JPS60187875A - Shaping of waveform - Google Patents

Abstract

PURPOSE:To remove a pile-up by shaping an input signal to a pulse signal with a short time range while the resolution and the counting rate characteristic are both improved based on a pulse signal of an exponential waveform. CONSTITUTION:Timing signals for generation of a pulse signal with the waveform decreasing exponentially are detected with a detection circuit 34 separately and each time a setup time Ts for an exponential function simulation circuit 41 passes, a simulation signal decreasing exponentially simulating these waveforms while the peak thereof is adjusted respectively is generated. The difference is determined between input signals and the simulation signal with an arithmetic unit 33 to shape the input signals to a group of pulse signals with a short time range thereby removing the pile-up.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a waveform shaping method for shaping a group of input signals into pulse signals each having a predetermined time width in the field of radiation measurement and the like.

[Technical background of the invention]

FIG. 1 shows the general configuration of a radiation measuring device to which the waveform shaping method of the present invention can be applied.

The radiation emitted from the radiation source 11 is transmitted to the detector 1 of this device.
Detected by 2. The detection output is as weak as μ■. The preamplifier 13 amplifies this to about mV and also shapes the waveform using a built-in RC circuit. A linear amplifier (main amplifier) 14 amplifies this to a voltage level suitable for a measuring device 15 such as a pulse height analyzer, and also shapes the waveform.

By the way, when such a radiation measuring device measures a strong radiation source, the radiation counting rate per unit time becomes high. As a result, a phenomenon occurs in which signal waveforms overlap each other. This phenomenon is called pileup. When pile-up occurs, the baseline shifts in the portion of the superimposed pulse signals, making it impossible to accurately measure the wave height. For this reason, there is a limit to the radiation counting rate.

In order to solve these problems, it has been proposed to perform waveform shaping using a linear amplifier. This includes (i)
There are a waveform shaping method using a delay line and (11) a waveform shaping method using a differentiating circuit.

(1) First, in the former method, a pulse signal is shaped by subtracting the delayed signal from the human input signal using a delay line whose ends are short-circuited.

FIG. 2 shows the principle of this waveform shaping method. Assuming that a pulse signal 17 as shown in a of the figure is generated, a delayed pulse signal 18 as shown in the figure is created based on this, and a pulse signal 19 as shown in c of the figure is created as the sum of both. obtain.

However, this waveform shaping method has the following problems.

(2) When noise is superimposed on the pulse signal, this will be added 57 times, resulting in a decrease in the S/N ratio.

(2) Although the details are omitted in FIG. 2, for example, in the case of an exponential function waveform, the baseline is not accurately corrected and undershoot occurs.

■It is difficult to arbitrarily change the pulse width.

(ii) Next, in the latter method, the human signal is differentiated using a differentiation circuit using CR.

FIG. 3 shows the principle of this waveform shaping method. Assuming that a pulse signal 20 as shown in the figure a is generated, this is differentiated to create a pulse signal 21 as shown in the figure, and the waveform is shaped.

However, this waveform shaping method also has the following problems.

■After differentiation, it is common to use a Gaussian filter to shape the waveform into a Gaussian waveform. At this time, the decrease in the S/N ratio due to the differentiation cannot be ignored.

■The waveform after differentiation becomes an exponential function with a differentiation time constant. As a result, the "tail" portion of the waveform appears to be trailing, and the time it takes for the signal to completely return to the baseline cannot be ignored. In other words, sharp characteristics cannot be obtained.

■Resolution and counting rate characteristics (counting rate limit) are in a contradictory relationship, and one must be prioritized and the other sacrificed depending on the purpose.

As explained above, the conventional waveform shaping method has a problem in that the accuracy, resolution, and count rate characteristics of the pulse signal obtained from the preamplifier are impaired by waveform shaping.

[Purpose of the invention]

In view of these circumstances, the present invention aims to improve both resolution and count rate characteristics based on an exponential waveform pulse signal, while shaping it into a pulse signal with a short time width and eliminating pile-up. The purpose is to provide a waveform shaping method that can

[Structure of the invention]

In the present invention, for an input signal consisting of a group of pulse signals with an exponentially decreasing waveform, the generation timing of each of these pulse signals is detected, and each time a predetermined delay time elapses, these waveforms are simulated and Generate exponentially decreasing simulated signals with respective wave heights adjusted. Then, the difference between the input signal and the simulated signal is taken, and the input signal is shaped into a group of pulse signals with a short time width. This makes it possible to eliminate pile-ups.

〔Example〕

The present invention will be described in detail below with reference to Examples.

FIG. 4 shows a waveform shaping device using the waveform g-shape method of this embodiment. This device is a device that shapes only pulse signals that decrease exponentially, and the input terminal 31
For example, a signal 32 as shown in FIG. 5a is input manually from a preamplifier. Two pulse signals 32, . 322 is time t. It's piled up as of now.

Now, as shown in FIG. 6a, the peak value of the first pulse signal 321 is set to A, and that of the next pulse signal 322 is set to Δ2.
shall be. In this case, the signal 32 input from the preamplifier
The voltage level vp(l) can be expressed by the following equation.

・・・・・・(1) Here τ. is the discharge time constant of the preamplifier.

The signal 32 is manually inputted to the signal detection circuit 34 via the arithmetic unit 33. The signal detection circuit 34 differentiates the signal 32 and separates each pulse signal 32 . ．． Detection signal synchronized with 32□ rising 35. ．． 35□ is output (Figure 5b). The signal detection circuit 34 may be configured with a differentiating circuit as described above, or may be configured with a comparator.

Detection signals 351 and 352 are single shot multi-high break (hereinafter referred to as single shot circuit) 3
6, and each pulse signal 37.6 has a minimum signal passing time Δ. ．． 372 is created (Figure 5C). These pulse signals 371 and 372 are input to the second single shot circuit 38, and the pulse signals 39 and 372 rise and fall at the falling edges of these pulse signals 371 and 372, respectively. ．． 392 is created (Figure 5 d).

These pulse signals 39. ．． The time width of 392 corresponds to the setup time T5 of the exponential function simulation circuit 41, which will be explained next.
is set to

Now, the exponential function simulation circuit 4I has a two-stage operational amplifier 4.
2.43 is arranged, and an RC charging/discharging circuit 44 is provided at the input end of the operational amplifier 43 in the latter stage. The output side of the operational amplifier 42 at the previous stage is connected to this charge/discharge circuit 44 via a switch circuit 45 . Charge/discharge circuit 4
The base side potential V r e I of 4 is set by an arithmetic error detection circuit 47. These potentials ``1'' and 8'' are set to the final value of the exponential function that is assumed to be output from the output terminal 48 via the arithmetic unit 33, and the ground level is corrected.

By the way, the switch circuit 45 receives pulse signals 391.392.
The contact is closed at each rising edge of , and the contact is opened at each falling edge. Therefore, the exponential function simulation circuit 41
As shown in FIG. 6, the pulse signal 511 rises after the minimum signal passing time Δ has elapsed from time 0, and creates a pulse signal 511 that simulates the first pulse signal 321 (a in the figure). The peak value of this pulse signal 51 is the pulse signal 32. is set to the estimated value A+' of the wave height at the same point in time. From time O to t. '' After 6 hours have elapsed, a pulse signal 512 that simulates the next pulse signal 322 is created.

Similarly, this peak value becomes A2'.

The voltage level VS(t) of the simulated signal 51 created by the exponential function simulation circuit 41 in this manner can be expressed by the following equation.

Vs(t)- ・－・・・(2) Here, τ is the discharge time constant of the exponential function simulation circuit.

The arithmetic unit 33 calculates the difference between the signal 32 input manually from the preamplifier and the simulated signal 51 described above, and creates an output signal 52. The voltage level of the output signal 52 is ■. (1), this is expressed by the following equation.

Vo (t) -Vp (t>-Vs (1:)...
...(3) Specifically, it is as follows.

Vo(t)- If you know the preamplifier to be used, you can find the discharge time constant τ.

The preamplifier's τ. It is possible to make it approximately equal to . In other words, when τ, , ″, τ0, equation (4) can be rewritten as follows.

......(5) Figure 6C shows the output signal when τs'''Iτ0.The manually generated pulse signal 321.322 (
In the same figure, pulse signals 52, . It can be seen that the image has been reshaped to 52□ and the pileup has been removed.

Now, in the above waveform shaping method, as shown in FIG.
．． 322 becomes Δ+Ts.

If we pay attention to the later molecules in these time periods, we can see that the pulse signal 51. ．． The rise of 512 is not necessarily satisfactorily steep (FIG. 5f). In such a case, the calculation result of the calculator 33 is as shown in FIG. 5g, and each pulse signal 52. ．． 52. The waveform appears to have a tail, albeit slightly.

In order to correct such waveform distortion, a signal passing gate may be provided on the output side of the exponential function simulating circuit 41 or on the output side of the arithmetic unit 33. This signal passing gate has a gate pulse 54. as shown in Figure 5. ．． 542 allows the pulse signals 521 and 52□ to pass through, and as a result, the pulse signals 521' and 52□' shown in FIG. 5 can be obtained.

〔Effect of the invention〕

As explained above, according to the present invention, since a short width pulse is created from an input signal by simulating an exponential function, it is possible to eliminate pile-up, and even in high count rate radiation measurements, accuracy and resolution are compromised. No.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the general configuration of a radiation measuring device, Fig. 2 is various waveform diagrams to explain a conventional waveform shaping method using a delay line, and Fig. 3 is the same (conventional 4 is a block diagram showing an example of a circuit for carrying out the waveform shaping method of the present invention; FIGS. 5 and 6 are various waveform diagrams for explaining the waveform shaping process in this circuit. 32... Signal (input signal), 33... Arithmetic unit, 34... Signal detection circuit, 36.38... Single shot circuit, 41...
...exponential function simulation circuit, 51 ...simulation signal, 52 ...output signal. Applicant Japan Atomic Energy Corporation Representative Patent Attorney Umeo Yamauchi Figure 1 Figure 6 Figure 2 Figure 3 Figure 2

Claims (1)

[Claims] For human signals consisting of a group of Noculus signals each having a waveform that decreases exponentially, the generation timing of these Noculus signals is detected, and each time a predetermined delay time elapses, these waveforms are simulated and the wave height is A waveform shaping method characterized in that the human-powered signal is shaped into a group of Noculus signals with a short time width by generating a simulated signal that decreases exponentially by adjusting each of the human-powered signals and the simulated signal, and by taking the difference between the human-powered signal and the simulated signal.