JPS63248181A - Integrated sensor - Google Patents

Abstract

PURPOSE:To obtain an accurate detecting value at all times by forming a plurality of sensors and a signal processing circuit onto the same chip, selecting an effective signal and arithmetically operating the detecting value. CONSTITUTION:When detecting values by a plurality of pressure sensors C1-C5 on a chip 1 exceed upper or lower limits by long-term usage, an arithmetic processing circuit 3 decides the sensor to be a defective, and writes the number of the defective sensor to a ROM 5. The circuit makes reference to the content of the ROM 5 and selects an effective signal from the effective sensors C1-C5 and writes the effective signal to a built-in RAM 3a, and reads it and computes a detecting value and outputs the detecting value. Accordingly, the method differs from the time when only one sensor is shaped, and the accurate detecting value can be acquired at all times.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a tJAfa sensor in which a sensor and a signal processing circuit for processing an output signal of the sensor are formed on the same chip.

(Prior Art and Problems) Conventionally, for example, an integrated pressure sensor has one pressure sensor and its signal processing circuit formed on a chip. The output of the signal processing circuit, that is, the detected value of the integrated pressure sensor had also become abnormal.

(Objective of the invention) This invention solves the above problems and always provides accurate detection values.
The object of the present invention is to provide a highly reliable integrated sensor that can be used for various purposes.

(Means for Solving the Problems) This invention was made to achieve the above object, and it forms m, l, and k sensors on the same chip, and inputs the output signals of each of the sensors. The gist of the sensor is an integrated sensor formed with a signal processing circuit that selects a valid signal from among the signals and calculates a detected value based on the selected signal.

(Operation) The signal processing circuit receives output signals from a plurality of sensors formed on the same chip, selects a valid signal from among the signals, and calculates a detected value based on the selected signal. As a result, the detected value is calculated based only on valid signals.

(Example) Hereinafter, an example in which the present invention is embodied in an integrated pressure sensor will be described with reference to the drawings.

As shown in FIG. 1, five identical pressure sensors C1 to C5 and amplifier circuits A1 to A5 corresponding to the pressure sensors C1 to C5 are formed on the same chip 1, respectively.

In this embodiment, the number of pressure sensors is five, but the number may be changed as appropriate.

Further, the multiplexer 2 formed on the same chip 1 connects the pressure sensors C1 to C via the amplifier circuits A1 to A5.
5, and selects one of the output signals of each of the pressure sensors C1 to C5 according to a command signal from the arithmetic processing circuit 3, which is also formed on the same chip 1. The selected signal is output to the A/D converter 4 formed on the same chip 1F.

The arithmetic processing circuit 3 inputs analog signals outputted from each pressure sensor C1 to C5 after being converted into digital signals by an A,.../D converter 4, and determines whether the output signal of the pressure sensor is an abnormal signal or not. At the same time, a signal indicating the detected value is outputted to the outside based on the output signals of the pressure sensors C1 to C5 determined to be normal.

Further, the opening sieve processing circuit 3 includes a readable and replaceable memory (RAM) 3a, and old detection data from the pressure sensor determined to be normal in the above-described determination is stored in the RAM 3a. Further, a non-volatile memory 5 is formed on the same chip 1, and the number of the defective pressure sensor determined to be normal in the above-described determination is written into the memory 5.

Next, the operation of the Song top pressure sensor constructed as described above will be explained based on FIG. 2.

First, the arithmetic processing circuit 3 determines whether the first pressure sensor C1 is defective by referring to the nonvolatile memory 5 (step 1). In addition, in the initial state, all pressure sensors C
1 to C5 are normal, so no numbers of defective sensors are written in the nonvolatile memory 5. As a result, the arithmetic processing circuit 3 determines that the pressure sensor C1 of the first whistle is normal (Step 2), and sends a signal to the multiplexer 2 instructing the pressure sensor C1, and converts the output signal from the pressure sensor C1 into an A/D converter. Capture via converter 4 (step 3). Then, the arithmetic processing circuit 3 determines whether the detected data is normal or not, that is, determines whether the detected data of the input signal is within a predetermined upper and lower limit range (step 4). At this time, since each of the pressure sensors 01 to C5 is normal in the initial state, the captured data is within the limited range.

The arithmetic processing circuit 3 stores the normal data in the memory 3a within the circuit (step 5), and after confirming that all sensors have not been selected (step 6), selects the next (second) pressure. Select sensor C2 (step 7). The processing circuit 3 of the performance repeats such processing until all the sensors C1
After carrying out the processing for ~C5, the average value of the data stored in the memory 3a is calculated (step 8) and output (step 9).

The arithmetic processing circuit 3 outputs data by repeating this routine It';'y.

If any of the pressure sensors C1 to C5 breaks down due to long-term use, etc., step 4 in the above routine
In other words, if it is determined whether the detected data based on the signal from the pressure sensor is within the limited range, the arithmetic processing circuit 3 determines that the pressure sensor is defective. Based on this determination, the arithmetic processing circuit 3 writes the number of the defective pressure sensor into the nonvolatile memory 5 (step 10). In the next routine operation of the pressure sensor stored in the non-volatile memory 5, the arithmetic processing circuit 3 does not take in data based on the output signal in step 1.2, and naturally performs the operation [1]. The data is excluded from being stored in the memory 3a in the logic circuit 3.

Then, the arithmetic processing circuit 3 stores only data from normal pressure sensors excluding the excluded pressure sensors in the memory 3a, and averages the data (step 8) and outputs the same data (step 8). 9)
.

In this way, in this embodiment, for the five same pressure sensors C1 to C5, it is determined whether each pressure sensor C1 to C5 is normal or defective, the output signal of the normal sensor is selected, and the data from the normal sensor is used. Since the values are averaged and output, the output detection values are always accurate and highly reliable.

In the above embodiment, the memory 3a of the II processing circuit 3
Although the data stored in the data is simply averaged and output, processing such as averaging after removing the maximum 1111 and minimum values of the data may be performed.

In addition, although the non-volatile memory 5 only stores the defective sensor number, it also stores the data of the defective sensor, and calculates the temperature characteristics, linearity, and offset for the output signal of each pressure sensor. Characteristics etc. may also be adjusted. That is, for example, a sensor circuit may be provided with several resistor arrays having various resistance values as adjustment resistors, and the connections of the resistor arrays may be changed based on the data in the nonvolatile memory 5, or the data in the nonvolatile memory 5 may be processed by arithmetic processing. However, the array may be adjusted. Furthermore, the difference in data from the pressure sensor between a certain point in time is read out from the non-volatile memory 5, the change over time is recorded, and the compensation is performed in the arithmetic processing circuit 3, and the connection of the resistor array is changed in the same manner as above. Alternatively, the change over time may be compensated for by directly performing the performance in the @Chuu processing circuit 3.

Furthermore, the present invention is not limited to the above embodiments, but may be implemented as follows.

(B) In the above embodiment, the same pressure sensor (pressure sensor with the same diaphragm size) was used, but pressure sensors with different diaphragm sizes were formed on the same chip, and the data output from each sensor was It may also be processed.

That is, when pressure sensors having diaphragms of one type of size are formed on the same chip, they may resonate at a certain frequency and a frequency that is an integral multiple thereof, and may not be able to output a normal signal. However, if the size of the diaphragm is formed so as not to be an integer ratio of that particular frequency,
Even if one of the plurality of pressure sensors causes resonance and ceases to output a normal signal, it is possible to selectively process only the signals from the other pressure sensors to obtain a normal output. Furthermore, if the relationship between the size of the diaphragm and the resonance frequency is investigated, it becomes possible to use it as a vibration frequency sensor.

(b) As in the embodiment of (a) above, pressure sensors with different diaphragms may be formed on the same chip, and signals may be selectively enabled by switching within a range that does not deteriorate the linearity of the pressure sensors.

That is, as shown in FIG. 3, for example, when the pressure is in the range of O to P1, the signal SGI from the first pressure sensor is selected (
When the pressure is in the range of P1 to P2, the signal SG2 from the second pressure sensor is selected (validated). This is because the sensitivity to pressure can be changed by changing the size of the diaphragm of the pressure sensor.Increasing the size of the diaphragm increases the pressure sensitivity, but the linearity deteriorates; This takes advantage of the fact that the pressure sensitivity decreases, but the linearity improves.

Therefore, by sequentially selecting the output signal of each pressure sensor for each pressure range in this way, accurate detection values can always be obtained with output signals having good linearity. At this time, the pressure sensor that is not being used can be moved as a backup for the pressure sensor that is in use, and can be switched to use if the signal of the pressure sensor that is in use becomes abnormal. You can do it like this.

Furthermore, at the time of switching, if the output of the pressure sensor to be switched next is corrected based on the output value of the pressure sensor before switching, a more accurate value can be obtained.

In other words, in FIG. 3, when switching from the signal SG1 from the first pressure sensor to the signal SG2 from the second pressure sensor, the output value from the second pressure sensor is operated to match the output value from the first pressure sensor. The numbness processing circuit 3 performs correction. In this way, the first pressure sensor has higher sensitivity and more accurate data than the second pressure sensor, so if the output of the second pressure sensor is corrected to match the output of the first pressure sensor, the second pressure sensor The output of the pressure sensor can be made more accurate.

Furthermore, the present invention is not limited to the integrated pressure sensor as in each of the above embodiments, but may be embodied in other integrated sensors, such as various types of sensors such as an integrated temperature sensor.

Effects of the Invention As detailed above, the integrated sensor of the present invention exhibits an excellent effect in that it can always obtain accurate detection values and is highly reliable.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing an integrated pressure sensor embodying the present invention, Fig. m2 is a flow chart diagram for explaining its operation, and Fig. 3 is a diagram for explaining cutting down. In the figure, 1 is a depth, 3 is a performance processing circuit, and C1 to C5 are pressure sensors.

Claims (1)

[Claims] 1. Forming a plurality of sensors on the same chip,
An integrated sensor comprising a signal processing circuit that inputs the output signals of each sensor, selects a valid signal from among the signals, and calculates a detected value based on the selected signal. 2. The plurality of sensors are each the same pressure sensor,
The signal processing circuit selects the output signals of each pressure sensor within a predetermined range, calculates the average value of the selected output signals, and uses it as a detected value. The integrated sensor described. 3. The integrated circuit according to claim 1, wherein the plurality of sensors are pressure sensors having different diaphragm sizes, and the signal processing circuit sequentially selects the output signal of each pressure sensor for each pressure range. sensor. 4. The integrated sensor according to claim 1, wherein the plurality of sensors are pressure sensors having diaphragms with different resonance frequencies.