Methods for measuring time intervals. Time interval meter Structural diagram of a digital time interval meter

There are two main methods for measuring the period and time intervals: oscillographic and electronic counting.

Measurement of time intervals using an oscilloscope is carried out according to the oscillogram of the investigated voltage using a linear sweep. Due to significant errors in counting the beginning and end of the interval, as well as due to the non-linearity of the sweep, the total error in measuring time intervals is a few percent. A much smaller error is inherent in specialized meters of time intervals with a spiral sweep.

At present, electronic counting methods for measuring the period and time interval are the most common. When measuring very small time intervals, conversion methods are convenient. Based on these methods, interval multipliers have been created - devices that allow expanding the measured interval by a given number of times. Multipliers are often used in conjunction with electronic counting devices.

10.1 Electronic counting time interval meter

The block diagram of the time interval meter is shown in fig. 6.1, . The investigated voltages U x 1 and U x 2 are fed through two channels to the forming devices. When these voltages reach the reference levels U 01 and (U 02 , short pulses U H and U K appear at the output of the forming devices, corresponding to the beginning and end of the measured time interval Tx. These pulses act on the trigger, the output pulse of which unlocks the selector for the time Tx.

During the duration of the pulse, counting pulses with a known period T 0 coming from the generator are recorded by the counter.

Their number N is proportional to the measured time interval and is read from the reading device,

The scheme of the period meter differs from the considered one in that the pulses of the beginning and end of the interval equal to the repetition period of the studied voltage are formed in one channel, and the second formation scheme is absent.

The period of the counting pulses T 0 is selected as a multiple of 10 - k , s, where k is an integer.

The systematic component of the instability of the counting pulses can be reduced by periodically adjusting the frequency of the generator.

The discretization error, to reduce it, it is necessary to increase the generator frequency, the maximum value of which is limited by the speed of the counter used. Currently, the best mass-produced counters operate up to frequencies of hundreds of megahertz. The discretization error can be somewhat reduced by using a shock-excited counting pulse generator triggered by a UH pulse.

If the device is designed to measure the delay time in the device under study, then the interval start pulse can be synchronized with counting pulses. The time interval meter includes a frequency divider triggered by counting pulses. The pulse from the output of the divider starts the device under study. Due to the instability of the time delay in the divider, it is not possible to completely eliminate the start error.

The accuracy of measurements can be greatly improved by applying the special methods discussed below.

If the measured interval is repeated, then the discretization error can be reduced by increasing the measured interval by an integer number of times or by making multiple measurements.

10.2 Frequency measurement

Frequency measurement is one of the most important tasks solved in radio engineering. Frequency can be measured with very high accuracy, therefore, methods for measuring various parameters with their preliminary conversion into frequency and measurement of the latter have become widespread.

There are the following main frequency measurement methods; electronic counting, capacitor charge and discharge, comparison of the measured frequency with the exemplary one, as well as with the help of selective passive circuits.

The electronic counting method consists in counting the number of periods of an unknown frequency during an exemplary time interval by an electronic counter, the speed of which limits the range of measured frequencies to 100 ... 500 MHz. Large frequencies have to be converted, lowering them to the specified limits. Digital frequency meters make it possible to obtain a relative frequency measurement error of the order of 10 -11 V or less. range up to hundreds of gigahertz.

The method of charging and discharging a capacitor consists in measuring the average value of the charge or discharge current of the capacitor, which is proportional to the frequency of the measured oscillation. The method is suitable for measuring frequencies up to hundreds of kilohertz with an error of about 1%.

Frequency measurement by comparison with the reference can be made in a wide frequency range, including microwave. The measurement error depends mainly on the error in determining the reference frequency and can be up to 10 -13 .

Frequency measurement using selective passive circuits: resonant circuits and resonators, is reduced to tuning the circuit to resonance, the value of the measured frequency is read from the scale of the tuning element. The measurement error is up to 10 -4 .

Thus, the most accurate results are obtained by the methods of electronic counting and comparison, which is due to the presence of quantum frequency standards, the best examples of which are characterized by frequency instability up to 10 -13 . For example, hydrogen frequency standards produced by the industry allow obtaining exemplary frequencies with an instability of 5 ... 10 -13 per day.

Carrying out accurate measurements requires knowing not only the nominal value of the reference frequency, but also some other parameters that characterize its instability.

10.3 Electronic counting method for measuring frequency

The electronic counting method is based on counting the number of pulses with an unknown repetition rate fx on a known, stable time interval. A simplified block diagram of the frequency meter (Fig. 8.2, a) is similar to the time interval meter circuit.

The frequency of the quartz oscillator is chosen equal to n*10 k Hz, where k is an integer, and the value of the division factor n is a multiple of ten. Therefore, the number of pulses recorded by the counter N corresponds to the value of the measured frequency in the selected units. The value f 0 is read from the reading device of the device.

Frequency measurement by charging and discharging a capacitor

This method is the basis for the operation of the frequency meter, the circuit of which is shown in. rice. 8.4, a. Voltage U g with a frequency f x is supplied to the limiting amplifier (Fig. 8.4, b). Its output voltage U 2, which is in the form of rectangular pulses, acts on a circuit consisting of a capacitor C and diodes D1 and D2. Let at the initial moment of time the voltage on the capacitor Uс = U2- The charge time constant is chosen to be much less than half of the period of the input voltage. The average value of the capacitor charge current passing through the diode D1 and the magnetoelectric device,

is proportional to the frequency fx, so the scale of the magnetoelectric device is calibrated in terms of the measured frequency.

Frequency counters of the considered type operate in the range from tens of hertz to units of megahertz. This frequency range is covered by several subranges with different measurement limits. The transition from limit to limit is achieved by changing the capacitance, which is chosen such that at the limiting frequencies of the subranges, the average current of the device is sufficient to deviate the arrow to the full scale.

Frequency measurement by comparison with reference

In this method, the measured frequency fx is compared with the known frequency f 0 of the reference frequency oscillator. By rearranging the latter one achieves equality

where Δσp1 is the frequency comparison error.

The frequency comparison error depends on the frequency equality indication method. In some devices, a mixer and headphones are used to indicate equality (Fig. 8.5, a). Under the action of oscillations of the reference and measured frequencies, oscillations of combination frequencies of the form mfx ± occur in the mixer. nf 0 , where m and n are integers. If the difference frequency signal falls within the headphone bandwidth, then the operator hears a tone of this frequency. By changing f 0, you should achieve the lowest tone, which for various types of headphones is tens of hertz.

Since the frequency is unknown during measurements, the method is ambiguous and before measurements it is necessary to know the approximate value of f x . The considered method of measuring frequencies is sometimes called the method of zero beats.

Measurements are made using the fork method. The comparison error in this case is 10...30 Hz.

10.4 Frequency measurement with selective passive circuits

Measurement in this way is reduced to tuning the selective circuit to the frequency of the signal. The frequency is counted by the position of the tuning element. Such circuits can be bridge circuits and oscillatory circuits. Currently, bridge frequency meters, the scope of which is limited low frequencies, have been completely superseded by other types of devices. Practical application was found only by frequency meters using a resonant circuit, called resonant wavemeters. These simple instruments cover the frequency range from hundreds of kilohertz to hundreds of gigahertz. A simplified diagram of a resonant wavemeter with a loop is shown in fig. 8.8. A voltage of unknown frequency fx is supplied through the coupling coil Lsv to a circuit consisting of a standard coil L and a variable capacitor C The circuit is tuned by changing the capacitance. The state of resonance is determined by a magnetoelectric device by the maximum voltage on a part of the coil. The value of the measured frequency is read from the scale of the capacitor.

The frequency measurement error using resonant wavemeters is determined by the following main factors: calibration error, instability of the resonant frequency of the oscillatory system, the influence of communication with the generator and indicator, inaccuracy of fixing the resonance. The calibration error can be large if there are malfunctions in the tuning mechanism, which has a rather complex design. This error increases due to the wear of the mechanism parts, the appearance of distortions and backlashes.

Due to the connection with the indicator and the source of the measured frequency, active and reactive resistances are introduced into the resonator. An increase in active losses reduces the quality factor, and the variability of the introduced reactances leads to a shift in the resonance. Reducing the errors due to the influence of the indicator and the signal source is achieved by reducing the connection. But in this case, the voltage supplied to the detector decreases, and amplifiers have to be introduced into the circuit after the detector.

There are two main methods for measuring the period and time intervals:

oscillographic;

electronic counting.

Measurement of time intervals using an oscilloscope is carried out according to the oscillogram of the investigated voltage using a linear sweep. Due to significant errors in counting the beginning and end of the interval, as well as due to the non-linearity of the sweep, the total error in measuring time intervals is a few percent. A much smaller error is inherent in specialized meters of time intervals with a spiral sweep.

At present, electronic counting methods for measuring the period and time interval are the most common. The main ones are:

digital method for measuring time intervals;

interpolation method;

vernier method.

Digital method for measuring time intervals

The principle of measuring the period of a harmonic signal by the digital method using a digital frequency meter is illustrated in fig. 17.1, which shows the block diagram of the device in the mode of measuring the period of harmonic oscillations and the timing diagrams corresponding to its operation.

Time interval measurement T x digital method is based on filling it with pulses following with an exemplary period T about, and counting the number M X these impulses.

All elements of the device and their action have been analyzed in matters related to frequency measurement. The structural composition of the reference frequency generator when measuring the period is discussed below.

Rice. 3.6. Digital method for measuring time intervals: a - block diagram; b - time charts

Harmonic signal, period T x which you want to measure, after passing the input device VU (u 1 - output signal VU) and pulse shaper F2 converted into a sequence of short pulses u 2 from the same period. In the device for the formation and control of the UFU, a strobe pulse is formed from them And h rectangular shape and duration T x, arriving at one of the inputs of the time selector Sun. Short pulses are applied to the second input of this selector. u 4 with exemplary follow-up period T about , created by shaper F1 from oscillations of the reference frequency generator GOC.

Time selector Sun skips to the counter MF M X counting pulses u 4 for a time T x, equal to the duration of the strobe pulse And h. Measured period T x, as follows from Fig. 17.1, b,

T x = M X T about + Δ t d , (3.6)

where Δ t d = Δ t to – Δ t n- total discretization error; Δ t n And Δ t to- discretization errors of the beginning and end of the period T X .

Without taking into account in the formula (17.1) the error Δ t d the number of pulses received by the counter M X = T x /T about, and the measured period is proportional to M X

T x = M X T about . (3.7)

Counter output code MF, issued to a digital reading device COU, corresponds to the number of counting pulses he counted M X, and the testimony COU- period T x, since the repetition period of the counting pulses And 5 is chosen from the ratio T about = 1 - n, where P - integer. So, for example, when P = 6 COU displays a number M X , corresponding to the period T x, expressed in µs.

Period measurement error T x, as in the frequency measurement, has systematic and random components.

Systematic component depends on stability δ sq. reference frequency GOCH(his crystal oscillator), and random determined mainly by the discretization error Δ t d discussed above. The maximum value of this error is conveniently taken into account through the equivalent change in the number of counting pulses M X by ±1.

Wherein maximum absolute discretization error can be determined by the difference of two period values T x obtained by formula (17.2) for M X± 1 and M X and equal to Δ T x = ± T about .

Relevant maximum relative error

δ = ± Δ T x /T x = ± 1/ M X= ±1/( T x f about),

where f about = 1/T about- the value of the exemplary frequency of the generator GOC.

The measurement error is also affected by noise in the channels of the formation of the strobe pulse And 3 and counting pulses And 4 (Fig. 17.1, but), introducing temporal modulation into their position according to random law. However, in real devices with a large signal-to-noise ratio, the measurement error due to the influence of noise is negligible compared to the discretization error.

The total relative error of the period measurement is determined as a percentage by the formula

(3.8)

From expression (17.3) it follows that due to discretization error period measurement errorT x increases sharply as it decreases.

Improving the accuracy of measurements can be achieved by increasing the frequency f about frequency generator (by multiplying the frequency of its crystal oscillator in Ku times), i.e. by increasing the number of counting pulses M X. For the same purpose, a frequency divider of the studied signal with a division factor is introduced into the circuit after the input device TO(in Fig. 17.1, but not shown). This takes the measurement TO periods T X and in TO times the relative discretization error decreases.

The discretization error can be reduced and method of measurements with multiple observations. However, this significantly increases the measurement time. In this regard, methods have been developed that reduce the discretization error with a significantly lower increase in measurement time. These include: interpolation method, vernier method.

There are the following methods of electronic measurement of time intervals according to the method of displaying information:

Oscilloscope;

Digital.

Digital methods for measuring time intervals include:

Sequential counting method;

Delayed match method;

Nonius method;

Methods with intermediate conversion.

Consider the features of each of the listed measurement methods.

Essence sequential counting method consists in representing the measured interval fmeas as a sequence of a certain number of pulses following one after another with a certain time interval fo. By the number of pulses of this sequence, called quantizing, judge the duration of the interval. The number of pulses of the quantizing sequence is a digital code of the time interval f meas. Figure 1.1 shows the timing diagram for the sequential counting method.

Figure 1.1 - Timing diagram for the sequential counting method

a) pulses of the quantizing sequence;

b) impulses that determine the beginning and end of the measured time interval;

c) control impulse;

d) pulses at the input of the selector

A device that implements this method is called a serial counting converter. The functional diagram of the device is shown in Figure 1.2. The algorithm of its work is the following. The time selector receives pulses from the quantizing sequence generator. The time selector is controlled by a rectangular pulse, the duration of which is equal to the measured interval f meas. The control pulse is generated by the formation unit.

Figure 1.2 - Functional diagram of the sequential counting converter

In the presence of a control pulse, the pulses of the quantizing sequence pass through the selector, which are then registered by the counter.

The disadvantage of the method is the lack of accuracy in many cases. To improve accuracy, it is necessary to reduce the interval f about or somehow take into account the intervals Df 1 and Df 2 . Reducing the interval f o requires an increase in the speed of recalculation schemes, which is difficult to implement. The interval Df 1 can be reduced to zero if you synchronize the pulses of the quantizing sequence with the starting pulse. To take into account the interval Df 2, there are various methods.

Nonius method. The vernier method has found wide application in the technique of measuring time intervals, both as a means of reducing the error of sequential counting converters, and as an independent method for constructing some measuring devices.

Figure 1.3 shows a functional diagram of a time interval meter with a vernier method for reducing the error Df 2 and with synchronization of the starting pulse (Df 1 = 0).

Figure 1.3 - Functional diagram of the vernier time interval meter

The scheme works as follows. The pulses from the quantizing sequence generator are fed to the inputs of the coincidence circuits and to the input of the frequency divider. The frequency divider generates pulses that are synchronous with the quantizing sequence and serve to trigger the devices under study. At the same time, the divider pulses open the coincidence circuit, the output pulses of which are recorded by a coarse counter.

The vernier pulse generator is triggered by a stop pulse. The pulses it generates with a period

f i \u003d (n-1) / n,

where n is an integer, arrive at the other input of the coincidence circuit and are simultaneously registered by the exact counting counter.

After a certain period of time, depending on the duration of the section f 0 -Df 2 , the pulses of the quantizing and vernier sequences will coincide. The coincidence circuit pulse blocks the vernier pulse generator. It is obvious that the number of pulses registered by the counter is proportional to the duration of the section f 0 -Df 2 .

The measured interval fmeas can be expressed as

Ф meas \u003d (N-N n) f 0 + N n Df n, (1.1)

where N is the reading of the coarse counter;

N n - indications of the exact counting counter;

Df n - vernier step equal to f 0 /n.

Thus, the vernier method makes it possible to reduce the absolute measurement error to the value f 0 /n. In this case, the value of n can reach quite large values ​​(several tens and even hundreds), which determines the wide distribution of the method.

The use of the vernier method for large values ​​of n imposes a number of requirements on the circuit nodes, the most significant of which are:

high frequency stability of the vernier sequence;

high stability of pulse parameters of both sequences;

high resolution coincidence circuits.

A significant disadvantage of the vernier method is the inconvenience of reading the measurement results from several scoreboards with subsequent calculations.

TO methods with intermediate conversion include the time-amplitude conversion method and the time-scale conversion method.

Time-amplitude conversion method is used to account for the section Df 2 in the sequential counting converter. Figure 1.4 shows the functional diagram of the measuring device.

The device operation algorithm is as follows. The pulses of the quantizing sequence from the generator are fed to the first inputs of the coincidence circuits 1 and 2, which are controlled by a trigger through the second inputs.

With the arrival of the start pulse, the flip-flop is flipped, and coincidence circuit 2 opens and coincidence circuit 1 closes. The coarse timing circuit, consisting of coincidence circuit 2 and a counter, begins to work.

Figure 1.4 - Functional diagram of the time interval meter according to the time-amplitude conversion method

The stop pulse returns the trigger to its original position, coincidence circuit 2 closes and coincidence circuit 1 opens. The stop pulse simultaneously enters the time-to-amplitude converter and starts it. The first pulse from the output of the coincidence circuit 1 stops the converter. In this case, a pulse appears at the output of the converter, the amplitude of which is proportional to the duration of the interval between two pulses - the stop and the first pulse from the output of the coincidence circuit 1, i.e., proportional to the section Df 2. As a time - amplitude converter, a linear sawtooth voltage generator is most often used, controlled by two pulses - starting and stopping.

Next, the pulse from the output of the converter is fed to the input of the n-channel amplitude analyzer. In the simplest case, the amplitude analyzer can be made in the form of n parallel-connected integral discriminators with equally spaced discrimination thresholds. Depending on the amplitude of the pulse at the output of the converter, the output of the analyzer will be a signal of one type or another (the type of signal depends on the type of analyzer used), which carries information about the duration of the interval Df 2 . This signal is fed to the decoding and display unit.

Time scale conversion method consists in the fact that the duration of the measured interval f meas is converted into a pulse with a duration kf meas, which is measured using a serial count converter. Typically, time scale conversion is done in two steps. The first of them consists in the time-amplitude type transformation, the second one - in the amplitude-time type transformation. Figure 1.5 shows a general functional diagram of the measuring device. The start and stop pulses, the interval fmeas between which you want to measure, are fed to the time scale converter. The pulse at the output of the converter, having a duration kf meas, controls the coincidence circuit, which, during the action of this pulse, passes quantizing pulses from the generator to the counter. Therefore, the generator, the coincidence circuit and the counter are a sequential counting converter, with the help of which the measurement of the interval kf meas.

Figure 1.5 - Functional diagram of the time interval meter according to the time scale conversion method

For the measured interval, we can write

f meas =Nf 0 /k,

where N is the number of pulses registered by the counter.

Thus, the method under consideration makes it possible to measure small time intervals without resorting to high-speed scaling circuits.

The error of the time scale conversion method is determined mainly by the value and constancy of the conversion factor k.

There are two main methods for measuring the period and time intervals: oscillographic and electronic counting.

Measurement of time intervals using an oscilloscope is carried out according to the oscillogram of the investigated voltage using a linear sweep. Due to significant errors in counting the beginning and end of the interval, as well as due to the non-linearity of the sweep, the total error in measuring time intervals is a few percent. A much smaller error is inherent in specialized meters of time intervals with a spiral sweep.

At present, electronic counting methods for measuring the period and time interval are the most common. When measuring very small time intervals, conversion methods are convenient. Based on these methods, interval multipliers have been created - devices that allow expanding the measured interval by a given number of times. Multipliers are often used in conjunction with electronic counting devices.

10.1 Electronic counting time interval meter

The block diagram of the time interval meter is shown in fig. 6.1, . The investigated voltages U x 1 and U x 2 are fed through two channels to the forming devices. When these voltages reach the reference levels U 01 and (U 02 , short pulses U H and U K appear at the output of the forming devices, corresponding to the beginning and end of the measured time interval Tx. These pulses act on the trigger, the output pulse of which unlocks the selector for the time Tx.

During the duration of the pulse, counting pulses with a known period T 0 coming from the generator are recorded by the counter.

Their number N is proportional to the measured time interval and is read from the reading device,

The scheme of the period meter differs from the considered one in that the pulses of the beginning and end of the interval equal to the repetition period of the studied voltage are formed in one channel, and the second formation scheme is absent.

The period of the counting pulses T 0 is selected as a multiple of 10 - k , s, where k is an integer.

The systematic component of the instability of the counting pulses can be reduced by periodically adjusting the frequency of the generator.

The discretization error, to reduce it, it is necessary to increase the generator frequency, the maximum value of which is limited by the speed of the counter used. Currently, the best mass-produced counters operate up to frequencies of hundreds of megahertz. The discretization error can be somewhat reduced by using a shock-excited counting pulse generator triggered by a UH pulse.

If the device is designed to measure the delay time in the device under study, then the interval start pulse can be synchronized with counting pulses. The time interval meter includes a frequency divider triggered by counting pulses. The pulse from the output of the divider starts the device under study. Due to the instability of the time delay in the divider, it is not possible to completely eliminate the start error.

The accuracy of measurements can be greatly improved by applying the special methods discussed below.

If the measured interval is repeated, then the discretization error can be reduced by increasing the measured interval by an integer number of times or by making multiple measurements.

10.2 Frequency measurement

Frequency measurement is one of the most important tasks solved in radio engineering. Frequency can be measured with very high accuracy, therefore, methods for measuring various parameters with their preliminary conversion into frequency and measurement of the latter have become widespread.

There are the following main frequency measurement methods; electronic counting, capacitor charge and discharge, comparison of the measured frequency with the exemplary one, as well as with the help of selective passive circuits.

The electronic counting method consists in counting the number of periods of an unknown frequency during an exemplary time interval by an electronic counter, the speed of which limits the range of measured frequencies to 100 ... 500 MHz. Large frequencies have to be converted, lowering them to the specified limits. Digital frequency meters make it possible to obtain a relative frequency measurement error of the order of 10 -11 V or less. range up to hundreds of gigahertz.

The method of charging and discharging a capacitor consists in measuring the average value of the charge or discharge current of the capacitor, which is proportional to the frequency of the measured oscillation. The method is suitable for measuring frequencies up to hundreds of kilohertz with an error of about 1%.

Frequency measurement by comparison with the reference can be made in a wide frequency range, including microwave. The measurement error depends mainly on the error in determining the reference frequency and can be up to 10 -13 .

Frequency measurement using selective passive circuits: resonant circuits and resonators, is reduced to tuning the circuit to resonance, the value of the measured frequency is read from the scale of the tuning element. The measurement error is up to 10 -4 .

Thus, the most accurate results are obtained by the methods of electronic counting and comparison, which is due to the presence of quantum frequency standards, the best examples of which are characterized by frequency instability up to 10 -13 . For example, hydrogen frequency standards produced by the industry allow obtaining exemplary frequencies with an instability of 5 ... 10 -13 per day.

Carrying out accurate measurements requires knowing not only the nominal value of the reference frequency, but also some other parameters that characterize its instability.

10.3 Electronic counting method for measuring frequency

The electronic counting method is based on counting the number of pulses with an unknown repetition rate fx on a known, stable time interval. A simplified block diagram of the frequency meter (Fig. 8.2, a) is similar to the time interval meter circuit.

The frequency of the quartz oscillator is chosen equal to n*10 k Hz, where k is an integer, and the value of the division factor n is a multiple of ten. Therefore, the number of pulses recorded by the counter N corresponds to the value of the measured frequency in the selected units. The value f 0 is read from the reading device of the device.

Frequency measurement by charging and discharging a capacitor

This method is the basis for the operation of the frequency meter, the circuit of which is shown in. rice. 8.4, a. Voltage U g with a frequency f x is supplied to the limiting amplifier (Fig. 8.4, b). Its output voltage U 2, which is in the form of rectangular pulses, acts on a circuit consisting of a capacitor C and diodes D1 and D2. Let at the initial moment of time the voltage on the capacitor Uс = U2- The charge time constant is chosen to be much less than half of the period of the input voltage. The average value of the capacitor charge current passing through the diode D1 and the magnetoelectric device,

is proportional to the frequency fx, so the scale of the magnetoelectric device is calibrated in terms of the measured frequency.

Frequency counters of the considered type operate in the range from tens of hertz to units of megahertz. This frequency range is covered by several subranges with different measurement limits. The transition from limit to limit is achieved by changing the capacitance, which is chosen such that at the limiting frequencies of the subranges, the average current of the device is sufficient to deviate the arrow to the full scale.

Frequency measurement by comparison with reference

In this method, the measured frequency fx is compared with the known frequency f 0 of the reference frequency oscillator. By rearranging the latter one achieves equality

where Δσp1 is the frequency comparison error.

The frequency comparison error depends on the frequency equality indication method. In some devices, a mixer and headphones are used to indicate equality (Fig. 8.5, a). Under the action of oscillations of the reference and measured frequencies, oscillations of combination frequencies of the form mfx ± occur in the mixer. nf 0 , where m and n are integers. If the difference frequency signal falls within the headphone bandwidth, then the operator hears a tone of this frequency. By changing f 0, you should achieve the lowest tone, which for various types of headphones is tens of hertz.

Since the frequency is unknown during measurements, the method is ambiguous and before measurements it is necessary to know the approximate value of f x . The considered method of measuring frequencies is sometimes called the method of zero beats.

Measurements are made using the fork method. The comparison error in this case is 10...30 Hz.

10.4 Frequency measurement with selective passive circuits

Measurement in this way is reduced to tuning the selective circuit to the frequency of the signal. The frequency is counted by the position of the tuning element. Such circuits can be bridge circuits and oscillatory circuits. Currently, bridge frequency meters, the scope of which is limited to low frequencies, have been completely replaced by other types of devices. Practical application was found only by frequency meters using a resonant circuit, called resonant wavemeters. These simple instruments cover the frequency range from hundreds of kilohertz to hundreds of gigahertz. A simplified diagram of a resonant wavemeter with a loop is shown in fig. 8.8. A voltage of unknown frequency fx is supplied through the coupling coil Lsv to a circuit consisting of a standard coil L and a variable capacitor C The circuit is tuned by changing the capacitance. The state of resonance is determined by a magnetoelectric device by the maximum voltage on a part of the coil. The value of the measured frequency is read from the scale of the capacitor.

The frequency measurement error using resonant wavemeters is determined by the following main factors: calibration error, instability of the resonant frequency of the oscillatory system, the influence of communication with the generator and indicator, inaccuracy of fixing the resonance. The calibration error can be large if there are malfunctions in the tuning mechanism, which has a rather complex design. This error increases due to the wear of the mechanism parts, the appearance of distortions and backlashes.

Due to the connection with the indicator and the source of the measured frequency, active and reactive resistances are introduced into the resonator. An increase in active losses reduces the quality factor, and the variability of the introduced reactances leads to a shift in the resonance. Reducing the errors due to the influence of the indicator and the signal source is achieved by reducing the connection. But in this case, the voltage supplied to the detector decreases, and amplifiers have to be introduced into the circuit after the detector.

In this article, a device for measuring time intervals was developed. According to the task, the time interval can lie within 1ms-32C.

To measure the time interval between two events, it is necessary to "fill" the measured interval with pulses, and then count the number of pulses.

For a microcontroller, this means:

By definition of an event corresponding to the beginning of the time interval, start a “generator” that produces a sequence of pulses of a certain duration,

Organize the counting of impulses of a given sequence,

On the event corresponding to the end of the time interval, stop the "generator",

- "issue" the value of the number of pulses to the specified ports,

- "reset" the value of the pulse counter

Functional diagram for measuring time intervals

Description of the device operation algorithm.

At the beginning of the program, all the interrupt vectors of this processor are listed, the first interrupt is the reset vector ( rjmp RESET ).

In this subroutine, the necessary peripheral nodes of the microcontroller are initialized, namely:

Port A is configured to output

Port C is configured to output

Port D configured for input

Interrupt configured int 1 (falling interrupt)

Interrupt configured int 0 (edge ​​interrupt)

The top of the stack is determined

The initializing part of the program ends with the command SEI - enable interrupts

Upon the arrival of the pulse front (at the output int 1(PD 3)), an interrupt is generated int 1, the instruction counter "leaves" their main loop to the interrupt vector table at address $0004, there is a command to jump to the interrupt handler EXT_INT 1.

In the interrupt routine, the timer-counter T0 is configured.

The timer is given a number to compare (125), a prescaler value (8), an operation mode (reset by coincidence). This means that for eight cycles of the processor, the value in the counter will increase. When it reaches 125, (125*8=1000, at 1MHz clock, the clock period reaches 1µs, 1000µs is 1ms), a T0 coincidence interrupt occurs. So every 1ms, T0 will trigger an interrupt. team reti , the interrupt handler ends, the program counter returns to the main loop (where it was before the interrupt).

Every 1ms T0 will trigger a TIM0_COMP interrupt. This interrupt performs one operation - increasing the register pair Z per unit. This ends the interruption.

Upon the onset of a falling pulse (at pin int0 (PD2)), an interrupt int0 is generated. In this routine, the contents of the index register Z is copied to ports (A and C), then the contents of the counting register are reset. Next, the timer-counter T0 stops (0 is entered into the control register of the counter). This is where the interrupt ends.

Circuit diagram