NZ223394A - Electric pulses treat micro-organisms: pulse width held constant at optimum value - Google Patents

Abstract

In the treatment of substances and/or microorganisms with electrical impulses, deviations from the optimum attainable effect occur again and again in continuously working systems, in particular if the composition (for example water content) of the substances to be treated changes. It is proposed to determine an optimum value of the time constant determined for the discharge, in which the desired effect of the electrical impulse on the substance to be treated is maximal. The capacitants or the impedance of the system is then adjusted to the theoretical time constant and the actual time constant is kept constant by appropriate matching of the impedance of the system or the capacitance of the discharge capacitor.

Description

. . -''J.f'Au'''' -?">S">' ...J.-., 22 3 394 r>. vj' V / ! O Priority Datcls^: ... .L"5.1 ^>1 .^7?.

Complete Specification Filed: .* Class: A?"?).

.QSA as. t^/<22.; A-\S&*?£/ S5?#DJr7j.C9.^.ir.\ Publication Date: ... 2.1.M1W P.O. Journal. No: . tag.-.

Patents Form No. 5 NEW ZEALAND PATENTS ACT 1953 »i i f i' t COMPLETE SPECIFICATION METHOD AND DEVICE FOR TREATING SUBSTANCES AND/OR MICROORGANISMS WITH ELECTRIC PULSES I/J^ HEINZ DOEVENSPECK a German citizen of Sigurdstr. 1, D-4950 Minden/ West Germany, hereby declare the invention, for which I/y£" pray that a patent may be granted to me/y£, and the method by which it is to be performed, to be particularly described in and by the following statement: (followed by page la) I $ „,,»s-i r*> C ia Description The invention relates to a method for treating substances and/or microorganisms, with electric pulses in a continuously operating system, the substances to be treated being exposed to pulses of defined field strength, the variation with time of which is determined by discharging at least one storage capacitor and by the impedance of a discharge path, and to a device for treating substances and/or microorganisms with electric pulses in a continuously operating system, with at least one capacitor which can be charged via a direct voltage and can be connected via a controlled switch to electrodes for discharging. 223394 I From German Patent Specification 1,237,541, an electric pulse method is \ known which is suitable for treating organic and inorganic substances in many I j cases. Particular applications of this electric pulse method lie in obtaining | ! individual components from disperse systems which mainly has a significant | importance for breaking down foodstuffs, for example in sugar extraction, and ; the separation of the components (protein, fat, water and so forth) of meat j ! products and so forth. It is also known from this printed document that | bacteria (microorganisms) can be destroyed by means of electric pulses. This | is of great significance, for example, for preserving foodstuffs or for drinking water and waste water processing. It is also known that chemical reactions can be started by electric pulses of the type described there and in this manner J molecule chains of organic and inorganic substances can also be grown.

With respect to the effect on microorganisms, New Zealand Patent Specification 219155, describes a stimulating effect on the metabolism of the 223394 microorganisms which is of great interest, for example, in the production of methane gas.

In this electric pulse method, the substances/microorganisms present in an electrolyte or the substances which themselves form an electrolyte are exposed to brief electric fields which are produced from the discharge of one or several capacitors along a discharge path formed by at least two electrodes. In this context, it has been found to be important that the electric pulses have steep rising edges for the production of which, for example, a circuit according to German Patent Specification 1,233,958 is suitable. The capacitor is discharged via a controlled switch, namely a thyratron. It is particularly advantageous in this connection if the substances to be treated are at the same time circulated as is described, for example in European Patent Specification A0063791.

It has hitherto been described as being particularly important or as the only important factor that the electric pulses exhibit a defined energy density, that is to say the maximum voltage has a quite decisive influence on the effectiveness of the method. An increase in effectiveness is additionally achieved by the fact that the electric pulses act on the substance to be treated in successive pulse groups (possibly of different amplitude) as is described, for example, in European Patent Specification AO 148380.

Although all the systems described above are quite operable and can be used with advantage, it has been found in continuous operation that changes in the "efficiency" of the systems or of the methods continue to occur in continuous operating mode.

On the basis of the above prior art, the present invention has the object of developing method and device of the type initially mentioned in such a manner that an increase in the effectiveness of the electric pulses can be achieved in continuous operation.

According to one embodiment of the invention the method, this object is achieved by the fact that an optimum value of the nominal time constant determining the discharge is determined at which the desired effect of the electric pulses on the treated substances occurs or at which the corresponding parameter is at a maximum, that the capacitance and/or impedance is set to the nominal time constant and that the time constant is kept constant independently of the impedance of the system.

Surprisingly, it has been found that the "decay constant" of the discharge represents a significant factor with respect to the effectiveness of the method. If then the unavoidable changes in the composition of the substance to be treated lead to a change in the impedance of the system, that is to say the resistance between the discharge electrodes, in continuous operation, this time constant is kept constant in spite of the change in impedance, in accordance with the present invention.

To keep the time constant constant in spite of a change in the impedance, it is then possible, in accordance with one embodiment of the invention, to generate the voltage via a current source the internal resistance of which is very high compared with the impedance between the discharge electrodes. However, since the discharge voltages are in the kV range, extensive equipment engineering measures are possible for this purpose if it is not intended to use a simple series resistance with its inherent efficiency-reducing characteristics. According to a preferred embodiment of the invention, therefore, a control process is performed in which the capacitance of the storage capacitor (or of the storage capacitors) is corrected to a value at which the desired discharge time constant is set, taking into consideration the instantaneous impedance of the system. In another preferred embodiment of the invention, the time constants are corrected by adding series resistors. Naturally, it is also possible to use both solutions in parallel or simultaneously, in which arrangement a coarse/fine calibration can be performed via the capacitors or — via the resistors.

The parameter to be maximized - for example the quantity of methane gas produced per unit time or the ratio between germs destroyed and germs not destroyed - is preferably measured at the system output and the time constant is changed until the parameter reaches its maximum value. The value then present is characteristic of the substance to be treated and is stored as nominal time constant and used as basis for the subsequent calibration processes. During the optimization, trial and error methods are advantageously used, small changes in the nominal time constant being tried more frequently than large changes (random distribution).

The time constant actually occurring with a current pulse is compared with the time constant thus obtained and from the difference a correction value is formed by means of which the capacitance of the capacitor and/or of the series resistance is changed before the next capacitor discharge occurs. By using this mode, the capacitors or the resistors can be switched in current-less condition which is much simpler than continuous control.

To obtain the most accurate possible control, it is of advantage if the mean value of the instantaneous time constant is formed over a particular number of pulses and this mean value is used as a basis for the further parameter determinations.

According to the device, the object is achieved by the fact that a calibration device is provided which is connected, at least during the discharge process, to the discharge circuit for sampling the progress of the discharge which exhibits means for comparing the discharge process with a predetermined value and means for matching the discharge process to the predetermined value. ... . ' 22 3 3 9 4 -4^ :f bs ■f $ - 'v ' 4 f*) .# Preferred embodiments of the invention are obtained from the subclaims and "the subsequent description of illustrative embodiments which are explained in greater detail with reference to figures, in which: J ) ! jO\ 5 Figure 1 shows a diagrammatic representation of a system for carrying out the electric pulse method, i Figure 2 shows a diagram sketch of a first embodiment of | a device for generating electric pulses, ,'i o: j Figure 3 shows a circuit of the calibration device indicated ' ; 10 in Figure 2, i | Figure 4 shows a diagram for explaining the operation of :j the device according to Figure 3, ] Figure 5 shows another preferred embodiment of the inven- . I ="! tion in a representation according to Figure 2 , '! I Figure 6 shows a circuit of the calibration device shown i in Figure 5, "j Figure 7 shows another preferred embodiment of the inven- .■"I /""> j } tion in a representation according to Figure 2, and Figure 8 shows the circuit of a calibration device as can 20 be used in the circuit according to Figure 7. i (^j In the system shown in Figure 1, a reactor 10 is provided via the inlet 15 of which a substance to be treated, for example waste water, can be pumped in a labyrinth path (see arrows) between electrodes 11, 12, 11', 12' through 25 to an outlet 16. A drain 17 projecting downward is attached in the vicinity of the outlet 16. At the roof of the reactor 10, a gas line 18 with a gas valve 19 is provided. The electrodes 11, 12, 11", 12* are introduced into the latter via insulating passages through the wall of the reactor 30 10.

The electrodes 11,12 are connected to a capacitor via a change-over switch 22 in its first switch position. In the second switch position of the change-over switch 22, the capacitor Cq is connected to a power supply 21 via which it can be charged up to a defined voltage with direct current. The change-over switch 22 is controlled via pulse control system 20 which determines the times at which pulses are emitted. In Figure 1, the change-over switch 22 is constructed as a mechanical switch but, in practice, it is advantageously constructed as an electrically controllable switch (ignitron, thyristor and so forth) corresponding to the high voltages used here. In the system shown in Figure 1, only a single capacitor is provided but it is of advantage to couple several electrode pairs along the path to various capacitors in order to allow various voltages to become effective. In this arrangement, the impedance of the system is primarily determined by the area of the electrodes 11, 12, their spacing AI and A2 from one another and by the conductivity of the medium pumped through. In this arrangement, the conductivity generally has a complex value.

In Figure I, the discharge capacitor CQ was represented as fixed unchangeable capacitor and the impedance of the system can also not be adjusted.

According to the present invention, an arrangement as described in Figure 2 is selected instead of the fixed capacitor. This illustration shows that, in parallel with the capacitor C^, a calibration device 25 is provided, one input A of which is connected to the output of the power supply 21 which, in the present case, is stabilized via a choke 23 and a filter capacitor 24 and the other input of which is connected to one electrode 12. In the text which follows, the operation of the calibration device 25 is described in a first embodiment with reference to Figure 3.

The input A of the calibration circuit 25 leads to a voltage divider Rj, R^, the dividing ratio of which is 1/e. The divided voltage is connected to the inverting input.- 22 3 3 9 4 of an operational amplifier 26. The second input B of the calibration circuit 25 is connected via a protective resistor Rj, which does not exhibit any voltage-changing influence due to the input impedance of the subsequent operational amplifier 26, to the non-inverting input of the operational amplifier 26. Between the two inputs of the operational amplifier 26, diodes 0-j and D 2 are connected in opposite directions of conduction so that the voltage difference at the input of the operational amplifier 26 cannot exceed the forward diode voltage. It should be noted here that, naturally, a voltage divider circuit (equal dividing ratio for both inputs) can be provided in front of the inputs A and 3 so that the very high capacitor voltages can be reduced to the usual low levels.

The output of the comoarator circuit 45 formed in this manner, that is to say the output of the operational a m p I i -■ fier 26 is connected to the input of a measuring circuit 46. The measuring circuit 46 comprises at its input .an AND gate 27, one input of which is connected to the output of the operational amplifier 26 and the other input of which is connected via an adjustable divider to a clock oscillator 34. The divider 32 is connected to a nominal-value adjuster 33 via which the dividing ratio of the divider 32 can be adjusted.

The output of the ANO gate 27 is connected to the counting input Up of a binary counter 31. The counting outputs 1 to n of the counter 31 are connected to the inputs of an OR gate 30 the output of which is connected to another AND gate 29. A second input of the AND gate 29 is connected via an inverter 28 to the output of the operational amplifier 26.

The output of the AND gate 29 is connected to the input of a delay section 35 which, after a certain period of time after a positive voltage step has been applied, outputs a pulse at its output. This pulse is supplied to the reset input R of the counter 31.

-Av- '■ ' . , 'j.. f 22 3 3 9 4 X—\ V_J- - 8) The counting outputs 1 to n\of the binary counter 31 are i| also in each case connected to a 6-type flip floD 3 6 -j — '5 | 36n the C inputs of which are all connected to the output | of the AND gate 29.

• \ O) 5 The Q outputs of the D-type flip flco are connected to the excitation coils of relays 3 7 -| - 3 7 n which exhibit normally open contacts which are connected, on the one hand, to a (virtual) earth and, on :he other hand, to first poles of capacitors. The other poles of the capacitors are connec-10 ted together and brought as common connections D out of the calibration device 25. Thus, when the relay contacts close, the capacitors are connected in parallel with the capacitor Cq. The capacitance values of capacitors are selected in such a manner that the lowest digit 1 of the 15 counter 31 connects a capacitor having capacitance C in parallel with the capacitor Cg whilst the most significant digit n of the counter 31 corresponds to a capacitor having the value 2n x C.

In the text which follows, the operation of this circuit 20 is explained in greater detail with reference to Figures 4.

Figure 4A shows the variation of the voltage U of a discharge pulse versus time t. This pulse variation is present at input B of the calibration device 25. At input A the charging voltage, that is to say the output voltage 25 of the power supply 21, is present so that a threshold voltage Us is■present at the inverting input of the operational amplifier 26, which voltage corresponds to the 1/e value of the maximum voltage across capacitor Cg in accordance with the dividing ratio.

If the voltage U at the electrode 12 suddenly rises at time t0 the output of the operational amplifer 26 jumps to its maximum (positive) output voltage as is shown in Figure 4B. As soon as the voltage at the electrode 12 drops below the threshold Us (time t ■)), the output at 35 the operational amplifier 26 drops to its zero value or O 'V. 22 3 3 9 4 0 "\9 " to its negative value- As\long as a positive output voltage appears at the output of the operational amplifier 26, the AND gate 27 allows the pulses generated by the clock oscillator 34 and divided down by the divider 32 5 into the counting input of the counter 31. This is shown in Figure 4C.

During the time the counter 31 is counting up, a zero level is present at the output of the inverter 28 so that the AND gate 29 is inhibited as is shown in Figure 40.

As soon as the counter 31 has counted the first pulse, the output of the OR gate 30 jumps to a positive value as is shown in Figure 4E.

If the voltage at the electrode 12 drops below the threshold lls at time t -j, the output of the AND gate 29 jumps to 15 a positive value since the output of the inverter 28 jumps to a positive value at this time and the output of the OR gate 30 is still at a positive value. This positive output value, which is shown in Figure 4 F, in turn reaches the C inputs of the D-type flip flops 36 -] — 36 n so that 20 these flip flops accept the values present at outputs 1 to n of the counter 31 via their D inputs. At the same U/ time, the delay circuit 35 is triggered with the voltage step of the.AND gate 29, which delay circuit outputs an output pulse of specified length after a delay period tv 25 as is shown in Figure 4G. The output pulse of the delay circuit 35 resets the counter 31 so that the next measuring i i VJ process can begin, The output values of the counter 31 accepted into the D -type flip flops 36— 36n lead to the capacitors being 30 connected or disconnected in accordance with the value counted. Thus, this circuit directly measures the time constant and "translates" it into a time-proportional capacitance value. In this arrangement, the capacitor Cg is designed in such a manner that the time constant defined 35 by it is shorter than the desired time constant. If then 22 3 3 94 the first voltage pulse is measured when the system is taken into operation, a number of counting pulses which can be defined via the divider 32 runs into the counter 31 so that some capacitors are connected in parallel with the capacitor Cg after the end of the first voltage pulse. If this calibration has lead to the correct result, the same number of pulses will accumulate in the counter 31 with the next voltage pulse so that the switch position of the relays 37 -j — 3 7 n is not changed. If the calibration was not yet correct, a changed value is accepted which, finally, leads to an approximation to the desired value. In this arrangement, the desired value can be preadjusted via the dividing ratio of the divider 32 or via the nominal-value adjusters 33.

In the embodiment of the invention described with reference to Figures 5 and 6 in the text which follows, the calibra-' tion device 125 is connected in series with the capacitor Cg. In this calibration device, the compensation circuit 47 is constructed as series circuit of resistors R, 2 x R, 2n x R. These resistors can be included into the series circuit or short circuited via the relays 37 -j—37n. Thus, the decay constant of the discharge circuit changes depending on the change in resistance.

In this preferred embodiment of the invention, the measuring circuit 46 is modified compared with the previously described form by the fact that the outputs 1 to n of the counter 31 are additionally connected to a digital comparator 38 the other inputs of which are connected to a limit value adjuster 39. The comparator 38 compares the values counted in the counter 31 with the value set in the limit value adjuster 39 and then provides a signal to a warning lamp 40 when the set value is exceeded. By means of this arrangement, a warning signal can be output when the compensation circuit 47, for example, has reached its limit, that is to say all resistors are in series or all resistors are short circuited. All resistors are connected in series, for example, when a short circuit occurs between "4 >.7 ..ft 22 3 3 S V.-' - 11 - '■it ,2' n O o electrodes 11, 12, whereas all resistors are short circuited when the resistance between the electrodes 11, 12 assumes a very high value which is the case, for example, with no-load operation of the system. Thus, a warning with faults 5 in the overall system is possible with this preferred embodiment of the invention.

Furthermore, the divider 32 in the preferred embodiment of the invention described in Figure 6 is set via a microprocessor 4 1 which can be programmed via an input unit £2. 10 The outputs of a measurement transducer 43, which converts the output signals of a measurement sensor 44 into digital words are present at one input of the microprocessor 41. The sensor 44 is constructed and mounted in the reactor 10 in such a manner that it measures the parameter which is 15 to be optimised. If, for example, the reactor 10 is used for generating methane (bio-gas), the sensor 44 can be a flow sensor which, measures the volumetric rate of flow flowing in the gas line 18. The optimization process proceeds in such a manner that the microprocessor 41 initially 20 stores a measurement value of the sensor 44 and then sets the counter 32 to a value selected at random. After a certain period of time which is determined by the operating speed of the system, the microprocessor 41 again samples the measurement value present at the measuring 25 sensor 44 and compares this value with the measurement value previously obtained. If the new measurement value has changed in the direction of an optimization, compared with the previous measurement value, the new measurement value is retained as basis, which is (slightly) changed, 30 during the next change attempt for the divider ratio. If the measurement value obtained first was "better", the microprocessor 41 uses this first measurement value as a basis. An optimization can be achieved which proceeds particularly quickly (in a few steps) if the values selec-35 ted at random are distributed around the previously found optimum value in accordance with a Gaussian distribution, to such an extent that small changes are more frequent than large changes (principle of evolution). 22 3 3 9 4 In the text which follows, another preferred embodiment of the invention is described in greater detail with reference to Figures 7 and 8. In this embodiment of the invention, a modified calibration device 225 is provided which is also connected in parallel with the capacitor Cg (as in Figure 2) but has no further inputs. In this circuit, the voltage across the capacitor Cg is supplied via a diode D 4 to a capacitor CT, an appropriate polarization of the power supply 21 being assumed in this case. Thus, as soon as the capacitor Cq is charged up, the same voltage (minus the forward voltage of the diode D 4 ) is present across capacitor CT. If the capacitor Cg discharges, the point 0 is at a lower potential than the corresponding junction of the capacitor CT so that the diode !> 4 cuts off. In parallel with the diode D 4, a high-resistance resistor R3 is connected via which the capacitor CT can discharge very slowly which is necessary to allow any change in the mains voltage to become effective.

The capacitor CT is again followed by the voltage divider R-j, R2, the output voltage of which is present at the inverting input of the operational amplifier 26.

The same potential point 0 is connected via a protective resistor R5 to a high-pass filter consisting of capacitor CH and resistor R 4. The output of the high-pass filter is connected to the cathode of a diode D3 the anode of which is connected to the non-inverting input of the operational amplifier 26. Once the capacitor is then charged up, the maximum voltage (corresponding to the power supply output voltage) across the capacitor CT is then retained even during the discharging (with appropriate dimensioning of the resistor R3) so that the threshold voltage Us initially described is present at the inverting input of the operational amplifier 26. However, the drop in voltage across the capacitor Cq is conveyed via the high-pass filter C H / R 4 and the diode D3 to the non-inverting input of the operational amplifier 26 so that the pulse shape of the capacitor discharge curve shown in $ t V I? %■ H 'fe f f . I 22 3394 Figure 4 A is again present at this input Furthermore, it is possible to use, for examole, a decimal counter instead of a binary counter, in which arrangement the capacitance values must then be selected in accordance (} 5 with the decimal digits. In this arrangement, it is of advantage (both with binary and with decimal setting), if the values of the compensation capacitors or resistors, with an essentially known capacitance value of Cg for a known nominal time constant, are only selected in accordance V„ 10 with a fine calibration which can considerably reduce the number of capacitors and relays required.

All individual features described by themselves are claimed as essential to the invention, as is a combination of them which particularly applies to the combination of possibili-15 ties for changing the time constants (parallel-connection of capacitances, series-connection of resistors). The same applies to keeping the discharge curve shape constant by direct means, not shown in the illustrative embodiment.

Claims (19)

223394 WHAT I CLAIM IS:

1. A method for treating substances and/or micro-organisms with electric pulses in a continuously operating system, the substances to be treated being exposed to pulses of defined field strength, the variation with time of which is determined by discharging at least one storage capacitor and by the impedance of a discharge path, characterized in that an optimum value of the nominal time constant fC) determining the discharge length or a corresponding parameter is determined at which a desired effect of the electric pulses on the substance to be treated is at a maximum, that capacitance and/or impedance is set to give the nominal time constant (C) and that the time constant is kept constant independently of the impedance of the discharge path.

2. A method according to Claim 1, characterized in that the desired effect of the electric pulses on the substance is continuously measured at a system output and a time constant (T~) is changed until the desired effect reaches its maximum value and the value of the time constant (*£) then x present is stored as the nominal time constant.

3. A method according to Claim 2, characterized in that the capacitance of said at least one storage capacitor is changed for setting the time constant (Yj. x

4. A method according to Claim 3, characterized in that a plurality of storage capacitors are provided and the capacitance of said storage capacitors is set by connecting defined numbers of capacitors in parallel.

5. A method according to any one of claims 2-4, characterized in that the time constant fjp is set by setting the resistance of a resistance connected in series with said at least one storage capacitor and the discharge path.

6. A method according to Claim 5, characterized in that the time constant is set by measuring the actual time constant -of. one pulse, comparing X 11""" Jil - J 4 ...v.. .. . j ' * ' 223394 CT? - 15 - c i' © calculating a correction value (£) from the difference fC -*£) and changing the capacitance of the storage capacitor and/or the value of the series resistance in accordance with the correction value Cg) before the next capacitor discharge occurs.

7. A method according to Claim 6, characterized in that the correction value (X) is formed from the difference between the nominal time constant K C^) and the mean value of the actual time constants over a defined number of pulses.

8. A device for treating substances and/or microorganisms with electric pulses in a continuously operating system, with at least one capacitor which can be charged via a direct voltage and can be connected via a controlled switch to electrodes for discharging and means for repeatedly charging and discharging said at least one capacitor to generate said pulses, characterized in that a calibration device is provided which is connected to the discharge circuit, at least during the discharge process, for sampling the progression of the discharge, which calibration device has means for comparing the discharge length or a corresponding parameter with a predetermined time constant (V) and means for matching the discharge length 8 or corresponding parameter to the predetermined time constant CC). v _s 3

9. A device according to Claim 8, characterized in that the calibration device comprises a) a comparator which compares the voltage across the storage capacitor s. ^ during the discharging with a threshold (0a) which corresponds to a defined fraction of the capacitor voltage at the beginning of the discharge, b) a measuring device for determining the period of time (£) between the beginning of discharge and when the voltage drops below the threshold (Og), c) a compensation circuit for changing capacitance and/or impedance of the discharge circuit in accordance with the difference between the predetermined time constant Q^) and the measured time . it

10. A device according to Claim 9, characterized in that {I <5 & * r 663544/5 - 16 - the input of the comparator is connected to the charging direct voltage via a voltage divider and to an electrode.

11. A device according to Claim 9, characterized in that the input of the comparator is connected via a rapidly charging and slowly discharging voltage store and a voltage divider circuit to one pole of the storage capacitor and, via a high-path filter with subsequent rectifier to the pole of the storage capacitor.

12. A device according to any one of Claims 9 to 11, characterized in that the measuring device comprises a counter, the counter input of which is connected via an AND gate to a clock generator, the second input of the AND gate being connected to the output of the comparator circuit so that clock pulses can be counted within the period of time (^) between the beginning of the discharge and when the voltage drops below the threshold (Ug).

13. A device according to Claim 12, characterized in that a frequency divider, the dividing ratio of which can be adjusted via adjusting means, is arranged between the clock generator and the AND gate.

14. A device according to Claim 13, characterized in that the adjusting means comprise a computing unit which can be supplied with output signals of a measuring sensor which measures a parameter of the substances and/or microorganisms to be tested and which can be adjusted via an input keyboard.

15. A device according to any one of Claims 9 to 14, characterized in that the compensation circuit comprises a plurality of compensation capacitors and/or compensation vv . 17 . 223394 resistors which can be connected in parallel and/or in series fspf with the storage capacitor.

16. A device according to Claim 15 when appended to Claim 12 characterized in that the counter is a binary counter an output from which is connected via storage circuits to switches for actuating these circuits in order to connect and disconnect the compensation capacitors and/or resistors, and that the compensation capacitors and/or resistors exhibit values corresponding to the binary counter digits.

17. A device according to either of Claims 15 or 16, characterized in that the total impedance of the compensation capacitors and/or resistors is less than that of said at least one capacitor or of the impedance of the discharge circuit.

18. A method as claimed in Claim 1, substantially as herein described with reference to the accompanying drawings.

19. A device as claimed in Claim 8, substantially as herein described with reference to the accompanying drawings. O HEINZ 450EVENSPE bj^yhis attorneys BALDWIN, SON & CAREY Tt r) O *5£t- <£l 0+T>s;