JP2566006B2 - Wave height detection circuit for radiation detector - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a radiation detector used for X-ray analysis and the like, and more particularly to a detection circuit for detecting the wave height of an output signal of the radiation detector.

[Prior Art] In radiation measurement, after converting the energy of the radiation incident on the radiation detector into a DC pulsed electrical signal, the energy of each radiation is measured by the method of measuring the wave height. At present, the wave height is converted into a digital signal by an A / D converter and handled, but in the field of radiation measurement, the main purpose is to obtain the spectral distribution of the radiation to be measured. At the time of -D conversion, it is necessary to accurately quantize at, for example, about 1024 to 4096 levels, and at the same time suppress the differential nonlinearity to 0.5% or less.

Conventionally, the wave height detection circuit used for this purpose uses a Wilkinson type AD converter, and the output signal from the radiation detector is shaped by a waveform shaping circuit such as an approximate Gaussian filter. In FIG. 6 (a), S
After holding the peak voltage of the Gaussian waveform signal in the capacitor, the accumulated charge is linearly discharged using the constant current circuit, and the time t required until the discharge ends is As shown in FIG. 6B, the clock pulse is counted and measured.

In FIG. 6, T _P is the time required for the detection signal to reach a peak from the time origin O at which the radiation arrives, and T _T is the time required for sufficient attenuation after the peak.

This method takes advantage of good differential non-linearity in time measurement performed by counting clock pulses of a constant period, and is mainly used for the purpose of improving differential non-linearity.

[Problems to be Solved by the Invention] However, this method has a drawback in that the A-D conversion takes a considerable amount of time. If the clock frequency is 50 MHz with the device configuration that is currently commonly used, 10
As shown in Fig. 6 (b), about 24 levels are counted.
20 μs is required. When the waveform is shaped into a Gaussian waveform as shown in FIG. 6 (a), it takes more time than the time for the waveform to rise until the waveform returns to the baseline. Therefore, even if such an A / D converter is used, counting is performed. It did not jeopardize the rate and the time required for this conversion was not considered a major obstacle.

By the way, there is an attempt to reduce the time until the waveform returns to the baseline by using the gated integrator, and thereby improve the counting rate.
Proposed by Mr. ka (V.Radeka, IEEE Trans.NS, N
S-15, No. 3 455 (1968)). When such a gated integrator is used, the signal waveform becomes as shown in FIG. 6 (c), and the waveform instantly returns to the baseline by the gate. In the figure, T _M is the measurement time. When such a signal waveform is used, the counting rate greatly depends on the AD conversion time, and the Wilkinson type AD converter having a long conversion time cannot be expected to improve the counting rate very much.

Therefore, it has been attempted to use a successive approximation A / D converter having a short conversion time. However, this A
Although the conversion time of the -D converter is as short as about 2 μsec, the differential nonlinearity is about 50% of the minimum unit, and the condition described above in this respect is not satisfied.

The present invention has been made in view of the above points,
An object of the present invention is to provide a wave height detection circuit for a radiation detector, which has good differential non-linearity and can dramatically improve the counting rate.

[Means for Solving the Problems] In order to achieve this object, the radiation detector wave height detection circuit of the present invention performs A-D conversion with an output signal of the radiation detector as an input signal at a predetermined clock cycle. -D
A converter, an integrating circuit that sequentially integrates the output signals of the AD converter, a register that holds the output signal of the integrating circuit, and the start and operation of the integrating circuit based on the output signal of the radiation detector. A wave height detection circuit for a radiation detector comprising a timing circuit that creates a timing signal for instructing termination, wherein a multiplier is arranged between the AD converter and the integrating circuit, To supply the output signal of the AD converter to the integrating circuit and to supply to the multiplier a time function signal whose value is predetermined according to the time from the start of operation of the integrating circuit. It is characterized in that the output signal of the converter is multiplied.

[Operation] In general, the pulse height is proportional to the area of the pulse signal.
Focusing on that point, the present invention detects the wave height by integrating the pulse signal and obtaining the area, rather than detecting the wave height of the pulse signal exactly as in the conventional case. That is, the output signal of the radiation detector supplied through the waveform shaping circuit is sampled and AD-converted at a predetermined clock cycle by the high-speed AD converter. The AD conversion values that are sequentially obtained are multiplied by the function signal in the multiplier, and then sent to the integrator to be integrated. A certain number of A-D
When the converted values are integrated, the integration is stopped, and the integrated value at that time is read out from the register as a peak value.

 An embodiment of the present invention will be described in detail below with reference to the drawings.

[Embodiment] FIG. 1 is a diagram showing a configuration of an embodiment of the present invention. The detection signal obtained from the radiation detector 1 in FIG.
For example, via the waveform shaping circuit 2 that shapes an approximate rectangular wave,
This AD converter 3 is sent to the parallel AD converter 3.
Is an A / D converter for the input signal based on the clock signal supplied from the clock oscillator 4, and the obtained digital signal is sent to an accumulator 5 and an integrating circuit 7 comprising a counter 6 for counting the overflow signal of the accumulator 5. Are calculated and accumulated. Reference numeral 8 is a register for holding the output value of the integrating circuit and sending it to the outside. A timing control circuit 9 is composed of a discriminator 10, a flip-flop 11, a gate 12, and a frequency divider 13.

In the above configuration, for example, an 8-bit video ADC is used as the parallel A / D converter 3, 8-bit one is used as the accumulator 5 and the counter 6, and 16-bit one is used as the register 8.

The operation of the apparatus having the above configuration will be described with reference to FIG. The waveform shaping circuit 2 outputs a pulse signal having an approximate rectangular waveform having a width of 40 μs as shown by the solid line in FIG. 2 (a), and the clock oscillator 4 has a frequency of 3.2 MHz.
Generates a clock signal (with a period of 0.3125 μs) and frequency divider 13
The division ratio of is assumed to be set to 1/128. A-D
The converter 3 converts the pulse signal based on the clock signal.
Sampling and AD conversion are performed every 0.3125 μsec. Second
7B is an output signal b of the discriminator 10, FIG. 8C is a signal c sent from the flip-flop 11 to the gate 12, and FIG. 8D is an integrated timing signal d obtained as an output of the gate 12. The figure (e) shows the output signal e of the frequency divider 13, respectively. As you can see, the pulse signal is generated 40
128 integration timing signals d are generated during μ seconds, and the integration circuit 7 sequentially integrates the 128 digital signals sent from the AD converter 3 based on this signal d.

The binary information in the integrating circuit 7 after the integration is the integrated value of the input signal, as indicated by the broken line in FIG. 2 (f).

In this case, the full-scale 8-bit data is 128 times (2
⁷ times), so it is 14 with the waveform shown in Fig. 2 (a).
Bit information is obtained. If the necessary information is 10 bits, the upper 10 bits of the 14-bit integration result may be read from the register 8 and used. In this way, the differential non-linearity error δ is reduced to 1/2 ⁴ by discarding the lower 4 bits.

Currently, a 10-bit or 12-bit video ADC is commercially available. 12 bit video ADC
Is used as the A-D converter 3, an integration result of 18 bits is obtained in the waveform as shown in FIG. 2 (a) by integrating 128 pieces of data.
The differential nonlinearity error is reduced to 1/2 ⁸ . A-D converter 3 of differential non linearity error δ is ± (1/2) LSB i.e. even though 50% is reduced to its half ⁸ i.e. about 0.2%, the aforementioned differential nonlinearity The error condition is satisfied.

Regarding the conversion time, it means that the A-D conversion is completed at the time when the final addition of the sampling data is completed.
The nanosecond to 100 nanosecond corresponds to the conversion time in the conventional method, which is a substantially negligible number for the pulse signal width of 40 μsec. Therefore, according to the wave height detecting circuit of the present invention, the detection of the next pulse signal can be dealt with immediately after the wave height is detected, and the count rate can be greatly improved.

Further, in the present invention, since the integration is performed, even if noise is superimposed on the pulse signal as shown in FIG. 2 (f), due to the averaging effect of the integration, the true waveform shown by the broken line is obtained. It is possible to obtain an integration result close to the integration of.
Therefore, improvement of the signal-to-noise ratio (SN ratio) can be expected.

FIG. 3 shows the configuration of another embodiment of the present invention, in which the same components as those of the embodiment of FIG. 1 are designated by the same reference numerals.
In this embodiment, a multiplier 14 is inserted between the A / D converter 3 and the integrating circuit 7, and a function is generated to generate a digital time function signal g (t) predetermined by the integrating timing signal d. 15 is provided and the digital signal output from the AD converter 3 is multiplied by the digital function signal g (t) and sent to the integrating circuit 7. As the function generator 15, for example, a configuration is used in which the time function data stored in the read-only memory (ROM) is sequentially read based on the integrated timing signal d.

Now, it is assumed that the input signal E to the AD converter 3 is an ideal rectangular wave having a width T _M as shown in FIG. g (t)
However, when g (t) = 1 as shown in FIG. 4 (b), that is, when it is equivalent to the absence of the multiplier 14, the output signal f (t of the integrating circuit 7 after A-D conversion and integration is obtained. ) becomes f (t) = (t / T M) as shown in FIG. 4 (c).

By the way, a step noise index is used as an index for evaluating the characteristics of the radiation detection circuit.
Ns ² , delta noise index
There are Nd ² and flicker noise index Nf ² . Goul
Ding proposes to use the Nd ² · Td value as a criterion for evaluation from the viewpoint of throughput when evaluating the detection circuit (FSGoulding, IEEE Trans.NS, NS−29, No. 1 412 (19
82)). This Td is a dead time, Nd ² ·
The smaller the Td value, the better. When g (t) = 1 above
When the Nd ² · Td value is calculated, the ideal value is 4.

Here, as shown in FIG. 4 (d), g (t) is changed to g
When (t) = (t / Tm) ^0.5 is set, the output signal f (t) of the integrating circuit 7 after AD conversion, multiplication and integration is g (t).
Therefore, as shown in FIG. 4 (e), f (t) = (t / Tm) ^1.5 . The Nd ² · Td value in this case is 4.5, which is slightly worse than the ideal value of 4, but the flicker noise index Nf ² is g
The noise figure is improved as compared with the case of (t) = 1. Therefore, this is more advantageous in the case where the resolution is more important than the throughput.

Although the detection signal is ideally assumed to be a rectangular wave in FIG. 4A, the actual signal has a blunted waveform at the rising portion as shown in FIG. 5A. If the integration is performed including this rising part, the error corresponding to the deviation (shaded part) from the rectangular wave affects the integration value, and as a result,
Fluctuations of signals appear due to so-called collection time. Therefore, if the time function signals shown in FIGS. 5 (b) and 5 (d), which are obtained by delaying the time function signals shown in FIGS. 4 (b) and 4 (d) by the rise time, are used, During the rising transition period, the function signal becomes "0" and the integration for that period can be stopped.
The output signal f (t) of is as shown in FIGS. 5 (c) and 5 (e). Therefore, the fluctuation of the signal due to the above-mentioned collection time is reduced, and the SN ratio is improved.

[Effect] As described in detail above, according to the present invention, when the wave height is obtained by A-D converting the output signal of the radiation detector at a predetermined clock cycle and sequentially integrating by an integrator. Since a multiplier is inserted between the D converter and the integrating circuit and the digital signal output from the AD converter is multiplied by the function signal and sent to the integrating circuit, radiation with an optimized noise figure A wave height detection circuit for the detector is realized.

[Brief description of drawings]

FIGS. 1 and 3 are diagrams showing the configuration of an embodiment of the present invention, FIGS. 2, 4 and 5 are waveform diagrams for explaining the operation of each embodiment, and FIG. It is a waveform diagram for explaining the conventional problem. 1: Radiation detector, 2: Waveform shaping circuit 3: Parallel AD converter, 4: Clock oscillator 5: Accumulator, 6: Counter 7: Integration circuit, 8: Register 9: Timing control circuit, 10: Discriminator 11: Flip-flop, 12: Gate 13: Divider, 14: Multiplier 15: Function generator

Claims (1)

(57) [Claims] 1. An A / D converter for A / D converting an output signal of a radiation detector as an input signal in a predetermined clock cycle, and an integrating circuit for sequentially integrating output signals of the A / D converter. A register for holding the output signal of the integrating circuit,
A wave height detection circuit for a radiation detector, comprising: a timing circuit that creates a timing signal for instructing the operation start and operation end of the integrating circuit based on the output signal of the radiation detector; A multiplier is arranged between the integrating circuit and the output signal of the AD converter is supplied to the integrating circuit via the multiplier, and a value is predetermined according to the time from the start of operation of the integrating circuit. A pulse height detection circuit for a radiation detector, characterized in that the time function signal thus obtained is supplied to the multiplier to multiply the output signal of the AD converter.