SU714302A1 - Pulse repetition frequency measuring method - Google Patents

Description

; ;one : ;- / - ; . The invention relates to the field of measurement of electrical quantities and can be used in the development of high-speed converters of pulse frequency to digital code for measuring systems with a frequency representation of the primary signals. There is a known method for measuring pulse frequency, in which the number of integer periods of the measured frequency is counted, depending on the measurement interval, and the number is determined. The values of the fractional parts of the periods adjacent to the beginning and to the end of the measurement shaft, and the value measured by frequency is calculated by the formula. N -I Hilbi,. ". where Ы is the number of integer periods in the measurement interval; H- (Ki is the fractional part of periods; TU is the measurement interval 1. The closest technical solution to this invention is a method for measuring the pulse frequency, in which the number of integer periods Ni is calculated in each of the contiguous measurement intervals Ti and the value of the fractional part of the & U period adjacent to the beginning of each measurement interval, and the value of the measured frequency in the 1st measurement interval is calculated by the formula {2. -.tlilitlillll siiL .:. The disadvantage of the known methods is that P averaging uncertainties when measuring time-varying frequency. In the known methods, the fractional part of the period is calculated as the ratio of the measured durations of the corresponding fractional part of the period and the whole period where tt and t (j is the duration of the fractional parts of the period adjacent to When the frequency in the vicinity of the beginning of the measurement interval is changed, the deist B ® r11a1 EGe shchina the other bar differs from the one calculated by the formula and leads to an error of averaging. .

The aim of the invention is to improve the accuracy of measuring Yemena with a higher pulse frequency.

This goal is achieved by the fact that in the method of measuring the pulse frequency, based on counting the number of integer periods and determining the durations of the fractional parts of the periods of the measured frequency, ranged from the measurement interval to; , and the values of Iactei periodis are determined by C Ratio

fc-K -4Ci-i

iaiil

rrct.t.

H Pc / Ch to iM bkg.

MM -HI I. 1- H KM H Where ukHJ is the fractional part

period in the i-th measurement interval;

tni is the duration of the fractional part of the period, following the start of the i-th interval: measurement;

t is the duration of the fractional part, the period preceding N. beginning of the 1st interval. la measurement;

 the duration of the last integer period in the (1-1) -th measurement interval. FIG. 1 and 2 are time diagrams explaining the essence of the method for measuring the pulse frequency; in fig. 3 - a structured electrical circuit is depicted: an arrangement of the 1B of the most efficient method; in fig. 4 - time di agramma of his work.

The device for inputting frequency signals into the memory of the digital computer (FIG. 3) is performed on logic elements and contains triggers 1.2 — controlling the operation of the entire device, elements AND 3-7, a high frequency pulse generator (pulses .t |,) 8 used when measuring time intervals T, tK, fen, counter 9 for measuring T and tK, counter 10, the number of integer periods N, memory register 11 for storing the contents of account 9 (duration T), counter 12 for measuring tbf, the element OR 13 Elec. delays 14–16, readout block 17 in the digital computer memory, input bus 18, which receives pulses determining the contiguous measuring intervals Tu (pulses η) and b 19 frequency signal bus (pulses fc). V

In yzMeritelnok systems with a frequency representation of the primary signals, the most widespread linear frequency 714302

impulse modulators of the primary signal under study x (t). Two adjacent pulses generated by a linear modulator cut off the constant areas So bounded by the time axis and the signal x (t) (Fig. 2). When measuring the frequency of a linear modulator of a torus on a time interval TU, the most informative measurement result will be that at which the obtained frequency value, averaged over the interleader frequency, will be proportional to the average value of the signal X (t) at this interval. From FIG. 2 it can be seen that the area indicated is equal to

That 5 'Г5иЫ-: By dividing the last expression by SaTuf, we will get the formula for the exact frequency value, the proportion of jibHoro to the mean value of the signal x (t) on the interval TU

j -; - --SHJ / V H; / S.

j - - .. I - - 1 I

Consequently, the fractional parts of the periods should be found, as the ratio of the corresponding areas

 -.

liu - e

With the known measurement method, they are found as a ratio of time intervals

/,.-

In the IB case of a varying signal x (t), the magnitude is not equal to the value of Cin, which leads to an error of averaging. .

We will show that the additional measurement of the duration of the period T- (Fig, 2) - and the calculation of the fractional value (Parts of the period using the formula adopted in the present invention, eliminates the averaging error.

For systems with a frequency module, the inequality is T + t - bt T i.

where Tpii is the period of the highest frequency useful component of the modulating signal spectrum x (t).

Therefore, with sufficient accuracy for the practice, the signal X (t) within the interval Tfttc -tK can be considered linearly varying x (t),

where x is the value of the signal x (t) corresponding to the middle of the interval T;

a is the rate of change of the signal x (t).

Taking into account the previous formulas

) H

) S. where xj. - value of input signal

corresponding to the middle of the interval ta. tH. It's obvious that

...

Then we get

"Te7; nT5T7Tu

After simple transformations, we find the true value of the fractional part of the period & C1 in the 4th measurement interval, /. . .

Therefore, the fractional part, as adopted in the present invention, is defined as the area ratio, which eliminates the averaging error, as a result of which the measurement accuracy of the varying pulse frequency increases.

The durations T, tx, and tn are measured in the device by the pulse counting method f-t of the high-frequency generator for the corresponding time intervals. -.

In the initial state, the device triggers 1 and 2 are set to O. From the output Q of the trigger 1, this gives permission for elements 4, 5 and 6. Pulses fb from generator 8, flow through element 5 to the input of counter 9, When fed to the input buses 18 and 19 pulses c and fc (Fig. 4 a, b), pulses fc through element 4 arrive at the input of counter 10 In addition, each pulse f arriving through element 6, at the input of the record, memory register 11, organizes the reading of information from counter 9 to the memory register (Fig. 4c). The passage further through the element 14 and the element OR 13 to the installation input of the counter 9, pulse fc, sets it to O (Fig. 4 g). As a result, each pulse causes a census of information about the duration of the finished period of the measured frequency to memory register 11, resetting counter 9 and re-counting pulses fb with counter 9. With the arrival of the first pulse fq on bus 18, trigger 1 and 2 are set in (Fig. 4 , e), what is resolved. elements 4 (end of the count N), 5 (end of the measurement tK) and 6 (forbidden, rewrite the information from counter 9 to memory register 11). at the same time, the resolution from the output Q of flip-flop 2 is applied to the element through which the pulses {are fed to the input of the counter 12 (the beginning of the measurement is t "). From the output Q of flip-flop 1, the permission to element 3 is applied. First After pulse f, pulse fc sets trigger 2 to O, which removes the resolution from element 7 (the end of measurement Lc). The same impulse, the passage through element 3 on the most important entrance of block 17, organizes a census

information from counters 9, 10, 12 and memory register 11 into an ECM memory cell) (Fig. 4 g), then counters 10 and 12 through the delay element 15 and counters 10 and 12 and through the OR element 13 counter 9 (Fig. 4 h) . Pulse fg. Also enters through delay element 16 to one of the inputs of trigger 1 and sets it to O. Next, N is measured, T and tic are measured in the next measurement interval

0 and so on.

Thus, information about the values of N-i, T, t, tn-i + i (frequency code) is recorded in each memory cell of the computer. After entering the frequency

The s signal in the digital computer is processed by the resulting code array — the calculation of the values of & n and the discrete values of the measured frequency.

The use of the present invention allows to significantly improve the accuracy of measurements in systems with a frequency representation of dynamic signals. Improving the accuracy of determining the values of the fractional parts of the periods accordingly increases the accuracy of measuring the pulse frequency.

The invention is also effective in the case of pre-recording frequency signals together with a fui signal on a magnetic tape. It is known that recording and playback are accompanied by fluctuations in the speed of the magnetic tape, which cause spurious frequency modulation when the signal passes through the magnetic recording path. This measurement method significantly reduces the averaging errors and more effectively suppresses the parasitic frequency modulation caused by the instability of the movement of the magnetic tape.

Claims (1)

Invention Formula The method of measuring the frequency of the following i pulse, ow, based on counting the number. entire periods and determining the durations of the fractional parts of the periods of the measured frequency that fall into the measurement interval, characterized in that, in order to improve the measurement accuracy of the varying frequency, the duration of the last integer period in each measurement interval is additionally measured, and the fractional values - parts of periods are determined by the ratio - -fcHJ r ,, - fc.JH (T} .rt.; - rtKO, t,. /% tTTTTrXr-Tt-T-j & H - fractional part where AH is the period in the i-th measurement interval;