WO1998026266A1

WO1998026266A1 - Judgement method of acceptance/rejection of pressure sensor

Info

Publication number: WO1998026266A1
Application number: PCT/JP1996/003601
Authority: WO
Inventors: Toshihisa Matsui; Tomio Matsui; Masahiro Ueda; Atsuo Irisa
Original assignee: Hokkei Industries L.T.D.
Priority date: 1996-12-09
Filing date: 1996-12-09
Publication date: 1998-06-18
Also published as: WO1998026267A1

Abstract

A method of judging acceptance/rejection of strain gauge type pressure sensors (1) depending on leakage, wherein the pressure sensor (1) measures the pressure by a strain voltage occurring from the difference between the vacuum inside the sensor container and a pressure applied to a pressure introduction portion. When inspection is carried out for judgement, a high D.C. voltage is applied between a terminal (2a) for short-circuiting all the electrodes (2) and a metallic container (6) by a power supply (8) for generating a high D.C. voltage. A resistor (7) having a high resistance value selected to keep discharge current from destroying an IC circuit inside the metallic container (6) is inserted between the positive terminal of the power supply (8) and the terminal (2a). The high D.C. voltage from the power supply (8) is controlled by a computer (13). The voltage is slowly changed in the beginning, and is turned OFF after a discharge of one pulse. The vacuum pressure inside the metallic container (6) is measured from the discharge current Id of one pulse in accordance with the Paschen's law. This measurement is repeated at two-week intervals, for example, and a mean pressure change ratio during this period is determined. If this pressure change ratio exceeds a predetermined value, the product is rejected because of leakage.

Description

Specification

Quality judgment method of pressure sensor

The present invention relates to a method for judging the quality of a vacuum leak of a strain gauge type pressure sensor. Background art

A strain gauge type pressure sensor is indispensable as a vacuum sensor for, for example, increasing the efficiency of an automobile engine. This type of normal pressure sensor detects the pressure difference between the internal high vacuum area and the external pressure introduction pipe by means of a piezoelectric element installed in a vacuum insulated manner between the two. The value is amplified as an electrical signal and extracted. The degree of vacuum inside the container during manufacturing is usually 6.7 Pa (0.05 Torr) or less, and the durability of the vacuum sensor is 20 years or more. That is, it is said that the maximum permissible vacuum leak rate of this kind of pressure sensor that is put into practical use must be less than 1.0 X 10 † —5 † Pa · ccZ seconds. By the way, in such a sensor structure, when the degree of vacuum inside the container is reduced to a certain value, the electric signal output from the piezoelectric element is not proportional to an external pressure change. This vacuum leak is fatal for a pressure sensor. For this reason, various vacuum leak inspections have been conducted. The current general inspection method is to enclose the pressure sensor in a radioactive material autoclave for several days, and if a vacuum leak occurs, the radioactive gas is sucked. It is usually referred to as the “radicro method,” in which measurements are made at power supply or the like to determine the quality or vacuum leak rate.

However, in the quality judgment by the radio method, it is very difficult to handle radioactive materials, and there is a problem of pollution and environmental problems due to radioactive contamination such as extremely poor working environment and poor workability of workers. Various determination methods are known in addition to such a radiography method. However, all of them have problems that the method is complicated, equipment and devices to be used are expensive, and the determination takes time.

The present invention has been made in view of the above-mentioned conventional problems, and has been made in consideration of the conventional determination method using radioactivity. Instead, it is possible to estimate practically a wide range from both discharge and electrical output characteristics for vacuum leaks and electrical circuit defects, and to determine the quality of pressure sensors easily, safely, accurately, non-polluting and economically. The purpose is to provide. Disclosure of the invention

The present invention proposes a new method for measuring the internal pressure of a pressure sensor container as described above, and the present invention is based on, in principle, Paschen's law that the firing voltage depends on the degree of vacuum inside the container. Based on (Paschen's Law), it is possible to instantaneously measure the degree of vacuum inside a sensor container from the discharge starting voltage of the first pulse, and pass / fail without giving any change to the state inside the sensor container It proposes a method that can judge.

The method for judging the quality of a pressure sensor according to the present invention is based on a strain gauge type pressure sensor that measures pressure by a strain voltage generated from a difference between a degree of vacuum inside a sensor and a pressure applied to a pressure introducing unit. This is a method for determining whether or not leakage has occurred. In the present invention, the positive electrode side of a power supply that generates a DC high voltage is connected to a terminal having all electrodes short-circuited through a resistor having a high resistance value, the sensor-one container is grounded, and the terminal and the cell are connected to each other. A high DC voltage is applied between the sensor container and the sensor, the vacuum pressure inside the sensor container is measured from a discharge current generated by the degree of vacuum in the sensor container, and the measurement is repeated after a predetermined period. If the average rate of change in pressure during the period exceeds a predetermined value, it is determined that a vacuum leak has occurred and is determined to be defective. The application of the DC high voltage is controlled such that the discharge is performed for one pulse. Therefore, at the start of voltage application, the applied voltage is slowly varied, and the voltage application is turned off by the discharge of one pulse. In addition, a resistor having a high resistance value is interposed between the positive electrode side of the power supply and the terminal in which all the electrodes are short-circuited. The resistance value of this resistor is selected so that the discharge current does not destroy the circuit in the sensor-one container. The principle adopted by the present invention for measuring the vacuum pressure of such a pressure sensor is Paschen's law as described above. As is well known, this law defines the minimum voltage required to cause a spark discharge as a function of the product of the distance between the electrodes and the pressure of the gas when the electric field is uniform and the temperature and humidity of the gas are constant. Is to be done. Use this rule Then, the vacuum leak rate is measured from the relationship between the discharge phenomenon and the pressure. That is, since the vacuum pressure can be estimated from the discharge starting voltage or discharge current of one discharge pulse caused by the difference in vacuum pressure, the internal pressure of the sensor container at a certain time is measured, and the internal pressure after a certain period of time is measured. If the pressure is measured, the average rate of change in pressure during this period can be determined, and the leak rate during that period can be determined in relation to the internal volume of the sensor and the container using the same data.

Therefore, the pass / fail judgment method of the pressure sensor according to the present invention requires a certain period of time until the judgment can be made, but the degree of vacuum can be instantaneously measured from the discharge starting voltage of one discharge pulse. It is a completely new method that can measure without changing the internal state. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a plan sectional view (A) and a partial side sectional view (B) showing the structure of a pressure sensor 1 to be measured.

FIG. 2 is a conceptual diagram showing a measurement system according to the present invention of the pressure sensor of FIG.

FIG. 3 is a cross-sectional view conceptually showing a configuration of an experimental apparatus for verifying the present invention. FIG. 4 is a graph showing a result of discharging by applying a voltage in the experimental apparatus of FIG. FIG. 5 is a graph showing an enlarged part of FIG.

FIG. 6 is a waveform diagram showing a discharge pulse waveform due to voltage application in the experimental apparatus of FIG.

FIG. 7 is a graph showing the relationship between the number of days and the pressure change when a leak occurs at the allowable maximum leak rate. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

First, a structural example of a strain gauge type pressure sensor to be determined in the present invention will be described with reference to FIG. The pressure sensor 11 includes a base 1a, a plurality of electrodes 2 and a pressure introduction pipe 3 provided electrically insulated from each other, a metal diaphragm 4 attached to a tip of the pressure introduction pipe 3, and a diaphragm 4 A piezoelectric element and an IC circuit (hereinafter simply IC) It is called a circuit. 5), wiring between the IC circuit 5 and the electrode 2, and the upper part of them is covered with the metal container 6.

The principle adopted by the present invention for measuring the vacuum pressure of the pressure sensor 1 is to measure the vacuum leak rate from the relationship between the discharge phenomenon and the pressure according to Paschen's law as described above. That is, since the vacuum pressure P 1 can be estimated from the discharge starting voltage or discharge current caused by the difference in vacuum pressure, the internal pressure P 1 at a certain time T 1 is measured, and after a certain period T (seconds), By measuring the pressure P 2 at this point, the average rate of pressure change R during this period is

R = (P 2 -P 1) / T (P a 'c cZ seconds)

Is required. In addition, the leak rate S at that time is expressed assuming that the internal volume of the sensor is V

S = (P 2 -P 1) v / T (P a · c cZ seconds)

Is required.

Therefore, as shown in FIG. 2, a terminal 2a, in which all the electrodes 2 of the pressure sensor 11 of FIG. 1 are short-circuited, is connected to a DC high-voltage power supply 8 via an external resistor 7, The metal container 6 is grounded via an external resistor 9 and an ammeter (A meter) 10, and a storage oscilloscope 11 is connected so that the voltage at both ends of the external resistor 9 can be observed and the discharge starting voltage can be recorded. The voltmeter 12 is connected so as to indicate the applied voltage V0 from the power supply 8. The power supply 8 sets the maximum applied voltage so as to generate an actual discharge voltage Vs (usually several hundred to several thousand) between the electrodes 2 and 2 of the pressure sensor 1. Further, a computer 13 is connected to the power supply 8 as means for varying the output voltage. Further, an electromagnetic relay 14 that is turned off by one discharge current is connected to the negative electrode side of the power supply 8, and this is grounded.

As described above, the present invention makes it possible to estimate the internal pressure of one sensor and one container from one discharge pulse without affecting the inside of the pressure sensor as much as possible. However, the discharge voltage and discharge current are detected as early as possible so that the discharge can be stopped instantaneously.

As a first measure for this purpose, a high resistance value of about 50ΜΩ is used as the external resistance 7. In this case, for example, if a discharge current of Id = 10 flows, VR = 500 V at both ends of the external resistor 7 (from the relation of VR = RI d, 50 X 10 0 6 X 10 0 X 1 0 † -6 † = 500V), V s = V 0 Due to the relationship of one VR, the discharge voltage between the terminal 2a and the metal container 6 drops by 500 V, and the discharge stops. When the discharge stops, the discharge current Id naturally stops flowing, so that the voltage V 0 applied by the power supply 8 is applied again between the terminal 2 a and the metal container 6, and the next discharge can be started. Needless to say, the discharge starting voltage itself does not depend at all on the size of the external resistor 7, and therefore, in practicing the present invention, it is necessary to actually connect and check resistors having various resistance values.

On the other hand, the repetition of such a pulse discharge greatly differs depending on the applied voltage V 0 from the power supply 8. No touch of the needle of the ammeter 10 is observed at all with only one pulse, but when this pulse becomes quite frequent, the current of the number; A is observed as an average value. However, even if the current is as small as 1 to several; A, if it is observed by the ammeter 10, the experiments conducted by the present inventors show that at least several tens to several hundreds of pulses are obtained at this time. After passing through the ammeter 10 and then checking the operation of the pressure sensor 11, it was found that the IC circuit 5 was completely damaged and lost its function as a sensor. Therefore, when one discharge pulse is detected, it is necessary to surely stop the generation of the next discharge pulse, and for that purpose, the voltage V 0 applied from the power supply 8 must be slowly increased.

Therefore, as a second means, the output voltage of the power supply 8 can be changed in a desired form by the computer 13, and the voltage application from the power supply 8 is turned on and off by the electromagnetic relay 14.

That is, a sharp rise in the applied voltage V 0 from the power supply 8 leads to large current discharge, that is, generation of a plurality of pulses in a short time. In this embodiment, if only one pulse is detected, the next discharge is performed. The circuit is automatically opened by the electromagnetic relay 14 before a pulse occurs. If the electromagnet used for the electromagnetic relay 14 has, for example, a dead time of about 0.5 ms and a rise time of about 1 ms, a delay of 1.5 ms occurs before operation starts. The increase control of the applied voltage V 0 from the power supply 8 is controlled by the computer 13 so as to correspond to this time delay.

In experiments performed by the present inventors, it was confirmed that if the discharge interval is set to, for example, 1 to 2 minutes in one-pulse discharge, the IC circuit 5 does not suffer any damage regardless of the number of discharges. The reproducibility of the experiment was also improved. By the way, it is clear that the electron emission from the electrode that triggers the discharge greatly depends on the electric field, that is, the discharge voltage if the electrode structure is determined. May also occur. Electron emission due to an electric field, ie, field emission, usually occurs at an electric field strength of 10 † 8 † VZm or more. If the shape of the electron emitting surface is like a needle tip having a diameter of about several i ^ m, the above-described strength can be easily reached even at a voltage of several hundred V. In particular, inside the metal container 6 of the pressure sensor 11, the end of the pin-shaped electrode 2 and the IC surrounded by the plurality of electrodes 2 and arranged at the center thereof are several 10 m or so. Since the metal wires are connected by a thin metal wire and the distance between such a metal wire and the inner surface of the metal container 6 is 1 mm or less, there is a high possibility that the field emission as described above will occur. In addition, even if the distance between the metal wire and the inner surface of the metal container 6 has an error of about 10%, the electric field strength may be sufficiently different by an order of magnitude, and the electron emission is an exponential function of the electric field strength. Differently.

Therefore, in order to eliminate the possibility of such field emission, the metal container 6 is grounded, and a positive electrode is connected to the terminal 2 a where the electrode 2 is short-circuited so that electrons are emitted from the metal container 6 to the electrode 2. Is applied. That is, by stabilizing the discharge start state, in other words, stabilizing the discharge start voltage by causing the electron emission to occur on the inner surface of the flat metal container 6.

As a comparative example with the present invention, when the resistance value of the external resistor 7 is about several ΩΩ used for ordinary experiments and observations of this kind, a current of about 100; However, the voltage drop due to the external resistor 7 is at most about several volts. As a result, the actual discharge voltage Vs between the electrodes 2 and 2 of the pressure sensor 1 hardly changes, resulting in a continuous discharge, and the IC circuit 5 inside the pressure sensor 11 is damaged. It does not work properly.

Next, an experimental example of the present invention will be described with reference to FIGS. 3 and 4 showing measurement results of the discharge starting voltage Vd and the discharge current Id when the pressure P1 in the metal container 6 is changed. .

The pressure measurement experimental device is configured as shown in FIG. First, the metal container 6 is removed, and the electrode 2, the pressure introducing pipe 3, the diaphragm 4, and the IC circuit 5 are mounted on the base la. It is attached to an acrylic plate 20 with holes as shown in the figure using an adhesive such as an arral die, covered with a vacuum bell jar 21, and inserted into one through hole of the vacuum bell jar 21. A Pirani vacuum gauge 22 is connected, a rotary pump (not shown) is connected to the other through-hole via a cock 23, and another one of the through-holes is connected to outside air through a cock 24. To allow the internal pressure of the pressure sensor to be adjusted freely. In the figure, reference numeral 25 denotes an insulating silicone agent, and reference numeral 26 denotes a lead wire.

Next, the relationship between the sensor internal pressure P and the firing voltage V 0 was examined using the measurement circuit of FIG. 2 described above. As the ammeter 10, a 10 full-scale ammeter was used. The results of this experiment and the details of the study will be described. First, the relationship between the firing voltage and the internal pressure of the sensor vessel in this experiment is shown in Figs. 4 and 5, and Fig. 6 shows an example of the pulse waveform at that time. FIG. 4 shows a wide range from Equation 1 OPa to Equation 1 00 Pa, and FIG. 5 shows a range from practical 50 Pa to 200 Pa in FIG. It is shown enlarged. The results shown in these figures show that the relationship between the discharge starting voltage and the internal pressure of the sensor vessel changes according to Passen's law, which can be the basis for estimating the internal pressure of the pressure sensor according to the present invention. Is well shown.

According to the results of this experiment, an error of about 10% is recognized even with the same sensor and the same pressure. This is because the speed at which the applied voltage is increased varies, and can be improved by automatically adjusting the voltage increase by control means such as a computer. Of course, the narrower the error range is, the shorter the pressure and the wider the pressure can be estimated.

Observing the waveform in Fig. 6, the peak value of the current in one-pulse discharge is about 2.5 mA (25 V / 10 X 10 13 † Ω), and the average time is about 5 mA. // seconds. This was converted to an average current of 12.5 nA. Naturally, no needle touch was observed even when measured with a 10-meter full-scale ammeter. Conversely, if pulses of the same waveform occur intermittently, for example, a discharge current of 1 mm is observed, then 80 discharge pulses are generated per second, and the average pulse in that case The interval time, that is, the period T was 12.5 milliseconds, and the next pulse discharge was almost certainly stopped by an electromagnetic relay with an operation time of 1.5 milliseconds. By the way, the energy per pulse at this time is 0.313 J, which is about this level. It is unlikely that the energy will damage the IC circuit.

A description will be given of a case where the above-described method of determining the quality of the pressure sensor is actually used for measuring a vacuum leak of the pressure sensor.

As already mentioned, the permissible maximum leak rate Sm is about 1.0X 10 † —5P (P a 'c cZ seconds) due to the life of this type of pressure sensor (about 20 years). However, Fig. 7 shows the relationship between the number of days and pressure change when a leak occurs at this allowable maximum leak rate Sm. From FIG. 7, it can be seen that if the internal volume V of the pressure sensor is, for example, 0.5 cc, there is a pressure rise of about 25 Pa or more in two weeks.

That is, in the actual vacuum leak measurement of the pressure sensor, it is sufficient that this 25 Pa pressure change can be detected by the discharge starting voltage. In other words, if the discharge starting voltage after leaving it for about two weeks does not overlap with an error, it is possible to judge whether or not the pressure has changed so much, that is, the quality of the sensor.

On the other hand, this is also the expression

S = (P 2 -P 1) v / T (P a 'c cZ seconds)

From the relationship between Fig. 5 and Fig. 5, in order to estimate the leak rate, it is desirable to set the pressure measurement range to about 20 Pa to 100 Pa. If the pressure is 20 Pa or less, the internal pressure of one sensor does not reach 50 Pa even after about 2 weeks, and even if a high voltage of about 1500 V is applied, no discharge occurs, and the vacuum leakage rate is reduced. Estimation may not be possible. Conversely, if the pressure is 100 Pa or more, even if the internal pressure of one sensor container becomes 125 Pa after about two weeks, the change in the firing voltage at that time is small and within the error range. However, there is a possibility that the vacuum leak rate cannot be estimated. Note that if the error in the firing voltage at a certain pressure is small, the measurement range of the above-mentioned pressure may be slightly wider. One possibility is that the internal pressure of the pressure sensor, which is sealed in a vacuum, is due to Boyle-Charles law,

P V = nRT

(P: Pressure, V: Sensor internal volume, n: Number of moles, R: Gas constant, T: Absolute temperature) In this case, V, n, and R are constant. Move P to the more convenient one. For example, if it is sealed at room temperature 27 ° C (T = 30 OK), heating to 57 ° C (T = 33 OK) will give a 10% temperature The pressure rises by 10% because of the rise, and if the minimum value of the sealed pressure per sensor is reduced to 1 OPa, and if it is cooled down to 0 ° C (T = 273 Κ), the maximum value of the pressure to be measured will be about 11 OPa It is possible to expand to.

Although the present invention has been described above, various modifications can be made within the scope of the claims, and the circuit components and the like shown in the drawings are merely examples. Thus, various alternatives can be employed.

Claims

The scope of the claims

1. For a strain gauge type pressure sensor that measures pressure by a strain voltage generated from the difference between the degree of vacuum inside the sensor container and the pressure applied to the pressure introducing unit, the quality of the pressure sensor is determined by the presence or absence of vacuum leakage. A method for determining, comprising: connecting the positive electrode side of a power supply that generates a DC high voltage to all the short-circuited terminals, grounding the sensor container, and connecting the terminal to the sensor container. DC high voltage is applied during the period, the vacuum pressure inside the sensor container is measured from the discharge current generated by the degree of vacuum in the sensor container, and the measurement is repeated after a predetermined period, and the average during the period is measured. If the actual pressure change rate exceeds a predetermined value, it is determined that a vacuum leak has occurred and is determined to be defective.The discharge current between the positive electrode side of the power supply and the terminal is the same as that of the sensor. Electricity in the vessel A high-resistance resistor selected so as not to destroy the slave circuit is interposed.When starting the application of the DC high voltage, the applied voltage is slowly varied, and the one-pulse discharge is performed. A method for judging the quality of a pressure sensor, which is characterized in that the voltage application is controlled to be turned off.