RU2263368C1 - Method for correcting frequency response characteristic of photoelectronic multiplier - Google Patents

Abstract

FIELD: measurement technology.

SUBSTANCE: proposed method includes selection of dynode by voltage drop across dynode characteristic. Working point of chosen dynode is set by potential selection on descending branch of dynode characteristic when nonflatness coefficient of output pulse is higher than unity or on its ascending branch when nonflatness coefficient is lower than unity. Voltage pulse is generated across chosen dynode simultaneously with arrival of light pulse, its value each time being proportional to charge picked off this dynode, and voltage amplitude by end of light pulse is set equal to voltage drop across chosen dynode thereby decreasing or increasing photoelectronic multiplier gain by value of nonflatness coefficient of photoelectronic multiplier.

EFFECT: reduced output pulse distortions.

4 cl, 4 dwg

Description

The invention relates to the field of measuring equipment, namely to measuring the shape of a light pulse using a photomultiplier tube (PMT), and can be used to correct the frequency response of a PMT in the low frequency region.

When measuring light pulses using a photomultiplier, with a large output current, distortions in the shape of the output pulse are observed. In these cases, correction of the frequency response of the PMT is necessary.

There is a method of suppressing high-frequency oscillations that appear at the pulse front and are caused by the inductance of the conclusions of the PMT (see, for example, US patent No. 2798165, CL 250-207, 1956) by introducing damping filters from low-resistance resistors into the chains of the last PMT dynodes.

But in the low-frequency region, a rise or fall of the peak of the PMT output pulse is observed when its photocathode is exposed to a rectangular pulse.

A known method (see GOST 11612.0-75), based on the stabilization of the voltage between the PMT diodes using capacitors shunting the last PMT dynodes, the capacitance values of which are calculated by the formula:

where Q is the anode current charge, C,

m is the gain of the cascade (dynode),

n is the total number of cascades,

U _i - voltage at the i-th cascade (between the i-th and (i-1) -th dynodes),

Q _∂ is the charge of the i-th dynode, C,

I _∂ - current of the i-th dynode, A,

t _u - the duration of the light pulse, s,

100 - coefficient entered under the assumption of an allowable change in the inter-single-voltage voltage of not more than 1%.

This method is adopted as a prototype.

When using this method, the shape of the output pulse is not provided for an output charge of more than 1 μC (or for currents of more than 0.1 A and pulse duration of more than 10 μs), which is associated with the Molter effect in the structure of the dynode (see Nuclear Instruments and Methods, 1963, v .24, No.2, 353-357), as a result of which the secondary emission coefficient increases with time and, despite the constant voltage between the dynodes, the peak of the output pulse begins to increase. In addition, when high currents are taken from photocathodes, especially those that are alkaline, due to the high resistance of the photocathode, the potential changes between the photocathode and the first dynode, as a result of which the top of the output pulse begins to decline. The use of passive correction filters at the output of the PMT, for example, pulse differentiation, leads both to a decrease in the output signal and to distortion of the shape of the output pulse at low output currents of the PMT, that is, in the absence of the above Molter effects.

The proposed method solves the problem of reducing distortion of the flat top of the output pulse in the entire range of linear output currents of the PMT.

This is achieved by the fact that in the method of correcting the frequency response of the photomultiplier tube by stabilizing the voltage between the last dynodes of the photomultiplier tube, one of the dynodes is selected, by selecting the potential, the operating point of the selected dynode is set on the falling branch of the dynode characteristic, if the output pulse non-flatness factor is greater than unity, or on the ascending branch if the non-flatness coefficient is less than unity, while simultaneously with the arrival of a light pulse at the selected dynode with create a voltage pulse, the value of which at each moment of time is proportional to the charge removed from this dynode, while the voltage amplitude at the time of the end of the light pulse is set equal to the voltage drop across the selected dynode, thereby reducing or increasing the gain of the photoelectronic multiplier by the value of the non-flatness coefficient the output pulse of the photoelectronic multiplier, while the voltage pulse is formed by integrating the current of the selected dynode on the capacitor, Connects the selected dynode and neighboring dynode, and the capacitance of the capacitor based on the parasitic capacitance set of conditions:

where I _d is the maximum current of the selected dynode corresponding to the linearity limit of the output characteristic of the photoelectronic multiplier,

t _u - the maximum duration of the light pulse,

ΔU is the voltage drop across the dynode characteristic, at which the gain of the photoelectronic multiplier changes to the value of the non-flatness coefficient of the output pulse, duration t _u ,

Q _d is the maximum charge of the selected dynode corresponding to the limiting charge of the output characteristic of the photoelectron multiplier.

The dynode and its operating point are selected from the condition of ensuring the minimum deviation of the shape of the output pulse from the rectangular one when the photocathode is illuminated by a rectangular light pulse, and the light pulse intensity is selected sufficient to obtain the maximum anode current.

Analysis of known solutions did not reveal signs similar to the hallmarks of the claimed method. This allows us to conclude that the claimed method meets the novelty criterion.

The invention is illustrated by drawings, in which Fig. 1 shows a typical shape of a photomultiplier current pulse with a standard voltage divider (in the prototype) when illuminated by a rectangular pulse, Fig. 2 shows a photomultiplier characteristic of a photomultiplier - the dependence of its gain on voltage between the selected dynode and the previous dynode, figure 3 shows the amplitude-time diagrams illustrating the proposed correction method: on figa - voltage pulse generated on the selected dynode, on fig.3b - change the gain of the PMT, on figv - PMT output pulse shape in the presence of correction.

Figure 4 presents a device that implements the proposed method.

The device contains: PMT - 1, which shows the selected dynode 2, four adjacent dynodes (i-1), i, (i + 1), (i + 2) and anode 3, a fragment of the resistor divider of the supply voltage of the PMT 1, consisting of resistors R _i-1 , R _d (tuning) and R _{i + 1} , connected between (i-1) and i, i and (i + 1), (i + 1) and (i + 2) diodes, respectively . Capacitors C _i-1 and C _{i + 1} shunt the resistors R _i-1 and R _{i + 1,} respectively, the capacitor C _{i is} connected between the i-th diode and the common point of the circuit, the integrating capacitor C _{and is} connected between the (i + 1) diode and the selected dynode 2, which is connected to the movable tap of the trimming resistor R _d . Resistors R _i-1 , R _d and R _{i + 1} set the necessary voltage between the PMT diodes. Capacitors C _i-1 , C _i and C _{i + 1} provide stabilization of inter-diode voltages during the passage of a current pulse, and their capacitances are selected in accordance with the well-known formula (1). The capacitor C _and a voltage pulse is generated for controlling the gain of the photomultiplier dynode 1 to 2 selected that provides a low frequency correction characteristics PMT and pulse shape of the output.

The proposed method is as follows.

When illuminating the photomultiplier photomultiplier at the output of the photomultiplier (from its anode) when powered from a known voltage divider with storage capacitors, a current pulse is removed, the shape of which is shown in figure 1.

The deviation of the shape of the output pulse from a rectangular shape can be characterized:

- non-flatness of the output pulse, the numerical value of which is determined by the formula:

where I _o - PMT current at the initial time,

I _k - PMT current at the end of the light pulse,

ΔI = I _k -I _o the current difference at the end and beginning of the light pulse.

- the flatness coefficient of the output pulse, the value of which is calculated by the formula:

The parameter δ can take a positive value (for a pulse with a positive slope of the peak of the output pulse) or a negative value (for a pulse with a negative slope of the peak of the output pulse). Accordingly, the coefficient K may be greater or less than unity.

The correction of the frequency response of the PMT in the low frequency region, that is, the reduction of the coefficient δ, using the proposed method is as follows.

As can be seen from figure 1, the coefficients δ and K increase with increasing duration of the light pulse, which corresponds to an increase in the output charge Q = I _a · t _u . Studies have shown that these coefficients are independent of the value of the anode current if the value of the output charge in the pulse is preserved, that is, with an increase in the output current, the same distortions (values of the coefficients δ and K) will be achieved with a proportionally reduced light pulse duration t _u .

As can be seen in figure 2, the dynode characteristic of the PMT has a pronounced plateau, the middle point of which corresponds to the typical (known) distribution of potentials between the dynodes. Two pairs of points are marked on this dependence: (A, B) and (A1, B1) corresponding to the operating points of the selected dynode, one of which requires its potential to be set. The choice of a specific operating point depends on the nature of the distortion of the shape of the output pulse (which is known in advance, before the start of the measurement) and on the polarity of the voltage pulse supplied to the selected dynode, which will be discussed below.

Consider how the correction of the characteristics of the PMT is carried out with distortion in the form of a pulse with a rising peak (see figure 1). With this distortion of the pulse shape, the potential of the selected dynode is set to point A on the falling branch of the dynode characteristic (see Fig. 2).

To increase the accuracy of the correction characteristics and visually observe its effectiveness, illuminate the PMT with a rectangular pulse of light. Simultaneously with the arrival of the light pulse (at the moment of the beginning of the light pulse), a pulse of positive polarity is fed to the selected dynode (see Fig. 3a), the voltage of which at each moment of time is increased in proportion to the output charge of the PMT. In this case, the gain of the PMT will decrease as the output charge increases (see Fig.3b) in accordance with its dynode characteristic (see Fig.2), as a result of which the output voltage of the PMT will decrease and the shape of the output pulse becomes flatter. In order to completely correct the non-flatness of the output pulse, it is necessary to create a voltage pulse with the amplitude ΔU on the selected dynode (see figa), under which the following condition is satisfied:

where M (U _a ) is the gain of the PMT at a voltage U _a at point A (see figure 2),

M (U _a + ΔU) is the gain of the PMT at voltage (U _a + ΔU),

K (t _u ) is the non-flatness coefficient for a pulse of duration t _u .

In order for compensation to occur over the entire duration of the pulse, the selection (adjustment) of the operating point A and the selection of the dynode to which the voltage pulse will be supplied are performed so that relation (5) is fulfilled at different light pulse durations with maximum accuracy over the entire pulse duration.

As is known, the current pulse from the PMT dynode has a positive polarity and is proportional to the anode current. This allows you to generate a voltage pulse on the selected dynode by integrating its current on the capacitor C _and , the second output of which is connected to an adjacent dynode, relative to which the voltage changes. Knowing the required ΔU pulse amplitude and pulse charge Q _d in the selected dynode can be calculated capacitance of the integrating capacitor _{C, and} the formula (2).

In the case of distortion of the pulse in the form of a negative slope, the working point of the dynode is set to point B (see figure 2). When a positive polarity pulse is applied to the selected dynode, the gain of the PMT increases, which compensates for the drop in the peak of the output pulse, similar to the process described above for distortion in the form of a rising peak.

The proposed method provides correction of the characteristics of the PMT and when a negative pulse is applied to the selected dynode. In this case, the operating points of the selected dynode should be taken A1 (instead of A) or B1 (instead of B) for distortions in the form of rising or falling peak of the pulse, respectively.

A device that provides the implementation of this method (see figure 4), works as follows.

In practice, the correction of the shape of the output pulse in this device is as follows. In the initial state (at the rated voltage of the PMT, the voltage of the selected dynode is set in the plateau region (see Fig. 2), and the capacitance of the capacitor Cu is set based on formula (1)). PMT is illuminated by a rectangular pulse of light of sufficient intensity and duration. The nature of the distortions in the shape of the output pulse and the values of the coefficients K and δ are determined at the maximum values of the output current of the PMT and the duration of the light pulse. Given that for the PMT of a given type, the nature of the distortion does not change, choose the operating point A or B (see figure 2) of the selected dynode 2 using a trimming resistor R _d . Then, choosing the capacitance of the integrating capacitor C _and (the initial value is calculated by the formula (2)), they achieve the minimum distortion of the shape of the output PMT pulse (see Fig. 3c). If necessary, re-build the working point A using a trimming resistor R _d and the capacitance of the integrating capacitor C _and . Note that the second terminal of the integrating capacitor _{C, and} can be connected to the i-th dynode (instead of the (i + 1) -th dynode), but in this case using the dynode power mode to the descending branch dynode characteristics (point A) to the operating voltage integrating capacitor C _and will increase, which will require the use of capacitors for a larger operating voltage.

The choice of PMT dynode 2, according to which the frequency response should be corrected, is determined by the permissible value of the capacitance of the integrating capacitor C and the proximity (correspondence) of the shape of the dynode characteristic (see Fig. 2, Fig. 3b) to the non-flatness of the output pulse (see. 1). For a PMT of a given type, it is enough to select dynode 2 once, by which to correct the frequency response of the PMT, and for individual PMT samples, select only the capacitance of the integrating capacitor C _and . A condition for choosing a dynode is also the requirement that the capacitance of the integrating capacitor significantly exceed the stray capacitance. In practice, this corresponds to capacitance C _and > 200 pF. Therefore, it can be included, starting with dynodes in the middle part of the photomultiplier tube multiplier system.

Dynode characteristics, as a rule, do not depend on the dynode number, however, individual dynodes may have design features that are advisable to use when choosing the dynode by which the correction will be performed. Moreover, with an insufficient degree of correction for one dynode, one more dynode with the same correction capacity can be used.

After selecting the dynode, its potential and the capacitance of the capacitor C, with the help of a rectangular light pulse, the photomultiplier is ready to detect light pulses of arbitrary shape without distortion if the output charge does not exceed the output charge for which the adjustment is made.

Note that, despite the circuit similarity of the inclusion of the integrating capacitor C _{and the} storage capacitors C _i-1 -C _{i + 2} , they fundamentally differ in their function. Capacitors C _i-1 -C _{i + 2} stabilize the voltage between the dynodes. The integrating capacitor Cu creates a voltage change at the selected dynode 2, which controls the gain of the PMT. In practice, the voltage drop ΔU at the end of the light pulse is tens of volts (tens of percent of the installed voltage at point A), and the technical effect of the proposed method, in comparison with the known solution, is expressed in several times increase in the output charge of the PMT at the same (specified) level of permissible PMT output waveform distortion.

Claims (4)

1. A method of correcting the frequency response of a photoelectronic multiplier by stabilizing the voltages between the last dynodes of the photoelectronic multiplier, characterized in that one of the dynodes is selected, by selecting the potential, the operating point of the selected dynode is set on the falling branch of the dynode characteristic, if the output pulse non-flatness factor is greater than one, or on the upward branches, if the non-flatness coefficient is less than unity, while simultaneously with the arrival of a light pulse at the selected dynode, They produce a voltage pulse, the value of which at each moment of time is proportional to the charge removed from this dynode, while the voltage amplitude at the time of the end of the light pulse is set equal to the voltage drop across the selected dynode, thereby reducing or increasing the gain of the photoelectronic multiplier by the value of the non-flatness coefficient output pulse of a photomultiplier tube. 2. The method of correcting the frequency response of the photoelectronic multiplier according to claim 1, characterized in that the voltage pulse is formed by integrating the current of the selected dynode on the capacitor connecting the selected dynode and its adjacent dynode, while the capacitance of the capacitor, taking into account the stray capacitance, is established from the condition

where I _d is the maximum current of the selected dynode corresponding to the linearity limit of the output characteristic of the photoelectronic multiplier, t _u - the maximum duration of the light pulse, ΔU is the voltage drop across the dynode characteristic, at which the gain of the photoelectronic multiplier is changed by the value of the non-flatness coefficient of the output pulse of duration t _u , Q _d is the maximum charge of the selected dynode corresponding to the limiting charge of the output characteristic of the photoelectron multiplier. 3. The method of correcting the frequency response of the photoelectronic multiplier according to claim 1, characterized in that the dynode and its operating point are selected from the condition of ensuring a minimum deviation of the shape of the output pulse from the rectangular when the photocathode is illuminated by a rectangular light pulse. 4. A method for correcting the frequency response of a photoelectronic multiplier according to claim 3, characterized in that the light pulse intensity is selected sufficient to obtain the maximum anode current.

Images