RU2457617C1 - Integrating voltage analogue-to-digital conversion method - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: method is based on generating a sequence of conversion time cycles of fixed duration, in each of which three unfolding functions are generated; the informative time interval is selected and converted to a digital code, wherein to generate a first unfolding function from the moment in time corresponding to the beginning of each conversion cycle, the sum of the input voltage and the reference voltage of positive polarity is integrated and the current integral valve is subtracted from the positive threshold voltage; in order to generate a second unfolding function, the sum of the input voltage and the reference voltage of negative polarity is integrated and from the moment in time corresponding to the equality of values of the unfolding functions, the value of the unfolding function is recorded, which is equal to the value of the positive threshold voltage, and to generate a second unfolding function, the sum of the input voltage and the reference voltage of positive polarity is integrated, from the moment in time corresponding to the middle of each cycle of fixed duration a third unfolding function is generated via integration of the sum of the input voltage and the reference voltage of negative polarity and summation with the negative threshold voltage of the current integral value, and from the moment in time corresponding to equality of values of the second and third unfolding functions, the value of the third unfolding function is recorded, which is equal to the value of the negative threshold voltage; the duration of informative time intervals is defined as the difference between time intervals for integrating reference voltages of positive and negative polarity, respectively.

EFFECT: faster operation.

4 dwg

Description

The invention relates to electrical engineering. The purpose of the invention is improving performance.

Known methods for integrating analog-to-digital voltage conversion: Germany patent No. 2214602, class. G01R 19/26, 09/04/75; Ornatsky P.P. Automatic measurements and instruments. Kiev: Vishka School, 1973, p. 435-437. In these methods, gaps in the input voltage information take place.

Among all the methods that perform analog-to-digital conversion without omissions of information and presented in the article, Shakhov E.K., Ashanin V.N., Nadev A.I. “Implementation of ΔΣ-ADC concepts in integrating ADCs with other types of pulse modulation” / News of higher educational institutions (Volga region). Penza: CCP, 2006, No. 6, S.226-237, the closest is the method of integrating analog-to-digital voltage conversion, shown in Figure 1 (a timing diagram explaining the method under consideration). Figure 2 shows a structural diagram of a device for its implementation.

This method of integrating analog-to-digital conversion is based on the formation of a sequence of time periods of conversion of a fixed duration, in each of which two development functions are formed, an informative time interval is allocated and informative time intervals are converted into a digital code, moreover, for the formation of the first development function from the time moment corresponding the beginning of each conversion cycle, integrate the sum of the input voltage and the reference voltage of positive polarity and subtracting the current integral value from the positive threshold voltage, and to form a second deployment function, integrate the sum of the input voltage and the reference voltage of negative polarity and, from the moment of time corresponding to the equality of the values of the deployment functions, fix the value of the first development function to the value of the positive threshold voltage, and for the formation of the second deployment function, integrate the sum of the input th reference voltage and the voltage of positive polarity, duration of time intervals informative defined as the difference integration time intervals of the reference voltages of the positive and negative polarity, respectively.

In accordance with Figure 1, R ₁ (t) is the first deployment function, R ₂ (t) is the second deployment function. In volume A [i], the values of the first and second expanding functions are equal to R ₁ (t) = R ₂ (t). Figure 1 shows the case of changing the input voltage from zero to a value of U _x , highlighted moments of switching the values of the reference voltages (+ U ₀ ) and (-U ₀ ): T ₁ [1] ... T ₁ [4] - time intervals from the beginning of the time step of converting a fixed duration (equal to T ₀ ) to the time instant of equality of the first and second developing functions, T ₂ [1] ... T ₂ [4] - time intervals from the time of equality of the first and second developing functions to the time of the beginning of the next temporary fixed duration conversion cycle. The duration of informative time intervals ΔT is defined as the difference between the intervals T ₂ [i], T ₁ [i] of the integration time of the reference voltages of positive and negative polarity, respectively (ΔT = T ₂ [i] -T ₁ [i], where i is the number of time cycle fixed duration conversions). Consider this method on the example of a device that measures DC voltage.

The device shown in figure 2, contains the main 1 and auxiliary 2 integrators, a controlled switch 3 of the reference voltage, a comparator 4, an operational amplifier 5, a driver 6 of the threshold level, the source of the reference voltage 7 (ION), the generator 8 of the reference frequency (GOCH), a control device 9, a controlled key 10, and a computing device 11 (WU). Moreover, the output of the switch 3 is connected to the input 1 of the main integrator 1, and its inputs to the source of the reference voltage 7 and UU 9; input 2 is supplied to input 2 of the main integrator 1 and to input 1 of the auxiliary integrator 2; the input 2 of the auxiliary 2 integrator is constantly supplied with a positive reference voltage from a reference voltage source 7; the outputs of the main and auxiliary integrators are connected respectively to the inputs 1 and 2 of the comparator 4, which supplies a control signal to the input 1 of the UU 9; UU 9 controls the operation of the switch 3, key 10 and VU 11; at the input 2 of the UU 9 and the input 1 of the VU 11 is the reference frequency with GOCH 8; the output of the OS 5 through the key 10 is connected to the input 3 of the integrator 2; the threshold level driver 6 supplies the threshold voltage value to the input 2 of the operational amplifier 5.

The expanding function R ₂ (t) is generated by integrator 1, and the expanding function R ₁ (t) is formed by integrator 2. The moment of equality of the developing functions (t. A) is fixed by comparator 4. At the same moment (t. A), the integrator 2 through key 10 covered by feedback and the value of the deployment function R ₁ (t) becomes equal to the threshold voltage U _n . The polarity of the reference voltage U _{0 is} switched at the beginning of each conversion step of a fixed duration T ₀ , as well as at the times of the comparator 4.

The reference frequency generator 8 generates a sequence of time conversion clock cycles of a fixed duration T ₀ , in each of which two development functions are formed. The first expanding function R ₁ (t) is formed by the integrator 2 and operational amplifier 5 as follows: from the beginning of the conversion cycle of a fixed duration, the values of R ₁ (t) are formed by subtracting from the positive threshold voltage U _{P the} current integral value of the sum of the input U _x and the reference U ₀ stresses; from the moment of the equality of the deployment functions (T.A) fixed by the comparator 4, R ₁ (t) is set equal to U _П until the end of the current measure T ₀ due to the closure of the key 10. The second expansion function R ₂ (t) is formed by the integrator 1 by integrating the difference input U _x and reference U ₀ voltages from the beginning of the conversion cycle of a fixed duration T ₀ until the equality of the deploying functions (t. A) and integration of the sum of the input U _x and reference U ₀ voltages from t. A and to the end of the current cycle T ₀ due to connecting negative or positive Nogo reference voltage source 7 via the key 3 times in the beginning of the next cycle T ₀ of the current or the response of the comparator 4, respectively. The duration of the informative time intervals ΔT ₀ is determined in the computing device 11 as the difference of the time intervals (T ₁ [i] and T ₂ [i]) of the integration of the reference voltages of positive and negative polarity in the process of forming the second deployment function, respectively.

The main disadvantage of this method and, therefore, the device is the presence of a transient process when the input voltage changes (see Shakhov E.K., Ashanin V.N., Nadev A.I. “Implementation of ΔΣ-ADC concepts in integrating ADCs with other types pulse modulation "/ News of higher educational institutions (Volga region). Penza: PSU, 2006, No. 6 S.226-237).

In the steady state U _x = const, the values of U _B [i] of the developing function R ₂ (t) at the points B (i) are equal, i.e. U _B [i-1] = U _B [i] and T ₁ [i-1] = T ₁ [i] and T ₂ [i-1] = T ₂ [i]. From the equation for the second expanding function R ₂ (t)

find the conversion function:

- интервал преобразования,

, n - количество тактов преобразования фиксированной длительности, τ - постоянная времени интеграторов. Полученное значение

Where

- conversion interval,

, n is the number of conversion steps of a fixed duration, τ is the time constant of integrators. Value obtained

converted to digital code in WU 11.

The transient in the time diagrams is formed by connecting to the input of the device at a random moment of time the standard effect in the form of a voltage surge. It can be seen that the points A [0], A [1], A [2], A [3], which determine the moment of switching the polarity of the reference voltage, i.e. the duration of informative time intervals, change their position in each step of the conversion of a fixed duration T ₀ . The duration of the transition process is two cycles of conversion of a fixed duration T ₀ . Thus, only in the 3rd step (T. A [3]) the conversion result will correspond to the input voltage.

To confirm this drawback, we find the values of the informative interval ΔT [i] (i is the number of the current conversion cycle) in the 0, 1, 2, and 3 conversion clock cycles. The values of U _B [i] we find from the equation

where t is the current time during a fixed-duration interval T ₀ , τ is the time constant of integrators 1 and 2 (since the time constants of integrators 1 and 2 are constants, therefore, they do not affect the duration of the transient process, to simplify the calculations, we can assume that R ₁ = R ₂ = R ₃ = R ₄ = R, C ₁ = C ₂ = C and τ = RC). Also in the time diagrams of FIG. 1, for definiteness, U _P = U ₀ and 2τ = T ₀ .

From formula (2) it follows that

The values of U _B [2], U _B [3] and U _B [4] will be found similarly from formula (2).

Since the change in U _x occurs at a random time, we denote the time interval from the beginning of the conversion step of a fixed duration T ₀ to the moment of change of U _x by Δt, then: Δt = kT ₀ , where 0≤k≤1.

Find the value of U _B [2]:

Starting from the 3rd step, k = 0, so the values of U _B [3] and U _B [4] will be equal to U _B [3] = U _B [4] = 2U _X -U ₀ .

For T.A, we can write the following equality of the deploying functions:

R ₁ (T ₁ [i]) = R ₂ (T ₁ [i]).

.Or in expanded form:

.

From here we express

Find the value of the interval ΔT ₀ = T ₁ [i] -T ₂ [i]. Given that T ₂ [i] = T ₀ -T ₁ [i]), we obtain

ΔT ₀ = T ₁ [i] -T ₀ + T ₁ [i] = 2T ₁ [i] -T ₀ .

Since T ₀ = const, U _n = const, then at T ₁ [i] = const for U _x = const the transition process ends, ie when U _B [3] = U _B [4] we get T ₁ [3] = T ₁ [4].

Therefore, the duration of the transition process is 2 cycles: a voltage jump U _x occurred in the 1st cycle, and the correct conversion result was obtained in the 3rd cycle, which limits the speed of this converter (described in the prototype method).

In order to improve performance, a method of integrating analog-to-digital conversion, based on the formation of a sequence of time periods of conversion of a fixed duration, in each of which two development functions are formed, allocate an informative time interval and convert informative time intervals to digital code, moreover, to form the first deployment function from the moment of time corresponding to the beginning of each conversion step, integrate the sum of the input voltage voltage and the reference voltage of positive polarity and subtracting the current integral value from the positive threshold voltage, and to form the second development function, the sum of the input voltage and the reference voltage of negative polarity is integrated, and from the moment of time corresponding to the equality of the values of the development functions, the value of the first development function is fixed equal to the value of the positive threshold voltage, and for the formation of the second developing function, integration of the sum of the input voltage and the reference voltage of positive polarity, the duration of the informative time intervals is determined as the difference between the time intervals of the integration of the reference voltages of positive and negative polarity, respectively, additionally, during the conversion process, they form a third deployment function, for this from the moment corresponding to the middle of each clock cycle of a fixed duration integrate the sum of the input voltage and the reference voltage of negative polarity and summing with a negative threshold voltage the current integral value, and from the moment of time corresponding to the equality of the values of the second and third development functions, fix the value of the third development function to the value of the negative threshold voltage, and the second development function is formed by integrating the sum of the input voltage and the reference voltage negative polarity.

Timing diagrams explaining the operation of the proposed method are shown in FIG. 3, where R ₁ (t) is the first deployment function, R ₂ (t) is the second deployment function, R ₃ (t) is the third deployment function. At points A [0] ... A [n], the values of the first and second developing functions are equal (R ₁ (t) = R ₂ (t)), at points B [0] ... At [n] the values of the second and third developing functions are equal (R ₂ (t) = R ₃ (t)). In the upper part of Figure 3 shows the moments of switching the values of the reference voltage U ₀ / -U ₀ . T ₁ [1] ... T ₁ [n] - time intervals from the beginning of the time step of the conversion of a fixed duration to the time of equality of the first and second developing functions. T ₂ [1] ... T ₂ [n] - time intervals from the time instant equal to the middle of the time interval of the conversion clock cycle of a fixed duration to the time of equality of the second and third developing functions, n is the number of conversion clock cycles of a fixed duration included in the conversion interval T _n .

A device that implements the proposed method and shown in Figure 4, contains the main 1 and auxiliary 2, 12 integrators, a controlled switch 3 of the reference voltage + U ₀ and -U ₀ , comparators 4, 14, operational amplifiers 5, 15, threshold level drivers 6 , 16, a reference voltage source 7 (ION), a reference frequency generator (GFC) 8, controlled keys 10, 13, and computing device 11 (WU). Moreover, the output of switch 3 is connected to the input 1 of the main integrator 1, and its inputs to the source of the reference voltage 7 and UU 9. At the input 2 of the main integrator 1, an input (measured) signal is supplied. The input voltage is constantly supplied to the input 1 of the auxiliary integrators 2 and 12 and the negative reference voltage -U ₀ with ION 7 is constantly supplied to the input 2 of the auxiliary integrator 2, and the positive reference voltage + U ₀ with ION 7 is input 2 of the auxiliary integrator 12 1 and auxiliary 2 integrators are connected respectively to the inputs 1 and 2 of the comparator 4, which is connected to the input 1 of the control unit 9. The outputs of the main 1 and auxiliary 12 integrators are connected respectively to the inputs 1 and 2 of the comparator 14, which is connected to the input ohm 3 UU 9. At the input 1 of the analog adders 18 and 19, output voltages from the auxiliary 2 and 12 integrators are supplied, respectively. The output voltage from the analog adder 18 is supplied to the operational amplifier 5, and from the analog adder 19 is supplied to the operational amplifier 15. Operational amplifiers 5 and 15 are connected to the inputs 2 of the comparators 4, 14. Pulses are fed to the input 2 of the UU 9 and to the input 1 of the VU 11 reference frequency with GOCH 8. Input 2 UU 9 controls the operation of switch 3, keys 10, 13 and VU 11.

The operation of the device is as follows. As in the device that implements the prototype method, a reference frequency generator 8 generates a sequence of conversion time steps of a fixed duration T ₀ . The first developing function R ₁ (t) is generated by integrator 2, the second developing function R ₂ (t) is formed by integrator 1, the third developing function R ₃ (t) is formed by integrator 12. On the time interval from the beginning of a measure of fixed duration to the moment of equal values of the first and the second expanding functions R ₁ (t) = R ₂ (t) t. A) the values of R ₁ (t) are formed by subtracting from the positive threshold voltage U _{P the} current integral value of the sum of the input voltage U _x and the reference voltage U ₀ ; the values of R ₂ (t) are formed as integral values of the difference between the input and reference voltages.

, как показано на Фиг.3. Момент равенства значений третьей и второй развертывающих функций фиксируется компаратором 14. На Фиг.3 эти моменты обозначены т.B [i], где i - номер такта фиксированной длительности. Значения напряжений, участвующих в формировании всех развертывающих функций, приведены на диаграмме.Additionally, the third deployment function R ₃ (t) is generated by the integrator 12 from the value corresponding to -U _P , and from the moment of time shifted relative to the beginning of the formation of the first deployment function for a time interval of duration

as shown in FIG. 3. The moment of equality of the values of the third and second developing functions is fixed by the comparator 14. In FIG. 3, these moments are indicated by T. B [i], where i is the measure number of a fixed duration. The stress values involved in the formation of all the deploying functions are shown in the diagram.

The result of the conversion is determined by the computing device 11 by calculating and digitizing the value

Steady-state conditions: as in the prototype method, the voltage values U _A [i] at points A [i] are equal (ie, U _A [i] = U _A [i + 1] = U _A [n]) and additionally at points B [i] U _B [i] = U _B [i + 1] = U _B [n], therefore, in the steady state, the time interval between points A [i] and A [i + 1] is T ₀ ; the interval between points B [i] and B [i + 1] is also equal to T ₀ and T ₁ [i] = T ₁ [i + 1] = T ₁ [n], T ₂ [i] = T ₂ [i + 1] = T ₂ [n]. From the equation of formation of the second developing function R ₂ (t) for the interval T ₀

U _x + (T ₊ [i] + T _- [i]) + U ₀ (T ₊ [i] -T _- [i]) = 0

you can find an expression for the conversion result

и

для интервала преобразования T_П, где T_П=nT₀.or

and

for the transformation interval T _P , where T _P = nT ₀ .

Similarly, as in the method of the prototype, we consider the abrupt change in U _x at a random point in time. Denote the time interval from the beginning of the conversion step of a fixed duration T ₀ to the moment of change of U _x through Δt, then: Δt = kT ₀ , where 0≤k≤1. We will also assume that U _P = U ₀ and 2τ = T ₀ .

Then the voltage value in T. A [1] will be determined by the expression:

U _A [1] = U _P -2 (U ₀ + U _x (1-k)).

From the diagram of FIG. 3 it can be seen that from the second step k = 0, and the values are U _A [2] = U _A [3], and U _B [2] = U _B [3].

Therefore, starting from the time interval T ₁ [2], the conversion process can be considered steady, therefore, the result of the conversion, starting from this time interval, must be determined in accordance with the expressions (1) for the prototype method and (3) for the proposed way.

до Т₀ (как показано на Фиг.3), аналогично, если скачок входного напряжения произошел между началом интервала T₂[1] формирования R₃(t) и началом интервала T₁[2] формирования R₁(t), то переходный процесс закончится до начала интервала Т₂[2] и также не превышает значения T₀.Thus, if the input voltage jump occurred between the beginning of the interval T ₁ [1] of the formation of R ₁ (t) and the beginning of the interval of T ₂ [1] of the formation of R ₃ (t), the transition process ends before the start of the interval T ₁ [2], t .e. its duration ranges from

to T ₀ (as shown in FIG. 3), similarly, if the input voltage jump occurred between the beginning of the interval T ₂ [1] of formation R ₃ (t) and the beginning of the interval T ₁ [2] of formation R ₁ (t), then the transition the process will end before the beginning of the interval T ₂ [2] and also does not exceed the value of T ₀ .

Since in the prototype method the duration of the transition process is 2 cycles of a fixed duration T ₀ , ceteris paribus, the speed of the proposed method is at least 2 times higher.

To confirm this, we write the equation for the increments of the second expanding function R ₂ (t) at the following time intervals:

1) from the beginning of T ₁ [0] to the beginning of T ₁ [1]:

Simplifying, we get:

U ₀ (T ₊ [0] + T _- [0] -T ₁ [0] -T ₁ [1]) = 0, as expected, since U _x = 0.

2) from the beginning of T ₁ [1] to the beginning of T ₁ [2] (there is a jump in the input voltage):

Simplifying, we get:

it can be seen that this expression differs from expression (3) for the steady state by the presence of kT ₀ .

3) from the beginning of T ₁ [2] to the beginning of T ₁ [3]

Simplifying, we get:

, может быть переписано в общем виде:Given that in the steady state T ₁ [2] = T ₁ [3] = T ₁ [i], we obtain the expression:

, can be rewritten in general form:

,что совпадает с выражением (3) для установившегося режима.

, which coincides with expression (3) for the steady state.

Thus, from the moment U _x changes to the beginning of the steady state, a time interval of duration not exceeding the value of T ₀ passes, which was to be proved.

Information sources

1. The patent of Germany No. 2214602, class. G01R 19/26, 09/04/75.

2. Ornatsky P.P. Automatic measurements and instruments. Kiev: Vishka School, 1973, p. 435-437.

3. Shakhov E.K., Ashanin V.N., Nadev A.I. Implementation of ΔΣ-ADC concepts in integrating ADCs with other types of pulse modulation. / News of higher education institutions (Volga region). Penza: PSU, 2006, No. 6 P.226-237.

Claims (1)

The method of integrating analog-to-digital conversion is based on the formation of a sequence of time periods of conversion of a fixed duration, in each of which two development functions are formed, an informative time interval is allocated and informative time intervals are converted to a digital code, moreover, for the formation of the first development function from the time moment corresponding to the beginning each conversion clock, integrate the sum of the input voltage and the reference voltage of the positive polarity and subtracting the current integral value from the positive threshold voltage, and to form the second deployment function, the sum of the input voltage and the reference voltage of negative polarity is integrated, and from the moment of time corresponding to the equality of the values of the deployment functions, the value of the first deployment function is fixed to the value of the positive threshold voltage, and for the formation of the second deployment function, integrate the amount of input positive voltage and positive polarity reference voltage, the duration of informative time intervals is defined as the difference between the integration time intervals of the positive and negative polarity reference voltages, respectively, characterized in that, in order to increase the speed, they form a third deployment function, for this, from the moment of time corresponding to the middle of each measure fixed durations integrate the sum of the input voltage and the reference voltage of negative polarity and summing with a negative threshold voltage the current integral value, from the moment of time corresponding to the equality of the values of the second and third development functions, fix the value of the third development function to the value of the negative threshold voltage, and the second development function is formed by integrating the sum of the input voltage and the reference voltage of negative polarity.

Images