Leakage resistance detection device for an on-board high-voltage device and leakage resistance detection method therefor

The leakage resistance detection device for on-board high-voltage devices uses a transition time-based method to quickly and accurately detect ground abnormalities, addressing inaccuracies and delays in existing technologies by varying the threshold setting and using a fixed cycle pulse signal.

DE102012220456B4Active Publication Date: 2024-12-05MITSUBISHI ELECTRIC CORP
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
DE102012220456
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2012-03-16
Filing Date
2012-11-09
Publication Date
2024-12-05
Estimated Expiration
2032-11-09

AI Technical Summary

Technical Problem

Existing leakage resistance detection methods for on-board high-voltage devices in vehicles suffer from inaccuracies and delays in detecting ground abnormalities due to high-frequency pulse signals, leading to incorrect determinations and prolonged response times.

Method used

A leakage resistance detection device that includes a coupling capacitor, repetitive signal output circuit, and arithmetic control circuit, which measures leakage resistance based on transition time during charge and discharge periods, allowing for quick detection of ground abnormalities by varying the limit leakage resistance setting and using a fixed cycle pulse signal.

Benefits of technology

Enables rapid and accurate detection of leakage resistance without waiting for the completion of a half cycle, reducing response time and improving detection accuracy by setting an abnormality threshold variably, thus enhancing the reliability of ground abnormality detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Leakage resistance detection device (50A, 50B, 50C, 50D, 50E) for an on-board high-voltage device, which is connectable to an on-board high-voltage device (60X, 60Y) comprising an on-board high-voltage DC source (61) and an electrical high-voltage load (64) which is to be supplied and driven by the on-board high-voltage DC source (61), wherein the on-board high-voltage device (60X, 60Y) has a leakage resistance Rx with respect to a vehicle body (11) as typified by an equivalent leakage resistance (65) on a positive potential side and an equivalent leakage resistance (66) on a negative potential side, wherein the leakage resistance detection device (50A, 50B, 50C, 50D, 50E) comprises a constant voltage control current source (25) which is supplied and driven by a low-voltage DC power source (10) connected to the vehicle body (11) via a negative terminal and which generates a control current supply voltage Vcc, wherein the leakage resistance detection device (50A, 50B, 50C, 50D, 50E) is configured to measure a value of the leakage resistance Rx via a coupling capacitor (51) one terminal B of which is connected to a predetermined position of the on-board high-voltage device (60X, 60Y), wherein the leakage resistance detection device (50A, 50B, 50C, 50D, 50E) further comprises: a repetitive signal output circuit (30A, 30B, 30C, 30D, 30E), a monitoring signal processing circuit (40A, 40B, 40C, 40D, 40E), and an arithmetic control circuit (20A, 20B, 20C, 20D, 20E), wherein the arithmetic control circuit (20A, 20B, 20C, 20D, 20E) comprises a microprocessor (21) and a program memory (24A, 24B, 24C, 24D, 24E) cooperating, wherein the repetitive signal output circuit (30A, 30B, 30C, 30D, 30E) alternately repeats a charging period and a discharging period in response to a switching operation of a charge / discharge switching element (31 or 39) operating in response to a repetitive command signal PLS, during which a measuring point A is intermittently connected to the control power supply voltage Vcc via a charge / discharge resistor (33 or 35), wherein the measuring point A is another terminal of the coupling capacitor (51) in order to gradually increase or gradually decrease a monitoring voltage Vx, which is a potential between the measuring point A and the vehicle body (11), where the monitoring signal processing circuit (40A, 40B, 40C, 40D, 40E) is configured to: to generate at least one of the inverted logic signals ANS, LOW and HIG when a value of the monitoring voltage Vx, which gradually increases or decreases with a gradient that becomes smoother as the charge / discharge time constant increases, passes a predetermined threshold voltage and to input at least one of the inverted logic signals ANS, LOW and HIG to the arithmetic control circuit (20A, 20B, 20C, 20D, 20E); or to generate an analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20A, 20B, 20C, 20D, 20E); where the arithmetic control circuit (20A, 20B, 20C, 20D, 20E) is configured to: to measure a transition time Tx, during which the value of the monitoring voltage Vx changes from one threshold voltage equal to or higher than 0 volts or from one threshold voltage equal to or lower than the control supply voltage Vcc to the other; to calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance (65) on the positive potential side and the equivalent leakage resistance (66) on the negative potential side, based on a functional formula or a data table relating to the transition time Tx and the leakage resistance Rx; and to generate a resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than a specified limit leakage resistance Rx0, or when the transition time Tx reaches a limit transition time Tx0 corresponding to the specified limit leakage resistance Rx0, and wherein the repetitive command signal PLS comprises a pulse signal with a fixed cycle having a charge period or discharge period longer than at least the limit transition time Tx0, or a pulse signal with a variable cycle in which an output pulse is inverted when the transition time Tx is reached.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention

[0001] The present invention relates to a leakage resistance detection device for an on-board high-voltage device, which is connectable to an on-board high-voltage device including an on-board high-voltage DC power source and a high-voltage electric load to be supplied and driven by the high-voltage DC power source, and which measures a leakage resistance with respect to a vehicle body as typified by equivalent leakage resistances on the positive and negative potential sides of the on-board high-voltage device, so as to notify an abnormality when the leakage resistance decreases, and also relates to a leakage resistance detection method for the on-board high-voltage device. 2. Description of related prior art

[0002] Generally, an electric vehicle, a hybrid electric vehicle, or other such vehicles use a combination of a 12V DC low-voltage battery and a high-voltage battery, such as a 400V DC battery pack. A negative terminal of the low-voltage battery is connected to a vehicle body, but the high-voltage battery and a high-voltage electrical load to be supplied and driven by the high-voltage battery are mounted in the vehicle body while being substantially completely insulated from the vehicle body. In this type of vehicle, a leakage resistance detection device supplied and driven by the low-voltage battery measures an insulation resistance (which has the same meaning as leakage resistance) of the entire high-voltage device to thereby detect the presence or absence of a grounding abnormality.

[0003] For example, a ground detection device for an electric vehicle described in Japanese Patent Laid-Open No. 2002-209331 is a ground detection device for an electric car that includes a high-voltage DC power source electrically isolated from the vehicle body and a three-phase AC motor driven by a DC voltage from the high-voltage DC power source. The ground detection device includes a microcomputer that supplies a ground detection signal representing a square wave to the high-voltage DC power source via a detection resistor and a coupling capacitor. The microcomputer detects a voltage amplitude value at the ground detection point corresponding to a connection point between the detection resistor and the coupling capacitor.Based on a preset relationship between a voltage amplitude value and an insulation resistance value, the microcomputer converts the detected voltage amplitude value into an insulation resistance value and compares the converted insulation resistance value with a predetermined grounding determination threshold to thereby detect the degree of degradation in the insulation resistance of the high-voltage DC power source. Thus, the circuit configuration can be simplified and the degree of reduction in the insulation resistance relative to the vehicle body can be detected with high accuracy.

[0004] First, the ground detection device of Japanese Patent Laid-Open No. 2002-209331 will be described in detail with reference to the Fig. 28 and Fig. 29. The name of each element is replaced by the name in the present invention, which will be described later. Referring to Fig. 28, which is an overall configuration diagram of the conventional device, a leakage resistance detection device 50 includes an arithmetic control circuit 20 mainly constituted by a microprocessor, and a repetitive signal output circuit 30 and a monitor signal processing circuit 40.

[0005] An on-board high-voltage device 60 includes a high-voltage electrical load (not shown) including a high-voltage direct-current power source 61. The on-board high-voltage device 60 is mounted in a vehicle body 11 while being insulated from it, but has equivalent leakage resistors R1 and R2. A coupling capacitor 51 has one terminal B connected to, for example, a negative terminal of the on-board high-voltage device 60, and another terminal B connected to an output terminal of the repetitive signal output circuit 30.

[0006] The arithmetic control circuit 20 generates a repetitive command signal PLS, which is a pulse train signal in which an "H" period T1 and an "L" period T2 satisfy the expression T1=T2 (=half cycle T). The repetitive signal output circuit 30 includes a charge / discharge switching element 39 and a series resistor R0. The charge / discharge switching element 39 is formed of a pair of upper and lower transistors, one of which becomes conductive in response to the repetitive command signal PLS. When the output logic level of the repetitive command signal PLS is "H," a charging current from a predetermined control power supply voltage Vcc is supplied to the coupling capacitor 51 via the upper transistor (not shown) and the series resistor R0. At this time, the lower transistor is open.

[0007] On the other hand, when the output logic level of the repetitive command signal PLS is "L," a discharge current of the coupling resistor 51 flows through the lower transistor (not shown) and the series resistor RO. At this time, the upper transistor is open. The monitor signal processing circuit 40 inputs the value of a monitor voltage Vx, which is the potential between the other terminal A of the coupling capacitor 51 and the vehicle body 11, where the other terminal A is a measurement point, to the arithmetic control circuit 20 via a noise filter (not shown) and an operational amplifier 49 as an analog signal voltage ANL.

[0008] In the first period T1=T, in which the output logic level of the repetitive command signal PLS is "H," the value of the analog signal voltage ANL gradually increases from an initial voltage Vx1 to a final voltage Vx2. In the second period T2=T, in which the output logic level of the repetitive command signal PLS is "L," the value of the analog signal voltage ANL gradually decreases from the initial voltage Vx2 to the final voltage Vx1. Note that in the case where the leakage resistance fluctuates due to voltage fluctuations of the high-voltage DC power source 61 or due to ON / OFF of a power supply switch for load driving, the voltage of the coupling capacitor 51 at the measurement point A becomes a value equal to or less than 0 volts, or equal to or greater than the power supply voltage Vcc relative to the vehicle body potential, thereby deviating from a correct voltage range.

[0009] When a grounding abnormality has occurred in the on-board high-voltage device 60, bypass diodes 815 and 816 and bypass diodes 817 and 818 feed the electric charges stored in the coupling capacitor 51 back to the vehicle body 11 via an output terminal and a negative terminal of a constant-voltage power source (not shown) that generates the control power supply voltage Vcc, regardless of the operating state of the charge / discharge switching element 39. Thus, the value of the monitor voltage Vx can be quickly returned to the appropriate voltage range from 0 to Vcc. However, if a leakage resistance Rx is too small, the bypass diodes 817 and 818 in the upstream stage may be disconnected by an overcurrent. In this case, a limited current is fed back via the series resistor RO and the bypass diodes 815 and 816.

[0010] In the state where the value of the equivalent leakage resistance R1 on the positive potential side and the value of the leakage resistance R2 on the negative potential side are equal to each other, and where no charge / discharge current flows through the coupling capacitor 51, the potential at the connection point B relative to the vehicle body is -Vh / 2 compared to a voltage Vh of the high-voltage DC power source 61. When the equivalent leakage resistance R1 is short-circuited, the potential at the connection point B relative to the vehicle body becomes -Vh. When the equivalent leakage resistance R2 is short-circuited, the potential at the connection point B relative to the vehicle body becomes 0. Such potential fluctuations cause a charge / discharge current to flow through the coupling capacitor 51, with the result that the potential at the measurement point A fluctuates significantly and transiently deviates outside the proper range of 0 to Vcc.

[0011] In Fig. 29 is a characteristic diagram of the as in Fig. 28 illustrates a conventional device configured as a leakage resistance coefficient β on the horizontal axis, a relationship between the value of a parallel combined resistance of the equivalent leakage resistances R1 and R2, namely, the value of the leakage resistance Rx=R1xR2 / (R1+R2), and the series resistance R0. The series resistance R0 is a known constant, and therefore, the value of the leakage resistance coefficient β=Rx / R0 is proportional to the leakage resistance Rx. A threshold voltage coefficient γ on the vertical axis is a relationship between a deviation voltage Vx2-Vx1 between the final voltage Vx2 and the initial voltage Vx1 described above and the control power supply voltage Vcc. The control power supply voltage Vcc is a known constant, and therefore, the value of the threshold voltage coefficient γ=(Vx2-Vx1) / Vcc is proportional to the measurement resistance of the deviation voltage Vx2-Vx1.

[0012] A number of characteristic curves of Fig. Figure 29 uses a reference time coefficient α0 as a parameter. The reference time coefficient α0 is a ratio between the half-cycle T of the repetitive command signal PLS and a charge / discharge time constant τ=(R0+Rx)xC=R0xC with respect to the coupling capacitor 51 when the value of the leakage resistance is Rx0. Both the half-cycle T and the charge / discharge time constant τ are known constants, and therefore the reference time coefficient α0=T / (R0xC) is a known constant.

[0013] For example, in a case where the series resistance R0 is selectively designed to have the same value as a limit leakage resistance Rx0, which is an allowable lower limit of the leakage resistance Rx, the value of the threshold voltage coefficient γ is of interest when the leakage resistance coefficient β is equal to 1. If the reference time coefficient α0 is selectively designed to be 0.5, the threshold voltage coefficient γ at a point of interest P1 is equal to 0.56.

[0014] Therefore, in the case where the control power supply voltage Vcc is, for example, 5 V, if the deviation voltage Vx2-Vx1 exceeds 5x0.56=2.8 V at the end of the "H" period of the repetitive command signal PLS, the leakage resistance Rx is in a safe range that exceeds the limit leakage resistance Rx0, and if the deviation voltage Vx2-Vx1 falls below 2.8 V, the leakage resistance Rx falls below the limit leakage resistance Rx0 and is in a dangerous range. Note that the characteristic diagram of Fig. 29 on the above in Fig. 29 and Expression III is satisfied in a stable condition in which the value of the monitoring voltage Vx falls within the correct range from 0 to Vcc.

[0015] The leak detection device of Japanese Patent Laid-Open No. 2002-209331 calculates, based on the characteristic diagram of Fig. 29, the deviation voltage Vx2-Vx1 between the initial voltage Vx1 and the final voltage Vx2 at a time at the end of the "H" period (or the end of the "L" period) of the repetitive command signal PLS and reads out the value of the leakage resistance coefficient β corresponding to the deviation voltage, thereby detecting the current value of the leakage resistance Rx.

[0016] The first problem with this method is that the leakage resistance Rx is calculated at each end of the generated pulse of the repetitive command signal PLS (at each logical change), and therefore, even if the deviation voltage is sufficiently high and the leakage resistance is normal at the time before the end of the generated pulse, a grounding abnormality determination cannot be made during the pulse generation period. The second problem with this method is that, as can be seen from Fig. 29, the characteristic curves for the reference time coefficient a0 exceeding 1.54 are concave in the center and therefore even the same threshold voltage coefficient γ has two solutions for the leakage resistance coefficient β and a correct solution cannot be obtained.

[0017] The reason is that the expression indicates that the value of the threshold voltage coefficient γ converges to 1 when the leakage resistance coefficient β is infinite, and the value of the threshold voltage coefficient γ converges to 1 along with the increase in the reference time coefficient α0 when the leakage resistance coefficient β is 0, and therefore the threshold voltage coefficient γ takes a value in the range of 0 to 1. Note that when the leakage resistance coefficient β is 0, the value of the threshold voltage coefficient γ decreases along with the decrease in the reference time coefficient α0, and γ converges to 0 when α0 reaches 0.Therefore, to avoid the above-mentioned binary problem, it is necessary that the reference time coefficient α0 to be used takes a value of 1.0 or less, and it is necessary that the half-cycle T of the pulse be a high-frequency pulse whose reference time constant τ0 is equal to or less than the product of the series resistance R0 and the electrostatic capacitance C of the coupling capacitor 51 (= R0xC). If the repetitive command signal PLS is not a high-frequency pulse, there is another problem in that the detection accuracy of the monitor voltage deteriorates because an abrupt change occurs between the monitor voltage at the end of the previous cycle and the monitor voltage at the end of the current cycle.

[0018] Therefore, in the case where the repetitive command signal PLS is a high-frequency signal, even after the monitor voltage Vx deviates from the correct range of 0 to Vcc due to an abrupt change in the leakage resistance Rx and then returns to the correct range through the bypass diodes 815 to 818, it is necessary to wait for a large number of operations of the repetitive command signal PLS until a stable initial voltage Vx1 and a stable final voltage Vx2 are obtained. Thus, the grounding abnormality determination cannot be made immediately. This is the second problem mentioned above.

[0019] This problem is caused because the coupling capacitor 51 is alternately charged and discharged by the repetitive command signal PLS, and therefore the next charge starts before the previous discharge is completed, and the monitor voltage Vx does not monotonically increase or decrease. The repetitive command signal PLS exhibits a smaller change per cycle as the frequency increases and the cycle becomes shorter. Therefore, the detection accuracy improves by increasing the frequency, but the response deteriorates.

[0020] The third problem of the above-mentioned method is related to the above-mentioned second problem and is that, in a transient delay period from when the monitor voltage Vx returns to the correct range after an abrupt change in the leakage resistance Rx until when a plurality of repetitive command signals PLS are operated to enable the measurement of the leakage resistance Rx, the measured value of the deviation voltages Vx1 and Vx2 is smaller than that in the steady state used as the determination condition. Therefore, the obtained leakage resistance coefficient β is a small value, and it may be erroneously determined that a grounding abnormality has occurred. Thus, a standby time for avoiding such an erroneous determination becomes longer.

[0021] US 2007 / 0 285 057 A1 describes a circuit that detects a leakage of mainly a charging / discharging circuit of a rechargeable battery, and a leakage detection circuit for an electric vehicle that detects a leakage of a power supply device that drives a motor for driving the electric vehicle.

[0022] US 5 382 946 A describes creepage path detection circuits and a method and apparatus for determining the resistance of a creepage path in an electric vehicle having an isolated high-voltage traction battery and an auxiliary battery grounded to the vehicle.

[0023] US 2010 / 0 237 872 A1 describes an apparatus and a method for detecting a leakage current of a battery and an apparatus and a method for detecting a leakage current of a battery used in a high-voltage battery power supply system such as electric vehicles and hybrid vehicles. SUMMARY OF THE INVENTION

[0024] It is a first object of the present invention to provide a leakage resistance detection device capable of variably and easily setting a value of a limit leakage resistance as a setting constant with respect to an arithmetic control circuit, and capable of solving the above-mentioned first problem of measuring a leakage resistance without waiting for a logic inversion end of a repeated command signal, thereby quickly detecting the occurrence of a grounding abnormality.

[0025] Furthermore, second and third objects of the present invention are to provide a leakage resistance detection device capable of solving the above-mentioned second and third problems to perform accurate measurement of a

[0026] Leakage resistance at a time of the 1.5th cycle of a repetitive command signal at the latest after a monitoring voltage deviates from a correct range due to an abrupt change in leakage resistance and then returns to the correct range.

[0027] According to an exemplary embodiment of the present invention, a leakage resistance detection device for an on-board high-voltage device is provided, which is connectable to an on-board high-voltage device including an on-board high-voltage DC power source and a high-voltage electrical load to be supplied and driven by the on-board high-voltage DC power source, wherein the on-board high-voltage device has a leakage resistance Rx with respect to a vehicle body, as typified by an equivalent leakage resistance on a positive potential side and an equivalent leakage resistance on a negative potential side, wherein the leakage resistance detection device includes a constant-voltage control power source which is supplied and driven by a low-voltage DC power source connected to the vehicle body with a negative terminal and which generates a control power supply voltage Vcc,The leakage resistance detection device is configured to measure a value of the leakage resistance Rx via a coupling capacitor, one terminal of which is connected to a predetermined position of the on-board high-voltage device. The leakage resistance detection device further includes a repetitive signal output circuit, a monitoring signal processing circuit, and an arithmetic control circuit, the arithmetic control circuit including a microprocessor and a program memory that cooperate with each other. The repetitive signal output circuit alternately repeats, in response to a switching operation of a charge / discharge switching element operating in response to a repetitive command signal PLS, a charging period and a discharging period in which a measurement point is intermittently connected to the control power supply voltage Vcc via a charging / discharging resistor, the measurement point being another terminal of the coupling capacitor.to thereby gradually increase or gradually decrease a monitoring voltage Vx, which is a potential between the measuring point and the vehicle body.

[0028] Furthermore, the monitor signal processing circuit is configured to: generate at least one of the inverted logic signals ANS, LOW and HIG when a value of the overvoltage Vx, which gradually increases or gradually decreases at a gradient that becomes smoother as the charge / discharge time constant becomes longer, passes a predetermined threshold voltage and input the at least one of the inverted logic signals ANS, LOW and HIG to the arithmetic control circuit, or generate an analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit.The arithmetic control circuit is configured to: measure a transition time Tx during which the value of the monitor voltage Vx changes from one threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to another; calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance on the positive potential side and the equivalent leakage resistance on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and generate a resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than a predetermined limit leakage resistance Rx0 or when the transition time Tx reaches a limit transition time Tx0 corresponding to the predetermined limit leakage resistance Rx0.The repetitive command signal PLS includes a pulse signal with a fixed cycle having the charge period or the discharge period longer than at least the boundary transition time Tx0, or a pulse signal with a variable cycle in which an output pulse is inverted when the transition time Tx is reached.

[0029] As described above, the leakage resistance detection device for an on-board high-voltage device according to the present invention includes a coupling capacitor having one terminal connected to the on-board high-voltage device and the other terminal connected to the repetitive signal output circuit. The arithmetic control circuit measures a leakage resistance as a function of the transition time during which the monitoring voltage, which is the potential at the other terminal of the coupling capacitor to be charged and discharged in response to the switching operation of the charge / discharge switching element, changes from a predetermined voltage to another predetermined voltage. When the measured leakage resistance becomes equal to or less than a predetermined limit leakage resistance, the arithmetic control circuit generates an abnormality determination output.

[0030] Therefore, such an abnormality determination threshold can be variably and easily set as a setting constant with respect to the arithmetic control circuit, making it possible to determine whether the leakage resistance has become equal to or smaller than the limit leakage resistance, and also to monitor the secular change in the leakage resistance. Furthermore, the calculation result of the leakage resistance can be obtained without waiting for the completion of a half-cycle of the charge / discharge pulse with respect to the coupling capacitor. In addition, after the monitoring voltage deviates from the correct voltage range due to an abrupt change in the leakage resistance and then returns to the correct voltage range, the leakage resistance measurement is performed by a charge / discharge pulse, at the latest after the half-cycle, without requiring multiple repetitive command signals.As a result, when the grounding abnormality has occurred, the resistance abnormality determination output can be quickly generated and therefore there is an effect of improving the search performance for the grounding abnormality that temporarily occurs due to vibration of the vehicle body. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In the attached drawings: Fig. 1 is an overall configuration diagram of an apparatus according to a first embodiment of the present invention; Fig. 2 is a timing diagram for describing the operation of the device of Fig. 1; Fig. 3 is a full range characteristic diagram for describing the operation of the device of Fig. 1; Fig. 4 is a low-range characteristic diagram for describing the operation of the device of Fig. 1; Fig. 5 is a high-range characteristic diagram for describing the operation of the device of Fig. 1; Fig. 6 a first flowchart for describing the device of Fig. 1; Fig. 7 is a second flowchart for describing the operation of the device of Fig. 1; Fig. 8 shows an overall configuration program of a device according to a second embodiment of the present invention; Fig. 9 is a timing diagram for describing the operation of the device of Fig. 8; Fig. 10 is a first flowchart for describing the operation of the device of Fig. 1; Fig. 11 is a second flowchart for describing the operation of the device of Fig. 8; Fig. 12 is an overall configuration diagram according to a third embodiment of the present invention; Fig. 13 is a timing chart for describing the operation of the device of Fig. 12; Fig. 14 is a first flowchart for describing the operation of the device of Fig. 12; Fig. 15 is a second flowchart for describing the operation of the device of Fig. 12; Fig. 16 is an overall configuration diagram of an apparatus according to the fourth embodiment of the present invention; Fig. 17 is a timing chart for describing the operation of the device of Fig. 16; Fig. 18 is a first flowchart for describing the operation of the device of Fig. 16; Fig. 19 is a second flowchart for describing the operation of the device of Fig. 16; Fig. 20 is an overall configuration diagram of an apparatus according to a fifth embodiment of the present invention. Fig. 21 is a timing chart for describing the operation of the device of Fig. 20; Fig. 22 is a full range characteristic diagram for describing the operation of the device of Fig. 20; Fig. 23 is a low-range characteristic diagram for describing the operation of the device of Fig. 20; Fig. 24 is a high-range characteristic diagram for describing the operation of the device of Fig. 20; Fig. 25 is a first flowchart for describing the operation of the apparatus of Fig. 20; Fig. 26 is a second flowchart for describing the operation of the apparatus of Fig. 20; Fig. 27 is a third flowchart for describing the operation of the apparatus of Fig. 20; Fig. 28 is an overall configuration diagram referring to a conventional known example; and Fig. 29 is a full range characteristic diagram for describing the operation of the device of Fig. 28. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0032] A leakage resistance detection device for an onboard high-voltage device and a leakage resistance detection method therefor according to exemplary embodiments of the present invention will be described below with reference to the accompanying drawings. In the drawings, the same or equivalent parts are denoted by the same reference numerals. First embodiment

[0033] Now, a description will be given with reference to an overall configuration diagram of Fig. 1, which illustrates a device according to a first embodiment of the present invention. Fig. 1 is a low-voltage DC power source 10 externally connected to a leakage resistance detection device 50A, an on-board battery such as a 12V lead-acid battery. The low-voltage DC power source 10 has a negative terminal grounded to a vehicle body 11. A positive terminal thereof supplies a first power supply voltage Vb via a low-voltage power supply switch 12 to a positive-side low-voltage power supply terminal 13 provided on the leakage resistance detection device 50A. The low-voltage power supply switch 12 uses an output contact of an electromagnetic relay that is energized when a manual power supply switch (not shown) is closed. When the manual power supply switch is opened, the low-voltage power supply switch 12 is opened after a short delay time, thus performing a delayed supply.

[0034] A negative-side low-voltage power supply terminal 14 provided on the leakage resistance detection device 50A is grounded to the vehicle body 11. A connection terminal 16 is connected to, for example, a negative-side power supply line 67 of the on-board high-voltage device 60X. A vehicle state signal 18 is information related to a leakage resistance measurement to be input to the leakage resistance detection device 50A, such as whether a high-voltage power supply switch 62 included in the on-board high-voltage device 60X, to be described below, is closed or whether a charger (not shown) is connected to a high-voltage DC power source 61. The vehicle state signal 18 is connected, for example, via a serial signal line.An abnormality alarm device 19 is, for example, a display device that operates in response to a resistance abnormality determination output ER1 or a pre-alarm output ER2 generated by the leakage resistance detection device 50A.

[0035] The high-voltage DC power source 61 included in the on-board high-voltage device 60X is a lithium battery with a nominal output voltage of, for example, 400 V. The high-voltage DC power source 61 supplies a second power supply voltage Vh to an inverter 63 via the high-voltage power supply switch 62. The inverter 63 is formed of a three-phase transistor bridge circuit and performs variable speed drive control on a high-voltage electrical load 64 such as a three-phase AC motor.

[0036] The on-board high-voltage device 60X, which includes the high-voltage DC power source 61, an electromagnetic relay (not shown) such as the high-voltage power supply switch 62, the inverter 63, and the high-voltage electrical load 64, is mounted in the vehicle body 11 while being electrically insulated from the vehicle body 11. In fact, however, a leakage resistance, typified by a positive-side equivalent leakage resistance 65 and a negative-side equivalent leakage resistance 66, is generated and distributed in the vehicle body.

[0037] If the value of an equivalent leakage resistance connected to a connection terminal 16 (corresponding to the equivalent leakage resistance 16 in Fig. 1), is represented by R2 and the value of the equivalent leakage resistance not connected to the connection terminal 16 (corresponding to the equivalent leakage resistance 65 in Fig. 1) is represented by R1, a voltage Vn generated between the intermediate terminal 16 connected to the negative-side power supply line 67 and the vehicle body 11 is calculated by expression (4x), which is the solution of a system of equations (1x) to (3x). I1=(Vh−Vn) / R1 I2=Vn / R2 I1=I0+I2 Vn=Vn0−I0×Rx where Vn0=Vh×R2 / (R1+R2)=Vh×Rx / R1Rx=R1×R2 / (R1+R2)

[0038] In the expressions, I1 represents a current flowing through the equivalent resistance R1 in the illustrated direction, I2 represents a current flowing through the equivalent resistance R2 in the illustrated direction, I0 represents a current flowing out of the connection terminal 16 in the illustrated direction, Rx represents a combined leakage resistance value, and Vn0 represents the value of Vn when I0 is 0 (stable divided voltage).

[0039] When the electrostatic capacitance of the coupling capacitor 51 is represented by C, a voltage across the coupling capacitor 51 (both-ended voltage) is represented by E, and a monitoring voltage is represented by Vx, equations (5x) and (6x) are established. C×(dE / dt)=I0 E=Vn+Vx

[0040] Expressions (4x) and (5x) are substituted into expression (6x) to obtain expression (7x). Vn0+Vx=E+C×Rx(dE / dt)

[0041] Next, an internal configuration of the leakage resistance detection device 50A will be described. The arithmetic control circuit 20A is an integrated circuit element that includes a microprocessor 21 as a main component, and a volatile RAM 22, a non-volatile data memory 23, and a non-volatile program memory 24A, such as a flash memory that can be electrically chip-erased in block units. The non-volatile data memory 23 uses a partial area of ​​the program memory 24A. Alternatively, the non-volatile data memory 23 may be installed outside the integrated circuit element and serially connected to the microprocessor 21 as a memory that can freely read and write data electrically in byte units.

[0042] A constant-voltage control power source 25 generates a stabilized control power supply voltage Vcc, such as 5 V DC, based on the first power supply voltage Vb supplied from the positive-side low-voltage power supply terminal 13 and fluctuating significantly. The constant-voltage control power source 25 supplies the control power supply voltage Vcc to the arithmetic control circuit 20A, as well as a repetitive signal output circuit 30A and a signal processing circuit 40A to be described below. The arithmetic control circuit 20A generates a repetitive command signal PLS, which is a pulse train signal with a fixed cycle, and inputs the repetitive command signal PLS to the repetitive signal output circuit 30A. The repetitive command signal PLS will be described below with reference to Fig. 2. The arithmetic control circuit 20A is connected to an inverted logic signal HIG received from the monitoring signal processing circuit 40A as an input signal. Furthermore, the arithmetic control circuit 20A generates the pre-alarm output ER2 when the measured value of the leakage resistance Rx becomes equal to or less than a predetermined preliminary leakage resistance Rxn, and generates the resistance abnormality determination output ER1 when the measured value of the leakage resistance Rx becomes equal to or less than a limit leakage resistance Rx0, which are output to the abnormality alarm device 19.

[0043] The repetitive signal output circuit 30A includes a charge / discharge switching element 31, a drive resistor 32, a charge / discharge resistor 33, and an open-circuit stabilizing resistor 34. The charge / discharge switching element 31 is, for example, an NPN transistor. The drive resistor 32 is connected to a base terminal of the charge / discharge switching element 31 and receives the repetitive command signal PLS as its input. The charge / discharge resistor 33 is connected to a collector terminal of the charge / discharge switching element 31, and the control power supply voltage Vcc is applied thereto. The open-circuit stabilizing resistor 34 is connected between the base terminal and an emitter terminal connected to the vehicle body 11.The collector terminal of the charge / discharge switching element 31 is further connected via a rapid charge / discharge resistor 35 to another terminal A of the coupling capacitor 51, which has a terminal B connected to the connection terminal 16.

[0044] The collector terminal of the charge / discharge switching element 31 is further connected to bypass diodes 37 and 36. When the potential of the other terminal A of the coupling capacitor 51 rises to be equal to or higher than the control power supply voltage Vcc, the bypass diode 37 feeds back a current to the constant-voltage control power source 25 through the fast charge / discharge resistor 35 and discharges the coupling capacitor 51, regardless of the operating state of the charge / discharge switching element 31. When the potential at the other terminal A of the coupling capacitor 51 decreases to be equal to or lower than the potential of the vehicle body, the bypass diode feeds back a current from the vehicle body 11 through the fast charge / discharge resistor 35 and charges the coupling capacitor 51, regardless of the operating state of the charge / discharge switching element 31.The potential between a measuring point A, which is the other terminal of the coupling capacitor 51, and the vehicle body 11 is input to the monitor signal processing circuit 40A as a monitor voltage Vx.

[0045] Note that a bypass leakage resistor 53 may be connected between the connection point B and the vehicle body 11, preventing the charge / discharge current from being too small with respect to the coupling capacitor 51 when the leakage resistance Rx is infinite. The value of an additional resistor R3, which is a resistance value of the bypass leakage resistor 53, is, for example, approximately ten times the limit leakage resistance Rx0 used for abnormality determination. The monitor signal processing circuit 40A, to which the monitor voltage Vx is input, includes a comparator 41, voltage dividing resistors 42 and 43, an input resistor 44, and a noise filter circuit formed by a smoothing resistor 45 and a smoothing capacitor 46.The voltage dividing resistors 42 and 43 divide the control power supply voltage Vcc to generate a threshold voltage VO, which is input to a negative input terminal of the comparator 41. The input resistor 44 is connected to a positive input terminal of the comparator 41. The smoothing resistor 45 has one terminal to which the monitoring voltage Vx is applied and another terminal connected to the positive input terminal of the comparator 41 via the input resistor 44.

[0046] The noise filter circuit prevents a noise signal having a much higher frequency than the repetition frequency of the repetitive command signal PLS from being input to the comparator 41. The influence of the noise filter appears as an error in which the measured transition time Tx becomes slightly longer. However, from the standpoint of actual control characteristics, the value of the monitor voltage Vx can be considered to be directly applied to the positive input terminal of the comparator 41. When the value of the monitor voltage Vx becomes equal to or greater than the set threshold voltage V0, the output logic of the comparator 41 becomes "H" and is input to the arithmetic control circuit 20A as an inverted logic signal HIG.

[0047] The following describes the effect and operation of the device according to the first embodiment of the present invention, which is shown in Fig. 1 illustrated, briefly with reference to a timing diagram of Fig. 2 as well as Fig. 1. First, in Fig. 1, when the manual power supply switch (not shown) is closed and the low-voltage power supply switch 12 is closed, the constant-voltage control power source 25 generates a predetermined control power supply voltage Vcc. The microprocessor 21 then starts its control operation and generates a signal shown in part A of Fig. 2. A first period (slow charge period) T1, in which the repetitive command signal PLS has a logic level "L," and a second period (fast discharge period) T2, in which the repetitive command signal PLS has a logic level "H," have the same half-cycle value of the total cycle T0=T1+T2. However, in actuality, T1≤T2 may be satisfied.

[0048] In Fig. 1, when the logic level of the repetitive command signal PLS is “L” (or “H”), the charge / discharge switching element 31 is interrupted (or becomes conductive) as shown in part B of Fig. 2. Then, in Fig. 1, when the charge / discharge switching element 31 is interrupted, the coupling capacitor 51 is charged from the constant voltage control power source 25 via a series connection of the charge / discharge resistor 33, the rapid charge / discharge resistor 35, and the leakage resistor Rx, with the result that the monitoring voltage Vx gradually increases. On the other hand, when in Fig. 1 the charge / discharge switching element 31 becomes conductive, the charged electric charges of the coupling capacitor 51 are discharged via a series connection of the rapid charge / discharge resistor 35 and the leakage resistor Rx, with the result that the monitoring voltage Vx decreases rapidly.

[0049] Part C of Fig. Figure 2 shows how the monitoring voltage Vx gradually increases or rapidly decreases. The gradient of the charge / discharge characteristics becomes smoother as the charge / discharge time constants expressed in expressions (8x) and (9x) increase. τ1=(R0+Rx)×C, R0=Rs+Rq τ2=(Rq+Rx)×C

[0050] In the expressions, Rs represents the resistance value of the charge / discharge resistor 33; Rq the resistance value of the fast charge / discharge resistor 35 (Rq< <Rs); R0=Rs+Rq≈Rs einen Reihenwiderstand; Rx einen parallelen kombinierten Widerstand der äquivalenten Leckwiderstände R1 und R2; C die elektrostatische Kapazität des Kopplungskondensators 51; τ1 eine langsame Ladezeitkonstante und τ2 eine Schnellentladezeitkonstante.

[0051] For example, the resistance value Rs is several hundred kΩ, while the resistance value Rq is several kΩ. When the value of the leakage resistance Rx is small, τ1>>t2 is established between the slow charge time constant τ1 and the fast discharge time constant τ2. However, when the value of the leakage resistance Rx is sufficiently larger than the resistance value Rs, τ1≈τ2 is established, and therefore the charge / discharge time constants have little difference, and fast discharge is not performed.

[0052] In part (C) of Fig. 2, the value of the monitoring voltage Vx at the end of the first period T1 after the slow charge is a first end voltage V2, and the value of the monitoring voltage Vx at the end of the second period T2 after the fast discharge is a second end voltage V1. The value of the second end voltage V1 must be almost 0. Therefore, the fast discharge constant τ2 must be smaller than the slow charge time constant τ1, and the second period T2 is desired to be longer than the first period T1. As a result, the initial value V1 of the gradually increasing monitoring voltage Vx is always 0 (or a value close to 0), and therefore the initial value does not fluctuate depending on the magnitude of the first end voltage V2.

[0053] Part (D) of Fig. Figure 2 shows the logic state of the inverted logic signal HIG. The inverted logic signal HIG has a logic level of “H” during a period in which the value of the monitoring voltage Vx, which is shown in part C of Fig. 2, is equal to or higher than a set threshold voltage VO. The arithmetic control circuit 20A measures a transition time Tx, which starts at the time of the logical inversion of the repetitive command signal PLS from "H" to "L" and continues until the logic of the inverted logic signal HIG changes from "L" to "H".

[0054] Part (E) of Fig. Figure 2 shows the waveform of the both-end voltage E of the coupling capacitor 51. A charge start voltage (=discharge end voltage) E1 is equal to the stable divided voltage Vn0 expressed in expression (4x). The value of a charge end voltage (=discharge start voltage) E2 increases to Vn0+Vcc in the case where the charge period T1 is infinite. The value of the transition time Tx in which the value of the gradually increasing monitor voltage Vx transitions from a first threshold value (V1=0) to a second threshold value (V0) is measured to thereby calculate the value of the leakage resistance Rx. Note that the first period T1 of the repetitive command signal PLS is set to be longer than a limit transition time Tx0, which corresponds to the value of the transition time Tx when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0, which is used for abnormality determination.

[0055] Hereinafter, based on the device according to the first embodiment of the present invention, which is shown in Fig. 1 illustrates a method for detecting the leakage resistance with reference to characteristic diagrams of the Fig. 3 to 5. In Fig. 3, the leakage resistance coefficient β on the horizontal axis is a ratio between the value of the parallel combined resistance of the equivalent leakage resistances R1 and R2, namely the value of the leakage resistance Rx=R1×R2 / (R1+R2), and the series resistance R0=Rs+Rq. The series resistance R0 is a known constant, and therefore the value of the leakage resistance coefficient β=Rx / R0 is proportional to the leakage resistance Rx.

[0056] A transition time coefficient α on the vertical axis is a ratio between a measured transition time Tx and a charge / discharge time constant τ=(R0+Rx)×C=R0×C with respect to the coupling capacitor 51 when the value of the leakage resistance Rx is 0. The charge / discharge time constant τ is a known constant, and therefore the value of the transition time coefficient α=Tx / (R0×C) is proportional to the transition time Tx.

[0057] A number of characteristic curves of Fig. 3 uses a threshold voltage coefficient γ as a parameter. In the case where the measurement of the transition time Tx is performed with respect to the monitoring voltage Vx, which gradually increases from 0 volts to V0 volts, the threshold voltage coefficient γ is V0 / Vcc. Even in the case where the measurement is performed with respect to the monitoring voltage Vx, which gradually decreases from Vcc volts to (Vcc-V0) volts, the threshold voltage coefficient γ satisfies the same expression. The threshold voltage V0 and the candidate reference value Vcc are known values, and therefore the value of the threshold voltage coefficient γ is a known value as a design constant.

[0058] The majority of characteristic curves using the threshold voltage coefficient γ as a parameter are classified into a negative gradient curve, as represented by curves 101, 102, and 103, in which the threshold voltage coefficient γ decreases along with the increase in the leakage resistance coefficient β, and a bell-shaped curve, as represented by curves 111, 112, and 113, which is convex in the center and has a positive gradient curve region and a negative gradient curve region.

[0059] Fig. 4 is an enlarged graph of curve 103 of Fig. 3 and shows a negative gradient curve for the threshold voltage coefficient γ of 0.6. In the case where the value of the series resistance R0 is designed to be 3.3 times the limit leakage resistance Rx0, the value of the leakage resistance coefficient β when the actual value of the leakage resistance Rx has become equal to the limit leakage resistance Rx0 is Rx / R0=Rx0 / (3.3Rx0)=0.3 and therefore the transition time coefficient α of 0.85 corresponding to a point of interest P4 of Fig. 4 the value corresponding to the measured transition time Tx.

[0060] Therefore, when the transition time coefficient α is 0.85 or more, the leakage resistance Rx is determined to be in a dangerous state of being equal to or less than the limit leakage resistance Rx0, and when the transition time coefficient α is less than 0.85, the leakage resistance Rx is determined to be in the safe state of exceeding the limit leakage resistance Rx0. For example, in a case where the limit leakage resistance Rx0 is set to 200 KΩ, the series resistance R0 is set to 200x3.3=660 KΩ, and the electrostatic capacitance C of the coupling capacitor 51 is set to 0.47 µF, the charge / discharge time constant τ is equal to 660 x 0.47=310 ms, and therefore the limit transition time Tx0 is α×(R0 ×C)=0.85 x 310=264 ms.

[0061] Furthermore, in Fig. 4. When the leakage resistance coefficient β is 0, the value of the transition time coefficient α is 0.92. Thus, the occurrence of a complete grounding abnormality can be detected as long as the loading period of the repetitive command signal PLS has a pulse width equal to or longer than 310 × 0.92 = 285 ms. On the other hand, a time-out abnormality is detected when an inverted logic signal cannot be obtained even after 250 ms because the logic of the repetitive command signal PLS has been inverted to the loading side.

[0062] Additionally, in Fig. 4. When the leakage resistance coefficient β of 1.4 (α=0.1) is the upper limit of measurement, the measurable value of the leakage resistance Rx is β×R0=1.4×(3.3×Rx0)=4.62Rx0. Therefore, it is difficult to accurately measure a leakage resistance exceeding 4.6 times the limit leakage resistance Rx0, but the normal state of the leakage resistance can be reliably detected.

[0063] Returning to Fig. 3, curve 112 can be used to measure the leakage resistance itself by excluding the positive gradient curve region and using only the negative gradient curve region. The horizontal axis of Fig. 3 exhibits a geometric scale, and therefore the transition time coefficient α appears to decrease rapidly. However, a smooth attenuation curve is obtained on an arithmetic scale.

[0064] In this case, attention is paid to the leakage resistance coefficient β of 6.4 corresponding to a point of interest P3 (α=2.23), where the value of the transition time coefficient α for the leakage resistance coefficient β of 0 is 2.3 or less. When the series resistance R0 is set to Rx0 / 6.4, the leakage resistance coefficient β when the leakage resistance Rx has decreased to the limit leakage resistance Rx0 is Rx0 / R0=6.4. Therefore, the value of the limit leakage resistance Rx0 can be detected based on the transition time coefficient α corresponding to a point of interest P3.

[0065] In this case, however, the leakage resistance equal to or less than the limit leakage resistance Rx0 cannot be measured, and if the transition time coefficient α exceeds 2.3, a binary problem occurs and a correct value of the leakage resistance cannot be specified. Furthermore, in the curve 112 of Fig. 3. When the leakage resistance coefficient β of 8.5 (α=0.49) is the upper limit of the measurement, the measurable value of the leakage resistance Rx is β×R0=8.5×(Rx0 / 6.4)=1.33Rx0. Therefore, it is difficult to accurately measure a leakage resistance exceeding 1.33 times the limit leakage resistance Rx0, but the normal state of the leakage resistance can be reliably detected.

[0066] If the resistance equal to or less than the limit leakage resistance can be measured, the measured resistance can be used as information to trace the cause of the drop in leakage resistance. However, in general, there is no problem even if the resistance equal to or less than the limit leakage resistance Rx0 cannot be measured. Furthermore, in general, as long as the leakage resistance for pre-alarming, which is about 1.3 times the limit leakage resistance Rx0, can be measured, it is unnecessary to accurately measure a larger leakage resistance. From this point of view, it is possible to use both the negative gradient curve 103 and the curve 112 with the negative gradient curve region following the positive gradient curve region. However, there is no advantage to using the negative gradient region of the bell-shaped curve, and it is more advantageous to use the curve 103 with only the negative gradient curve.

[0067] Fig. 5 is an enlarged graph of curve 113 of Fig. 3 and shows a positive gradient curve for the threshold voltage coefficient γ of 0.95. However, the value of the series resistance R0 is set to twice the value of the limit leakage resistance Rx0, and the additional resistance R3=10×Rx0 is connected in parallel with the bypass leakage resistance 53. Therefore, even if the actual leakage resistance Rx is infinite, the upper limit of the leakage resistance coefficient β is R3 / R0=10×Rx0 / (2×Rx0)=5, which is within the positive gradient curve range of the curve 113 of Fig. 3 lies.

[0068] In case of Fig. 5, the value of the leakage resistance coefficient β when the actual value of the leakage resistance Rx has become equal to the limit leakage resistance Rx0 is (Rx0 / / R3) / R0=(Rx0 / / 10Rx0) / 2Rx0)=0.45, and therefore the transition time coefficient α of 3.8 corresponding to a point of interest P5 of Fig. 5 a value corresponding to the measured transition time Tx.

[0069] For example, in a case where the limit leakage resistance Rx0 is set to 200 kΩ, the additional resistance R3 is 2 mΩ, and the series resistance R0 is 400 kΩ. Thus, as a reference, when the electrostatic capacitance C of the coupling capacitor 51 is set to 0.15 μF, the charge / discharge time constant is R0 ×C=400×0.15=60 ms. Therefore, the limit transition time Tx0 when the leakage resistance Rx has decreased to the limit leakage resistance Rx0 is α×(R0×C)=3.8×60=228 ms. Thus, the leakage resistance Rx can be determined to be in the dangerous state when the transition time Tx becomes 228 ms or less, and in the safe state when the transition time Tx exceeds 228 ms.

[0070] Furthermore, in Fig. 5, the value of the transition time coefficient α at the upper limit of the leakage resistance coefficient β is 5.0 7.2. Thus, theoretically, the leakage resistance can be detected up to an infinite value as long as the charging period of the repetitive command signal PLS has a pulse width equal to or longer than 60 × 7.2 = 432 ms. On the other hand, it is determined that an off-time abnormality exists if an inverted logic signal cannot be obtained even after 432 ms since the logic of the repetitive command signal PLS has been inverted to the charging side.

[0071] In the above description, the upper limit of the leakage resistance coefficient β through the bypass leakage resistance 53 for the limited use of the positive gradient curve region of the bell-shaped curve in Fig. 3. The bypass leakage resistance 53 has a known resistance value. Thus, at the time of product delivery inspection, by measuring the leakage resistance in the state where the onboard high-voltage device 60X is not connected, the individual fluctuations in the electrostatic capacitance C of the coupling capacitor 51 can be measured, so that a calibrated value of the electrostatic capacitance C can be calculated and stored.

[0072] The Fig. The characteristic curves shown in Figure 3 are based on the Fig. Expression 1 shown in Figure 3, and Expression 1 is established when the monitor voltage Vx falls within a proper range from 0 to Vcc. The establishment of Expression 1 is demonstrated as follows. Charging period

[0073] In the charging period in which the logic level of the repetitive command signal PLS in Fig. 1 is at "L" and the transistor serving as the charge / discharge switching element 31 is open, expression (10a) is established. VCC=R0×C(dE / dt)+Vx where R0=Rs+Rq≈Rs>>Rq Vx in expression (10a) is substituted into expression (7x) above to obtain expression (11a). Vcc+Vn0=E+τ1×(dE / dt) where τ1=(R0+Rx)C≈(Rs+Rx)C

[0074] In differential equation (11a), the initial value of E at time t=0 is represented by E1, and the value of E at infinity is represented by Vcc+Vn0. Then, a solution expressed by expression (12a) is obtained. E=E1exp(−t / τ1)+(Vcc+Vn0){1−exp(−t / τ1)}

[0075] In addition, the value of (dE / dt) in expression (10a) is substituted into expression (7x) above to obtain expression (13a). Vx / (R0 / / Rx)=Vcc / R0+(E−Vn0) / Rx where R0 / / Rx represents the parallel combined resistance of R0 and Rx. Unloading period

[0076] In the discharge period in which the logic level of the repetitive command signal PLS in Fig. 1 is "H" and the transistor serving as the charge / discharge switching element 31 is closed, expression (10b) is established. 0=C×Rq(dE / dt)+Vx

[0077] Vx in expression (10b) is substituted into expression (7x) above to obtain expression (11b). Vn0=τ2(dE / dt)+E where τ2=(Rq+Rx)×C

[0078] In differential equation (11b), the initial value of E at time t = 0 is represented by E2, and the value of E at infinite time is represented by Vn0. Then, a solution expressed by expression (12b) is obtained. E=E2exp(−t / τ2)+Vn0{1−exp(−t / τ2)}

[0079] In addition, the value of (dE / dt) in expression (10b) is substituted into expression (7x) above to obtain expression (13b). Vx / (Rq / / Rx)=(E−Vn0) / Rx where Rq / / Rx represents the parallel combined resistance of Rq and Rx. Stable condition

[0080] In the stable state in which the logic level of the repetitive command signal PLS in Fig. 1 is alternately inverted into the charging period T1 and the discharging period T2, and there is no change in the value of the leakage resistance Rx and the output Vh of the high-voltage DC power source 61, the initial values ​​E1 and E2 in expressions (12a) and (12b) are calculated as follows. First, in expression (12a), at the charging end time t=T1, E=E2 is established, and therefore expression (14a) is established. E2=E1×K1+(Vcc+Vn0)(1−K1) where K1=exp(−T1 / τ1)

[0081] In addition, in expression (12b), at the discharge end time t=T2, E=E1 is established and therefore expression (14b) is established. E1=E2×K2+Vn0(1−K2) where K2=exp(−T2 / τ2)

[0082] From expressions (14a) and (14b) the expressions (15a) and (15b) are obtained. E1−Vn0=Vcc×K0×K2 E2−Vn0=Vcc×K0 where K0=(1−K1) / (1−K1×K2) Conventional case

[0083] In Fig. In the conventional case illustrated in Figure 28, the charge / discharge time constants are τ1=τ2=τ=C×(R0+Rx), and the charge period T1 and the discharge period T2 are T1=T2=T. Thus, K1=K2=K=exp(-T / τ) is established, and KO is simplified to 1 / (1+K). A monitoring voltage value Vx2 at the end of charge (time t=T1=T) is calculated by Expression (16a) using Expressions (13a) and (15b). Vx2 / (R0 / / Rx)=Vcc / (E2−Vn0) / Rx=Vcc / R0+Vcc / {(1+K)×Rx}

[0084] A monitoring voltage value Vx1 at the end of discharge (time t=T2=T) is calculated by expression (16b) using expressions (13b) and (15a). However, in expression (13b), the fast charge / discharge resistance Rq is replaced by the charge / discharge resistance R0. Vx1 / (R0 / / Rx)=(E1−Vn0) / Rx=VccK / {(1+K)×Rx}

[0085] Therefore, a deviation voltage Vx2-Vx1 is calculated by expression (17) using expressions (16a) and (16b). Vx2−Vx1=(R0 / / Rx)[Vcc / R0+(Vcc / Rx)(1−K) / (1+K)]

[0086] In this case, if the threshold voltage coefficient γ is set to (Vx2-Vx1) / Vcc, the leakage resistance coefficient β is set to Rx / RO, and a reference time coefficient α is set to T / (R0×C), Expression (17) becomes a simplified unitless expression as in Expression (18). γ{β+(1−K) / (1+K) / (1+β)} where K=exp[−T / {(R0+Rx)C}]=exp{−α0 / (1+β)}

[0087] This is the one above in Fig. Expression III shown in Figure 29. In the case of the first embodiment

[0088] In Fig. 1 and Fig. 2, V1=Vx1=0 and V2=Vx2 are established at the charging start time t=0. At time Tx=0 to T1, if the value of Vx becomes equal to the set threshold voltage V0, Expression (19) for time t=Tx is calculated from Expression (12a). Note that a rapid discharge is performed during the discharge period, and therefore, E1=Vn0 is established and Kx=exp(-Tx / τ1) is established. Ex=Vn0exp(−Tx / τ1)+(Vcc+Vn0){1−exp(−Tx / τ1)} =Vcc(1−Kx)+Vn0

[0089] Expression (20) is obtained from expressions (13a) and (19). V0 / (R0 / / Rx)=Vcc / R0+(Ex−Vn0) / Rx=Vcc / R0+Vcc(1−Kx) / Rx

[0090] Then, the threshold voltage coefficient γ of V0 / Vcc, the leakage resistance coefficient β of Rx / RO, and the transition time coefficient α of Tx / (ROC) are substituted into Expression (20), and Expression (21) is obtained. γ=(1+β−Kx) / (1+β)∴Kx=(1+β)×(1−γ)

[0091] Kx=Exp[-Tx / {(R0+Rx)C}]=exp{-α / (1+β)} is established and therefore expression (22) is obtained by a natural logarithm transformation. α=(1+β)LOGe(1 / Kx)

[0092] Expressions (21) and (22) correspond to the above in Fig. Expression I shown in Figure 3.

[0093] Next, the action and operation of the device is described by Fig. 1 detailed with reference to the Fig. 6 and Fig. 7 illustrated operational explanation flowcharts. In the first flowchart of Fig. 6, step 1600 is a step in which the microprocessor 21 starts the operation of detecting the leakage resistance Rx, and the subsequent step 1600a is a determination step of determining whether the current operation is the first operation after the low-voltage power supply switch 12 has been closed. If it is the first cycle, it is determined as YES, and the flow proceeds to step 1600b. If the cyclic cycle continues, it is determined as NO, and the flow proceeds to step 1601.

[0094] Step 1600b is an initialization step in which the current value of a counter or timer described below is set to 0. The subsequent step 1601 is a step in which a ring counter that periodically counts a clock signal starts its counting operation. The subsequent step 1602 is a determination step in which, if the current value of the ring counter counted in step 1601 is in the first half range, it is determined as YES and the flow proceeds to step 1603a. If the current value is in the second half range, it is determined as NO and the flow proceeds to step 1603b.

[0095] Step 1603a is a step in which the output logic of the repetitive command signal is set to "H." Step 1603b is a step in which the output logic of the repetitive command signal PLS is set to "L." A step block 1604, which includes steps 1601 to 1603b, is a repetitive command signal generating means.

[0096] Step 1605, which is executed after step 1603a or step 1603b, is a determination step of determining whether or not the logic of the inverted logic signal HIG input to the microprocessor 21 has changed. If the logic has changed within one cycle of the repetitive command signal PLS, it is determined as YES, and the flow proceeds to step 1606c. If the logic has not changed, it is determined as NO, and the flow proceeds to step 1606a. The determination of YES in step 1605 indicates that the value of the monitor voltage Vx is within the proper range of 0 to Vcc and has increased or decreased to exceed the set threshold voltage V0.

[0097] In step 1606c, a timer 0, which has been activated in step 1606a, described below, is initialized and the flow leads via a continuation A to a step block 1700 of Fig. 7. In step 1606a, the timer 0 for transient abnormality determination starts its counting operation, and the flow proceeds to step 1606b. Step 1606b is a step serving as transient abnormality determination means, in which the current value of counter 0 that started counting in step 1606a is read to determine whether or not a predetermined period of time has elapsed. If the logic of the inverted logic signal HIG remains unchanged even after the predetermined period of time has elapsed, it is determined as YES, and the flow proceeds to step 1607a. If the logic has changed within the predetermined period of time, it is determined as NO, and the flow proceeds to step 1607b.

[0098] Step 1607a is transient characteristic abnormality processing means. For example, if the power supply voltage Vh of the high-voltage DC power source 61 has changed abruptly, or if the positive-side leakage resistance 65 or the negative-side leakage resistance 66 has changed to abruptly change the potential of the negative-side power supply line 67 of the high-voltage DC power source 61 relative to the vehicle body at a connection point B of the coupling capacitor 51, then the potential of the measurement point A temporarily becomes equal to or lower than the vehicle body potential (0 volts) or equal to or higher than the control power supply voltage Vcc, and the coupling capacitor 51 is charged or discharged through the bypass diode 36 or the bypass diode 37.In this case, when a transient charge / discharge time during which the monitoring voltage Vx corresponding to the voltage of the measuring point A has elapsed to the proper range of 0 to Vcc for measuring the leakage resistance Rx, it is determined that transient characteristic abnormality ERR2 exists, and a system abnormality alarm (not shown) is performed, followed by transition to a retreat operation mode (emergency or limp-home operation).

[0099] Note that in the case where the vehicle state signal 18 is predicted to change and the monitor voltage Vx temporarily deviates from the proper range, the determination of the transient characteristic abnormality ERR2 is avoided at least for a predetermined period of time.

[0100] Note that in this case, where no system abnormality occurs, such as a decrease in internal leakage resistance due to degradation of the coupling capacitor 51, disconnection of other wires, or short circuit, the maximum value of the transient charge / discharge time is shorter than the predetermined determination period in step 1506b. Normally, this is determined as NO in step 1606b, and the flow proceeds to step 1607b.

[0101] Step 1607b is performed in the case where NO is determined in step 1605, and it is determined in step 1606b that the logic has changed in the predetermined period. In step 1607b, the calculation of the Fig. 7 is suspended. The process proceeds via step 1607a or step 1607b or via a Fig. 7 shown continuation B to step 1608.

[0102] Step 1608 is a determination step of determining whether the current timing is in a period of retracting and saving the value of the leakage resistance Rx calculated in a step described below, the calculated leakage resistance Rx, or abnormality occurrence information in the non-volatile data memory 23. For example, if the current timing is in the delayed supply period in which the manual power supply switch (not shown) is opened and then the low-voltage power supply switch 12 is interrupted, this is determined as YES, and the flow proceeds to step 1609a. If the current timing is not the retraction period, this is determined as NO, and the flow proceeds to an operation completion step 1610. Note that in step 1608, the determination of YES may be made periodically at a predetermined time interval.

[0103] In step 1609a, moving average values ​​of the leakage resistance Rx calculated in step 1706, described below, are sequentially written and saved while updating the address of the non-volatile data memory. In the subsequent step 1609b, based on the determination information of the transient characteristic abnormality ERR2 written into the RAM 22 in step 1607a or based on the abnormality occurrence information determined in step 1709, described below, the accumulated number of abnormality occurrences is updated according to the type of abnormality occurrence, and then written and saved to a predetermined address of the non-volatile data memory 23. After that, the flow proceeds to the operation completion step 1610.

[0104] In operation completion step 1610, the microprocessor 21 executes another control program, and the flow advances again to the operation start step 1600 after a waiting time of, for example, 10 ms. Therefore, immediately after the start of operation, in the state where the initial charging of the coupling capacitor 51 has not been performed, the microprocessor 21 executes the steps in the first cycle in the order of step 1600, step 1600a (determination of YES), step 1600b, step block 1604, step 1605 (determination of NO), step 1606a, step 1606b (determination of NO), step 1607b, step 1608 (determination of NO), step 1610, and step 1600 after a waiting time of 10 ms. In the next and subsequent cycles, the microprocessor 21 performs a series of processing cyclically in a state where the determination of step 1600a is NO and step 1600b is not executed.

[0105] In the process, if the inversion operation of the inverted logic signal HIG is detected in step 1605, it is determined as YES, and the flow proceeds to step 1701 to be described below. On the other hand, in a period in which it is determined as NO in step 1605 and the predetermined time has not been reached in step 1606b, the steps in the order of step 1607b, step 1608 (emphasize NO), and step 1610 are executed, and the cyclic operation continues. Then, if it is determined as YES in step 1606b, step 1607a is executed.

[0106] Referring to the second flowchart of Fig. 7, step 1700 is a step block equivalent to steps 2700a to 2700c described with reference to Fig. 11. Step 1700 is a shipping calculation means for measuring the value of the electrostatic capacitance C of the coupling capacitor 51 at the time of shipping adjustment and storing a calibrated value thereof.

[0107] The following step 1701 is a determination step. In a period in which the repetitive command signal PLS, whose logic level is “L”, is Fig. 6 is generated so that the transistor serving as the charge / discharge switching element 31 is interrupted to slowly charge the coupling capacitor 51 via the charge / discharge resistors 33 and 35, this is determined as YES and the flow advances to step 1702. In a period in which the repetitive command signal PLS, whose logic level is "H", in step 1603a of Fig. 6 is generated so that the transistor serving as the charge / discharge switching element 31 becomes conductive to quickly discharge the coupling capacitor 51 via the rapid charge / discharge resistor 35, NO is determined and the flow advances to step 1608 of Fig. 6 via continuation B.

[0108] In step 1702, a timer for measuring a transition time Tx is activated to start the counting operation, and the flow proceeds to step 1703. Step 1703 is a determination step of determining whether the logic level of the inverted logic signal HIG has changed from "L" to "H." If the logic level is "H," it is determined as YES, and the flow proceeds to step 1704a. If the logic level is "L," it is determined as NO, and the flow proceeds to step 1704b.

[0109] In step 1704a, the current value of timer 1, which has been activated to start counting in step 1702, is read out and stored in RAM 22. The subsequent step 1705 is a step serving as leakage resistance calculation means for calculating the leakage resistance Rx based on the value of the transmission time Tx read out and stored in step 1704a.

[0110] In the subsequent step 1706, the current value of the leakage resistance Rx calculated in step 1705 is input to the first stage of the shifter (shift register) constituted by the RAM 22, and pieces of past data stored in the shift register are sequentially migrated to the subsequent stage. The past data of the leakage resistance Rx stored in the last stage are supplied and cleared, and the sum of the values ​​of the leakage resistance Rx remaining in the shift register is divided by the number of values ​​stored in the shift register to thereby calculate the moving average value. In this way, the measurement can be prevented from being directly affected by a temporal measurement error of the leakage resistance Rx caused by temporal fluctuations in the power supply voltage Vh of the high-voltage DC power source 61 or, for example, noise.

[0111] The subsequent step 1707a is a step that serves as resistance abnormality determination means. If the moving average value of the leakage resistance Rx calculated in step 1706 becomes equal to or less than the predetermined leakage resistance Rxn or the limit leakage resistance Rx0, this is determined as YES and the flow proceeds to step 1709. If there is no abnormality, this is determined as NO and the flow proceeds to step 1708.

[0112] Note that in the case where there is a difference of a predetermined value or more between the moving average value of the leakage resistance Rx calculated in step 1706 and the leakage resistance Rx calculated in step 1609a of Fig. 6 in the non-volatile data memory 23, and if it is detected that the leakage resistance Rx has abruptly decreased, if not to the predetermined leakage resistance Rx0, this is determined as YES in step 1707a and the flow proceeds to step 1709. Step 1708 is a step in which the timer 1 activated in step 1702 is initialized and suspended, and the flow proceeds to step 1608 of Fig. 6 via forwarding B.

[0113] Step 1704b is performed in a case where NO is determined in step 1703. In step 1704b, in order to measure the elapsed time from when the logic level of the repetitive command signal PLS changes to "L" until the logic level of the inverted logic signal HIG changes to "H", the current value of the timer 1 activated in step 1702 is updated and read out. The subsequent step 1707b is a step serving as time-lapse abnormality determination means. If the elapsed time read out and stored in step 1704b is excessive and exceeds a predetermined period, this is determined as YES and the flow advances to step 1709. If the elapsed time is within the predetermined period, this is determined as NO and the flow advances to step 1708 of Fig. 6 via forwarding B.

[0114] Step 1709 is a step serving as measurement abnormality processing means. If determined as YES in step 1707a, the measurement abnormality processing means generates the pre-alarm output ER2 or the resistance abnormality determination output ER1 and stores abnormality determination information for the state change abnormality ERR3. If determined as YES in step 1707b, the measurement abnormality processing means stores abnormality determination information for the time-out abnormality determination ERR1. After step 1709, the flow proceeds to step 1608 of Fig. 6 via forwarding B.

[0115] In step 1709, when the pre-alarm output ER2 or the resistance abnormality determination output ER1 is generated, the abnormality alarm device 19 is activated. When the abnormality determination information for the time-out abnormality determination ERR1 or the state change abnormality ERR3 is stored, the stored abnormality determination information is separated from other abnormality information. In step 1609b of Fig. 6, the accumulated number of abnormality occurrences is written and saved in the non-volatile data memory 23, which is used for maintenance and inspection operation.

[0116] Note that the calculation of the leakage resistance Rx in step 1705 is performed in three ways. First, the calculation is performed based on the characteristic curve whose overall region is the negative gradient curve region, as shown by the curve 103 of Fig. 3. Second, the calculation is performed using the negative gradient curve region of a bell-shaped curve as represented by curve 112 of Fig. 3. Third, the calculation is performed using the positive gradient curve region of a bell-shaped curve, as represented by curve 113 of Fig. 3. In each case, the value of the leakage resistance Rx is calculated from the transition time measurement resistance Tx based on a functional formula or a data table between the transition time coefficient α and the leakage resistance coefficient β using the threshold voltage coefficient γ as a parameter.

[0117] As an alternative abnormality determination method, instead of comparing the detected leakage resistance Rx with the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, the preliminary transition time Txn or limit transition time Tx0 corresponding to the preliminary leakage resistance Rxn or the limit leakage resistance Rx0 can be calculated in advance, and the actual transition time Tx can be compared with the preliminary transition time Txn or the limit transition time Tx0.

[0118] Furthermore, in a case where the value of the reference time constant R0xC, which is the denominator of the transition time coefficient α, the value of the series resistance R0, which is the denominator of the leakage resistance coefficient β, or the value of the threshold voltage coefficient γ is considered a fixed value, a formula and a data table between the remaining time Tx and the leakage resistance Rx can be directly used as a functional formula or data table. In this case, the leakage resistance Rx can be directly calculated from the measured transition time Tx without performing complicated calculation processing during operation.

[0119] However, in the case of varying the value of the threshold voltage coefficient γ, the denominator value of the transition time coefficient α, or the denominator value of the leakage resistance coefficient β according to the applied vehicle types or the operating condition, it is advantageous to set the functional formula or the data table using α, β, or γ, which are unitless index values. Alternatively, in a case where the arithmetic control circuit 20A has a built-in multi-channel A / D converter and the number of analog inputs is sufficient, an operational amplifier may be used instead of the comparator 41, and an analog signal voltage proportional to the monitor voltage Vx may be input to the arithmetic control circuit 20A so that the microprocessor 21 generates a signal equivalent to the inverted logic signal HIG. In addition, in the first embodiment of Fig. 1, the connection point B is connected to the negative-side power supply line 67 of the on-board high-voltage device 60X, but even if the connection point B is connected to a positive-side power supply line, no change occurs in the expression I.

[0120] As is clear from the above description, according to the first embodiment of the present invention, the leakage resistance detection device 50A for an on-board high-voltage device is provided, which is connectable to the on-board high-voltage device 60X including the high-voltage DC power source 61 and the high-voltage electrical load 64 to be supplied and driven by the high-voltage DC power source 61, the on-board high-voltage device 60X having the leakage resistance Rx with respect to the vehicle body 11, as typified by the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side, the leakage resistance detection device 50A including the constant-voltage control power source 25 provided by the low-voltage DC power source 10 having a negative terminal connected to the vehicle body 11 and generating the control power supply voltage Vcc,is supplied and driven, wherein the leakage resistance detection device 50A is configured to detect a value of the leakage resistance Rx via the coupling capacitor 51 having a terminal B connected to a predetermined position of the on-board high-voltage device 60X, the leakage resistance detection device 50A further includes the repetitive signal output circuit 30A, the monitoring signal processing circuit 40A, and the arithmetic control circuit 20A, wherein the arithmetic control circuit 20A includes the microprocessor 21 and the program memory 24, which cooperate with each other.

[0121] Furthermore, the repetitive signal output circuit 30A alternately repeats, in response to the switching operation of the charge / discharge switching element 31 operating in response to the repetitive command signal PLS, a charge period and a discharge period in which the measuring point A is intermittently connected to the control power supply voltage Vcc via the charge / discharge resistor 33 or 35, the measuring point A being another terminal of the coupling capacitor 51, to thereby gradually increase or gradually decrease the monitor voltage Vx, which is a potential between the measuring point A and the vehicle body 11.The monitor signal processing circuit 40A is configured to: generate the inverted logic signal HIG when a value of the monitor voltage Vx, which gradually increases or gradually decreases at a gradient that becomes smoother as the charge / discharge time constant becomes longer, passes a predetermined threshold voltage, and input the inverted logic signal HIG to the arithmetic control circuit 20A, or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage level ANL to the arithmetic control circuit 20A.The arithmetic control circuit 20A is configured to: measure the transition time Tx during which the value of the monitor voltage Vx changes from a threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to each other; calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance 65 on the positive potential side and the negative-side equivalent leakage resistance 66 on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and generate the resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than the predetermined limit leakage resistance Rx0 or when the transition time Tx reaches the limit transition time Tx0 corresponding to the predetermined limit leakage resistance Rx0.The repetitive command signal PLS is a pulse signal with a fixed cycle that has a charging period or discharging period longer than at least the boundary transition time Tx0.

[0122] The charge / discharge resistor includes the reference resistor 33 with the resistance value Rs and the quick charge / discharge resistor 35 with the resistance value Rq sufficiently smaller than the resistance value Rs. The quick charge / discharge resistor 35 is connected to a negative terminal and a positive terminal of the constant-voltage control power source 25 via the pair of bypass diodes 36 and 37. The reference resistor 33 is connected in series with the leakage resistor Rx as a charge / discharge resistor for the coupling capacitor 51 when the arithmetic control circuit 20A measures the transition time Tx of the gradually increasing monitor voltage Vx or the gradually decreasing monitor voltage Vx. The reference resistor 33 is removed when a potential of the monitor voltage Vx relative to the vehicle body has changed to 0 volts or lower or the control power supply voltage Vcc or higher.due to an abrupt change in the power supply voltage of the high-voltage direct current source 61 or an abrupt change in the equivalent leakage resistance 65 on the positive potential side or the equivalent leakage resistance 66 on the negative potential side, because the high-voltage power supply switch 62 for the high-voltage electrical load 64 is closed or opened, or because an abnormality has occurred so as to charge and discharge the coupling capacitor 51 via the rapid charge / discharge resistor 35 and the bypass diode 36 or the bypass diode 37 regardless of the operating state of the charge / discharge switching element 31.

[0123] As described above, in conjunction with claim 2 of the present invention, the repetitive signal output circuit includes the reference resistor connected in series with the coupling capacitor and the rapid charge / discharge resistor having a resistance sufficiently smaller than that of the reference resistor. When the potential of the coupling capacitor changes abruptly at the connection point relative to the vehicle body, the reference resistor is removed to charge / discharge the coupling capacitor.

[0124] Therefore, during the abrupt change transition period, the charge / discharge time constant for the coupling capacitor has a value determined by the product of the sum of the fast charge / discharge resistance and the leakage resistance, and the electrostatic capacitance of the coupling capacitor. Especially when the leakage resistance has decreased to near the dangerous state, the coupling capacitor is rapidly charged / discharged, allowing the monitoring voltage Vx to return to the correct range for monitoring. Thus, the leakage resistance measurement can be performed quickly, and excessive current is suppressed by the fast charge / discharge resistor.

[0125] The repetitive signal output circuit 30A is configured to, in response to the switching operation of the charge / discharge switching element 31, connect the measuring point A, which is another terminal of the coupling capacitor 51, to the output terminal of the constant-voltage control power source 25 via the reference resistor 33 having the reference value Rs, or to connect the measuring point A of the negative terminal corresponding to the vehicle body potential via the rapid charge / discharge resistor 35 having the resistance value Rq sufficiently smaller than the resistance value Rs, so as to gradually increase or rapidly decrease the monitor voltage Vx, which is the potential between the measuring point A and the vehicle body 11.The monitor signal processing circuit 40A is configured to: generate the inverted logic signal HIG when the value of the monitor voltage Vx, which gradually increases at a gradient that becomes smoother as the time constant (Rs+Rx)C, which is a product of the sum of a resistance value Rs (Rs>>Rq), the leakage resistance Rx, and the electrostatic capacitance C of the coupling capacitor 51, increases, has increased to pass the predetermined threshold voltage V0 and input the inverted logic signal HIG to the arithmetic control circuit 20A; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20A.The arithmetic control circuit 20A inverts an output of the repetitive command signal PLS, which is a pulse train signal having a fixed cycle, in a state where the value of the monitor voltage Vx has abruptly decreased to approach 0 volts, through the charge / discharge switching element 31, and inverts the output of the repetitive command signal PLS after receiving the inverted logic signal HIG as an input or after measuring, at the transition time Tx, a period from when the inverted logic signal HIG is generated based on the input analog voltage ANL until when the inverted logic signal HIG is obtained.

[0126] As described above, in conjunction with claim 3 of the present invention, the repetitive signal output circuit charges and discharges the coupling capacitor, and the arithmetic control circuit measures, as the transition time Tx, a period during which the monitor voltage Vx rises to the predetermined threshold voltage V0 from a threshold voltage of 0 volts. Therefore, the comparator determination circuit or the comparator determination processing with respect to one of the pairs of threshold voltages can be omitted, and the transition time can be measured by a simple structure.

[0127] As is clear from the above description, according to the first embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used as the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30A, is represented by R0, and when a period during which the monitor voltage Vx gradually increases from a state of 0 to the predetermined threshold voltage V0,represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to V0 / VCC: calculating, by the arithmetic control circuit 20A, the value of the leakage resistance Rx corresponding to the measured transition time Tx by using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a resistance value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; selectively using a negative gradient curve region following the positive gradient curve region,or a negative gradient curve region occupying an entire region, and determining the threshold voltage coefficient y such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the negative gradient curve region; determining, when the transition time Tx has become a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds a short-circuit transition time Tx00 corresponding to a case where the leakage resistance Rx becomes 0.

[0128] As described above, in conjunction with claim 8 of the present invention, in the characteristic curve using the threshold voltage coefficient y proportional to the set threshold voltage as a parameter and showing the relationship between the leakage resistance coefficient β proportional to the leakage resistance Rx and the transition time coefficient α proportional to the transition time Tx, the negative gradient curve region in which the transition time coefficient α gradually decreases along with the increase of the leakage resistance coefficient β is used. When the measured transition time Tx reaches a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, it is determined that the leakage resistance Rx is too small, and the resistance abnormality determination output ER1 is generated.When the transition time Tx exceeds the short-circuit transition time Tx00 at which the leakage resistance Rx0 is, the off-time abnormality determination ERR1 is performed.

[0129] Therefore, if the leakage resistance Rx is within the normal range, the measurement of the transition time Tx is completed in a short period of time, confirming the normal state. If the transition time Tx becomes longer, the resistance abnormality determination output ER1 is generated. If the measurement of the transition time Tx has not been performed even after the short-circuit transition time Tx00 has been exceeded, the off-time abnormality determination ERR1 is performed. In this way, some kind of determination result is obtained each time the repetitive command signal is repeatedly operated.

[0130] The characteristic curve has a negative gradient curve region as an overall region, and a low threshold voltage coefficient γ is applied so that the characteristic curve does not include a positive gradient curve region. As described above, in conjunction with claim 9 of the present invention, the threshold voltage coefficient is low, and the characteristic curve to be applied has the negative gradient curve region as the overall region. Therefore, the leakage resistance equal to or lower than the limit leakage resistance can be accurately measured, and the leakage resistance can be measured in the abnormal state including the short-circuit state. Thus, information effective for tracking the occurrence of an abnormality can be obtained.

[0131] As is clear from the above description, according to the first embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used as the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30A, is represented by R0, and when a period,during which the monitor voltage Vx gradually increases from a state of 0 to the predetermined threshold voltage V0 represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / RO, and the threshold voltage coefficient y is set to VO / Vcc: calculating, by the arithmetic control circuit 20A, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient y as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; Exclude, in advance,by a characteristic curve correction means, applying a negative gradient curve region following the positive gradient curve region; determining the threshold voltage coefficient y such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the corrected positive gradient curve region; determining, when the transition time Tx has become a value equal to or less than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the time-off abnormality determination ERR1 when the transition time Tx exceeds an open-circuit transition time Txm corresponding to a case where the leakage resistance Rx is infinite.

[0132] As described above, in conjunction with claim 10 of the present invention, in the characteristic curve using the threshold voltage coefficient y proportional to the set threshold voltage as a parameter and showing the relationship between the leakage resistance coefficient β proportional to the leakage resistance Rx and the transition time coefficient α proportional to the transition time Tx, the corrected positive gradient curve region in which the transition time coefficient α gradually increases is used together with the increase in the leakage resistance coefficient β. When the measured transition time Tx has reached a value equal to or less than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, it is determined that the leakage resistance Rx is too small, and the resistance abnormality determination output ER1 is generated.When the transition time Tx exceeds the open circuit transition time Txm at which the leakage resistance Rx is infinite, the off-time abnormality determination is performed.

[0133] Therefore, when the leakage resistance Rx is within the abnormal range, the measurement of the transition time Tx is completed in a short period of time, and the resistance abnormality determination output ER1 is generated. The case where the transition time Tx becomes longer is determined to be normal. If the measurement of the transition time Tx is not performed even after the open-circuit transition time Txm has been exceeded, the time-out abnormality determination ERR1 is performed. In this way, a certain determination result is obtained each time the repetitive command signal operates repetitively.

[0134] The leakage resistance detection method further includes: connecting, in advance, by the characteristic curve correcting means, the bypass leakage resistor 53 between the connection point B of the coupling capacitor 51 and the on-board high-voltage device 60X, and the vehicle body 11, the bypass leakage resistor 53 having the additional resistance R3 sufficiently larger than the limit leakage resistance Rx0, and suppressing, by the bypass leakage resistor, the value of the leakage resistance coefficient β to R3 / R0 when the leakage resistance Rx is infinite, thereby preventing the transition time coefficient α from becoming excessive.

[0135] As described above, in conjunction with claim 11 of the present invention, the bypass leakage resistance is used so that in the characteristic curve of the transition time coefficient α relative to the leakage resistance coefficient β having the positive gradient curve region, the positive gradient characteristic curve can avoid being a negative gradient curve along with the increase in the leakage resistance coefficient β.

[0136] Therefore, it is possible to prevent two kinds of large and small values ​​of the leakage resistance Rx from being calculated for the measured transition time Tx, and it is possible to measure the value of the leakage resistance Rx which varies in a wide range.

[0137] The leakage resistance detection device further includes the bypass leakage resistor 53 connected between the connection point B between the coupling capacitor 51 and the on-board high-voltage device 60X, and the vehicle body 11. The leakage resistance detection method further includes measuring, by the arithmetic control circuit 20A, the leakage resistance Rx in a state where the coupling capacitor 51 and the on-board high-voltage device 60X are not connected to each other, and calibrating and storing the value of the electrostatic capacitance C of the coupling capacitor 51 so that an obtained result becomes the additional resistance R3 of the bypass leakage resistor 53.

[0138] As described above, in connection with claim 12 of the present invention, the value of the electrostatic capacitance C of the coupling capacitor is recalculated using the bypass leakage resistance having a known resistance value.

[0139] Therefore, it is possible to reduce the occurrence of a calculation error of the leakage resistance caused by the influence of individual fluctuations in the electrostatic update period, and it is possible to suppress a calculation error of the leakage resistance by suppressing the fluctuation range of the leakage resistance coefficient β.

[0140] The leakage resistance detection method further includes, by the arithmetic control circuit 20A: measuring, in response to the inverted logic signal HIG or the analog signal voltage ANL operating in response to the monitor voltage Vx input from the monitor signal processing circuit 40A, a period during which the monitor voltage Vx deviates from a correct voltage range determined by a threshold voltage and another threshold voltage for measuring the transition time Tx; and interrupting the measurement of the leakage resistance Rx when the measurement period is within a predetermined allowable determination period, and determining the transient characteristic abnormality ERR2 when the measured period exceeds a predetermined abnormality determination period.

[0141] As described above, in conjunction with claim 13 of the present invention, the arithmetic control circuit suspends the measurement of the leakage resistance in a predetermined period of time in which the monitor voltage Vx is outside the proper voltage range, and after elapse of the predetermined period of time, the arithmetic control circuit performs transient characteristic abnormality determination.

[0142] Therefore, in the case where the power supply voltage of the high-voltage DC power source changes abruptly or the leakage resistance changes abruptly in response to the connection or disconnection of the high-voltage electrical load, it is possible to prevent incorrect calculation of the leakage resistance and it is possible to detect the occurrence of system abnormality, including a problem that the leakage resistance within the coupling capacitor has reduced.

[0143] The leakage resistance detection method further includes: inputting the vehicle state signal 18 to the arithmetic control circuit 20A, the vehicle state signal 18 being a signal for identifying the occurrence of a condition responsible for fluctuations in the vehicle body potential at a connection point between the coupling capacitor 51 and the on-board high-voltage device 60X; and detecting, by the arithmetic control circuit 20A, a change in the state of the vehicle state signal 18 to predict an abrupt change in the leakage resistance, and avoiding the determination of the transient characteristic abnormality immediately after the change in the state of the vehicle state signal 18.

[0144] As described above, in conjunction with claim 14 of the present invention, the leakage resistance detection device receives the vehicle state signal used for predicting the abrupt change in the leakage resistance as an input, and neglects the transient characteristic abnormality immediately after the state of the vehicle state signal has changed.

[0145] Therefore, the occurrence of system abnormality, which never occurs under normal circumstances, can be detected due to the occurrence of the transient characteristic abnormality.

[0146] The leakage resistance detection method further includes: causing the non-volatile data memory 23, which is a portion of the program memory 24A or a memory provided together with the program memory 24A, to preliminarily write and store a data table related to a characteristic curve of the transition time coefficient α of Tx / (R0C) relative to the leakage resistance coefficient β of Rx / RO using the threshold voltage coefficient γ of VO / Vcc as a parameter, and a value of the threshold voltage coefficient γ, values ​​of the reference value Rs and the rapid charge / discharge resistance Rq that determine the value of the series resistance R0, a value of the electrostatic capacitance C, a value of the limit leakage resistance Rx0, and a value of the preliminary leakage resistance Rxn to be preliminarily notified;and write and store the value of the leakage resistance Rx measured during operation and abnormality occurrence history information periodically or immediately before the operation is stopped; and read out, by the arithmetic control circuit 20A, the value of the leakage resistance coefficient β from the data table in correspondence with the transition time coefficient α calculated by the transition time Tx, to thereby calculate a current leakage resistance Rx, and compare the current leakage resistance Rx with the value of the limit leakage resistance Rx0 or the value of the preliminary load resistance Rxn, thereby generating a pre-alarm output ER2 in addition to the resistance abnormality determination output ER1, or performing state change abnormality determination ERR3 when the leakage resistance has abruptly decreased in a time series.

[0147] As described above, in connection with claim 15 of the present invention, in the non-volatile data memory, the data table for measuring the leakage resistance and the basic data for performing the abnormality determination are written and saved, and the generation of the pre-alarm output is performed in addition to the resistance abnormality output or the determination of the state change abnormality.

[0148] Therefore, in the case where the leakage resistance has gradually decreased due to aging degradation of electrical insulation or the leakage resistance has abruptly decreased due to an immersion accident or the like, it is possible to request maintenance and inspection before the leakage resistance reduces to the limit leakage resistance. Second embodiment

[0149] Referring to Fig. 8, which is an overall configuration diagram of an apparatus of a second embodiment of the present invention, mainly differences from Fig. 1 below. In the drawings, the same reference numerals designate the same or equivalent parts. In Fig. 8, a leakage resistance detection device 50B includes an arithmetic control circuit 20B, a repetitive signal output circuit 30B, and a monitor signal processing circuit 40B, and detects a leakage resistance of the on-board high voltage device 60Y.

[0150] The first major difference is that in the on-board high-voltage device 60Y, a positive-side power supply line 68 is connected to a terminal B of the coupling capacitor 51 via the connection terminal 16, and the positive-side leakage resistor 65 has an equivalent leakage resistance R2 on the connection side, while the negative-side equivalent leakage resistor 66 has an equivalent leakage resistance R1 on the non-connection side.

[0151] The second main difference is that the repetitive signal output circuit 30B includes a charge / discharge switching element 31 such as a PNP transistor, and when the output logic level of the repetitive command signal PLS is "H", an auxiliary switching element 31A becomes conductive via an auxiliary drive resistor 32A so that the charge / discharge switching element 31 becomes conductive via the drive resistor 32, and when the output logic level of the repetitive command signal PLS is "L", the auxiliary switching element 31 becomes non-conductive through the auxiliary stabilizing resistor 34A so that the auxiliary switching element 31 becomes non-conductive through an open-circuit stabilizing resistor 34.

[0152] The coupling capacitor 51 is charged by a voltage Vn across the equivalent leakage resistance R2 (both-ended voltage Vn), and the value of the both-ended voltage Vn is expressed by expression (4y). I1=(Vh−Vn) / R1 I2=Vn / R2 I1=I0+I2 Vn=Vn0−I0×Rx where Vn0=Vh×R2 / (R1+R2)=Vh×Rx / R1Rx=R1×R2 / (R1+R2)

[0153] In the expressions, I1 represents a current flowing through the equivalent resistance R1 in the illustrated direction, I2 represents a current flowing through the equivalent resistance R2 in the illustrated direction, I0 represents a current flowing out of the connection terminal 16 in the illustrated direction, Rx represents a combined leakage resistance value, and Vn0 represents the value of Vn when I0 is 0 (stable divided voltage).

[0154] When the electrostatic capacitance of the coupling capacitor 51 is represented by C, a voltage across the coupling capacitor 51 (both-ended voltage) is represented by E, and a monitor voltage is represented by Vx, equations (5y) and (6y) are established. C×(dE / dt)=I0… (5y)=(5x) E=Vn−Vx… (6y)≠(6x)

[0155] Expressions (4y) and (5y) are substituted into expression (6y) to obtain expression (7y). Vn0−Vx=E+C×Rx(dE / dt)… (7y)≠(7x)

[0156] Therefore, when the transistor serving as the auxiliary switching element 31 becomes continuously conductive, the charging voltage E of the coupling capacitor 51 decreases to VnO-Vcc, and when the transistor is continuously interrupted, the charging voltage E of the coupling capacitor 51 increases to Vn0. Thus, the charging / discharging concept is opposite to the case of Fig. 1 and the discharge is carried out when the transistor is conductive and the charge is carried out when the transistor is open.

[0157] The third main difference is that the monitor signal processing circuit 40B uses an operational amplifier 49 connected to a negative feedback resistor 47 instead of using the comparator 41. The operational amplifier 49 generates an analog signal voltage ANL proportional to the monitor voltage Vx instead of generating the inverted logic signal HIG and outputs the analog signal voltage ANL to the arithmetic control circuit 20B.

[0158] The arithmetic control circuit 20B generates a repetitive command signal PLS, which is a pulse train signal with a fixed cycle, and outputs the repetitive command signal PLS to the repetitive signal output circuit 30B. The repetitive command signal PLS is described below with reference to Fig. 9. When the value of the monitor voltage Vx reaches (Vcc-V0) or less, which is obtained by subtracting the set threshold voltage V0 from the control power supply voltage Vcc, the microprocessor 21 operates with a program memory 24B to generate an inverted logic signal LOW within the arithmetic control circuit 20B based on the analog signal voltage ANL obtained from the monitor signal processing circuit 40B.

[0159] Furthermore, the arithmetic control circuit 20B generates a pre-alarm output ER2 when the measured value of the leakage resistance Rx becomes equal to or less than a predetermined leakage resistance Rxn, and generates a resistance abnormality determination output ER1 when the measured value of the leakage resistance Rx becomes equal to or less than a limit leakage resistance Rx0. The resistance abnormality determination output ER1 and the pre-alarm output ER2 are outputs to the abnormality alarm device 19 and are fed back to be input to the arithmetic control circuit 20B as the report command confirmation signal RET to check whether the abnormality alarm output has been correctly generated.

[0160] The fourth major difference is that a series circuit of a test leakage resistor and a test switching element 55 is connected between the connection terminal 16 and the vehicle body 11. The test switching element 55, such as an optical isolation transistor, receives a conduction command via a drive resistor 56 from a test operation command TST provided to the arithmetic control circuit 20B. A test resistor R4, which is a resistance value of the test leakage resistor 54, is equal to, for example, the limit leakage resistance Rx0 or the preliminary leakage resistance Rxm. The test switching element 55 temporarily becomes conductive at the start of operation, immediately after the manual power supply switch (not shown) is closed and the low-voltage power supply switch 12 is closed.

[0161] In this way, a pre-operation inspection is performed to check whether the detection control of the leakage resistance is performed correctly or not, and it is also possible to detect individual fluctuation in the electrostatic capacitance C of the coupling capacitor 51 and its secular change so that the calibrated value of the electrostatic capacitance C can be used. Note that similar to Fig. 1, the low-voltage DC power source 10, the low-voltage power supply switch 12, the vehicle state signal 18, and the abnormality alarm device 19 are connected outside the leakage resistance detection device 50B, and the constant-voltage control power source 25, which generates the control power supply voltage Vcc, is provided in the leakage resistance detection device 50B.

[0162] The following describes the effect and operation of the device according to the second embodiment of the present invention, which is shown in Fig. 8 illustrated, briefly with reference to a timing diagram of Fig. 9 as well as Fig. 8. First, in Fig. 8, when the manual power supply switch (not shown) is closed and the low-voltage power supply switch 12 is closed, the constant voltage control power source 25 generates a predetermined control power supply voltage Vcc. The microprocessor 21 then starts its control operation and generates a signal shown in part A of Fig. 9. A first period (slow charge period) T1, in which the repetitive command signal PLS has a logic "L," and a second period (fast discharge period) T2, in which the repetitive command signal PLS has a logic "H," have the same half-cycle value of the total cycle TO=T1+T2. However, in actuality, T1≤T2 may be satisfied.

[0163] In Fig. 8, when the logic level of the repetitive command signal PLS is “L” (or “H”), the charge / discharge switching element 31 is interrupted (or becomes conductive) as shown in part B of Fig. 9. Then, in Fig. 8, when the charge / discharge switching element 31 is interrupted, the coupling capacitor 51 from the on-board high-voltage device 60Y side is slowly charged via a series connection of the leakage resistor Rx, the rapid charge / discharge resistor 35, and the charge / discharge resistor 33, resulting in the monitor voltage Vx gradually decreasing along with the decrease in the charging current. On the other hand, when in Fig. 8 the charge / discharge switching element 31 becomes conductive, the coupling capacitor 51 is discharged from the constant voltage control power source 25 via a series connection of the rapid charge / discharge resistor 35 and the leakage resistor Rx, with the result that the monitor voltage Vx decreases rapidly.

[0164] Part C of Fig. Figure 9 shows how the monitoring voltage Vx gradually decreases or rapidly increases. The gradient of the charge / discharge characteristics becomes smoother as the charge / discharge time constants expressed in expressions (8x) and (9x) increase. τ1=(R0+Rx)×C,R0=Rs+Rq… (8y)=(8x) τ2(Rq+Rx)×C… (9y)=(9x)

[0165] In the expressions, Rs represents the resistance value of the charge / discharge resistor 33; Rq the resistance value of the fast charge / discharge resistor 35 (Rq< <Rs); R0=Rs+Rq≈Rs einen Reihenwiderstand; Rx einen parallelen kombinierten Widerstand der äquivalenten Leckwiderstände R1 und R2; C die elektrostatische Kapazität des Kopplungskondensators 51; τ1 eine langsame Ladezeitkonstante und τ2 eine Schnellentladezeitkonstante.

[0166] For example, the resistance value Rs is several hundred kΩ, while the resistance value Rq is several kΩ. When the value of the leakage resistance Rx is small, τ1>>τ2 is established between the slow charge time constant τ1 and the fast discharge time constant τ2. However, when the value of the leakage resistance Rx is sufficiently larger than the resistance value Rs, τ1≈τ2 is established, and therefore the charge / discharge time constants have little difference.

[0167] In part (C) of Fig. 9, the value of the monitoring voltage Vx at the end of the first period T1 after the slow charge is a first end voltage V1, and the value of the monitoring voltage Vx at the end of the second period T2 after the fast discharge is a second end voltage V2. The value of the second end voltage V2 must be closest to the control power supply voltage Vcc. Therefore, the fast discharge constant τ2 must be smaller than the slow charge time constant τ1, and the second period T2 is desired to be longer than the first period T1. As a result, the initial value V2 of the gradually decreasing monitoring voltage Vx is always Vcc (or a value closest to Vcc), and therefore the initial value does not fluctuate depending on the magnitude of the first end voltage V1. Note that the set threshold voltage V0 is a reduced voltage value compared to the control power supply voltage Vcc.

[0168] Part (D) of Fig. Figure 9 shows the logic state of the inverted logic signal LOW generated by the microprocessor 21. The inverted logic signal LOW has a logic level “H” in a period in which the value of the monitoring voltage Vx, which is shown in part C of Fig. 2, is equal to or lower than (Vcc-V0). The arithmetic control circuit 20B measures a transition time Tx that starts at the time of the logical inversion of the repetitive command signal PLS from "H" to "L" and continues until the logic of the inverted logic signal LOW changes from "L" to "H."

[0169] Part (E) of Fig. Figure 9 shows the waveform of the both-end voltage E of the coupling capacitor 51. A charge start voltage (=discharge end voltage) E1 is equal to a value obtained by subtracting the control power supply voltage Vcc from the stable divided voltage Vn0 expressed in expression (4y). The value of a charge end voltage (=discharge start voltage) E2 increases to Vn0 in the case where the charging period T1 is infinite. Similar to the case of Fig. 1, the first period T1 of the repetitive command signal PLS is set to be longer than a limit transition time Tx0 corresponding to the value of the transition time Tx when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 used for abnormality determination.

[0170] However, in the case of Fig. 8, the value of the transition time Tx, in which the value of the gradually changing monitoring voltage Vx transitions from a first threshold value (V2=Vcc) to a second threshold value (Vcc-V0), is measured to thereby calculate the value of the leakage resistance Rx. The following demonstrates that the calculation using the threshold voltage coefficient y of VO / Vcc is similar to the case of Fig. 1 can be carried out so that the characteristic diagrams of the Fig. 3 to 5 can be applied as they are.

[0171] The following demonstrates that the above Fig. 3 shown expression I similarly in the second embodiment of Fig. 8 is applicable. Charging period

[0172] In the slow loading period, in which the logic level of the repetitive command signal PLS in Fig. 8 is at "L" and the transistor serving as the charge / discharge switching element 31 is open, expression (10aa) is established. R0×C(dE / dt)=Vx where R0=Rs+Rq≈Rs>>Rq

[0173] Vx in expression (10aa) is substituted into expression (7y) above to obtain expression (11aa). Vn0=E+τ1×(dE / dt) where τ1=(R0+Rx)C≈(Rs+Rx)C

[0174] In the differential equation (11aa), the initial value of E at time t = 0 is represented by E1, and the value of E at infinite time is represented by Vn0. Then, a solution expressed by expression (12sa) is obtained. E=E1exp(−t / τ1)+Vn0{1−exp(−t / τ1)}

[0175] In addition, the value of (dE / dt) in expression (10aa) is substituted into expression (7y) above to obtain expression (13aa). Vx / (R0 / / Rx)=(Vn0−E) / Rx where R0 / / Rx represents the parallel combined resistance of R0 and Rx. Unloading period

[0176] In the discharge period in which the logic level of the repetitive command signal PLS in Fig. 8 is at "H" and the transistor serving as the charge / discharge switching element 31 is closed, expression (10bb) is established. C×Rq(dE / dt)=Vx−Vcc

[0177] Vx in expression (10bb) is substituted into expression (7y) above to obtain expression (11bb). Vn0−Vcc=τ2(dE / dt)+E where τ2=(Rq+Rx)×C

[0178] In differential equation (11bb), the initial value of E at time t=0 is represented by E2, and the value of E at infinity is represented by Vn0-Vcc. Then, a solution expressed by expression (12bb) is obtained. E=E2exp(−t / τ2)+(Vn0−Vcc){1−exp(−t / τ2)}

[0179] In addition, the value of (dE / dt) in expression (10bb) is substituted into expression (7y) above to obtain expression (13bb). Vx / (Rq / / Rx)=(E−Vn0) / Rx+Vcc / Rq where Rq / / Rx represents the parallel combined resistance of Rq and Rx. Stable condition

[0180] In the stable state in which the logic level of the repetitive command signal PLS in Fig. 8 is alternately inverted into the charging period T1 and the discharging period T2, and there is no change in the value of the leakage resistance Rx and the output Vh of the high-voltage DC power source 61, the initial values ​​E1 and E2 in expressions (12aa) and (12bb) are calculated as follows. First, in expression (12aa), at the charging end time t=T1, E=E2 is established, and therefore expression (14aa) is established. E2=E1×K1+Vn0(1−K1) where K1=exp(−T1 / τ1)

[0181] In addition, in expression (12bb), at the discharge end time t=T2, E=E1 is established and therefore expression (14bb) is established. E1=E2×K2+(Vn0−Vcc)(1−K2) where K2=exp(−T2 / τ2)

[0182] From the expressions (14aa) and (14bb) the expressions (15aa) and (15bb) are obtained. Vn0−E1=Vcc×K0 Vn0−E2=Vcc×K0×K1 where K0=(1−K2) / (1−K1×K2) In the case of the second embodiment

[0183] In Fig. 8 and Fig. 9, at the start of charging time t=0, Vx1=V2≈Vcc is established. At time Tx=0 to T1, if the value of Vx becomes equal to the set threshold voltage (Vcc-V0), Expression (19a) for time t=Tx is calculated from Expression (12aa). Note that a rapid discharge is performed during the discharge period, and therefore, E1=Vn0-Vcc is established, and Kx=exp(-Tx / τ1) is established. Ex =(Vn0 - Vcc)exp(-Tx / τ1)+Vn0{1−exp(−Tx / τ1)}=Vn0 - VccKx+

[0184] Expression (20a) is obtained from expressions (13aa) and (19a). (Vcc - V0) / (R0 / / Rx)=(Vn0-Ex) / Rx=VccKx / Rx

[0185] Then, the threshold voltage coefficient γ of VO / Vcc, the leakage resistance coefficient β of Rx / RO, and the transition time coefficient α of Tx / (R0C) are substituted into Expression (20a), and Expression (21) is obtained. γ=(1+β−Kx) / (1+β)∴Kx = (1+β)×(1−γ)

[0186] Ex=Exp [-Tx / {(R0+Rx)C}]=exp{-α / (1+β)} is established and therefore expression (22) is obtained by a natural logarithm transformation. α=(1+β)LOGe(1 / Kx)

[0187] Expressions (21) and (22) correspond to the above in Fig. Expression I shown in Figure 3.

[0188] Next, the action and operation of the device is described by Fig. 8 detailed with reference to the Fig. 10 and Fig. 11 illustrated operational explanation flow diagrams, mainly pointing out differences compared to Fig. 6 and Fig. 7. Note that the step numbers of the Fig. 6 and Fig. 7 are in the thousands, while the step numbers of the Fig. 10 and Fig. 11 are in the two thousandths, and the same three digits or lower numbers represent the same or equivalent parts. However, with respect to the equivalent part, the different parts requiring additional description are described below.

[0189] Referring to the first flowchart of Fig. 10, in step 2605, which is executed after step 2603a or step 2603b, the analog signal voltage ANL input to the microprocessor 21 is monitored, and it is determined whether the value of the monitor voltage Vx is within the proper range of 0 to Vcc. If the value of the monitor voltage Vx is outside the proper range, it is determined as NO, and the flow proceeds to step 2606a. If the value of the monitor voltage Vx is within the proper range, it is determined as YES, and the flow proceeds to step 2606c.

[0190] Next, referring to the second flowchart of Fig. 11, step 2700a is a step of determining whether a calibration operation is to be performed. If a calibrated value of the electrostatic capacitance C of the coupling capacitor 51 is measured at the time of the delivery inspection of the leakage resistance detection device 50 or immediately after the low-voltage power supply switch 12 is closed to supply the leakage resistance detection device 50B at the time of actual operation of the leakage resistance detection device 50B, this is determined as YES and the flow proceeds to step 2700c. After the calibration operation is completed, this is determined as NO and the flow proceeds to step 2700b.

[0191] Step 2700c is a step serving as operation calibration means in which a test flag is set in the case of a delivery adjustment operation, while a test operation command TST is generated at the start of actual operation, and the flow proceeds to step 2701. Step 2700b is a step in which the test flag generated in step 2700c is reset and the test operation command TST is suspended, and the flow proceeds to step 2701.

[0192] In the case of the delivery adjustment operation, the arithmetic control circuit 20B measures the leakage resistance Rx in step 2706 to be described below in the state where the coupling capacitor 51 and the on-board high-voltage device 60Y are not connected to each other. The arithmetic control circuit 20B calibrates and stores the value of the electrostatic capacitance C of the coupling capacitor 51 so that the obtained result can become the additional resistance R3 of the bypass leakage resistor 53.

[0193] At the start of the actual operation, the arithmetic control circuit 20B temporarily closes the test switching element 55 to start the operation and determines whether the resistance abnormality determination output ER1 is momentarily activated or the pre-alarm output ER2 is momentarily activated when the leakage resistance Rx becomes equal to or less than the preliminary leakage resistance Rxm, thereby inspecting whether the operation of detecting the leakage resistance Rx can be performed normally. If the inspection result shows that the detection operation is abnormal, the resistance abnormality determination output ER1 or the pre-alarm output ER2 is generated continuously or intermittently.When the inspection result shows that the detection operation is normal, the output generation is suspended for such a short period of time that the abnormality alarm device 19 to be driven by the resistance abnormality determination output ER1 or the pre-alarm output ER2 does not operate.

[0194] Furthermore, the arithmetic control circuit 20B closes the test switching element 55 at the start of operation to measure the value of a first combined leakage resistance in the state where the test leakage resistance 54, the bypass leakage resistance 53, and the leakage resistance Rx are connected in parallel, and measures the value of a second combined leakage resistance in the state where the test switching element 55 is open and the bypass leakage resistance 53 and the leakage resistance Rx are connected in parallel. The arithmetic control circuit 20B recalculates the electrostatic capacitance C of the coupling capacitor 61 based on the values ​​of the first and second leakage resistances and stores the calculated electrostatic capacitance C as a calibrated value.The calibrated value is calculated so that the actual leakage resistances Rx in the open circuit and closed circuit states of the test switching element 55 can have the same value, due to the calibrated and stored electrostatic capacitance C.

[0195] Step 2703 following step 2702 is a determination step of determining whether the logic level of the internally generated inverted logic signal LOW has changed from "L" to "H" when the value of the analog signal voltage ANL has decreased to be equal to or lower than a predetermined set voltage (Vcc-V0). In a case where the logic level of the inverted logic signal LOW has changed when the value of the analog signal voltage ANL has become equal to or lower than the set voltage (Vcc-V0), it is determined as YES, and the flow proceeds to step 2704a. In the case where the logic level of the inverted logic signal LOW has changed when the value of the analog signal voltage ANL is higher than the set voltage (Vcc-V0), it is determined as NO, and the flow proceeds to step 2704b.

[0196] Note that the calculation of the leakage resistance Rx in step 2704 is performed in three ways. First, the calculation is performed based on the characteristic curve whose entire area is the negative gradient curve region, as shown by the curve 103 of Fig. 3. Secondly, the calculation is performed using the negative gradient curve region of a bell-shaped curve, as represented by the curve 112 of Fig. 3. Thirdly, the calculation is performed using the positive gradient curve region of a bell-shaped curve, as represented by curve 113 of Fig. 3. In any case, the value of the leakage resistance Rx is calculated from the measured value of the transition time Tx based on a functional formula or a data table between the transition time coefficient α and the leakage resistance coefficient β, using the threshold voltage coefficient y = VO / Vcc as a parameter. However, in the second embodiment, the set threshold voltage V0 is a voltage value lower than the control power supply voltage Vcc and is the set voltage (Vcc-V0).

[0197] As an alternative abnormality determination method, instead of comparing the leakage resistance Rx with the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, the transition time Txn or the limit transition time Tx0 may be calculated in advance according to the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, and the actual transition time Tx may be compared with the preliminary transition time Txn or the limit transition time Tx0.

[0198] Furthermore, in the case where the value of the reference time constant R0×C, which is a dominant factor of the transition time coefficient α, the value of the series resistance R0, which is the dominant factor of the leakage resistance coefficient β, or the value of the threshold voltage coefficient γ is considered a fixed value, a formula or a data table between the transition time Tx and the leakage resistance Rx can be directly used as a functional formula or the data table. In this case, the leakage resistance Rx can be directly calculated from the measured transition time Tx without performing complicated calculation processing during operation.

[0199] However, in the case of varying the value of the threshold voltage coefficient y, the denominator value of the transition time coefficient α, or the denominator value of the leakage resistance coefficient β based on the applied vehicle types or the operating condition, it is advantageous to set the functional formula or the data table using α, β, or γ, which are unitless index values. Alternatively, in a case where the arithmetic control circuit 20B does not have an analog input terminal or an analog input terminal is occupied, which is Fig. 1 may be used instead of the operational amplifier 49. In this case, a signal equivalent to the inverted logic signal LOW may be input to the microprocessor 21.

[0200] In addition, in the embodiment of Fig. 8, the connection point B is connected to the positive-side power supply line 68 of the on-board high-voltage device 60X, but even if the connection point B is connected to a negative-side power supply line, no change occurs in Expression I. Similarly, in the first embodiment of Fig. 1, the connection point B is connected to the negative-side power supply line 67 with the on-board high-voltage device 60X, but even if the connection point B is connected to a positive-side power supply line, no change occurs in Expression I.

[0201] As is clear from the above description, according to the second embodiment of the present invention, the leakage resistance detection device 50B for an on-board high-voltage device is provided, which is connectable to the on-board high-voltage device 60Y including the on-board high-voltage DC power source 61 and the high-voltage electrical load 65 to be fed and driven by the on-board high-voltage DC power source 61. The on-board high-voltage device 60Y has the leakage resistance Rx with respect to the vehicle body 11, as typified by the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side. The leakage resistance detection device 50B includes the constant-voltage control power source 25 fed and driven by the low-voltage DC power source 10, which has a negative terminal connected to the vehicle body 11.and which generates the control power supply voltage Vcc, the leakage resistance detection device 50B is configured to measure a value of the leakage resistance Rx via the coupling capacitor 51 having a terminal B connected to a predetermined position of the on-board high-voltage device 60Y, wherein the leakage resistance detection device 50B further includes the repetitive signal output circuit 30B, the monitoring signal processing circuit 40B, and the arithmetic control circuit 20B, wherein the arithmetic control circuit 20B includes the microprocessor 21 and the program memory 24B, which cooperate with each other.

[0202] Further, in response to the switching operation of the charge / discharge switching element 31 operating in response to the repetitive command signal PLS, the repetitive signal output circuit 30B alternately repeats a charge period and a discharge period in which the measurement point A is intermittently connected to the control power supply voltage Vcc via the charge / discharge resistor 33 or 35, the measurement point A being another terminal of the coupling capacitor 51, to thereby gradually increase or gradually decrease the monitor voltage Vx, which is a potential between the measurement point A and the vehicle body 11.The monitor signal processing circuit 40B is configured to: generate the inverted logic signal LOW when a value of the monitor voltage Vx, which gradually increases or gradually decreases with a gradient that becomes smoother as a charge / discharge time constant becomes longer, passes a predetermined threshold voltage, and input the inverted logic signal LOW to the arithmetic control circuit 20B; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20B.The arithmetic control circuit 20B is configured to: measure the transition time Tx while the value of the monitor voltage Vx changes from one threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to another; calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and generate the resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than the predetermined limit leakage resistance Rx0, or when the transition time Tx reaches the limit transition time Tx0 corresponding to the limit leakage resistance Rx0.The repetitive command signal PLS is a pulse signal with a fixed cycle having the charging period or discharging period longer than at least the boundary transition time Tx0.

[0203] The repetitive signal output circuit 30B is configured, in response to the switching operation of the charge / discharge switching element 31, to connect the measuring point A between the other terminal of the coupling capacitor 51 to the output terminal of the constant-voltage control power source 25 via the rapid charge / discharge resistor 35 having the resistance value Rq, or to connect the measuring point A to the negative terminal corresponding to the vehicle body potential via the reference resistor 33 having the reference value Rs sufficiently larger than the resistance value Rq, to ​​thereby rapidly increase or gradually decrease the monitor voltage Vx, which is the potential between the measuring point A and the vehicle body 11.The monitor signal processing circuit 40B is configured to: generate the inverted logic signal LOW when the value of the monitor voltage Vx, which gradually decreases at a gradient that becomes gentler as the time constant (Rx+Rx)C, which is a product of a sum of the resistance value Rs (Rs>>Rq) and the leakage resistance Rx with the electrostatic capacitance C of the coupling capacitor 51, increases, has decreased to pass a value obtained by subtracting the predetermined threshold voltage V0 from the control power supply voltage Vcc, and input the inverted logic signal LOW to the arithmetic control circuit 20B; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20B.The arithmetic control circuit 20B inverts an output of the repetitive command signal PLS, which is a pulse train signal having a fixed cycle, in a state where the value of the monitor voltage Vx has abruptly increased to approach the control power supply voltage Vcc, through the auxiliary switching element 31, and inverts the output of the repetitive command signal PLS after receiving the inverted logic signal LOW as an input or after measuring, as the transition time Tx, a period from when the inverted logic signal LOW is generated based on the input analog signal voltage ANL to when the inverted logic signal LOW is obtained.

[0204] As described above, the repetitive signal output circuit charges and discharges the coupling capacitor in conjunction with claim 4 of the present invention, and the arithmetic control circuit measures, as the transition time Tx, a period during which the monitor voltage Vx rises to a predetermined threshold voltage (Vcc-VO) from a threshold voltage set to a value close to the control power supply voltage Vcc. Therefore, the comparator determination circuit or the comparator determination processing with respect to one of the pair of threshold voltages can be omitted, and the transition time can be measured by a simple structure.

[0205] As is clear from the above description, according to the second embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistances 33 and 35 provided in the repetitive signal output circuit 30B, is represented by R0, and when a period,during which the monitoring voltage Vx gradually decreases from a state of the control power supply voltage Vcc to the predetermined threshold voltage (Vcc-V0) represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0 ×C), the leakage resistance coefficient β is set to Rx / RO, and the threshold voltage coefficient y is set to VO / Vcc or (V2-V1) / Vcc; Calculating, by the arithmetic control circuit 20B, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases,along with an increase in the leakage resistance coefficient β; selectively using a negative gradient curve region following the positive gradient curve region or a negative gradient curve region occupying an entire region, and determining the threshold voltage coefficient γ such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the negative gradient curve region; determining, when the transition time Tx has become a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the short-circuit transition time Tx00 corresponding to a case where the limit leakage resistance Rx becomes 0.

[0206] As is clear from the above description, according to the second embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30B, is represented by R0, and when a period,during which the monitoring voltage Vx gradually decreases from a state of the control power supply voltage Vcc to a predetermined threshold voltage (Vcc-V0) represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient y is set to V0 / Vcc: calculating, by the arithmetic control circuit 20B, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient y as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region,in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; excluding, in advance, by the characteristic curve correcting means, the application of a negative gradient curve region following the positive gradient curve region; determining the threshold voltage coefficient γ so that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the corrected positive gradient curve region; determining, when the transition time Tx has become a value equal to or less than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1, and performing the time-off abnormality determination ERR1 when the transition time Tx exceeds the open-circuit transition time Txm, which corresponds to a case,where the leakage resistance Rx is infinite.,

[0207] The leakage resistance detection device further includes the test leakage resistor 54 and the test switching element 55 connected between the connection point B between the coupling capacitor 51 and the on-board high-voltage device 60Y and the vehicle body 11. The test leakage resistor 54 has a resistance value equal to or less than the limit leakage resistance Rx0 or a resistance value equal to or less than the preliminary leakage resistance Rxn, which is close to the limit leakage resistance Rx0 and greater than the limit leakage resistance Rx0, and the test switching element 55 is connected in series with the test leakage resistor 54. The leakage resistance detection method further includes, by the arithmetic control circuit 20B: temporarily closing the test switching element 55 at the start of operation and determining whether the resistance abnormality determination output ER1 operates momentarily or the pre-alarm output ER2 is generated.When the leakage resistance Rx has become equal to or smaller than the preliminary leakage resistance Rxn, it operates for a moment to inspect whether an operation of detecting the leakage resistance Rx is being performed normally; and when an inspection result shows that the detection operation is abnormal, generating the resistance abnormality determination output ER1 or the pre-alarm output ER2 continuously or intermittently; and when the inspection result shows that the detection operation is normal, suspending the output generation for such a short period that the abnormality alarm device 19 to be driven by the resistance abnormality determination output ER1 or the pre-alarm output ER2 does not operate.

[0208] As described above, in connection with claim 16 of the present invention, the leakage resistance detection device further includes the test leakage resistance temporarily connected through the test switching element so that the arithmetic control circuit performs a preliminary inspection related to the detection operation of the leakage resistance.

[0209] Therefore, at the start of operation, it is possible to inspect whether the resistance abnormality determination output or the pre-alarm output, which is never activated if the leakage resistance is normal, is generated normally. If the inspection reveals an abnormality, the abnormality can be notified through the resistance abnormality determination output or the pre-alarm output.

[0210] The leakage resistance detection device further includes the bypass leakage resistance 53 connected between the connection point B between the coupling capacitor 51 and the on-board high-voltage device 60Y, and the vehicle body 11. The leakage resistance detection method further includes, by the arithmetic control circuit 20B: closing the test switching element 55 at the start of operation and measuring a first combined leakage resistance value in a state where the test leakage resistance 54, the bypass leakage resistance 53, and an actual leakage resistance Rx are connected in parallel, and measuring a second combined leakage resistance value in a state where the test switching element 55 is open and the bypass leakage resistance 53 and the actual leakage resistance Rx are connected in parallel;and recalculating the electrostatic capacitance C of the coupling capacitor 51 from the first combined leakage resistance value and the second combined leakage resistance value and storing the recalculated electrostatic capacitance C as a calibrated value, wherein the calibrated value is calculated such that the actual leakage resistances Rx in open-circuit and closed-circuit states of the test switching element 55 have the same values ​​due to the calibrated and stored electrostatic capacitance C.;

[0211] As described above, in conjunction with claim 17 of the present invention, the leakage resistance detection device further includes the bypass leakage resistance and the test leakage resistance. Two types of combined leakage resistance are detected in the state where the test leakage resistance is connected or opened by the test switching element, thereby calculating and calibrating the electrostatic capacitance of the coupling capacitor.

[0212] Therefore, the electrostatic capacitance calibration can be performed at each start-up in the state where the onboard high-voltage device is connected to an unknown leakage resistance. Thus, the correct electrostatic capacitance can always be determined with respect to individual fluctuations in the electrostatic capacitance from initial items and fluctuations caused by secular changes in the electrostatic capacitance. Accordingly, the leakage resistance can be accurately measured. Third embodiment

[0213] Referring to Fig. 12, which is an overall configuration diagram of an apparatus of a third embodiment of the present invention, mainly differences from Fig. 1 below. In the drawings, the same reference numerals designate the same or equivalent parts. In Fig. 12, a leakage resistance detection device 50C includes an arithmetic control circuit 20C, a repetitive signal output circuit 30C, and a monitor signal processing circuit 40C, and detects a leakage resistance of the onboard high voltage device 60X.

[0214] The first major difference is that the repetitive signal output circuit 30C includes an auxiliary switching element 31 such as a PNP transistor, and when the output logic level of the repetitive command signal PLS is "H", an auxiliary switching element 31a becomes conductive via an auxiliary drive resistor 32a so that the auxiliary switching element 31 becomes conductive via the drive resistor 32, and when the output logic level of the repetitive command signal PLS is "L", the auxiliary switching element 31a becomes non-conductive through an auxiliary stabilizing resistor 34a so that the auxiliary switching element 31 becomes non-conductive through an open-circuit stabilizing resistor 34.

[0215] The second main difference is that the repetitive command signal PLS uses an output signal of an inversion memory circuit 52 provided in the monitor signal processing circuit 40C and is a pulse train signal having a variable cycle, as described below with reference to Fig. 13 described.

[0216] The third main difference is that the monitor signal processing circuit 40C includes first and second comparators 41a and 41b and the inversion memory circuit 52 instead of the comparator 41.

[0217] The first comparator 41a has a positive input terminal connected to voltage divider resistors 42a and 43a. The voltage divider resistors 42a and 43a divide the control power supply voltage Vcc to generate a first threshold voltage V1. The first comparator 41a has a negative input terminal connected to a noise filter circuit formed of a smoothing resistor 45 and a smoothing resistor 46 via an input resistor 44a. The smoothing resistor 45 has one terminal connected to a monitor voltage Vx.

[0218] The second comparator 41b has a negative input terminal connected to voltage divider resistors 42b and 43b. The voltage divider resistors 42b and 43b divide the control power supply voltage Vcc to generate a second threshold voltage V2. The second comparator 41b has a positive input terminal connected to the noise filter circuit formed by the smoothing resistor 45 and the smoothing capacitor 46 via an input resistor 44b.

[0219] The first comparator 41a has an output logic level "H" when the monitor voltage Vx is equal to or less than the first threshold voltage V1. The output logic level "H" is input to the arithmetic control circuit 20C as a first inverted logic signal LOW. The first comparator 41a further provides a "SET" input to the inversion storage circuit 52, such as a flip-flop circuit.

[0220] The second comparator 41b has an output logic level "H" when the monitor voltage Vx is equal to or greater than the second threshold voltage V2. The output logic level "H" is input to the arithmetic control circuit 20C as a second inverted logic signal HIG. The first comparator 41a further provides a RESET input to the inversion memory circuit 52, such as a flip-flop circuit.

[0221] A SET output of the inversion memory circuit 52 is input to the repetitive signal output circuit 30C as a repetitive command signal PLS and is further input to the arithmetic control circuit 20C as an inverted logic signal ANS serving as a tape history logic signal.

[0222] A microprocessor 21 provided in the arithmetic control circuit 20C cooperates with a program memory 24C and measures the value of a leakage resistance Rx. The microprocessor 21 generates a pre-alarm output ER2 when the measured value of the leakage resistance Rx becomes equal to or less than a predetermined preliminary leakage resistance Rxn, and generates a resistance abnormality determination output ER1 when the measured value becomes equal to or less than a limit leakage resistance Rx0, which are input to the abnormality alarm device 19.

[0223] Note that similar to Fig. 1, the low-voltage DC power source 10, the low-voltage power supply switch 12, the vehicle state signal 18, and the abnormality alarm 19 are connected outside the leakage resistance detection device 50C, and the constant-voltage control power source 25 which generates the control power supply voltage Vcc is provided in the leakage resistance detection device 50C.

[0224] The following describes the effect and operation of the device according to the third embodiment of the present invention as shown in Fig. 12, briefly referring to a timing diagram of Fig. 13 as well as Fig. 12. First created in Fig. 12, when the manual power supply switch (not shown) is closed and the low-voltage power supply switch 12 is closed, the constant-voltage control power source 25 supplies a predetermined control power supply voltage Vcc. The microprocessor 21 then starts its control operation.

[0225] Part (A) of Fig. 13 shows the waveform of the repetitive command signal PLS generated by the monitoring signal processing circuit 40C. A first period (slow discharge period) T1, in which the repetitive command signal PLS has a logic "L," and a second period (fast discharge period) T2, in which the repetitive command signal PLS has a logic "H," have different values, each occupying a portion of the total cycle T0=T1+T2.

[0226] In Fig. 12, when the logic level of the repetitive command signal PLS is “L” (or “H”), the auxiliary switching element 31 is interrupted (or becomes conductive) as shown in part (B) of Fig. 13. Then in Fig. 12, when the auxiliary switching element 31 is interrupted, the coupling capacitor 51 is slowly discharged via a series connection of the leakage resistor Rx, the rapid charge / discharge resistor 35, and the charge / discharge resistor 33, with the result that the monitoring voltage Vx gradually decreases along with the decrease in the discharge current. On the other hand, in Fig. 12, when the auxiliary switching element 31 becomes conductive, the coupling capacitor 51 is rapidly charged from the constant voltage control power source 25 via a series connection of the rapid charge / discharge resistor 35 and the leakage resistor Rx, with the result that the monitoring voltage Vx rises rapidly.

[0227] Part (C) of Fig. Figure 13 shows how the monitoring voltage Vx decreases gradually or increases rapidly. The gradient of the charge / discharge characteristics becomes smoother as the charge / discharge time constants expressed in expressions (8x) and (9x) above increase. τ1=(R0+Rx)C is a slow discharge time constant, and τ2=(Rq+Rx)C is a fast discharge time constant.

[0228] In part (C) of Fig. 13, the value of the slowly discharged monitoring voltage Vx gradually decreases to the first final voltage V1 at the end of the first period T1. At this time, the logic level of the first inverted logic signal LOW, which is the output of the first comparator 41a, changes from "L" to "H," and the logic level of the SET output of the inversion memory circuit 52 becomes "H." As a result, the auxiliary switching element 31 becomes conductive, starting the rapid charging of the coupling capacitor 51.

[0229] The value of the rapid charging monitor voltage Vx rapidly rises to the second end voltage V2, which is a value substantially equal to the control power supply voltage Vcc at the end of the second period T2. At this time, the logic level of the second inverted logic signal HIG, which is the output of the second comparator 41b, changes from "L" to "H," and the output level of the SET output of the inversion memory circuit 52 becomes "L." As a result, the auxiliary switching element 31 is interrupted, starting the slow discharge of the coupling capacitor 51.

[0230] Note that in order to avoid waiting for the monitor voltage Vx to completely converge to the control power supply voltage Vcc at the end of the second period T2, a residual error ΔV of about Vcc-V2-0.03 Vcc is added to the set threshold voltage V0, which is the value obtained by subtracting the first end voltage V1 from the control power supply voltage Vcc.

[0231] Part (D) of Fig. Figure 13 shows the output waveform of the first inverted logic signal LOW. Part (E) of Fig. 13 shows the output waveform of the second inverted logic signal HIG. Part (F) of Fig. Figure 13 shows the generated waveform of the inverted logic signal ANS, which is the output of the inversion memory circuit 52. The inverted logic signal ANS and the repetitive command signal PLS are the same signal. A transition time Tx is a period during which the logic level of the inverted logic signal ANS is "L" and the monitor voltage Vx gradually decreases from the control power supply voltage Vcc to a voltage corresponding to the set threshold voltage V0 higher than the first end voltage V1.

[0232] Part (J) of Fig. Figure 13 shows the waveform of the two-ended voltage E of the coupling capacitor 51. A discharge start voltage (= charge end voltage) E1 is equal to a value obtained by adding the control power supply voltage Vcc to the stable divided voltage Vn0 expressed in expression (4x). The value of a discharge end voltage (= charge start voltage) E2 decreases to Vn0 in the case where the discharge period T1 is infinite.

[0233] The following shows that the calculation using the threshold voltage coefficient y of VO / Vcc is similar to the case of Fig. 1 can be carried out so that the characteristic diagrams of Fig. 3 to 5 can be applied as they are. Note that the generation frequency of the repetitive command signal PLS is a fixed value in the first and second embodiments, but in the third embodiment, the first period T1 and the second period T2 are automatically changed according to the value of the leakage resistance Rx so that the next repetitive command signal PLS can be generated quickly after the completion of the measurement of the transition time Tx. Furthermore, the transition to the next first period T1 is performed after confirming that the monitor voltage Vx has risen to a value close to the control power supply voltage Vcc, and thus, it is unnecessary to provide a delay time to the second period T2 with respect to the second end voltage V2, unlike part (C) of Fig. 9.

[0234] The following shows that the above Fig. 3 shown expression I similarly in the third embodiment of Fig. 12 is applicable. Discharge period

[0235] In the slow discharge period, in which the logic level of the repetitive command signal PLS in Fig. 12 is “L” and the transistor serving as auxiliary switching element 31 is open, the expression (30b) is established. Vx + R0x(dE / dt)=0 where R0 = Rs + Rq≈Rs>>Rq

[0236] Vx in expression (30b) is substituted into expression (7x) above to obtain expression (31b) Vn0 = E+τ1×(dE / dt) where τ1=(R0 + Rx)C≈(Rs + Rx)C

[0237] In differential equation (31b), the initial value of E at time t = 0 is represented by E1, and the value of E at infinite time is represented by Vn0. Then, a solution expressed by expression (32b) is obtained. E = E1exp(−t / τ1)+Vn0{1−exp(−t / τ1)}

[0238] In addition, the value of (dE / dt) in expression (30b) is substituted into expression (7x) above to obtain expression (33b). Vx / (R0 / / Rx)=(E - Vn0) / Rx where R0 / / Rx represents the parallel combined resistance of R0 and Rx. Charging period

[0239] In the fast charging period, in which the logic level of the repetitive command signal PLS in Fig. 12 is “H” and the transistor serving as auxiliary switching element 31 is closed, expression (30a) is established. C×Rq(dE / dt)=Vcc - Vx

[0240] Vx in expression (30a) is substituted into expression (7x) above to obtain expression (31a). Vn0 + Vcc = τ2(dE / dt)+E where τ2 = (Rq + Rx)×C

[0241] In differential equation (31a), the initial value of E at time t=0 is represented by E2, and the value of E at infinite time is represented by Vn0+Vcc. Then, a solution expressed by expression (32a) is obtained. E = E2exp(-t / τ2)+Vn0+Vcc){1−exp(-t / τ2)}

[0242] In addition, the value of (dE / dt) in expression (30a) is substituted into expression (7x) above to obtain expression (33a). Vx / (Rq / / Rx)=(E - Vn0) / Rx+Vcc / Rq where Rq / / Rx represents the parallel combined resistance of Rq and Rx. Stable condition

[0243] In the stable state in which the logic level of the repetitive command signal PLS in Fig. 12 is alternately inverted in the discharge period T1 and the charge period T2, and there is no change in the value of the leakage resistance Rx and the output voltage Vh of the high-voltage DC power source 61, the initial values ​​E1 and E2 in expressions (32b) and (32a) are calculated as follows. First, in expression (32b), at the discharge end time t=T1, E=E2 is established, and therefore expression (34b) is established. E2 = E1×K1+Vn0(1−K1) where K1=exp(−T1 / τ1)

[0244] In addition, in expression (32a), at the discharge end time T=T2, E=E1 is established and therefore expression (34a) is established. E1=E2×K2+(Vn0+Vcc)(1−K2) where K2=exp(−T2 / τ2)

[0245] From expressions (34a) and (34b) the expressions (35a) and (35b) are obtained. E1−Vn0=Vcc×K0 E2−Vn0=Vcc×K0×K1 where K0=(1−K2) / (1−K1×K2) In the case of the third embodiment

[0246] In the Fig. 12 and Fig. 13, at the discharge start time t=0, Vx2=V2≈Vcc is established. At time Tx=T1, if the value of Vx becomes equal to the set threshold voltage V1P(Vcc-V0), Expression (39b) for time t=Tx is calculated from Expression (32b). Note that the value of the discharge start voltage E1 is Vn0+VccKo≈Vn0+Vcc by Expression (35b), and Kx=exp(-Tx / τ1)=k1 is established. Ex=(Vn0+Vcc)exp(−Tx / τ1)+Vn0{1−exp(−Tx / τ1)}=Vn0+VccKx

[0247] Expression (40b) is obtained by expressions (33b) and (39b). (Vcc−V0) / (R0 / / Rx)=(Ex−Vn0) / Rx =VccKx / Rx

[0248] Then, the threshold voltage coefficient γ of V0 / Vcc, the leakage resistance coefficient β of Rx / R0 and the transition time coefficient α of Tx / (R0C) are substituted into Expression (40b) and Expression (41) is obtained. γ=(1+β−Kx) / (1+β)∴Kx=(1+β)×(1−γ)

[0249] Kx=exp[-Tx / {(R0+Rx)C}]=exp{-α / (1+β)} is established and therefore expression (42) is obtained by natural logarithm transformation. α=(1+β)LOGe(1 / Kx)

[0250] Expressions (41) and (42) correspond to the above in Fig. Expression I shown in Figure 3.

[0251] Next, the effect and operation of the device will be described by Fig. 12 in detail with reference to the Fig. 14 and Fig. 15 illustrated operational explanation flow charts are described in detail, mainly focusing on differences compared to the Fig. 6 and Fig. 7. Note that the step numbers of the Fig. 6 and Fig. 7 are in the 1000s, while the step numbers of Fig. 14 and Fig. 15 are in the 3000s, and the same three-digit or fewer numbers represent the same or equivalent parts. However, regarding the equivalent parts, the differing parts that require additional description are described below.

[0252] Referring to the first flowchart of Fig. In Fig. 14, step 3604 indicated by a dashed line is a step necessary for generating the repetitive command signal PLS from the arithmetic control circuit 20C by causing the microprocessor 21 to execute logic processing equivalent to the first comparator 41a, the second comparator 41b, and the inversion memory circuit 52 illustrated in the monitor signal processing circuit 40C in the state where the analog signal voltage ANL proportional to the monitor voltage Vx is input to the arithmetic control circuit 20C. In this case, the monitor signal processing circuit 40C only needs to include the noise filter constituted by the smoothing resistor 45 and the smoothing capacitor 56, and an operational amplifier.

[0253] A subsequent step block 3605 includes step 3611 and step 3612. Step 3611 is a determination step of monitoring the logic level of the first inverted logic signal LOW. If the monitoring voltage Vx is equal to or higher than the first threshold V1, this is determined as YES and the flow proceeds to step 3612. If the monitoring voltage Vx is lower than the first threshold V1, this is determined as NO and the flow proceeds to step 3606a.

[0254] Step 3612 is a determination step of monitoring the logic level of the second inverted logic signal HIG. If the monitor voltage Vx is equal to or lower than the second threshold V2, it is determined as YES, and the flow proceeds to step 3606c. If the monitor voltage Vx exceeds the second threshold V2, it is determined as NO, and the flow proceeds to step 3606a. Therefore, if the monitor voltage Vx is within the proper range of the first threshold V1 or higher and the second threshold V2 or lower, the flow proceeds to step 3606c, and if the monitor voltage Vx deviates from the range V1 to V2, the flow proceeds to step 3606a.

[0255] Next, referring to the second flowchart of Fig. 15, step 3700 is a step corresponding to steps 2700a to 2700c of the above-described Fig. 11 equivalent step block. Step 3700 is a factory calibration means for measuring the value of the electrostatic capacitance C of the coupling capacitor 51 at the time of factory calibration and storing a calibrated value thereof.

[0256] Following step 3701 is a determination step of monitoring the logic level of the second inverted logic signal HIG. If the logic level is "H," this is determined as YES, and the flow proceeds to step 3702. If the logic level is "L," this is determined as NO, and the flow proceeds to step 3608 of Fig. 14 via a relay terminal B.

[0257] Note that when YES is determined in step 3701, the logic level of the repetitive command signal PLS is "L" and the charge / discharge switching element 31 becomes non-conductive, and therefore the monitor voltage Vx starts to gradually decrease, and its transition time is measured in the subsequent step 3702.

[0258] The subsequent step 3703 is a determination step of monitoring the logic level of the first inverted logic signal LOW. If the logic level is "H," it is determined as YES, and the flow proceeds to step 3704a. If the logic level is "L," it is determined as NO, and the flow proceeds to step 3704b. Note that if it is determined as YES in step 3703, the logic level of the repetitive command signal PLS is "H," and the charge / discharge switching element 31 becomes non-conductive. Therefore, the monitor voltage Vx begins to rise rapidly, and its transition time Tx is read out and stored in the subsequent step 3704a.

[0259] As described above, in steps 3701 and 3703, the start timing of the gradual decrease of the monitor voltage Vx and the completion timing of the gradual decrease are determined, respectively. Instead of using the first inverted logic signal HIG and the second inverted logic signal LOW, an inverted logic signal ANS serving as a band history logic signal may be used.

[0260] Note that the calculation of the leakage resistance Rx in step 3705 is performed in three ways. First, the calculation is performed based on the characteristic curve whose overall region is the negative gradient curve region, as shown by the curve 103 of Fig. 3. Secondly, the calculation is performed using the negative gradient curve region of a bell-shaped curve, as represented by the curve 112 of Fig. 3. Thirdly, the calculation is performed using the positive gradient curve region of a bell-shaped curve as represented by curve 113 of Fig. 3. In any case, the value of the leakage resistance Rx is calculated from the measured value of the transition time Tx based on a functional formula or a data table between the transition time coefficient α and the leakage resistance coefficient β, using the threshold voltage coefficient γ=V0 / Vcc as a parameter. However, in the third embodiment, the set threshold voltage V0 is a voltage value lower than the control power supply voltage Vcc.

[0261] As an alternative abnormality determination method, instead of comparing the leakage resistance Rx with the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, the limit transition time Tx0 or the limit transition time Tx0 may be calculated in advance according to the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, and the actual transition time Tx may be compared with the preliminary transition time Txn or the limit transition time Tx0.

[0262] Furthermore, in the case where the value of the reference time constant R0×C, which is the denominator of the transition time coefficient α, the value of the series resistance R0, which is the denominator of the leakage resistance coefficient β, or the value of the threshold voltage coefficient γ is treated as a fixed value, a formula or a data table between the transition time Tx and the leakage resistance Rx can be directly used as the functional formula or data table. In this case, the leakage resistance Rx can be directly calculated from the measured transition time Tx without performing complicated calculation processing during operation.

[0263] However, in the case of varying the value of the threshold voltage coefficient γ, the denominator value of the transition time coefficient α, or the denominator value of the leakage resistance coefficient β according to the applied types of vehicle or the operating condition, it is advantageous to set the functional formula or the data table using α, β, or γ, which are unitless index values.

[0264] Alternatively, in the case where the arithmetic control circuit 20C has an analog input port, an operational amplifier may be provided instead of the first comparator 41a, the second comparator 41b, and the inversion memory circuit 52 of the monitor signal processing circuit 40C. In this case, the analog signal voltage ANL proportional to the monitor voltage Vx may be input to the arithmetic control circuit 20C, so that at least one of the inverted logic signals LOW, HIG, and ANS is generated by the microprocessor 21, and therefore the repetitive command signal PLS can be generated.

[0265] Additionally, in the third embodiment of Fig. 12, the connection point B is connected to the negative-side power supply line 67 of the on-board high-voltage device 60X, but even if the connection point B is connected to the positive-side power supply line, no change occurs in Expression I.

[0266] As is clear from the above description, according to the third embodiment of the present invention, the leakage resistance detection device 50C for an on-board high-voltage device is provided, which is connectable to the on-board high-voltage device 60X including the high-voltage DC power source 61 and the high-voltage electrical load 64 to be supplied and driven by the on-board high-voltage DC power source 61, wherein the on-board high-voltage device 60X has the leakage resistance Rx with respect to the vehicle body 11, as typified by the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side, wherein the leakage resistance detection device 50C includes the constant-voltage control power source 25 supplied and driven by the low-voltage DC power source 10,which has the negative terminal connected to the vehicle body 11 and which generates the control power supply voltage Vcc, wherein the leakage resistance detection device 50C is configured to measure a value of the leakage resistance Rx via the coupling capacitor 51 having a terminal B connected to a predetermined position of the on-board high-voltage device 60X, wherein the leakage resistance detection device 50C further includes the repetitive signal output circuit 30C, the monitoring signal processing circuit 40C, and the arithmetic control circuit 20C, wherein the arithmetic control circuit 20C includes the microprocessor 21 and the program memory 24C, which cooperate with each other.

[0267] Furthermore, the repetitive signal output circuit 30C alternately repeats, in response to the switching operation of the charge / discharge switching element 31 operating in response to the repetitive command signal PLS, a discharge period and a charge period in which the measurement point A is intermittently connected to the control power supply voltage Vcc via the charge / discharge resistor 33 or 35, the measurement point A being another terminal of the coupling capacitor 51, to thereby gradually raise or gradually lower the monitor voltage Vx, which is a potential between the measurement point A and the vehicle body 11.The monitor signal processing circuit 40C is configured to: generate at least one of the inverted logic signals ANS, LOW, and HIG when a value of the monitor voltage Vx, which gradually increases or gradually decreases at a gradient that becomes smoother as the charge / discharge time constant becomes longer, passes a predetermined threshold voltage, and input at least one of the inverted logic signals ANS, LOW, and HIG to the arithmetic control circuit 20C, or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20C.The arithmetic control circuit 20C is configured to: measure the transition time Tx during which the value of the monitor voltage Vx changes from one threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to another; calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and generate the resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than the predetermined leakage resistance Rx0, or when the transition time Tx reaches the limit transition time Tx0 corresponding to the predetermined limit leakage resistance Rx0.The repetitive command signal PLS is a pulse signal with a variable cycle in which an output pulse is inverted when the transition time Tx is reached.

[0268] The repetitive signal output circuit 30C is configured to, in response to the switching operation of the charge / discharge switching element 31, connect the measurement point A, which is the other terminal of the coupling capacitor 51, to an output terminal of the constant-voltage control power source 25 via the rapid charge / discharge resistor 35 having the reference value Rq, or connect the measurement point A to the negative terminal of a vehicle body potential via the reference resistor 33 having the resistance value Rs sufficiently larger than the resistance value Rq, to ​​thereby quickly increase or gradually decrease the monitor voltage Vx, which is the potential between the measurement point A and the vehicle body 11.The monitor signal processing circuit 40C is configured to: generate the first inverted logic signal ANS or LOW while the value of the monitor voltage Vx, which gradually decreases at a gradient that becomes smoother as the time constant (Rs+Rx)C, which is a product of a sum of the resistance value Rs (Rs>>Rq) and the leakage resistance Rx, and the electrostatic capacitance C of the coupling capacitor 51, has decreased to pass the first threshold value V1, and input the first inverted logic signal ANS or LOW to the arithmetic control circuit 20C; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20C.The monitor signal processing circuit 40C is further configured to: generate the second inverted logic signal ANS or HIG when the value of the monitor voltage Vx, which rapidly increases at a gradient that becomes steeper as the time constant (Rq+Rx)C becomes smaller, has increased to pass the second threshold value V2, which is close to the control power supply voltage Vcc, and input the second inverted logic signal ANS or HIG to the arithmetic control circuit 20C; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20C.

[0269] The arithmetic control circuit 20C receives the second inverted logic signal ANS or HIG as an input or generates the second inverted logic signal ANS or HIG based on the input analog signal voltage ANL, and the arithmetic control circuit receives the first inverted logic signal ANS or LOW as an input or generates the first inverted logic signal ANS or LOW based on the input analog signal voltage ANL. The arithmetic control circuit 20C measures, as the transition time Tx, a period of time from when the second inverted logic signal ANS or HIG is generated until when the first inverted logic signal ANS or LOW is received. The arithmetic control circuit 20C or the monitor signal processing circuit 40C inverts the repetitive command signal PLS when the first inverted logic signal ANS or LOW and the second inverted logic signal ANS or HIG are input or generated.The first inverted logic signal and the second inverted logic signal are a pair of logic signals of the first logic signal LOW whose output logic changes before and after the first threshold value V1, and the second logic signal HIG whose output logic changes before and after the second threshold value V2, or the band history logic signal ANS that comes to a first logic state of "L" or "H" when the value of the monitor voltage Vx has increased to the second threshold value V2 or higher after decreasing to the first threshold value V1 or lower, and becomes a second logic state of "H" or "L" when the value of the monitor voltage Vx has decreased to the first threshold value V1 or lower after increasing to the second threshold value V2 or higher.

[0270] As described above, in conjunction with claim 5, the repetitive signal output circuit charges and discharges the coupling capacitor, and the arithmetic control circuit measures, at the transition time Tx, a period during which the monitor voltage Vx decreases to the first threshold V1 from the second threshold V2, which is set to a value close to the control power supply voltage Vcc. Furthermore, the monitor signal processing circuit or the arithmetic control circuit inverts the pulse output when the monitor voltage Vx has decreased to pass the first threshold V1 and when the monitor voltage Vx has increased to pass the second threshold V2.

[0271] Therefore, the output pulse is inverted along with the completion of the transition time Tx measurement, and the output pulse is inverted immediately after the completion of charge / discharge is confirmed. Therefore, the occurrence frequency of the repetitive command signal PLS is shortened as the leakage resistance decreases. Thus, an abnormality can be quickly detected.

[0272] As is clear from the above description, according to the third embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 65 on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30C, is represented by R0, and when a period,during which the monitoring voltage Vx gradually decreases from a state of the control power supply voltage Vcc to a predetermined threshold voltage (Vcc-V0) represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to V0 / Vcc: calculating, by the arithmetic control circuit 20C, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases,along with an increase in the leakage resistance coefficient β; selectively using a negative gradient curve region following the positive gradient curve region or a negative gradient curve region occupying the entire region, and determining the threshold voltage coefficient γ such that a value of the transition time coefficient α when the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the negative gradient curve region; determining, when the transition time Tx has become a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the short-circuit transition time Tx00 corresponding to a case where the leakage resistance Rx becomes 0.

[0273] As is clear from the above description, according to the third embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is the total resistance of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30C, is represented by R0, and when a period,during which the monitoring voltage Vx gradually decreases from a state of the control power supply voltage Vcc to a predetermined threshold voltage (Vcc-V0) represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to V0 / Vcc: calculating, by the arithmetic control circuit 20C, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases,along with an increase in the leakage resistance coefficient β; excluding, in advance, by the characteristic curve correcting means, the application of a negative gradient curve region following the positive gradient curve region; determining the threshold voltage coefficient γ so that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the corrected positive gradient curve region; determining, when the transition time Tx has become a value equal to or less than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the open-circuit transition time Txm corresponding to a case where the leakage resistance Rx is infinite. Fourth embodiment

[0274] Referring to Fig. 16, which is an overall configuration diagram of the device according to a fourth embodiment of the present invention, below mainly the differences from Fig. 1. In the drawings, the same reference numerals designate the same or equivalent parts. In Fig. 16, a leakage resistance detection device 50D includes an arithmetic control circuit 20D, a repetitive signal output circuit 30D, and a monitoring signal processing circuit 40D, and detects a leakage resistance of an onboard high voltage device 60Y.

[0275] The first major difference is that in the on-board high-voltage device 60Y, a positive-side power supply line 68 is connected to a terminal B of the coupling capacitor 51 via an interconnection terminal 16, and the positive-side equivalent leakage resistor 65 has an equivalent leakage resistance R2 on the connection side, while the negative-side equivalent leakage resistor 66 has an equivalent leakage resistance R1 on the non-connection side.

[0276] The second main difference is that the repetitive signal output circuit 30D is supplied with a repetitive command signal PLS having a variable cycle from the monitor signal processing circuit 40D.

[0277] The third major difference is that in the monitor signal processing circuit 40D, a comparator 41c has an output terminal and a positive input terminal connected to each other via a positive feedback resistor 48, and the positive input terminal is connected to a measurement point A via an input resistor 44c and a noise filter composed of a smoothing resistor 45 and a smoothing capacitor 46. The comparator 41c has a negative input terminal to which a comparison reference voltage Vs is applied, which is obtained by dividing the control power supply voltage Vcc by the voltage dividing resistors 42c and 43c. An output of the comparator 41c, configured as described above, is input to the arithmetic control circuit 20D as an inverted logic signal ANS, which serves as a band history logic signal.

[0278] When the set threshold voltage V0 is set to V2, a series-combined resistance value of the input resistor 44c and the smoothing resistor 45 is represented by R44, and a resistance value of the positive feedback resistor 48 is represented by R48, a value expressed by expression (8) is applied to the value of the comparison reference voltage Vs. Vs=V0×R48 / (R44+R48)

[0279] The value of the monitoring voltage Vx at which the output logic of the comparator 41c, which forms a hysteresis circuit, returns from "L" to "H", is set to V1≈0. In this case, it is only necessary to determine the resistance values ​​R44 and R48 so that they have the relationship of Expression (9). Vs=Vcc×R44 / (R44+R48)

[0280] According to expressions (8) and (9), at the time when the value of the monitor voltage Vx rises to the determination threshold V0=V2, the logic level of the inverted logic signal ANS, which is the output of the comparator 41c, becomes "H." Once the logic level of the inverted logic signal ANS becomes "H," the comparator 41c performs a hysteresis operation so that the current output logic can be maintained until the value of the monitor voltage Vx decreases to V1≈0.

[0281] The fourth main difference is that a series circuit of a test leakage resistor 54 and a test switching element 55 is connected between the connection terminal 16 and the vehicle body 11. The test switching element 55, such as an optical isolation transistor, receives a conduction command via a drive resistor 56 from a test operation command TST provided to the arithmetic control circuit 20D. A test resistor R4, which is a resistance value of the test leakage resistor 54, is equal to, for example, the limit leakage resistance Rx0 or the preliminary leakage resistance Rxn. The test switching element 55 temporarily becomes conductive at the start of operation immediately after the manual power supply switch (not shown) is closed and the low-voltage power supply switch 12 is closed.

[0282] In this way, a pre-operation inspection is performed to check whether detection control of the leakage resistance is performed correctly or not, and it is also possible to detect individual fluctuations in the electrostatic capacitance C of the coupling capacitor 51 and a secular change thereof, so that the calibrated value of the electrostatic capacitance C can be used.

[0283] A microprocessor 21 provided in the arithmetic control circuit 20D cooperates with a program memory 24D and measures the value of a leakage resistance Rx. The microprocessor 21 generates a pre-alarm output ER2 when the measured value of the leakage resistance Rx becomes equal to or less than a predetermined preliminary leakage resistance Rxn, and generates a resistance abnormality determination output ER1 when the measured resistance becomes equal to or less than a limit leakage resistance Rx0. The resistance abnormality determination output ER1 and the pre-alarm output ER2 are output to the abnormality alarm device 19 and fed back to be input to the arithmetic control circuit 20D as a report command confirmation signal RET to check whether the abnormality alarm output has been correctly generated.

[0284] Note that similar to Fig. 1, the low-voltage DC power source 10, the low-voltage power supply switch 12, the vehicle state signal 18, and the abnormality alarm device 19 are connected outside the leakage resistance detection device 50D, and the constant-voltage control power source 25, which generates the control power supply voltage Vcc, is provided in the leakage resistance detection device 50D.

[0285] The following describes the action and operation of the device according to the fourth embodiment of the present invention shown in Fig. 16, briefly referring to a timing diagram of Fig. 17 as well as Fig. 16. First created in Fig. 16, when the manual power supply switch (not shown) is closed and the low-voltage power supply switch 12 is closed, the constant-voltage control power source 25 supplies a predetermined control power supply voltage Vcc. Then, the microprocessor 21 begins its control operation.

[0286] Part (A) of Fig. Figure 17 shows the waveform of the repetitive command signal PLS generated by the monitor signal processing circuit 40D. A first period (slow discharge period) T1, in which the repetitive command signal PLS has a logic "L," and a second period (fast discharge period) T2, in which the repetitive command signal PLS has a logic "H," have different values, each occupying a portion of the entire cycle T0=T1+T2.

[0287] In Fig. 16, when the logic level of the repetitive command signal PLS is “L” (or “H”), the auxiliary switching element 31 is interrupted (or becomes conductive) as shown in part (B) of Fig. 17. Then in Fig. 16, when the charge / discharge switching element 31 is interrupted, the coupling capacitor 51 is slowly discharged laterally via a series connection of the leakage resistor Rx, the rapid charge / discharge resistor 35, and the charge / discharge resistor 33, with the result that the monitoring voltage Vx gradually increases along with the decrease in the charging current. On the other hand, when in Fig. 16 the charge / discharge switching element 31 becomes conductive, the coupling capacitor 51 is rapidly charged via a series connection of the rapid charge / discharge resistor 35 and the leakage resistor Rx, with the result that the monitoring voltage Vx drops rapidly.

[0288] Part (C) of Fig. Figure 17 shows how the monitoring voltage Vx gradually increases or rapidly decreases. The gradient of the charge / discharge characteristic becomes smoother as the charge / discharge time constants expressed in expressions (8y) and (9y) above increase. τ1=(R0+Rx)C is a slow discharge time constant, and τ2=(Rq+Rx)C is a fast charge time constant.

[0289] In part (C) of Fig. 17, the value of the slowly discharged monitoring voltage Vx gradually rises to the first final voltage V2 at the end of the first period T1. At this time, the logic level of the inverted logic signal ANS, which is the output of the comparator 41c, changes from "L" to "H." As a result, the charge / discharge switching element 31 becomes conductive to begin the rapid charging of the coupling capacitor 51.

[0290] The value of the fast-charged monitor voltage Vx rapidly decreases to the second end voltage V1≈0. At this time, the logic level of the inverted logic signal ANS, which is the output of the comparator 41c, changes from "H" to "L," and the logic level of the inversion memory circuit 52 becomes "L." As a result, the charge / discharge switching element 31 is interrupted to start the slow discharge of the coupling capacitor 51.

[0291] Note that in order to avoid waiting for the monitoring voltage Vx to completely converge to 0 volts at the end of the second period T2, a residual error ΔV of about V1=0.03Vcc is added to the set threshold voltage V0.

[0292] Part (D) of Fig. Figure 17 shows the generated waveform of the inverted logic signal ANS, which is the output of the comparator 41c. The inverted logic signal ANS and the repetitive command signal PLS are the same signal. A measurement time Tx is a period during which the logic level of the inverted logic signal ANS is "L" and the monitor voltage Vx gradually increases from the first initial voltage V1≈0 to the first final voltage V2=V0.

[0293] Part (E ) of Fig. Figure 17 shows the waveform of the two-ended voltage E of the coupling capacitor 51. The discharge start voltage (= charge end voltage) E1 is equal to the stable divided voltage Vn0 expressed in expression (4y). The value of a discharge end voltage (= charge start voltage) E2 drops to Vn0-Vcc in the case where the discharge period T1 is infinite.

[0294] The following demonstrates that the calculation using the threshold voltage coefficient γ of V0 / Vcc is similar to the case of Fig. 1 can be carried out so that the characteristic diagrams of Fig. 3 to 5 can be applied as they are. Note that the generation frequency of the repetitive command signal PLS is a fixed value in the first and second embodiments, but in the third and fourth embodiments, the first period T1 and the second period T2 are automatically changed according to the value of the leakage resistance Rx, so that the next repetitive command signal PLS can be quickly generated after completion of the measurement of the transition time Tx. Furthermore, a transition to the next first period T1 is performed after confirmation that the monitor voltage Vx has dropped to a value close to V1≈0, and therefore, it is unnecessary to provide a delay time to the second period T2 with respect to the second end voltage V1, unlike in part (C) of Fig. 2.

[0295] The following shows that the above Fig. 3 shows expression I in the fourth embodiment of Fig. 16 is similarly applicable. Unloading period

[0296] In the slow discharge period, in which the logic level of the repetitive command signal PLS in Fig. 16 is "L" and the transistor serving as the auxiliary switching element 31 is opened, expression (30bb) is established. C×R0(dE / dt)=Vx−Vcc where R0=Rs+Rq≈Rs>>Rq

[0297] Vx in expression (30bb) is substituted into expression (7y) above to obtain expression (31bb) Vn0−Vcc=τ1(dE / dt)+E where τ1=(R0+Rx)×c

[0298] In differential equation (31bb), the initial value of E at time t=0 is represented by E1, and the value of E at infinity is represented by Vn0-Vcc. Then, a solution obtained by expression (32bb) is obtained. E=E1exp(−t / τ1)+(Vn0−Vcc){1−t / τ1}

[0299] In addition, the value of (dE / dt) in expression (30bb) is substituted into expression (7y) above to obtain expression (33bb). Vx / (R0 / / Rx)=(Vn0−E) / Rx+Vcc / R0 where R0 / / Rx represents the parallel combined resistance of R0 and Rx. Charging period

[0300] In the fast charging period in which the logic level of the repetitive command signal PLS in Fig. 16 is “H” and the transistor serving as the auxiliary switching element 31 is closed, expression (30aa) is established. Vx=Rq×C(dE / dt)=0

[0301] Vx in expression (30a) is substituted into expression (7y) above to obtain expression (31aa). Vn0=E+τ2×(dE / dt) where τ=(Rq+Rx)C

[0302] In differential equation (31aa), the initial value of E at time t=0 is represented by E2, and the value of E at infinite time is represented by Vnc-Vcc. Then, a solution expressed by expression (32aa) is obtained. E=E2exp(−t / τ2)+(Vn0−Vcc){1−exp(−t / τ2)}

[0303] In addition, the value of (dE / dt) in expression (30aa) is substituted into expression (7y) above to obtain expression (33aa). Vx / (Rq / / Rx)=(Vn0−E) / Rx where Rq / / Rx represents the parallel combined resistance of Rq and Rx. Stable condition

[0304] In the stable state where the logic level of the repetitive command signal PLS is Fig. 16 is alternately inverted during the discharge period T1 and the charge period T2, and there is no change in the leakage resistance Rx and the output voltage Vh of the high-voltage DC power source 61, the output values ​​E1 and E2 are calculated in expressions (32bb) and (32aa) as follows. First, in expression (32bb), at the discharge end time t=T1, E=E2 is established, and therefore expression (34bb) is established. E2=E1×K1+(Vn0−Vcc)(1−K1) where K1=exp(−T1 / τ1)

[0305] Additionally, in expression (32aa), at the loading end time T=T2, E=E1 is established and therefore expression (34aa) is established. E1=E2×K2+Vn0(1−K2) where K2=exp(−T2 / τ2)

[0306] From the expressions (34bb) and (34aa) expressions (35bb) and (35aa) are obtained. Vn0−E1=Vcc×K0×K2 Vn0−E2=Vcc×K0 where K0=(1−K1) / (1−K1×K2) In the case of the fourth embodiment

[0307] In the Fig. 16 and Fig. 17, at the discharge start time t=0, Vx1=V1≈0 is established. At time Tx=T1, if the value of Vx becomes equal to the set threshold voltage V2=V0, Expression (39bb) for time t=Tx is calculated from Expression (32bb). Note that the value of the discharge start voltage E1 is Vn0-VccKoK2≈Vn0 by Expression (35b), and Kx=exp(-T1 / z1)=K1 is established. Ex=Vn0exp(−Tx / τ1)+(Vn0−Vcc){1−exp(−Tx / τ1)}=Vn0−Vcc(1−Kx)

[0308] Expression (40bb) is obtained from expressions (33bb) and (39bb). V0 / (R0 / / Rx)=(Vn0−Ex) / Rx+Vcc / R0=Vcc / (R0 / / Rx)−VccKx / Rx

[0309] Then, the threshold voltage coefficient γ of V0 / Vcc, the leakage resistance coefficient β of Rx / R0 and the transition time coefficient α of Tx / (R0C) are substituted into Expression (40bb) and Expression (41) is obtained. γ=(1+β−Kx) / (1+β)∴Kx=(1 + β)×(1-γ)

[0310] Kx=exp[-Tx / {(R0+Rx)C}]=exp{-α / (1+β)} is established and hence expression (42) is obtained via natural logarithm conversion. Ex = Vn0exp(-Tx / τ1)+(Vn0 - Vcc){1 - exp(-Tx / τ1)}=Vn0 - Vcc(1−Kx)

[0311] Expressions (41) and (42) correspond to the above in Fig. Expression I shown in Figure 3.

[0312] Next, the action and operation of the device are described by Fig. 16 in detail with reference to the Fig. 18 and Fig. 19 illustrated operational explanation flow charts, mainly pointing out differences compared to the Fig. 6 and Fig. 7. Note that the step numbers of Fig. 6 and Fig. 7 are in the 1000s, while the step numbers of Fig. 18 and Fig. 19 are in the 4000s, and the same 3-digit or lower numbers represent the same or equivalent parts. However, regarding the equivalent parts, the different parts that require additional description are described below.

[0313] Referring to the first flowchart of Fig. In Fig. 18, step 4604 indicated by a dashed line is a step necessary for generating the repetitive command signal PLS from the arithmetic control circuit 20D by causing the microprocessor 21 to execute logic processing equivalent to the hysteresis circuit formed by the comparator 41c illustrated in the monitor signal processing circuit 40D in the state where the analog signal voltage ANL proportional to the monitor voltage Vx is input to the arithmetic control circuit 20D. In this case, the monitor signal processing circuit 40D only needs to include the noise filter formed by the smoothing resistor 45 and the smoothing capacitor 46, and an operational amplifier.

[0314] The subsequent step 4605 is a determination step of determining whether the logic of the inverted logic signal ANS input to the microprocessor 21 has changed or not. If the logic has changed within one cycle of the repetitive command signal PLS, it is determined as YES, and the flow proceeds to step 4606c. If the logic has not changed, it is determined as NO, and the flow proceeds to step 4606a. The determination of YES in step 4605 indicates that the value of the monitor voltage Vx is in the proper range from 0 to Vcc and has increased or decreased to pass the set threshold voltage V0.

[0315] Next, referring to the second flowchart of Fig. 19 Step 4700a is a step of determining whether a calibration operation is to be performed. If a calibrated value of the electrostatic capacitance C of the coupling capacitor 51 is measured at the time of the delivery inspection of the leakage resistance detection device 50D or immediately after the low-voltage power supply switch 12 is closed to supply the leakage resistance detection device 50D at the time of actual operation of the leakage resistance detection device 50D, this is determined as YES and the flow proceeds to step 4700c. After the calibration operation is completed, this is determined as NO and the flow proceeds to step 4700b.

[0316] Step 4700c is a step serving as operation calibration means in which a test flag is set in the case of the delivery adjustment operation, while a test operation command TST is generated at the start of the actual operation, and the flow proceeds to step 4701. Step 4700b is a step in which the test flag generated in step 4700c is reset and the test operation command TST is suspended, and the flow proceeds to step 4701.

[0317] In the case of the pre-shipment adjustment operation, the arithmetic control circuit 20D measures the leakage resistance Rx in step 4706 to be described below in a state where the coupling capacitor 51 and the on-board high-voltage device 60Y are not connected to each other. The arithmetic control circuit 20B calibrates and stores the value of the electrostatic capacitance C of the coupling capacitor 51 so that the obtained result can become the additional resistance R3 of the bypass leakage resistor 53.

[0318] At the start of actual operation, the arithmetic control circuit 20D temporarily closes the test switching element 55 at the start of operation and determines whether the resistance abnormality determination output ER1 is momentarily activated or whether the pre-alarm output ER2 operates momentarily when the leakage resistance Rx becomes equal to or less than the preliminary leakage resistance Rxn, thereby inspecting whether the operation of detecting the leakage resistance Rx can be performed normally. If the inspection result shows that the detection operation is abnormal, the resistance abnormality determination output ER1 or the pre-alarm output ER2 is generated continuously or intermittently.When the inspection result shows that the detection operation is normal, the output generation is suspended for such a short period of time that the abnormality alarm device 19 to be driven by the resistance abnormality determination output ER1 or the pre-alarm output ER2 does not operate.

[0319] Furthermore, the arithmetic control circuit 20D closes the test switching element 55 at the start of operation to measure the value of the first combined leakage resistance in the state where the test leakage resistance 54, the bypass leakage resistance 53, and the actual leakage resistance Rx are connected in parallel, and measures the value of the second combined leakage resistance in the state where the test switching element 55 is open and the bypass leakage resistance 53 and the actual leakage resistance Rx are connected in parallel. The arithmetic control unit 20D recalculates the electrostatic capacitance C of the coupling capacitor 51 based on the values ​​of the first and second leakage resistances and stores the calculated electrostatic capacitance C as a calibrated value.The calibrated value is calculated so that the actual leakage resistance Rx in the open circuit and closed circuit states of the test switching element 55 can have the same value due to the calibrated and stored electrostatic capacitance C.

[0320] The subsequent step 4701 is a determination step of monitoring the logic level of the inverted logic signal ANS. If the logic level is "L," it is determined as YES, and the flow proceeds to step 4702. If the logic level is "H," it is determined as NO, and the flow proceeds to step 4608 of Fig. 18 via continuation B.

[0321] Note that when it is determined YES in step 4701 that the logic level of the repetitive command signal PLS is "L" and the charge / discharge switching element 31 becomes non-conductive, and therefore the monitor voltage Vx gradually starts to decrease, the transition time thereof is measured in the subsequent step 4702.

[0322] The subsequent step 4703 is a determination step of monitoring the logic level of the inverted logic signal ANS. If the logic level is "H," it is determined as YES, and the flow proceeds to step 4704. If the logic level is "L," it is determined as NO, and the flow proceeds to step 4704b of Fig. 14 continued.

[0323] Note that if YES is determined in step 4703, the logic level of the repetitive command signal PLS is "H" and the charge / discharge switching element 31 becomes conductive, and therefore the monitor voltage Vx starts to rise rapidly, and its transition time Tx is read out and stored in the subsequent step 4704a. As described above, in step 4701 and step 4703, the start time of the gradual rise of the monitor voltage Vx and the completion time of its gradual rise are determined, respectively.

[0324] Note that the calculation of the leakage resistance Rx in step 4705 is performed in three ways. First, the calculation is performed based on the characteristic curve whose overall region is the negative gradient curve region, as shown by the curve 103 of Fig. 3. Secondly, the calculation is performed using the negative gradient curve region of a bell-shaped curve, as represented by the curve 112 of Fig. 3. Thirdly, the calculation is performed using the positive gradient curve region of a bell-shaped curve, as represented by curve 113 of Fig. 3. In any case, the value of the leakage resistance Rx is calculated from the measured value of the transition time Tx based on a functional formula or a data table between the transition time coefficient α and the leakage resistance coefficient β using the threshold voltage coefficient γ=V0 / Vcc as a parameter.

[0325] As an alternative abnormality determination method, instead of comparing the detected leakage resistance Rx with the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, the preliminary transition time Txn or the limit transition time Tx0 can be calculated in advance according to the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, and the actual transition time Tx can be compared with the preliminary transition time Txn or the limit transition time Tx0.

[0326] Furthermore, in the case where the value of the reference time constant R0xC, which is the denominator of the transition time coefficient α, the value of the series resistance R0, which is the denominator of the leakage resistance coefficient β, or the value of the threshold voltage coefficient γ, is considered a fixed value, a formula or a data table between the transition time Tx and the leakage resistance Rx can be directly used as a functional formula or data table. In this case, the leakage resistance Rx can be directly calculated from the measured transition time Tx without performing complicated calculation processing during operation.

[0327] However, in the case of varying the value of the threshold voltage coefficient γ, the denominator value of the transition time coefficient α, or the denominator value of the leakage resistance coefficient β based on the selected vehicle types or operating conditions, it is advantageous to set the functional formula or the data table using α, β, or γ, which are unitless index values. Alternatively, in the case where the arithmetic control circuit 20D has an analog input terminal, an operational amplifier may be provided instead of the hysteresis circuit formed by the comparator 41c of the monitor signal processing circuit 40D.In this case, the analog signal voltage ANL proportional to the monitor voltage Vx can be input to the arithmetic control circuit 20D, so that at least one of the inverted logic signals LOW, HIG and ANS is generated by the microprocessor 21 and therefore the repetitive command signal PLS can be generated.

[0328] Additionally, in the fourth embodiment of Fig. 16, the connection point B is connected to the positive-side power supply line 68 of the on-board high-voltage device 60Y, but even if the connection point B is connected to a negative-side power supply line, no change occurs in Expression I.

[0329] As is clear from the above description, according to the fourth embodiment of the present invention, the leakage resistance detection device 50D for an on-board high-voltage device is provided, which is connectable to the on-board high-voltage device 60Y including the on-board high-voltage DC power source 61 and the high-voltage electrical load 64 to be supplied and driven by the high-voltage DC power source 61. The on-board high-voltage device 60Y has the leakage resistance Rx with respect to the vehicle body 11, as typified by the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side. The leakage resistance detection device 50D includes the constant-voltage control power source 25, which is supplied and driven by the low-voltage DC power source 10 having the negative terminal connected to the vehicle body 11.and which generates the control power supply voltage Vcc, wherein the leakage resistance detection device 50D is configured to measure a value of the leakage resistance Rx via the coupling capacitor 51 having a terminal B connected to a predetermined position of the on-board high-voltage device 60Y, wherein the leakage resistance detection device 50D further includes the repetitive signal output circuit 30D, the monitoring signal processing circuit 40D, and the arithmetic control circuit 20D, wherein the arithmetic control circuit 20D includes the microprocessor 21 and the program memory 24D, which cooperate with each other.

[0330] Further, in response to the switching operation of the charge / discharge switching element 31, the repetitive signal output circuit 30D alternately repeats a charge period and a discharge period in which the measurement point A is intermittently connected to the control power supply voltage Vcc via the charge / discharge resistor 33 or 35, the measurement point A being another terminal of the coupling capacitor 51, to thereby gradually increase or gradually decrease the monitor voltage Vx, which is a potential between the measurement point A and the vehicle body 11.The monitor signal processing circuit 40D is configured to: generate the inverted logic signal ANS when a value of the monitor voltage Vx, which gradually increases or gradually decreases with a gradient that becomes smoother as the charge / discharge constant becomes larger, passes a predetermined threshold voltage, and input the inverted logic signal ANS to the arithmetic control circuit 20D; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20D.The arithmetic control circuit 20D is configured to: measure the transition time Tx during which the value of the monitor voltage Vx changes from one threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to another; calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and generate the resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than the predetermined limit leakage resistance Rx0 or when the transition time Tx reaches the limit transition time Tx0 corresponding to the predetermined limit leakage resistance Rx0.The repetitive command signal PLS is a pulse signal with a variable cycle in which an output pulse is inverted when the transition time Tx is reached.

[0331] The repetitive signal output circuit 30D is configured to, in response to the switching operation of the charge / discharge switching element 31, connect the measurement point A, which is the other terminal of the coupling capacitor 51, to an output terminal of the constant-voltage control power source 25 via the reference resistor 33 having the resistance value Rs, or connect the measurement point A to the negative terminal having a vehicle body potential via the rapid charge / discharge resistor 35 having the reference value Rq sufficiently smaller than the resistance value Rs, to thereby gradually increase or quickly decrease the monitor voltage Vx, which is the potential between the measurement point A and the vehicle body 11.The monitor signal processing circuit 40D is configured to: generate the second inverted logic signal ANS or HIG when the value of the monitor voltage Vx, which gradually increases at a gradient that becomes smoother as the time constant (Rs+Rx)C, which is a product of the sum of the resistance value Rs (Rs>>Rq), the leakage resistance Rx, and the electrostatic capacitance C of the coupling capacitor 51, increases, has increased to pass the second threshold value V2 and input the second inverted logic signal ANS or HIG to the arithmetic control circuit 20D; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20D.The monitor signal processing circuit 40D is further configured to: generate the first inverted logic signal ANS or LOW when the value of the monitor voltage Vx, which falls rapidly at a gradient that becomes steeper as the time constant (Rq+Rx)C becomes smaller, has decreased to pass the first threshold value V1, which is close to 0 volts, and input the first inverted logic signal ANS or LOW to the arithmetic control circuit 20D; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20D.

[0332] The arithmetic control circuit 20D receives the second inverted logic signal ANS or HIG as an input or generates the second inverted logic signal ANS or HIG based on the input analog signal voltage ANL, and the arithmetic control circuit 20D receives the first inverted logic signal ANS or LOW as an input or generates the first inverted logic signal ANS or LOW based on the input analog signal voltage ANL. The arithmetic control circuit 20D measures, as the transition time Tx, a period from when the first inverted logic signal ANS or LOW is generated until when the second inverted logic signal ANS or HIG is received. The arithmetic control circuit 20D or the monitor signal processing circuit 40D inverts the repetitive command signal PLS when the first inverted logic signal ANS or LOW and the second inverted logic signal ANS or HIG are input or generated.The first inverted logic signal and the second inverted logic signal are a pair of logic signals of the first logic signal LOW whose output logic changes before and after the first threshold value V1, and a second logic signal HIG whose output logic changes before and after the second threshold value V2, or the band history logic signal ANS, which takes a first logic state of "L" or "H" when the value of the monitor voltage Vx has increased to the second threshold value V2 or higher after decreasing to the first threshold value V1 or lower, and which becomes a second logic state of "H" or "L" when the value of the monitor voltage Vx has decreased to the first threshold value V1 or lower after increasing to the second threshold value V2 or higher.

[0333] As described above, in conjunction with claim 6 of the present invention, the repetitive signal output circuit charges and discharges the coupling capacitor, and the arithmetic control circuit measures, at the transition time Tx, a period during which the monitor voltage Vx rises from the first threshold V1, which is set to a value close to 0 volts, to the second threshold V2. Furthermore, the monitor signal processing circuit or arithmetic control circuit inverts the pulse output when the monitor voltage Vx has decreased to pass the first threshold V1 and when the monitor voltage Vx has increased to pass the second threshold V2.

[0334] Therefore, the output pulse is inverted along with the completion of the transition time Tx measurement, and the output pulse is inverted immediately after the completion of charging / discharging is confirmed. Therefore, the occurrence frequency of the repetitive command signal PLS is shortened and the leakage resistance is reduced. Thus, abnormalities can be detected quickly.

[0335] As is clear from the above description, according to the fourth embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is the total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30D, is represented by R0, and when a period during which the monitor voltage Vx gradually increases from a state of 0 to the predetermined threshold voltage V0,represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to V0 / Vcc: calculating, by the arithmetic control circuit 20D, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; selectively using a negative gradient curve region following the positive gradient curve region or a negative gradient curve region,which occupies the entire region, and determining the threshold voltage coefficient γ such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the negative gradient curve region; determining, when the transition time Tx has become a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the short-circuit transition time Tx00 corresponding to a case where the leakage resistance Rx becomes 0.

[0336] As is clear from the above description, according to the fourth embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by RX, and a series resistance value, which is a total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30D, is represented by R0, and when a period during which the monitor voltage Vx gradually increases from a state of 0 to the predetermined threshold voltage V0,represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to V0 / Vcc: calculating, by the arithmetic control circuit 20D, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; excluding, in advance, by the characteristic curve correcting means, the application of a negative gradient curve region,which follows the positive gradient curve region; determining the threshold voltage coefficient γ such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the corrected positive gradient curve region; determining, when the transition time Tx has become a value equal to or less than the limit transition time Tx0, corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the open-circuit transition time Txm corresponding to a case where the leakage resistance Rx is infinite. Fifth embodiment

[0337] Referring to Fig. 20, which is an overall configuration diagram of an apparatus according to a fifth embodiment of the present invention, below mainly differences from Fig. 1. In the drawings, the same reference numerals designate the same or equivalent parts. In Fig. 20, a leakage resistance detection device 50E includes an arithmetic control circuit 20E, a repetitive signal output circuit 30E, and a monitor signal processing circuit 40E, and detects a leakage resistance of an onboard high voltage device 60X.

[0338] The first major difference is that the repetitive signal output circuit 30E includes a charge / discharge switching element 39 with a pair of upper and lower transistors 31a and 31b (not shown). The upper resistor 31a becomes conductive when the output logic level of the repetitive command signal PLS becomes "H," while the lower transistor 31b becomes conductive when the output logic level of the repetitive command signal PLS becomes "L." Thus, the charge / discharge switching element 39 is controlled so that when one of the upper and lower transistors is conductive, the other transistor is non-conductive.

[0339] When the upper transistor 31a becomes conductive, the coupling capacitor 51 is slowly charged via the charge / discharge resistor 33 and a fast charge / discharge resistor 35. When the lower transistor 31b becomes conductive, the coupling capacitor 51 is slowly discharged via the charge / discharge resistor 33 and the fast charge / discharge resistor 35.

[0340] The fast charge / discharge resistor 35 is a current limiting resistor to cause positive and negative circuits of the constant voltage control power source 25 to charge and discharge the coupling capacitor 51 via bypass diodes 36 and 37 when the value of the monitoring voltage Vx deviates from the correct range of 0 to Vcc. Fig. 20 may be short-circuited or eliminated, and instead, the fast charge / discharge resistors 35a and 35b, illustrated by dashed lines, may be connected in series with the bypass diodes 36 and 37.

[0341] The second main difference is that the repetitive command signal PLS generated by the arithmetic control circuit 20E is a pulse train signal having a variable cycle, as described below with reference to part (A) of Fig. 21 described.

[0342] The third main difference is that the monitor signal processing circuit 40E includes an operational amplifier 49 instead of the comparator 41 and inputs an analog signal voltage ANL proportional to the monitor voltage Vx to the arithmetic control circuit 20E.

[0343] A microprocessor 21 provided in the arithmetic control circuit 20E cooperates with a program memory 24E and measures the value of the leakage resistance Rx. The microprocessor 21 generates a pre-alarm output ER2 when the measured value of the leakage resistance Rx becomes equal to or less than a predetermined preliminary leakage resistance Rxn, and generates a resistance abnormality determination output ER1 when the measured resistance becomes equal to or less than a limit leakage resistance Rx0, which are output to the abnormality alarm device 19.

[0344] Note that similar to Fig. 1, the low-voltage DC power source 10, the low-voltage power supply switch 12, the vehicle state signal 18, and the abnormality alarm 19 are connected outside the leakage resistance detection device 50E, and the constant-voltage control power source 25, which generates the control power supply voltage Vcc, is provided to the leakage resistance detection device 50E.

[0345] The following describes the effect and operation of the device according to the fifth embodiment of the present invention, which is shown in Fig. 20 illustrated, briefly with reference to a timing diagram of Fig. 21 as well as Fig. 20. First created in Fig. 20, when the unillustrated (manual) power supply switch is closed and the low-voltage power supply switch 12 is closed, the constant-voltage control power source 25 supplies a predetermined control power supply voltage Vcc. The microprocessor 21 then starts its control operation.

[0346] Part (A) of Fig. 21 shows the waveform of the repetitive command signal PLS generated by the arithmetic control circuit 20E. A first period (slow discharge period) T1, in which the repetitive command signal PLS has a logic "L," and a second period (slow charge period) T2, in which the repetitive command signal PLS has a logic "H," may have different values, each occupying a part of the total cycle of T0=T1+T2, but it is preferable that the first period T1 and the second period T2 have the same value.

[0347] In Fig. 20, when the logic level of the repetitive command signal PLS is “L” (or “H”), the lower transistor 31b (or the upper transistor 31a) of the charge / discharge switching element 39 becomes conductive and the other transistor becomes non-conductive, as shown in part (B) of Fig. 21 shown.

[0348] Then in Fig. 20, when the lower transistor 31b of the charge / discharge switching element 39 becomes conductive, the coupling capacitor 51 is slowly discharged through a series connection of the leakage resistor Rx, the rapid charge / discharge resistor 35, and the charge / discharge resistor 33, with the result that the monitoring voltage Vx gradually decreases along with the decrease in the discharge current. On the other hand, when in Fig. 20 the upper transistor 31a of the charge / discharge switching element 39 becomes conductive, the coupling capacitor 51 is slowly charged through the series connection of the leakage resistor Rx, the quick charge / discharge resistor 35 and the charge / discharge resistor 33, with the result that the monitor voltage Vx gradually increases along with the drop in the charging current.

[0349] Part (C) of Fig. Figure 21 shows how the monitoring voltage Vx gradually decreases or gradually increases. The gradient of the charge / discharge characteristics becomes smoother as the charge / discharge time constants expressed in expressions (8x) and (9x) above increase. t1=(R0+Rx)C is a slow discharge time constant, and τ2=τ1=(R0+Rx)C is a slow charge time constant.

[0350] In part (C) of Fig. 21, the value of the slow-discharge monitoring voltage Vx gradually decreases to the first end voltage V1 at the end of the first period T1. At this time, the output logic of the repetitive command signal PLS is inverted, so that the upper transistor 31a of the charge / discharge switching element 39 becomes conductive to start the slow charging of the coupling capacitor 51.

[0351] The value of the slow-charge monitoring voltage Vx gradually rises to the second final voltage V2 at the end of the second period T2. At this time, the output logic of the repetitive command signal PLS is inverted so that the lower transistor 31b of the charge / discharge switching element 39 becomes conductive to start the slow discharge of the coupling capacitor 51.

[0352] Note that the values ​​of the first threshold V1 and the second threshold V2 have the relationship of 0 <V1<V2<Vcc aufweisen und der erste Zeitraum T1 und der zweite Zeitraum T2 dieselbe Zeitbreite aufweisen, solange wie die Beziehung von V1+V2=Vcc erfüllt ist.

[0353] Part (D) of Fig. Figure 21 shows the waveform of a switching preparation signal described below with reference to Fig. 25. The logic of the repetitive command signal PLS is inverted when a first transition time Tx1 or a second transition time Tx2 is calculated after the switching preparation signal is generated. The first transition time Tx1 is a period during which the value of the monitoring voltage Vx gradually decreases from the second threshold value V2 to the first threshold value V1. The second transition time Tx2 is a period during which the value of the monitoring voltage Vx gradually increases from the first threshold value V1 to the second threshold value V2. When V1+V2=Vcc is established, Tx1=Tx2 is established.

[0354] Part (E) of Fig. Figure 21 shows the waveform of a voltage across the coupling capacitor 51 (both-ended voltage E). If the second period T2 is infinite, a discharge start voltage (=charge end voltage) E1 rises to a value obtained by adding the control power supply voltage Vcc to the stable divided voltage Vn0 expressed in expression (4x). If the first period T1 is infinite, the value of a discharge end voltage (=charge start voltage) E2 decreases to the stable divided voltage Vn0.

[0355] Hereinafter, based on the apparatus according to the fifth embodiment of the present invention, which is shown in Fig. 20, a method for detecting the leakage resistance will be explained with reference to characteristic diagrams of Fig. 22 to 24. In Fig. 22, a leakage resistance coefficient β on the horizontal axis is a ratio between the value of the parallel combined resistance of the equivalent leakage resistances R1 and R2, namely the value of the leakage resistance Rx=R1×R2 / (R1+R2) and the series resistance R0=Rs+Rq. The series resistance R0 is a known constant, and therefore the value of the leakage resistance coefficient β=Rx / R0 is proportional to the leakage resistance Rx.

[0356] A transition time coefficient α on the vertical axis is a ratio between a measured transition time Tx and a charge / discharge time constant τ=(R0+Rx)×C=R0×C with respect to the coupling capacitor 51 when the value of the leakage resistance Rx is zero. The charge / discharge time constant τ is a known constant, and therefore the value of the transition time coefficient α=Tx / (R0×C) is proportional to the transition time Tx.

[0357] A number of characteristic curves of Fig. 22 uses a threshold voltage coefficient γ as a parameter. In the case where the value of the transition time Tx is set to an average value between the second transition time Tx2, in which the value of the monitoring voltage Vx gradually increases and shifts from the first threshold value V1 to the second threshold value V2, and the first transition time Tx1, in which the value of the monitoring voltage Vx gradually decreases and shifts from the second threshold value V2 to the first threshold value V1, the threshold voltage coefficient γ is γ(V2-V1) / Vcc. The threshold voltage V2-V1 and the control power supply voltage Vcc are known values, and therefore the value of the threshold voltage coefficient γ is a known value as a design constant.

[0358] The plurality of characteristic curves using the threshold voltage coefficient γ as a parameter are classified into a negative gradient curve, as represented by curves 501 and 502, in which the transition time coefficient α decreases along with the increase in the leakage resistance coefficient β, and a bell-shaped curve, as represented by curves 511, 512, and 513, which is convex in the center and has a positive gradient curve region and a negative gradient curve region.

[0359] Fig. 23 is an enlarged graph of curve 501 of Fig. 22 and shows a negative gradient curve for the threshold voltage coefficient γ of 0.5. In the case where the value of the series resistance R0 is designed to be 5 times the limit leakage resistance Rx0, the value of the leakage resistance coefficient β when the actual value of the leakage resistance Rx has become equal to the limit leakage resistance Rx0 is Rx / R0=Rx0(5Rx0)=0.2 and therefore the value of the leakage resistance coefficient β corresponding to a point of interest P14 of Fig. 23 corresponding transition time coefficient α of 1.0 a value corresponding to the measured transition time Tx.

[0360] Therefore, when the transition time coefficient α is 1.0 or greater, the leakage resistance Rx is determined to be in the dangerous state in which it is equal to or less than the limit leakage resistance Rx0, and when the transition time coefficient α is less than 1.0, the leakage resistance Rx is determined to be in the safe state in which it exceeds the limit leakage resistance Rx0. For example, in the case where the limit leakage resistance Rx0 is set to 200 KΩ, the series resistance R0 is set to 200x5=1000 KΩ, and the electrostatic capacitance C of the coupling capacitor 51 is set to 0.22 µF, the charge / discharge time constant τ is 1000×0.22=220 ms, and therefore the limit transition time Tx0 = α×(R0×C)=1.0×220=220 ms.

[0361] Furthermore, Fig. 23, when the leakage resistance coefficient β is 0, the value of the transition time coefficient α is 1.1. Thus, the occurrence of a complete grounding abnormality can be detected as long as the charge / discharge period of the repetitive command signal PLS has a pulse width equal to or greater than 220 × 1.1 = 242 ms. In contrast, an off-time abnormality is detected when an inverted logic signal can be obtained even 242 ms after the logic of the repetitive command signal PLS is inverted to the charge side.

[0362] In addition, Fig. 23, when the leakage resistance coefficient β of 0.9 (α=0.2) is the upper limit of measurement, the measurable value of the leakage resistance Rx is β×R0=0.9R0=4.5×Rx0. Therefore, it is difficult to accurately measure a leakage resistance that exceeds 4.5 times the limit leakage resistance Rx0, but the normal state of the leakage resistance can be reliably detected.

[0363] Returning to Fig. 22, curve 511 can be used to measure the leakage resistance even by excluding the positive gradient curve region and using only the negative gradient curve region. The horizontal axis of Fig. 22 has a geometric scale, and therefore the transition time coefficient α appears to attenuate rapidly. However, on an arithmetic scale, a smooth attenuation curve is obtained.

[0364] In this case, attention is paid to the leakage resistance coefficient β of 1.0, corresponding to a point of interest P13 (α=1.7), at which the value of the transition time coefficient α for the leakage resistance coefficient β of 0 is 1.7 or less. When the series resistance R0 is set to Rx0, the leakage resistance coefficient β when the leakage resistance Rx has decreased to the limit leakage resistance Rx0 is Rx0 / R0=1.0. Therefore, the value of the limit leakage resistance Rx0 can be detected based on the transition time coefficient α corresponding to the point of interest P13.

[0365] In this case, however, the leakage resistance equal to or less than the limit leakage resistance Rx0 cannot be measured, and if the transition time coefficient α exceeds 1.7, a binary problem occurs and a correct value of the leakage resistance cannot be specified. Furthermore, in the curve 511 of Fig. 22, when the leakage resistance coefficient β of 2.0 (α=0.6) is the upper limit of measurement, the measurable value of the leakage resistance Rx is β×R0=2Rx0. Therefore, it is difficult to accurately measure a leakage resistance that exceeds twice the limit leakage resistance Rx0, but the normal state of the leakage resistance can be reliably detected.

[0366] If a resistance equal to or less than the limit leakage resistance can be measured, the measured resistance can be used as information for searching for the cause of a decrease in leakage resistance. However, in general, there is no problem even if the resistance equal to or less than the limit leakage resistance Rx0 cannot be measured. Furthermore, in general, as long as the pre-alarm leakage resistance, which is about 1.3 times the limit leakage resistance Rx0, can be measured, it is unnecessary to accurately measure a larger leakage resistance. From this point of view, it is possible to use both the negative gradient curve 501 and the curve 511 with the negative gradient curve region following the positive gradient curve region. However, there is no advantage in using the negative gradient region of the bell-shaped curve, and it is more advantageous to use the curve 501 with only the negative gradient curve.

[0367] Fig. Figure 24 is an enlarged graph of curve 513 of Fig. 22 and shows a positive gradient curve for the threshold voltage coefficient γ of 0.92. However, the value of the series resistance R0 is set to 2.5 times the value of the limit leakage resistance Rx0, and the additional resistance R3=10×Rx0 is connected in parallel through the bypass leakage resistance 53. Therefore, even if the actual leakage resistance Rx is infinite, the upper limit of the leakage resistance coefficient β is R3 / R0=10×Rx0 / (2.5×Rx0)=4, which is within the positive gradient curve region of the curve 513 of Fig. 22 is.

[0368] In case of Fig. 24, the value of the leakage resistance coefficient β when the actual value of the leakage resistance Rx has become equal to the limit leakage resistance Rx0 is (Rx0 / / R3) / R0=(Rx0 / / 10Rx0) / (2.5Rx0)=0.36, and therefore the transition time coefficient α of 3.88 corresponding to a point of interest P15 of Fig. 24 a value corresponding to the measured transition time Tx.

[0369] For example, in the case where the limit leakage resistance Rx0 is set to 200 kΩ, the additional resistance R3 is 2 mΩ, and the series resistance R0 is 500 kΩ. Thus, the charge / discharge time constant as a reference when the electrostatic capacitance C of the coupling capacitor 51 is set to 0.15 μF becomes R0xC=500x0.15=75 ms. Therefore, the limit transition time Tx0 when the leakage resistance Rx has decreased to the limit leakage resistance Rx0 is α×(R0×C)=3.88×75=291 ms. Thus, the leakage resistance Rx can be determined to be in a dangerous state when the transition time Tx becomes 291 ms or less, and in the safe state when the transition time Tx exceeds 291 ms.

[0370] Furthermore, in Fig. 24, the value of the transition time coefficient α is at an upper limit of the leakage resistance coefficient β of 4.0 7.0. Thus, theoretically, the leakage resistance can be detected up to an infinite value as long as the loading period of the repetitive command signal PLS has a pulse width equal to or longer than 75 x 7.0 = 525 ms. On the other hand, an off-time abnormality is detected if an inverted logic signal cannot be obtained even after 525 ms since the logic of the repetitive command signal was inverted.

[0371] In the above description, the upper limit of the leakage resistance coefficient β through the bypass leakage resistance 53 for the limited use of the positive gradient curve region of the bell-shaped curve in Fig. 22. The bypass leakage resistor 53 has a known resistance value. Thus, at the time of product delivery inspection, by measuring the leakage resistance in the state in which the onboard high-voltage device 60X is not connected, the individual fluctuations in the electrostatic capacitance C of the coupling capacitor 51 can be measured, so that a calibrated value of the electrostatic refresh period C can be calculated and stored.

[0372] The Fig. The characteristic curves shown in Figure 22 are based on the Fig. Expression II shown in Figure 22, and Expression II is established when the monitor voltage Vx falls within a proper range from 0 to Vcc. The establishment of Expression II is demonstrated as follows. Unloading period

[0373] In the slow discharge period, in which the logic level of the repetitive command signal PLS in Fig. 20 is "L" and the lower transistor 31b of the charge / discharge switching element 39 is closed, expression (50b) is established. V×R0×C(dE / dt)=0 where R0 = Rs + Rq≈>>Rq

[0374] Vx in expression (50b) is substituted into expression (7x) above to obtain expression (51b) Vn0 - E + τ1×(dE / dt) where τ1 = (R0 + Rx)C≈(Rx + Rx)C

[0375] In differential equation (51b), the initial value of E at time t = 0 is represented by E1, and the value of E at infinite time is represented by Vn0. Then, a solution obtained by expression (52b) is obtained. E = E1exp(-t / τ1)+(Vn0{1 - exp(−t / τ1)}

[0376] In addition, the value of (dE / dt) in expression (50b) is substituted into expression (7x) above to obtain expression (43b). Vx / (R0 / / Rx)=(E-Vn0) / Rx where R0 / / Rx represents the parallel combined resistance of R0 and Rx. Charging period

[0377] In the slow loading period, in which the logic level of the repetitive command signal PLS in Fig. 20 is "H" and the upper transistor 31a of the charge / discharge switching element 39 is closed, expression (50a) is established. C×R0(dE / dt)=Vcc - Vx

[0378] Vx in expression (50a) is substituted into expression (7x) above to obtain expression (51a). Vn0 + Vcc = τ2(dE / dt)+E where τ2 = τ1 = (R0 + Rx)×C.

[0379] In differential equation (51a), the initial value of E at time t=0 is represented by E2, and the value of E at infinite time is represented by Vn0+Vcc. Then, a solution expressed by expression (52a) is obtained. E = E2exp(-t / τ2)+(Vn0 + Vcc){1−exp(-t / τ2)}

[0380] In addition, the value of (dE / dt) in expression (50a) is substituted into expression (7x) to obtain expression (53a). Vx / (R0 / / Rx)=(E-Vn0) / Rx + Vcc / R0 where R0 / / Rx represent the parallel combined resistance of R0 and Rx. Stable condition

[0381] In the stable state where the logic level of the repetitive command signal PLS is Fig. 20 is alternately inverted in the discharge period T1 and the charge period T2, and there is no change in the value of the leakage resistance Rx in the output voltage Vh of the high-voltage DC power source 61, the initial values ​​E1 and E2 in expressions (52b) and (52a) are calculated as follows. First, in expression (52b), at the discharge end time t=T1, E=E2 is established, and therefore expression (54b) is established. E2=E1×K1+Vn0(1−K1) where K1=exp(−T2 / τ1)

[0382] Additionally, in expression (52a), at the loading end time T=T2, E=E1 is established and therefore expression (54a) is established. E1=E2×K2+(Vn0+Vcc)(1−K2) where K2 = exp(-T2 / τ2)

[0383] From expressions (54b) and (54a) expressions (55b) and (55a) are obtained. E1 - Vn0 = Vcc×K0 E2 - Vn0 = Vcc×K0×K1 where K0 = (1-K2) / (1−K1×K2) In the case of the fifth embodiment

[0384] In the Fig. 20 and Fig. 21 is established at the discharge start time t=0, Vx=V2. At time t=Tx1=T1, if the value of Vx becomes equal to the set threshold voltage V1, expressions (59b) and (60b) are calculated from expressions (53b) and (55a). V1 / (R0 / / Rx)=(E2-Vn0) / Rx = VccK0K1 / Rx ∴V1 / Vcc =(R0 / / Rx)K0K1 / Rx

[0385] At the charging start time t=0, Vx is equal to V1. At time t=Tx2=T2, if the value of Vx becomes V2, expression (59a) is obtained from expressions (53a) and (55a). V2 / R0 / / Rx)=(E1 - Vn0) / Rx + Vcc / R0 =VccK0 / Rx+Vcc / R0 ∴V2 / Vcc =(R0 / / Rx)(K0 / Rx + 1 / R0)

[0386] Then, the threshold voltage coefficient γ1 of V1 / Vcc and a leakage resistance coefficient β2 of V2 / Vcc, the leakage resistance coefficient β of Rx / R0 and the transition time coefficient α of Tx / (R0C) are substituted into expressions (60a) and (60b), and expressions (61a) and (61b) are obtained. γ=K0K1 / (1+β) γ2 =(k0 + β) / (1+β)

[0387] If γ1+γ2=1 is established, K1=K2 is obtained from expressions (61a) and (61b). If K1=K2=K is established, K0=1 / (1+K) is established. Therefore, expressions (61b) and (61a) are modified to expressions (62b) and (62a). γ1=K / {(1+K) (1+β)} γ2 ={1+β(1+K)} / {(1 + K)}(1 + β)

[0388] Therefore, when y=y2-y1 is established, expression (63) is obtained from expressions (62a) and (62b). γ=γ2−γ1=1−2γ1=1−2K / {(1+K)(1+β) 1 / K is recalculated from expression (63) to obtain expression (64). 1 / K =[2 / {(1−γ)(1−β)}]−1

[0389] On the other hand, if Tx1=T1=Tx2=T2=Tx is established and the transition time coefficient α is set to Tx / (ROC), expression (65) is obtained. K = exp[−Tx / {(R0 + Rx)C}]=exp{-α / (1+β)}

[0390] Expression (65) is converted to a natural logarithm LOGe to obtain expression (66). α =(1+β)LOGe(1 / Kx)

[0391] Expressions (64) and (66) correspond to the above in Fig. 22. Note that expression (64) is calculated using γ=1-2γ1 in expression (62b), but alternatively, γ=2γ2-1 can be used in expression (62a) or γ=γ2-γ1 can be used in both expression (62a) and expression (62b) to obtain expressions (64) and (65).

[0392] Next, the action and operation of the device will be Fig. 20 in detail with reference to operational flow charts included in the Fig. 25 to 27. Note that the step numbers of Fig. 6 and Fig. 7 are in the 1000s, while the step numbers of Fig. 25 to 27 are in the 5000s, and that the same 3-digit or lower numbers represent the same or equivalent parts. Thus, some of the description of the same parts is omitted.

[0393] First, referring to the first flowchart of Fig. 25, step 5600 is a step in which the microprocessor 21 starts the operation of detecting the leakage resistance Rx. The subsequent step 5600a is a determination step of determining whether the current operation is the first operation after the low-voltage power supply switch 12 has been closed. In the case of the first cycle, this is determined as YES, and the flow proceeds to step 5600b. In the case of a continuous cyclic cycle, it is determined as NO, and the flow proceeds to step 5611. Step 5600b is an initialization step in which the current value of a timer to be described below is set to 0.

[0394] Following step 5611 is a determination step of monitoring the analog signal voltage ANL to determine whether the monitor voltage Vx is equal to or higher than the first threshold V1. If the monitor voltage Vx is equal to or higher than the first threshold V1, this is determined as YES and the flow proceeds to step 5613b. If the monitor voltage Vx is lower than the first threshold V1, this is determined as NO and the flow proceeds to step 5613a. Step 5611 is equivalent to means for generating the first inverted logic signal LOW.

[0395] Step 5613a is a step of temporarily storing an inversion switching command for inverting the logic level of the repetitive command signal PLS to "H" so that the upper transistor 31a of the charge / discharge switching element 39 becomes conductive. Actually, the inversion switching is executed in the next calculated cycle in step 5604b, to be described below.

[0396] In step 5613b, the current logic state of the repetitive command signal PLS is maintained, and the flow proceeds to step 5612. The subsequent step 5612 is a determination step of monitoring the analog signal voltage ANL to determine whether the monitor voltage Vx is equal to or lower than the second threshold V2. If the monitor voltage Vx is equal to or lower than the second threshold V2, this is determined as YES, and the flow proceeds to step 5614b. If the monitor voltage Vx exceeds the second threshold V2, this is determined as NO, and the flow proceeds to step 5614a. Step 5612 is equivalent to means for generating the second inverted logic signal HIg.

[0397] Step 5614a is a step of temporarily storing an inversion switching command for inverting the logic level of the repetitive command signal PLS to "L" so that the lower transistor 31b of the charge / discharge switching element 39 becomes conductive. Actually, the inversion switching is performed in the next calculation step in step 5604b, to be described below. In step 5614b, the current logic state of the repetitive command signal PLS is maintained, and the flow proceeds to step 5606c.

[0398] Step 5606c is executed if it is determined in step 5611 and step 5612 that the value of the monitoring voltage Vx is within the correct range of the first threshold value V1 or higher and the second threshold value V2 or lower. In step 5606c, a timer 0, which was activated in step 5606a to be described below, is initialized, and the flow proceeds to step 5700 of Fig. 27 via a continuation A. A series of steps from step 5611 to step 5614b forms a step block 5605.

[0399] Step 5606a is executed when it is determined in step 5611 or step 5612 that the value of the monitoring voltage Vx is within the deviated range of lower than the first threshold V1 or higher than the second threshold V2. In step 5606a, the transient abnormality determination timer 0 is activated to start the counting operation, and the flow proceeds to step 5604a.

[0400] Step 5604a is a determination step. If the current calculation cycle is the same as that immediately after the execution of step 5613a or step 5614a as described above, this is determined as NO and the flow advances to step 5606b of Fig. 26 via a continuation C. When the flow advances to step 5604a after executing a series of control operations up to an operation termination step 5610 to be described later and re-executing the flow of the operation start step 5600 and subsequent steps to step 5604a, it is determined as YES and the flow advances to step 5604b.

[0401] In step 5604b, the output logic of the repetitive command signal PLS is actually inverted based on the storage state of the shift preparation flag preliminarily stored in step 5613a or step 5614a described above, and the shift preparation flag is reset. The flow then proceeds to step 5606b of Fig. 26 via continuation C. A step block 5604, which includes step 5604a and step 5604b, is a repetitive command signal generating means.

[0402] Referring to the second flowchart of Fig. 26, step 5606b is a step serving as transient abnormality determination means for reading the current value of timer 0 that started counting in step 5606a to determine whether a predetermined period of time has elapsed. If the monitor voltage Vx has not yet returned to the proper range of V1 to V2 after the predetermined period of time has elapsed, it is determined as YES, and the flow proceeds to step 5607a. If the monitor voltage Vx has returned to the proper range of V1 to V2 within the predetermined period of time, it is determined as NO, and the flow proceeds to step 5607b.

[0403] Step 5607a is transient characteristic abnormality processing means. For example, if the power supply voltage Vh of the onboard high-voltage DC power source 61 has changed abruptly, or if the positive-side leakage resistance 65 or the negative-side leakage resistance 66 has changed abruptly to change the potential of the negative-side power supply line 67 of the onboard high-voltage DC power source 61 relative to the vehicle body at the connection point B of the coupling capacitor 51, then the potential of the measurement point A transiently becomes equal to or lower than the vehicle body potential (0 volts) or equal to or higher than the control power supply voltage Vcc, and the coupling capacitor 51 is charged or discharged through the bypass diode 36 or the bypass diode 37.In this case, when a transient charge / discharge time during which the monitoring voltage Vx returns to the proper range of V1 to V2 corresponding to the voltage of the measuring point A for measuring the leakage resistance Rx has elapsed, this is detected as the transient characteristic abnormality ERR2 and the abnormality alarm (not shown) is performed, followed by the transition to a return operation mode (limp home operation, emergency operation program).

[0404] Note that in the case where the vehicle state signal 18 is predicted to change and the monitor voltage Vx temporarily deviates from the proper range, the determination of the transient characteristic abnormality ERR2 is avoided at least for a predetermined period of time.

[0405] Note that in the case where no system abnormality occurs, such as a decrease in internal leakage resistance due to degradation of the coupling capacitor 51, disconnection of other wiring, and short circuit, the maximum value of the transient charge / discharge time is shorter than the predetermined determination period in step 5606b. In this case, it is determined as NO in step 5606b, and the flow proceeds to step 5607b.

[0406] Step 5607b is performed in the case where it is determined in step 5606b that a period of time that has elapsed since the timer 0 was activated in step 5606a is within the predetermined period. In step 5607b, the Fig. 27 is suspended. The flow proceeds to step 5607a or step 5607b or to the step shown in Fig. 27, the continuation B proceeds to step 5608.

[0407] Step 5608 is a determination step of determining whether the current timing is within a return period, and saving the value of the leakage resistance Rx calculated in a step to be described below, or abnormality occurrence information, into the non-volatile data memory 23. For example, if the current timing is in the delayed supply period in which the manual power supply switch (not shown) is opened and then the low-voltage power supply switch 12 is interrupted, this is determined as YES, and the flow proceeds to step 5609a. If the current timing is not within the return period, this is determined as NO, and the flow proceeds to an operation end step 5610. Note that in step 5608, the determination of YES may be made periodically at a predetermined time interval.

[0408] In step 5609a, moving average values ​​of the leakage resistance Rx calculated in steps 5706 and 5716 to be described below are sequentially written and stored while updating the address of the non-volatile data memory 23. In the subsequent step 5609b, based on determination information about the transient characteristic abnormality ERR2 written into the RAM 22 in step 5607a or based on the abnormality occurrence information determined in steps 5709 and 5719 to be described below, the accumulated number of abnormality occurrences is updated according to the type of abnormality occurrence and then written and stored at a predetermined address of the non-volatile data memory. Thereafter, the flow proceeds to the operation end step 5610.

[0409] In operation end step 5610, the microprocessor 21 executes another control program, and the flow advances again to operation start step 5600 after a standby time of, for example, 10 ms. Therefore, immediately after the start of operation and in a state where the initial charge of the coupling capacitor 51 has not been performed, the microprocessor 21 executes the steps in the first cycle in the order of step 5600, step 5600a (determination of YES), step 5600b, step block 5605, step 5606a, step block 5604, step 5606b (determination of NO), step 5607b, step 5608 (determination of NO), step 5610, and step 5600 after a standby time of 10 ms. In the next and subsequent cycles, the microprocessor 21 executes a series of processing in a cyclic manner in a state in which the determination of step 5600a is NO and step 5600b is not executed.

[0410] In the process, if it is determined in step block 5605 that the monitor voltage Vx falls within the proper range, timer 0 is initialized in step 5606c, and the flow proceeds to step 5700 to be described below. On the other hand, in a period in which it is still determined in step block 5605 that the monitor voltage Vx falls outside the proper range and it is determined in step 5606b that the predetermined time has not been reached, the steps in the order of step 5607b, step 5608 (determination of NO), and step 5610 are executed, and the cyclic operation continues. Then, if YES is determined in step 5606b, step 5607a is executed.

[0411] Referring to the third flowchart of Fig. 27, step 5700 is a step block equivalent to steps 2700a to 2700c described above with reference to Fig. 11. Step 5700 is a shipment calibration means for measuring the value of the electrostatic capacitance C of the coupling capacitor 51 at the time of shipment adjustment and storing a calibrated value thereof.

[0412] The subsequent step 5701 is a determination step of monitoring the logic level of the repetitive command signal PLS obtained in step 5604b of Fig. 25 is executed. In the period in which the logic level of the repetitive command signal PLS is "H" and the upper transistor 31a of the charge / discharge switching element 39 becomes conductive to charge the coupling capacitor 51 via the charge / discharge resistors 33 and 35, this is determined as YES and the flow proceeds to step 5702. In the period when the logic level of the repetitive command signal PLS is "L" and the lower transistor 31b of the charge / discharge switching element 39 becomes conductive to slowly discharge the coupling capacitor 51 via the charge / discharge resistors 33 and 35, this is determined as NO and the flow proceeds to step 5712.

[0413] In step 5702, a timer 1 for measuring the transition time Tx is activated to start the counting operation, and the flow proceeds to step 5703. Step 5703 is a determination step. If logic inversion preparation information of the repetitive command signal PLS is temporarily stored in step 5614a of Fig. 25 have been stored, it is determined as YES and the flow proceeds to step 5704a. If the logic inversion preparation information has not been temporarily stored in step 5614a, it is determined as NO and the flow proceeds to step 5704b.

[0414] In step 5704a, the current value of timer 1, which has been activated to start counting in step 5702, is read out and stored in RAM 22. The following step 5705 is a step serving as leakage resistance calculation means for calculating the leakage resistance Rx based on the value of the transition time Tx read out and stored in step 5704a.

[0415] In the subsequent step 5706, the current value of the leakage resistance Rx calculated in step 5705 is input to the first stage of the shift register formed by the RAM 22, and portions of past data stored in the shift register are sequentially migrated to the subsequent stage. The past data of the leakage resistance Rx stored in the last stage are output and cleared, and the sum of the values ​​of the leakage resistance Rx remaining in the shift register is divided by the number of values ​​in the shift register divided by the stored number, thereby calculating the moving average value.In this way, it is possible to prevent the measurement from being directly affected by a temporary measurement error of the leakage resistance Rx caused, for example, by the temporal fluctuations in the power supply voltage Vh of the on-board high-voltage DC power source 61 or by noise.

[0416] The subsequent step 5707a is a step that serves as resistance abnormality determination means. If the moving average value of the leakage resistance Rx calculated in step 5706 becomes equal to or less than the provisional leakage resistance Rxn or the limit leakage resistance Rx0, this is determined as YES and the flow proceeds to step 5709. If there is no abnormality, this is determined as NO and the flow proceeds to step 5708.

[0417] Note that in the case where there is a difference of a predetermined value or more between the moving average value of the leakage resistance Rx calculated in step 5706 and that calculated in step 5609a of Fig. 26 in the non-volatile data memory 23, and if it is detected that the leakage resistance Rx has abruptly decreased, if not to the limit leakage resistance Rx0, this is determined as YES in step 5707a and the flow proceeds to step 5709. Step 5708 is a step in which the timer 1, which has been activated in step 5702, is initialized and suspended, and the flow proceeds via continuation B to step 5608 of Fig. 26 continued.

[0418] Step 5704b is executed in the case where NO is determined in step 5703. In step 5704b, in order to measure the elapsed time from when the logic level of the repetitive command signal PLS has changed to "H" until when the switching command for inverting the logic is stored in step 5714a, the current value of timer 1 activated in step 5702 is updated and read out. The subsequent step 5707b serves as a timeout abnormality determination step. If the elapsed time read out and stored in step 5704b is excessive and exceeds a predetermined period, this is determined as YES, and the flow proceeds to step 5709. If the elapsed time is within the predetermined period, this is determined as NO, and the flow proceeds to step 5704 in Fig. 26 on the continuation B.

[0419] Step 5709 is a step serving as measurement abnormality processing means. If YES is determined in step 5707a, the measurement abnormality processing means generates the pre-alarm output ER2 or the resistance abnormality determination output ER1 and stores abnormality determination information for a state change abnormality ERR3. If YES is determined in step 5707b, the measurement abnormality processing means stores the abnormality determination information for the time-out abnormality determination ERR1. After step 5709, the flow proceeds to step 5608 of Fig. 26 on the continuation B.

[0420] In step 5709, when the pre-alarm output ER2 or the resistance abnormality determination output ER1 is generated, the abnormality alarm device 19 is activated. If the abnormality determination information for the time-out abnormality determination ERR1 or the transient state change abnormality ERR3 is stored, the stored abnormality determination information is separated from other abnormality information. In step 5609b of Fig. 26, the accumulated number of abnormality occurrences is written and saved in the non-volatile data memory 23, which is used for maintenance and inspection operations.

[0421] A sequence of operations from step 5712 to step 5719 has a one-to-one correspondence with the sequence from step 5702 to step 5709, and the 2-digit numbers are changed from 0 to 1. Steps 5702 and 5709 described above are performed when the determination in step 5701 is YES. In steps 5702 to 5709, the value of the leakage resistance Rx is measured or the presence or absence of an abnormality is monitored in the slow charge period in which the output logic of the repetitive command signal PLS is "H".

[0422] On the other hand, if the determination in step 5701 is NO, steps 5712 to 5719 are performed. In steps 5712 to 5719, the value of the leakage resistance Rx is measured or the presence or absence of an abnormality is monitored in the slow discharge period in which the output logic of the repetitive command signal PLS is "L." Note that the measurement of the leakage resistance Rx may be performed in only one of the series processes from step 5712 to step 5719 and the series processes from step 5702 to step 5709.

[0423] Alternatively, the leakage resistance may be calculated as a moving average based on an average value between the transition time measured in step 5704 and the transition time measured in step 5714a. Regarding the shift register for calculating the moving average value, the same shift register may alternately and sequentially store the leakage resistance value measured in step 5705 and the leakage resistance value measured in step 5715, thereby calculating a single moving average value.

[0424] Note that the calculation of the leakage resistance Rx in steps 5705 and 5175 is performed in three ways. First, the calculation is performed on the characteristic curve whose overall region is the negative gradient curve region, as shown by the curve 501 in Fig. 22 represents based on. Second, the calculation is performed using the negative gradient curve region of a bell-shaped curve, as represented by the curve 511 of Fig. 22. Thirdly, the calculation is performed using the positive gradient curve region of a bell-shaped curve as represented by the curve 513 of Fig. 22. In any case, the value of the leakage resistance Rx is calculated from the measured value at the transition time Tx based on a functional formula or a data table between the transition time coefficient α and the leakage resistance coefficient β using the threshold voltage coefficient γ as a parameter.

[0425] As an alternative abnormality determination method, instead of comparing the detected leakage resistance Rx with the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, the preliminary transition time Txn or the limit transition time Tx0 may be calculated in advance according to the preliminary leakage resistance Rxn or the limit leakage resistance Rx0, and the transition time Tx may be compared with the limit transition time Txn or the limit transition time Tx0.

[0426] Furthermore, in the case where the value of the reference time constant R0×C, which is the denominator of the transition time coefficient α, the value of the series resistance R0, which is the denominator of the leakage resistance coefficient β, or the value of the threshold voltage coefficient γ is considered a fixed value, a formula or a data table between the transition time Tx and the leakage resistance Rx can be directly used as the functional formula or data table. In this case, the leakage resistance Rx can be directly calculated from the measured transition time Tx without performing complicated calculation processing during operation.

[0427] However, in the case of varying the value of the threshold voltage coefficient γ, the denominator value of the transition time coefficient α, or the denominator value of the leakage resistance coefficient β according to the applied vehicle types or the operating condition, it is advantageous to set the functional formula or the data table using α, β, or γ, which are unitless index values. Alternatively, in the case where the number of analog input terminals of the arithmetic control circuit 20E is insufficient, the operational amplifier 49 of the monitor signal processing circuit 40E may be replaced by a pair of comparators or an inversion memory circuit, as shown in Fig. 12, or a comparator with a hysteresis operation function, as in Fig. 16. In this case, the leakage resistance can be measured via slow charge and slow discharge using the charge / discharge switching element 39. In addition, in the fifth embodiment of Fig. 20, the connection point B is connected to the negative-side power supply line 67 of the on-board high-voltage device 60X, but even if the connection point B is connected to a positive-side power supply line, no change occurs in Expression II.

[0428] As is clear from the above description, according to the fifth embodiment of the present invention, the leakage resistance detection device 50E is provided for an on-board high-voltage device connectable to the on-board high-voltage device 60X, including the on-board high-voltage DC power source 61 and the high-voltage electrical load to be fed and driven by the on-board high-voltage DC power source 61, wherein the on-board high-voltage device 60X has the leakage resistance Rx with respect to the vehicle body 11, as typified by the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side, wherein the leakage resistance detection device 50E includes the constant-voltage control power source 25 fed and driven by the low-voltage DC power source 10,which has the negative terminal connected to the vehicle body 11 and which generates the control power supply voltage Vcc, wherein the leakage resistance detection device 50E is configured to measure a value of the leakage resistance Rx via the coupling capacitor 51 having a terminal B connected to a predetermined position of the on-board high-voltage device 60X, wherein the leakage resistance detection device 50E further includes the repetitive signal output circuit 30E, the monitoring signal processing circuit 40E, and the arithmetic control circuit 20E, wherein the arithmetic control circuit 20E includes the microprocessor 21 and the program memory 24E, which cooperate with each other.

[0429] Furthermore, the repetitive signal output circuit 30E alternately repeats, in response to the switching operation of the charge / discharge switching element 39 operating in response to the repetitive command signal PLS, a charge period and a discharge period in which the measuring point A is intermittently connected to the control power supply voltage Vcc via the charge / discharge resistor 33 or 35, the measuring point A being another terminal of the coupling capacitor 51, to thereby gradually increase or gradually decrease the monitor voltage Vx, which is a potential between the measuring point A and the vehicle body 11.The monitor signal processing circuit 40E is configured to: generate at least one of the inverted logic signals ANS, LOW, and HIG when a value of the monitor voltage Vx, which gradually increases or gradually decreases, passes a predetermined threshold voltage at a gradient that becomes smoother as the charge / discharge time constant increases, and input at least one of the inverted logic signals ANS, LOW, and HIG to the arithmetic control circuit 20E; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20E.The arithmetic control circuit 20E is configured to: measure the transition time Tx during which the value of the monitor voltage Vx changes from a threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to another; calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and generate the resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than the limit leakage resistance Rx0 or when the transition time Tx reaches the limit transition time Tx0 corresponding to the predetermined limit leakage resistance Rx0.The repetitive command signal PLS is a pulse signal with a variable cycle in which an output pulse is inverted when the transition time Tx is reached.

[0430] The repetitive signal output circuit 30E is configured to, in response to the switching operation of the charge / discharge switching element 39, connect the measurement point A, which is the other terminal of the coupling capacitor 51, to an output terminal of the constant-voltage control power source 25 via the reference resistor 33 having the resistance value Rs, or to connect the measurement point A to the negative terminal having a vehicle body potential via the same reference resistor 33, thereby gradually increasing or gradually decreasing the monitor voltage Vx, which is the potential between the measurement point A and the vehicle body 11. The monitor signal processing circuit 40E is configured to: generate the second inverted logic signal ANS or HIG when the value of the monitor voltage Vx, which gradually increases or gradually decreases at a gradient that becomes smoother as the time constant (Rs+Rx)C,which is a product of the sum of the resistance value Rs and the leakage resistance Rx with the electrostatic capacitance C of the coupling capacitor 51, increases, has increased to pass the second threshold value V2, and generate the first inverted logic signal ANS or LOW when the value of the monitor voltage Vx has decreased to pass the first threshold value V1, and input the second inverted logic signal ANS or HIG and the first inverted logic signal ANS or LOW to the arithmetic control circuit 20E; or generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit 20E. The arithmetic control circuit 20E receives the second inverted logic signal ANS or HIG as an input and generates the second inverted logic signal ANS or HIG,based on the input analog signal voltage ANL, and the arithmetic control circuit 20E receives the first inverted logic signal ANS or LOW as an input, or generates the first inverted logic signal ANS or LOW based on the input analog signal voltage ANL.

[0431] The arithmetic control circuit 20E is configured to: measure, at the second transition time Tx2, a period from when the first inverted logic signal ANS or LOW is generated until the second inverted logic signal ANS or HIG is received; measure, at the first transition time Tx1, a period from when the second inverted logic signal ANS or HIG is generated until the first inverted logic signal ANS or LOW is received; and set the first transition time Tx1 or the second transition time Tx2 or an average value of both the first transition time Tx1 and the second transition time Tx2 as the transition time Tx. The arithmetic control circuit 20E or the monitor signal processing circuit 40E inverts the repetitive command signal PLS when the first inverted logic signal ANS or LOW and the second inverted logic signal ANS or HIG are input or generated.The first inverted logic signal and the second inverted logic signal are a pair of logic signals of the first logic signal LOW whose output logic changes before and after the first threshold value V1, and the second logic signal HIG whose output logic changes before and after the second threshold value V2, or the band history logic signal ANS that becomes a first logic state of "L" or "H" when the value of the monitor voltage Vx has risen to the second threshold value V2 or higher after falling to the first threshold value V1 or lower, and that becomes a second logic state of "H" or "L" when the value of the monitor voltage Vx has fallen to the first threshold value V1 or lower after falling to the second threshold value V2 or higher.

[0432] As described above, in conjunction with claim 7 of the present invention, the repetitive signal output circuit charges and discharges the coupling capacitor, and the arithmetic control circuit measures, as the transition time Tx, the second transition time Tx2 during which the monitor voltage Vx increases from the first threshold value V1 to the second threshold value V2, or the first transition time Tx1 during which the monitor voltage Vx decreases from the second threshold value V2 to the first threshold value V1, or the average value of both the second transition time Tx2 and the first transition time Tx1. Further, the monitor signal processing circuit or the arithmetic control circuit inverts the pulse output when the monitor voltage Vx has decreased to pass the first threshold value V1 and when the monitor voltage Vx has increased to pass the second threshold value V2.

[0433] Therefore, the output pulse is alternately inverted along with the completion of the transition time Tx measurement, and therefore the occurrence frequency of the repetitive command signal PLS is shortened as the leakage resistance decreases. Thus, an abnormality can be quickly detected. Furthermore, by using both the first pulse and the second pulse of the repetitive command signal PLS, the leakage resistance measurement can be repeated over and over again.

[0434] As is clear from the above description, according to the fifth embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is the total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30E, is represented by R0, and when a period,during which the monitoring voltage Vx alternately gradually increases and gradually decreases between the first threshold value V1 and the second threshold value V2, represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to (V2-V1) / Vcc: calculating, by the arithmetic control circuit 20E, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a resistance value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region,in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; selectively using a negative gradient curve region following the positive gradient curve region or a negative gradient curve region occupying an entire region, and determining the threshold voltage coefficient y such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the negative gradient curve region; determining, when the transition time Tx has become a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1 and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the short-circuit transition time Tx00, corresponding to a case,where the leakage resistance Rx becomes 0.,

[0435] As is clear from the above description, according to the fifth embodiment of the present invention, the leakage resistance detection method for an on-board high-voltage device is provided, which is used for the leakage resistance detection device for an on-board high-voltage device, the leakage resistance detection method including when the electrostatic capacitance of the coupling capacitor 51 is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance 65 on the positive potential side and the equivalent leakage resistance 66 on the negative potential side is represented by Rx, and a series resistance value, which is the total value of the charge / discharge resistors 33 and 35 provided in the repetitive signal output circuit 30E, is represented by R0, and when a period,during which the monitoring voltage Vx alternately gradually increases and gradually decreases between the first threshold value V1 and the second threshold value V2, represented by the transition time Tx, the transition time coefficient α is set to Tx / (R0×C), the leakage resistance coefficient β is set to Rx / R0, and the threshold voltage coefficient γ is set to (V2-V1) / Vcc: calculating, by the arithmetic control circuit 20E, the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a resistance value of the leakage resistance coefficient β, the characteristic curve having a negative gradient curve region or a positive gradient curve region,in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; precluding, by the characteristic curve correcting means, the application of a negative gradient curve region following the positive gradient curve region; determining the threshold voltage coefficient γ so that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the corrected positive gradient curve region; determining, when the transition time Tx has become a value equal to or less than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and performing the off-time abnormality determination ERR1 when the transition time Tx exceeds the open-circuit transition time Txm corresponding to a case,where the leakage resistance Rx is infinite.,

Claims

[1] Leakage resistance detection device (50A, 50B, 50C, 50D, 50E) for an on-board high-voltage device, which is connectable to an on-board high-voltage device (60X, 60Y) which includes an on-board high-voltage direct-current power source (61) and a high-voltage electrical load (64) to be supplied and driven by the on-board high-voltage direct-current power source (61), wherein the on-board high-voltage device (60X, 60Y) has a leakage resistance Rx with respect to a vehicle body (11) as typified by an equivalent leakage resistance (65) on a positive potential side and an equivalent leakage resistance (66) on a negative potential side, wherein the leakage resistance detection device (50A, 50B, 50C, 50D, 50E) comprises a constant-voltage control power source (25) which is supplied and driven by a low-voltage direct-current source (10) connected to a negative terminal of the vehicle body (11) and which generates a control power supply voltage Vcc, wherein the leakage resistance detection device (50A, 50B, 50C, 50D, 50E) is configured to measure a value of the leakage resistance Rx via a coupling capacitor (51), one terminal B of which is connected to a predetermined position of the on-board high-voltage device (60X, 60Y), wherein the leakage resistance detection device (50A, 50B, 50C, 50D, 50E) further comprises: a repetitive signal output circuit (30A, 30B, 30C, 30D, 30E), a monitoring signal processing circuit (40A, 40B, 40C, 40D, 40E), and an arithmetic control circuit (20A, 20B, 20C, 20D, 20E), wherein the arithmetic control circuit (20A, 20B, 20C, 20D, 20E) comprises a microprocessor (21) and a program memory (24A, 24B, 24C, 24D, 24E) which cooperate with each other, wherein the repetitive signal output circuit (30A, 30B, 30C, 30D, 30E) alternately repeats, in response to a switching operation of a charge / discharge switching element (31 or 39) operating in response to a repetitive command signal PLS, a charge period and a discharge period in which a measuring point A is intermittently connected to the control power supply voltage Vcc via a charge / discharge resistor (33 or 35), the measuring point A being another terminal of the coupling capacitor (51), to thereby gradually increase or gradually decrease a monitor voltage Vx, which is a potential between the measuring point A and the vehicle body (11), wherein the monitoring signal processing circuit (40A, 40B, 40C, 40D, 40E) is configured to: generating at least one of the inverted logic signals ANS, LOW and HIG when a value of the monitor voltage Vx, which gradually increases or gradually decreases with a gradient that becomes smoother as the charge / discharge time constant becomes longer, passes a predetermined threshold voltage and inputting the at least one of the inverted logic signals ANS, LOW and HIG to the arithmetic control circuit (20A, 20B, 20C, 20D, 20E); or to generate an analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit (20A, 20B, 20C, 20D, 20E); wherein the arithmetic control circuit (20A, 20B, 20C, 20D, 20E) is configured to: to measure a transition time Tx during which the value of the monitoring voltage Vx changes from one threshold voltage equal to or higher than 0 volts or a threshold voltage equal to or lower than the control power supply voltage Vcc to another; to calculate the leakage resistance Rx, which is a parallel combined resistance of the equivalent leakage resistance (65) on the positive potential side and the equivalent leakage resistance (66) on the negative potential side, based on a functional formula or a data table related to the transition time Tx and the leakage resistance Rx; and to generate a resistance abnormality determination output ER1 when the leakage resistance Rx becomes equal to or less than a predetermined limit leakage resistance Rx0 or when the transition time Tx reaches a limit transition time Tx0 corresponding to the predetermined limit leakage resistance Rx0, and wherein the repetitive command signal PLS comprises a pulse signal with a fixed cycle having the charge period or the discharge period longer than at least the boundary transition time Tx0, or a pulse signal with a variable cycle in which an output pulse is inverted when the transition time Tx is reached. [2] Leakage resistance detection device for an on-board high-voltage device according to claim 1, wherein the charge / discharge resistor (33) includes a reference resistor having a resistance value Rs and a rapid charge / discharge resistor (35, 35a, 35b) having the resistance value Rq which is sufficiently smaller than the resistance value Rs, and the rapid charge / discharge resistor (35, 35a, 35b) is connected to a negative terminal and a positive terminal of the constant voltage control power source (25) via the pair of bypass diodes (36 and 37), and wherein the reference resistor (33) is connected in series with the leakage resistor Rx as a charge / discharge resistor for the coupling capacitor (51) when the arithmetic control circuit (20A, 20B, 20C, 20D, 20E) measures the transition time Tx of the gradually increasing monitor voltage Vx or the gradually decreasing monitor voltage Vx, and the reference resistor (33) is removed when a potential of the monitor voltage Vx relative to the vehicle body has changed to be 0 volts or lower or the control power supply voltage Vcc or higher due to an abrupt change in the power supply voltage of the high-voltage direct current source (61) or an abrupt change in the equivalent leakage resistor (65) on the positive potential side or the equivalent leakage resistor (66) on the negative potential side because the high-voltage power supply switch (62) for the high-voltage electrical load (64) is closed or opened,or because an abnormality has occurred, so as to charge and discharge the coupling capacitor (51) via the rapid charge / discharge resistor (35, 35a, 35b) and the pair of bypass diodes (36 or 37) regardless of an operating state of the charge / discharge switching element (31). [3] Leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the repetitive signal output circuit (30A) is configured, in response to the switching operation of the charge / discharge switching element (31), to connect the measuring point A, which is another terminal of the coupling capacitor (51), to an output terminal of the constant-voltage control power source (25) via the reference resistor (33) having the reference value Rs, or to connect the measuring point A of the negative terminal corresponding to the vehicle body potential via the rapid charge / discharge resistor (35) having the resistance value Rq sufficiently smaller than the resistance value Rs, so as to gradually increase or rapidly decrease the monitoring voltage Vx, which is the potential between the measuring point A and the vehicle body (11), wherein the monitoring signal processing circuit 40A is configured to: to generate the inverted logic signal HIG when the value of the monitor voltage Vx, which gradually increases with a gradient that becomes smoother as the time constant (Rs+Rx)C, which is a product of the sum of a resistance value Rs (Rs>>Rq) and the leakage resistance Rx and the electrostatic capacitance C of the coupling capacitor (51), increases, has increased to pass the predetermined threshold voltage V0 and input the inverted logic signal HIG to the arithmetic control circuit (20A); or to generate the analog signal voltage ANL proportional to the monitor voltage Vx and input the analog signal voltage ANL to the arithmetic control circuit (20A), and wherein the arithmetic control circuit (20A) inverts an output of the repetitive command signal PLS, which is a pulse train signal having a fixed cycle, in a state where the value of the monitor voltage Vx has abruptly decreased to approach 0 volts, through the charge / discharge switching element (31), and inverts the output of the repetitive command signal PLS, upon receiving the inverted logic signal HIG as an input or upon measuring, at the transition time Tx, a period from when the inverted logic signal HIG is generated based on the input analog voltage ANL until when the inverted logic signal HIG is obtained. [4] Leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the repetitive signal output circuit (30B) is configured, in response to the switching operation of the charge / discharge switching element (31), to connect the measuring point A, which is the other terminal of the coupling capacitor (51), to an output terminal of the constant-voltage control power source (25) via the rapid charge / discharge resistor (35) having the resistance value Rq, or to connect the measuring point A to the negative terminal corresponding to the vehicle body potential via the reference resistor (33) having the reference value Rs sufficiently larger than the resistance value Rq, to ​​thereby rapidly increase or gradually decrease the monitoring voltage Vx, which is the potential between the measuring point A and the vehicle body (11), wherein the monitoring signal processing circuit (40B) is configured to: to generate the inverted logic signal LOW when the value of the monitor voltage Vx, which gradually decreases with a gradient that becomes smoother as a time constant (Rx+Rx)C, which is a product of a sum of the resistance value Rs (Rs>>Rq) and the leakage resistance Rx with an electrostatic capacitance C of the coupling capacitor (51), becomes larger, has decreased to pass a value obtained by subtracting the predetermined threshold voltage V0 from the control power supply voltage Vcc and input the inverted logic signal LOW to the arithmetic control circuit (20B); or to generate the analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20B), and wherein the arithmetic control circuit (20B) inverts an output of the repetitive command signal PLS, which is a pulse train signal having a fixed cycle, in a state where the value of the monitor voltage Vx has abruptly increased to approach the control power supply voltage Vcc, through the charge / discharge switching element (31), and inverts the output of the repetitive command signal PLS after receiving the inverted logic signal LOW as an input or after measuring, as the transition time Tx, a period from when the inverted logic signal LOW is generated based on the input analog signal voltage ANL to when the inverted logic signal LOW is obtained. [5] Leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the repetitive signal output circuit (30C) is configured to, in response to the switching operation of the charge / discharge switching element (31), connect the measuring point A, which is the other terminal of the coupling capacitor (51), to an output terminal of the constant-voltage control power source (25) via the rapid charge / discharge resistor (35) having the reference value Rq, or to connect the measuring point A to the negative terminal of a vehicle body potential via the reference resistor (33) having the resistance value Rs sufficiently larger than the resistance value Rq, thereby rapidly increasing or gradually decreasing the monitoring voltage Vx, which is the potential between the measuring point A and the vehicle body (11), wherein the monitoring signal processing circuit (40C) is configured to: to generate the first inverted logic signal ANS or LOW while the value of the monitor voltage Vx, which gradually decreases with a gradient that becomes smoother as the time constant (Rs+Rx)C, which is a product of a sum of the resistance value Rs (Rs>>Rq) and the leakage resistance Rx, and the electrostatic capacitance C, of ​​the coupling capacitor (51) becomes larger, has decreased to pass the first threshold value V1, and to input the first inverted logic signal ANS or LOW to the arithmetic control circuit (20C); or to generate the analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20C), and wherein the monitoring signal processing circuit (40C) is further configured to: generating the second inverted logic signal ANS or HIG when the value of the monitor voltage Vx, which increases rapidly with a gradient that becomes steeper as the time constant (Rq+Rx)C becomes smaller, has increased so that it passes the second threshold value V2, which is close to the control power supply voltage Vcc, and inputting the second inverted logic signal ANS or HIG to the arithmetic control circuit (20C); or to generate the analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20C), wherein the arithmetic control circuit (20C) receives the second inverted logic signal ANS or HIG as an input or generates the second inverted logic signal ANS or HIG based on the input analog signal voltage ANL, and the arithmetic control circuit receives the first inverted logic signal ANS or LOW as an input or generates the first inverted logic signal ANS or LOW based on the input analog signal voltage ANL, wherein the arithmetic control circuit (20C) measures, as transition time Tx, a period from when the second inverted logic signal ANS or HIG is generated until when the first inverted logic signal ANS or LOW is received, wherein the arithmetic control circuit (20C) or the monitor signal processing circuit (40C) inverts the repetitive command signal PLS when the first inverted logic signal ANS or LOW and the second inverted logic signal ANS or HIG are input or generated, and wherein the first inverted logic signal and the second inverted logic signal are a pair of logic signals of the first logic signal LOW whose output logic changes before and after the first threshold value V1, and the second logic signal HIG whose output logic changes before and after the second threshold value V2, or the band history logic signal ANS which comes to a first logic state of "L" or "H" when the value of the monitor voltage Vx has increased to the second threshold value V2 or higher after decreasing to the first threshold value V1 or lower, and becomes a second logic state of "H" or "L" when the value of the monitor voltage Vx has decreased to the first threshold value V1 or lower after increasing to the second threshold value V2 or higher. [6] Leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the repetitive signal output circuit 30D is configured, in response to the switching operation of the charge / discharge switching element (31), to connect the measuring point A, which is the other terminal of the coupling capacitor (51), to an output terminal of the constant-voltage control power source (25) via the reference resistor (33) having the resistance value Rs, or to connect the measuring point A to the negative terminal having a vehicle body potential via the rapid charge / discharge resistor (35) having the reference value Rq sufficiently smaller than the resistance value Rs, to thereby gradually increase or quickly decrease the monitoring voltage Vx, which is the potential between the measuring point A and the vehicle body (11), wherein the monitoring signal processing circuit (40D) is configured: to generate the second inverted logic signal ANS or HIG when the value of the monitor voltage Vx, which gradually increases with a gradient that becomes smoother as the time constant (Rs+Rx)C, which is a product of the sum of the resistance value Rs (Rs>>Rq) and the leakage resistance Rx and the electrostatic capacitance C of the coupling capacitor (51), becomes larger, has increased to pass the second threshold value V2 and input the second inverted logic signal ANS or HIG to the arithmetic control circuit (20D); or to generate the analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20D), and the monitoring signal processing circuit (40D) is further configured to: to generate the first inverted logic signal ANS or LOW when the value of the monitor voltage Vx, which falls rapidly with a gradient that becomes steeper as the time constant (Rq+Rx)C becomes smaller, has decreased to pass the first threshold value V1, which is close to 0 volts, and to input the first inverted logic signal ANS or LOW to the arithmetic control circuit (20D); or to generate the analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20D), wherein the arithmetic control circuit (20D) receives the second inverted logic signal ANS or HIG as an input or generates the second inverted logic signal ANS or HIG based on the input analog signal voltage ANL, and the arithmetic control circuit (20D) receives the first inverted logic signal ANS or LOW as an input or generates the first inverted logic signal ANS or LOW based on the input analog signal voltage ANL, wherein the arithmetic control circuit (20D) measures as the transition time Tx a period from when the first inverted logic signal ANS or LOW is generated until when the second inverted logic signal ANS or HIG is received, wherein the arithmetic control circuit (20D) or the monitor signal processing circuit (40D) inverts the repetitive command signal PLS when the first inverted logic signal ANS or LOW and the second inverted logic signal ANS or HIG are input or generated, and wherein the first inverted logic signal and the second inverted logic signal are a pair of logic signals of a first logic signal LOW whose output logic changes before and after the first threshold value V1, and a second logic signal HIG whose output logic changes before and after the second threshold value V2, or a band history logic signal ANS that assumes a first logic state of "L" or "H" when the value of the monitor voltage Vx has increased to the second threshold value V2 or higher after decreasing to the first threshold value V1 or lower, and that becomes a second logic state of "H" or "L" when the value of the monitor voltage Vx has decreased to the first threshold value V1 or lower after increasing to the second threshold value V2 or higher. [7] Leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the repetitive signal output circuit (30E) is configured to, in response to the switching operation of the charge / discharge switching element (39), connect the measuring point A, which is the other terminal of the coupling capacitor (51), to an output terminal of the constant-voltage control power source (25) via the reference resistor (33) having the resistance value Rs, or to connect the measuring point A to the negative terminal having a vehicle body potential via the same reference resistor (33), thereby gradually increasing or gradually decreasing the monitoring voltage Vx, which is the potential between the measuring point A and the vehicle body (11), wherein the monitoring signal processing circuit (40E) is configured to: to generate second inverted logic signals ANS or HIG when the value of the monitor voltage Vx, which gradually increases or gradually decreases with a gradient that becomes smoother as a time constant (Rs+Rx)C, which is a product of the sum of the resistance value Rs and the leakage resistance Rx with an electrostatic capacitance C of the coupling capacitor (51), increases, has increased to pass the second threshold value V2, and to generate first inverted logic signals ANS or LOW when the value of the monitor voltage Vx has decreased to pass a first threshold value V1, and to input the second inverted logic signal ANS or HIG and the first inverted logic signal ANS or LOW to the arithmetic control circuit (20E); or to generate the analog signal voltage ANL proportional to the monitoring voltage Vx and to input the analog signal voltage ANL to the arithmetic control circuit (20E), wherein the arithmetic control circuit (20E) receives the second inverted logic signal ANS or HIG as an input and generates the second inverted logic signal ANS or HIG based on the input analog signal voltage ANL, and the arithmetic control circuit (20E) receives the first inverted logic signal ANS or LOW as an input, or generates the first inverted logic signal ANS or LOW based on the input analog signal voltage ANL, wherein the arithmetic control circuit (20E) is configured to: to measure as a second transition time Tx2 a period from when the first inverted logic signal ANS or LOW is generated until when the second inverted logic signal ANS or HIG is received, to measure as a first transition time Tx1 a period from when the second inverted logic signal ANS or HIG is generated until when the first inverted logic signal ANS or LOW is received, and to set the first transition time Tx1 or the second transition time Tx2 or an average value of both the first transition time Tx1 and the second transition time Tx2 as the transition time Tx, wherein the arithmetic control circuit (20E) or the monitor signal processing circuit (40E) inverts the repetitive command signal PLS when the first inverted logic signal ANS or LOW and the second inverted logic signal ANS or HIG are input or generated, and wherein the first inverted logic signal and the second inverted logic signal are a pair of logic signals of the first logic signal LOW whose output logic changes before and after the first threshold value V1, and a second logic signal HIG whose output logic changes before and after the second threshold value V2, or a band history logic signal ANS that becomes a first logic state of "L" or "H" when the value of the monitor voltage Vx has risen to the second threshold value V2 or higher after falling to the first threshold value V1 or lower, and that becomes a second logic state of "H" or "L" when the value of the monitor voltage Vx has fallen to the first threshold value V1 or lower after falling to the second threshold value V2 or higher. [8] A leakage resistance detection method for an on-board high-voltage device used as the leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the leakage resistance detection method comprises, when an electrostatic capacitance of a coupling capacitor (51) is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance (65) on the positive potential side and the equivalent leakage resistance (66) on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistors (33 and 35) provided in the repetitive signal output circuit (30A, 30B, 30C, 30D, 30E), is represented by R0, and when either a period during which the monitor voltage Vx gradually increases from a state of 0 to the predetermined threshold voltage V0, a ​​periodduring which the monitoring voltage Vx gradually decreases from a state of the control power supply voltage Vcc to a predetermined threshold voltage (Vcc-V0), or a period during which the monitoring voltage Vx gradually increases and gradually decreases alternately between a first threshold value V1 and a second threshold value V2, is represented by a transition time Tx, a transition time coefficient α is set to Tx / (R0×C), a leakage resistance coefficient β is set to Rx / R0, and a threshold voltage coefficient γ is set to V0 / Vcc or (V2-V1) / Vcc:, Calculating, by an arithmetic control circuit (20A, 20B, 20C, 20D, 20E), the value of the leakage resistance Rx corresponding to the measured transition time Tx by using the threshold voltage coefficient γ as a parameter, and based on a characteristic curve of the transition time coefficient α corresponding to a resistance value of the leakage resistance coefficient β, wherein the characteristic curve has a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases, together with an increase in the leakage resistance coefficient β; selectively using a negative gradient curve region following the positive gradient curve region or a negative gradient curve region occupying an entire region, and determining the threshold voltage coefficient γ such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the negative gradient curve region; Determining, when the transition time Tx has become a value equal to or greater than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and Performing the off-time abnormality determination ERR1 when the transition time Tx exceeds a short-circuit transition time Tx00 corresponding to a case where the leakage resistance Rx becomes 0. [9] A leakage resistance detection method for an on-board high-voltage device according to claim 1, wherein the characteristic curve has a negative gradient curve region as an entire region, and a low threshold voltage coefficient γ is applied so that the characteristic curve does not include a positive gradient curve region. [10] A leakage resistance detection method for an on-board high-voltage device used for the leakage resistance detection device for an on-board high-voltage device according to claim 1 or 2, wherein the leakage resistance detection method comprises, when the electrostatic capacitance of the coupling capacitor (51) is represented by C, a parallel combined leakage resistance value of the equivalent leakage resistance (65) on the positive potential side and the equivalent leakage resistance (66) on the negative potential side is represented by Rx, and a series resistance value, which is a total value of the charge / discharge resistors (33 and 35) provided in the repetitive signal output circuit (30A, 30B, 30C, 30D, 30E), is represented by R0, and when either a period during which the monitor voltage Vx gradually increases from a state of 0 to the predetermined threshold voltage V0, a ​​periodduring which the monitoring voltage Vx gradually decreases from a state of the control power supply voltage Vcc to a predetermined threshold voltage (Vcc-V0), or a period during which the monitoring voltage Vx gradually increases and gradually decreases alternately between a first threshold value V1 and a second threshold value V2, is represented by a transition time Tx, a transition time coefficient α is set to Tx / (R0×C), a leakage resistance coefficient β is set to Rx / R0, and a threshold voltage coefficient γ is set to V0 / Vcc or (V2-V1) / Vcc:, Calculating, by the arithmetic control circuit (20A, 20B, 20C, 20D, 20E), the value of the leakage resistance Rx corresponding to the measured transition time Tx using the threshold voltage coefficient γ as a parameter and based on a characteristic curve of the transition time coefficient α corresponding to a value of the leakage resistance coefficient β, wherein the characteristic curve has a negative gradient curve region or a positive gradient curve region in which the transition time coefficient α gradually increases or gradually decreases along with an increase in the leakage resistance coefficient β; Excluding, in advance, by a characteristic curve correction means, the application of a negative gradient curve region following the positive gradient curve region; Determining the threshold voltage coefficient y such that a value of the transition time coefficient α when the value of the leakage resistance Rx has reached the limit leakage resistance Rx0 is included in the corrected positive gradient curve region; Determining, when the transition time Tx has become a value equal to or less than the limit transition time Tx0 corresponding to the limit leakage resistance Rx0, that the leakage resistance Rx is too small, and generating the resistance abnormality determination output ER1; and Performing a time-off abnormality determination ERR1 when the transition time Tx exceeds an open-circuit transition time Txm corresponding to a case where the leakage resistance Rx is infinite. [11] A leakage resistance detection method for an on-board high-voltage device according to claim 10, further comprising: connecting, in advance, by the characteristic curve correcting means, the bypass leakage resistance (53) between the connection point B of the coupling capacitor (51) and the on-board high-voltage device (60X, 60Y), and the vehicle body (11), wherein the bypass leakage resistance (53) has the additional resistance R3 which is sufficiently larger than the limit leakage resistance Rx0, and, Suppressing, through the bypass leakage resistance, the value of the leakage resistance coefficient β on R3 / R0 when the leakage resistance Rx is infinite, thereby preventing the transition time coefficient α from becoming excessive. [12] Leakage resistance detection method for an on-board high-voltage device according to claim 8 or 10, wherein the leakage resistance detection device further comprises a bypass leakage resistor (53) connected between the connection point B between the coupling capacitor (51) and the on-board high-voltage device (60X, 60Y), and the vehicle body (11), and wherein the leakage resistance detection method further comprises measuring, by the arithmetic control circuit (20A, 20B, 20C, 20D, 20E), the leakage resistance Rx in a state where the coupling capacitor (51) and the on-board high-voltage device (60X, 60Y) are not connected to each other, and calibrating and storing the value of the electrostatic capacitance C of the coupling capacitor (51) so that an obtained result becomes the additional resistance R3 of the bypass leakage resistor (53). [13] A leakage resistance detection method for an on-board high-voltage device according to claim 8 or 10, further comprising, by the arithmetic control circuit (20A, 20B, 20C, 20D, 20E): Measuring, in response to the inverted logic signal HIG or the analog signal voltage ANL operating in response to the monitor voltage Vx input from the monitor signal processing circuit (40A, 40B, 40C, 40D, 40E), a period of time during which the monitor voltage Vx deviates from a correct voltage range determined by a threshold voltage and another threshold voltage for measuring the transition time Tx; and Interrupting the measurement of the leakage resistance Rx when the measurement period is within a predetermined allowable determination period, and detecting the transient characteristic abnormality ERR2 when the measured period exceeds a predetermined abnormality determination period. [14] A leakage resistance detection method for an on-board high-voltage device according to claim 13, further comprising: Inputting a vehicle condition signal (18) to the arithmetic control circuit (20A, 20B, 20C, 20D, 20E), the vehicle condition signal (18) being a signal for identifying the occurrence of a condition responsible for fluctuations in the vehicle body potential at a connection point between the coupling capacitor (51) and the on-board high-voltage device (60X, 60Y); and Detecting, by the arithmetic control circuit (20A, 20B, 20C, 20D, 20E), a change in the state of the vehicle state signal (18) to predict an abrupt change in the leakage resistance, and avoiding the determination of the transient characteristic abnormality immediately after the change in the state of the vehicle state signal (18). [15] A leakage resistance detection method for an on-board high-voltage device according to claim 8 or 10, further comprising: Initiating a non-volatile data memory (23) which is a sub-area of ​​the program memory (24A, 24B, 24C, 24D, 24E) or a memory provided together with the program memory (24A, 24B, 24C, 24D, 24E): to write and store in advance a data table relating to a characteristic curve of the transition time coefficient α of Tx / (R0C) relative to the leakage resistance coefficient β of Rx / R0 using the threshold voltage coefficient y of V0 / Vcc or (V2-V1) / Vcc as a parameter, and a value of the threshold voltage coefficient γ, values ​​of the reference value Rs and the rapid charge / discharge resistance Rq that determine the value of the series resistance R0, a value of the electrostatic capacitance C, a value of the limit leakage resistance Rx0, and a value of the preliminary leakage resistance Rxn to be provisionally notified; and to write and store the value of the leakage resistance Rx measured during operation and abnormality occurrence history information periodically or immediately before the stop of the operation; and to read out, by the arithmetic control circuit (20A, 20B, 20C, 20D, 20E), the value of the leakage resistance coefficient β from the data table in correspondence with the transition time coefficient α calculated by the transition time Tx, to thereby calculate a current leakage resistance Rx, and to compare the current leakage resistance Rx with the value of the limit leakage resistance Rx0 or the value of the preliminary load resistance Rxn, to thereby generate a pre-alarm output ER2 in addition to the resistance abnormality determination output ER1, or to perform determination of the state change abnormality ERR3 when the leakage resistance has abruptly lowered in a time series. [16] Leakage resistance detection method for an on-board high-voltage device according to claim 8 or 10, wherein the leakage resistance detection device further comprises a test leakage resistor (54) and a test switching element (55) connected between a connection point B between the coupling capacitor (51) and the on-board high-voltage device (60X, 60Y), and the vehicle body (11), wherein the test leakage resistance (54) has a resistance value equal to or less than the limit leakage resistance Rx0 or a resistance value equal to or less than the preliminary leakage resistance Rxn, which is close to the limit leakage resistance Rx0 and greater than the limit leakage resistance Rx0, wherein the test switching element (55) is connected in series with the test leakage resistor (54), wherein the leakage resistance detection method further comprises, by the arithmetic control circuit (20B, 20D): Closing the test switching element (55) temporarily at the start of operation and determining whether the resistance abnormality determination output ER1 operates for a moment or the pre-alarm output ER2, which is generated when the leakage resistance Rx has become equal to or smaller than the preliminary leakage resistance Rxn, operates for a moment, to thereby inspect whether an operation of detecting the leakage resistance Rx is performed normally; and when an inspection result shows that the detection operation is abnormal, generating the resistance abnormality determination output ER1 or the pre-alarm output ER2 continuously or intermittently, and when the inspection result shows that the detection operation is normal, suspending the output generation for such a short period of time that the abnormality alarm device (19) to be driven by the resistance abnormality determination output ER1 or the pre-alarm output ER2 does not operate. [17] Leakage resistance detection method for an on-board high-voltage device according to claim 16, wherein the leakage resistance detection device further comprises a bypass leakage resistor (53) connected between the connection point B between the coupling capacitor (51) and the on-board high-voltage device (60X, 60Y), and the vehicle body (11), and wherein the leakage resistance detection method further comprises, by the arithmetic control circuit (20B, 20D): closing the test switching element (55) at the start of operation and measuring a first combined leakage resistance value in a state in which the test leakage resistance (54), the bypass leakage resistance (53) and an actual leakage resistance Rx are connected in parallel, and measuring a second combined leakage resistance value in a state in which the test switching element (55) is open and the bypass leakage resistance (53) and the actual leakage resistance Rx are connected in parallel; and recalculating the electrostatic capacitance C of the coupling capacitor (51) from the first combined leakage resistance value and the second combined leakage resistance value and storing the recalculated electrostatic capacitance C as a calibrated value, wherein the calibrated value is calculated such that the actual leakage resistances Rx in open-circuit and closed-circuit states of the test switching element (55) have the same values, due to the calibrated and stored electrostatic capacitance C.

Citation Information

Patent Citations

  • Ground detecting device for electric vehicle

    JP2002209331A

  • Leakage detection circuit for electric vehicle

    US20070285057A1

  • Apparatus and method for sensing leakage current of battery

    US20100237872A1

  • Method and apparatus for detecting leakage resistance in an electric vehicle

    US5382946A