GB2080944A

GB2080944A - Radiation Intensity Counting System

Info

Publication number: GB2080944A
Application number: GB8120859A
Authority: GB
Original assignee: UK Secretary of State for Defence
Current assignee: UK Secretary of State for Defence
Priority date: 1980-07-10
Filing date: 1981-07-06
Publication date: 1982-02-10
Also published as: GB2080944B

Abstract

A radiation ratemeter circuit excludes the natural dead-time of the detector, eg a Geiger-Muller tube, from the measurement and hence dead-time correction is not needed. Following each detector-actuating event in tube 1 an artificial dead-time, longer than the natural dead-time is applied by pulse generator 4. The circuit aggregates those periods of time other than the artificial dead- times, eg by counting clock-pulses, from clock 7 in register 8 and after a given period of continuous real time, divides the number of counted events in register 9 by the thus aggregated "livetimes" to give a count-rate value free from dead-time error. This value is updated after periods of real time. The latter period may be a defined period of elapsed real time, or may be the elapsed time taken to reach a defined aggregated time. Preferably the detector cannot be activated during the artificial dead time. <IMAGE>

Description

SPECIFICATION Improvements in or Relating to Radiation Counting Methods and Circuits This invention relates to radiation counting methods and circuits in which the radiation detector gives a discrete output signal for each radiated entity (e.g. each beta particle, gamma photon, neutron, etc) which actuates the detector, and has one application in ratemeters employing detectors which are actuated by the ionisation produced therein by such entities, e.g. Geiger Muller tubes.

In conventional radiation counting circuits it is normal that after an ionising entity has actuated the detector, e.g. a Geiger-Muller (G-M) tube, the detector remains in an ionised condition for a period of time, during which it is insensitive to further incident radiation. The number of counts recorded in a given time will be reduced because of the time during which the detector was in the insensitive or "dead" condition.

It is normal present practice to introduce a correction factor if an accurate result is required.

Denoting the counting period as T and the detector "dead-time" as T, for n counts recorded the detector will have been dead for a total time nT. Hence, the actual sensitivetime during the counting period was T-n T, and a correction may be introduced thus: Corrected count in time T T=nx T-n The corrected count is, of course, still subject to statistical variation due to the random time distribution of ionising entities.

When the count is recorded by a scaler, it is usual to apply the correction manually; where a ratemeter is used, the correction can be incorporated into the instrument scale-marking, and for some purposes the resulting scale compression of the higher readings is useful.

In any case, the correction can only be derived if the dead-time T is known. As the dead-time of a G-M tube itself is somewhat uncertain, it is hormal to use the tube in a circuit which introduces an artificial dead-time. The dead-time can then be made more precise, thus improving the validity of the correction. In such a circuit it may be arranged, for example, that each actuating entity causes the polarising voltage on the G-M tube to fall to a low level at which the tube is insensitive, for a defined time. Such circuits are commonly known as "quenching circuits" even when used, for the purpose of accurately defining the dead-time, with tubes which would otherwise be self-quenching by virtue of the composition of the gas-filling.In such circuits it may be arranged that each tube output pulse immediately generates a dead-time defining pulse which is applied to reduce the tube polarising voltage to an insensitive level (as for quenching those tubes which are not self-quenching) for a period defined by the duration of the pulse and exceeding the tube's natural dead-time. Alternatively or additionally, the artificial dead-time can be produced by disregarding the tube output for such a period immediately following each output pulse, as by gating or clamping the subsequent signal channel.

The above-described method of correction has several limitations, however, including the following: (a) The need to apply dead-time correction is a complication if it is required to present the corrected count or count-rate in digital form.

(b) At high radiation levels the detector is dead for a high proportion of the time and a large correction factor is needed. Under these conditions the corrected count is highly dependent upon the value determined for the correction factor; however as the radiation level rises, so does the uncertainty in the derived correction factor. This can be seen from the expression for this factor, i.e. T/(T-nT), given above. At very high radiation levels, n tends towards (but cannot exceed) TIT. As this happens, nT approaches T and the denominator is determined by the small difference between the two terms. The uncertainty therefore increases even if the dead-time T is known accurately, which may not always be the case.

It is an object of the present invention to overcome or alleviate the above difficulties, and to permit a true count or count-rate to be displayed linearly and directly, without the need to introduce dead-time correction. As a result, useful measurements may be made (using a given type of detector) at higher count-rates than hitherto: alternatively, for the same upper limit of measurement a more responsive detector may be used, thus extending the lower limit of measurement.

According to the present invention, a radiation counting method using a detector which produces a discrete output signal for each radiated entity which actuates the detector, comprises: creating an artificial dead-time immediately following each actuation of the detector; Counting the number of output signals produced by the detector in a given period of continuous real time: aggregating those periods of time during said period of continuous real time which are other than artificial dead-times; and treating said aggregated time as the true period of time during which the counter number of output signals occurred.

Preferably the method comprises arranging that the detector cannot be actuated by a said entity during the artificial dead-time.

The given period of continuous real time during which said periods of time are aggregated may be a defined period of elapsed real time (e.g. one second by the clock). Alternatively, said given period of continuous real time may be the elapsed real time taken to reach a defined aggregated time (e.g. the elapsed real time taken for the aggregated periods of time to total one second, which will be longer than one second by the clock).

The true count-rate, i.e. the countrate free from dead-time error, may be obtained by dividing said counted number of output signals by said aggregated time, or directly as the number of output signals counted in a unit of said aggregated time.

The method may comprise obtaining the true count-rate as aforesaid repetitively over a succession of said given periods of continuous real time, expressing the thus-obtained count-rate for each said period at the end of each said period,and successively updating the expressed count-rate as the thus-obtained count-rate varies.

Also according to the present invention, a radiation counting circuit for use with a radiation detector which gives a discrete output pulse for each radiated entity which actuates the detector comprises: means operable by each said output pulse for producing an artificial dead-time immediately following each output pulse; and means for aggregating those periods of time during a counting period which are other than artifical dead-times.

Preferably the means for producing the artificial dead-time is adapted to reduce an energising supply for the detector to a level at which the detector cannot be actuated.

The circuit may additionally comprise means for terminating the counting of said pulse either after a predetermined time by the clock, or after the aggregated periods of time have reached a predetermined total.

A ratemeter circuit according to the present invention comprises: means connectable to a detector which gives a discrete output pulse for each radiated entity which actuates the detector, said means being operable by each output pulse to produce an artificial dead-time immediately following each output pulse; means for counting the output pulses and storing the count; means for aggregating the periods of time during said pulse-counting which are other than artificial dead-times and for storing the aggregated time; means for repetitively terminating said pulsecounting either after a predetermined elapsed time, or after the aggregated time reaches a predetermined value, and, after each transmission, for restoring said count-storing means and said time-aggregating means to their initial conditions and restarting said pulsecounting and time-aggregation;; and means for repetitively obtaining, after each said termination, a quantity representing said stored pulse-count divided by said aggregated time, said quantity constituting the count-rate.

One form of ratemeter circuit embodying the present invention, for use with a Geiger-Muller (G-M) detector, comprises: a pulse-generator operable by each G-M detector output pulse to generate a dead-time pulse whose duration defines a given dead-time; means for applying said dead-time pulses to reduce the detector HT supply during each deadtime pulse to a level which renders the detector insensitive to radiation; a clock-pulse source and a first register arranged to receive clock pulses under the control of said dead-time pulses so that the register does not receive clock pulses occuring during said dead-time pulses; a second register for receiving the detector output pulses; a divide circuit connected to divide the contents of the second register by the contents of the first register;; and connections from the clock-pulse source for repetitively performing the following functions after each successive equal period of elapsed time, viz terminating the supply of detector output pulses to the second register, dividing the contents of the second register by the contents of the first register, clearing both registers, and restoring the supply of detector output pulses to the second register.

The means for applying the dead-time pulses to reduce the detector HT supply may comprise an HT gate connectable between the G-M detector and an HT supply and controlled by said dead-time pulses; said HT gate may also be controlled by said clock-pulse source to effect said termination of the supply of detector output pulses to the second register by likewise reducing the detector HT supply.

Usually the G-M detector will be of the selfquenching type, in which case the duration of the generated dead-time pulses will exceed the natural dead-time of the detector when operating in a self-quenching mode, as in the dead-time defining arrangements described in the present introduction. However, the G-M detector may alternatively be one which is not self-quenching, in which case the generated dead-time pulses provide the conventional quenching pulses necessary for such tubes.

The circuit may further include a multiply circuit which receives the output of the divide circuit and multiplies it by a calibration factor, a store for receiving the calibrated output, and a display unit for displaying, during each successive period, the output entered into the store at the end of the preceding period.

The circuit may also include the said HT supply.

To enable the nature of the present invention to be more readily understood, attention is directed, by way of example, to the accompanying drawings wherein: Fig. 1 is a block schematic diagram of a Geiger- Muller ratemeter circuit embodying the present invention; Fig. 2 is a diagrammatic pulse-train illustrating the operation of the circuit of Fig. 1.

Fig. 3 is a plot of corrected against observed count-rate in a conventional GeigerMuller ratemeter circuit.

In Fig. 1 a G-M tube 1 is polarised by an HT supply 2 via an HT gate 3. The GM output pulses are fed to a pulse-generator 4 which generates a dead-time pulse for each G-M output pulse received, and controls gate 3 to reduce the G-M polarising voltage to a low value, which may be zero, thereby-rendering it insensitive to actuation by radiation during the artificial dead-time defined by the deadtime pulse. The latter dead-time is made longer than the natural dead-time of the G M detector (assumed to be of the usual selfquenching type) due to its ionised condition. The artificial dead-time is indicated by the duration D of the pulses in Fig 2.

The pulses from the circuit 4 are also fed, via a logic inversion circuit 5, to an AND circuit 6 which also receives clock pulses from a source 7. The output of circuit 6 thus consists of those clock pulses occurring other than during the dead-time pulses, i.e. those clock pulses occurring during the periods A in Fig. 2. These clock pulses are received by a first register 8, while the G-M output pulses are received by a second register 9.

The outputs of both registers are fed to a divide circuit 10 arranged to divide the contents of register 9 by those of register 8.

A connection from clock source 7 to gate 3 causes the G-M polarising voltage to be reduced (as during each dead-time pulse) momentarily at the end of successive equal periods of elapsed time (the periods C in Fig. 2), thereby interrupting the counting of ionising events by tube 1. At the same time a connection from source 7 to divider circuit 10 causes the latter to divide the contents of register 9 (i.e. the number of G-M pulses received during a period C) by the contents of register 8 (i.e the number of clock pulses representing the aggregated periods A during the same period C that tube 1 was sensitive to radiation); also, connections from source 7 to the two registers clear the latter of the received G-M and clock pulses, ready for those arriving during the next period C.

It will be understood that Fig. 2 is diagrammatic and not to scale; in particular there may be a much larger number of dead-time pulses D per counting period C than the average of four shown for clarity.

The quotient produced by circuit 10 is passed to a multiply circuit 11 where it is multiplied by a calibration factor to correct it to a convenient unit, e.g. Roentgens per hour (R/h), and thence passed to a digital store 12 whose contents are continuously displayed on a digital display 13.

The latter thus displays during each period C the count-rate (converted to a convenient unit) obtained during the immediately preceding period C, the displayed value being updated at the end of each such period. The calibration multiplier 11 can be omitted if not required.

After the momentary reduction of the G-M polarising voltage at the end of period C to allow the above operations to occur, the clock source 7 restores the polarising voltage to commence the next counting period C.

Suitably the duration of the period C may be of the order of one or a few seconds.

The quotient obtained from divide circuit 10 is a true count-rate which does not require a deadtime correction and requires no further manipulation before display other than the optional introduction of a calibration factor.

As seen above, the quantity representative of the count-rate has the form of a quotient, viz G-M pulses counted during period C aggregated periods A during period C.

At low count-rates the denominator approximates to period C and is substantially constant, i.e.

independent of the radiation level, and the quotient is thus approximately proportional to the number of G-M pulses counted. At high countrates the numerator tends to saturate-out, approaching a value C/D; however in this regime the denominator becomes an inverse function becoming very small at extreme radiation levels.

Thus the quotient becomes correspondingly large at the latter levels and so remains substantially a linear function of the true count-rate.

At high radiation levels there is a significant probability that an ionising event will coincide with the restoration of the polarising voltage to the G-M tube at the beginning of each period C.

Desirably the circuit design is such that in such a case the ionising event is recognised and the dead-time pulse correctly generated; otherwise the tube might become locked into its conductive state until the end of the relevant period C or until its self-quench action operates. In either case a false reading could result. For this reason it may be preferable to aggregate the "live-times" A by arranging that the circuit senses the detector polarising voltage and counts those clock pulses occurring when the full polarising voltage is present on the detector, rather than those clock pulses occurring other than during the generated dead-time pulses as in the above-described embodiment.

Another way of alleviating this difficulty is to apply an arbitrary train of HT-reducing pulses to the tube, e.g. generated from the clockpulse source, in addition to those generated by the ionising entities: this method can be applied in conjunction with, or instead of, using a selfquenching tube. The train is, of course, similarly applied to prevent the receipt of clockpulses by register 8 during the arbitrary pulses, since they effectively constitute further artificial dead-time pulses.

In the described embodiment, the dead-time after each ionisation is defined by rendering the detector itself insensitive, viz by reducing its HT supply. An alternative is to define the dead-time by simply disregarding the detector output for a defined time after each ionisation, e.g. by gating the detector output or clamping a subsequent pulse-amplifier (not shown) which receives its output. It is preferred, however, to de-sensitise the detector itself, by reducing its HT supply as described, because otherwise the dead-time must be made longer than the self-quenching time of the detector, which may be undesirably long.

However, it may be advantageous to de-sensitise the detector and also to gate or clamp its output simultaneously, in order to avoid the registration of any spurious pulses arising from some other unintentional mechanism. The preference for desensitising the detector does not extend to terminating the count at the end of each period C, which can equally be done by gating the input to register 9.

By way of.comparison, a small G-M tube may have a natural dead-time of about 25 ,usec and a response of 70 pulses per sec at a radiation level of 1 R/h. If this tube is operated in a conventional ratemeter circuit with an artificial dead-time of 50 ,usec then at 1000 R/h the correction factor becomes 4.5, i.e. the tube actually delivers about 15,000 pulses per second whereas the true count would be 70,000.

Fig. 3 shows the corrected count-rate plotted against the observed count-rate under these conditions. It is apparent that the result of using so large a correction factor will not be very reliable and will be sensitive to any variation of the dead-time. The present invention avoids these difficulties; in particular the constancy of the artificial dead-time is not important.

As seen, the present invention requires aggregation of the periods A between the end of each dead-time pulse and the beginning of the next. The durations of such periods may be estimated using the Poisson distribution for the occurrence of random events in a scale of time. It can thereby be shown that for the case considered with reference to Fig. 3, i.e. a true count-rate of 70,000/sec at 1000 R/h, the resolution of the aggregated time should be of the order of one ysec. It is also necessary that the circuit should respond rapidly to an ionising event, otherwise the end of each period A may be prolonged, excessive aggregated time thus recorded, and low readings obtained. This effect may limit the high-level capability of a given circuit.

In the described embodiment the periods A are aggregated over successive fixed, equal, counting periods C. In other forms of the invention the periods A may be aggregated until a fixed aggregated time is reached, say one second; the denominator in the aforesaid quotient then becomes a constant but the counting periods C vary in length, increasing as the count-rate rises.

In such an arrangement the count-rate may be displayed without the need for division, e.g. as the number of pulses registered in one second of aggregated time.

Although described with reference to a Geiger Muller tube circuit, the invention may also be applicable to ratemeter circuits usingother forms of radiation detector where a correction for deadtime would otherwise be required.

Claims

1. A radiation counting method using a detector which produces a discrete output signal for each radiated entity which actuates the detector comprising: creating an artificial dead-time immediately following each actuation of the detector; counting the number of output signals produced by the detector in a given period of continuous real time; aggregating those periods of time during said period of continuous real time which are other than artificial dead-times; and treating said aggregated time as the true period of time during which the counted number of output signals occurred.

2. A method as claimed in Claim 1 wherein it is arranged that the detector cannot be actuated by a said entity during the artificial dead-time.

3. A method as claimed in Claim 1 or Claim 2 wherein said given period of continuous real time is a defined period of elapsed real time.

4. A method as claimed in Claim 1 or Claim 2 wherein said given period of continuous real time is the elapsed real time taken to reach a defined aggregated time.

5. A method as claimed in any preceding claim wherein a true count-rate is obtained by dividing said counted number of output pulses by said aggregated time.

6. A method as claimed in any of Claims 1 to 4 wherein a true count-rate is obtained as the number of output signals counted in a unit of said aggregated time.

7. A method as claimed in Claim 5 or Claim 6 wherein the count-rate is obtained repetitively over a succession of said given periods of continuous real time and the thus-obtained count-rate for each said period is expressed at the end of each said period, the expressed count-rate being updated as the thus-obtained count-rate varies.

8. A radiation counting circuit for use with a radiation detector which gives a discrete output pulse for each radiated entity which actuates the detector comprising: means operable by each said output pulse for producing an artificial dead-time immediately following each output pulse; and means for aggregating those periods of time during a counting period which are other than artificial dead-times.

9. A circuit as claimed in claim 8 wherein the means for producing the artificial dead-time Is adapted to reduce an energising supply for the detector to a level at which the detector cannot be actuated.

10. A circuit as claimed in Claim 8 or Claim 9 wherein the circuit additionally comprises means for terminating the counting of said pulses after a predetermined period of elapsed real time.

1 A circuit as claimed in Claim 8 or Claim 9 wherein the circuit additionally comprises means for terminating the counting of said pulses after the aggregated periods oftime have reached a predetermined total.

12. A ratemeter circuit comprising: means connectable to a detector which gives a discrete output pulse for each radiated entity which actuates the detector, said means being operable by each output pulse to produce an artificial dead-time immediately following each output pulse: means for counting the output pulses and storing the count; means for aggregating the periods of time during said pulse counting which are other than artificial dead-times and for storing the aggregated time; means for repetitively terminating said pulsecounting either after a predetermined elapsed time, or after the aggregated time reaches a predetermined value, and, after each termination, for restoring said count-storing means and said time-aggregating means to their initial conditions and restarting said pulse-counting and timeaggregation;; and means for repetitively obtaining, after each said termination, a quantity representing said stored pulse-count divided by said aggregated time, said quantity constituting the count-rate.

13. A circuit as claimed in Claim 12 wherein the means for producing the artifical dead-time is adapted to reduce an energising supply for the detector to a level at which the detector cannot be actuated.

1 4. A ratemeter circuit for use with a Geiger Muller (G-M) detector comprising: a pulse-generator operable by each G-M detector output pulse to generate a dead-time pulse whose duration defines a given dead-time; means for applying said dead-time pulses to reduce the detector HT supply during each deadtime pulse to a level which renders the detector insensitive to radiation; a clock-pulse source and a first register arranged to receive clock pulses under the control of said dead-time pulses so that the register does not receive clock pulses occurring during said dead-time pulses; a second register for receiving the detector output pulses; a divide circuit connected to divide the contents of the second register by the contents of the first register;; and connections from the clock-pulse source for repetitely performing the following functions after each successive equal period of elapsed time, viz terminating the supply of detector output pulses to the second register, dividing the contents of the second register by the contents of the first register, clearing both registers, and restoring the supply of detector output pulses to the second register.

1 5. A circuit as claimed in Claim 14 wherein the means for applying the dead-time pulses to reduce the detector HT supply comprises an HT gate connectable between the G-M detector and an HT supply and controlled by said dead-time pulses.

1 6. A circuit as claimed in Claim 1 5 wherein said HT gate is also controlled by said clock-pulse source to effect said termination of the supply of detector output pulses to the second register by likewise reducing the detector HT supply.

1 7. A circuit as claimed in any of Claims 14 to 1 6 including a G-M detector of the self-quenching type and wherein the duration of the generated dead-time pulses exceeds the natural dead-time of the detector when operating in the selfquenching mode.

18. A circuit as claimed in any of Claims 14 to 1 7 further including a multiply circuit which receives the output of the divide circuit and multiplies it by a calibratipn factor, a store for receiving the calibrated output, and a display unit for displaying, during each successive period, the output entered into the store at the end of the preceding period.

19. A circuit as claimed in any of Claims 14 to 18 which also includes the said HT supply.

20. A radiation counting method substantially as hereinbefore described with reference to the accompanying drawings.

21. A ratemeter circuit substantially as hereinbefore described with reference to the accompanying drawings.