SU1465810A1 - Method of checking phase meters - Google Patents

Abstract

This invention relates to a measurement technique. The purpose of the invention is. increasing the accuracy of calibration by eliminating both systematic and random errors of the phase meter. Each time after the initial phase shift is set, a phase shift of approximately 180 is additionally inputted to the signal input. The discrete value of the initial phase shift of each of the n steps is arbitrary. A phase interval equal to the period is covered in n steps. Each measurement step is performed m times. In this case, the increment of the phase meter hits when replacing with the same tag is determined, the average value of the phase shift increments at each step, the average value of the phase meter reading increments. The parameters of the phase meter to be verified are determined by the design formulas presented in the description text. 2 Il.

Description

one

 The invention relates to a measurement technique and can be used to determine the errors of phase meters intended for operation in a wide frequency range of input signals.

The purpose of the invention is to improve the accuracy of verification.

Figure 1 shows the dependence of the display function of the phase meter on the phase difference between the input signals; FIG. 2 is a block diagram of a phase meter calibration.

The verification circuit block contains a generator of standard signals 1, the output of which is connected to a power divider 2, one output of which through a series-connected variable attenuator 3, an adjustable phase switch 2

The clamp 4 and the discrete phase shifter 5 are connected to the signal input of the phase pick-up 6, the reference input of which is connected to the output of the power divider 2.

The essence of the method is to use the internal standard of the phase meters themselves to detect their errors. Such a reference is the frequency of the indication function 9 (&) of the output recorder of the phase meter 6, which periodically repeats every 360 when the phase difference 9 between the input measured signals changes, i.e.

F (b) P (0) WITH

where K is a natural number.

In the general case, the reading function of the phase meter P (0) in the absence of the case

cl

00

3146

It looks like it is shown in FIG. 1 (a thick solid line), where oL is a certain phase difference between the input signals, at which the reading of the phase meter is zero,

Approximate the reading function of the phase meter (0) with the function

/. G. / O-ci-7

(6) r + 2arctg

G, t8 (-2 -) J.

(2)

thirty

which is depicted in FIG. bold dashed line.

The function of the phase meter reading f (c) is then represented as

(0) -2arct8 | tg () -; (e) (3)

where A (9) is the deviation of the display function Φ (Э) from the linearly approximated function f (0).

From expression (2) it follows that 20 the function of the deviation 71 9) is a periodic function with a period of 2T, besides A (9) is a continuous function, since at the points of discontinuity, i.e. when ± 2 it K, the approximated function and the function of the instrument readings are the same, therefore, the function of the deviation 1 ((e) at 0 ot ± 2 K is zero.

Let the input phase meter with the initial phase difference between the input signals at which the instrument reading is equal to f (0) is additionally introduced some phase shift qj, where 35 1 27, the true value: not known, As a result, at the phase meter output increment of the reading (fj (b, c), which in the general case depends on both the initial 40 phase difference b and the input phase difference cf

 р (е + ч „) Ф (е) + Ф (0дЧ о)

  ir + 2arctgj tg () l + (6 + Q,). (4) 45

Taking into account expressions (3) and (4), as well as expressions (2), for incrementing the reading of the phase meter depending on the initial phase difference 9, there are possible §0 two cases:

a) ((Q,, 0 o) W o- (9. (e) 5) with f (e, + tf) -f (v,), this is shown in Fig. 1. In this case, the reading of the phase meter when introducing a phase shift of q at the inlet, it moves from the initial position (point a of fig. 0 to the final position (point b of fig, j). Herewith, the increment

g

0

thirty

20 25

35 40

45

§0

55

O4

phase meter readings with some accuracy equal to the input phase difference

, b) (t (9j, q,) c -2 uV 02-4,) - (02)

(6)

at f (er + C / "): - f (Qj). In this case, the reading of the phase meter from the initial position (point with fig 1) is thrown to the final position (point d of fig 1), having suffered a jump, the magnitude of which is equal to 2 ", therefore, in order to obtain the true value of the phase meter increment in the right-hand side of expression (6), it is necessary to add a value equal to the jump, i.e. If, cast the expression (6) to the form (5).

Thus, in the general case, the true increment of the phase meter reading when introducing an incremental phase shift ifg is expressed as.

V (e, (f) C, -b A (0 + Cf) - A (0). C)

This also means that there is some divergence between the true value of the change in the phase difference (at the input and increment of the phase meter reading p (0, tpp))

8 (5 О „) ((в, cf) -tf, l ((o) - A (0)

(eight)

This discrepancy is due to the systematic errors of the phase meter ,, 1

Let us find the average for the period value of the increment of the readings of the phase meter with a smooth change of the initial phase difference 9 from a certain value of 0 to 8 + 2 n at the same value of the input phase difference Cf,

Taking into account the expression (7) V

V.Cfo) V 6 4 o in

-A (0), + 2V I) - (0) d6 0 So (9)

Due to the periodicity of the function “A (b), the last integral in expression (9) is zero, which means that the period average value of systematic errors with S (O, avg) O.

The physical interpretation of expression (9) is as follows: the average for the period value of the increment of the readings of the phase meter is equal to the unknown phase shift Cfg entered at the input of the phase meter and does not depend on the quality of the phase meter itself.

Know the true value of the phase shift (f and having the form of the function

For the input phase shift cf, you can find: (a) the average absolute systematic error

| 8 (9) 1

| YCI,; b) the mean of the systematic error - - th (c, (,) c) the maximum absolute error

18 (eLe) „„ ,:

| S (e.iio) I.) (0.sr „). I d0; ° 00)

Phase meters are calibrated

R. . itObt; rLS1 fsa J i .. j-1 f G “h15 on the same location, the block diagram of which is shown in J.2.

 -10 ifl J i. ,,, 0 pppgrnosnostei. is assigned to ,,

It is recommended between them to eliminate errors associated with the mismatch of path elements 5

- (, i () 7dQ5

(AND)

It is recommended between them to eliminate errors associated with the mismatch of path elements 5

 f To jtraktE; Recommended -

ne.a.L ,, “.с ,,) (“ ..1. „.- -yy” Гс ™ Gr: y; Gr; :::: -Phaz.r.

To determine the value of the incremental phase shifter with the limits of the change in the introduced phase shift C |), with a phase shift of 0-360, and as a result of the smoothness of 9 in the preg g discrete phase shifter 5 - electrons of one period takes place controlled phase shifter is the larger absolute value of the system — phase shift Cf, whose value is the phase error of the phase meter; consider the case unknown, but for two cases: the detection of the largest errors

a) .30 it is desirable that it be an example

Taking into account the theorem on a finite near-equals 180 °.

Lagrange from the expression (8) error-. variable attenuator serves for

setting the desired siggal level. In practice, together with systematic errors, there are also occasional errors and, despite their smallness as compared with systematic errors, they can play a significant role, as, for example, in the well-known method. In addition, the initial phase shift 8 does not change smoothly, but is set using an adjustable phase shifter with a certain pitch, smallpox / TL. t p is

ka measure -),.

8 (b, c,) (0 + H o) - (

where9b, 60 + tfo,

Absolute maximum error for a given input initial phase shift

l8 (e, 4,) L ,, - pfh (,)

It is obvious that the greatest value of the absolute error of the phase meter has .. Oh - V P i

MY Phase Iusi itiJi -

approximately equal to 360 / p. i.

. J. - - .t t W TTTfl

place at 3H eHHHCj T .. e. phase 9 where

supfl8 (0.4 o) ll, i 1.2, .... p.

   - (15), with some initial

b).-i- 2U.

Considering that L (9) D (b + 2 |)

Empty1 at some initial phase 0 of the controlled phase shifter, m state switches are performed.

sc9,.). “e.c,.) -“ “. (9.,.) -“ (, “-r rv: aZr :: r a; Hr ™“: a-.

Institute of phase meter expression (7) dl on

„“ N, 0 t.t.

+2 it) - N (P (2 r-q,).

where 9 + W about | 2 2

Obviously, even in this case, the largest value of the absolute error of the phase meter occurs when the value of avg II t "

(0, M)

Scientific research institute of the aolya V- / phase phase meter 0; will record

, ((

- meter with one measurement;

y.- random deviation of the phase meter increment,, 2, ..., ni

The outcome of the expression (U), OJ) and (16) can be written at (

lHe, cf,) U "(e,) b T

(17)

The relationship between the rms: koi systematic error for a given phase shift tf, and the largest value of the rms error

follow a:

(9,)

 )four-

where I q I 6.,.

Phase meters are calibrated

itObt; rLS1 fsa J i .. j- on the location, the block diagram of which is shown in FIG. 2.

,,, 0 ППгпршностей. is assigned to ,,

It is recommended between them to eliminate errors associated with the mismatch of path elements 5

tractE; Recommended -

-yy "Gt ™ Gr: g; Gr; :::: -Phazo.r.installation of the desired siggal level. In practice, together with systematic errors, there are also occasional errors and, despite their smallness as compared with systematic errors, they can play a significant role, as, for example, in the well-known method. In addition, the initial phase shift 8 does not change smoothly, but is set using an adjustable phase shifter with a certain pn / t step. t p is

MY Phase Iusi itiJi -

approximately equal to 360 / p. i.

. J. - - .t t W TTTfl

 phase 9 where

Institute of phase meter expression (7) dl on

„“ N, 0 t.t.

Scientific research institute of the aolya V- / phase phase meter 0; will record

, ((

   - meter with one measurement;

y.- random deviation of the phase meter increment,, 2, ..., ni

714

The average value of the increments of the indications i-ro of the phase meter pitch for the initial phase shift Q,

Vi Va - fo + 5 (0; APo) (20)

m

For sufficiently large values of m, the last term of expression (20) tends to zero.

Has experimental values and (f, find

a) average absolute random

Error f: iV, -.- Q. (2J) (ra-l)

b) mean square random error

m

.,.

ta-j

(22)

c) random maximum absolute error

Н1

man

ls il; K-ViLc ,,: (23)

Now it is possible to find the average value of the phase shifts at each step S.Su; .

Kz expressions (20), neglecting the last term due to its smallness, have:

 +

.o)

n

(24)

The last term in expression (24) at sufficiently large values of n also tends to zero.

According to the experimental values (; and

.- a) the mean absolute systematic error

| 8 | -SJ iilijid. (25)

l | n (n-1)

b) mean square systematic error

(26)

k8S L i :: ff:: il-.

  n-l

c) maximum absolute matic error

Y1

max

-rylV: - V; U (27)

Based on the data obtained from expressions (21) - (23) and (25) - (27), we find:

a) average absolute error

z: ivi, -v; i, i: iVi- (

() jnXn-l)

j (28)

b) the total root mean square error

D

|: (V;, - 9i) 2 :()

m-l

n-j

ten

(29)

c) full maximum absolute error

 that

 max

15

one

moix

(thirty)

0

five

0

Thus, with the help of the proposed method, both random and systematic errors of the phase meter being calibrated can be detected with great accuracy, and the need to use exemplary phase shift measures is completely eliminated, since during the measurement the magnitude of the unknown phase shift can be found with high accuracy.

The advantages of the method are the high precision of the verification achieved with the help of simple non-precision equipment. In addition, the accuracy of verification increases with an increase in the number of steps, and a given measure of accuracy is determined by specifying the number of steps.

Claims (1)

Invention Formula The method of phase meter calibration, which consists in applying a signal to the signal input of the phase meter being turned on with an initial phase shift relative to the reference signal, changing the phase shift of the input signal each time by a discrete value, which in n steps covering the faults The limits of the phase meter to be calibrated, and the measurement of the initial phase shift at each step are different, and so that, in order to improve the accuracy of the calibration, every time the initial phase shift is set, the phase shift is added to the signal input of the phase meter to be changed, approximately equal to J80, the discrete value of the initial phase shift of each of the ptags is chosen arbitrary so that the phase interval is equal to the period in n steps, each measurement step is performed m times, and 1A the phase shift at each step and the measured phase shift and the initial phase shift determine the increment of the phase meter readings when measuring at the same step, the average value of the phase shift increments at each step, and the average value of the increment of the phase meter readings averaged over n steps, These phase meters are determined.  ™ | te ": | iK- V, H 7m (m-l) Xn-J) D l | ((, () l. -T n-1 ten de (1) | - IIKS . H  Iukk Max MesKC ; x- V: H: MO COP full average absolute mistake; full of standard error; maximum absolute mistake; incrementing the phase meter reading with one measurement; the average value of the increment reading of the i-ro phase meter; averaged over n steps the average value of the increment of the phase meter readings. ot 01  9i ffs fieL fi8.2