JP2001255345A - Method and device for measuring capacitance of capacitor and equivalent series resistance - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and device for measuring the capacitance and equivalent series resistance of a large-capacity electrical double-layer capacitor. SOLUTION: A rectangular wave constant current of ±Ic with a frequency of (f) is allowed to flow to an electrical double-layer capacitor, voltage data exceeding one period is examined, discharge and charge waveforms excluding a transition part are approximated by a straight line, a capacity is obtained from the inclination of charge regression and discharge regression straight lines, and an equivalent series resistance ESR is obtained from the amount of rise and fall at the discontinuity of the regression straight lines. By changing the frequency, the frequency characteristics of the capacitance C and the resistance ESR can be obtained. The C and ESR can also be measured by applying an arbitrary bias voltage.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for measuring a frequency-dependent equivalent resistance R and a frequency-dependent capacitance C of a large-capacity capacitor such as an electric double-layer capacitor (capacitor). A capacitor is an element that has two electrodes and stores a proportional charge when a voltage is applied between the electrodes. The ratio of the voltage V to the charge Q is the capacitance C.
C is given by C = εS / d. That is, the larger the electrode area S, the higher the dielectric constant ε, and the smaller the distance d between the electrodes, the larger the capacitance C.

There are many types of capacitors (capacitors). Ceramic capacitors, mylar capacitors,
There are many things such as tantalum capacitors, mica capacitors, and electrolytic capacitors. The origin of the name often depends on the material of the dielectric, but it is not the only one. Those having a small capacity are represented by pF (10 ⁻¹² F). Above it is expressed in μF (10 ⁻⁶ F). nF ^{(10 -9} F) and mF
The unit of (10 ⁻³ F) is not used much. The characteristics of a capacitor (capacitor) are expressed by capacitance, withstand voltage, frequency characteristics, and the like. A small-capacity capacitor with good high-frequency characteristics,
Large-capacity capacitors with poor frequency characteristics have different applications. The latter is used for stabilizing the power supply section of the electric circuit. The electrolytic capacitor generally has a low withstand voltage and a large capacity (several tens to several hundreds of μF) and is suitable for a power supply unit. The electrolytic capacitor has a large capacitance,
^0-4F . Electrolytic capacitors have polarity and are destroyed when voltage is applied in the opposite direction. Smaller capacitors for high frequency use have no polarity, so there is no such concern.

[0003] These capacitors have good frequency characteristics, and the value of the capacitance C does not change much within the operating frequency range. The present invention is concerned with electric double layer capacitors (DLC)
It is. This is a capacitor having a low withstand voltage but an extremely large capacity. The capacity is about 1F. By increasing the capacity, a capacitor having a large capacity of 10F or 100F can be manufactured. However, the withstand voltage is low, 2.5
About V. It's just like 5V by connecting two stages. The structure is such that fine particles of activated carbon are dispersed and fixed in a solution such as dilute sulfuric acid.
Is extremely wide. Although a double layer is formed on the surface layer of the activated carbon, d is small because the layer is extremely thin. Therefore, the capacitance C increases. Instead of increasing C by increasing the dielectric constant ε, it is intended to realize a huge capacitance C by decreasing d and increasing S.

In some cases, the conventional capacitor technology cannot be used because the capacitance is much larger than that of a normal capacitor. Further, when the frequency rises due to poor frequency characteristics, the capacity is extremely reduced. Therefore, there are times when only DC use is possible. Further, the DC resistance R becomes a problem. There are various differences from ordinary capacitors. Here, it is called a capacitor to distinguish it from a normal capacitor.

[0005]

2. Description of the Related Art An electric double layer capacitor has an intermediate property between a capacitor and a battery. In terms of storing energy, it has properties similar to batteries. It is also similar to batteries in that it has poor frequency characteristics and lacks high speed. The former means that the stored energy E = 0.5 CV ² is large because the capacitance C is large. Since the withstand voltage is low, it is necessary to connect a plurality of stages in series. The latter comes from the fact that the fine activated carbon is dispersed in the solution.
The current flow depends on not only electrons but also cations and anions of the electrolyte. The movement of these ions is slow. Can't follow fast exchange. As the frequency increases, the capacitance C decreases to a very small value. Therefore, it is thought that electric double layer capacitors can be used only for DC applications.

[0006] On the other hand, however, the capacitor can be regarded as a capacitor because the charge is held by the voltage between the electrodes instead of storing the charge by a chemical change. Response speed is slower than battery. Thus, the electric double layer capacitor is intermediate between a capacitor and a battery.

The electric double layer capacitor itself is not new.
It has been practically used for more than 20 years. An electric double layer capacitor having a huge capacitance C is a computer,
Some are used for memory backup of electronic devices. Works as an emergency power source during a power outage. It is useful when the commercial voltage drops even if there is no power outage. However, there are many suitable batteries for backup, and batteries are still mainstream. One year since backup
It is used only once every few years and only rarely.
Static use. For this purpose, a small DLC capacitor having a withstand voltage of 3 V or 5 V and 1 F or less is used. The production scale was small and the number of elements used was small. The device was aside and used in a DC manner. Frequency characteristics do not matter because of DC specifications.

However, in recent years, new applications are being opened for electric double layer capacitors. This is an application as a capacitor for retaining energy such as electric vehicles and wind power generation. Electric vehicles use rechargeable batteries as energy sources. Batteries generate electric current using a chemical reaction. Charging / discharging means that the chemical reaction proceeds reversibly to both sides. The chemical reaction takes time, so it takes time to charge or discharge. The battery is used to turn the motor, but in the case of rapid acceleration, the battery cannot follow. It is being sought to improve the performance of the battery to enable rapid acceleration and start.

However, there is another way. That is, a capacitor is used in parallel with the battery to supplement the battery function. The capacitor is Q = C without chemical reaction
Only the charge is stored by the relation of V. Rapid discharge and rapid charge are possible. Therefore, it is only necessary to draw current from the capacitor at the time of sudden acceleration, starting, etc. Capacitors store less charge and less stored energy. Energy storage depends on the battery. In other words, capacitors are used dynamically, and batteries are used statically. Since the characteristics of a battery are significantly deteriorated when it is rapidly discharged and charged, it is desirable to insert a capacitor in parallel in order to avoid this.

In braking, a current flows in the opposite direction. If the current is absorbed by a capacitor, energy loss can be advantageously reduced. For this reason, a combination of a chargeable / dischargeable battery and a large-capacity capacitor has been studied as an energy source for an electric vehicle. There is a demand for using a large-capacity electric double layer capacitor.

Another application that is aimed at is an energy retention device for wind power generation. In wind power generation, the wind turns a windmill and turns a generator to generate and transmit electric power, and a portion is stored in a battery. However, wind power fluctuates every moment. The amount of power generation also changes significantly. Batteries are static in nature. Not suitable for severe current-voltage fluctuations. When charging a battery, it is desirable that the current and voltage are constant. Fluctuations in voltage and current damage the battery and shorten its life.

When a large-capacity capacitor is used in combination, it is possible to absorb a rapid change in voltage and current. There has been a growing demand for using electric double layer capacitors. With a large-capacity capacitor, a battery can be charged with a constant current and voltage. Its purpose is to always work and complement the battery. It is not a backup that always sleeps. It has the important purpose of continuously working remarkably and absorbing the fluctuation of the current and voltage to protect the battery. This is a completely new use of an electric double layer capacitor that has been used only in small ways such as backup. That is why attention was paid to the point that the capacitance was large.

As described above, the electric double layer capacitor is expected to be promising in two attractive fields such as electric vehicles and wind power generation. However, it has not yet been realized. There is not yet an electric double layer capacitor that meets the requirements. The same purpose may be achieved by improving the performance of the battery. This is because it is only necessary to have a battery that can withstand severe voltage and current changes. Whether the bright future of electric double layer capacitors will open or not will be a race against the speed of battery development.

There are a number of problems that need to be overcome. One is the pressure resistance. Since the activated carbon is dispersed in an electrolyte solution such as dilute sulfuric acid, the pressure resistance is low. It is as low as 1V or 2.5V. 300 for electric vehicles
V power supply voltage is required. Higher power supply voltages are needed for energy storage for wind power. Such a high withstand voltage electric double layer capacitor does not yet exist. 2.5
If V elements are connected in 120 stages, the voltage should be 300 V. However, connecting such a large number of capacitors is dangerous. Voltage fluctuations occur between the elements, and become unstable. Some ingenuity is needed.

Another problem is the capacitance C. One by one
Even if there is a capacity of F, it becomes 0.01F when 100 pieces are connected in series.
Will drop. However, the key for storing energy
Because it is a capita, it is not enough on the first floor. Capacitor
Holding energy is 0.5CV ²But this is greatly
To do so, the capacity C must be large. So 10F
Or electric double layer cap with a large capacity such as 100F
Pasita is desired. When connected in parallel, the capacitance of the capacitor
That increases. 2.5V breakdown voltage and 1F capacitance in the previous example
Of 100 capacitors in series and 100 in parallel
(10000 pieces) withstand voltage of 250V and only 1F
No. It takes 100,000 to make 10F
Will be. Such a large number of elements are arranged in series and parallel
They are expensive and cannot be used in large volumes. Endurance
It is necessary to overcome pressure and capacity problems.

[0016] That alone is not the end of the story. The frequency characteristics of the electric double layer capacitor also become a problem. The response speed of an electric double layer capacitor is extremely slow because heavy ions carry charges. A little better than a battery.
In the case of an old usage such as a backup at the time of a power failure of an electric device, the frequency characteristic is not a problem because the direct current (f = 0) is used. However, if it is used for power storage such as electric vehicles and wind power generation, it must be able to cope with fluctuations in current and voltage, so frequency characteristics become a problem. An electric double layer capacitor using an ion flow, unlike a capacitor using a normal electron flow, is extremely disadvantageous in that respect. In the case of an electric double layer capacitor, even if it is 1F in direct current, the capacity drops sharply in the case of alternating current. In a new application, it is necessary to evaluate how much C (f) falls with respect to the alternating current f. It cannot be used just because the DC capacity C is large. There must be a desired capacitance C up to a certain frequency. The frequency characteristic is a new evaluation performance for the electric double layer capacitor.

There is no problem in the case of DC, but there is another factor that affects the performance in the case of AC specification. It is a series resistor. Since ions are made of activated carbon and a solution and move in an electric field, there is a time delay and a DC resistance appears. Although each activated carbon particle k is considered to be a series body of C _k + R _k , the electric double layer capacitor as a set thereof can also be considered to be a series body of C + R. This is a series resistance, not a parallel resistance. So it does not lead to permanent power loss. However, the charging current flows only slightly when the polarity is reversed. In other words, this is not a problem in the case of DC, but a problem in the case of AC use. If the series resistance is large and the frequency is high, the power loss due to the series resistance cannot be ignored. This means that the series resistance R is also subject to performance evaluation. In the case of electric vehicles and wind power storage, alternating current is used. Therefore, the dependence of the series resistance R on the frequency f must be examined. Thus, R (f) can be written as a function of frequency. pF
In the case of ordinary capacitors of the order and μF order, the series resistance R is very small and does not matter much, but in the case of an electric double layer capacitor, it may have a size that cannot be ignored. So it can be said that the measurement of R is more important. This is Equivalent Series Resistance (ESR)
Also called.

[0018]

In order to use an electric double layer capacitor for electric vehicles and wind power generation, it is essential to improve the capacitor itself by increasing the breakdown voltage and capacity. That's what matters. Must be able to evaluate whether the performance has improved or is actually getting better. The evaluation must be able to examine not only DC characteristics but also dynamic characteristics, that is, frequency dependence. In an electric double layer capacitor, the capacitance C,
The withstand voltage V, the series resistance R and the like are important factors. Among them, it is an object of the present invention to provide a method for measuring the dynamic characteristics (frequency characteristics) of the capacitance C and the series resistance R except for the withstand voltage V. .

The capacitance C of the capacitor and the equivalent series resistance ESR
Is measured using an impedance meter. It is also called a vector impedance meter, LCR meter, or impedance analyzer. In each case, the C and ESR of the capacitor can be obtained as functions of the frequency f based on the same principle.

In this method, an AC oscillator is built in, an AC voltage V is applied to a target capacitor, and the relationship between the current I and the current is examined. The impedance is determined from the current-voltage ratio and the phase delay δ of the current. The principle is the same whether the object is the coil L, the resistor R, or the capacitor C. The current can be determined from the voltage of a small resistor connected in series with the capacitor.
The voltage can be determined with a digital oscilloscope. It is called the AC impedance method. Voltage _{V = V} p si
A Enuomegati, current is assumed to be _{I = I p sin (Ωt +} δ). (V _p and I _p are amplitudes.) Δ is a phase lag. The absolute value | Z | of the impedance is V _p / I _p
It is. If the capacitor has no DC resistance, the phase delay δ is just 90 degrees, but the phase delay δ is 9 because of the DC resistance R.
It is not 0 degrees. Since Z = R + 1 / jωC,

V _p / I _p = {R ² + (1 / ωC) ² } ^1/2 (1) tan δ = 1 / ωCR (2)

The method for obtaining the capacitance C and the series resistance R in this manner is called an AC impedance method. It is a method that adheres to the principles. This is a commonly used method, and its implementation in the device is commercially available as an impedance meter. Since the frequency f is variable, the frequency characteristics of the capacitance and the resistance can be understood. However, there is a simplified type in which the frequency is fixed at 1 kHz and C, L, and R at 1 kHz are obtained. If the element has no frequency dependence, it is often sufficient. However, in general, the AC impedance method can be measured by changing the frequency.

In this apparatus, the capacitance can be easily measured with a pF level capacitor and a μF level capacitor. However, it is difficult to measure with 1F. The CR product is a time constant and is a good measure of time. This can be said to be the time required to charge and discharge C through the resistor R. Since C is large, the time constant becomes extremely long, and the measurement takes time. Even when the voltage is low, a large amount of charge is required for charging, so that a large current needs to flow.
Since the phase delay δ is close to 90 degrees, tan δ is close to divergence, and it is necessary to obtain the current-voltage phase difference δ for many cycles and perform an averaging operation for accurate measurement. Since one cycle is long, it is necessary to repeat the measurement many times, so that it takes more time.

Therefore, it is difficult to measure with a small meter which is widely used at present. In principle, it would be possible to measure a 1F capacitor by increasing the size of the device, increasing the current that can be passed, and taking time. However, such devices are large and expensive and cannot be easily purchased.

If used as a capacitor for supplementing the battery of an electric vehicle, it is desirable that the vehicle owner can easily purchase and own the battery. Where to put equipment that is as bulky as a car, even if you can buy it? At 100F, it is no longer practical to make a measuring device according to the concept of the AC impedance method.

It is not an AC impedance method but a large capacity
A method for statically measuring the capacitance of an electric double layer capacitor
There are several. One is the constant current discharge method shown in FIG.
That. The capacitor is set to 5.
Charge up to 5V. Maintain the state of charge for 30 minutes. Power supply
From the capacitor and in the case of 1F through a resistive load
Discharge at 1 mA. 3.0 V (V₁)
T is the time when it falls ₁, 2.5V (V₂Dropped to
Time of T₂And The capacity C is

[0027]

(Equation 1)

[0028] Discharge by 1 mA
The time (T ₂ −T ₁ ) during which the voltage decreases by 00 mV is measured. Since the capacitor is at 1F, the measurement takes 500 seconds. Since the constant current I is increased in proportion to the capacitor, the time is always about 500 seconds. This is only one measurement, and it takes time to repeat the measurement. Moreover, this measuring method only measures DC characteristics. Naturally the capacity C
Is not known, and the ESR is not known. The other is a so-called constant resistance charging method shown in FIG. 0.632 constant resistance that time seek time τ taken to charge up times R _c of shorting the capacitor was left for 30 minutes or longer to charge the capacitor through a constant resistance R _c and the electric potential to the 0 power supply voltage Is the capacitor capacity C.

[0029]

(Equation 2)

[0030] _{R c} is a low resistance of about 100Ω~2000Ω. This can measure a large capacity, but it is static (f = 0) and the AC characteristics are unknown. At present, the capacity of an electric double layer capacitor is determined by a constant current discharging method and a constant resistance charging method. In each case, only the static characteristics are known. I do not understand the AC characteristics at all. ESR is also unknown. Further, the value of C strongly depends on the power supply voltage, V ₁ (3 V), V ₂ (2.5 V), etc., and the value of C changes depending on conditions.
Poor reliability, time consuming, and unclear frequency characteristics. There is a disadvantage that ESR is not known. In either case, the ESR is determined by applying a constant current of 1 mA at 1 kHz to obtain the voltage across the capacitor, and multiplying the voltage by 100 times the ESR
I have decided. In each case, C is measured by direct current (f = 0) and E
SR is measured at 1 kHz. The ESR greatly fluctuates at a low frequency, and C and ESR measurement conditions are different.

One AC measurement method and two DC measurement methods have been introduced for large-capacity capacitors. Each has its own drawbacks and is not a good solution.

It is a first object of the present invention to provide an apparatus which can measure the C and ESR of a large-capacity capacitor with a smaller size, a lower cost, a shorter time and a higher accuracy. It is a second object of the present invention to provide a method capable of measuring C and ESR of a large-capacity capacitor with high accuracy without taking time. It is a third object of the present invention to provide a method capable of measuring the frequency characteristic of C of a large-capacity capacitor. 10F-1
It is a fourth object of the present invention to provide a method capable of measuring C accurately even at 00F. It is a fifth object of the present invention to provide a method capable of measuring the frequency characteristic of the equivalent series resistance ESR of a large-capacity capacitor.

[0033]

Measurement method [SUMMARY OF invention, capacitor and resistor connecting the serial body into bipolar power square wave constant current positive and negative ± I _c from Bipolar Power (repetition frequency f) is allowed to capacitors generate The rise (charge) and fall (discharge) voltages are observed, and the capacitance C of the capacitor is calculated from the rise (charge) fall (discharge) gradient.
This is a method in which a discontinuous voltage drop rise and fall is divided by a constant current value to obtain an equivalent series resistance ESR.

Further, the frequency characteristics of C and ESR can be known by repeatedly measuring the rectangular wave frequency f. The capacitance C and the resistance ESR can be measured even when an arbitrary bias voltage is applied to the capacitor. Another is that we can know the changes over time. Since electric double layer capacitors change over time,
This needs to be clearly understood.

The use of a bipolar power supply is a distinctive feature of the present invention. The term bipolar power supply is the name of the power supply manufacturer and not a commonly used name. It is a special power source not listed in the electrical engineering dictionary or handbook. It is bipolar because the current flow is bidirectional. Don't be confused because it is completely different from a transistor bipolar.

An ordinary DC power supply has a positive relationship between the voltage V and the current I. In a coordinate system in which voltage V is plotted on the horizontal axis and current I is plotted on the vertical axis (FIG. 1), the DC power supply is
There is a voltage-current relationship in the quadrant. Normally, since a resistive load is connected, the direction of the voltage and the direction of the current are the same, and the voltage and current are determined at one point by the resistance of the load. That is, it is limited to only one point in the first quadrant in the voltage / current coordinate system.

In the case of an AC power supply, a straight line passing through the first quadrant, the origin, and the third quadrant indicates the voltage-current relationship. The slope of the straight line is the reciprocal of the resistance value. The voltage and current values reciprocate on this straight line. Even in an AC power supply, the relationship between voltage and current is fixed, and only a part of the first and third quadrants moves.

However, a bipolar power supply can flow not only a positive current but also a negative current. Even though the voltage is positive, a negative current can flow. That means that it is absorbing energy at that time. At that time, it means that energy is being actively taken from the load, not the power supply. Whether this is possible depends, of course, on the load.

In the fourth quadrant of FIG. 1, the voltage is positive but the current is negative. In the second quadrant, the current is positive and the voltage is negative. In the second and fourth quadrants, energy is absorbed. A bipolar power supply is a special power supply capable of providing a current-voltage relationship as in the second quadrant or the fourth quadrant. Moreover, a constant current can flow. This is not because the load has negative resistance, but because the power supply has a mechanism. However, when the resistance is a load, such a thing is impossible. It is impossible to squeeze energy from the resistor R. Power consumption in the resistance can be written as RI ^2, which is always positive. There is no formula for energy storage with respect to resistance. On the other hand the energy storage capacitor can be written as 0.5CV ^2. The power consumption is CV (dV / dt), which alternates between positive and negative and has an average value of zero. Therefore, when the capacity is a load, the power can absorb energy from the load.

In FIG. 1, if the load is a capacitor, the current and voltage values can draw a circular loop even with a normal AC power supply. Since I = C (dV / dt), if V = sinωt, then I = Cωcosωt.
If Cω = 1 is normalized for simplicity, the locus of the current / voltage becomes a circle such as o, la, mu, u, o. Even in that case, the AC power supply absorbs energy in the second and fourth quadrants.

Unlike the AC power supply, the bipolar power supply allows a negative current to flow even if the voltage is positive. In FIG. 1, a current-voltage relationship that draws a locus of positive constant current / positive / negative voltage such as "no", "mi" or "ku" can be given, and a negative constant current / positive / negative voltage such as "q", "shi" or "ya" can be given. It is also possible to draw a simple trajectory. Such a breakthrough is something that cannot be done with conventional ordinary power supplies.

What voltage,
Whether to apply the current to the load can be freely specified by, for example, a function generator.
In the embodiment of the present invention, a current square wave is applied to the capacitor by the function generator. Since the function generator has flexibility and generality, the rectangular wave current I and the frequency f can be freely changed. Function generator function is an IC with a dedicated circuit
And circuit devices.

If the capacitor is a load, it is not a resistive load, so that a current-voltage relationship as shown in the second and fourth quadrants is allowed. The present invention uses such a bipolar power supply as shown in FIG.
A positive / negative rectangular current wave, such as shown in (1), is applied to a load capacitor. The current is constant, not the voltage. Note that it is a constant current. In addition, it is a constant current whose polarity is reversed.

Let T be the period. fT = 1. A rectangular wave of a constant current, in the first half of the period T + _{I c,} in the second half of the period T is -I _c. When m is an arbitrary integer, (a) mT ≦ t <(m + 0.5) T I = + I _c (5) (b) (m + 0.5) T ≦ t <(m + 1) T I = −I _c (6 )

Switching between positive and negative is performed instantaneously. Don't take the time to switch. The switching points themselves also provide important measurements, as described below. Naturally, when the current is integrated in one cycle, it is zero.

[0046]

(Equation 3)

(7) is essential, but (5) and (6) are not. The integral of one cycle of the current is 0 and may be a rectangular wave, and may not be a half cycle as in (5) and (6). The rectangular wave is used to charge the capacitor with a constant current and discharge the capacitor with a constant current. The time for flowing each as + I _c1 and −I _c2 is
I _c2 : It may be I _c1 . In that case, the current values dividing the charge straight line and the discharge straight line differ as _Ic1 and _Ic2 . That's it.

The change in the voltage V is examined by passing this rectangular wave current through the capacitor. If the capacitor has no series resistance, it should be a triangular wave with no discontinuity, such as e, f, g, h, i in FIG. 2 (2). Voltage is obtained by integrating the current, current ± I _c So voltage is a triangular wave.

However, in practice, since there is an equivalent series resistance ESR, the current flowing through the resistance at the time of switching the current is inverted, so that a double voltage drop occurs. Therefore, the voltage waveform as shown in FIG. 2 (2), W, C, Y, T, R, S, T, N ...
become. There is a voltage discontinuity during switching. Actually, there is a short transition area at the time of switching, and {1-exp
(−t / CR)}. Since the change in this part includes CR (R is ESR), it is possible to calculate ESR therefrom, but the present invention does not take it. The transition region is too short to measure the voltage correctly. Therefore, the present invention obtains C and ESR from the linear portion by focusing on only the linear portion excluding the short transition region.

The present invention tracks the capacitor voltage change for at least one period (Luwayakayota). Preferably, a voltage change for two cycles (Luwayakayoteretsutsune) is obtained. When charging (Waka,
Since the current is positive (+ I _c ), the voltage V increases.
The rate of increase is inversely proportional to capacity. Since Q = CV, both sides of this are differentiated with time,

(DV / dt) = (I / C) (8) Because it is I = + _{I c} in the half cycle of the charge (a),

(During charging) (dV / dt) = (+ I _c / C) (9) Since I = −I _c in the half cycle (b) of discharging, (during discharging) (dV / dt) = _{(-I c / C) (10} )

Is as follows. That charging (Waka, Lesotho) is the gradient of the capacitor voltage of a + I c _{/ C.} This is a constant. That is, the voltage change is a straight line. Since a rectangular wave current is applied, the voltage gradient becomes a straight line. It gives a rectangular wave to make it straight. The capacity C is obtained by dividing I _c by the gradient.

The same can be said for the voltage gradient at the time of discharge (spot, fox). Gradient during discharge is _-I c / C. The voltage change is a straight line. The voltage gradient is constant and -I
_The capacity C is obtained by dividing _c by the gradient. The gradient of discharge and the gradient of charge should be about the same. Since there is discharge and charge in one cycle, two data can be obtained. 2
Taking the period, four data are obtained. These four
Are averaged to obtain the value of C.

A positive / negative rectangular wave is given by the function generator, but charge / discharge is instantaneously switched at the time of positive / negative conversion. The electric double layer capacitor includes an equivalent series resistance. Therefore, the voltage changes discontinuously in switching between charging and discharging and switching between discharging and charging. In FIG. 2 (2), it is a discontinuous point such as luwa, cayo, sauce and sotsu. This is a discontinuity resulting from a sudden reversal of the current direction. Equivalent series resistance R, the current of the charging + _{I c} is generated a voltage of the flow + RI _c. Current -I _c Conversely generates a voltage of -RI _c flows through during discharging. So Luwa, Cayo, Sauce, and Sotsu should all be equal to 2RI _c . There are four discontinuities in two cycles, but this voltage drop,
Increase since 2RI _c, can be calculated series resistance ESR if we voltage discontinuity at 2I _c.

That is, in the capacitor voltage change shown in FIG. 2B, the capacitance C is obtained from the gradient of the continuous portion, and the ESR is obtained from the discontinuous jump. This is the method of the present invention. Measurement can be performed in at least one cycle time. Highly accurate measurement can be performed in two cycles.

When such a measurement can be performed at a certain frequency f, the same C and ESR measurements are repeated while changing the frequency.
Measurement values of C and ESR for an arbitrary frequency f can be obtained. An important frequency band is a very low frequency region from DC to 1 Hz. Since the measurement is performed using a square wave, it takes too much time at a very low frequency. Therefore, a practical measurement range is 0.01 Hz to 1 Hz. However, in the examples of the present invention, the measurement results from 0.01 Hz to 1000 Hz are shown in detail. Actually, such detailed data is not necessary, and 0.01 Hz, 0.1 Hz, 1H
In many cases, it is sufficient if C and ESR at about four frequencies of 10 Hz can be measured.

In this way, the frequency characteristics of C and ESR can be known, which is completely different from the large-capacity direct current (constant current, constant resistance) measuring method described above. The AC impedance method requires a large device and takes a long time. However, the present invention can obtain necessary data by sampling in about one to two cycles, so that the time may be short. For example, even at an extremely low frequency of 0.01 Hz, one cycle is 100 seconds and two cycles are 200 seconds. Therefore, it is sufficient to obtain the data in a minimum of 100 seconds, and even in case of carefulness, 200 seconds is sufficient.

[0059]

FIG. 3 is a diagram showing a schematic configuration of a measuring apparatus according to the present invention. A terminal of function generator 1 is connected to bipolar power supply 2. The bipolar power supply 2 can generate a constant current rectangular wave as described above. What kind of current waveform should be generated is defined by the function generator. Arbitrary current waveforms and voltage waveforms can be generated by the function generator. Here, it is assumed that generates (offset) + (rectangular wave of ± I _c).

The capacitor 3 (electric double layer capacitor) to be measured and the current detecting resistor 4 (r) are connected in series. The + terminal of the bipolar power supply 2 is connected to one end of the capacitor 3 for measurement. The negative terminal of the bipolar power supply is connected to the current detection resistor 4. The current i flows from the + terminal of the bipolar power supply 2 to the capacitor 3 and the resistor 4 and returns to the − terminal.

The digital oscilloscope 5 has a plurality of channels. For example, it has 2 to 10 channels. Two channels can be used for this measurement. CH1 is connected to both ends of the capacitor. CH1 can detect the capacitor voltage. Since CH2 detects the voltage ri at the resistor, it can be divided by r to find the current i flowing through the capacitor. That is, the digital oscilloscope 5 monitors the voltage and current of the capacitor 3.

The current is a rectangular wave alternating between positive and negative as shown in FIG. Such a waveform is generated in the bipolar power supply by the function generator. Bipolar power supplies can also provide an offset V _of . Since the offset does not affect the resistance, it is entirely applied to the capacitor. The current and voltage of the capacitor are

I = + I _c (mT ≦ t (m + 0.5) T) (11) −I _c ((m + 0.5) T ≦ t <(m + 1) T) (12)

[0064]

(Equation 4)It is. In the voltage expression, ir is the voltage of the current detection resistor.
Appears at pressure CH2. Otherwise, (V_of + ∫i
dt + iR) is the voltage of the capacitor. R is ESR
It is. iR changes at the time of current reversal at the voltage of ESR.
± 2 RI _cIt is. V_ofIs the offset,
Are specified by the function generator
Source occurs.

In FIG. 1, a bipolar power supply can apply a current and a voltage which change like a nocturnal voltage to a capacitor. If there is no offset and the current is a rectangular wave with the maximum amplitude, a change such as Nomike Keshiya Nomike Keshiya is made. Nomic is a line segment drawn within the pressure resistance.

When a bias is applied, a locus that is asymmetric with respect to voltage can be designated. For example, when a bias voltage such as d and t is applied, the locus of the current and voltage becomes as follows.

[0067]

Example 1 Example 1 (1F, 2.5 V, f = 1 Hz,
Example 1 is a measurement of C and ESR of an electric double layer capacitor of 1F at a frequency of 1 Hz. The used function generator, bipolar power supply, digital oscilloscope, capacitors, etc. are shown below.

Function generator: FG120 (706012-M) manufactured by Yokogawa Electric Corporation Bipolar power supply: Polarized unit PS-07 manufactured by Toho Giken
(Potentiometer / galvanostat) Digital oscilloscope: Digital scope DL708E (701820, with 3.5-inch FDD) made by Yokogawa Electric High-resolution insulation module: High-resolution insulation module 70
1853 (voltage sensitivity 5 mV / div to 20 V / div,
A / D conversion resolution 16 bits, input impedance 1M
Ω) Current detection resistor: 1 Ω, 2.5 W, accuracy 0.1% Capacitor to be measured: Electric double layer capacitor (2.5 V,
1F)

As shown in FIG. 3, a resistor (1Ω, 2.5 W, accuracy 0.1%) was connected to the negative terminal of the capacitor to be measured. The + terminal (+ current line and + voltage line) of the bipolar power supply was connected to the + terminal of the capacitor. The-terminal (-current line and-voltage line) of the bipolar power supply was connected to the other end of the resistor.

The output of the function generator and the input terminal of the bipolar power supply were connected by a coaxial cable. Two channels of high-resolution insulation module were attached to a digital oscilloscope. As shown in FIG.
1 and a capacitor, and CH2 and a resistor.

Each device was started up and set as follows. ○ Function generator FG120 Output: off Output mode: cont Output waveform: Rectangular wave Frequency: 1 Hz Output voltage: 2 V (± 1 V) Offset: 0 mV

This sets the condition of current (not voltage) generated by the function generator by the bipolar power supply. The generated current is a square wave with a frequency of 1H
z is defined. The output voltage from the function generator is ± 1V. This value determines the value of the charge / discharge current generated in relation to the polarization unit. Since the voltage is +1 V in a half cycle, the charging current is 1A. Since the remaining half cycle is -1V, the discharge current is 1A.

When the offset value of the function generator is set to be positive, the charging current is increased and the discharging current is reduced. Thus, a bias voltage can be applied to the capacitor. However, the relationship between the offset value and the bias voltage is not known in advance. Here, the case where there is no bias voltage (there is no offset) is shown first.
It is necessary to check the performance by applying a bias, and this is done. However, since the relationship between the two is not known, the relationship between the offset and the bias is obtained by trial and error.

Polarizing unit / bipolar power supply (potentiometer / galvanostat) PS-07 Function: off Input: on Current range: 1A Initial voltage (INIT): 0 mV

When the initial voltage is set to be positive, similarly to the offset value of the function generator, the charging current increases and the discharging current decreases. Thus, a bias voltage can be applied to the capacitor. However, the relationship between the bipolar power supply initial voltage and the bias voltage is not known in advance. Here, the case where there is no bias voltage (the initial voltage is 0 V) is shown first. It is necessary to check the performance by applying a bias and actually do it, but since the relationship between the two is not known, the relationship between the initial voltage of the bipolar power supply and the bias is obtained by trial and error.

Digital Scope DL708E Sampling: off ACQ Record Length: 1k ACQ Mode: Box Average Channels Display CH1, CH2: on, CH3 to CH8: off Vertical V / Divich V1: DIVCH2 , CH2: 0 Div Coupling: DC Horizontal TIME / DIV: 200 ms / div Trigger Simple: on Mode: Auto Source: CH2 Level: 0 mV Slope: 0% V0 Auto Naming: on, file nam e: test Media: FD

The amount of sampling data is determined by the Re
Determined by cord Length. This time is 100
0 (1k) data can be obtained by one measurement.
Since voltage data is taken for two cycles, 250 data (which can be considered as points representing time and voltage) correspond to charging and discharging. First charge 250 points, discharge 250 points,
The second charge is 250 points and the discharge is 250 points. Since data collection is completed in two cycles, there is an advantage that the measurement time is short.

Here, the frequency f of the applied current change is 1
In Hz, the period T is 1 second (T = 1 / f). Two seconds because of two cycles. Hori so that the measurement time is 2 seconds
200ms / div of zone TIME / DIV
Is set to The div is the interval between the vertical and horizontal grid sections of the oscilloscope screen. 10 seconds is 2 seconds.
CH1 detects the capacitor voltage. The capacitor voltage is in the range of -1 to + 3V. To be able to sense this,
Vertical V / DIV was set to 500 mV / div. Position (the height of the origin) was -2 div. In other words, the setting allows the user to see -1V to + 3V.

Since CH 2 is connected to both ends of the current detection resistor (1Ω) 4, it checks the current flowing through the capacitor. Since the current is -1A to + 1A, it is -1V to + 1V in terms of voltage. VerticalV / DIV is set to 500 mV so that this can be detected.
/ Div and Position were set to 0 div. That is, the setting allows the user to see the upper and lower 2V from 0V. Each value of Trigger was set so that the measurement was started when switching from discharging to charging.

The above is the initial setting. Next, the measurement operation will be described. Digital oscilloscope Sampl
ing turned on. Next, the output of the function generator was turned on. Function of polarization unit
n rotary switch with a galvanostat (G-STA).
T, current control). That is, the mode was set such that a constant current of ± 1 A flows. Digital oscilloscope screen (C
H1, CH2).

CH1 indicates the capacitor voltage, and the voltage sharply rises immediately after the start of charging. Then rise gently. Immediately after the start of discharge, the voltage drops sharply, and then the voltage drops slowly. CH2 indicates a capacitor current.
It becomes + 1A (+ 1V) at the time of charge, and -1A (-) at the time of discharge.
1V). That is, the current takes only two values of ± 1A.

As the offset of the function generator is increased, the waveform of CH1 (capacitor voltage) increases. The offset of the function generator was adjusted so that the average voltage of this waveform was 2V.

When the trigger mode of the digital oscilloscope is changed to normal and waits for about 3 seconds, a waveform of two cycles of charging / discharging appears in the resistance voltage (capacitor current).
This data was recorded on the floppy disk FD. The FD record was imported to a personal computer. The extension of the obtained file was changed to “csv”. I read it with spreadsheet software. This file consists of about 1000 data in total. The first 250 points are the first charge, the next 250 points
The points are data of the first discharge, the next 250 points are data of the second charge, and the next 250 points are data of the second discharge.

Since charging and discharging are performed with a constant current, the capacitor voltage rises almost linearly and falls linearly. Since the gradient of the rising line and the falling line is proportional to 1 / C, the capacitance C is obtained from the gradient. The ascending and descending lines are strictly out of line. Therefore, the ascending and descending lines are approximated by straight lines. The least squares method is used to approximate with a straight line. The straight line obtained by the least square approximation is called a “regression line”.

The data of CH1 (capacitor voltage) from the eleventh point to the 240th point was plotted against time (FIG. 4), and a regression line was obtained by linear approximation. It is the first charging regression line. Similarly, data of capacitor voltages from the 261st point to the 490th point were plotted against time to obtain a regression line. It is a 1st discharge regression line.

A second charge regression line was determined from the capacitor voltage data at the 511-th to 740-th points. 761th ~
A second discharge regression line was determined from the capacitor voltage data at the 990th point. As a result, four regression lines are obtained. Two regression lines for discharging and two for charging.

As described above, the capacity is C and the charging / discharging current is I
(I is positive for charging and I is negative for discharging), and if the slope of the regression line is dV / dt, C = I /
(DV / dt).

At the time of charging, the current is +1 A, so that the charging linear slope gives the reciprocal of C. It was first and the linear up to 240 goal 11 goal during charging approximate least squares method charging regression line _{V a = + 0.8428t + 1.9310 (} 14). The reason for excluding the first 10 points in the data for each 250 points is that they deviate significantly from the straight line at the time of transition. It is better to remove the first 10 points because the variation in data is reduced. However, there are cases where it is not necessary to remove the first 10 points. The proportionality constant of t gives the capacitance C, and the zero-order term independent of t gives the ESR. In this example, the capacitance C is 0.842
Since it is the reciprocal of 8, it is calculated as 1.186F.

Since the current is -1 A at the time of discharging, the capacity C is obtained by adding a minus sign to the reciprocal of the slope of the discharge line.
When the discharge straight lines from the 261st point to the 490th point of the first discharge are approximated by the least squares,

The discharge regression line was V _b = −0.8475 t + 2.5218 (15). As in the previous example, the proportionality constant of t gives the capacitance C, and the 0th order gives the ESR. In this example, the reciprocal of 0.8475,
C = 1.17F is calculated. The capacity C at the time of charging and at the time of discharging have almost the same value. Actually, the capacity C is determined by an average value of reciprocals of four regression lines. C = 1.182F (16) as an average value.

The calculation of the series resistance ESR is more complicated than the calculation of the capacitance. This will be described with reference to FIG. The voltage drop when switching from charging to discharging first is 2I
_cR . Since the direction of the current _{I c} flowing through the R is inverted voltage drop is so 2I _c R. Since the drop speed itself at this time is exp (-t / CR), it may be possible to obtain R from the voltage drop in the transition region. To do so, from 251 to 260
A remarkable change such as the 501st point to the 510th point must be measured.

However, the present invention does not take such a step.
Otherwise, the ESR is obtained from the relationship that the total amount of dip = 2I _c R. The transition region is excluded from the judgment. It suffices if the drop amount from the charge regression line to the discharge regression line can be accurately obtained. It switched to the -I _c current from + _{I c} at 250 goal. The distance between the points is 2 ms, which is quite long. It is believed that the voltage dip will occur between 250 and 251 points. So 250 points and 2
Find the ESR at the 51st point and take the average value.

The charge regression linear voltage at 250 points (t = 0.498 s) was V _a = 2.3507 V, and the discharge regression linear voltage was V _b = 2.0997 V. The difference (2R
Corresponding to I _c) is ΔV = 0.251V. This is 2A
Dividing by (2I _c ), ESR = 0.1255Ω is calculated.

At 251 points (t = 0.500 s), the charge regression linear voltage was V _a = 2.3524 V, and the discharge regression linear voltage was V _b = 2.0980 V. The difference (2R
Corresponding to I _c) is ΔV = 0.2544V. This is 2
Dividing by A (2I _c ), ESR = 0.1272Ω is calculated. Switching from charge to discharge is 250 points and 251
Do this calculation because it will be between the points.

This time, the switching point 500 from discharge to charge is
Do the opposite calculation for points -501. The discharge regression line has already been obtained from the data at points 261 to 490, and a new charge regression line is obtained (points 511 to 740). 5
At the point 00 (t = 0.998 s), the voltage difference between the discharge regression straight line voltage and the charge regression straight line was calculated and divided by 2 ESR = 0.12
It was calculated to be 19Ω. Next 501 points (t = 1.0
00s), the voltage difference between the discharge regression line voltage and the charge regression line was calculated and divided by 2 to calculate ESR = 0.1236Ω.

The average value of the ESR at these four points was taken as ESR = 0.125Ω (17). Thus, the capacitance C (1.182F) at f = 1 Hz of the electric double layer capacitor and the series resistance ESR (0.12
5 Ω).

Example 2 (1F, 2.5 V, bias,
(Frequency characteristics, change with time)] The polarization unit was changed to a bipolar power supply BWA25-1 manufactured by Takasago Seisakusho, which was capable of controlling a constant current using an operational amplifier as shown in FIG. In order to automatically perform measurement after the set time has passed, CH8 of the digital oscilloscope DL708E is used.
A clock for generating a 1 V signal after a lapse of a set time was created using a personal computer, and connected to CH8.

The electric double layer capacitor to be measured has a withstand voltage of 2.5 V and 1F. This electric double layer capacitor was placed in a thermostat at 60 ° C. This is for examining the temperature characteristics. It can be housed in a thermostat at an arbitrary temperature to check temperature characteristics in a desired temperature range. The same function generator and digital oscilloscope as in Example 1 were used. However, the setting conditions are different from the first embodiment. Only the differences are listed.

○ Function generator FG120 Output voltage: 1.73V Offset: 60mV

It has been described that the relationship between the DC bias voltage of the capacitor and the function generator offset voltage is not known in advance. Therefore, the average voltage (capacitor DC bias) is 2 V using a capacitor different from the target capacitor.
An offset of 60 mV is provided so as to achieve the following. Example 2 differs from Example 1 (offset is 0) in that an offset is given. Since the withstand voltage of the capacitor is 2.5 V, an offset of 2 V, which is 80% thereof, is given.
Since a voltage is applied above and below the offset, the amplitude is 0.5 V at the maximum. Repeating charging and discharging by applying a bias of 80% is in accordance with the conditions of actual use. 60 mV offset of function generator to give 2 V offset with capacitor
That is. The offset of the function generator and the offset of the capacitor must not be confused.

○ Bipolar power supply BWA25-1 Power supply: off ○ Digital scope DL708E Each channel Display CH8: on Vertical CH8: 500mV / div Position CH8: 0div Coupling: DC Triggering: Enhanced Access Enhancement: DC Triggering : 0 mV Slope: Rise MISC Action on trigger Mode: on Set action: save to HD Sequence: cont File Media: HD

Digital oscilloscope Sampli
ng was turned on, and the output of the function generator was turned on. The clock was operated simultaneously with turning on the power switch of the bipolar power supply. This clock generates a pulse after a lapse of a set time.

After confirming that the charging and discharging were normally performed by the digital oscilloscope, the trigger of the digital oscilloscope was set to normal. When a predetermined time comes, a clock generates a signal, and the digital oscilloscope takes in data. The data is sent to the oscilloscope's HD and automatically stored on the hard disk.

After all the measurements were completed, these data were copied to a floppy disk and taken into a personal computer. Change the extension of the obtained file to "csv",
The data was read by spreadsheet software and subjected to the same data processing as in Example 1. Then, the capacitance C of the electric double layer capacitor and the equivalent series resistance ESR were calculated.

The charging / discharging frequency is 0.01 Hz to 1000
The same operation was performed while changing between Hz. Further, changes in C and ESR over time were also examined as a function of frequency. The frequencies at that time are 0.1 Hz, 1 Hz, and 10 Hz. Then, not only a change in the frequency f but also a change with time can be known.

FIG. 5 is a graph showing the measurement results obtained by changing the frequency of the capacitance C and the resistance ESR. The horizontal axis is 0.01 Hz to 1000 Hz in frequency f. The vertical axis represents the capacitance C (black diamond point) and the resistance ESR (square point). The capacity C is 0.
It is 1.33F at 01Hz and decreases with frequency.
At 0.1 Hz, it drops to 1.22F. 0.8 at 1 Hz
8F, which drops to 0.1F at 100 Hz.
Even if the capacitor has a capacity of 1F, it is a DC measurement, and it can be seen that the capacity actually decreases significantly with frequency.

The ESR is as large as 1.23Ω at 0.01 Hz, but sharply decreases as the frequency increases, and becomes 0.29 at 1 Hz.
Drops to Ω. At 1000 Hz, it is about 0.09Ω.
FIG. 6 shows 0.1 Hz (black diamond point), 1 Hz (square point), 1 Hz.
Capacitance C when frequency is measured at 0 Hz (triangular point)
Shows the change over time. The horizontal axis is the charge / discharge time t (second), and the vertical axis is the capacity C (F). Since the measurement was repeated under the same conditions, the range in which the value exists was represented by a bar, and the average value was indicated by a point.
Similarly, the electric double layer capacitor referred to as 1F has a significantly different value depending on the measurement frequency and the measurement time.

At 0.1 Hz, 1.45 F in 100 seconds
However, it drops to 1.37F after 1000 seconds. 10
After 0000 seconds, it is 1.27F. The capacity decreases over time. 1.36H at 200Hz for 1Hz
Although it is z, it is 1.32 Hz after 1000 seconds. 10
1.19 Hz after 0000 seconds. Similarly, the capacity decreases over time.

At 10 Hz, it is 0.93F after 100 seconds. It is 0.89F after 10,000 seconds. The capacity decreases over time, but more slowly. The higher the frequency, the smaller the capacity. It can be clearly seen that the capacity decreases with time at any frequency. That is, according to the present invention, the frequency characteristics and the aging characteristics can be clearly understood. Although the same electric double layer capacitor is used, the capacitance measurement values differ in this way. An electric double layer capacitor is a capacitor whose capacity is easily changed. Generally, the higher the frequency, the smaller the capacity. The capacity decreases over time. Depending on the purpose of use, the life may be shortened by a certain reduction in capacity. In this case, the life of the capacitor can be accurately evaluated by this measurement. Such a double time change is very hard to understand by the conventional DC measurement method.

FIG. 7 shows the series resistance E for the same capacitor.
The measurement results of SR are shown. The top is a graph of ESR measurement at 0.1 Hz, the middle is the ESR measurement result at 1 Hz,
The lower part shows the result of ESR measurement at 10 Hz. The lower the frequency, the greater the ESR. However, it does not seem to decrease with time, and the change with time is not uniform. This can be understood for the first time by the present invention.

Example 3 (Capacity 100F, withstand voltage 2.5
V)] Next, a measurement example of an electric double layer capacitor of 100F will be described. As already described, it can be said that it is almost impossible to measure the capacitance of such a large-capacity capacitor by the conventional AC impedance method. It can be measured by the direct current method (constant current method, constant resistance method), but it takes time and the frequency characteristics are unknown. However, the present invention can easily measure the C, ESR, frequency characteristics, and the like of such a large-capacity capacitor.

A commercially available electric double layer capacitor (2.5 V, 100 F) was measured by the method of the present invention. The same device as in Example 1 was used. A square wave of a constant current of ± I _c to discharge from flowing charged in the capacitor, is switched to the charge from the discharge direction of the instantaneous current measuring two periods a voltage change of the capacitor. The charge regression line and the discharge regression line are obtained by linear approximation except for the data in the transition region, and the capacitance C is obtained from the slope, and the ESR is obtained from the voltage jump at the time of switching.
Ask for.

FIG. 8 shows that the electric double layer capacitor sample of 100F is charged (0 to 0.5 s) and discharged (0.5 s to 1 s).
s) A graph showing data of the capacitor voltage every 2 ms by dots. The charge regression line is obtained by linearly approximating the charge voltage. The discharge regression line is obtained by linearly approximating the change in the discharge voltage. Square wave current is ± 1 A, frequency is f =
1 Hz. Since the withstand voltage is a 2.5 V capacitor, 1.63 V was applied as a bias voltage. Under these conditions, the capacitance and resistance of the capacitor were measured.

The capacity was C = 62.55F. The equivalent series resistance was ESR = 0.0093Ω. Nominally 10
Although it says OF, it is a DC capacity measured under limited conditions. Bias in the measurement method of the present invention
Since it is 62.55F when the frequency is 63V and the frequency is 1Hz, it is only about 60%. An electric double layer capacitor has such a large frequency and bias dependency. Such is not known by the conventional method. This clearly shows that the present invention, in which the frequency characteristics, the aging characteristics, and the bias dependency, are superior, is excellent.

For the same electric double layer capacitor of 100 F and 2.5 V, the frequency f is set to 0.01 Hz to 1000 H under the common conditions of ± 1 A rectangular wave current and bias 2 V.
C and ESR were measured in place of z. FIG. 9 shows the result. 100F at 0.01Hz, 97F at 0.1F, 1
It drops to 69F at Hz. Beyond this, the capacity decreases rapidly and drops to 22F at 10Hz and 3F at 100Hz. It can be seen that the capacitance decreases significantly with frequency.

The ESR is 0.0136Ω at 0.01 Hz,
It is 0.01Ω at 0.5 Hz. 0.00 at 500 Hz
75Ω. ESR also decreases with frequency. However, it turns out that it is not so remarkable as the capacity. It has never been possible to measure the frequency characteristics of C and ESR of a capacitor having a large capacity of 100F. Further, although a bias is applied here, such a thing cannot be imagined conventionally for a 100 F class large capacity capacitor.

[Embodiment 4 (Measurement of C and ESR of Lead Battery)] By the method of the present invention, the capacity C and equivalent series resistance ESR of a battery can be measured. A battery can carry electric charge in one direction by a chemical reaction and continuously generate an electric current. The voltage is determined by the chemical reaction. Rechargeable batteries have a reversible chemical reaction.
When an electric current is applied, a chemical reaction proceeds to change the amount of the substance. This does not mean that charges are accumulated on the electrodes. As mentioned earlier, batteries have a slow response speed and cannot follow fast current-voltage changes in electric vehicles and wind power. So how slow? If a capacitor and a battery are used together, it is desirable that their speeds can be compared specifically.

Fortunately, the method of the present invention can measure C and ESR while changing the frequency while applying a bias. Since the voltage of the battery can be regarded as a bias, the method of the present invention can be used to measure a battery such as C or ESR. FIG. 10 shows the result of the measurement. The horizontal axis is frequency 0.01Hz ~
1000 Hz. The capacity of this lead battery is 0.01Hz
Is 222F, but decreases rapidly with frequency. 1 Hz
Then it drops to 25F. ESR is 0.08 at 0.01 Hz
Ω. At 1 Hz, it is 0.01Ω. Since batteries also have capacity, this must also be taken into account when designing. The concept of impedance also exists in batteries.
If C and R obtained by the method of the present invention are used as in the following equation, Z = R + 1 / (jωC) (18) Expressing the impedance on the complex plane is Cole−Co.
It is called le Plot and is well known as a method for examining characteristics of a battery or the like. Therefore, using this method,
The characteristics of the battery can also be evaluated. The characteristics of the lead storage battery obtained this time have an inclination of 45 degrees in all the measured regions. Therefore, it can be inferred that diffusion control is performed in this frequency region. FIG. 12 shows the Col of this battery.
Shows e-Cole Plot.

[0119]

According to the present invention, the capacitance C and the equivalent series resistance ESR of a large-capacity capacitor exceeding 1F can be measured as a function of the frequency f. Large-capacity capacitors have had little use so far, and therefore have low production volumes. However, in recent years, large-capacity capacitors for storage of electric vehicles and wind power have been expected. This is the use of a capacitor as a supplement to a battery to store energy. Batteries can store large amounts of energy because they store energy through chemical action, but they are difficult to keep up with fast voltage and current changes. Rapid charging and discharging significantly reduce battery life. There is a strong expectation for large-capacity capacitors to absorb fast voltage-current fluctuations and complement battery action.

To manufacture a device using a large-capacity capacitor, it is necessary to examine the characteristics of the capacitor in detail. The electric double layer capacitor is a large-capacity capacitor suitable for power storage, but C is not stable and has a strong frequency dependency like an ordinary capacitor, and has an equivalent series resistance ESR.
have. If you try to use such a thing, C
Frequency dependence and accurate ESR measurement. The present invention can meet such a purpose well.

Although a large-capacity capacitor for power storage is often used by applying a bias voltage (average voltage), the present invention can measure C and ESR with a bias of an arbitrary height applied. . That is, the present invention can measure the frequency characteristics and bias characteristics of a large-capacity capacitor.

When a converter is used, the voltage range in which the capacitor is used is an intermediate range between the case where the capacitor is charged to a withstand voltage and the case where the capacitor is charged to about half. The energy intermediate in this range is approximately 80% of the breakdown voltage when converted into voltage. This means that the bias voltage is about 80%.
Although the setting is quite high, the width of the voltage fluctuation does not exceed the withstand voltage,
Moreover, if the capacity of the capacitor is to be used effectively, the capacitor is used under such severe conditions. In such a case, it is necessary to examine not only the static (DC-like) characteristic but also the frequency characteristic and the change with time in detail. Conventional methods are completely useless for such purposes. The present invention can meet such an object for the first time.
In this sense, the present invention is significant.

[Brief description of the drawings]

FIG. 1 is a coordinate system diagram showing a voltage-current relationship of a bipolar power supply used in the present invention.

Graph showing the square wave constant current and voltage waveforms of the capacitor thereby resulting in ± I _c is applied to the capacitor in the present invention; FIG. FIG. 2A shows a current waveform, and FIG. 2B shows a voltage waveform.

FIG. 3 is a schematic configuration diagram of a capacitor measuring device according to the present invention.

FIG. 4 is a diagram showing a part of a voltage waveform for one cycle in which switching from charge to discharge is performed when a rectangular wave current of ± 1 A is supplied to the electric double layer capacitor on the 1F according to the first embodiment. The horizontal axis represents time (seconds), and the vertical axis represents voltage (V).

FIG. 5 shows a ± 1 A rectangular wave current in a 1F electric double layer capacitor with a frequency of 0.01 Hz to
9 is a graph showing the results of measurement and calculation of the capacitance C and the equivalent series resistance ESR according to the method of the present invention at 1000 Hz. The horizontal axis is frequency (Hz), and the vertical axis is capacitance C (F) and ESR
(Ω).

FIG. 6 shows a capacitance C (F) obtained by applying a rectangular wave current of ± 1 A at a frequency of 0.1 Hz, 1 Hz, and 10 Hz to a 1F electric double-layer capacitor in Example 2 to determine a capacitance C (F), and changes with time. The graph which shows the result of having investigated.

FIG. 7 is a graph showing an equivalent series resistance E obtained by applying a rectangular wave current of ± 1 A at a frequency of 0.1 Hz, 1 Hz, and 10 Hz to a 1F electric double layer capacitor according to Example 2;
7 is a graph showing the result of determining SR (Ω) and examining the change over time.

FIG. 8 is a diagram showing voltage point data every 2 ms when a bias voltage of 1.63 V is applied to an electric double layer capacitor of 100 F and a rectangular wave constant current of ± 1 A (frequency 1 Hz) is applied to Example 3; . From the point sequence during charging,
A discharge regression line is obtained from the point sequence at the time of discharge.

FIG. 9 relates to Example 3 and applies a bias of 2 V to an electric double layer capacitor of 100 F to change ± 1 A at different frequencies.
, A capacitance C (F) and an equivalent series resistance ES
9 is a graph showing the result of measuring R (Ω). The horizontal axis represents frequency (Hz), and the vertical axis represents C (F) and ESR (Ω).

FIG. 10 shows Example 4 in which the method of the present invention was applied to a battery.

FIG. 11 is an example of a control circuit of a polarization unit using an operational amplifier.

FIG. 12 is a Cole-Cole Plot diagram of a lead battery.

FIG. 13 shows a constant resistance charging method for measuring the capacitance of a large-capacity electric double layer capacitor according to a conventional example. FIG. 11 (1) is a schematic circuit diagram thereof, and FIG. 5 is a graph showing a change over time in voltage (V).

FIG. 14 shows a constant current discharging method for measuring the capacitance of a large-capacity electric double layer capacitor according to a conventional example. FIG. 12 (1) is a schematic circuit diagram thereof, and FIG. 5 is a graph showing a change over time in voltage (V).

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Function generator 2 Bipolar power supply 3 Capacitor 4 Resistance 5 Digital oscilloscope

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2G028 AA01 BB06 CG01 CG02 CG07 CG20 DH06 DH10 DH11 DH13 EJ06 FK01 LR10 5E082 AB09 BC40 MM35 MM36

Claims (4)

[Claims] To 1. A subject of the capacitor, and ± I square wave constant current charging regression line approximated by a straight line charging waveform and discharging waveform examine more than one period of the voltage data flow in the capacitor of the _c a frequency f A method for measuring a capacitance and an equivalent series resistance of a capacitor, wherein a capacitance is obtained from a gradient of a discharge regression line, and an equivalent series resistance ESR is obtained from a rising amount and a falling amount at a discontinuous point of the regression line. In a state that wherein gave a bias to the target of the capacitor, the charging waveform except transition portion examine one or more periods of voltage data flow in the capacitor square wave constant current of ± I _c a frequency f discharge Approximate the waveform with a straight line, find the capacitance from the gradient of the charge regression line and the discharge regression line, and find the equivalent series resistance ESR from the amount of rise and fall at the discontinuous point of the regression line. How to measure equivalent series resistance. 3. The frequency characteristic of the capacitance C and the equivalent series resistance ESR is obtained by changing the frequency f.
Or the measuring method of the capacitance and equivalent series resistance of the capacitor according to 2. 4. A bipolar power supply capable of generating a positive / negative constant current and allowing a positive / negative constant current to flow through a target capacitor, and a device for generating a rectangular wave constant current including a positive / negative constant current at a frequency f in the bipolar power supply. A device for measuring and storing a voltage change of a capacitor, and a device for approximating a voltage change at the time of charging and a voltage change at the time of discharging by a straight line.