SU1405111A1 - Method and device for conversion of short pulses of known shape to number - Google Patents

Abstract

The invention relates to a physical experiment technique and can be used in nuclear physics in the development of high-speed analog-digital converters (ADC), for example, to measure the amplitude of pulses in drift-based ionization spectrometers. The aim of the invention is to improve the accuracy of the conversion. The method of converting short pulses of known shape into a code is that simultaneously with the determination of the higher bits, the duration of the converted pulse at the upper nth reference level is measured and the amplitude of the pulse uh is exceeded over the nth reference level and corresponding to it the lower bits of the output code using the formula LF. (n, ut). A device for implementing this method comprises a parallel ADC, two RS flip-flops, an encoder, anti-matching elements, an OR element, delay lines, a chronotron, and a persistent storage device. 2 sec. f-ly, 1 ill. with (L

Description

4;:

about ate

The invention relates to a technique of physical experiment and can be used in nuclear physics by developing high-speed analog-to-digital converters (ADCs), for example, to measure the amplitude of pulses in drift-based ionization spectrometers.

The aim of the invention is to improve the accuracy of the conversion.

The drawing shows a functional diagram of the device for implementing this method.

The essence of the method of converting short pulses of known shape into a code is that using a set of reference levels of a parallel A / D converter, a preliminary (rough) measurement of the amplitude of the input signal is performed. In this case, the higher bits of the output code are determined. Based on the results of operation of a parallel ADC, the number n of the highest achieved reference level is determined, against which the duration of the input pulse / ut is measured at this nth level. In this case, the pulse amplitude is exceeded over the nth reference level and the corresponding lower bits of the output code are found by the formula (n, & t), based on the previously known form of the input signal. When using this method in nuclear electronics, the popularity of the input signal form I is determined by the fact that I receive signals from the spectrometric amplifier at the ADC where they are formed.

In the particular case for an input signal having the shape of an isosceles triangle, the dependence LL f

(n, ut) looks like lh 2 nh ---

where h is the value of one quantization level, T is the duration of the input signal at the base „

For a Gaussian signal, this dependence can be obtained by expanding the Gaussian curve in Taylor's series near the point corresponding to the maximum of the curve:

uh p-2 (-) -12 (-t |) + ... 55

where G is the standard deviation of the gaussian curve.

five

0

five

0

five

0

five

O

five

A device for implementing this conversion method comprises a parallel ADC 1, two RS flip-flops 2, delay lines 3, an encoder 4, anti-matching elements 5, G1 6, a hronotron 7 and a read-only memory (ROM) 8.

The input of the A / D converter 1 is connected to the input 9 of the device, the bus 10 Reset the device is connected to the RF inputs of the RS flip-flops 2, the outputs of the encoder 4 are the first group of outputs 11 of the device, and the outputs of the ROM 8 are the second group of outputs 12 of the device.

The device works as follows.

Let an input impulse, for example, of a triangular shape, enter the device input 9. At the initial stage of operation of the device, parallel APP 1 is triggered, this will lead to the fact that at certain outputs of its output, depending on the amplitude of the input signal, the code of its highest bits will appear. This code will trigger the corresponding RS-flip-flops 2, previously through the bus 10 Reset set to the zero state. According to the state of the RS flip-flops 2 connected to the inputs of the a 4 encoder, on. The first group of device outputs 11 generates a code for the higher bits of the amplitude of the input signal. At the same time, regardless of the voltage levels at the outputs of the parallel A / D converter 1, the information in the RS flip-flops 2 is stored for the entire measurement cycle and multiplied only by the following signal. Reset on bus 10 Reset device.

The duration of the output signal from the oldest triggered output of the parallel A / D converter 1 corresponds to the base duration of the remainder of the input pulse. This signal passes through the corresponding delay line 3 and the anti-coincidence element 5 on the element OR 6. The delay lines 3 are included in order to prevent short signals that OR can cause on the OR element 6 to occur due to a delayed operation in the parallel ADC 1 circuits. delays for the lowest output of the parallel A / D converter 1 must be selected in the order of half the rise time of the input pulse, for more high outputs

31

parallel ADC 1 it gradually decreases) and for the most senior output of parallel ADC 1, it can be chosen n times smaller, where n is the number of outputs of parallel ADC 1.

Thus, the duration of the signal at the output of the element OR 6 exactly corresponds to the duration of the base of the residue. For input pulses with a peak tope, for example, for a Gaussian form, it is better to measure the response time of the parallel ADC 1 not at the highest level for a given pulse, but at the second, third level from the top. In this case, the measurement accuracy will be higher.

Chronotron 7 is used as a time interval meter in the device. ROM 8 is included to correct nonlinearity, which occurs due to the fact that the duration of the residual depends on amplitude nonlinearly. The first group of inputs of the ROM B receives the code of the higher bits from the output of the encoder 4, the second group of inputs contains the code for measuring the duration of the remainder from the chronotron 7, while the output of the ROM 8 and therefore the second group of outputs 12 of the device is generated lower-order amplitude bits of the input signal.

Claims (2)

Claim 1, A method for converting short pulses of known shape into a code, including comparing the pulse amplitude with a set of reference levels, the results of which determine the number n of the upper reached level and the corresponding high bits of the output code that distinguish Q n 5 0 five 0 114 so that, in order to increase the accuracy of the conversion, simultaneously with the determination of the higher bits, the duration of the input pulse is measured at the nth level, and the excess amplitude of the pulse Lh above the nth level and the corresponding lower bits of the output code are based on the known form of the input pulse 2. A device for converting short pulses of known shape into a code containing n comparators, the first inputs of which are combined and connected to the input signal bus, and the second inputs to the corresponding resistors of a voltage divider connected to the reference voltage bus, the outputs of the Comparators are connected respectively to the S-inputs of the RS-flip-flops, the number of which is equal to n, the R-inputs of the RS-flip-flops are combined and connected to the Reset bus, and the single outputs are connected to the corresponding inputs of the encoder, the outputs of which are connected to the senior bus x device bits, S-inputs of RS-flip-flops, except for the first one, are connected via delay lines, respectively, with direct inputs (n-1) of anti-coincidence elements, inverse inputs of which are connected respectively with single outputs of RS-flip-flops, except for n-th, outputs the anti-coincidence elements are connected to the inputs of the OR element, the output of which is connected to the input of the chronotron, the outputs of which are connected to one group of inputs of a permanent storage device, the other group of inputs of which are connected to the outputs of the encoder, and the outputs are connected to buses low-order device.