand, when using a material with low rigidity, such as a resin material, in the resin sealing step, the dimensional variation in the mounting height of the components clamped between the upper mold UM and the lower mold BM is reduced to a plate-like structure. It can also be absorbed by deformation of PLT.

  The plate-like structure PLT is, for example, a thermoplastic resin such as PBT resin, ABS resin, PC resin, nylon resin, PS resin, PP resin, or fluorine resin, or thermosetting resin such as epoxy resin, phenol resin, or urethane resin. It can consist of In this case, the plate-like structure PLT can mainly function as a buffer material that protects the semiconductor chip CHP1 and the semiconductor chip CHP2 from external impacts.

  On the other hand, the plate-like structure PLT can be formed by pressing a metal material such as an iron alloy, an aluminum alloy, or a copper alloy, or can be formed from a glass material. In particular, when the plate-like structure PLT is formed from a metal material, the rigidity of the flow sensor FS1 can be increased. Furthermore, the plate-like structure PLT can be electrically connected to the semiconductor chip CHP1 and the semiconductor chip CHP2, and the plate-like structure PLT can be used for supplying the ground potential and stabilizing the ground potential.

  When the plate-like structure PLT is composed of a thermoplastic resin or a thermosetting resin, an inorganic filler such as glass, talc, silica, mica, or an organic filler such as carbon is used for the thermoplastic resin or thermosetting resin. Can be filled. The plate-like structure PLT can be molded by filling a mold with a resin by a transfer molding method, or can be formed by arbitrarily laminating sheet-shaped products by roll processing.

  Further, an adhesive ADH1 that bonds the semiconductor chip CHP1, the semiconductor chip CHP2, and the plate-like structure PLT, or an adhesive ADH2 that bonds the plate-like structure PLT, the chip mounting portion TAB1, and the chip mounting portion TAB2. For example, an adhesive having a thermosetting resin such as an epoxy resin or a polyurethane resin as a component, or an adhesive having a thermoplastic resin such as a polyimide resin, an acrylic resin, or a fluorine resin as a component can be used. In addition, it has conductivity by mixing metallic materials such as gold, silver, copper, tin, and inorganic materials such as silica, quartz, carbon, mica, talc, etc. with thermosetting resin or thermoplastic resin as the main component. Or the linear expansion coefficient can be controlled.

  Also in the flow rate sensor FS1 in the second modification configured as described above, the variation in the mounting height of the components can be absorbed by the deformation of the adhesive ADH1, as in the first embodiment.

  For example, according to the second modification, the adhesive ADH1 that bonds the semiconductor chip CHP1 and the plate-like structure PLT is made of a material having a lower elastic modulus than the semiconductor chip CHP1. For this reason, for example, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the adhesive ADH1 has a thickness of the adhesive ADH1 so as to absorb variations in the mounting dimensions of components. The vertical dimension changes.

  At this time, the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the plate-like structure PLT is compressed, so that a part of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the plate-like structure PLT is obtained. In plan view, it is pushed out to the area overlapping the diaphragm DF and the outer area of the semiconductor chip CHP1.

  As a result, according to the second modification, in the flow rate sensor FS1 after resin sealing, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the plate-like structure PLT is equal to the diaphragm DF. It becomes smaller than the maximum value HDF of the thickness of the adhesive ADH1 formed in the inner space.

  From the above, even when the mounting height is higher than the average value, it is possible to prevent an unnecessary force from being applied to the semiconductor chip CHP1, and as a result, it is possible to prevent the semiconductor chip CHP1 from being broken. . On the other hand, even when the mounting height is low, the dimension between the upper mold UM and the lower mold BM is set so that the adhesive ADH1 bonding the semiconductor chip CHP1 and the plate-like structure PLT is compressed. By adjusting, it is possible to apply a compressive load to the end of the semiconductor chip CHP1, thereby preventing resin leakage onto the semiconductor chip CHP1.

  That is, the dimensional variation in the mounting height of the components due to the variation in the thickness of the semiconductor chip CHP1 and the lead frame LF is absorbed by the change in the thickness of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the plate-like structure PLT. be able to. Thus, according to the second modification, the clamping force applied to the semiconductor chip CHP1 can be relaxed. As a result, it is possible to prevent breakage such as cracks, chips or cracks of the semiconductor chip CHP1.

  Further, in the second modification, without using the elastic film LAF, only the dimensional change of the adhesive material ADH1 bonding the semiconductor chip CHP1 and the plate-like structure PLT causes the dimensional variation in the mounting height of the component. Although it is assumed that it absorbs, for example, as in Modification 1, the elastic film LAF is used, and the dimensional variation of the mounting height of the component is also absorbed by the dimensional change caused by the elastic film LAF. You may comprise as follows. That is, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the dimensional change of the elastic film LAF is bonded to the semiconductor chip CHP1 and the plate-like structure PLT. By using the dimensional change of the material ADH1 together, it is possible to cope with a dimensional variation of a larger mounting height.

  Further, in the second modification, the variation in the mounting height of the component is also absorbed by the change in the thickness of the adhesive ADH2 that bonds the plate-like structure PLT to the chip mounting portion TAB1 and the chip mounting portion TAB2. be able to.

  At this time, the adhesive material ADH1 and the adhesive material ADH2 are more elastic than the semiconductor chip CHP1 in order to absorb the dimensional variation in the mounting height of the components due to the dimensional change in the thickness direction of the adhesive material ADH1 and the adhesive material ADH2. Need to be low. In addition, when the thickness of the adhesive material ADH1 and the adhesive material ADH2 is thin, the amount of dimensional change in the thickness direction is small, and the dimensional variation in the mounting height cannot be sufficiently absorbed. Therefore, the adhesive material ADH1 and the adhesive material ADH2 The thickness needs to be an appropriate dimension corresponding to the dimensional variation of the mounting height.

(Embodiment 2)
In the first embodiment, for example, as illustrated in FIG. 7B, the flow sensor FS1 having a two-chip structure including the semiconductor chip CHP1 and the semiconductor chip CHP2 has been described as an example. The technical idea of the present invention is not limited to this. For example, the technical idea of the present invention can be applied to a flow sensor having a one-chip structure including one semiconductor chip in which a flow rate detection unit and a control unit (control circuit) are integrally formed. . In the second embodiment, a case where the technical idea of the present invention is applied to a flow sensor having a one-chip structure will be described as an example.

<Mounting configuration of flow sensor in embodiment 2>
FIG. 20 is a diagram illustrating a mounting configuration of the flow rate sensor FS2 in the second embodiment, and is a diagram illustrating a configuration after sealing with resin. In particular, FIG. 20A is a plan view showing a mounting configuration of the flow rate sensor FS2 in the second embodiment. 20B is a cross-sectional view taken along the line AA in FIG. 20A, and FIG. 20C is a cross-sectional view taken along the line BB in FIG. 20A. In particular, FIG. 20B shows a cross section parallel to the traveling direction of the gas flowing on the exposed flow rate detection unit FDU. In FIG. 20B, the gas is, for example, left on the X axis. It flows from the right to the right.

  First, as shown in FIG. 20A, the flow sensor FS2 in the second embodiment has a sealing body including a resin MR having a rectangular shape, and a lead LD2 protrudes from the resin MR. A part of the semiconductor chip CHP1 is exposed from the upper surface (surface) of the resin MR. In particular, the semiconductor chip CHP1 is formed with a flow rate detection unit FDU and a control unit that controls the flow rate detection unit FDU. Specifically, the flow rate detection unit FDU formed in the semiconductor chip CHP1 is electrically connected to the control unit by the wiring WL1. In FIG. 20A, this control unit is covered with the resin MR, and is not shown, but is disposed inside the resin MR. That is, the flow rate sensor FS2 according to the second embodiment includes the semiconductor chip CHP1 in which the flow rate detection unit FDU and the control unit are integrally formed, and the flow rate detection unit FDU is exposed from the resin MR. It will be.

  Next, as shown in FIG. 20B, it can be seen that the flow rate sensor FS2 in the present second embodiment has the semiconductor chip CHP1 mounted on the chip mounting portion TAB1 via the adhesive ADH1. At this time, a flow rate detection unit FDU is formed on the upper surface (front surface, main surface) of the semiconductor chip CHP1, and a diaphragm DF (thin plate portion) is formed on the back surface of the semiconductor chip CHP1 facing the flow rate detection unit FDU. ing. On the other hand, an opening OP1 is formed at the bottom of the chip mounting portion TAB1 existing below the diaphragm DF.

  The adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 is, for example, an adhesive containing a thermosetting resin such as an epoxy resin or a polyurethane resin, a polyimide resin, an acrylic resin, a fluorine resin, or the like. It is possible to use an adhesive comprising as a component a thermoplastic resin. In addition, it has conductivity by mixing metallic materials such as gold, silver, copper, tin, and inorganic materials such as silica, quartz, carbon, mica, talc, etc. with thermosetting resin or thermoplastic resin as the main component. Or the linear expansion coefficient can be controlled.

  Here, as shown in FIG. 20B, in the flow rate sensor FS2 in the second embodiment, the resin MR is formed so as to cover a part of the side surface and upper surface of the semiconductor chip CHP1 and a part of the chip mounting portion TAB1. Has been.

  At this time, in the second embodiment, the opening OP1 is formed at the bottom of the chip mounting portion TAB1 below the diaphragm DF formed on the back surface of the semiconductor chip CHP1, and the resin that covers the back surface of the chip mounting portion TAB1. An opening OP2 is provided in the MR.

  Thereby, according to the flow rate sensor FS2 according to the second embodiment, the internal space of the diaphragm DF flows through the opening OP1 formed in the bottom of the chip mounting portion TAB1 and the opening OP2 formed in the resin MR. It communicates with the external space of the sensor FS2. As a result, the pressure in the inner space of the diaphragm DF and the pressure in the outer space of the flow rate sensor FS2 can be equalized, and stress can be suppressed from being applied to the diaphragm DF. Further, also in the second embodiment, as shown in FIG. 20B, the upper surface SUR (MR) of the resin MR is formed to be higher than the upper surface SUR (CHP) of the semiconductor chip CHP1.

  As shown in FIG. 20C, the semiconductor chip CHP1 is mounted on the chip mounting portion TAB1 via the adhesive ADH1, and the flow rate detection unit FDU and the control unit CU are formed on the upper surface of the semiconductor chip CHP1. It can be seen that it is formed. That is, in the second embodiment, it can be seen that the flow rate detection unit FDU and the control unit CU are integrally formed on the semiconductor chip CHP1. Further, a pad PD is formed on the upper surface of the semiconductor chip CHP1, and the pad PD and the lead LD2 are electrically connected by a wire W. The control unit CU and the pad PD formed on the upper surface of the semiconductor chip CHP1 and the wire W are sealed with a resin MR.

  Also in the flow rate sensor FS2 according to the second embodiment configured as described above, a resin sealing step that absorbs the dimensional variation in the mounting height of the components can be realized as in the first embodiment.

  For example, according to the second embodiment, the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 uses a material having a lower elastic modulus than the semiconductor chip CHP1. For this reason, for example, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the adhesive ADH1 has a thickness of the adhesive ADH1 so as to absorb variations in the mounting dimensions of components. The vertical dimension changes.

  At this time, the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed, so that a part of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is flat. In view, it is pushed out to the area overlapping the diaphragm DF and the outer area of the semiconductor chip CHP1.

  As a result, according to the second embodiment, in the flow rate sensor FS2 after resin sealing, the thickness HCT of the adhesive ADH1 sandwiched between the semiconductor chip CHP1 and the chip mounting portion TAB1 is equal to the diaphragm DF. It becomes smaller than the maximum value HDF of the thickness of the adhesive ADH1 formed in the inner space.

  From the above, even when the mounting height is higher than the average value, it is possible to prevent an unnecessary force from being applied to the semiconductor chip CHP1, and as a result, it is possible to prevent the semiconductor chip CHP1 from being broken. . On the other hand, even when the mounting height is low, the dimension between the upper mold UM and the lower mold BM is adjusted so that the adhesive ADH1 bonding the semiconductor chip CHP1 and the chip mounting portion TAB1 is compressed. By doing so, a compressive load can be applied to the end of the semiconductor chip CHP1, thereby preventing resin leakage onto the semiconductor chip CHP1.

  That is, the dimensional variation in the mounting height of the components due to the variation in the thickness of the semiconductor chip CHP1 and the lead frame LF is absorbed by the change in the thickness of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1. Can do. Thus, according to the second embodiment, the clamping force applied to the semiconductor chip CHP1 can be relaxed. As a result, it is possible to prevent breakage such as cracks, chips or cracks of the semiconductor chip CHP1.

  Further, in the second embodiment, without using the elastic film LAF, only the dimensional change of the adhesive ADH1 that bonds the semiconductor chip CHP1 and the chip mounting portion TAB1 causes a variation in the mounting height of the component. Although it is assumed that it absorbs, for example, as in Modification 1, the elastic film LAF is used, and the dimensional variation of the mounting height of the component is also absorbed by the dimensional change caused by the elastic film LAF. You may comprise as follows. That is, when the lead frame LF on which the semiconductor chip CHP1 is mounted is sandwiched between the upper mold UM and the lower mold BM, the adhesive material that bonds the dimensional change of the elastic film LAF and the semiconductor chip CHP1 to the chip mounting portion TAB1. By using the dimensional change of ADH1 together, it is possible to cope with a dimensional variation of a larger mounting height.

  Note that the adhesive ADH1 needs to have a lower elastic modulus than that of the semiconductor chip CHP1 in order to absorb the dimensional variation in the mounting height of the component due to the dimensional change in the thickness direction of the adhesive ADH1. Further, when the thickness of the adhesive ADH1 is small, the amount of dimensional change in the thickness direction is small, and the dimensional variation in the mounting height cannot be sufficiently absorbed. It is necessary to make the dimensions appropriate for the dimensional variation.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  For example, the opening OPT1 of the chip mounting portion TAB1 and the groove DPLT of the plate-like structure PLT installed in the region immediately below the diaphragm DF formed in the semiconductor chip CHP1 need not be installed. In addition, components such as a capacitor, thermistor, control circuit, memory, transistor, resistor, and heater can be mounted on the lead frame LF.

The flow sensor described in the above-described embodiment is a device that measures the flow rate of gas, but specific types of gas are not limited, and air, LP gas, carbon dioxide (CO ₂ gas) It can be widely applied to devices for measuring the flow rate of any gas such as chlorofluorocarbon.

  In the above-described embodiment, the flow sensor for measuring the flow rate of gas has been described. However, the technical idea of the present invention is not limited to this, and a part of a semiconductor element such as a humidity sensor is exposed. The present invention can be widely applied to semiconductor devices that are resin-sealed in such a state.

1 CPU
2 Input circuit 3 Output circuit 4 Memory A node ADH Adhesive material ADH1 Adhesive material ADH2 Adhesive material B Node BM Lower mold BR1 Downstream RTD BR2 Downstream RTD C node CHP1 Semiconductor chip CHP2 Semiconductor chip CU Controller D node DF Diaphragm DM Dam bar DPLT Groove EJPN Ejection pin FDU Flow rate detection part FSP Flow rate sensor FS1 Flow rate sensor FS2 Flow rate sensor H Mounting height HCB Heater control bridge HCT thickness HCT2 thickness HDF maximum value HR Heating resistor IP1 Insert film LAF LD1 lead LD2 lead LF lead frame MR resin OPA opening OP1 opening OP2 opening OP3 opening PD1 pad PD2 pad PD3 pad PLT plate-like structure PS power supply Q gas flow rate DESCRIPTION OF SYMBOLS 1 Resistor R2 Resistor R3 Resistor R4 Resistor SP1 1st space SUR (CHP) Upper surface SUR (MR) Upper surface TAB1 Chip mounting part TAB2 Chip mounting part Tr Transistor TSB Temperature sensor bridge UM Upper mold UR1 Upstream resistance temperature sensor UR2 upstream resistance temperature detector Vref1 reference voltage Vref2 reference voltage W wire WL1 wiring W1 wire W2 wire W3 wire

Claims (14)

A first chip mounting portion;
A first semiconductor chip in which a detection unit is formed and bonded to the first chip mounting unit via a first adhesive;
A second chip mounting portion;
A second semiconductor chip bonded on the second chip mounting portion via a second adhesive;
With
In a state where the detection unit formed on the first semiconductor chip is exposed, a part of the first semiconductor chip is fixed with resin,
The thickness of the first adhesive material sandwiched between the first semiconductor chip and the first chip mounting portion is the thickness of the first adhesive material sandwiched between the second semiconductor chip and the second chip mounting portion. 2 A semiconductor device smaller than the thickness of the adhesive. The semiconductor device according to claim 1,
A thin portion is formed in the first semiconductor chip,
A semiconductor device in which the detection part is formed in the thin part. The semiconductor device according to claim 2,
The semiconductor device, wherein the first adhesive is bonded to a part of the thin portion. The semiconductor device according to claim 3,
The thickness of the first adhesive material sandwiched between the first semiconductor chip and the first chip mounting portion is the maximum thickness of the first adhesive material bonded to a part of the thin portion. A semiconductor device smaller than the value. The semiconductor device according to claim 1,
The semiconductor device, wherein an elastic modulus of the first adhesive is smaller than an elastic modulus of the first semiconductor chip. The semiconductor device according to claim 5,
The first adhesive material includes a thermosetting resin or a thermoplastic resin,
The semiconductor device, wherein the first semiconductor chip includes silicon. The semiconductor device according to claim 1,
The second semiconductor chip includes a control unit that controls the detection unit. The semiconductor device according to claim 1,
A plate-like structure is inserted between the first chip mounting portion and the first semiconductor chip,
The first semiconductor chip and the plate-like structure are bonded via the first adhesive material,
The semiconductor device, wherein the first chip mounting portion and the plate-like structure are bonded via a third adhesive. The semiconductor device according to claim 8,
The semiconductor device, wherein an elastic modulus of the first adhesive is smaller than an elastic modulus of the first semiconductor chip. The semiconductor device according to claim 2,
A semiconductor device in which at least one through hole is formed in the first adhesive formed in a region overlapping with the thin portion in plan view. A method of manufacturing a semiconductor device in which a first semiconductor chip disposed on a first chip mounting portion and a second semiconductor chip disposed on a second chip mounting portion are fixed with a resin,
In a state where a first adhesive having a smaller elastic modulus than the first semiconductor chip is disposed between the first chip mounting portion and the first semiconductor chip, the first chip mounting portion and the first chip are formed by a mold. A method of manufacturing a semiconductor device, wherein the second semiconductor chip and a part of the first semiconductor chip are sealed with the resin by pressurizing the semiconductor chip and pouring the resin. A method for manufacturing a semiconductor device according to claim 11, comprising:
A second adhesive is disposed between the second chip mounting portion and the second semiconductor chip,
The thickness of the first adhesive material between the first chip mounting portion and the first semiconductor chip is greater than the thickness of the second adhesive material between the second chip mounting portion and the second semiconductor chip. A method for manufacturing a semiconductor device, wherein the pressurization is performed so as to be small. A first chip mounting portion;
A first semiconductor chip in which a detection unit is formed and bonded to the first chip mounting unit via a first adhesive;
A second chip mounting portion;
A second semiconductor chip bonded on the second chip mounting portion via a second adhesive;
With
In a state where the detection unit formed on the first semiconductor chip is exposed, a part of the first semiconductor chip is fixed with resin,
The thickness of the first adhesive material sandwiched between the first semiconductor chip and the first chip mounting portion is the thickness of the first adhesive material sandwiched between the second semiconductor chip and the second chip mounting portion. 2 Flow sensor smaller than the thickness of the adhesive. A first chip mounting portion;
A first semiconductor chip in which a detection unit is formed and bonded to the first chip mounting unit via a first adhesive;
A second chip mounting portion;
A second semiconductor chip bonded on the second chip mounting portion via a second adhesive;
With
In a state where the detection unit formed on the first semiconductor chip is exposed, a part of the first semiconductor chip is fixed with resin,
The thickness of the first adhesive material sandwiched between the first semiconductor chip and the first chip mounting portion is the thickness of the first adhesive material sandwiched between the second semiconductor chip and the second chip mounting portion. 2 Humidity sensor smaller than the thickness of the adhesive.

Images