JPH0455273B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION A. Object of the Invention [Field of Industrial Application] The present invention relates to a time width measuring device. More specifically, the present invention relates to a time width measuring device that can accurately measure even so-called fractional hours that are less than the period of a reference clock signal.

[Conventional technology]

Universal counters are widely used as devices for measuring the frequency, period, etc. of signals. In addition, not only counters like this, but also counters such as
Devices such as LSI testers use devices that measure the time width from a certain point to a certain point in a signal to be measured.

BACKGROUND ART In recent years, with the development of the telecommunications field, the frequencies of signals handled have increased, and there has also been a demand for measuring the time width of signals with high precision (high resolution).

Generally, the following principle is employed to measure time width with high precision. Measured time width T _x
A clock signal with a period T ₀ is passed through a gate that is opened at , and the number N of the clocks passing through is counted. And, NT ₀ is the time width. Although this method improves resolution as the clock frequency increases, there is actually a limit to the speed of the circuit elements. That is, this means cannot measure with a resolution greater than the clock period.

Strictly speaking, in the above method, T _x =NT ₀ does not hold, but T _x NT ₀ . This is usually T _x
This is because T ₀ is not divisible and there are small fractional times. This is shown in FIG. In FIG. 4, ΔT ₁ is the fractional time from the rising edge of T _x to the immediately following clock;
ΔT ₂ is the fractional time from the falling edge of T _x to the immediately following clock. and,
If the gate is open only during the falling period of both ΔT ₁ and ΔT ₂ [see H in Figure 4], and the number of clocks in that period is N [FIG. 4 F], the time width T _x
is expressed by equation (1).

T _x = NT ₀ + ΔT ₁ − ΔT ₂ (1) Therefore, if we measure the fractional times ΔT ₁ and ΔT ₂ , we get
It can be seen from equation (1) that the time width T _x can be measured with a resolution greater than the clock period T ₀ .

Therefore, as a conventional technique, it is possible to measure the fractional time ΔT. In other words, it is possible to measure the fractional time ΔT.
The following means that can measure time width with a resolution of T ₀ or higher is known.

<> Time vernier method This is an application of the caliper principle to the time axis, and will be explained using Figure 5. This method uses period
In addition to the main clock at T ₀ , there is a vernier clock with period T ₀ ′ (T ₀ ′>T ₀ ) that occurs at the beginning of fractional time ΔT.
A clock is required. By counting the number of clocks N until the phases of both clocks match, ΔT can be found as ΔT=N(T ₀ '−T ₀ ). The resolution is given by the period difference (T ₀ '−T ₀ ) between the two clocks.

＜＞ Time Expansion (time
(expansion) method will be explained using Fig. 6. This method uses an integrator to expand the fractional time using the charge or voltage stored in the capacitor as an intermediary, and measures it using a clock. FIG. 6 is a diagram showing the case where charges are used as mediators. If the integrating capacitor is charged with a current I ₁ between ΔT and then discharged with a current I ₂ , ΔT _E =I ₁ /I ₂ ·ΔT, and the expansion ratio is given by I ₁ /I ₂ .

[Problem that the invention seeks to solve]

However, the above-mentioned means have the following problems.

<> The time vernier method has the problem that it takes time for the main clock and the vernier clock to match, as shown in FIG. 5, and high-speed repeated measurements or real-time measurements are not possible.

<> The time expansion method is the 6th
As shown in the figure, since an additional ΔT _E time is required to measure the fractional time ΔT, there is a problem that, like the time vernier method, high-speed repeated measurements or real-time measurements cannot be performed.

An object of the present invention is to provide a time width measuring device capable of high-speed repetitive measurements, real-time measurements, and high-resolution measurements.

B ``Structure of the Invention'' [Means for Solving the Problems] In order to solve the above problems, the present invention provides a clock generator that generates a clock signal, a counter that counts signals based on this clock signal, and a clock generator that generates a clock signal and a counter that counts signals based on the clock signal. In a device that measures time width based on the value of a counter, it is synchronized with this clock signal and whose phases are mutual
A two-phase oscillator that outputs two signals s1 and s2 that differ by 90 degrees, and an AD converter that converts the amplitude values of the signals s1 and s2 of the two-phase oscillator into digital values when the time width measurement start signal and end signal are generated. , a control circuit that controls the operating timing of the AD converter and the start and stop of the counter, and the amplitude values (digital values) of the signals s1 and s2, as well as the calculation that calculates the time width by introducing the count signal from the counter. The device is equipped with an arithmetic circuit.

〔Example〕

Hereinafter, the present invention will be explained in detail using the drawings.
FIG. 1 is a diagram showing an embodiment of the present invention.
In the figure, 1 is a two-phase oscillator, which is synchronized with the clock with a phase difference φ (φ = 0), and generates two signals s1, 90° out of phase with each other,
It outputs s2. In the following explanation, the operation of these two signals whose phases are shifted by 90 degrees will be explained using an example in which a sine wave is used as the signal s1 and a cosine wave is used as the signal s2. Note that the periods of the signals s1 and s2 do not need to be the same as the clock period _T0 , and may be an integral multiple or a fraction of _T0 .

3 is a clock generator (hereinafter simply referred to as clock), which generates a clock signal serving as a time reference with a period _T0 . For example, it is 10MHz.

5a to 5d are AD converters, which convert introduced analog signals into digital signals.
Since the time width measurement device according to the present invention aims at high-speed operation, the AD converters 5a to 5d
It is also desirable that the operation be fast. Therefore, in the embodiment, a flash type AD converter was used as the AD converter. However, the AD converters 5a to 5d are not limited to flash type AD converters.

7 is a control circuit, which introduces a start pulse signal (time width measurement start signal) and a stop pulse signal (time width measurement end signal) for measuring time from input terminals p1 and p2, and in synchronization with this, CLK1. It outputs a signal, a CLK2 signal, and a signal for starting and stopping the operation of a counter, which will be described later. Although it has been explained that the time width measurement start signal and end signal are a start pulse signal and a stop pulse signal, the present invention is not limited to this description, and may be regarded as other signals generated in synchronization with these signals.

A counter 9 counts the number of times the introduced signal crosses a certain level.

Reference numeral 11 denotes an arithmetic circuit, which has an arithmetic function to be described later. As the output of this arithmetic circuit 11, the terminal p3
A signal corresponding to the time width of the measurement target is extracted from.

The clock signal from clock 3 is sent to two-phase oscillator 1.
and is added to the control circuit 7. The signal s1 (sin ωt) from the two-phase oscillator 1 is sent to the AD converter 5a,
5c, and the signal s2 (cos ωt) is added to AD
applied to converters 5b, 5d and counter 9.
A start pulse and a stop pulse are applied to the control circuit 7. Further, from the control circuit 7, a signal CLK1 is applied to the AD converters 5a and 5b, a signal CLK2 is applied to the AD converters 5c and 5d, and a signal to start and stop this operation is sent from the control circuit 7 to the counter 9. is added. The outputs of each AD converter 5a to 5d and the output of the counter 9 are applied to an arithmetic circuit 11.

FIG. 2 is a time chart showing the operation of each part of the apparatus shown in FIG.

The operation of the apparatus according to the present invention shown in FIG. 1 and configured as described above will be explained with reference to FIG. 2.

The device shown in FIG. 1 is a device for measuring the time T _x between the start pulse and the stop pulse shown in FIG. The fractional time between the rising edge of the clock and the rising edges of the start and stop pulses is expressed as ΔT ₁ as shown in FIG. Let ΔT be ₂ . At this time, the above-mentioned equation (1) holds true.

When a start pulse is applied to the control circuit 7, the signal CLK1 is sent to the AD converter 5a, in synchronization with this start pulse from the control circuit 7.
Added to 5b.

On the other hand, the two-phase oscillator 1 outputs signals s1 and s2 as shown in FIG. 2 to AD converters 5a and 5b. Then, at the falling edge of CLK1, the AD converter 5a outputs the instantaneous value of the signal s1.
Convert V _s1 to an n-bit digital value, and
The AD converter 5b converts the instantaneous value V _c1 of the signal s2 into an n-bit digital value.

V _s1 = Asin (-ω・ΔT ₁ +φ) (2) V _c1 = Acos (−ω・ΔT ₁ +φ) (3) φ is the phase difference between the clock and signals s1 and s2,
In FIG. 2, it is drawn as φ=0.

These digital values V _s1 and V _c1 are introduced into the arithmetic circuit 11, and the amplitude A of the signals s1 and s2 is calculated by the calculation of equation (4).

A=√ _s1 ² + _c1 ² (4) However, if the amplitude A is fixed in the two-phase oscillator 1, the calculation of equation (4) is not necessary.

Fractional time ΔT ₁ can be found from equation (2).

ΔT ₁ =−1/ω{arc sin (V _s1 /A)−φ} (5) Fractional time ΔT ₁ can also be obtained from equation (3).

ΔT ₁ =−1/ω{arc cos (V _c1 /A)−φ} (6) Note that φ is measured in advance by the following means when time is not being measured. At timing synchronized with the clock, that is, ΔT ₁ = 0, the sin wave voltage V _s0
and measure the voltage V _c0 of the cos wave.

From equation (2), φ=arc sin (V _s0 /A) (7) and from equation (3), φ=arc cos (V _c0 /A) (8) can be obtained.

Although φ can be obtained using either equation (7) or (8), the accuracy is higher if the smaller of |V _s0 | and |V _c0 | is used.

When φ is determined as described above, ΔT ₁ is determined from equation (5) or equation (6). In this case as well, |V _s1 |, |V _c1 |
The accuracy is higher if the smaller one is used.

Specifically, |V _s1 /A| or |V _c1 /A|≦0.707 is used.

The counter 9 receives the start signal from the control circuit 7 and receives the signal s2 from the two-phase oscillator 1.
Count the number of times (cos) crosses a certain level. Of course, the signal counted by the counter 9 is not limited to s2, and may be s1. The counter 9 counts the signal s2 until the control circuit 7 outputs a stop signal.

Next, when a stop pulse is applied to the input terminal p2 as shown in FIG. 2, a stop signal is applied from the control circuit 7 to the counter 9.
stops its operation.

Further, the control circuit 7 outputs CLK2 to the AD converters 5c and 5d in synchronization with the stop pulse. Then, the AD converters 5c and 5d convert the signals s1 and s at the falling edge of CLK2 to
2 instants V _s2 and V _c2 are converted into digital values in the same manner as described above, and the fractional time ΔT ₂ is determined in the same manner as described above.

From the above, ΔT ₁ , ΔT ₂ , N (counter 9
output) and T ₀ (clock period) are known, so using equation (1) above, the arithmetic circuit 11 calculates
It is possible to output the time width T _x .

In addition, in the above, 5a to 5d are used as AD converters.
The explanation was given using an example using the above four pieces. That is, the AD converters 5a and 5b serve to convert into digital values the instantaneous values of the signals s1 and s2 when the start pulse is applied. In addition, AD converters 5c and 5d
are the signals s1,s when the stop pulse is applied
It plays the role of converting the instantaneous value of 2 into digital. As described above, the reason for using four AD converters is that the time that the time width measuring device according to the present invention is trying to measure is the time width of a very high-speed signal. This is because, unless separate AD converters are prepared for the stop pulse and the stop pulse, there is a risk that the purpose of high speed cannot be fully achieved.

However, there is a high possibility that technology will be developed in the future that improves the operating speed of AD converters, and in that case, it will not be necessary to have as many as four AD converters as shown in Figure 1. For example, AD in Figure 1
Converters 5a and 5c are shared, and AD converters 5b and 5
d can also be shared, resulting in a total of two AD converters. Alternatively, it is also possible to deal with this problem with one AD converter.

FIG. 3 is a diagram showing another configuration example of the present invention. The device shown in Figure 3 is equipped with a sample-and-hold circuit, so the conversion time is slightly longer than the device shown in Figure 1, but this makes it more expensive than the AD converter used in Figure 1. It has the advantage of being able to use a high-precision AD converter. The difference between the device in FIG. 3 and the device in FIG. 1 is that the AD converters 5a and 5 in FIG.
c has been replaced with one AD converter 4a in FIG. 3, and AD converters 5b and 5 in FIG.
d is replaced with one AD converter 4b in FIG. 3, and the output signals s1 and s of the two-phase oscillator 1
Sample and hold circuit 2a that temporarily holds 2
and 2b were provided.

The rest of the configuration is the same as the device shown in FIG.
In the device shown in FIG. 3, the instantaneous values V _s1 , V _c1 and V s2 , V _c2 of the signals s1, _s2 described above are temporarily stored in sample-and-hold circuits 2a, 2b, and their contents are read into AD converters 4a, 4b. Then, the arithmetic circuit 11 calculates the time width T _x using the same operation as the apparatus shown in FIG.

As a calculation method, calculate V _s1 /V _c1 = tan (-ω・ΔT ₁ +φ) or V _c1 /V _s1 = cot (−ω・ΔT ₁ +φ) from equations (2) and (3), ΔT ₁ may be determined from one of the above equations. At this time, the denominator on the left side is |V _s1 |, |
The error is small when the larger V _c1 | is used.

Also, the two AD converters in FIG. 3 may be combined into one, and this may be used by switching. In this case, the high speed is inferior to that in Figure 3, but AD
This has the effect of eliminating errors due to instrumental differences in the converter.

In addition, in the above, the signals s1 and s1, which differ in phase by 90°,
The explanation has been given using an example using sin and cos waves as s2, but the invention is not limited to this. For example, a triangular wave can be used as the signal s1,
It may also be used for s2.

C. “Effects of the Present Invention” As described above, according to the present invention, the following effects can be obtained.

Conventional devices, such as the time vernier system, require time for the main clock and vernier clock to match. Furthermore, the time expansion method always requires ΔT _E time.
Even if the operating speed of the AD converter becomes faster in the future than it is now, these required times are theoretically necessary times and there is no room for improvement.

On the other hand, the device according to the present invention requires time in the AD converter and the arithmetic circuit 11, but this has the potential to be further speeded up in the future due to advances in devices, and the operating principle Therefore, it can be faster than conventional means.

As shown in FIGS. 1 and 3, a fractional time less than the clock period can be measured with a relatively simple configuration.

A two-phase oscillator is relatively easy to implement and has high accuracy.

By increasing the resolution of the AD converter,
The resolution of the measurement time width can be improved without increasing the clock frequency or lengthening the measurement time.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a configuration example of a time width measuring device according to the present invention, FIG. 2 is a timing chart of the device shown in FIG. 1, FIG. 3 is a diagram showing another configuration example of the present invention, and FIG. Diagram showing the general principle of measuring time width, No. 5
The figure is a diagram for explaining the operation of the time vernier method, and FIG. 6 is a diagram for explaining the operation of the time expansion method. 1... Two-phase oscillator, 2a, 2b... Sample hold circuit, 3... Clock generator, 4a, 4
b, 5a to 5d...AD converter, 7...control circuit, 9...counter, 11...arithmetic circuit.

Claims (1)

[Claims] 1. A clock generator that generates a clock signal;
A device that is equipped with a counter that counts signals based on this clock signal and that measures time width based on the value of the counter is synchronized with this clock signal and whose phases are mutually mutual.
A two-phase oscillator that outputs two signals s1 and s2 that differ by 90 degrees, and an AD converter that converts the amplitude values of the signals s1 and s2 of the two-phase oscillator into digital values when the time width measurement start signal and end signal are generated. , a control circuit that controls the operating timing of the AD converter and the start and stop of the counter, and an operation that calculates the time width by introducing the digital values of the amplitude values of the signals s1 and s2 and the count signal from the counter. A time width measuring device characterized by comprising a circuit. 2. The time width measuring device according to claim 1, wherein a sine wave and a cosine wave are used as the two signals. 3. The time width measuring device according to claim 1, wherein triangular waves whose phases differ by 90 degrees from each other are used as the two signals. 4. Claim 1, characterized in that a sample and hold circuit is provided at the next stage of the two-phase oscillator, and the value held in this sample and hold circuit is converted into a digital width by an AD converter.
The time width measuring device described in section.