JPH0330830B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION When the present invention performs phase detection on a phase-detected signal,
The present invention relates to a phase detection method that removes errors generated in an output signal due to harmonic components contained in a phase-detected signal.

Phase detection or synchronous rectification is used in a wide range of fields to extract the phase-detected signal component that is in phase with the phase-detected signal. However, if the phase-detected signal contains harmonic components, the output signal may contain harmonic components. The influence of the harmonic components occurs, and the magnitude of the in-phase component is not accurately represented. For example, in a circuit constant measuring device that measures capacitance values and inductance values of capacitors, inductors, etc., the output signal does not accurately represent these values due to the influence of harmonic components.

Conventionally, the following methods are known as methods for removing the influence of harmonic components. FIG. 1 is a diagram showing the relationship between input and output signals of a phase detector, and FIG. 2 is a waveform diagram of a signal applied to the phase detector in a conventional phase detection method. In Fig. 1, phase detector 1
The phase-detected signal V _X is applied to one input terminal of the , and the phase-detected signal V _R is applied to the other input terminal, and a phase-detected output signal V _O is obtained at the output terminal. Second
Figure A shows the phase detected signal V _X (hereinafter referred to as V _X ), and Figure 2 B shows the phase detected signal V _R (hereinafter referred to as V _R ). V _R divides one period of _V (1+
√2) times the square wave signal. like this
When _{phase detecting V} (For details, see Japanese Patent Application Laid-Open No. 133423 of 1970). FIG. 3 is a waveform diagram when the signal shown in FIG. 2B is divided into three signals. signal B is signal A
signal C is 45° ahead of signal A, and signal C is 45° behind signal A.
The amplitudes of these signals B and C are each 1/√2 of the amplitude of signal A. Therefore, signal A,
When B and C are added, the signal shown in FIG. 2B is obtained. Therefore, in other conventional methods, signals A of equal amplitude,
Time-division phase detection is performed using B and C, and a digital value proportional to the magnitude of each phase detection output signal is obtained separately, and only the digital value corresponding to the output signal of the phase detection using signal A is multiplied by √2. and 3
The output signal is obtained by adding the digital values. This method is of course equivalent to the method shown in Figure 2 (for details, refer to the U.S. patent
(See No. 4181949).

It is an object of the invention to provide a phase detection method different from the methods described above. First, in order to facilitate understanding of the present invention, the principle of the present invention will be explained using mathematical formulas.

Phase-detected signal V _X = asin (ωt + θ _a ) + bsin (3ωt + θ _b ) + csin (5ωt + θ _c ) +... Here, θa, θb, and θc represent phase differences.

Phase detection signal V _R (0°) (square wave signal) = A _∞ 〓 ^K=1 sin (2k-1) ωt/2k-1 V _R (+45°) (signal 45° ahead of V _R ) = A _{∞ =} ^A _{_ _} ^_ (ωt−45°)}/2k−1. From these, V _X・V _R (0°)={asin(ωt+θ _a )+bsin
(3ωt+θ _b )+csin(5ωt+ _θc )+……} ×A・(sinωt+1/3sin3ωt+1/
5sin5ωt+……) Here, {asin(ωt+θ _a )}×Asinωt+A/3sin3
ωt+A/5sin5ωt+...) =aAcosθ _a /2−aAcos(2ωt+ _θa )/2
+aAcos(-2ωt+ _θa )/6-aAcos(4ωt+ _θa )/6
+... {bsin(3ωt+θ _b )}×Asinωt+A/3sin3
ωt+A/5sin5ωt+……) = bAcos(2ωt+θ _b )/2−bAcos(4ωt
+θ _b )/2+bAcosθ _b )/6−bAcos(6ωt+θ _b )/
6+... Do the same below.

Furthermore, V _X・V _R (+45°) and V _X・V _R (−
Do the same for 45°). Here, we focus on the DC output component that occurs when the frequencies of the _phase -detected signal _V Then, V _X・V _R (0°)≒ _aA /2cosθ _a +bA _/ 6 cosθ _b +cA/10cosθ _{c V b −135} ° ₎ + _cA /10cos (θ _c −225 _° ) _V +225_{゜ ）} ＋＋ _√2 ＝ _√2Aacosθ
By adding the multiplied signal, the phase detection output is _V
An output signal proportional to cosθ _a can be obtained without being affected by harmonic components contained in the output signal. In this case V _R (0°)
is the reference phase detection signal. Based on the above formula, the present invention performs phase detection as follows. FIG. 4 is a waveform diagram showing the phase detection method according to the present invention. In the figure, first, V _X is changed to V _R for a certain period Tc.
(+45°) and integrate the output. Next, phase detection is performed at V _R (-45°) for the same period Tc, and the output is integrated. Furthermore, √
Phase detection is performed at V _R (0°) for 2Tc period and the output is integrated. At this time, let the time constant of integration be τ,
If τ is selected as a time constant that sufficiently suppresses the AC signal component, only the DC component will be integrated, and the output voltage = 1/τ・Tc・V _X・V _R (+45°) +1/τ・Tc_・V _X・V _R (−45°) +1/τ・√2_・Tc・V 2+√2) The integrated output after the Tc period is 1/τ・Tc・(√2aAcosθ _a ) from the formula. Therefore, if you measure this output signal with a voltmeter,
Not affected by harmonic components contained in _V
It is possible to obtain an output proportional to cosθ _a . For example, if V _X =P+jQ, P is proportional to acosθ _a , so the real component of V _X can be accurately measured. Note that V _R (−
There is no problem with the sequence of applying V _R (+45°), V R (+45°), and V _R (−45°); in short, it is sufficient to carry out continuous measurements using three phase detection signals. In addition, in the above example, the signal V _R (−0°) which is in phase with V _X (assuming θ _a = 0°)
was used as the reference phase detection signal, and V _X (assuming θ _a = 0°)
A signal with a phase difference of 90° (for example, a signal delayed by -90°) V _R (-90°) is used as the reference phase detection signal, and V _R (-
90°) and _the V _R (−90°), which has a phase difference of ±45°
45゜) and V _R (-135゜), the integral output becomes -√2aA(sinθ _a )Tc (the time constant τ of integration is assumed to be 1 for convenience. The same applies hereafter), and the above-mentioned Q is
Since it is proportional to asinθ _a , the imaginary component of V _X can be measured accurately. Next, an application example of the present invention will be described.

FIG. 5 is a block diagram of a vector voltmeter to which the present invention is applied. The phase detection signal V _X is applied to one input terminal of the phase detector 1, and the output signal of the variable phase signal generator 9 is applied to the other input terminal. The variable phase signal generator 9 responds to the control signal of the logical operation controller 5 to generate a signal _VR (0°) which is in phase with the reference signal V _REF .
Signal V _R (+45°) out of phase with the signal by ±45°, V _R
(-45°) etc. The integrator 3 is also connected via a switch 4 (or further via a filter) to the output signal of the phase detector 1 or a DC reference signal (voltage
Er)6 is selectively applied. The output signal of the integrator 3 is applied to the logic operation controller 5. Controller 5
generates the aforementioned control signal, and also performs comparison operations, control operations for switch 4, counting operations, etc.
Control is performed so that a so-called dual slope integral operation is performed. This voltmeter works as follows.

We will explain the case of measuring the V _X component that is in phase with V _R (0°). During the charging period, the switch 4 is connected to receive the output signal of the phase detector 1. and signal generator 9 as described above with respect to FIG.
The signals V _R (0°), V _R (+45°), and V _R (−45°) are sequentially generated and phase-detected (synchronous rectification), resulting in (2+√
2) After the Tc period, the output signal of integrator 3 is √2
Aacosθ _a Tc. After this period, switch 4
is switched, voltage Er is applied to integrator 3, and a discharging operation is performed. A so-called dual slope integral operation is performed. Therefore, the discharge period _T
Since it is proportional to acosθ _a , T _X = √2Aacosθ _a Tc/Er
The real component of V _X can be found by finding . Here, V _X is _the voltage obtained by applying the reference signal V _REF to a capacitor, inductor, etc.
By setting (X+jY)V _REF , the loss can be measured accurately without being affected by harmonics. Also
If you perform phase detection using V _R (-90°), V _R (-45°), and V _R (-135°) and perform the same operation as above, _T
Since it is proportional to asinθ _a , T _X =√2Aasinθ _a Tc/
By determining Er, the imaginary component of _VX , that is, the capacitance value of the capacitor can be accurately measured. Note that when measuring the loss of the capacitor mentioned above, the phase detection signal does not necessarily need to be derived from V _REF , and may be obtained from another reference symbol. In this case, if there is basically a phase difference between V _REF and the phase detection signal, a final correction calculation may be performed by the logical operation controller 5 (see US Pat. No. 4,196,475).

FIG. 6 is a block diagram of a vector voltage ratio meter to which the present invention is applied. The difference between the voltage ratio meter in Figure 6 and the voltmeter in Figure 5 is that the voltage ratio meter in Figure 6 is
V _X , V _REF are introduced, switch 4 and DC reference signal 6 are absent. The difference in operation is that during the discharging operation, switch 2 is switched to receive V _REF , and V _R (0°) is generated from variable phase signal generator 9 and phase detected. Therefore, the vector voltage ratio of the X component, Y component, etc. can be directly determined by the discharge time T _X =√2 a cos θ _a Tc, etc. In this case, the influence of harmonics may be considered during discharging, but if the charging voltage is small and the discharging voltage is large (for example, θ _a is small, V _R (-90°), V _R
(-45°), charging at V _R (-135°) and discharging at V _R (0°)), the effect of harmonics is so small that it can be ignored.

As is clear from the above description, according to the present invention, the phase detection output is not affected by harmonic components contained in the phase detected signal. Therefore, the present invention can be used with great effect in vector voltmeters and circuit constant measuring instruments.

[Brief explanation of drawings]

Figure 1 is a diagram showing the relationship between input and output signals of a phase detector, Figure 2 is a waveform diagram of the signal applied to the phase detector in the conventional phase detection method, and Figure 3 is the signal shown in Figure 2B. Figure 4 is a waveform diagram showing the phase detection method according to the present invention, Figure 5 is a block diagram of a vector voltmeter to which the present invention is applied, and Figure 6 is a diagram of a vector voltage ratio meter to which the present invention is applied. It is a block diagram. 1: Phase detector, 2, 4: Switch, 3: Integrator, 5: Logic operation controller, 9: Variable phase signal generator.

Claims (1)

[Claims] 1 comprising a phase detection means for phase-detecting a phase-detected signal with a phase detection signal, and an integration means for integrating the output of the phase detection means, and as the phase detection signal, a first signal and the first signal, respectively. +45°, -
The second and third signals having 45° different phases are sequentially applied one signal at a time to the phase detection means, and the output of the phase detection means is applied only for a corresponding period while each signal is being applied. is integrated by the integrating means, and the output of the integrating means accumulated sequentially through three integrations is used as a final phase detection output, the method comprising: A phase detection method characterized in that both integration times are equal to Tc, and the integration time during application of the first signal is √2·Tc.