JPS62197781A - High counting rate and high resolution preamplifier - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a high-resolution preamplifier used as a preamplifier for semiconductor radiation detectors such as Regarding amplifiers.

[Prior Art] A charge amplifier with an FET input is generally used as a preamplifier for a semiconductor detector for radiation such as low-energy γ-rays or X-rays. F E 1- has an extremely high input resistance of <100 GΩ. Therefore, when used in the input circuit of a radiation detector, charges are accumulated in the gate, and this charge must be discharged. Several methods have been put into practical use, examples of which are as follows.

(1) Resistance feedback type This is the method shown in Figure 3 (a).

1 is a radiation detector, 2 is an FET, 3 is an amplifier, 4 is a feedback capacitor (hereinafter referred to as C"), and 5 is a feedback resistor R".

t for radiation input! i-ray detector 1 emits a charge,
Enter the FET2 gate. This changes the drain-source current ids, and the signal passes through the amplifier 3 and sends the output vout to the next stage. In this circuit, resistor 5 is Cf4
Since the resistor 5 is connected in parallel with the resistor 5, the gate charge of the FET can be discharged, but connecting the resistor 5 causes noise. If the resistance value is made high, the noise will be reduced, but if the resistance value is large, the time constant of the discharge circuit of Cf Rf will become large and fast discharge will not occur, so there will be a limit to the count rate characteristics. Conversely, when the count rate characteristics are the same, the resolution is @10e compared to the optical pulse feedback method described later.
V is bad.

(2) Continuous light return)! Method: Shown in Figure 3 (b).

The DC level of yout is detected by amplifier 6, and V out
The LED7 generates an amount of light proportional to the FET2
to discharge the charge accumulated in the gate. This causes noise because a leakage current constantly flows due to light irradiation. Therefore, this method is also several times smaller than the optical pulse feedback method.
0eV resolution is poor.

Since all of the above methods have flaws, an optical pulse feedback method was devised. It is shown in Figure 4 (a). In the figure, when X-rays are incident on the radiation tA-ray detector 1, Q=qE/ε(
C) A charge is generated [q: unit charge, ε: ? Energy required to create an l-child-hole pair, E; energy of X-rays]
. This is FET2°Cf4. The charge type preamplifier constituted by the amplifier 3 provides an output VOut. vou
t is the following formula% formula% (AO is the voltage amplification degree, Ct is the capacitance of the input stray capacitance 9) Here, Ao is adjusted so that Cf >> Ct / Ao.
and C", so Vout-Q/Cf becomes Vout-Q/Cf. When the gate potential of FET2 decreases due to this signal charge (due to X-ray incidence) and leakage current, yout increases in potential with each X-ray pulse and becomes Vout-Q/Cf. The waveform becomes as shown in Figure (1) (a), and yout gradually increases.If you leave it as it is, vo
ut is finally limited by the power supply voltage and becomes inoperable, so when VOut reaches the upper threshold ■tH, the window comparator 8 works to make the LED 7 emit light, and the FET 2
receives the light from LED 7 and releases the charge accumulated in the gate to the source. When the charge on the gate decreases, vout decreases rapidly. When VOut reaches the lower gmvt L, the window comparator 8 detects this, stops the LED 7 from emitting light, and starts the charging cycle again. This method uses LED7
Noise will be generated when light is coming in, but in that case you can stop the measurement. The resolution of this method is 140 to 150 e<2> at 5.9 KeV. There is also a drain pulse feedback method.

This is as shown in Figure 4 (B), and Figures 3 and 4 (A).
The same parts are given the same symbols. 20 is a relay,
21 is a resistor. The gate leakage current of the FET rapidly increases with an increase in the drain voltage Vd, so this characteristic is utilized. That is, the comparator 8 operates in accordance with the upper limit of vout, activates the relay 20, and connects the resistor 2.
1 is shorted. Therefore, the drain voltage rises rapidly, and 9 increases, discharging the gate charge.

[Problems to be solved by the invention] In the above-mentioned optical pulse system, the light of EO7 is
The time required to irradiate 2 and discharge the charge accumulated in the FET gate is several μsec, whereas the rapid voltage change accompanying the discharge of the gate charge has an extremely large degree of amplification in the fli ray measurement system. Therefore, a large disturbance is caused to the mock membrane, and it takes 100 p sea before it recovers to a steady state.
It takes ~1 re sec. Therefore, Figure 2 (
1) A signal is generated to inhibit the measurement shown in (c) to prevent the detection of erroneous signals. This disturbance is not a problem when the X-rays are weak. This is because the baseline rise of VOut is slow for weak inputs, so the LED
3 is discharged, and the dead time described above is sufficiently short compared to the open time during measurement. However, in an electron microscope, it is necessary to irradiate a strong electron beam to emit the desired X-rays in a short time and shorten the working time.
As the energy of the line increases, the baseline of vout rises quickly, thus requiring frequent resets, and the proportion of disturbance time becomes relatively large. The same applies to the drain pulse feedback method. Ultimately, the limit of high count rate measurement is determined by the dead time until the disturbance in this long system recovers.

The present invention has been made in view of the above problems, and its purpose is to reduce the number of times the gate charge is discharged, thereby increasing the measurement efficiency.

[Means for Solving the Problems 1] The present invention for solving the above-mentioned problems is a radiation detection system in which the output charge of a semiconductor radiation detector due to incidence of radiation is received at the gate of an FET and the output voltage is sent to the next stage. Make it into a circuit,
In a high-resolution device amplifier in which a comparator drives a reset circuit based on a set threshold value and discharges the FET gate charge, a frequency component lower than the signal frequency band of the amplifier output is selectively fed back negatively to the previous stage. It is characterized by being equipped with a circuit for.

[Operation] The low frequency component of the output voltage VOut generated when the output charge of the radiation detector due to radiation incidence is added to the gate of the FET is extracted and negatively fed back to the front stage of the amplifier, thereby stabilizing the output baseline.

[Example] Examples of the present invention will be described below. FIG. 1 is a circuit diagram of an embodiment of the present invention. The same parts as in FIGS. 3 and 4 are given the same reference numerals. When X-rays enter the radiation detector 1, a signal enters the gate of the FET 2, passes through the transistor 12, passes through the amplifier 3, and sends the output as VOut to the next stage.

10. When there is no feedback circuit in 13, this Vout rises due to the accumulation of gate charge in FET2 as shown in Figure 2 (2) (a) above, but what is required as a signal is the rise of each incident X-ray. It's only a minute. A signal enters the gate of FET2 due to the incidence of X-rays, and the gate potential decreases due to the charge, and the current ids between the drain and source decreases. The potential at point 0 15 in the figure is kept constant without responding to sudden changes because there are many capacitors 16. Since the power supply 17 is constant and the emitter-base voltage Feb of the transistor 12 is constant at about 0.6V, the potential of the emitter of the transistor 12 is constant.

When ids of FET2 decreases in this state, resistor 18
Since the current flowing through is constant ((potential at 0 point 15 - emitter potential of transistor 12)/resistance 18), the emitter-collector current iec of transistor 12 increases accordingly. This passes through the amplifier 3 and is output as vout. This waveform is supposed to be the waveform shown in FIG. 2(1)(a). Here, we want to eliminate only the increase in unnecessary DC components and prevent Vout from rising. To do this, it is sufficient to correct the potential at point 0 15 by lowering it by the amount of change in ids of FET2. Then, since the DC component or the low frequency component is a potential drop at the 0 point 15, the iec of the transistor 12 will not increase, and it will not be output to VOut, but only the signal pulse that is the output of the radiation detector 1 will be output to Volt. It will appear in For that reason you
The medium and low frequency components of the output are passed through a low pass filter (LPF) 1.
0, this extracted component is passed through the amplifier 13 and the transistor 11, and negative feedback is applied to the 0 point 15, and the potential of the 0 point 15 is lowered based only on the low frequency component. Eliminating the increase in the current iec of transistor 12, thereby reducing the aforementioned yout
Stabilize the baseline. Therefore, in the waveform of VOut, the potential rise due to the DC component does not occur as in the waveform (a) of FIG. 2(2). However, the accumulation of charge at the gate of FET2 progresses, and eventually the channel is closed and 1cls no longer flows.As mentioned above, the potential at point 015 decreases every time radiation is input, so the potential at point 015 becomes lower. When the threshold value is reached, the comparator 8 operates to prevent the LED 7 from emitting light and to increase the gate charge to fi. The above-mentioned disturbances still occur due to this discharge, but their frequency is reduced. That is, the fourth
In the conventional pulsed light feedback circuit shown in the figure, Vout −Q/ (
As shown by Cf +Ct /Ao >, A
If ○ is not sufficiently large, Ct cannot be ignored, resulting in a decrease in the S/N ratio and deterioration in linearity. Therefore, the bare gain Δ0 of the amplifier is usually designed to be 105 or more. However, since this gain is large, vout changes greatly in response to a small change in the FET gate potential. This means that the dynamic range of the amplifier is not limited by the input, but the output (vout) is limited by the power supply voltage of the amplifier.

Therefore, as described above, in the past, the comparator was activated to perform a reset before the output was saturated.

Now, if the voltage gain Ao of the amplifier is 105, and in the conventional amplifier, the difference between the upper and lower thresholds is 1V, then the change in gate potential ΔVgs is only 104V. In this embodiment, the value of the resistor 18 is set to IOKΩ, and the potential of the 0 point is set to 20V and 16V, respectively, to the threshold values of the comparator. FE
Assuming that the mutual conductance g- of T2 is 41 and the drain-source voltage Vds of FET2 is 5V, it is expressed as Δgs-(1/9m)Δids in the FET. Therefore, in this embodiment, the value of ΔVos is as follows.

ΔV9s-(1/ (4x 10°3))×[((20
-5)/104) - ((16-5>/104)] -0,1(Volt) From the above results, when the same signal is input, the number of resets of the amplifier in this example is compared to the conventional one. It becomes 1/10,000th.

In this way, the output is stabilized at a constant level, which was conventionally limited by VOut, but in principle it can be used until the FET is cut off, and in the amplifier of the present invention, the number of resets can be reduced. can be significantly reduced.

FIG. 2 compares the waveforms of various parts of the conventional pulsed light feedback circuit and the circuit of the present invention. FIG. 2(1) shows the waveform of the conventional method, and FIG. 2(2) shows the waveform of the present invention. VOut is (1)
In this case, a voltage based on the leakage current from the detector is added to the step voltage of the signal, and Vout has a sawtooth shape. Then, when Vout reaches the upper threshold Vt+, the comparator 8 is turned on to lower the baseline to the lower threshold. On the other hand, in (2), the step pulse caused by the signal becomes an attenuated waveform due to the time constant of the feedback circuit, but the baseline is kept substantially constant. Further, in the comparator output (1), the width of the output waveform is narrow because the accumulation of gate charge of the FET is small, but in the case of the present invention, the pulse width is wide because the accumulation of gate charge is large. However, this does not become a problem compared to the disturbance time, and since the number of discharges is small, the ratio of the measurement inhibition time to the total measurement time becomes small, so that measurement efficiency can be improved.

Although the optical pulse feedback method is taken as an example here, it has no particular relation to the reset method, and a drain pulse feedback method may also be used. Also, although we have shown the method of returning to the drain,
Feedback may be applied to the source of the FET. or,
The negative feedback method is not limited to one using an LPF, but a DC servo or the like may be used as long as it does not pass the signal frequency component but passes the low frequency component, and it does not matter whether it is a passive circuit or an active circuit.

Furthermore, by making the resistance value of the low-pass filter 10 switchable, the time constant of the low-pass filter 10 can be switched, and when applying optical feedback based on the comparator output, it is possible to use a foot coupler. The reset time may be shortened by changing the resistance value and decreasing the time constant.

〔Effect of the invention〕

As mentioned above, according to the present invention, the number of gate charge discharges due to light irradiation is greatly reduced compared to the conventional pulsed light feedback method, which is said to have the best resolution, and the number of times measurement is stopped due to disturbance is reduced. The measurement efficiency increased significantly.

[Brief explanation of drawings]

Fig. 1 is a circuit diagram of an embodiment of the present invention, Fig. 2 is a waveform diagram of each part of a conventional circuit and a waveform diagram of each part of an embodiment of the present invention, Fig. 3 is a conventional circuit diagram, and (A) is a resistor feedback Figure 4 (B) is a circuit diagram of a continuous optical feedback type, Figure 4 (A) is a conventional optical pulse feedback type, and Figure 4 (2) is a circuit diagram of a drain pulse feedback type. 1...Radiation detector 2...FET3.6...
Amplifier 4...l)! Kanyo-eup 5...Resistor
7...LED8...Comparator 9...
・Input stray capacitance 10...LPF 11.12...Transistor 13...Amplifier 15...O point 16...Capacitor 17...Power supply 18.21...Resistor
20...Relay patent applicant: JEOL Co., Ltd. Agent: Patent attorney: Fujijigai Ijima 1 Figure 1], Radiator 1 FET 3.13i amplifier 4I feedback capacitor 1 7i LED 8 + comparator 9; input 5 ``Tatami roll I capacitor 1711 ring negative wisdom 3 diagram (b) 1; radiation species earthenware 1 FET skin 6;

Claims (1)

[Claims] The output charge of the semiconductor radiation detector due to the incidence of radiation is F
In the high-resolution preamplifier, the radiation detection circuit receives the output voltage at the gate of the ET and sends the output voltage to the next stage, and the comparator drives the reset circuit based on a set threshold value and discharges the FET gate charge. A high counting rate, high resolution preamplifier comprising a circuit for selectively negative feedback of frequency components lower than the signal frequency band of the amplifier output to the previous stage.