RU2012125C1 - Phase-sensitive rectifier - Google Patents

Abstract

FIELD: electrical measurements. SUBSTANCE: rectifier has multiplier circuit with sources of input and reference signals, (n-1) integrating circuits, adder, (n/2) inverters, (n-1) scaling circuits. Proper selection of gains of scaling circuits makes the rectifier non-sensitive to the specified population of derivatives of the processed signal envelope. EFFECT: reduced dynamic error. 1 dwg

Description

 The invention relates to the field of electrical measurements in the ranges of infra-low and sound frequencies, can be used in the experimental determination of the amplitude-phase frequency characteristics of quadripoles in acoustic measurements, studies in the field of automatic control and regulation, in geophysics, oceanology, biophysics, etc. - analog calculation of the scalar product of the vectors of the reference and measured signals.

 The purpose of the invention is to reduce the dynamic error arising due to changes in the amplitude of the processed signal over time.

 The drawing shows a structural diagram of a rectifier.

 The device contains sources of input 1 and reference 2 signals, a multiplying link 3 with inputs 4, 5 and output 6, n of low-pass filters in the form of integrating circuits 7-1,. . . , 7-n, ent (n / 2) inverters 8-1, 8-m; n-1 scaling links 9-1,. . . , 9-n-1 and n-input adder 10 with inputs 11-1,. . . , 11-n and output 12.

 The rectifier operates as follows.

At inputs 5 and 4 of the multiplying link 3, respectively, the reference U _o = = sin sinω t and the processed signals U = U _m (1 + k sinΩt) sin (ωt + φ) are received, where U _m is the amplitude of the processed signal;
ω is its frequency;
φ is the phase shift between the input signal and the first harmonic of the reference signal;
Ω is the frequency of the envelope;
k is the amplitude modulation coefficient.

[cosφ-cos(2ωt+φ)] (1+ksin Ωt) (1)
Выходной сигнал U_вых ^q q-й интегрирующей цепи с постоянной времени q, возбуждаемой сигналом Е по выражению (1), определяется дифференциальным уравнением
τ_q(dU $\genfrac{}{}{0ex}{}{\text{q}}{\text{в}}$ _ых/dt)+U $\genfrac{}{}{0ex}{}{\text{q}}{\text{в}}$ _ых= E (2)
В установившемся режиме при Ω<<ω и (2ωτ_q)²>>1, опуская высокочастотную составляющую, получаем
U $\genfrac{}{}{0ex}{}{\text{q}}{\text{в}}$ _ых=

1+

sin(Ωt-arctgΩτ_q)

cosφ (3)
Образуем линейную комбинацию U_Σ выходных напряжений U_вых ^q n интегрирующих цепей 7-1, . . . , 7-n
U_Σ=

U $\genfrac{}{}{0ex}{}{\text{q}}{\text{в}}$ _ых=

+
+k

sin (Ωt-arctgΩτ_q)

, (4) где α_q - весовые коэффициенты и α₁ = 1.The output voltage E of the multiplying link, taking into account only the first harmonic of the reference signal
E = UU _o =

[cosφ-cos (2ωt + φ)] (1 + ksin Ωt) (1)
The output signal U _oi ^{q of the} qth integrating circuit with a time constant q excited by signal E according to expression (1) is determined by the differential equation
τ _q (dU $\genfrac{}{}{0ex}{}{\text{q}}{\text{at}}$ _s / dt) + U $\genfrac{}{}{0ex}{}{\text{q}}{\text{at}}$ _s = E (2)
In the steady state, for Ω << ω and (2ωτ _q ) ² >> 1, omitting the high-frequency component, we obtain
U $\genfrac{}{}{0ex}{}{\text{q}}{\text{at}}$ _oops =

1+

sin (Ωt-arctgΩτ _q )

cosφ (3)
We form a linear combination of U _Σ output voltages U _output ^q n integrating circuits 7-1,. . . , 7-n
U _Σ =

U $\genfrac{}{}{0ex}{}{\text{q}}{\text{at}}$ _oops =

+
+ k

sin (Ωt-arctgΩτ _q )

, (4) where α _q are weight coefficients and α ₁ = 1.

=

= 0, (5) где γ_q= τ_q/τ₁, то напряжение U_Σ не зависит от 1,2, . . . , n-1 производных Е^(р) низкочастотной компоненты Е по формуле (1). В этом случае эффект фазочувствительного выпрямления определяется мгновенным значением указанной компоненты и ее производными с порядками p = n, n+1, . . .If the quantities α _{i are} determined from the system of equations

=

= 0, (5) where γ _q = τ _q / τ ₁ , then the voltage U _Σ does not depend on 1.2,. . . , n-1 derivatives of E ^{(p) of the} low-frequency component E by the formula (1). In this case, the effect of phase-sensitive rectification is determined by the instantaneous value of the indicated component and its derivatives with orders p = n, n + 1,. . .

Solutions of system (5), for n = 2 and p = 1, α ₁ = 1, α ₂ = -1 / γ ₂ . If n = 3 and p = 1.2, then α ₁ = 1, α ₂ = - (1-γ ₃ ) / γ ₂ (γ _2- γ ₃ ), α ₃ = (1-γ ₂ ) / γ ₃ (γ ₂ -γ ₃ ), etc.

Weights are positive for odd indices q and negative for even q. Therefore, the output signals of the corresponding integrating circuits are inverted by the nodes 8-1,. . . , 8-m. In this case, the transmission coefficients of the scaling links are positive and equal to | α _q | .

For example, with γ = 2 and γ ₃ = 4, α ₂ = -0.75 and α ₃ = 0.125. The dynamic error of the proposed rectifier in comparison with the similar characteristic of the prototype containing only one integrating circuit decreases by 4.3 times at Ωτ ₁ = 0.2, 13.7 times at Ωτ ₁ = 0.1 and 51.2 times at Ωτ ₁ = 0.05.

Claims (1)

 A PHASE-SENSITIVE RECTIFIER containing a multiplying unit, the first and second inputs of which are connected to the outputs of the sources of the input and reference signals, respectively, and a low-pass filter in the form of an integrating circuit, characterized in that, in order to reduce the dynamic error, n - 1 integrating circuits are introduced, eut (n / 2) inverters, n - 1 scaling links and an n-input adder, while the inputs of n integrating circuits are combined and connected to the output of the multiplying link, the output of the first integrating circuit is connected directly to the first the input of the n-input adder, the outputs of the even integrating circuits are connected to the corresponding inputs of the n-input adder via the corresponding inverter and the scaling link, and the outputs of the subsequent odd integrating circuits through the corresponding scaling link to the corresponding inputs of the n-input adder, the output of which is the output phase sensitive rectifier.

Images