JP6545503B2 - Power supply device and current-voltage characteristic measuring apparatus - Google Patents

Description

  The present invention relates to a power supply device and a current-voltage characteristic measurement device including the power supply device.

  The current-voltage characteristics (hereinafter referred to as IV characteristics) of the measurement object such as a substance, an element, a measuring device, or a power device depend on the temperature of the measurement object. Therefore, in order to accurately measure the I-V characteristics, it is necessary to perform the measurement while keeping the temperature of the measurement object itself as constant as possible. To meet such requirements, it is necessary to measure the non-linear characteristics of the electrical resistance near the critical temperature of the superconducting material, to measure the precise resistance in a general resistance measuring instrument, to perform phase transformations of metals, aging phenomena, recrystallization reactions, etc. It is important in the evaluation of physical properties when involved, and in parts such as power circuit breakers for large currents.

  In order to measure the I-V characteristics at a predetermined temperature, basically, the temperature of the measurement environment in which the measurement object is stored is controlled. In this state, a plurality of different values of current are sequentially supplied to the object to be measured, and voltage drops in the object to be measured corresponding to the respective currents are measured or voltages of different values are sequentially applied to the object to be measured Apply and measure the current flowing to the object corresponding to each. At this time, power is supplied to the object to be measured and electrodes, lead wires and the like for passing current therethrough, and Joule heat generated thereby has an effect of raising the temperature of the object itself. As a result, the temperature of the object to be measured in the process of measuring the I-V characteristics changes in accordance with the supplied current or the applied voltage. It is ideal to realize active control such that the temperature of the measurement object is maintained at a predetermined temperature, including the effect of Joule heating. However, the effect of Joule heating directly affects the object to be measured, whereas the temperature sensor for temperature control is provided outside the object to be measured. Therefore, even if the temperature is controlled to match the temperature obtained by the measurement with the temperature sensor to a predetermined value, it is inevitable that the temperature of the object to be measured changes depending on the voltage or current, and the IV characteristic is measured The IV characteristic can not be measured in a state where the temperature of the object to be measured is substantially constant at each set temperature in the temperature range required for the measurement.

  The current-voltage characteristic measuring device disclosed in Patent Document 1 uses a voltage supply device as a power supply device, and continues to apply a certain voltage to an object to be measured, which is lower than the certain voltage, A pulse voltage whose voltage value changes in time series is applied as a time series signal, and current measurement is performed by the current measuring device in synchronization with the application of this time series signal. The purpose is to measure the IV characteristics with the temperature of the substance kept constant.

JP-A-6-102310

  However, in the power supply device disclosed in Patent Document 1, the peak value of the voltage pulse applied to the object to be measured changes in the process of measuring the IV characteristics, and therefore, the lead wire or electrode for applying a voltage The Joule heat per time generated in the measurement object including the H level fluctuates in the process of measuring the IV characteristics. Such temperature fluctuation of the measurement object due to fluctuation of heat generation per hour becomes a problem when accurate measurement of IV characteristics is required.

  In order to alleviate this problem to a negligible extent, Joule heating per hour can be further reduced by further reducing the applied voltage and voltage application time width causing temperature fluctuation, ie, the peak value and pulse width of the voltage pulse. Just do it. However, since the peak value of the voltage pulse must be a value that can cover the measured voltage range of the IV characteristic, it can not be made smaller than a predetermined value. Also, the pulse width must also have a value greater than or equal to a certain time width, because after the response time determined by the presence of the time constant of the current measurement device, the required measurement time must be ensured. . Therefore, there is a limit to the idea of lowering the value of Joule heating per hour. That is, when the I-V characteristic of the measurement object is measured by the power supply device disclosed in Patent Document 1, the Joule heat per time generated in the measurement object essentially fluctuates in the measurement process. It is difficult to measure the accurate I-V characteristics because the temperature fluctuation of the object is inevitable.

  Even when the IV characteristic is measured by using the power supply device as a current supply device, supplying a current to the object to be measured, and measuring the response voltage by the voltage measurement device, the response at the time of measurement of the voltage measurement device The same problems as described above occur because of the restriction of sex.

  The present invention has been made to solve the above problems, and provides a power supply apparatus that outputs a time-series pulse signal in which a value obtained by averaging Joule heat per time generated by adjacent pulse signals is substantially constant. It is an object of the present invention to provide a current-voltage characteristic measuring device provided with the power supply device.

The power supply apparatus according to the present invention is a power supply apparatus that outputs two series of time-series pulse signals of current or voltage, and F (t) and G (t) represent respective squares of time t. Where t is a continuous natural number in a predetermined range including 0, i is a measurement start time t ₀ , and a preset time width is Δt. Of the first series of time-series pulse signals among the two series of time-series pulse signals, where t _2i = t ₀ + Δt × 2i and t _{2i + 1} = t ₀ + Δt × (2i + 1) Is a value equal to the absolute value of F (t _2i ), and the absolute value of the peak value of the pulse of the second series of time series pulse signals adjacent to each pulse of the first series of time series pulse signals is G (t _{2i + 1)} is the absolute value equal to the value of the Δt is, F (t the sum of the square and the square of G (t 2i _{+ 1)} of the _i), the difference between the constant value, regardless of the value of the i, preset to be equal to or less than the preset allowable error It is characterized in that it is a value.

  In the current-voltage characteristic measuring device according to the present invention, voltage drop or current at the measurement target point when the time series pulse signal is input to the measurement target point of the I-V characteristic of the measurement object. And a response signal measurement unit that measures in synchronization with the time-series pulse signal as a time-series response pulse signal, and outputs a net peak value of each pulse of the time-series response pulse signal. Do.

  According to the power supply device of the present invention, when the time-series pulse signal output from the power supply device is input to the object to be measured, the adjacent pulse signals average the Joule heat generated per time of the object to be measured. Is approximately constant. Therefore, the temperature fluctuation of the measurement object to which the time-series pulse signal is input can be suppressed as compared with the case where the signal output from the conventional power supply device is input to the measurement object. Moreover, the measuring apparatus of the current-voltage characteristic provided with such a power supply device makes control which keeps the temperature of a measuring object substantially constant easy, and as a result, measures the current-voltage characteristic more correctly than the conventional system. can do.

It is a figure which shows the structural example of the current-voltage characteristic measuring apparatus provided with the power supply device which concerns on embodiment of this invention. It is a figure which shows an example of the 2 function time-sequential pulse pattern used as the basis of the time-sequential signal input into a measurement object from the power supply device which concerns on embodiment. It is a figure which shows the example of a hardware constitutions of the setting * analysis part of the current-voltage characteristic measuring apparatus provided with the power supply device which concerns on embodiment. (A) shows the current-voltage characteristic curve of the object to be measured, (B) shows an example of the case where the current input to the object to be measured is a sine waveform, and (C) shows to (A). It is a figure which shows the example of a waveform of the response voltage when the electric current of the sine waveform shown to (B) is input to the measurement object which has the current-voltage characteristic shown by the current-voltage characteristic curve shown. The figure which shows the example of the time-sequential current pulse signal input into a measurement object from the power supply device which concerns on embodiment, and a time-sequential response voltage pulse signal, (A) corresponds to the time-axis area | region K of FIG. The figure which expanded the series electric current pulse signal to the time-axis direction, (B) is an enlarged view of the time-series response voltage pulse signal corresponding to (A). It is a figure which shows an example of the measurement result of the current-voltage characteristic by the current-voltage characteristic measuring apparatus provided with the power supply device which concerns on embodiment. It is a figure which shows the 1st modification of the time-sequential pattern of the current pulse input into a measurement object from the power supply device which concerns on embodiment. It is a figure which shows the 2nd modification of the time-sequential pattern of the current pulse input into a measurement object from the power supply device which concerns on embodiment. It is a figure which shows the modification of a structure of the current-voltage characteristic measuring apparatus provided with the power supply device which concerns on embodiment.

Embodiment
FIG. 1 shows a configuration example of a current-voltage characteristic measuring device provided with a power supply device according to an embodiment of the present invention. The current-voltage characteristic measuring device 2 measures the value of voltage drop in the measurement object 1 when the current signal is input to the measurement object 1 and measures the IV characteristic of the measurement object 1 It is an example of composition.

Before describing the current-voltage characteristic measurement device 2, the measurement object 1 will be described in more detail. In FIG. 1, the measurement object 1 is represented by _three electrical resistances K _{1 to} K ₃ . Electrical resistance K ₁ indicates a site of interest to measure the I-V characteristic (hereinafter referred to as the measurement site K ₁ in.). Electrical resistance K ₂ is a combination of the resistance of the resistor and lead wires connected to the electrodes of the installed one of the electrodes to the measurement site K ₁ to energizing current to sample K _1. Electrical resistance K ₃ is a combination of the resistance of the resistor and lead wires connected to the electrodes of the installed other electrode measurement site K ₁ to energizing current to sample K _1. In the following, the value of the voltage drop at the measuring object 1, unless otherwise stated, the voltage drop value at measurement site K _1, i.e. the voltage across the measurement site K _{1 (electrical} resistance K ₁₎ I shall mean. Moreover, the IV characteristic of the measurement object 1 means the IV characteristic of the measurement site K ₁ .

Although the lead wires and electrodes, that is, the electrical resistances K ₂ and K ₃ do not necessarily constitute the measurement object 1, the measurement is performed when the measurement site K ₁ is a substance with a low electrical resistance such as a superconductor. since the Joule heat generated at the site K ₁ near the lead wire or electrode is the main cause of the temperature variation of the measurement site K _1, 1, and an electrical resistance K ₂ and K ₃ to display these on the measurement object 1 It included and showed. Note that if the measurement site K ₁ is a high-resistance material such as a semiconductor material or insulating material, since Joule heat generated when current in the measurement object 1 is given by the measuring site K ₁ is the main, the measurement site K The temperature _{1 is determined by} the temperature of the installation environment and the Joule heat generated at the measurement site K ₁ , and the influence of the Joule heat generated at the lead wire and the electrode may be ignored. Therefore, at this time, the electrical resistances K ₂ and K ₃ in FIG. 1 can be ignored from the viewpoint of the temperature of the measurement object 1.

  The current-voltage characteristic measuring apparatus 2 includes a power supply device 20 that outputs a time-series current pulse signal to be input to the measurement object 1, and a voltage drop in the measurement object 1 when the time-series current pulse signal is input. The voltage measurement unit 21 which measures the value, data which is the basis of the time-series current pulse signal is set and output to the power supply device 20, and the peak value information of the time-series current pulse signal and the measurement result of the voltage measurement unit 21 And a setting / analyzing unit 22 for acquiring information on the IV characteristic of the measurement object 1, and a display unit 23 for displaying the analysis result of the setting / analysis unit 22.

  The power supply device 20 shown in FIG. 1 is a current supply source that outputs a current signal in the form of a time series current pulse signal. The power supply device 20 supplies (inputs) a time-series current pulse signal to the measurement object 1, and outputs a trigger signal for synchronization to the voltage measurement unit 21 when outputting the time-series current pulse signal. . Further, a current value corresponding to the peak value of each pulse of the time-series current pulse signal to be output is output to the setting / analysis unit 22 as analysis data.

  The power supply device 20 has a two-function time-series pulse pattern storage unit 200 that stores and stores two-function time-series pulse patterns formed based on two functions, and a maximum amplitude setting unit that inputs the maximum amplitude value of the output electric signal. A time-series pulse signal generation / output unit 202 that generates and outputs a time-series pulse signal based on the two-function time-series pulse pattern and the maximum amplitude value, in the example of FIG. 1, a time-series current pulse signal.

  The two-function time-series pulse pattern storage unit 200 is configured by a normal memory, and stores and stores the two-function time-series pulse pattern input from the setting / analysis unit 22. The two-function time-series pulse pattern is used by the time-series signal generation / output unit 202 as basic data for time-series current pulse signal generation. The two-function time-series pulse pattern is a time-series pattern of pulses formed based on two functions f (t) and g (t), where t is a variable indicating time. The function f (t) and the function g (t) are a pair of functions such that the sum of squares of the respective functions at time t has a constant value.

  As a specific example, FIG. 2 is a diagram showing the concept of a bifunctional time-series pulse pattern when f (t) is a cosine function and g (t) is a sine function. f (t) and g (t) have the same amplitude 1, the same period T, and the same phase. At this time, when each of f (t) and g (t) is squared and added, it becomes a constant value 1 regardless of the value of t. As shown in FIG. 2, the bifunctional time series pulse pattern divides the time axis t into a plurality of sections, and for each section, f (t) and g (t) for the value of t representative of the section Values are alternately acquired to form a time-series pulse pattern having a constant pulse width and a constant mutual interval. That is, the two-function time-series pulse pattern is composed of a first time-series pulse pattern whose envelope is a function f (t) and a second time-series pulse pattern whose envelope is a function g (t) Ru. The pulses of the first time-series pulse pattern and the pulses of the second time-series pulse pattern are alternately arranged. Usually, the pulse width is 5 to 100 ms (more preferably 5 to 50 ms), and the mutual interval is at least as large as the pulse width. Details of this two-function time-series pulse pattern will be described later.

  Here, the two-function time-series pulse pattern storage unit 200 stores and stores a two-function time-series pulse pattern generated using a cosine function as f (t) and a sine function as g (t). Although an example has been described, as long as the sum of squares is a fixed two function, a two-function time-series pulse pattern generated based on the other two functions may be stored and stored.

The maximum amplitude setting unit 201 receives the maximum amplitude value I ₀ of the time-series current pulse signal via the input unit (not shown) of the power supply device 20 and generates the maximum amplitude value I ₀ as a time-series pulse signal Output to the output unit 202. Measurement limit of the voltage measurement unit that measures a voltage drop value of the maximum amplitude value value measured object 1 I ₀ In this case, the measuring object 1, occurs may increase in temperature, change in characteristics and energization of the measuring object 1 It is determined in advance in consideration of the destruction limit of the current value. The input method of the input unit of the power supply device 20 may be any method such as a dial or a keyboard.

The time-series signal generation / output unit 202 is based on the two-function time-series pulse pattern read from the two-function time-series pulse pattern storage unit 200 and the maximum amplitude value I ₀ (usually, the maximum amplitude of the two-function time-series pulse pattern A time-series current pulse signal is generated by multiplying the value I _0, and is output to the measurement object 1, and at the same time, a trigger signal for synchronization is output to the voltage measurement unit 21. Further, the peak value of each current pulse of the time-series current pulse signal to be output, that is, the current value is output to the setting / analysis unit 22 as time-series data. Note that if F (t) = I ₀ × f (t) and G (t) = I ₀ × g (t), the time-series current pulse signal divides the time axis t into a plurality of sections, and each section Every time, the values of F (t) and G (t) for the value of t representative of the section are alternately acquired and each is taken as a peak value, and pulses having a constant pulse width are time-series at constant intervals. It can be said that it is a time-series signal which is arranged and formed.

The voltage measurement unit 21 is a device that measures the voltage drop value of the measurement object 1. Specifically, the trigger signal output from the time-series signal generation / output unit 202 synchronizes with the time-series current pulse signal. Te, measuring the voltage across the measurement site K _1. This voltage is measured as a time-series response voltage pulse signal corresponding to the time-series current pulse signal input to the measurement target 1, and the voltage measuring unit 21 removes the drift component of the signal from the signal as an offset and The pulse peak value of is calculated and output. Derivation of a specific net pulse peak value will be described later.

  The setting / analysis unit 22 is composed of a CPU 300, an internal memory 310, an external memory 320, an input unit 330, an output unit 340, and a bus line 350 connecting these as hardware as shown in FIG. In the second embodiment, as shown in FIG. 1, a 2-function time-series pulse pattern setting unit 220 and an analysis unit 221 are provided. The external memory 320 stores a program, which is read and executed by the CPU 300 to realize the functions of the two-function time-series pulse pattern setting unit 220 and the analysis unit 221.

  The two-function time-series pulse pattern setting unit 220 sets a two-function time-series pulse pattern, which is basic data for generating time-series current pulse signals, and outputs the two-function time series pulse pattern to the power supply device 20 through the output unit 340. The two-function time-series pulse pattern setting unit 220 receives the functions f (t) and g (t) that the sum of the respective squares inputted through the input unit 330 becomes a constant value, via the CPU 300 and the internal memory 310. To generate a two-function time-series pulse pattern.

Next, the analysis unit 221 will be described. The analysis unit 221 receives the current value which is the peak value of each current pulse of the time-series current pulse signal received from the time-series pulse signal generation / output unit 202 via the input unit 330, and the input unit 330 from the voltage measurement unit 21. Based on the voltage value which is the net pulse wave height value of each voltage pulse of the time-series response voltage pulse signal received via the CPU 300 and the internal memory 310, the information of the IV characteristic is generated by analysis or the like. The information on the IV characteristics is the current I supplied to the object 1 and the voltage drop value at the object 1 due to the current, that is, the voltage V across the measurement point K ₁ (electrical resistance K ₁ ) Information that indicates a relationship. Further, the analysis unit 221 outputs the generated information of the I-V characteristic to the display unit 23 through the output unit 340.

  The IV characteristics are schematically shown in FIG. FIG. 4A shows the IV characteristic in the form of an IV characteristic curve. FIG. 4B shows the time change of the current I input to the measurement object 1. The example of FIG. 4B is an example in which the current I changes with time in accordance with a sine wave. The time-series current pulse signal corresponds to the envelope of only the peak value generated based on G (t) represented by a sine function. FIG. 4C shows a temporal change of the voltage V which is a voltage drop value of the object to be measured 1 when the current I shown in FIG. 4B is inputted to the object to be measured 1. As for the time-series response voltage pulse signal, only the envelope of the net pulse peak value of each voltage pulse of the time-series response voltage pulse signal corresponding to the peak value of the time-series current pulse signal corresponding to G (t) is shown It is

When the voltage V is a function of the current I, the IV characteristic curve is characterized by the coefficient of each term when Taylor is expanded by I. V is usually represented only by the odd term of I, and as a simple example, when V is represented by only the first and third terms of I, it becomes as shown in Formula (1).
_{V (I) = R 0 ×} I + R 2 × I 3 (1)
R ₀ ; Linear resistance
R ₂ ; nonlinear resistance

The analysis unit 221 determines R ₀ and R ₂ in equation (1) from the pulse wave height value corresponding to G (t) of the time series current pulse signal and the pulse wave height value of the time series response voltage pulse signal corresponding thereto. Ask. Equation (1) and the values of R ₀ and R ₂ may be included in the information of the IV characteristics. The method of obtaining the R ₀ and R ₂ have various known methods, such as Fourier analysis techniques, R ₀ and R ₂ may be determined in any way. Note that obtaining R ₀ and R ₂ corresponds to performing numerical differentiation or difference calculation in a normal method. Since noise components are always superimposed on the measurement data, the effects of noise increase when numerical differentiation or difference calculation is performed, and often significant analysis results can not be obtained. However, if input data is taken as a waveform and Fourier analysis is applied to this, R ₀ and R ₂ can be obtained in a state where the influence of noise is small. This Fourier analysis is basically an integral operation, so the influence of noise is reduced. When applying Fourier analysis, measurement data of current and voltage to be used uses data of one cycle. Further, f (t) and g (t) are preferably cosine function and sine function, respectively.

  The analysis unit 221 may obtain the I-V characteristic in the form of equation (1) by a method of polynomial fitting. Also, the IV characteristics need not necessarily be determined in the form of equation (1), and may be generated simply as a current value and a corresponding response voltage value displayed in a tabular form or a graph. In any case, the net response pulse peak value corresponding to all pulse peak values of the time-series current pulse signal input to the measurement target 1 can be utilized for grasping the IV characteristics, and the measurement target object There is an advantage that the signal input to 1 is not wasted.

  Even when Fourier analysis is performed, by devising the whole response pulse peak value of the time-series current pulse signal input to the measurement object 1 and the corresponding net response pulse peak value to grasp the IV characteristics can do. The method uses the time-series current pulse signal corresponding to the first time-series pulse pattern whose envelope is f (t), that is, the phase of the pulse peak value of the time-series current pulse signal corresponding to F (t), The time-series current pulse signal corresponding to the second time-series pulse pattern whose envelope is g (t), that is, delayed by 90 degrees with respect to the phase of the pulse peak value of the time-series current pulse signal corresponding to G (t) Ie shift by T / 4 on the time axis. Due to this phase shift, the envelope of the time-series current pulse signal corresponding to the first time-series pulse pattern becomes a sine wave. As a result, the time-series current pulse signal can be regarded as having its envelope G (t). The same phase shift processing is performed on the net response pulse peak value obtained corresponding to the time-series response voltage pulse signal. The Fourier analysis is performed by regarding time series data of the current value and the response voltage value obtained by this processing as waveform information. Thereby, even when obtaining an IV characteristic from Fourier analysis, it is possible to utilize all pulse peak values of the time-series current pulse signal input to the measurement object 1 and the corresponding net response pulse peak values. .

  The display unit 23 is configured by an output device such as a printer or a display device, and displays the information by outputting the IV characteristic information of the measurement object 1 obtained by the setting analysis unit 22.

  The operation of the current-voltage characteristic measuring device 2 provided with the power supply device 20 according to the present embodiment will be described.

The time-series pulse signal generation / output unit 202 of the power supply device 20 reads the two-function time-series pulse pattern stored in the two-function time-series pulse pattern storage unit 200. The data of the two-function time-series pulse pattern is generated and output by the method already described in the two-function time-series pulse pattern setting unit 220 of the setting and analysis unit 22 and is input and stored in the power supply device 20. Further, the maximum amplitude setting unit 201 sets a value input from the outside as the maximum amplitude value. The time-series pulse signal generation / output unit 202 generates a time-series current pulse signal of the maximum amplitude value I ₀ based on the two-function time-series pulse pattern and the maximum amplitude value I ₀ and outputs it to the measurement object 1 At the time of output, a trigger signal is output to the voltage measurement unit 21 as a signal for synchronization. Also, the peak value of each pulse of the time-series current pulse signal is output to the analysis unit 221 of the setting / analysis unit 22 as time-series information.

  When a time-series current pulse signal is output from the power supply device 20 to the measurement object 1 and a trigger signal is input to the voltage measurement unit 21, the voltage measurement unit 21 performs a time-series response of voltage drop at the measurement object 1. The measurement is performed in the form of the net pulse peak value of each response voltage pulse of the voltage pulse signal, and is output to the analysis unit 221.

  The analysis unit 221 is a peak value of each pulse of the time-series current pulse signal input from the power supply device 20 and a net pulse of each response voltage pulse of the time-series response voltage pulse signal input from the voltage measurement unit 21 corresponding thereto. Information of the I-V characteristic is determined based on the peak value, and the result is output to the display unit 23. The display unit 23 displays the analysis result input from the analysis unit 221.

  The two-function time-series pulse pattern setting unit 220 of the setting / analysis unit 22 generates data of the two-function time-series pulse pattern based on the two functions f (t) and g (t) input from the outside. Output to The power supply device 20 stores this data in the two-function time-series pulse pattern storage unit 200.

Two series of function time series pulse patterns generated based on two functions f (t) and g (t) with which each sum of squares becomes a constant value, that is, function f The first time-series pulse pattern generated based on (t) and the second time-series pulse pattern generated based on the function g (t) are composed for the following reasons. The time-series current pulse signal generated by multiplying the 2-function time-series pulse pattern by I ₀ is based on F (t) and G (t) obtained by multiplying f (t) and g (t) by I ₀ respectively. It can be regarded as generated. Therefore, if the sum of the squares of two functions f (t) and g (t) becomes a constant value regardless of t, F (t) and G (t) also have their sum of squares t It becomes a constant value regardless of it. If the peak value of each current pulse of the time-series current pulse signal is alternately selected and set from F (t) and G (t), the time interval of adjacent current pulses becomes smaller, the square of each peak value That is, the sum of squares of current values is substantially constant regardless of time. Since Joule heat is proportional to the square of the current, when the time-series current pulse signal set as described above is input to the measurement object 1, the Joule heat generated in the measurement object 1 by the adjacent current pulses is averaged. The value does not depend on the time t, that is, it becomes substantially constant in the process of changing the current for measurement of the IV characteristic. Therefore, it is possible in principle to prevent the temperature of the measurement object 1 from being systematically correlated with the current.

  The 2-function time-series pulse pattern is the same as the pattern of the time-series current pulse signal except for its peak value. Therefore, the details of the time-series current pulse signal will be described below in comparison with the time-series response voltage pulse signal based on FIG. 5.

FIG. 5A shows an enlarged view of a time series current pulse signal corresponding to two series of function time series pulse patterns included in the region K of the time axis of FIG. In order to clarify the relationship between the time-series current pulse signal and the function f (t) and the function g (t), the function f (t) and the function g (t) multiplied by I ₀ respectively The function G (t) is shown by a broken line. FIG. 5 (B) shows a measurement example of the time-series response voltage pulse signal of the measurement site K ₁ corresponding to the time-series current pulse signal shown in FIG. 5 (A), aligning the time axis with FIG. 5 (A). It is a thing.

The time-series current pulse signal is a train of pulse signals aligned at time intervals of Δt. Each pulse has a time width (pulse width) Δt _p and a wave height value set based on F (t) or G (t), and between adjacent pulses, a wave height value of time width W 0 There is a section of Δt = Δt _p + W.

The peak value of each pulse is set by alternately selecting F (t) and G (t) as follows. t ₀ is the start of measurement, i is the upper limit _{i n} a natural number of consecutive containing 0, with respect to time _{t, t 2i = t 0 +} Δt × 2i, and _{t 2i + 1 = t 0 +} Δt × (2i + 1) and when defining As shown by points A and B in FIG. 5A, the crest values of adjacent pulses of the time-series current pulse signal are F (t _2i ) and G (t _{2i + 1} ) for each i, respectively. It is set equally. 2i and 2i + 1 denote even-numbered and odd-numbered sections when the time axis is divided by the time interval Δt between pulses, in other words, the section width Δt, and the time to represent each section is t _2i and t _{2i + 1} It is. That is, the peak values of the even-numbered sections are set according to F (t), and the peak values of the odd-numbered sections are set according to G (t). Points A and C indicating the peak values of the pulses corresponding to the even-numbered sections in FIG. 5A are the values of F (t _2i ) and F (t _{2 (i + 1)} ), respectively, and the odd-numbered sections The point B indicating the peak value of the pulse corresponding to the is the value of G (t _{2i + 1} ). That is, the time-series current pulse signal has a first time-series current pulse signal whose envelope of peak value is represented by F (t) and a second time-series whose envelope of peak value is represented by G (t) It can be said that it is composed of two series of signals having a current pulse signal. Of course, the order of peak value selection for F (t) and G (t) may be reversed.

In FIG. 5A, points such as points A, B and C for obtaining values of F (t) and G (t) are located at the middle points of the respective pulse widths, but are limited thereto I will not. The points A, B, C, etc. may be located within the range of the pulse width Δt _p of the corresponding pulse.

In order to determine the pulse width Δt _p and the time width W between pulses, it is necessary to consider the pulse waveform of the time-series response voltage pulse signal. The pulse width of the time series response voltage pulse signal, as shown in FIG. 5 (B), the same as the pulse width Delta] t _p of the time series current pulse signal. In addition, the pulse waveform shown in FIG. 5B shows not a sharp linear shape but a rounded shape like the rising and falling portions of the pulse shown in FIG. 5A at the rising and falling portions thereof. the is because the voltage measuring unit 21, has a response time Delta] t _r by time constant. Although a time delay may occur in the time-series response voltage pulse signal with respect to the time-series current pulse signal, this time delay can be grasped in advance, and it is only parallel movement on the time axis. Ignored and displayed.

Since the pulse width of the time-series current pulse signal and the pulse width of the corresponding time-series voltage pulse signal have the same value, the values of Δt _p and W are measured by the voltage measuring unit 21 under the following conditions: It is set to run. The measurement in the pulse region requiring the measurement time Δt _m by the voltage measurement unit 21 is after the response time Δt _r has elapsed from the rise of the pulse signal and when the fall of the voltage pulse signal starts, ie, the rise of the pulse signal It must be performed within the time of Δt _p from the start time. In addition, the measurement that requires measurement time Δt _m by the voltage measurement unit 21 in the region between adjacent pulses, that is, in the offset region, the voltage pulse after the response time Δt _r has passed from the start of the fall of the pulse signal It must be carried out within the time of W from the start of the rise of the pulse signal, that is, from the start of the fall of the pulse signal. The Δt _p and W that satisfy these conditions can be expressed by the following equations (2) and (3). Since Δt _r and Δt _m can be set as known values once the voltage measurement unit 21 is determined, the values of Δt _p and W are set according to the equations (2) and (3), respectively.
Δt _p ΔΔt _r + Δt _m (2)
W Δ Δt _r + Δt _m (3)
The Δt _p and W may be set to a time width obtained by adding a predetermined time margin to the right sides of the equations (2) and (3). Further, since Δt = Δt _p + W, Δt _p and W are set in consideration of how Δt is set.

  The above description holds true if F (t) and G (t) are replaced with f (t) and g (t) for the bifunctional time-series pulse pattern, respectively.

  The setting of the interval Δt of each pulse of the 2-function time-series pulse pattern will be described. In the present invention, f (t) and g (t) are selected as two functions such that the sum of the respective squares does not depend on the time t and becomes a constant value H. In the example shown in FIG. 2 and FIG. 5A, the constant value H is “1”. As described above, in order for the sum of squares to be a constant value H, t must be exactly the same value in two functions.

However, as can be seen from F (t), G (t) (proportional to f (t), g (t)) shown in FIG. 5 (A), f (t) and g (t) are mutually t Values at positions different by Δt, ie, f (t _2i ) and g (t _{2i + 1} ) (t _{2i + 1} −t _2i = Δt, i; continuous natural numbers including 0; up to a preset upper limit value Pulse peak value is set. The value J (i) obtained by adding f (t _2i ) ² and g (t _{2i + 1} ) ² usually changes depending on Δt and i, and becomes a constant value regardless of the value of i. Absent.

However, if f (t) and g (t) are selected such that the sum of their squares is constant, the degree of change depending on i of J (i) as Δt becomes smaller, ie, The degree of time-dependent change is smaller and can be regarded as a substantially constant value. The determination criterion as to whether it can be regarded as a substantially constant value is determined by the permissible time change of Joule heating corresponding to the permissible temperature change of the measurement object 1. Since Joule heat is proportional to the square of the current, an allowable variation value of J (i) can be calculated from the change of allowable time of Joule heat, and the calculated value is a judgment standard regarding substantially constant. The fluctuation value of J (i) becomes large depending on the error ε with respect to the constant value (“1” in the example shown in FIG. 2) of the value of J (i). _Assuming that the error ε _th with respect to the fixed value of the allowable fluctuation value of J (i) is a threshold error or an allowable error, the difference ε between the value of J (i) and the fixed value becomes smaller than the allowable error ε _th Is set as That is, if the Delta] t in the case of epsilon = epsilon _th and Delta] t _max, Delta] t the square sum can be regarded as substantially constant may be set to the following values Delta] t _max. Δt is a value larger than the sum of Δt _p and W obtained from the expressions (1) and (2), and may be any value as long as it satisfies the condition of a value of Δt _max or less. As a result of considering the above conditions, Δt _p and W are usually set to about 5 to 100 ms (more preferably about 5 to 50 ms) in consideration of the measurement time, the temperature of the object to be measured, and the like.

The time range T of the 2-function time-series pulse pattern will be described. This time range T is a time range in which the amplitude value of the time-series current pulse signal after setting s maximum amplitude value I ₀ can cover the current value necessary for acquiring the information of the I-V characteristic of the measuring object 1 Set to The value of i described in the setting of the time-series current pulse signal has its maximum value set corresponding to the time range set in this way. The time range T shown in FIG. 2 indicates the time range from time t = 0 to one period T when f (t) = COS (2πt / T) and g (t) = SIN (2πt / T) . When Fourier analysis is performed by the setting / analysis unit 22, the data to be used corresponds to a periodic function, and at least one period is necessary, so f (t) and g (t) are functions of the period T '. Then, T = n x T '(n = natural number).

The calculation method of the net voltage pulse wave peak value by the voltage measurement unit 21 will be described using FIG. 5 (B). Although the peak value of each pulse of the time-series response voltage pulse signal is measured at the measurement time Δt _m , the measurement result is the peak value at a point representing this measurement time. In FIG. 5B, the representative point is, for example, a time point (displayed at a, b, c, etc.) of Δt _m / 2, which is a middle point of measurement time in each pulse. Since the drift component of the time-series response voltage pulse signal is indicated by the peak value in the offset region, the measurement of the drift component can be performed on both sides of points a, b, c, etc., as shown in FIG. It carries out by measurement time (DELTA) t _m centering | focusing on the time position each distant by time (DELTA) t _s . In FIG. 5B, the time positions on both sides of the point b are shown by the point d and the point e. Note that the measurement time of the peak value in the offset region may be a value different from the measurement time Δt _m of the pulse peak value.

Taking the point b as an example, a method of calculating the corresponding net pulse wave peak value V _x will be described. The voltage measurement unit 21 obtains measurement values V _b , V _d , and Ve of the representative point b of the pulse peak value measurement, the representative point d of the peak value measurement of the offset region, and the point _e by measurement. The voltage measuring unit 21 calculates the drift component V _y at the point b as approximately equal to (V _d + V _e ) / 2, and calculates the net pulse wave height value V _x corresponding to the point _b from V _b to V _y . Obtained by removing. This method is called a three-point method and is a method used in a commercially available voltage measurement device.

In FIG. 6, a sintered body of the high-temperature superconductor YBa ₂ Cu ₄ O ₈ is used as a sample constituting the object to be measured 1, that is, a measurement site K ₁ by the current-voltage characteristic measuring device 2 described above. The example which measured the I-V characteristic of is shown. In this example, equation (1) is not obtained, and the temperature of the installation environment of the measurement object 1 is used as a parameter, and the current value input to the measurement object 1 and the voltage between both ends of the measurement site K ₁ are Is a plot of From such a display, the degree of variation of the data of the I-V characteristic for each temperature can be known. The greater the degree of variation, the greater the temperature fluctuation of the object 1 to be measured. The temperature of the installation environment is 23.50K, 27.40K, 29.50K and 31.40K. Although the mutual temperature difference is as small as about 2 K to 3 K, it can be seen that the variation in the IV characteristic data of the measurement site K ₁ at each temperature is small, and the mutual data can be distinguished well. From this result, the power supply 20 current comprises according to this embodiment - According to the voltage characteristic measuring device 2, can be obtained the I-V characteristic of the measurement site K ₁ of the measuring object 1 with high precision and It could be demonstrated that Therefore, it is understood that when the power supply device 20 according to the present embodiment is used, the temperature change of the measurement object 1 becomes so small that the IV characteristic can be measured with high accuracy to the extent shown in FIG.

Thus, power supply device 20 according to the present embodiment generates a time-series current pulse signal based on functions F (t) and G (t) such that the sum of the respective squares becomes a constant value, Since the output is performed, according to the power supply device 20, when the output signal is input to the measurement object 1, it is possible to suppress the value obtained by averaging the Joule heat generated in the measurement object 1 by the adjacent current pulses substantially constant. . Furthermore, the current including the power supply 20 - According to the voltage characteristic measuring apparatus 2, since the temperature change of the measured object 1 in the measurement can be made smaller than conventional measurement site K ₁ of the measuring object 1 It is possible to obtain the I-V characteristic of V with higher accuracy.

In the above explanation, a two-function time-series pulse pattern generated by using f (t) as a cosine function and g (t) as a sine function has been described, but f (t) and g (t) are functions shown in FIG. It may be The example shown in FIG. 7 is a first modification of the two-function time-series pulse pattern. The two functions at this time are f (t) = 2t / T−1 and g (t) = SQR (1- (2t / T−1) ² ), where T is a time range or period. That is, in the time range of t = 0-T, f (t) is a linear function that zero-crosses from negative to positive at T / 2, g (t) has its center at the position of T / 2 on the time axis, It is a function that represents the positive part of the equation that represents an ellipse in which the axis length in the time axis direction is T and the axis length in the g (t) direction is 2. Also in the case of such functions f (t) and g (t), the sum of the respective squares becomes a constant value regardless of t.

  At this time, data used for Fourier analysis in the analysis unit 221 is data of current pulse peak values corresponding to the linear function and net voltage pulse peak values corresponding thereto. The reason is that the I-V characteristic is required for both positive and negative values. Also in this case, except for the difference in the function form, the contents of the description so far hold true, and the power supply device 20 and the current-voltage characteristic measuring device 2 including the same exhibit the same effects as the effects described above.

  The analysis unit 221 may obtain the I-V characteristic in the form of Equation (1) by a polynomial fitting method. Also, the IV characteristics need not necessarily be determined in the form of equation (1), and may be generated simply as a current value and a corresponding response voltage value displayed in a tabular form or a graph. In this case, all pulse peak values of the time-series current pulse signal input to the measurement target 1 and the net response voltage pulse peak value corresponding thereto can be utilized for grasping the IV characteristic, which is a waste. There is an advantage that there is no

Although the maximum value of the absolute value on the vertical axis in FIG. 7 is 1, it is not limited thereto. It may be set to an arbitrary value I _a . However, in this case, when generating the time-series current pulse signal so that the maximum value of the time-series current pulse signal is equal to I ₀ set by the maximum amplitude setting unit 201, a value by which the two-function time series pulse pattern is multiplied Is not I ₀ but I ₀ / I _a .

  A second modification of the two-function time-series pulse pattern is shown in FIG. The point that f (t) at this time is a cosine function is the same as the case where f (t) described above is a cosine function. However, the difference is that the time range of f (t) used to generate a bifunctional time-series pulse pattern is 1/2 of the period T. Furthermore, for g (t), a sine function having the same period T as f (t) and the same amplitude and phase as g (t), the time range of g (t) used to generate the bifunctional time-series pulse pattern has a period T And t is a function obtained by inverting the sign in the range from T / 4 to T / 2. The two-function time-series pulse pattern shown in FIG. 8 is generated as described above based on such f (t) and g (t). Also in this case, the sum of the square of f (t) and the square of g (t) is constant.

  When obtaining the IV characteristic in the form of equation (1), the analysis unit 221 converts the phase of data corresponding to the discontinuous function g (t) and combines it with the data corresponding to f (t). Do. The synthesized data is used as data of one cycle of one continuous function. That is, the current pulse peak value of the time-series current pulse signal corresponding to g (t) and the net voltage pulse peak value of the corresponding time-series response voltage pulse signal at t = 0 to T / 4 迄The phase of the data is shifted on the time axis by -T / 4 corresponding to -90 degrees, and the phase of the corresponding data from t = T / 4 to T / 2 is only + T / 4 corresponding to +90 degrees The current pulse peak value of the time-series current pulse signal corresponding to f (t) and the net voltage pulse peak value of the time-series response voltage pulse signal corresponding to this are connected on the axis, and synthesized. As a result, the current pulse peak value of the time series current pulse signal corresponding to one sine function of one period and the net voltage pulse peak value of the time series response voltage pulse signal corresponding thereto can be obtained as measurement data. It will be.

  The analysis unit 221 can apply Fourier analysis to the data of one period generated by combining to obtain information of the IV characteristic according to Equation (1). According to this method, all time-series current pulse signals can be utilized, so that there is an effect that there is no waste. Furthermore, since the measurement time of the data can be T / 2, the efficiency of the measurement can be improved.

  The time-series current pulse is also generated when the analysis unit 221 generates the IV characteristic not in the form of equation (1) but as a current value and a corresponding response voltage value displayed in a table format or a graph. Since all the signals can be used, there is no waste, and the measurement time of data can be T / 2, so that the measurement can be shortened. Further, as can be seen from FIG. 8, since the sign of the current input to the measurement object 1 can be obtained in a short time with good balance between positive and negative, in that sense, the IV characteristics can be accurately obtained in a short time.

  There are various methods for combining one periodic function by shifting the phase of f (t) and g (t) with respect to one of the other. But you can get the same effect.

The two-function time-series pulse pattern is generated based on the functions f (t) and g (t) in which the sum of the squares becomes constant, but the phase of the function g (t) is shifted by Δt Define the function g '(t) as g' (t) = g (t-.DELTA.t), and the bifunction time series pulse pattern is the same as before based on the functions f (t) and g '(t) It may be generated by a method. As for the time-series current pulse signal, the time-series current pulse signal is obtained based on F (t) and G ′ (t) t obtained by multiplying f (t) and g ′ (t) by I ₀ respectively. It may be regarded as generated.

  When time-series current pulse signals are generated based on F (t) and G (t), the representative times t of adjacent pulses are shifted by Δt, and the simultaneity of pulses targeted for sum of squares is lost And that the pulse height value of the pulse is not synchronized at the same time due to the deviation of the synchronization, that is, the values are different, so that the sum of squares becomes an error from a constant value not depending on time. The thing is as already explained. When the time-series current pulse signal is generated based on F (t) and G '(t), the representative time t of adjacent pulses is shifted by Δt, and the value itself to be subjected to the square sum is simultaneity It will be secured. Therefore, since the error of that amount is reduced, according to the power supply device 20 which outputs such a time-series current pulse signal, in the measurement target 1 at the output destination, two adjacent pulses of the time-series current pulse signal to be output The uniformity of the averaged value of Joule heat is further improved compared to the cases described so far. According to the current-voltage characteristic measuring apparatus provided with such a power supply device 20, the improvement in the accuracy of the measurement of the IV characteristic of the measurement object 1 is further improved.

  In a fourth modification (not shown) of the bifunctional time-series pulse pattern, the absolute value of the peak value of each adjacent pulse of the bifunctional time-series pulse pattern is the absolute value of f (t) and g (t) It is assumed that they are set alternately based on the absolute value of, and the sign thereof is freely set to either positive or negative. This is because the generation of a bifunctional time-series pulse pattern only involves the sum of the squares of the relied on bifunctions, and the square becomes irrelevant to the sign when it is squared. However, since it is necessary to input positive and negative currents to the measurement object 1 in the measurement of the I-V characteristic, it is necessary to set a two-function time-series pulse pattern in consideration of that point. The above-mentioned second modification may be regarded as a special case of this third modification. In the case of such a two-function time-series pulse pattern, normally, the analysis unit 221 uses the method other than the Fourier analysis to analyze the information of the IV characteristics except when combining the periodic function as in the second modification. get. At this time as well, the time variation of the temperature of the measurement object 1 can be neglected, and the measurement accuracy of the IV characteristic is improved.

  Heretofore, the case has been described in which the voltage drop at the measurement target location of the measurement target 1 when the current is supplied to the measurement target 1 is measured to obtain the IV characteristic. However, contrary to this, an IV characteristic may be obtained by applying a voltage to the measurement target portion of the measurement target 1 and measuring the current flowing to the measurement target 1 at that time. A configuration example of the current-voltage characteristic measuring device 2 at such a time is shown in FIG.

9, the difference from FIG. 1 is that the power supply device 20 is a voltage supply source, and outputs a time-series voltage pulse signal to the measurement target unit 1, and the voltage measurement unit 21 of FIG. Measuring the current flowing to the measurement object 1 to which the time-series voltage pulse signal is applied as a time-series response current pulse signal, the connection between the current measurement unit 21 and the measurement object 1 is the measurement site K of the measurement object 1 It is wired so that the current flowing to ₁ can be measured. The current measurement unit 21 and the voltage measurement unit 21 illustrated in FIG. 1 have the same reference numerals because they can be combined as a response signal measurement unit 21 that is the upper concept.

The voltage is applied to the object to be measured 1 to measure the I-V characteristic, in general, when the resistances of the electrical resistances K ₂ and K ₃ can be neglected as compared with the resistance of the measurement site K ₁ . In such a case, the voltage applied to the measuring object 1 is equal to it a voltage drop across the measurement site K _1. Therefore, even without measuring the voltage drop value of the measurement site K _1, it can be transmitted as a voltage value to the setting and analysis unit 22 each pulse height of the time-series voltage pulse signal output by the power supply 20. The data set analysis output from the current measurement unit 21 is a net current pulse peak value.

  With respect to the apparatus having the configuration shown in FIG. 9, on the premise of the above, if the description regarding FIG. 1 is read by replacing the current and the voltage, the description regarding FIG. The same applies to the effects described above including the modified example. Therefore, the power supply device 20 shown in FIG. 9 can output a time-series voltage pulse signal that becomes substantially constant regardless of time when Joule heat generated in the measurement object 1 by adjacent pulses is averaged. According to the current-voltage characteristic measuring apparatus provided with such a power supply device 20, since the temperature fluctuation in the measuring object 1 in the process of measuring the IV characteristic can be ignored, I-- with high accuracy. It becomes possible to measure V characteristics.

Even in the case of measuring a current by inputting a voltage in the measurement of IV characteristics, the voltage across the measurement site K ₁ can be measured by the response signal measuring unit 21 shown in FIG. 9 or a voltage measuring unit separately provided. If, by inputting the voltage measurements in the analysis unit 221, regardless of the resistance value of the measurement site K _1, the the I-V characteristic of the measuring object 1 can be measured with high accuracy.

  The output of the power supply device 20 is a time-series current pulse signal in FIG. 1 and a time-series voltage pulse signal in FIG. These can be generically referred to as time-series pulse signals as shown in FIG. 1 and FIG. Further, although it has already been described that the voltage measurement unit 21 of FIG. 1 and the current measurement unit 21 of FIG. 9 can be collectively referred to as the response signal measurement unit 21, time series response voltage pulse signals of the voltage measurement unit 21, The time-series response current pulse signals of the current measurement unit 21 can be collectively referred to as time-series response pulse signals.

1 measurement object 2 current-voltage characteristic measuring device 20 power supply device 21 response signal measuring unit (voltage measuring unit, current measuring unit)
22 setting / analysis unit 23 display unit 200 2 function time series pulse pattern storage unit 201 maximum amplitude setting unit 202 time series signal generation / output unit 220 2 function time series pulse pattern setting unit 221 analysis unit 300 central processing unit (CPU)
310 internal memory 320 external memory 330 input section 340 output section 350 bus line K ₁ measurement site (electrical resistance)
_{K 2,} K ₃ electrical resistance

Claims (7)

What is claimed is: 1. A power supply device for outputting two series of time-series pulse signals of current or voltage, comprising:
For time t, F (t) and G (t) are functions satisfying the condition that the sum of the respective squares becomes a constant value,
Let t be a continuous natural number in a predetermined range including 0, i be a measurement start time to be t ₀ , and a predetermined time width be Δt. t _2i = t ₀ + Δt × 2i and t _{2i + 1} = t _{When 0} + Δt × (2i + 1) ,
Of the two series of time series pulse signals, the absolute value of the pulse _peak value of the first series of time series pulse signals is equal to the absolute value of F (t _2i ), and the first series of time series pulse signals The absolute value of the pulse height value of the pulse of the second series of time-series pulse signals adjacent to each of the pulses in the above is equal to the absolute value of G (t _{2i + 1} ),
The difference between the sum of the square of F ( _t2i ) and the square of G ( _{t2i + 1} ) and the constant value does not depend on the value of i, and? The value is preset to be
A power supply device characterized by F (t) is a cosine function, and G (t) is a sine function having the same period, the same amplitude, and the same phase as F (t),
The power supply device according to claim 1, characterized in that: F (t) is a linear function that has a period T and becomes 0 at time T / 2, and G (t) has a center at T / 2 and an axial length in the direction indicating time t T is a function representing either positive or negative of an equation representing an ellipse whose axis length in the direction indicating the value of G (t) is equal to F (0),
The power supply device according to claim 1, characterized in that: As G ′ (t) = G (t−Δt),
The absolute value of the pulse peak value of the second series of time-series pulse signals is a value equal to the absolute value of G ′ (t _{2i + 1} ),
The power supply device according to any one of claims 1 to 3, characterized in that. The power supply device according to any one of claims 1 to 4.
The voltage drop or current at the measurement target point when the time-series pulse signal is input to the measurement target point of the IV characteristic of the measurement object is synchronized with the time-series pulse signal as a time-series response pulse signal A response signal measurement unit that measures and outputs the net peak value of each pulse of the time-series response pulse signal;
A current-voltage characteristic measuring apparatus comprising: The power supply device is the power supply device according to claim 2;
The preset period, the first information the a peak value of each pulse of the time series pulse signal of the first series, second is the peak value of each pulse of the time-based Retsupa pulse signal of the second series inputs the second information, based on the first information and the second information to the input, by performing a Fourier analysis, the position to be measured of the current - analysis unit to obtain information about voltage characteristics,
The current-voltage characteristic measurement device according to claim 5, comprising:   The power supply device is the power supply device according to claim 2;
  A first information which is a peak value of each pulse of the first series of time-series pulse signals in a preset period, and a second value which is a peak value of each pulse of the second series of time-series pulse signal Information is input, and Fourier analysis is performed on the basis of the input first information and the time-series response pulse signal measured corresponding to the second information, thereby current of the measurement target point Analysis unit to obtain information on voltage characteristics,
  The current-voltage characteristic measurement device according to claim 5, comprising:

Images