JP3332596B2 - Radiation measurement device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radiation measuring device,
In particular, the present invention relates to a radiation measurement device capable of detecting an abnormality even during operation of a radiation detector.

[0002]

2. Description of the Related Art A conventional radiation measuring apparatus will be described by taking a case where a radiation detector is a semiconductor detector as an example. The radiation spectrometer for measuring the energy spectrum of radiation is shown in the block diagram of FIG. This radiation spectrometer consists of a semiconductor detector 1, a preamplifier 2, a linear amplifier 3, an analog-to-digital converter 4 (Analog to Digi
tal Converter (hereinafter abbreviated as ADC), and a process memory 5, and a computer 6 and a spectrum display device 7 as necessary.

The signal waveforms shown in FIG.
The waveform of the output signal 8 of the detector 1 is shown in FIG.
Output signal 9 and the output signal of the linear amplifier 3
10 shows waveforms. The semiconductor detector 1 converts the radiation into an electric signal. At this time, the energy of the radiation incident on the detector 1 is finally converted into energy for generating an electron-hole pair inside the semiconductor detector 1. You.

Here, since the energy for generating one electron and hole pair is substantially constant, the number of generated electrons and hole pairs is substantially equal to the energy of incident radiation lost in the detector 1. . A high voltage called a bias is applied to the semiconductor detector 1, and electrons generated in the detector 1
The hole pairs move toward the positive and negative bias application electrodes.

Accordingly, electric charges are generated at the output of the detector 1 during a period from the generation of the electron and hole pairs to the arrival at the electrodes (charge collection time). The waveform of the output signal 8 of the detector 1 can be predicted from the shape and size of the detector, the electric field inside the detector, and the like.

[0006] The preamplifier 2 amplifies a minute electric signal output from the semiconductor detector 1 to amplify it to a voltage of an appropriate magnitude, and performs waveform shaping to output a low impedance electric signal. The total number of electron and hole pairs generated inside the semiconductor detector 1 is obtained by integrating the output signal 8 of the detector.

[0007] In order to realize the integration function using an electronic circuit, the electric charge output from the detector may be stored in a capacitor. The preamplifier 2 based on such a principle is called a charge preamplifier. Next, the basic circuit of the charge preamplifier is shown.
This is shown in the circuit diagram of FIG.

The basic circuit 11 of the charge type preamplifier includes an inverting amplifier 12 and a feedback capacitor 13. However, in the basic circuit shown in FIG. 45A, the output signal voltage rises every time the electric charge from the detector 1 is input, and eventually saturates near the power supply voltage of the preamplifier 2 and stops operating. .

In order to solve such a problem, the electric charge stored in the feedback capacitor 13 must be discharged under appropriate conditions. For this reason, in a general preamplifier, a resistor is connected in parallel with the feedback capacitor 13 to perform discharging. Such a preamplifier is called a resistance feedback preamplifier.

FIG. 45 (b) is a circuit diagram of the resistance feedback type preamplifier. The output signal of the resistance feedback type preamplifier 14 rises from the time when radiation is incident on the detector until the charge is collected. At the same time, the waveform exponentially decays. The time constant of the attenuation is determined by the feedback capacitor 16 and the feedback resistor in parallel with the inverting amplifier 15.
It is determined by the product of the power factor and 17 and is generally about 50 to 100 μs in the preamplifier used for the semiconductor detector 1.

The linear amplifier 3 performs waveform shaping for the purpose of removing noise from the output signal 9 of the preamplifier 2 and amplifies the input signal to a voltage within the conversion range of the ADC 4. Since the generation of radiation is random, the time intervals of the radiation incident on the semiconductor detector 1 are statistically distributed.

Therefore, when radiation is incident at short time intervals, the signal incident on the previous signal may overlap. This phenomenon is called pile-up, and energy information is not stored in the piled-up signal. In order to solve this problem, it is necessary to return the pulse signal generated by one radiation to the ground level as quickly as possible to reduce the probability of pile-up.

For this purpose, a CR differentiating circuit is provided at the first stage in the linear amplifier 3, and the time constant of 50 to 100 μs, which is the time constant of the output signal 9 of the preamplifier 2, is changed to 2 to 2.
After converting to a time constant of 5 μs and differentiating, integration is performed to reduce noise included in the output signal 9 of the preamplifier 2.

The differentiation / integration circuit built in the linear amplifier 3 is called a waveform shaper, and a waveform shaper called a Gaussian filter is generally used. FIG. 44 (c)
2 shows a waveform of an output signal 10 of a linear amplifier having a built-in Gaussian filter. The ADC 4 converts the peak portion of the waveform of the output signal 10 of the linear amplifier 3 into a digital value, and outputs the conversion result to the process memory 5.

The process memory 5 creates a histogram of the conversion result of the ADC 4 and stores it in a built-in memory. The horizontal axis of the obtained histogram corresponds to the conversion result of the ADC 4, and the vertical axis corresponds to the number of conversions. Usually, the horizontal axis is channel,
The vertical axis is called count, where the channel corresponds to the radiation energy and the count corresponds to the intensity of the radiation.

Therefore, when radiation of a single energy is incident on the detector 1, the conversion result of the ADC 4 becomes almost the same value, so that a spectral peak having a large specific energy intensity can be observed. However, due to the fact that the amount of charge generated inside the radiation detector is not always constant for single-energy radiation, and the influence of noise in the circuit system, the spectrum peak has a broad shape, The smaller the energy, the higher the ability (resolution) to discriminate between two radiations with close energy.

This resolution is an index for indicating the performance of the radiation spectrometer, and usually uses a spread (half width) at half the count value at the front end of the spectrum peak.

The computer 6 reads out the radiation energy spectrum recorded in the process memory 5 as needed, and performs analysis such as qualitative and quantitative radiation. Further, the spectrum display device 7 displays information recorded in the process memory 5 as a spectrum.

Conventionally, when the radiation detector 1 is diagnosed with respect to the above-mentioned radiation spectrometer, tests relating to leak current and resolution are performed. First, the leak current increases due to deterioration of the surface and the inside of the radiation detector 1 and is an important parameter for diagnosing the soundness of the semiconductor detector 1.

In the resistance feedback type preamplifier 14 shown in FIG. 45B, since the feedback resistance 17 is connected to the feedback circuit of the inverting amplifier 15,
The leak current from the detector 1 is converted into a voltage by a voltage drop of the feedback resistor 17 and can be detected as an output voltage of the resistance feedback preamplifier 14.

However, the output voltage of the preamplifier 14 is
Since the current is detected as the sum of the offset current of the inverting amplifier 15 and the leak current of the detector 1, a means for distinguishing the two is necessary. Since a leak current hardly flows unless a voltage is applied to the electrode of the semiconductor detector 1, an output voltage due to an offset of the inverting amplifier 15 is obtained by measuring an output voltage of the preamplifier 14 without applying a voltage. Can be.

The leak current of the semiconductor detector 1 itself can be detected by measuring the output voltage of the preamplifier when the voltage is applied to the semiconductor detector 1 with reference to the output voltage. Specifically, after the power of the preamplifier 14 is turned on with no radiation incident on the detector 1, the output voltage of the preamplifier 14 is measured using a measuring instrument such as a tester capable of measuring a DC voltage. While increasing the bias voltage of the detector 1 little by little.
Observe the change in the output voltage of 14.

Next, when measuring the resolution, the measurement is performed using an energy called a standard source and a source whose radiation emission rate is known. The resolution of a radiation spectrometer varies depending on the type of radiation detector, and the effects of noise and external noise in the circuit system. Can be measured. Therefore, the deterioration of the semiconductor detector 1 is diagnosed by periodically measuring the energy resolution by the standard radiation source.

[0024]

In a conventional method of detecting a leak current for a radiation detector in a radiation measuring apparatus, as shown in FIG. Since the output signal 18 of the preamplifier when the radiation is not incident on the semiconductor detector 1 is almost a DC signal, the leak current of the detector can be determined by measuring the DC voltage using a tester.

Further, as shown in FIG.
In other words, the output signal 19 of the preamplifier when the radiation enters the detector at a time interval sufficiently larger than the decay time constant of the preamplifier output signal, although the instruction of the tester may change intermittently, Measurement, the leakage current of the detector can be determined by measuring the DC voltage with a tester.

Further, at the time of the medium counting rate shown in FIG. 46 (c), that is, the output voltage which has risen rapidly due to the incidence of radiation,
Time to sufficiently decay (about 10 times the decay time constant)
With the output signal 20 of the preamplifier until the radiation enters the detector at the same time interval, it is difficult to measure the DC voltage with a tester because the DC component is small, so the waveform of the output signal 20 can be observed with an oscilloscope. Need to be done.

Since the interval at which the radiation enters the detector has a statistical variation, the radiation incident before the output voltage of the preamplifier is sufficiently attenuated and the radiation incident after the attenuation are reduced by the display of the oscilloscope. Observed above.

However, by properly adjusting the time axis of the oscilloscope and observing it carefully, it is possible to predict the level at which the output voltage from the preamplifier reaches after the attenuation, so that the leak current of the detector 1 is known. It was possible.

However, the output signal 21 of the preamplifier in the case of the high counting rate shown in FIG. 46D, that is, in the state where the probability that the radiation is incident at a time interval smaller than the time during which the output voltage of the preamplifier is sufficiently attenuated is high. However, since the output voltage of the preamplifier hardly attenuates sufficiently, even if it is observed using an oscilloscope, it is not possible to measure the level reached when the output voltage is sufficiently attenuated.

Therefore, in order to solve this problem, there has been proposed a method of estimating an attained level when the signal is sufficiently attenuated by an arithmetic circuit, assuming that the decay time constant of the output signal pulse of the preamplifier is known (see FIG. 1). Tokuhei 5-3549
Publication “True ground level estimation method”).

According to such a method, the leak current of the detector can be known even at a high counting rate, but the arithmetic circuit is complicated, it is difficult to adjust, and there is a problem in arithmetic processing accuracy and the like.

Regarding the deterioration diagnosis of the detector by the following resolution measurement, the purpose of the radiation spectroscopy is to qualitatively and quantitatively the radionuclide. That is, the purpose is to measure what radionuclides are contained and how much.

This means that if the type of radionuclide contained in the object to be measured is single or small, the radionuclide can be qualitatively and quantitatively determined using a detector with low resolution. If there are many types of nuclides, or even if the types of nuclides are small and the energy of the emitted radiation is close, the object cannot be achieved unless a detector with the highest possible resolution is used. As described above, in an application requiring a high resolution, a decrease in the resolution due to an abnormality in the detector cannot fulfill the original role of the radiation spectrometer.

Therefore, it is important to periodically check the resolution of the detector, but depending on the use of the radiation spectrometer, it is often the case that the periodic inspection can be performed only once a year. For such a case, it has been desired to be able to detect that the resolution has decreased during the operation of the detector.

Further, since the decrease in the resolution of the detector is information that can be detected only after performing data processing on the radiation spectrum stored in the process memory 5, when the decrease in the resolution is recognized, the measurement is performed up to that point. There was a problem that the radiation spectrum data had to be discarded.

The object of the present invention is to provide a diagnostic unit and a signal processing unit in a radiation measuring apparatus, to stop the measuring apparatus for inspection, or to perform a normal measurement without performing a resolution test using a standard source. An object of the present invention is to provide a radiation measuring device capable of diagnosing a radiation detector during operation.

[0037]

[0038]

In order to achieve the above object, a radiation measuring apparatus according to the first aspect of the present invention provides a radiation measuring apparatus.
Input the signal detected by the transmitter and converted to an electrical signal.
A radiation measurement device for measuring radiation, wherein the radiation
Eliminates the trajectory loss of the output signal of the line detector
A diagnostic unit for diagnosing the presence or absence of an abnormality in the output signal after the
Diagnosed as abnormal in the output signal by the diagnostic unit
In this case, a signal for determining whether the radiation detector is abnormal
And a processing unit for diagnosing a time interval by inputting an output signal of the radiation detector, extracting a rapidly changing portion of the signal after removing the trajectory defect and measuring an interval. In addition, the signal processing unit removes the preceding signal or the current signal when the time interval of the signal approaches, and removes the preceding signal and the current signal when the time interval approaches, based on the diagnosis result. And

In the radiation measuring apparatus according to the second aspect of the present invention, the signal detected by the radiation detector and converted into an electric signal can be used.
This is a radiation measurement device that inputs radiation signals and measures radiation.
The trajectory loss of the output signal of the radiation detector
The output signal after removing the ballistic defect
Diagnostic unit, and the output signal is abnormal in the diagnostic unit
If it is diagnosed, the presence or absence of abnormality of the radiation detector
A signal processing unit for determining
The diagnostic unit inputs the output signal of the radiation detector, extracts the rising portion of the signal after removing the trajectory defect, measures the rising speed, and diagnoses the rising speed. Detects the presence of a charge trapping area inside the detector and continues measurement, stops measurement as a detector error, or detects that radiation has entered during charge collection and removes the signal Alternatively, it is characterized in that the detection of the presence of a charge recombination region inside the detector removes the signal or stops the radiation measurement as an abnormality of the detector.

In the radiation measuring apparatus according to the third aspect of the present invention, the signal detected by the radiation detector and converted into an electric signal is provided.
This is a radiation measurement device that inputs radiation signals and measures radiation.
The trajectory loss of the output signal of the radiation detector
The output signal after removing the ballistic defect
Diagnostic unit, and the output signal is abnormal in the diagnostic unit
If it is diagnosed, the presence or absence of abnormality of the radiation detector
A signal processing unit for determining
The diagnostic unit inputs the output signal of the radiation detector, extracts the rising portion of the signal after removing the trajectory defect, measures the rising time, diagnoses the rising time, and based on the diagnosis result, the signal processing unit Detects the presence of a charge trapping area inside the detector and continues measurement, stops measurement as a detector error, or detects that radiation has entered during charge collection and removes the signal Alternatively, it is characterized in that the detection of the presence of a charge recombination region inside the detector removes the signal or stops the radiation measurement as an abnormality of the detector.

In the radiation measuring apparatus according to the fourth aspect of the present invention, the signal detected by the radiation detector and converted into an electric signal is provided.
This is a radiation measurement device that inputs radiation signals and measures radiation.
The trajectory loss of the output signal of the radiation detector
The output signal after removing the ballistic defect
Diagnostic unit, and the output signal is abnormal in the diagnostic unit
If it is diagnosed, the presence or absence of abnormality of the radiation detector
A signal processing unit for determining
The diagnostic unit receives the output signal of the radiation detector, extracts a flat portion of the signal after removing the trajectory defect, measures the slope of the flat portion, diagnoses the slope, and performs a signal processing unit based on the diagnosis result. Is characterized by detecting the abnormality sign of the detector and continuing the measurement or stopping the radiation measurement as an abnormality of the detector.

The radiation measuring apparatus according to the fifth aspect of the present invention is the radiation measuring apparatus according to the second or third aspect.
The diagnostic unit provided in the radiation measuring apparatus uses a complete differentiating circuit for signal processing after inputting the output signal of the radiation detector and removing a ballistic defect when measuring the rise speed and rise time of the signal. It is characterized by the following.

According to a sixth aspect of the present invention, there is provided a radiation measuring apparatus according to any one of the first to fifth aspects.
In the measuring device, the diagnostic unit and the signal processing unit provided in the radiation measuring device convert the analog signal from which the trajectory defect has been removed into digital time-series data and perform digital numerical processing.

[0044]

According to the first aspect of the present invention, in the diagnostic section, radiation is provided.
Eliminates the trajectory loss of the output signal of the line detector
Diagnosis of abnormalities in the output signal after leaving. further,
When the diagnostic unit diagnoses that the output signal is abnormal
Indicates whether the signal processing unit has an abnormality in the radiation detector.
judge. This allows the signal processing unit to remove the signal when the signal abnormality is minor and continue the radiation measurement, but it is also possible to stop the measurement depending on the degree of the abnormality.
You . In addition, this makes it possible to easily detect abnormality of the detector even during the radiation measurement operation.

[0045] In the diagnostic part, after removal of the ballistic defect on the output signal of the detector, to diagnose the time interval of rapid change of the signal, the stop signal is removed or the measurement of the signal processing unit from the results I do.

According to a second aspect of the present invention, the diagnosing unit removes a trajectory defect from the output signal of the detector, and then diagnoses the rising speed of the signal. When the number is small , the measurement is continued, and when the number is large, it is determined that the detector is abnormal.

When the next radiation is incident during charge collection, this signal is removed and removed from the object to be measured. If the rise speed is faster and there is a recombination region of the charge, but the signal generation is small , remove this signal and continue the measurement.
When the number of signals is large, it is determined that the detector is abnormal.

According to a third aspect of the present invention, the diagnosing unit removes a trajectory defect from the output signal of the detector and diagnoses the rise time of the signal. When the occurrence is small , the measurement is continued, and when the occurrence is large, it is determined that the detector is abnormal.

When the next radiation enters during charge collection, this signal is removed and removed from the object to be measured. When the rise time is short and the signal generation is small, the signal is removed and the measurement is continued. However, when the signal is large, it is determined that the detector is abnormal.

According to a fourth aspect of the present invention, after the diagnosing unit removes the trajectory defect from the output signal of the detector, the inclination of the flat portion of the output voltage of the trajectory defect removing circuit is diagnosed.
From this result, an increase in the slope is regarded as an abnormality of the detector. If the increase is slight, the measurement is continued, and if the increase is large, it is determined that the detector is abnormal.

According to a fifth aspect of the present invention, a signal obtained by removing a trajectory defect from an output signal of a radiation detector is pulsed by using a complete differentiating circuit to measure a rise speed and a rise time of the signal. Since the effect of extracting the portion where the signal rapidly changes is good, the accuracy of signal detection is improved.

According to the sixth aspect of the present invention, since the diagnosis unit and the signal processing unit after removing the trajectory defect from the output signal of the radiation detector are digitized, circuit construction and integration are facilitated, and the processing speed is increased. be able to.

[0053]

An embodiment of the present invention will be described with reference to the drawings. The same components as those of the above-described conventional technology are denoted by the same reference numerals, and detailed description is omitted. As shown schematically in the block diagram of FIG. 1, the radiation measuring apparatus includes a detector 1, a preamplifier 2, a diagnostic unit 22, and a signal processing unit for a diagnostic result.
23, a linear amplifier 3 and an ADC 24, a process memory 5 and a computer 6, and a display device 25 for displaying a spectrum, a detector abnormality, and the like.

In the first invention, the diagnosis is made by measuring the interval of the rapid change of the detector output signal. As shown in the block diagram of FIG. , An interval measurement 28 of rapid change, and a time interval diagnosis 29, and outputs a diagnosis result A30. The signal processing unit for the diagnosis result A30 is configured by signal processing of removal 31 of the preceding signal or current signal and removal 32 of the preceding signal and current signal.

In the second invention, the diagnosis is made by measuring the rising speed of the detector output signal. As shown in the block diagram of FIG. 3, the diagnosis section removes the trajectory defect 26, extracts the rising portion 33, and sets the rising portion. Measuring speed 34, and diagnosis of rise speed
It outputs a diagnosis result B36 composed of 35 elements.

The diagnosis result B36 is based on the judgment 38 that there is a charge trapping region inside the detector and the judgment result 38 with respect to the signal diagnosed as the signal 37 whose rising speed of the signal processing unit is slow. Based on measurement continuation 39 and signal generation 40 indicating that the detector is abnormal, similarly to the diagnosis that the rising speed is a slow signal 37, judgment 41 that the next radiation has entered during charge collection, The signal removal signal 42 based on this determination result
Generation and the rise rate of the signal 43 is determined to be a signal recombination region inside the detector 44
And a signal processing for generating a signal removal signal 45 based on the determination result and a signal 46 indicating that the detector is abnormal.

[0057] The third invention, detector diagnosed by measurement of the output signal rise time, the diagnosis unit shown in block diagram in Figure 4, removal of the ballistic defect 26 and the rising portion of the extraction 33,
It comprises a rise time measurement 47 and a rise time diagnosis 48 and outputs a diagnosis result C49.

Based on the diagnosis result C49, it is determined that the signal 50 has a long rise time in the signal processing section 50, and a determination 51 that there is a charge trapping region inside the detector and a measurement continuation 52 based on the determination result are generated. The signal 53 indicating that the detector is abnormal is generated, the signal 50 is diagnosed to be a signal 50 having a long rise time, and it is determined that the next radiation is incident during the charge collection, and the signal is removed based on the determination result. It is determined that the signal 55 has a short rise time and that the signal has a recombination region inside the detector 57, and the signal is removed based on the determination result.
Each signal processing includes generation of 58 and generation of a signal 59 indicating that the detector is abnormal.

The fourth invention is a diagnosis based on the measurement of the inclination of the flat portion of the detector output signal. As shown in the block diagram of FIG. 5, the diagnosis section removes the trajectory defect 26 and extracts the flat portion 6
0, which is composed of the elements of the measurement 60 of the inclination of the flat portion and the diagnosis 62 of the inclination, and outputs a diagnosis result D63.

The diagnosis result D63 is based on the result of the diagnosis 64 indicating that the inclination of the signal processing section has increased, the determination 65 of the sign of abnormality of the detector, the occurrence of the measurement continuation 66 based on the determination result, and the abnormality of the detector. It is constituted by signal processing of generation of a signal 67 indicating that there is.

The operation of the structure according to the first invention is as follows. First, regarding the elimination 26 of the trajectory defect, the influence of the trajectory defect and its removal are described in the basic circuit 11 of the charge type preamplifier shown in FIG. When the output signal 68 of the square wave detector is input as shown in FIG. 6A, the waveform of the output signal 69 of the charge type preamplifier basic circuit 11 rises linearly as shown in FIG. Then, the voltage reaches an output voltage E proportional to the area Q of the square wave.

The output signal 70 of the detector having a different peak value and pulse width from the output signal 68 of the detector and having the same area is used.
, The output signal 71 of the basic circuit 11 of the charge preamplifier also reaches the output voltage E.

Therefore, in the basic circuit 11 of the charge type preamplifier shown in FIG. 45A, even if the output waveform of the detector changes, the area is constant, that is, the energy of the radiation incident on the detector 1 is constant. If it is, the output voltage E is always obtained.

However, in practice, a resistance feedback type preamplifier 14 shown in FIG. 45B is often used as a radiation detector preamplifier. In the resistance feedback type preamplifier 14, the feedback capacitor 16
At the same time, the electric charge from the detector 1 is accumulated, and at the same time, the discharge by the feedback resistor 17 is performed.

Therefore, the output of the resistance feedback type preamplifier 14 with respect to the output signal 68 of the detector is shown in FIG.
As shown by the output signal 72 of (c), the attenuation starts before the voltage E is reached.

Further, the output signal 73 of the resistance feedback type preamplifier 14 for the output signal 70 having the same area as the detector output signal 68 and the long pulse width is more than that when the output signal 68 having the short pulse width is input. Start decay from low voltage. In this way, a portion that does not reach the voltage E to be originally reached is called a ballistic defect 74.

In an actual detector, even if radiation having the same energy is incident, the output signal waveform of the detector often does not become the same depending on the incident position, so that the trajectory defect 74 has a large effect on the energy resolution. Furthermore, preamplifier 2
Since the output signal 9 is normally differentiated into a time constant of about several μs at the first stage of the linear amplifier 3, the decrease in energy resolution due to the trajectory defect 74 becomes remarkable.

In order to eliminate the influence of this ballistic defect, the output of the resistance feedback type preamplifier 14 shown in FIG. It is sufficient that there is a circuit for converting the signal waveform into the same signal waveform as the basic circuit 11 of the charge preamplifier shown in FIG. Such a circuit is hereinafter referred to as a ballistic loss removing circuit, and its basic circuit and input / output waveforms are shown in FIG.

When an output signal 76 of the exponentially attenuating resistance feedback type preamplifier 14 shown in the signal waveform diagram of FIG. 7B is input to the ballistic loss removing circuit 75 shown in the circuit diagram of FIG. The waveform of the output signal 77 of the removal circuit 75 has a step shape.

In this state, as in the basic circuit 11 of the charge type preamplifier shown in FIG. 45A, the output signal saturates to the power supply voltage of the preamplifier and does not operate.
A reset input 78 is provided in the trajectory defect removing circuit 75 shown in (a), and plays a role of rapidly discharging the electric charge accumulated in the feedback capacitor 79.

The input / output signals of the trajectory defect removing circuit 75 are signals whose signs are inverted by the inverting amplifier 80. The direction that changes when radiation is incident on the detector is expressed as positive or rising. An input resistor 81 is connected to the input side of the inverting amplifier 80, and a feedback resistor 82 is connected to the feedback capacitor 79.

Since the trajectory defect removing circuit 75 converts the waveform of the output signal to the same as that of the basic circuit 11 of the charge type preamplifier shown in FIG. 45 (a), the trajectory defect generated by using the resistance feedback type preamplifier is removed. In addition to this, it has an effect of removing signal distortion generated by using the resistance feedback type preamplifier. Therefore, an accurate diagnosis result can be obtained when the diagnosis is performed on the signal waveform of the detector.

The characteristic diagram of FIG. 8 shows the extraction timing signals of the signals in extraction 27 of the rapidly changing portion in FIG. 2, extraction 33 of the rising portion in FIGS. 3 and 4, and extraction 60 of the flat portion in FIG. Show.

The input signal waveform 76 shown in FIG. 8A rises every time radiation enters the detector, and has a time constant determined by the feedback resistance and feedback capacitor of the preamplifier, and attenuates exponentially. The waveform of the output signal 77 of the ballistic loss removing circuit 75 of FIG. 7A shown in FIG. 8B rises every time radiation enters the detector, and is reset before the output signal is saturated with the power supply voltage. It is.

The extraction timing signal 83 of the abruptly changing portion shown in FIG. 8E is obtained from the waveform of the output signal 77 of the trajectory defect removing circuit 75 due to the radiation incident on the detector shown in FIG. This is obtained by calculating the logical sum of the change timing signal 84 and the sudden change timing signal 85 due to the reset of the trajectory defect removal circuit 75 in FIG.

The extraction timing signal 86 of the flat portion shown in FIG. 8F is used to remove a sudden change from the waveform of the output signal 77 of the trajectory defect removal circuit 75, that is, the extraction timing signal 83 of the sudden change is logically calculated. It is an inverted one. Also,
The rising portion extraction timing signal 86 is the same as the rising portion from the waveform of the output signal 77 of the trajectory loss removing circuit 75, that is, the same timing signal 84 as the timing signal 84 of a sudden change due to the incidence of radiation on the detector.

FIG. 8 (g) shows the details of the rising edge extraction timing signal. The signal 77 and the signal 84 are shown in an enlarged manner. The pulse width of the timing signal 84 which changes abruptly due to the radiation incident on the detector is shown. Is the same width as the rise time of the output signal 77 after removal of the trajectory defect, and
The pulse width of the timing signal 85 of an abrupt change due to the reset is the same as the fall time of the reset of the output signal 77 after removing the trajectory defect.

First, the operation by the diagnosis of the interval of the rapid change shown in FIG. 2 and the signal processing based on the result of the diagnosis are shown in FIG.
In the waveform of the output signal 77 of the ballistic defect removing circuit 75 shown in (b), the energy of the radiation incident on the detector is obtained by measuring the rising height 87 generated each time the radiation is incident on the detector. Can be.

However, in order to improve the energy resolution, the noise included in the output signal 77 of the trajectory defect removal circuit 75 must be reduced by a suitable filter.
Specifically, it is necessary to obtain the rising height 87 in a state where the noise components of the flat portion 88 before the rising and the flat portion 89 after the rising are reduced.

As described above, it takes time to remove the noise component before and after the rise, and when radiation enters the detector at intervals that cannot secure this time, it is removed from the object to be measured. Is performed.

FIG. 9 shows in detail how the signal approaches the characteristic diagram of FIG. Conditions to determine whether to remove from the object of measurement is required to determine the height of each of the preceding rise 90 and the current rise 91, the time 92 required to remove the noise component of the flat portion before the rise, It is determined by the magnitude relation with the time 93 required for removing the noise component of the flat portion after the rise.

When the noise removal time 92 before the rise is shorter than the noise removal time 93 after the rise shown in FIG. 9A, when the signal approaches, first, the noise in the flat portion after the rise of the preceding rise 90 Since the time 90a required for removal cannot be secured, the preceding signal is removed from the measurement target. When approaching further, the time 93b required for noise removal of the flat portion before the current rising 91 cannot be secured, so that the current signal is also removed from the measurement target.

If the noise removal time 92 before the rise shown in FIG. 9B is longer than the noise removal time 93 after the rise, when the signal approaches, first the noise removal of the flat part before the rise of the current rise 91 is performed. Since the time 93b required for the measurement cannot be secured, the current signal is removed from the measurement target. When the distance further approaches, the time 93a required for noise removal of the flat portion after the leading rise 90 cannot be secured, so that the leading signal is also removed from the measurement target.

When the noise removal time 92 before the rise and the noise removal time 93 after the rise shown in FIG. 9C are equal to each other, the time required for removing the noise component after the rise of the preceding rise 90 when the signal approaches. Since the time 93a required for removing the noise component before the rising of the current rising 91 cannot be secured at the same time, both the preceding signal and the current signal are removed.

If a sudden change occurs due to the reset of the ballistic defect removing circuit 75, the energy information of the radiation incident on the detector immediately before or immediately after the reset operation may be lost. Remove signals during reset and before and after reset.

Further, despite the low frequency of occurrence of the timing signal 84 of the rapid change shown in FIG. 8C due to the incidence of radiation on the detector, the ballistic defect removal circuit 75 of FIG. If the frequency of occurrence of the timing signal 85 of a sudden change due to reset is high, it can be clearly determined that the leak current of the detector has increased. Therefore, the determination interval for stopping the measurement, replacing the detector, and the like can be obtained based on the interval between the timing signal 84 and the timing signal 85.

As described above, the signal processing shown in FIG. 2 is for removing a signal based on the above-described principle, performing an interval measurement 28 of a rapid change, and setting the signal interval to a time interval necessary for noise component removal. Diagnosing time intervals for greater than
It operates according to the result of 29.

FIG. 10 is a characteristic diagram showing a state of signal processing for a diagnosis result at time intervals. The radiation energy spectrum 94 when the signal processing for the diagnosis result A30 is not performed is
The resolution of the spectrum peak portion is poor, and the radiation shape is detected even in the energy portion which does not originally exist.

On the other hand, the radiation energy spectrum 95 obtained when the signal processing is performed on the diagnosis result A30 has a good energy resolution at the spectrum peak portion and no radiation is counted at the energy portion where no radiation exists. Measurement of a radiation energy spectrum close to the spectrum is possible.

With respect to the configuration shown in FIG. 3 of the second invention, the operation relating to the diagnosis of the rising speed and the signal processing based on the diagnosis result will be described. When radiation is incident on the detector, charges are generated inside the detector, and the output signal of the preamplifier continues to rise until the generated charges move to the bias electrode and are collected by the bias electrode.

The time until the generated charges are collected by the bias electrode is called a charge collection time, and the charge collection time is determined by the electric field in the path along which the charges move. Therefore, once the shape, size, and bias voltage of the detector are determined, the charge collection time can be obtained as a function of the incident position of the radiation.

Assuming that the charge collection time is within a range determined by the incident position of the radiation, the rise rate of the signal is measured.
34, it is possible to predict the attained voltage after the elapse of the charge collection time, and compare the predicted attained voltage with the actual attained voltage to determine whether the actual signal rise speed is slow or fast. You can do speed diagnosis 35.

That is, when the actual attained voltage is higher than the predicted attained voltage, it is slower than the original rising speed,
If the actual reach voltage is lower than the predicted reach voltage,
It can be determined that it is faster than the original rising speed.

As shown in FIG. 3, when the diagnostic result B36 in the diagnostic section has a slow signal rising speed 37, there is a charge trapping region 38 inside the detector, or the next radiation enters during charge collection. If there is a possibility of 41 and the rising speed of the signal is 43, it is considered that there is a charge recombination region 44 inside the detector.

When there is a charge trapping region inside the detector 38
Since energy generated inside the detector is captured and re-emitted after moving to the bias electrode, energy information is stored. Therefore, the voltage after the rise reaches a voltage proportional to the energy, but it takes time to reach the voltage.

As shown in FIG. 6, the output waveforms 72 and 73 of the resistance feedback type preamplifier 14 cause a ballistic defect 74 due to a change in the output signal waveform of the detector, but the output waveforms 69 and 71 of the charge type preamplifier basic circuit 11. That is, the output waveform of the trajectory defect removing circuit 75 can obtain a voltage proportional to the energy irrespective of a change in the output signal waveform of the detector.

Therefore, if the energy information of the signal from the detector is stored, it is possible to continue the measurement 39 even if the rising speed is low. If the capture region is part of the detector, the frequency of the slow rise signal is low, but if the capture region extends over the entire detector, the frequency of the slow rise signal will increase. .

Therefore, when the frequency of occurrence of a signal having a slow rising speed tends to increase, it is possible to perform processing such as stopping the measurement or replacing the detector as an abnormality 40 of the detector. When the next radiation is incident 41 during charge collection, the energy information is lost, so the signal is removed 42 and excluded from the measurement target.

In the signal 37 having a slow rising speed, a case 38 in which a charge trapping region exists inside the detector and a case 41 in which the next radiation enters during the charge collection are distinguished by the radiation counting rate. The former occurrence frequency does not depend on the counting rate, and the latter occurrence frequency depends on the counting rate. Therefore, when a signal having a slow rising speed is observed at a low counting rate, it can be determined that there is a charge trapping region inside the detector.

When there is a charge recombination region inside the detector
Since the energy information is not stored because the moving charges are combined and no longer exists, the signal 44 is removed from the measurement target by removing the signal 45.

When the recombination region is a part of the detector, a signal with a fast rise speed is less frequent, but when the recombination region is spread over the entire detector, the signal with a fast rise speed is small. When the frequency of occurrence of a signal having a fast rise speed tends to increase, it is possible to determine that the detector is abnormal 46 and perform processing such as measurement stop or replacement of the detector. As described above, the abnormality inside the detector can be detected based on the diagnosis result B36, and the determination for stopping the measurement, replacing the detector, or the like can be performed.

FIG. 11 is a characteristic diagram showing the operation of the signal processing for the diagnosis result of the rise speed. In the case where the next radiation is incident 41 during the charge collection shown in FIG. 3, the signal removal 42 is performed to remove the sum peak 98 caused by the simultaneous incidence on the energy portions of the spectrum peak 96 and the spectrum peak 97. Is possible.

In addition, it is possible to remove the swell 99 on the low energy side which occurs when the charge recombination region exists inside the detector. Thus, even if it is determined that there is an abnormality inside the detector, it is diagnosed whether each signal is normal or abnormal, and normal energy spectrum measurement is continued by removing the abnormal signal. be able to.

The operation relating to the diagnosis of the rising speed and the signal processing based on the result of the diagnosis shown in FIG.
When radiation is incident on the detector, charges are generated inside the detector, and the output signal of the preamplifier continues to rise until the generated charges move to the bias electrode and are collected by the bias electrode. The time until the generated charge is collected by the bias electrode is called a charge collection time, and the charge collection time is determined by the electric field of the path along which the charge moves.

Therefore, once the shape, size, and bias voltage of the detector are determined, the charge collection time can be predicted as a function of the incident position of the radiation. By measuring the rise time of the signal 47 and comparing it to the expected value,
Whether the rise time of the actual signal is long or short can be diagnosed 48 as the rise time.

As shown in FIG. 4, when the rise time of the signal is long based on the diagnosis result C49, if the signal capture region is inside the detector 51, or there is a possibility that the next radiation has entered during the charge collection. In the case where the rise time of the signal is short 56, it is considered that there is a charge recombination region 57 inside the detector.

When there is a charge trapping region inside the detector 51
Since energy generated inside the detector is captured and re-emitted after moving to the bias electrode, energy information is stored. Therefore, the voltage after the rise reaches a voltage proportional to the energy, but it takes time to reach the voltage.

As shown in FIG. 6C, the output waveforms 72 and 73 of the resistance feedback type preamplifier 14 cause a ballistic defect 74 due to a change in the output signal waveform of the detector. Regarding the output waveforms 69 and 71, that is, the output waveform of the ballistic loss removing circuit 75, a voltage proportional to the energy can be obtained regardless of a change in the output signal waveform of the detector. Therefore, if the energy information of the signal from the detector is stored, the measurement can be continued even if the rise time is long.

When the capture region is a part of the detector, the signal 50 with a long rise time occurs less frequently, but when the capture region extends over the entire detector, the signal 50 with a long rise time occurs. More frequently.

Therefore, when the frequency of occurrence of a signal having a long rise time tends to increase, it is possible to perform processing such as stopping the measurement or replacing the detector as a detector abnormality 53. If the next radiation enters during the charge collection, the energy information is lost, so the signal is removed 55 and excluded from the measurement target.

In the signal 50 having a long rise time, a case 51 in which a charge trapping region exists inside the detector and a case 54 in which the next radiation enters during the charge collection are distinguished by the radiation counting rate. The former occurrence frequency does not depend on the counting rate, and the latter occurrence frequency depends on the counting rate. Therefore, when the signal 50 having a long rise time is observed at the low counting rate, it can be determined that the signal capture region 51 exists inside the detector.

When there is a charge recombination region inside the detector
In 57, since energy information is not stored because the moving charges are combined and no longer exists, the signal is removed 58 and excluded from the measurement target.

When the recombination region is a part of the detector, the frequency of the signal 56 having a short rise time is low, but
When the recombination region is spread over the entire detector, the frequency of occurrence of the signal 56 having a short rise time increases.Therefore, when the frequency of occurrence of the signal 56 having a short rise time tends to increase,
When it is determined that the detector is abnormal 59, processing such as measurement stop or detector replacement can be performed. As described above, an abnormality in the detector can be detected based on the diagnosis result C49, and a determination for performing processing such as measurement stop or replacement of the detector can be performed.

FIG. 11 shows the effect of signal processing on the rise time diagnosis result. When the next radiation enters during the charge collection shown in FIG. 4, 54 is performed by removing the signal.
It is possible to remove the sum peak 98 caused by simultaneously entering the energy portions of the spectrum peak 96 and the spectrum peak 97.

In addition, it is possible to remove the swell 99 on the low energy side, which occurs when there is a charge recombination region inside the detector. Thus, even if it is determined that there is an abnormality inside the detector, it is diagnosed whether each signal is normal or abnormal, and normal energy spectrum measurement is continued by removing the abnormal signal. be able to.

The operation relating to the diagnosis of the inclination of the flat portion and the signal processing based on the diagnosis result shown in FIG. 5 in the fourth invention is as follows. In the ballistic loss removing circuit 75 shown in FIG. It will flow through the input resistance 81.

Since the current flowing through the input resistor 81 also flows through the feedback circuit, a current having the same value also flows through the feedback resistor 82, so that the charge is continuously stored in the feedback capacitor 79, and the output voltage leaks. It changes according to the amount of current. Therefore, the output voltage 77 of the flat portion, which should be constant, changes with a slope corresponding to the amount of leakage current. Therefore, measurement of the inclination of the flat part
By performing 61, the leak current of the detector can be measured during the operation of the detector.

Since the generation of radiation is a random event, the extraction timing signal of the flat portion shown in FIG.
The 86 high-level pulse widths are statistically distributed. Even when the counting rate is high, it is possible to obtain a pulse width enough to measure the inclination of the flat portion, and therefore, it is possible to detect the leak current of the detector even in a situation where the operation is performed at a high counting rate.

Although the detector originally has a small leak current, if the leak current tends to increase, it can be determined that the detector is an abnormal sign 65. By measuring the leak current of the detector during operation, if the abnormality sign 65 is minor, the measurement is continued 66, and if the amount of the leak current further increases, it is determined that the detector is abnormal 67 and the measurement is stopped. Processing such as stopping or replacing the detector can be performed.

The first embodiment is based on the analog signal processing system, and is shown in the block diagram of FIG. FIG. 12 collectively shows the diagnostic units of the radiation measuring apparatus shown in FIGS. 2 to 5 for convenience of explanation. In the diagnostic section of this radiation measuring apparatus, the trajectory defect removal 26 shown in FIG. 2 of the first invention is a trajectory defect removal circuit 100, and the rapid change portion extraction 27 is a rapid rise detection circuit 101 and a rapid rise detection circuit. 102,
The rapid change interval measurement 28 corresponds to the time interval measurement circuit 103, and the time interval diagnosis 29 corresponds to the time interval diagnosis circuit 104, and a diagnosis result A30 is obtained.

The trajectory defect removal 26 shown in FIG. 3 of the second invention is a trajectory defect removal circuit 100, a rising portion extraction 33 is a sharp rise detection circuit 101, a rise speed measurement 34 is a rise speed measurement circuit 105, and a rise speed. Diagnosis 35 is the rise speed diagnostic circuit
A diagnosis result B36 is obtained corresponding to 106.

The trajectory defect removal 26 shown in FIG. 4 of the third invention 26 is a trajectory defect removal circuit 100, a rising part extraction 33 is a sharp rise detection circuit 101, a rise time measurement 47 is a rise time measurement circuit 107, and a rise time. Diagnostic 48 is a rise time diagnostic circuit
A diagnostic result C49 is obtained corresponding to 108.

The trajectory defect removal 26 shown in FIG. 5 of the fourth invention is a trajectory defect removal circuit 100, and the extraction of a flat portion 60 is a logical sum 109 of the rapid rise detection circuit 101 and the rapid fall detection circuit 102, and a flat portion. 61 is a flat part tilt measurement circuit
In step 110, the inclination diagnosis 62 obtains a diagnosis result D63 corresponding to the flat portion inclination diagnosis circuit 111. These diagnostic results A to D are mainly subjected to signal processing by an analog circuit.

As the basic circuit of the trajectory defect removing circuit 100, the trajectory defect removing circuit 75 shown in FIG. 7A is used. Figure
13 (a) is a circuit diagram, and FIG. 13 (b) is a characteristic diagram, showing a reset method of the ballistic loss removing circuit and signal waveforms of various parts.

The basic circuit of the trajectory defect removing circuit 100 includes an inverting amplifier 80, an input resistor 81, a feedback resistor 82, and a feedback capacitor 79. When an input signal 112 composed of a pulse train that attenuates exponentially is input to the trajectory defect removing circuit 100, a waveform 114 of an output signal 113 that rises in a step-like manner every time a pulse is input is obtained. Ballistic defect removal circuit when pulse signal is continuously input
The output 113 of 100 rises to near the power supply voltage of the inverting amplifier 80, saturates, and does not operate.

The comparator 115 has a reference power supply Vr116 and a resistor
Reference voltage 119 and output signal 11 determined by 117 and resistor 118
The switch A120 is turned on and the current source is turned on before the voltage of the output 113 rises to near the power supply voltage.
121 is connected to the input terminal of the inverting amplifier 80. This current source
Since the current flowing through 121 also flows through the feedback circuit of the inverting amplifier 80, the electric charge stored in the feedback capacitor 79 is discharged.

As a means for discharging the electric charge stored in the feedback capacitor 79, a method of connecting the switch B122 in parallel with the feedback capacitor 79 as shown by a dotted line in FIG.
Since the output voltage range of 80 is narrower than when switch A120 is used, there are restrictions on use.

As the switch A120, a transistor,
An FET or a dedicated switch IC can be used. If necessary, the signal level conversion between the output voltage range of the comparator 115 and the control input range of the switch A120 is performed.
123 is required.

When the switch A120 is turned on, the voltage of the output 113 falls. However, if the switch A120 is kept as it is, the voltage drops to near the power supply voltage of the inverting amplifier 80, saturates, and does not operate. No. In order to automatically perform such a control operation of the switch A120, the reference voltage 119 of the comparator 115 is set to a reference power supply.
Hysteresis is set by Vr 116, resistor 117, and resistor 118.

That is, the waveform 114 of the output signal shown in FIG.
As shown by the dotted line in the figure, when the output waveform 114 of the ballistic loss removing circuit 100 rises stepwise and reaches the high-level hysteresis voltage 124, the output voltage of the comparator 115 falls and turns on the switch A120, Reference voltage 119
Transitions to a low level hysteresis voltage 125.

When the electric charge accumulated in the feedback capacitor 79 is discharged and the output waveform 114 of the ballistic loss removing circuit 100 reaches the low-level hysteresis voltage 125, the output voltage of the comparator 115 rises, turning off the switch A120 and turning off the switch A120. , The reference voltage 119 returns to the high-level hysteresis voltage 124 again.

The control timing signal 126 for the switch A120 is a timing signal for turning on and off the switch A120, and can be used as a timing signal for the sharp fall detection circuit 102.

FIG. 14A is a block diagram of the signal detection circuit. Output signal of the ballistic loss removal circuit 100 shown in FIG.
In order to detect a signal from the waveform 114, a pulse 127 for separating a rapidly changing portion is first formed, and then a signal shaping 128 for using a signal detection result as a timing signal in another circuit is performed. I do. Normally, waveform shaping is performed to match the input level of a standard logic element such as TTL.

The circuit diagrams shown in FIGS.
FIG. 14 (b) shows pulsing by a CR differentiating circuit, FIG. 14 (c) shows pulsing by an active filter, and both output signal waveforms 129 suddenly change to an input signal waveform 130. Outputs a pulse that decays exponentially when a change occurs.

FIG. 14D shows the case of pulse formation by a complete differentiating circuit. When a sudden change occurs in the input signal waveform 132, the output signal waveform 131 outputs a pulse having a peak value corresponding to the change speed. I do. Further, the pulse width of the output signal waveform 131 indicates a time during which the input signal waveform 132 is rapidly changing.

FIG. 14E shows pulsing by the delay line (DL), and the output signal waveform 133 is the input signal waveform 13.
4 outputs a pulse having a peak value corresponding to the changed voltage when a sudden change occurs. The output signal waveform 13
The pulse width of 3 depends on the delay time of the delay line.

Next, the circuit of the waveform shaping 128 and its description will be described with reference to FIG.
(A) to (d). FIG. 15A is a waveform shaping circuit diagram of a rising signal, in which a comparator 135 supplies a positive reference voltage Vr + 13
6 and the voltage level of input signal 137 is positive reference voltage.
Output 138 goes high when Vr + 136 is exceeded.

FIG. 15B is a waveform shaping circuit diagram of the falling signal, in which a negative reference voltage Vr-140 is connected to the comparator 139 and the voltage level of the input signal 141 is a negative reference voltage Vr-140.
The output 142 goes high when it is less than

The positive or negative reference voltage needs to be set to a value close to the base line to detect a signal as small as possible. However, the CR differentiating circuit shown in FIGS. Output signal waveform of pulse by active filter
Since 129 is an exponentially attenuating pulse, the pulse width after waveform shaping becomes longer when the reference voltage is reduced.

This not only reduces the time resolution of the pulse, but also as shown in the waveform shaping of the output signal of the CR differentiating circuit in FIG. 15C, the pulse 144 before the preceding pulse 143 is sufficiently attenuated. There is one pulse 1 even though two pulses must be output.
Only 43a can be output. Therefore, it is desirable that the circuit for pulsing has a pulse output having no attenuation as shown in FIGS. 14 (d) and 14 (e).

In the pulse formation by the complete differentiating circuit shown in FIG. 14D, since the peak value of the output signal 131 is proportional to the rising speed of the input signal 132, the same ultimate voltage as shown in FIG. Even if the input signal has a high rise speed, the peak value of the output signal is high for the input signal 145 having a fast rising speed, and the peak value 148 of the output signal is low for the input signal 147 having a slow rising speed. Occurs.

That is, even if the signals have the same energy, the signal detection sensitivity differs depending on the rise speed. Therefore, when the purpose is to detect a signal, pulsing by the delay line (DL) in FIG.

However, in the pulse formation by the delay line shown in FIG. 14E, since the pulse width depends on the delay time of the delay line, it is not possible to know the time of the rapid change. On the other hand, in the pulsing by the fully differentiating circuit in FIG. 14D, a positive or negative signal is output while the input signal is changing, so that the time during which the signal changes rapidly can be known.

That is, when the purpose is to extract a portion that changes rapidly, pulsing by a complete differentiating circuit shown in FIG. 14D is used. The output signal of the trajectory defect removal circuit 100 accurately reflects the output signal waveform of the detector or has a rising edge, like the output waveforms 69 and 71 of the charge type preamplifier shown in FIG. The result of pulsing by the complete differentiating circuit shown in FIG. 14 (d) can reproduce a waveform similar to the output signal waveform of the detector.

FIG. 16 shows (a) a circuit diagram corresponding to the abrupt rising detection circuit 101 shown in FIG. 12 and (b) a signal waveform diagram. An input waveform 153 is obtained by differentiating the input signal 149 changing in a step-like manner by a complete differentiating circuit constituted by the inverting amplifier 150, the input capacitor 151, and the feedback resistor 152.

The output pulse crest value of this completely differentiating circuit is
When using a radiation detector in which the rising speed changes greatly because it is proportional to the rising speed of the input signal 149, the inverting amplifier 154, the input resistor 155, and the feedback resistor are used.
156 to improve the signal detection sensitivity.

The waveform of the output 157 of this amplifying circuit is obtained by amplifying a pulse having a small peak value, and the waveform can be easily shaped by the comparator 158. The waveform of the output 159 of the comparator 158 corresponds to the output voltage 157 of the amplifier circuit.
Is higher than the positive reference voltage Vr + 160, a high level pulse train is obtained. Each pulse width of the waveform of the output 159 of the comparator 158 is the rise time of the input signal 149.

FIG. 17 shows (a) a circuit diagram corresponding to the abrupt falling detection circuit 102 shown in FIG. 12 and (b) a signal waveform diagram. An output 165 is obtained by differentiating the input signal 161 by a complete differentiating circuit including an inverting amplifier 162, an input capacitor 163, and a feedback resistor 164.

This complete differentiating circuit can be used in common with the rise detecting circuit. However, when the rising speed and the falling speed are significantly different, the rising detection circuit is used to obtain an output waveform optimal for the falling speed. It is better to use a completely differentiating circuit independent of the circuit.

The output 167 of the comparator 166 is an inverting amplifier.
It goes high when the output voltage 165 at 162 is less than the negative reference voltage Vr-168. Since the sharp fall occurs due to the reset of the ballistic loss removing circuit 100, the fall speed is almost constant, and the amplitude is larger than the sharp rise, so an amplification circuit such as a rise detection circuit is used. No need.

However, in the trajectory defect removing circuit 100 shown in FIG.
Since the output signal 114 has detected that the output signal 114 has exceeded the reset start level, there is a delay time from when the switch A120 is actually turned ON and the discharge of the feedback capacitor 79 is started. The true reset start time cannot be obtained by the method of detecting the falling part of the output waveform of the circuit.

Therefore, in the sharp fall detection circuit,
Output signal 167 of comparator 166 and trajectory defect removal circuit 10
After calculating the logical sum 170 with the switch control timing signal 169 of 0, the output signal 171 is output.

FIG. 18 shows the time interval measuring circuit 10 shown in FIG.
(A) Circuit diagram and (b) signal waveform diagram corresponding to FIG. In the circuit that constantly stores the charge in the capacitor 173 from the constant current source 172, the charge in the capacitor 173 is discharged using the switch 174 each time a pulse is input, so that the time until the next pulse is input is reduced. Capacitor 17
3 is obtained as the voltage between both ends.

In this case, if the time during which the switch 174 is ON is constant, the voltage between both ends of the capacitor 173 and the pulse interval are proportional. The time during which the switch 174 is ON is not only a constant term when the pulse interval is calculated from the voltage between both ends of the capacitor 173, but also becomes the time resolution of the pulse interval measurement. .

Therefore, a sudden rising detection pulse train
175 and the CR integration circuit after inverting this pulse train 175
The pulse train 178 after the pulse width shaping is extracted by the logical product 177 with the signal delayed by 176. Note that the pulse width after the pulse width shaping, that is, the minimum value of the time during which the switch 174 is turned on, is limited by the time required for discharging the charge stored in the capacitor 173.

While the rapid falling detection pulse train 179 is at the high level, the ballistic loss removing circuit 100 is in the reset state, and the measurement of the pulse interval needs to be started after the reset of the ballistic loss removing circuit 100 is released. There is. Therefore, the pulse width of the rapidly falling detection pulse train 179 is not adjusted, and the OR 180 with the pulse train 178 after the pulse width shaping is directly performed to obtain the OR output 181. The OR output 181 is delayed by a delay line 183 and output 184 in consideration of using the output signal 182 of the time interval measuring circuit 103 in another circuit.

The output signal 182 of the time interval measuring circuit 103 is intended to detect that the pulse interval is short.
Need not be proportional to the pulse interval. Also,
Since the time required to discharge the charge stored in the capacitor 173 increases depending on the amount of stored charge,
It is necessary to set an appropriate limit 185 as shown in FIG.

FIG. 19 shows the time interval diagnostic circuit 10 shown in FIG.
4A shows a circuit diagram and FIG. 4B shows a signal waveform diagram. By sampling the output signal 186 of the time interval measuring circuit 103 at the rising edge of the rapid rising detection pulse train 187 output from the rapid rising detecting circuit 101, a time interval from the preceding pulse can be obtained. .

The function of the time interval diagnosis circuit 104 is to precede the irradiation of radiation at a time interval shorter than the time required for processing for obtaining energy information from the output signal of the ballistic defect removal circuit 100. Consists in removing the signal or the current signal.

Specifically, the radiation incident on the detector in a time shorter than the time 92 required to remove the noise in the flat portion before the rise or the time 93 required to remove the noise in the flat portion after the rise shown in FIG. Is to output an error signal for performing the removal process on the signal of

The reference voltage A188 shown in FIG. 19 is a voltage corresponding to the time required for removing the noise in the flat portion before the rising, and the reference voltage B189 is the voltage corresponding to the time required for removing the noise in the flat portion after the rising. It is the corresponding time.

The comparator A190 keeps comparing the reference voltage A188 with the output signal 186 of the time interval measuring circuit 103, and the comparator B191 keeps comparing the reference voltage B189 with the output signal 186 of the time interval measuring circuit 103, and outputs 192 and 19, respectively.
Outputs 3.

The result of this comparison is recorded in the D-type flip-flops A194 and B195, respectively, when the output signal of the trajectory defect removing circuit 100 suddenly rises, that is, when radiation enters the detector. , Error A196 or error B197
Is output as

When the error A 196 becomes high level, the time required for removing the noise in the flat portion before the rising edge is obtained. When the error B 197 becomes high level, the time required for removing the noise in the flat portion after the rising edge is obtained. It is not enough time. If the time required for noise removal before and after the rise is the same, only one comparator and one D-type flip-flop may be used.

FIG. 20 shows (a) a circuit diagram for directly outputting a time interval diagnostic result without measuring the time interval,
(B) shows a timing characteristic diagram of the signal. The time interval obtained by the time interval measuring circuit 103 is not used for the purpose of performing a two-dimensional spectrum measurement depending on the time interval. Outputting can be realized with a simple circuit.

A detection pulse train 198 of a sharp rise indicating that radiation has entered the detector is applied to a single shot A19.
9 and input to the trigger of single shot B200.
The single shot A 199 generates a pulse 201 having the same width as the time required to remove noise in the flat portion before the rising of the output signal of the ballistic loss removing circuit 100, and the single shot B 200 generates the noise in the flat portion after the rising. A pulse 202 having a width approximately equal to the time required for removal is generated.

The rapid rising detection pulse train 198 is also input to the D-type flip-flop A203 and the D-type flip-flop B204, and samples the output 201 of the single shot A199 and the output 202 of the single shot B200 at the rising edge, respectively. .

Thus, when the time interval from the preceding signal is shorter than the time required for noise removal, that is, when the output signal of the single shot is at the high level, the error A205 or the error B206 becomes the high level and the diagnosis result is obtained. Is obtained.

When the rapid falling detection pulse train 207 indicating that the trajectory defect removing circuit 100 is reset is at a high level and a sudden rising detection pulse is generated, energy cannot be measured. Error A205
And an error B206 are generated by using the logical sum A208 and the logical sum B209 to output the respective outputs 210 and 211.
Is output.

Further, in order to obtain the energy of the radiation incident on the detector after the reset of the ballistic loss removing circuit 100 is completed, it is necessary to secure the noise removal time before and after the rise of the ballistic loss removing circuit 100. The sharp falling detection pulse train 207 is input to the clear terminals of the single shots 199 and 200, and when the clearing is released, a pulse having a time width necessary for removing noise before and after the rising is generated.
Note that if the time required for noise removal before and after the rise is the same, only one single shot, OR, and one D-type flip-flop are required.

FIG. 21 shows the rise speed measuring circuit shown in FIG.
FIG. 22 shows a signal waveform diagram corresponding to 105. The rise speed is obtained by dividing the rise voltage by the time required for the rise.

The rising voltage 212 is obtained from the difference between the voltage at the rising start and the voltage at the rising end. Specifically, the rising edge of the sudden rising detection pulse train 213,
Then, the difference is obtained after holding the voltage of the output 214 of the ballistic loss removing circuit 100 at the falling time. The trace & hold circuit A 215 holds the voltage before rising, and the trace & hall circuit B 216 holds the voltage after rising. The difference between the voltage 218 after the rise obtained by the subtraction circuit 217 and the voltage 219 before the rise is a rise voltage 212.

The D-type flip-flop A220 and the D-type flip-flop B221 are used to output a timing signal for maintaining the hold state of the output 214 of the ballistic loss removing circuit 100 and for releasing the hold state.

The switch A 224 is connected between the constant current source 222 and the capacitor 223 during the time required for the rise, and the current from the constant current source 222 is supplied to the capacitor while the rapid rise detection pulse train 213 is at the high level. 223.

The single shot 225 generates a pulse that is longer than the pulse width of the sudden rising detection pulse train 213 by about the time required for the operation result of the subtraction circuit 217 and the division circuit 226 to stabilize. The output 227 of this single shot 225 is
The signals are input to the clear terminals of the D-type flip-flop A220 and the D-type flip-flop B221 to release the hold state, and are connected to a switch B228 connected in parallel with the capacitor 223 to discharge the electric charge.

Since the output 229 of the division circuit 226 does not output accurate results unless the rising voltage 212 and the time 230 required for the rising become stable, the output signal 231 of the D-type flip-flop B221 becomes high level. At the time
A signal is output only when the result of division is determined by turning on switch C232. Note that the D-type flip-flop A
The output signal 233 of 220 is output to the trace & hold circuit A215.

FIG. 23 shows a rising speed diagnostic circuit shown in FIG.
(A) Circuit diagram and (b) signal waveform diagram corresponding to FIG. Since the rise time is determined for each detector, the rise voltage can be estimated by measuring the rise speed. If the actual rising voltage is higher than the estimated rising voltage, the diagnosis result is that the rising speed is slow, and if the actual rising voltage is lower than the estimated rising voltage, the diagnosis result is that the rising speed is fast. The output is the role of the rise speed diagnostic circuit 106.

First, the ratio α 237 of the rising speed 236 to the actual rising voltage 235 is obtained by the dividing circuit 234.
The actual rising voltage 235 is obtained from the output 212 of the subtraction circuit 217 of the rising speed measuring circuit 105 shown in FIG.
The rising speed 236 is an output 229 of the measurement result in the rising speed measuring circuit 105 shown in FIG.

Since the ratio α237 is a value within a range determined for each detector, the ratio α237 is compared with the minimum value 239 of the expected rise speed by using the comparator A238, and the comparator B240
Is used to compare the ratio α237 with the maximum value 241 of the expected rising speed.

Rise speed 236 and rise voltage 235
Is stable while the switch C232 in FIG. 21 is ON, the comparison result of the comparator A238 and the comparator B240 is output at the timing of the falling edge of the output control signal 231, respectively. A242 and D-type flip-flop B24
3 and error A244 and error B24, respectively.
5 is output as the diagnostic result.

FIG. 24 shows the rise time measuring circuit shown in FIG.
(A) Circuit diagram and (b) signal waveform diagram corresponding to FIG. The rise time is measured by turning on the switch A 247 while the rapid rise detection pulse train 246 output from the rapid rise detection circuit 101 is at a high level.
This is done by charging the capacitor 249 with the current from 8. At both ends of the capacitor 249, a voltage 250 proportional to the width of each pulse of the sudden rising detection pulse train 246 is obtained.

The voltage 250 across the capacitor 249 becomes a value proportional to the rise time when each pulse of the rapid rise detection pulse train 246 changes from the high level to the low level. From the falling edge of the rapid rise detection pulse train 246, the switch C252 is turned ON for a necessary time in the next rise time diagnosis circuit 108, and the rise time measurement result 253 is output.

After outputting the rise time measurement result 253, the switch B254 is turned on to discharge the electric charge accumulated in the capacitor 249. The switch B254 is turned off by using the logical sum 255 when the rapid rising pulse train 246 is at a high level and while the rise time measurement result is being output.
I have to. Also, single shot 251 is output signal 25
6 and the logical sum 255 outputs the control signal 257 of the switch B254.

FIG. 25 shows the rise time diagnostic circuit shown in FIG.
The comparator A258 compares the rise time 259 with the minimum value 260 of the rise time, and shows the (a) circuit diagram and the (b) signal waveform diagram corresponding to 108.
When the rise time 259 is shorter than the above, the output 261 becomes high level.

The comparator B262 has a rise time of 25.
9 and the maximum rise time 263, and when the rise time 259 is greater than the maximum rise time 263,
Output 264 goes high. Where the rise time is 259
Is the output 253 of the rise time measuring circuit 107 shown in FIG.
It is.

The output control signal 265 of the rise time measuring circuit 107 shown in FIG. 25 is a signal connected to the output signal 256 of the single shot 251 shown in FIG. 24. When this signal is at a high level, the rise time shown in FIG. The output 253 of the measurement circuit 107,
That is, the rise time 259 shown in FIG. 25 is effective.

Output control signal 26 of rise time measuring circuit 107
5 changes from high level to low level,
Output 26 of comparator A258 and comparator B262
By reading 1 and 264 into the D-type flip-flop A266 and the D-type flip-flop B267, respectively, the error A268 goes high when the rise time is short, and the error B269 goes high when the rise time is long.

FIG. 26 shows (a) a circuit diagram in the case of directly outputting the rise time diagnosis result without measuring the rise time,
(B) A characteristic diagram showing the timing of the signal, which is not used for an application in which a two-dimensional spectrum measurement depending on the rise time is performed using the rise time signal obtained by the rise time measurement circuit 107. When only the diagnosis result is required, it is possible to output the diagnosis result directly with a simple circuit.

The width of each pulse of the sudden rising detection pulse train 270 is the time of the sudden rising portion. Therefore,
By checking the length of each pulse width, a diagnosis result can be directly output.

The single shot A271 outputs a pulse 272 having a width corresponding to the minimum rise time. Therefore, by directly sampling the sudden rising detection pulse train 270 at the timing of the rising edge of the output 272 of the single shot A271,
Can be diagnosed whether the pulse width is shorter than the pulse width set in.

That is, the output of the single shot A271
If the sudden rising detection pulse train 270 is at the low level at the timing of the rising edge of the signal 272, the error A274 output from the D-type flip-flop A273 goes to the high level.

The single shot B275 outputs a pulse signal 276 having a width corresponding to the maximum value of the rise time.
Accordingly, a sharp rising detection pulse train 27 is generated at the rising edge of the output 276 of the single shot B275.
By directly sampling 0, a single shot B
It is possible to diagnose whether the pulse width is longer than the pulse width set in 275.

If the sudden rising detection pulse train 270 is at the high level at the timing of the rising edge of the output 276 of the single shot B275, the error B278 output from the D-type flip-flop B277 becomes the high level.

FIG. 27 shows a flat portion inclination measuring circuit shown in FIG.
110A is a circuit diagram corresponding to FIG. 110, and FIG. The extraction pulse train 279 of the flat part is the logical sum 10 shown in FIG.
9 output signal, the sudden rise detection circuit shown in Figure 16 above
This signal indicates that neither the output 159 of 101 nor the output 171 of the abrupt falling detection circuit 102 shown in FIG. 17 is generated.

The output signal 280 of the trajectory defect removing circuit 100 is
The output 114 of the trajectory defect removing circuit 100 shown in FIG. 13 is a signal which rises every time radiation is incident on the detector and falls by reset. When this output 280 is input to a fully differentiating circuit composed of an inverting amplifier 281, an input capacitor 282, and a feedback resistor 283, the output 284 has a positive value according to the rising speed of the output 280 of the ballistic loss removing circuit 100. , And a negative signal is obtained according to the falling speed.

The switch 285 is turned ON only when the extraction pulse train 279 of the flat portion is at the high level, and only the flat portion is extracted from the output 284 of the complete differentiating circuit, whereby the output 286 of the flat portion inclination measurement result is obtained.

FIG. 28 shows a flat portion inclination diagnostic circuit 111 shown in FIG.
In the (a) circuit diagram and (b) signal waveform diagram corresponding to (a), the inclination of the flat portion corresponds to the leak current of the detector. The comparator 287 outputs an error when the output 288 of the flat portion inclination measuring circuit 110 exceeds a previously measured maximum value 289 of the flat portion inclination, that is, when the output 288 exceeds a preset leak current value.
Outputs 290.

The second embodiment relates to a digital signal processing system, and is shown in the block diagram of FIG. FIG. 29 shows the diagnostic units of the radiation measuring apparatus shown in FIGS. 2 to 5 collectively for convenience of explanation. FIG. 29 shows a hardware method. The trajectory defect removal 26 shown in FIG. 2 of the first invention is a trajectory defect removal circuit 291, and a rapid change portion extraction 27 is a rapid rise detection circuit 292 and a rapid fall detection. The circuit 293 corresponds to the time interval measurement circuit 294 for the rapid change interval measurement 28, and the time interval diagnosis circuit 295 corresponds to the time interval diagnosis 29.

The trajectory defect removal 26 shown in FIG. 3 of the second invention is a trajectory defect removal circuit 291, a rising portion extraction 33 is a sharp rise detection circuit 292, a rise speed measurement 34 is a rise speed measurement circuit 296, The rising speed diagnosis 35 corresponds to the rising speed diagnosis circuit 297.

The trajectory defect removal 26 shown in FIG. 4 of the third invention is a trajectory defect removal circuit 291, a rising portion extraction 33 is a sharp rise detection circuit 292, a rise time measurement 47 is a rise time measurement circuit 298, a rise time Diagnostic 48 is a rise time diagnostic circuit
Corresponds to 299.

Furthermore, the trajectory defect removal 26 shown in FIG. 5 of the fourth invention is a trajectory defect removal circuit 291, and the extraction of a flat portion 60 is a rapid rise detection circuit 292 and a rapid fall detection circuit 293.
The logical sum 300 of the flat portion inclination measurement 61 is a flat portion inclination measurement circuit 301, and the inclination diagnosis 62 is a flat portion inclination diagnosis circuit 302.
In response to the above, each of the circuits converts the output signal of the trajectory loss removing circuit 291 to a digital result series digital signal converted by the high-speed ADC 303 by numerical processing, and diagnoses A30 and B30.
36, C49 and D63 are obtained. The trajectory defect removing circuit 291 is the same as the trajectory defect removing circuit 100 shown in FIG. 13, that is, the same as the case of analog signal processing.

FIG. 30 is a block diagram (a) of the high-speed ADC,
(B) shows a signal waveform diagram. The output 113 of the trajectory defect removing circuit 100 shown in FIG. 13 is a continuous analog signal. The high-speed ADC 303 inputs the output 304 of the trajectory defect removal circuit 291 having a continuous time and voltage, converts the time and voltage into digital time-series data having a discrete time and voltage, and outputs an output signal 305. It has a conversion period and conversion accuracy that can reproduce the rapidly changing portion of the trajectory loss removing circuit 291.

The circuit diagram of FIG. 31 shows a circuit corresponding to the rapid rise detection circuit 292 and the rapid fall detection circuit 293 shown in FIG. The shift register 306 is a high-speed ADC 30
3 outputs digital time-series data 307 delayed from the converted digital time-series data 305 by the number of stages of the shift register 306.

By subtracting the digital time series data 307 output from the shift register 306 from the digital time series data 305 using the subtracter 308, a signal 309 obtained by extracting only the changed part of the digital time series data 305 is obtained.

This shift register 306 and subtractor 308
Is equivalent to the fully differentiating circuit of FIG. 14D in the analog signal processing method, and the subtractor 308
If it is assumed that the output 309 of the full differential circuit can be observed as a waveform, the output waveforms 153 and 16 of the fully differentiating circuit shown in FIG. 16 or FIG.
Waveform similar to 5.

Further, the digital comparator 310
The output 309 of the subtractor 308 is compared with the positive detection level set value 311 and the output 309 of the subtractor 308 is set to the positive detection level set value.
When it exceeds 311, a sudden rising detection pulse train 312
To a high level.

Further, the output 309 of the subtractor 308 is compared with the negative detection level set value 313, and when the output 309 of the subtractor 308 becomes smaller than the negative detection level set value 313, a sharp fall occurs. Is set to a high level. Accordingly, the pulse width of each pulse of the sudden rising detection pulse train 312 corresponds to the rising time of the digital time series data 305, and the pulse width of each pulse of the sudden falling detection pulse train 314 corresponds to the digital time series data 305. Fall time.

FIG. 32 shows the time interval measuring circuit 29 shown in FIG.
In the circuit diagram corresponding to Fig. 4, a sudden rising detection pulse train 31
2 is the output of the rapid rise detection circuit 292 shown in FIG.
Reference numeral 312 denotes an abrupt falling detection pulse train 314 which is an output 314 of the abrupt falling detection circuit 292 shown in FIG. For the time interval, the periodic clock signal 315 is
The output 317 is obtained by counting using 6.

With respect to the detection pulse train 312 of the sudden rising, the counter 316 is cleared at the rising portion by the shift register 318 and the logical product 319 via the logical product 320, and the detection pulse train 314 of the rapid falling is at the high level. During that time, the counter 316 is cleared. This time interval measurement circuit
If the output 317 of the 294 can be observed, the output signal 18 of the time interval measurement circuit 103 of the analog signal processing method shown in FIG.
A waveform equivalent to 2 is obtained.

In the output signal 182 in the analog signal processing method shown in FIG. 18, the output voltage is limited 185 for the purpose of shortening the discharge time of the capacitor 173 and the power supply voltage of the circuit. In the signal processing method, the reset time does not depend on the number of bits of the counter 316, and measurement of long time intervals can be performed by increasing the number of bits of the counter 316. Therefore, it is not necessary to particularly provide a limiter.

However, since the purpose of measuring the signal interval is to diagnose that the time interval is shorter than a predetermined time interval, an overflow output 321 is provided if time measurement of more bits than necessary is unnecessary. A counter 316 having a small number of bits is used.

FIG. 33 shows the time interval diagnostic circuit 29 shown in FIG.
A circuit diagram corresponding to Fig. 5 is shown. Output of time interval measurement circuit 31
7 is an output 317 of the time interval circuit 294 shown in FIG.
The set value A322 of the signal interval is the time required for noise removal before rising, and the set value B323 of the signal interval is the time required for noise removal after rising.

The digital comparator 324 compares the output 317 of the time interval measuring circuit 294 with the set value A322 of the signal interval and the set value B323 of the signal interval, and outputs them as the output A325 and the output B326 of the digital comparator 324, respectively.

The output result of the digital comparator 324 is immediately before the contents of the counter 316 shown in FIG.
That is, at the timing of the rising edge of the detection pulse train 312 of the sudden rising, the D-type flip-flop A327
And D-type flip-flop B328, and outputs them as error A329 and error B330, respectively.

FIG. 34 is a circuit diagram corresponding to (a) the rise speed measurement circuit 296 shown in FIG. 29 and (b) a characteristic diagram showing timing. The rise speed is the trajectory defect removal circuit shown in FIG. It is obtained by dividing the rising voltage of the output 114 of 100 by the time required for the rising.

The rising voltage is obtained from the difference between the digital value at the start of the rise and the digital value at the end of the rise. To be more specific, the digital time series data is obtained by the rise and fall of the sudden rise detection pulse train 312. After latching the digital value of 305, the difference is obtained.

A latch A331 shown in FIG. 34A latches a digital value before rising, and a latch B332 latches a digital value after rising. The subtractor 333 obtains the difference between the digital value 334 after the rising and the digital value 335 before the rising, that is, the voltage at the rising. The timing signal generator 336 is used to generate an output A337 for the latch A331 and an output B338 for the latch B332 to maintain the latched state of the digital time series data 305 and release the latched state.

The time required for the rise is determined by using a counter 339 having a gate of the detection pulse train 312 of the rapid rise,
It is obtained by counting the periodic clock signal 340. Also, the output C341 of the timing signal generator 336
Resets the counter 339 after a lapse of time when the operation result of the subtractor 333 and the divider 342 stabilizes after the pulse of the detection pulse train 312 of the rapid rise becomes low level.

The output of the divider 342 indicates the change 344 in the digital value at the rise and the time required for the rise.
345 will not produce accurate results unless it is stable,
The division result 347 is latched by the latch C348 at the timing when the operation result is stabilized by the output D346 of the timing generator 336, and then output 349.

FIG. 35 shows a rising speed diagnostic circuit shown in FIG.
Since the rise time is determined for each detector in the circuit diagram corresponding to 297, the rise voltage can be estimated by measuring the rise speed. If the voltage that actually rises is higher than the estimated rise voltage, a diagnosis result that the rise speed is slow is given.If the voltage that actually rises is lower than the estimated rise voltage, a diagnosis result that the rise speed is fast is given. The output is the rise speed diagnostic circuit 297
Role.

First, the ratio α 352 of the rising speed 351 to the digital value 344 of the actual rising is obtained by the divider 350. The digital value 344 of the actual rise is shown in the figure above.
The output 34 of the subtractor 333 of the rise speed measuring circuit 296 shown in 34
Obtained from 4. The rising speed 351 is an output 349 of the measurement result of the rising speed measuring circuit 296 shown in FIG.

Since the ratio α352 is a value determined for each detector, the digital comparator 353 compares the ratio α352 with the minimum value 354 of the expected rise speed, and compares the ratio α352 with the maximum value 355 of the expected rise speed.

The time during which the rising speed 351 and the rising digital value 344 are stable is the timing at which the output signal of the divider 342 is latched by the latch C348 in FIG. 34, so that the output latch control of the rising time measuring circuit 298 is performed. Signal 346
At the timing of the rising edge of, the comparison result of the digital comparator 353 is read into the D-type flip-flop A356 and the D-type flip-flop B357, respectively, and the error A358 and the error B359 are output as the diagnosis result, respectively.

FIG. 36 shows the rise time measuring circuit shown in FIG.
298 is a circuit diagram corresponding to FIG. 298, and a characteristic diagram showing the timing of FIG. The rise time is measured by counting the clock signal 361 using a counter 360 having a gate of the detection pulse train 312 of the rapid rise. At the output 362 of the counter 360, a digital value proportional to the width of each pulse of the sudden rising detection pulse train 312 is obtained.

The output 362 of the counter 360 becomes a value proportional to the rise time when each pulse of the sudden rising detection pulse train 312 changes from the high level to the low level. The output 362 is held as a count result by the rise time diagnosis circuit 299 for a necessary time from the falling edge of the detection pulse train 312 of the sudden rise. The timing signal generator 36
When the output A364 of 3 goes high, the count value is initialized.

FIG. 37 is a circuit diagram corresponding to the rise time diagnosis circuit 299 shown in FIG.
5 compares the rise time 362 with the minimum rise time 366, and when the rise time 362 is shorter than the minimum rise time 366, the output A367 of the digital comparator 365 goes high.

The digital comparator 365 compares the rise time 362 with the maximum value 368 of the rise time. When the rise time 362 is longer than the maximum value 368 of the rise time, the output B369 of the digital comparator 365 becomes high level. Become. Here, the rise time 362 is the output 362 of the rise time measurement circuit 298 shown in FIG.

The output A 364 of the timing signal generator of the rise time measuring circuit is a signal output from the output A of the timing signal generator 363 shown in FIG. 36. When this signal 364 changes from a low level to a high level. 37 shows that the output 362 of the rise time measuring circuit 298, that is, the rise time 362 shown in FIG. 37 is valid.

When the output A364 of the timing signal generator 363 of the rise time measuring circuit 298 changes from the low level to the high level, the output of the digital comparator 365 is sent to the D-type flip-flops A370 and B371, respectively. By reading, when the rise time is short, the error A372 goes high, and when the rise time is long, the error B373 goes high.

FIG. 38 shows the flat portion inclination measuring circuit shown in FIG. 29.
In the circuit diagram corresponding to 301, the extraction pulse train 374
The output signal of the OR 300 shown in FIG. 29 is a signal indicating that neither the sudden rising detection pulse train 312 nor the sharp falling detection pulse train 314 shown in FIG. 31 is generated.

The digital time series data 305 is an output signal 305 of the high-speed ADC 303 shown in FIG. 30, and is a digital value which rises every time radiation is incident on the detector and falls by reset.

When the digital time-series data 305 is input to a circuit composed of a shift register 375 and a subtractor 376, an output 377 outputs the digital time-series data.
A positive digital value is obtained according to the rising speed of 305 and a negative digital value is obtained according to the falling speed. Extraction pulse of flat part
On the falling edge of 374, the output 377 of the subtractor 376 is latched.
By reading the result into 378, an output 379 of the measurement result of the inclination of the flat portion is obtained.

FIG. 39 shows the flat portion inclination diagnostic circuit shown in FIG.
In the circuit diagram corresponding to 302, the slope of the flat portion corresponds to the leak current of the detector. The digital comparator 380 outputs an error 382 when the output 379 of the flat portion inclination measuring circuit 301 exceeds a preset maximum value 381 of the flat portion inclination, that is, when it exceeds a preset leak current value.

A first modification of the second embodiment is shown in the block diagram of FIG. The diagnostic unit shown in FIG. 29 is converted into software to form a central processing unit (CP).
U).

In the method using the CPU, the trajectory defect removal circuit 291 shown in FIGS. 2 to 5 of the first to fourth inventions is provided to the trajectory defect removal circuit 291 to extract the rapidly changing portion of FIG. 2 of the first invention. 27, rapid change interval measurement 28, time interval diagnosis 29
Also, the extraction of the rising portion 33, the measurement of the rising speed 34, and the diagnosis of the rising speed 35 in FIG. 3 of the second invention, the measurement of the rising time 47, the diagnosis of the rising time 48 in FIG. 4 of the third invention, and the fourth invention The extraction 60 of the flat part, the measurement 61 of the inclination of the flat part, and the diagnosis 62 of the inclination in FIG. 5 all correspond to the memory 383, the CPU and the peripheral circuit 384.

The output signal of the trajectory defect removing circuit 291 is high-speed A
The digital time series data is converted into a digital signal by the DC 303, and the digital time series data is subjected to numerical processing using a CPU to obtain each diagnosis result A to D. The trajectory defect removal circuit 29
FIG. 13 shows the first embodiment, and FIG. 30 shows the high-speed ADC 303.

In the method using this CPU, the digital time-series data 305 converted by the high-speed ADC 303 shown in FIG.
Is temporarily stored in the memory 383 shown in FIG. 40, and then numerical processing is performed by the CPU and the peripheral circuit 384 to obtain diagnosis results A to D. According to this, counting is performed due to the slow processing time of the CPU, and processing for all radiation incident on the detector cannot be performed. However, there is a feature that the signal processing content can be easily changed without changing the hardware configuration. .

A second modification of the second embodiment is shown in the block diagram of FIG. In this modification, a digital signal processor (Digital Signal Processor,
The ballistic defect removal shown in FIGS. 2 to 5 of the first to fourth inventions will be described below.
In the trajectory defect removing circuit 291, an abrupt change portion extraction 27 shown in FIG. 2 of the first invention, an abrupt change interval measurement 28, a time interval diagnosis 29, and a rising portion shown in FIG. 3 of the second invention are extracted. 33,
Rise speed measurement 34, rise speed diagnosis 35, and rise time measurement 47, rise time diagnosis 48 of FIG. 4 of the third invention,
The extraction of the flat part 60, the measurement 61 of the inclination of the flat part, and the diagnosis 62 of the inclination in FIG.
And the DSP 385 and the CPU and the peripheral circuit 386.

The output signal of the trajectory defect removal circuit 291 is
The digital time series data converted by the DC 303 is subjected to numerical processing using the DSP 385 to carry out diagnostic results A to D.
Have gained. FIG. 13 shows an embodiment of the trajectory defect removing circuit 291 and FIG. 30 shows an embodiment of the high-speed ADC 303.

In this method using the DSP, the digital time-series data 305 converted by the high-speed ADC 303 shown in FIG. 30 is temporarily stored in the memory 383, then subjected to numerical processing by the DSP 385, and the CPU and the peripheral circuit 386 analyze the diagnostic results A to D. What you get. The DSP 385 has a CPU and peripheral circuits 386
Thus, since the processing contents can be programmed in advance, the signal processing contents can be changed without changing the hardware as in the method by the CPU.

Further, the DSP is a processor dedicated to digital signal processing, and therefore has a higher processing speed than the method using the CPU. If the processing speed of the DSP 385 is sufficiently faster than the rising speed of the output signal of the preamplifier, it is possible to perform the processing directly without storing the digital time series data 305 in the memory 385.

FIG. 42 is a block diagram showing an example of signal processing for a diagnosis result in the radiation measuring apparatus. In this example, the diagnostic unit is built in the preamplifier 387, and the normal signal flow in the radiation spectrometer is from the preamplifier 387 to the linear amplifier 3 and the ADC 388.

Each of the diagnostic results A to D obtained from the diagnostic section shown in FIGS. 2 to 5 built in the preamplifier 387 is as follows.
The processing result is processed by the diagnosis result processing unit 389, and this processing result is
Connect to DC388 gate input. The gate input of the ADC 388 is provided for controlling whether to convert the input signal.

The signal processing unit 289 is obtained from the diagnosis result of the diagnosis unit.
Can remove signals or detect abnormalities in the detector. Therefore, in the case of signal removal, the ADC 38
The gate of 8 is controlled so that the ADC 388 does not convert the signal and removes the signal.

When an abnormality of the detector is detected, an abnormality is displayed on the abnormality display device 390. From this result, the operator stops the measurement work using the abnormal detector and arranges replacement of the detector. In addition, as for the display of the abnormality of the detector, there is a method of connecting to the computer 291 and automatically performing the abnormality processing of the detector. The computer 291 may be a dedicated or shared device.

In the present invention, by measuring the time interval from the signal of the radiation detector, the signal for which the time for noise removal is not enough can be removed, so that the resolution can be improved and the radiation can be accurately measured. In addition, since signal processing is performed based on the diagnosis result of the rise speed and abnormality of the detector during operation can be detected, accurate radiation measurement can be continued if the degree of abnormality is small.

Since the next radiation can be detected during the charge collection, the sum peak can be easily removed. Furthermore, if the recombination region is a part of the detector, the measurement can be continued while removing the signal whose energy information has been lost, so that the resolution on the low energy side can be improved. The same effect can be obtained by the diagnosis result of the rise time.

Further, since the leak current of the detector can be measured during the operation of the detector, it is possible to confirm that the data is measurement data obtained from a normal detector, thereby improving the reliability of the measurement data. .

Furthermore, if the digital signal processing system is used, it is not affected by environmental changes such as aging and temperature changes. Since the signal distortion is removed by using the trajectory defect removing circuit, the output of the complete differentiating circuit improves the measurement accuracy when measuring the rise time and the rise speed.

[0250]

As described above, according to the present invention, minor abnormalities can be detected at an early stage from the results of the diagnosis of the radiation detector, and the failure of the detector can be predicted, so that replacement such as replacement at the time of periodic inspection can be easily performed. . In addition, when a result of diagnosis of each signal output from the detector determines that some of the signals are abnormal, only the normal signals are employed, and the detector which has been conventionally treated as a failure is used. Even so, it is possible to use the detector as if it were a normal detector until the replacement, which has the effect of improving reliability as well as maintenance.

[Brief description of the drawings]

FIG. 1 is a block diagram of a radiation measuring apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram of the radiation measuring apparatus according to the first invention.

FIG. 3 is a block diagram of a radiation measuring apparatus according to a second invention.

FIG. 4 is a block diagram of a radiation measuring apparatus according to a third invention.

FIG. 5 is a block diagram of a radiation measuring apparatus according to a fourth invention.

FIG. 6 is a characteristic diagram showing a trajectory defect, in which (a) an output of a detector, (b) an output of a charge type preamplifier, and (c) an output of a resistance feedback type preamplifier.

7A is a circuit diagram of a trajectory defect removing circuit, and FIG. 7B is an output signal waveform diagram.

FIG. 8 is a characteristic diagram of various extraction timing signals, wherein (a) an input signal waveform, (b) a waveform after removal of a trajectory defect, (c) a signal of a sudden change due to radiation incidence, and (d) a sudden change by a reset. (E) an extracted signal of a rapidly changing portion,
(F) Extracted signal of flat part, (g) Extracted signal of rising part.

FIG. 9 is a characteristic diagram of signal approach, in which (a) before rising is smaller than time after rising, (b) before rising is longer than time after rising, and (c) time before rising is equal to time after rising.

FIG. 10 is a characteristic diagram of signal processing for a diagnosis result of a time interval.

FIG. 11 is a characteristic diagram of signal processing for a diagnosis result of a rise speed or a rise time.

FIG. 12 is a block diagram of a diagnosis unit according to the first embodiment.

FIG. 13A is a reset circuit diagram of the trajectory defect removing circuit,
(B) Signal waveform diagram.

FIG. 14A is a block diagram of a signal detection circuit;
(B) CR differentiation circuit diagram, (c) active filter circuit diagram, (d) complete differentiation circuit diagram, (e) delay line circuit diagram.

15A is a waveform shaping circuit diagram, and FIG.
Fall signal, (c) CR differentiator output signal, (d) complete differentiator output signal.

FIG. 16A is a circuit diagram of an abrupt rising detection circuit,
(B) Signal waveform diagram.

FIG. 17A is a circuit diagram of an abrupt falling detection circuit,
(B) Signal waveform diagram.

18A is a circuit diagram of a time interval measurement circuit, and FIG. 18B is a signal waveform diagram.

19A is a circuit diagram of a time interval diagnosis circuit, and FIG. 19B is a signal waveform diagram.

20A is a circuit diagram of a direct output circuit of a time interval diagnosis result, and FIG. 20B is a timing characteristic diagram of a signal.

FIG. 21 is a circuit diagram of a rise speed measurement circuit.

FIG. 22 is a signal waveform diagram of a rise speed measurement circuit.

FIG. 23A is a circuit diagram of a rising speed diagnostic circuit, and FIG.
Signal waveform diagram.

24A is a circuit diagram of a rise time measurement circuit, and FIG.
Signal waveform diagram.

FIG. 25A is a circuit diagram of a rise time diagnosis circuit, and FIG.
Signal waveform diagram.

FIG. 26A shows a direct output circuit of the rise time diagnosis result.
FIG. 2 is a circuit diagram, and FIG.

27A is a circuit diagram of a flat portion inclination measuring circuit, and FIG.
Signal waveform diagram.

28A is a circuit diagram of a flat portion tilt diagnosis circuit, and FIG.
Signal waveform diagram.

FIG. 29 is a block diagram of a diagnosis unit according to the second embodiment.

30A is a block diagram of a high-speed ADC, and FIG. 30B is a signal waveform diagram.

FIG. 31 is a circuit diagram of a sudden rising / falling detection circuit.

FIG. 32 is a time interval measurement circuit diagram.

FIG. 33 is a time interval diagnosis circuit diagram.

34A is a circuit diagram of a rise time measurement circuit, and FIG.
FIG. 4 is a timing characteristic diagram of a signal.

FIG. 35 is a diagram of a rise speed diagnosis circuit.

36A is a circuit diagram of a rise time measurement circuit, and FIG.
FIG. 4 is a timing characteristic diagram of a signal.

FIG. 37 is a rise time diagnosis circuit diagram.

FIG. 38 is a circuit diagram of a flat portion inclination measurement circuit.

FIG. 39 is a circuit diagram of a flat portion tilt diagnosis circuit.

FIG. 40 is a block diagram of a CPU.

FIG. 41 is a block diagram of a DSP.

FIG. 42 is a block diagram showing an example of signal processing for a diagnosis result.

FIG. 43 is a schematic block diagram of a conventional radiation measuring apparatus.

FIG. 44 is a conventional signal waveform diagram, in which (a) an output of a detector;
(B) Output of preamplifier, (c) Output of linear amplifier.

FIG. 45 is a circuit diagram of a conventional charge type preamplifier;
Basic circuit, (b) resistance feedback type.

FIG. 46 is a signal waveform characteristic diagram of a conventional preamplifier.
(A) No signal, (b) low count rate, (c) medium count rate, (d) high count rate.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Semiconductor detector (detector), 2 ... Preamplifier, 3 ... Linear amplifier, 4, 24,388 ... ADC, 5 ... Process memory, 6,291 ... Computer, 7 ... Spectrum display device, 11 ...
Basic circuit of charge type preamplifier, 12, 15, 80… Inverting amplifier, 13, 16, 79… Feedback capacitor, 14… Resistance feedback type preamplifier, 17… Feedback resistor, 22… Diagnosis unit, 23,389… Diagnosis result Signal processing unit, 25
... Display of spectrum and detector abnormalities, etc., 26 ... Removal of ballistic deficiency, 27 ... Extraction of sudden changes, 28 ... Measurement of intervals of rapid changes, 29 ... Diagnosis of time intervals, 30 ... Diagnosis result A, 31 ...
Removal of preceding signal or current signal, 32 ... Removal of preceding signal and current signal, 33 ... Extraction of rising part, 34
... Measurement of rise speed, 35 ... Diagnosis of rise speed, 36 ... Diagnosis result B, 37 ... Slow rise speed signal, 38,51 ... Electric charge capture area inside detector, 39,52,66 ... Continuation of measurement ,Four
0, 46, 53, 59, 67 ... detector malfunction, 41, 54 ... the next radiation incident during charge collection, 42, 45, 55, 58 ... signal removal, 43 ... signal with slow rise speed, 44, 57: There is a charge recombination region inside the detector, 47: Rise time measurement, 48 ...
Rise time diagnosis, 49: diagnosis result C, 50: long rise time signal, 56: short rise time signal, 60: extraction of flat part, 61: measurement of inclination of flat part, 62: diagnosis of inclination, 63
... Diagnosis result D, 64: Increasing inclination, 65: Detector abnormality sign, 74: Ballistic loss, 75, 100, 291 ... Ballistic loss removal circuit, 78: Reset input, 87: Rising height, 88: Rising Previous flat portion, 89: Flat portion after rising, 90: Previous rising, 91: Current rising, 92, 93: Time required for noise component removal, 101, 292: Sudden rising detecting circuit, 102, 29
3: rapid fall detection circuit, 103, 294: time interval measurement circuit, 104, 295: time interval diagnosis circuit, 105, 296: rising speed measurement circuit, 106, 297: rising speed diagnosis circuit,
107, 298: Rise time measuring circuit, 108, 299: Rise time diagnostic circuit, 110, 301: Flat portion inclination measuring circuit, 111
, 302 ... flat part tilt diagnosis circuit, 127 ... pulse, 128
… Waveform shaping, 303… high-speed ADC, 383… memory, 384,
386: CPU and peripheral circuits, 390: DSP.

Continuation of the front page (56) References JP-A-61-35383 (JP, A) JP-A-61-155887 (JP, A) JP-A-3-17587 (JP, A) JP-A-63-236988 (JP) JP-A-6-123778 (JP, A) JP-A-6-123779 (JP, A) JP-A-61-14590 (JP, A) JP-A-5-256951 (JP, A) 60-187877 (JP, A) JP-A-64-15688 (JP, A) US Patent 5,307,299 (US, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01T 1/24 G01T 1 / 17 G01T 7/00

Claims (6)

(57) [Claims] An electric signal detected by a radiation detector and converted into an electric signal.
Radiation measurement that inputs the converted signal and measures radiation
An apparatus, wherein the output signal of the radiation detector is a ballistic defect.
The output signal after removing the ballistic defect
A diagnostic unit for diagnosing absence, and
If an abnormality is diagnosed, the radiation detector
Radiation measurement having a signal processing unit for determining presence or absence of normal
In the apparatus, together with the diagnostic unit for the diagnosis of the measurement to the time interval extraction and spacing of rapid change portion of the signal after ballistic defect removal by inputting the output signal of the radiation detector, signal processing from the diagnosis result parts are first preceding signal or eliminate current signal when the time interval of the signal approaches, further when approaching release you <br/> and removing the signals and the current signals preceding Ray measuring device. 2. The method according to claim 1, wherein the detection is performed by a radiation detector and converted into an electric signal.
Radiation measurement that inputs the converted signal and measures radiation
An apparatus, wherein the output signal of the radiation detector is a ballistic defect.
The output signal after removing the ballistic defect
A diagnostic unit for diagnosing absence, and
If an abnormality is diagnosed, the radiation detector
Radiation measurement having a signal processing unit for determining presence or absence of normal
In the apparatus, together with the diagnosis unit inputs the output signal of the radiation detector to measure the extraction and rising velocity of the rising portion of the signal after ballistic defect removal to the diagnosis of the rising rate, the signal processing unit from the diagnostic result Detects that there is a charge trapping area inside the detector and continues the measurement, stops the measurement as a detector abnormality, or detects that radiation has entered during charge collection and removes the signal. performing either or detector inside the charge detect and stop ray measuring device release <br/> characterized in that the radiation measurement as abnormal if the detector to remove the signal that there is recombination region. 3. An electric signal detected by a radiation detector and converted into an electric signal.
Radiation measurement that inputs the converted signal and measures radiation
An apparatus, wherein the output signal of the radiation detector is a ballistic defect.
The output signal after removing the ballistic defect
A diagnostic unit for diagnosing absence, and
If an abnormality is diagnosed, the radiation detector
Radiation measurement having a signal processing unit for determining presence or absence of normal
In the apparatus, together with the diagnosis unit inputs the output signal of the radiation detector to measure the extraction and rise time of the rising portion of the signal after ballistic defect removal to the diagnosis of rise time, the signal processing unit from the diagnostic result Detects that there is a charge trapping area inside the detector and continues the measurement, stops the measurement as a detector abnormality, or detects that radiation has entered during charge collection and removes the signal. performing either or detector inside the charge detect and stop ray measuring device release <br/> characterized in that the radiation measurement as abnormal if the detector to remove the signal that there is recombination region. 4. An electric signal detected by a radiation detector and converted into an electric signal.
Radiation measurement that inputs the converted signal and measures radiation
An apparatus, wherein the output signal of the radiation detector is a ballistic defect.
The output signal after removing the ballistic defect
A diagnostic unit for diagnosing absence, and
If an abnormality is diagnosed, the radiation detector
Radiation measurement having a signal processing unit for determining presence or absence of normal
In the apparatus, together with the diagnosis unit inputs the output signal of the radiation detector to measure the inclination of the extraction and the flat portion of the flat portion of the signal after ballistic defect removal to the diagnosis of inclination, signal processing from the diagnosis result parts are detectors abnormal signs or to continue the measurement and detection, or detector ray measuring device release you characterized by stopping the radiation measurement as abnormalities. 5. A diagnostic unit provided in the radiation measuring device,
Claim 2 or claim 3, characterized in that using a full differential circuit to the signal processing after removing the ballistic defect during the rising speed and the rise time measurements of the input to signal the output signal of the radiation detector The radiation measuring device according to claim 1. 6. The diagnostic unit and the signal processing unit provided in the radiation measuring apparatus convert an analog signal after removing a trajectory defect into digital time-series data and perform digital numerical processing. Any one of claims 5
The radiation measuring device according to the item.