GB1587473A

GB1587473A - Electronic maximum recording device

Info

Publication number: GB1587473A
Application number: GB22860/78A
Authority: GB
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 1977-07-07
Filing date: 1978-05-25
Publication date: 1981-04-01
Also published as: IT1097463B; DE2730776A1; JPS5420769A; DE2730776C3; IT7825348A0; DE2730776B2; CH628147A5

Abstract

The mechanism essentially consists of two shift registers (5, 8) to which an adder (6) and comparator (9) are allocated, a control mechanism (4), an indicating device (13) and a mechanical cumulative counting mechanism (19) and restoring-type counting mechanism (20). A method for operating the mechanism in work phases is specified. <IMAGE>

Description

(54) ELECTRONIC MAXIMUM RECORDING DEVICE (71) We, SIEMENS AKTIENGESELLSCHAFT, a German company of Berlin and Munich, Germany, do hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: This invention relates to a device for, and a method of, determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods.

According to one aspect of the present invention, there is provided a device for determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which device comprises: a) a first electronic store for recording the number of pulses in each metering period; b) a second electronic store for recording the maximum number of pulses recorded by the first store over the preceding metering periods; c) comparison means for comparing the value recorded in the first store with the value recorded in the second store at the end of each metering period; and d) transfer means (i) for transferring the value in the first store to the second store and subsequently clearing the first store, in the event of the value in the first store exceeding the value in the second store at the end of each metering period, and (ii) for simply clearing the first store, in the event of the value in the first store not exceeding the value in the second store.

According to another aspect of the present invention, there is provided a method of determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which method comprises: (1) recording the number of pulses in each metering period in a first electronic store; (2), at the end of each metering period, comparing the value recorded in the first store with a value recorded in a second electronic store representing the maximum number of pulses recorded by the first store in the preceding metering periods; (3), if the value in the first store exceeds the value in the second store, transferring the value in the first store to the second store and clearing the first store; and (4), if the value recorded in the first store does not exceed the value recorded in the second store, simply clearing the first store.

It is known to provide a pulse generating electricity meter for transmitting pulses whose frequency is proportional to the power metered with an electronic maximum recording attachment (or maximum-demand indicator and recorder), wherein the maximum mean power value of a number of such values recorded over a number of chronologically equal metering periods is added at the end of a read-off period to the maximum mean power values of preceding read-off periods.

Each read-off period may be, for example, one month. The maximum recording attachment is not, as a rule, assembled directly with the meter, but receives from the meter a pulse sequence proportional to the power metered. Known maximum recording attachments have a receiving device for converting the pulses to angle-steps. These steps then control rotary movement of a pointer via a transmission gear and a coupling. The deflection of the pointer corresponds to the electrical energy consumed during the metering period concerned. At the end of each metering period, the pointer is uncoupled and returned to its starting position. The pointer pushes in front of it a maximum pointer which then comes to rest at the maximum deflection reached in each metering period.

Simultaneously with the movement of the maximum pointer, a mechanical counting mechanism for numerically displaying the deflection of the maximum pointer is generally actuated. At the end of a read-off period, the maximum pointer and optionally also the maximum counting mechanism are restored to zero, the largest maximum value which has been recorded in the read-off period being fed to a cumulative counter. A modern maximum recording attachment has at least three counters, i.e. a metering period counter for numerically displaying the particular state of the first-mentioned pointer, a maximum counter for displaying the deflection of the maximum pointer in a particular metering period within a particular read-off period, and a cumulative counter in which the maxima of the preceding read-off periods are accumulated. However, there is generally provided also a further counter for recording the number of monthly accumulations. The outlay in respect of mechanical means in the case of the known mechanical maximum recording attachements is, therefore, very considerable, and it is generally also necessary to provide supplementary storage devices in order that pulses which are transmitted during the uncoupling time (which may amount to as much as 1% of the metering time) are not lost.

Furthermore, the mechanical elements are subject to wear, so that the service life of the known maximum recording attachments is short relative to the associated electricity meters.

In order that the present invention may be more fully understood, an embodiment of the device according to the present invention will now be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a block diagram of the device; Figure 2 is a more detailed diagram of a part of the device; Figure 3 is a more detailed diagram of a further part of the device; Figure 4 is an explanatory diagram; Figure 5 is a diagram representing a current supply for the device; Figures 6 and 7 are circuit diagrams of still further parts of the device; and Figures 8 to 15 are pulse diagrams for illustrating the operation of the device.

Referring to Figure 1, the embodiment illustrated is an electronic maximum recording attachment to which a pulse generating electricity meter 1 supplies pulses whose frequency is proportional to the power metered. The pulses JE from the meter 1 are supplied via a pulse shaper 2 for pulse value adjustment to a preferably electronic pulse transducer 3 for normalising the pulses. The pulses El which have been normalised by the transducer 3 are then processed in a control system 4 and added in a first store 5 with the aid of an adder 6.

They represent the "running" means power value. The period of time over which this summation takes place (15 minutes, 30 minutes or some other multiple of 15 minutes) is determined by a metering period control system 7. When the metering period has elapsed, a metering period pulse MP is supplied by the metering period control system 7 to the control system 4.

In addition to the first store 5, there is a second store 8 which contains the maximum value determined from previous metering periods. When the metering period has elapsed and the metering period pulse MP is supplied to the control system 4, a comparison is effected in comparison means in the form of a comparator 9 of the mean power value in the store 5 with the value in the store 8. If the mean power value stored in the first store 5 is greater than the value in the second store 8, then the value in the first store 5 is transferred as a new maximum value to the second store 8. Store 5 is subsequently cleared ready for the determination of the next mean power value. If the value in the store 5 is less than the value in the store 8, then the store 8 receives no new maximum value; the value stored in the store 8 therefore remains unchanged and the store 5 is cleared. Control of these operations is effected by the control system 4 via transfer means in the form of multiplexers 10 and 11 associated with the stores 5 and 8, as will be discussed in greater detail hereinbelow.

The values stored in the two stores 5 and 8 can be selectively displayed by a switch 12 on a 7-segment LED (light-emitting diode) display 13 with the aid of a further multiplexer 14 and a BCD (binary-coded-decimal) 7-segment decoder 15. In order to increase the service life of the display, it is switched off via a line 17 at the end of each metering period by the metering period pulse MP. The display is switched on by a display flip-flop 16 which is activated via an OR-gate 18 on change-over of the switch 12.

At the end of a read-off period comprising a plurality of metering periods, for example after a period of one month. the maximum value stored in the store 8 is supplied to an electromechanical cumulative counter 19, and an electromechanical reset counter 20 recording the number of values accumulated by the counter 19 is stepped-on by one step.

Triggering of the two counters 19, 20 which are preferably roller counters is effected by the control system 4 via a time-period monitoring system 21 in acknowledgement signal (Hand Shake) mode in response to a pulse MR for monthly resetting. In this mode the counters supply a (Hand Shake) signal to the system 21 when a signal supplied by the system 21 has been received and accepted by the counters. Thus the counters determine the rate of data transmission.

Since the end of a metering period does not necessarily coincide with the end of a read-off period, prior to transfer of the maximum value to the cumulative counter 19, the maximum value is compared with the mean power value of the metering period which has not yet elapsed. Thereby, is is ensured that the latest, if not yet fully ascertained, mean power value is taken into account in the formation of the maximum value. Subsequent to the accumulation step, the stores 5 and 8 are cleared. Monthly triggering of the accumulation step can be effected in various ways, for example (i) by manual change-over of a switch 22 or (ii) under the action of change-over means 23 forming part of a receiver which is controlled by a centralised multi-service control unit or (iii) by an electronic monthly reset unit 24 via an electronic control system 25.

By means of a control input 26, the result can be achieved that the maximum recording attachment effects maximum detection only at fixed times. These times may be pre-set, for example by an external time switch. As with monthly resetting, here again the end of a read-off period does not necessarily coincide with the end of a metering period. The mean power value of the metering period which has not yet been concluded is therefore compared with the maximum value and, if appropriate, adopted as a new maximum value.

Timing pulses T necessary for operating the maximum recording attachment are supplied by timing means in the form of an oscillator 27 acting on a frequency divider 28 constituted by a counter made up of bistable switches which counts modulo 4 (up to divide by 8). From this frequency divider 28, the signals for the control system 4 and the time-period monitoring system 21 are obtained. A four output decoder 29 is connected to the LED display 13 and activates this display in multiplex opertion.

As Figure 2 shows, there are employed for the stores 5 and 8 4x4 bit shift registers in which the stored data is cycled in the form of four tetrads, each tetrad representing a decimal digit and having its bits in parallel and the tetrads being in series, by means of pulses T. The frequency divider 28 determines the valency of the tetrads appearing at the outputs of the registers. Thus, in state 0, there appear at the shift register outputs tetrads having valency 100, in state 1 there appear tetrads having valency 101, in state 2 there appear tetrads having valency 102, and in states 3 there appear tetrads having valency 103.

As long as no input pulse EI arrives at the control system 4, the outputs of the stores 5 and 8 are short-circuited with their own inputs due to correspondence of the signals at the inputs m4 and m6 of the multiplexers 10 and 11. The continuously running pulses T thereby merely cycle the data stored in the values 5 and 8. For detecting the mean power value the outputs of the store 5 are connected each to a respective input of the 4bit adder 6. The other inputs of the 4-bit adder are held at "0". A "1" is added to the value in the store 5 by supplying to the "carry" input of the 4-bit adder 6 a signal C from a C-flip-flop 30 (carry flip-flop) (see Figure 3). The C-flip-flop 30 (first flip-flop) is actuated by an input pulse EI, as is apparent from Figure 3. If the output of the C-flip-flop is equal to "0", i.e. if there is no input pulse EI, then in the normal case, a "0" is continuously added in the adder 6. If a pulse which is to be added arrives, then it is added with the aid of the C-flip-flop 30 and the adder 6 to the value in the shift register 5.

If the tetrad present at the output of the store 5 is already "9" when a "1" is added, then are are written into the store inputs and the output of the C-flip-flop 30 is added to the subsequent tetrad. To this end outputs MSo (valency 1) and MS3 valency8) of the store 5 which carry the "9" are connected with corresponding inputs of the control system 4, as shown in Figures 2 and 3.

The comparator 9 is connected to the outputs of the two stores 5 and 8. The outputs of the comparator 9 carry signals ml and m2, depending on whether the value in the store 5 is greater or less than the value in the store 8. As Figure 3 shows, these signals serve to set the C-flip-flop 30 which, via the multiplexers 10 and 11, causes the outputs of the store 5 to be connected to the inputs of the store 8 for transferring data if the value in the C-flip-flop subsequent to the comparison is "1".

As Figure 3, shows, there is associated with the C-flip-flop 30 a W-flip-flop 31 (second flip-flop) intended to prevent a particularly long lasting input pulse El from being taken into account several times by the C-flip-flop 30.

An important part of the control system 4 is constituted by an X-flip-flop 32 and a Y-flip-flop 33. These two flip-flops define four states (so-called phases). These four phases 800, 810, 811 and 801 serve to control all the control procedures of the maximum recording attachment. In each case, the digits following the 8 denote particular states of the two flip-flops 32 and 33. Thus, for example, 810 indicates that the output of the X-flip-flop 32 carries an "L"-signal and the output of the Y-flip-flop 33 a "0"-signal. Figure 4 illustrates the phase development.

In the first phase 800 (also called the "waiting phase") the input pulses EI normalised by the electronic pulse transducer 3 are accumulated in the store 5. The store 5 is thus an instantaneous store. It is important to appreciate that operation of the maximum recording attachment is not based on counting processes performed by means of binary or BCD counters, but purely and simply on adding processes. Thus, on arrival of a pulse El, a "1" is added to the value in the store 5 by the C-flip-flop 30. As previously explained, the pulse EI initiates a "carry" pulse which is added to the lowest valency tetrad.

On arrival of a metering period pulse MP or a pulse MR for monthly resetting, the phase 800 is replaced by the phase 810. This phase 810 is divided into two portions characterised by the state of an H-flip-flop. The H-flip-flop constitutes part of the frequency divider 28 and, in fact, represents practically speaking the last link before the output which is designated H. As can be seen in Figure 8, during the first four pulse periods of the pulses T H = "0" and over the next four pulse periods H = "L". In the first portion of the phase 810 H = "0" and the value in the store 5 is compared with the value HM in the store 8 by means of the C-flip-flop. The output of the C-flip-flop 30 is initially "0". If the output of the C-flip-flop 30 is finally "0", then the value in the store 5 is equal to or less than that in the store 8 (MS S HM); if, on the other hand, the output of the C-flip-flop is finally "L", then the value in the store 5 is greater than the value in the store 8 (MS > HM). In this case, the value in the store 5 must be transferred into the store 8 in the second portion of the phase 810 (i.e. with H = "L"). As this is done, the store 5 must be cleared. Once this has been done, then in the event of a metering period pulse MP there will be a return to phase 800.

If, however, a monthly preset pulse MR occurs, then there will be a change to phase 811.

This phase is again subdivided into two portions according to the state of the H-flip-flop.

With the aid of the C-flip-flop 30, a "1" is then added to the empty store 5 in the first portion in which H = "0". Subsequently, there is a change to phase 801 in which the value in the store 5 is again compared with the value in the store 8. If the value in the store 5 is less than the value in the store 8, then a pulse is transmitted to the electromechanical counter 19 and there is a return to the phase 811, and, with the aid of the C-flip-flop 30, a further "1" is added to the value in the store 5. Subsequently, there is a return to the phase 801, in which the value in the store 5 is compared with the value in the store 8. If the value in the store 5 has not yet reached the value in the store 8, the counter 19 is once again incremented and there is a return once again to the phase 811. This cycle is repeated until the value in the store 5 is greater by one than the value in the store 8. There is then a change from phase 801 to phase 800 and the complete process is repeated.

Figure 4 indicates, by means of a downwardly directed arrow, that there is transfer from phase 800 to phase 810 when the X-flip-flop 32 is set by a metering period pulse MP or a pulse MR. The corresponding upwardly directed arrow indicates that there is return from phase 810 to phase 800 when the pulse MR has disappeared. A horizontal arrow indicates transfer from phase 810 to phase 811 when a pulse MR is available for monthly re-setting, which pulse sets the Y-flip-flop 33 and the C-flip-flop 30. There is transfer from phase 811 to phase 801 when the W-flip-flop 31 carries no signal and the X-flip-flop 32 is set at "0", as indicated by an upwardly directed arrow. From phase 801, there is a return to phase 811 when H = "L" and C = "0", as indicated by the corresponding downwardly directed arrow.

C = "0" signifies (as previously stated) that the value in the store 5 is less than the value in the store 8. As indicated by a horizontal arrow, there is a return from phase 801 to the waiting phase 800 when H and C are equal to "L". In this case, the Y-flip-flop is reset.

As Figure 3 shows, the output of the X-flip-flop 32 and the output of the Y-flip-flop 33 are carried to a four output decoder 34, for evaluating the states of these two flip-flops in the following manner: lit()() = X.Y, 8()1 = X.Y. 810 = X.Y. (1111 = X.Y. Hereinbelow. the Bool equations for setting and resetting the flip-flops 32 and 33 are indicated.

"X = (MP + MR).#00 + H.C.801 = = MR.#10 + W.011 = = MR.10 "Y = H.C.00l 801 Pulse W, X, Y = TO.Tl.H positive edge SetW = 200 = 811) "W = El.000 + BS.#11 Hereinbelow, the further equations for the maximum recording attachment are indicated, which equations are carried into effect by the control system 4 according to Figure 3.

Msvin = Sv. MS().MS3.C. (000+011)+MSv.000+011+H.010+H.C.001 HMVin + MS#.H.C.#10+HM#.H.C.#10+H.C.#01 v = 0, 1, 2 and 3 (v designates the bit valencies) "C = T0.T1.H. (EI.W. (MP+MR). 000+MR.010+HC.001) +MSV > HM, H.000+011 "C = T0.T1.H. (MR.010+H.C.001)+MS0.MS3.C. (800+811) +MSv < HMv, .H.000+011 Pulse C = T positive edge D1 = T0.T1 rMR = W.811+m" D2 = T0.T1 rMP + H.810 D3 = T0.T1 IAK = MR.#10 D4 = T0.T1 IKM = H.C.#01 Pulse AZ = Sa.Sb (display flip-flop 16) B1 = AZ.v.O0O (switch on 7 segment display) Figure 3 shows the oscillator 27 which is adapted to transmit pulses at a frequency of substantially 26 Hz. The output signal of oscillator 27 is designated cl. This output signal cl is transmitted to the frequency divider 28 via a NAND gate 36. The second input of the NAND gate 36 has applied to it a signal VT generated by a voltage monitoring circuit 37.

The voltage monitoring circuit 37 serves to switch off the pulses T before a battery 38 provided with a DC/DC transducer 39 (see Figure 5) takes over current supply of the most important elements of the maximum recording attachment in the event of mains voltage breakdown. vnD designates the auxiliary voltage supplied by the battery. VUG designates the voltage derived from the mains voltage. VUG is approximately 12 volts and is applied to the voltage monitoring circuit 37. As long as the voltage VUG is available, a transistor 40 (see Figure 2) is conductive, so that the signal voltage vT can be tapped at a resistor 41.

However, as soon as the voltage VUG falls below the threshold value of a zener diode 42, the transistor 40 is blocked and the signal VT disappears. On disappearance of the signal voltage VT, the LED display is cleared via a resistor 43 and the decoder 15. Since the signal VT disappears at the NAND gate 36, the frequency divider 28 is brought to a standstill, so that the pulses T are no longer supplied. The LED display is, as is known, an extremely high, current consumer.

Figure 3 also shows a circuit arrangement 45 with which a pulse mO is transmitted when the maximum recording attachment is initially supplied with voltage in order that the two shift registers 5 and 8 may be reliably cleared. As will be apparent from Figure 3, the X-flip-flop 32 is set to "0" with the pulse mO and the Y-flip-flop 33 and the C-flip-flop 30 are set. Thus, as soon as a signal mO appears, there is passage into the phase 801. When HC = "L", the stores 5 and 8 are cleared.

The following apply:

Set Y = mO with mO, with mO, 801 is set. Here, during HC = Reset X = stores MS stores MS and HM are quenched.

Set C = mO ) BS = BS1 + BS2 Referring to Figure 2, reference numeral 54 designates an arrangement for setting an MP latch.

The input 26 carries a signal MZ if a switch 46 is closed. Only then is the prerequisite for maximum measurement satisfied. The switch 46 is opened if maximum measurement is to be interrupted. The switch 46 may be a manually-actuable switch or alternatively it may be controlled by a time switch. Via an input 47, there is transmitted a metering period pulse MP which is supplied by the metering period control system 7 (Figure 1). The flip-flop 16 is actuated by the metering period pulse MP via lines 48 and 49 in such a manner that the LED display is switched off during entry into the phase 810. The MP latch comprises two variously dimensioned differentiating elements 50 and 51, so that, in the event of failure of the mains voltage and restoring of the mains voltage, the MP latch adopts the preferred output position. Due to a signal mp in the case of a measuring period pulse MP (as Figure 3 shows) the X-type flip-flop is set, i.e. there is switching into the phase 810 in which comparison between the values in the two stores 5 and 8 takes place. Thus the voltage failure is treated as a metering period pulse transmission.

Figure 6 shows a two-way switching circuit for triggering a pulse MR for automatic resetting at the end of a read-off period, for example one month. Resetting can be effected by the manually actuable change-over switch 22 or via the change-over means 23 under the control of the centralised multiservice control unit. In the case of a signal being supplied to the two-way switching circuit, a flip-flop 52 is set and transmits a pulse mr via a NAND gate 53. The circuit is such that, in the event of voltage failure, subsequent to restoration of the main voltage monthly resetting is continued and monthly resetting can also be initiated if there is no mains voltage.

As Figure 3 shows, the monthly reset pulse mr is transmitted to the control mechanism, whilst in phase 810 pulses IAK and in phase 801 pulses IKM are transmitted. Referring to Figure 7, the pulses IAK and IKM are each fed to a respective flip-flop 55 and 56 set at instant in time TO,T1,H. These pulses are also fed via a NOR-gate 57 to a counter 58 which, on the set value being reached, passes on these pulses to the cumulative counter 19 or to the reset counter 20. After disappearance of a pulse, the two flip-flops 55 and 56 are returned to their starting position via a differentiating element 60. The outputs of the two flip-flops 55 and 56 are on the one hand fed via lines BS1 and BS2 to the control system 4, and also via a further NOR-gate 61 to one input of the NOR-gate 57, so that the counter 58 only receives pulses if either a pulse IAK or a pulse IKM is present. The output pulse of the counter 58 and also the signals of the flip-flops 55 and 56 are fed via further NOR-gates 62 and 63 and via transistors 64 and 65 to the triggering lines R and KW of the cumulative counter 20 and the reset counter 19, this being done in such manner that in each case only one of the two counters is operated.

Figure 7 shows also the pulse shaper 2 and the pulse transducer 3; these are known per se and do not therefore require to be further described. Both the counter 58 and the pulse transducer 3 receive a pulse mO in the event of the initial application of the mains voltage or in the event of excessively long non-arrival of the mains voltage. Therewith, the counter 58 and the counting chain of the pulse transducer 3 are restored to 0. Due to a suitable selection of the counting stages of the counter 58, it is ensured that the electromechanical counters 19 and 20 can be operated at adequate speed.

The mode of operation of the maximum recording attachment described above will now be discussed in detail with reference to Figures 8 to 14. In all the figures, there indicated in the first four lines the pulse signal T and the signals TO, T1 and H derived therefrom. As already stated hereinabove, the stores 5 and 8 comprise 4x4 bits, so that the word cycle comprises four tetrads each comprising four bits. As is apparent from Figure 8, each set of four bits is repeated, so that the machine cycle comprises 8 bit times. With the aid of the signal H, the machine cycle is divided into two portions, i.e. into a time H and a time H.

With the aid of a NAND-gate 66 according to Figure 3, the signal T0,T1,H is formed. With this signal, not only are the two flip-flops 55 and 56 and the counter 58 operated, but also the flip-flops 31, 32 and 33.

Figure 8 further shows the initial setting of the system by means of the pulse mO. As already previously stated, on emergence of a pulse mO there is passage into the phase 801, since the Y-flip-flop 33 is set to "L" and the X-flip-flop 32 is set to "0". Finally, due to the pulse mO, the C-flip-flop is set. From the sixth line in Figure 8 it will be gathered that the output 801 of the four output decoder 34 initially carries a signal, whereas the seventh line shows that the output 800 does not initially carry a signal. In the ninth and tenth lines, there is indicated the fact that values (HM) and (MS) are present in the store outputs of the store 5 (MS) are shown, from which it may be gathered that the value in the store 5 is 7743. In similar manner, there are shown in the fifteenth to eighteenth lines the outputs of the store 8 carrying the highest maximum - these outputs being designated HMo, HUMI, HM2 and HM3. The value in the store 8 is 1623. In the nineteenth and twentieth lines, the outputs ml and m2 of the comparator 9 are shown, a signal at ml signifying that MS < HM, whereas a signal at m2 signifies that MS > HM.

As may be seen in the sixth line, the circuit has hitherto been in phase 800 in which each input pulse is added with the aid of the C-flip-flop. As soon as a metering period pulse MP appears, due to the positive edge of TO.T1.H the X-flip-flop 32 is set in accordance with the relationship "X = (MP + My).000, whereby there is transfer into the phase 801. In this phase, there is determined with the aid of the C-flip-flop whether the value in the store 5 (MS) is or is not to be transferred into the store 8. The value in the store 5 is only to be transferred if this value is larger than the value in the store 8 (HM). As will be apparent from the twenty-first line, the C-flip-flop 30 is set under the condition "C = (MS < HM) H.000+000. At H = "L", the inputs of the store 8 are connected with the outputs of the store 5 and, after transfer of the data at the end of the second cycle, zeros are written into the store 5 in the third cycle. The value in the store 5 (MS) is now transferred into the store 8. At the end of the second cycle, the C-flip-flop 30 is reset, in accordance with the condition "C = TO.T1.H.MR.B10. Therewith, there is a return once more to the phase 800.

Figures 12 to 14 serve to illustrate the accumulation of the highest maximum HM which is stored in the store 8 in the cumulative counter 19 in response to a reset pulse MR, and the recording of the number of monthly resets. The first to fifth lines of these figures are identical to the corresponding lines of Figures 8 to 11 whilst the sixth to ninth lines show the four phases. The tenth line shows the metering period pulse MP, and the eleventh line 11 shows the monthly reset pulse MR. The twelfth to fifteenth lines show the inputs of the store 5, the sixteenth and seventeenth lines show the conditions of the C-flip-flop and W-flip-flop, and the eighteenth line shows the signal IAK for triggering an accumulation pulse for the electromechanical counter 20. In the nineteenth line pulses IKM for actuating the electromechanical counter 19 are shown, and, in the twentieth line, the signal BS reproducing the condition of the flip-flops 55 and 56 in the circuit of Figure 7 is shown.

Lastly, in the twenty-first to twenty-fourth lines, the outputs of the store 8 are shown.

The process begins at phase 800. From the tenth and eleventh lines, it may be seen that a metering period pulse MP chronologically coincides with a monthly reset pulse MR.

Furthermore, it is assumed that the value in the store is again 7743 and the value in the store 8 is again 1623. In the first two cycles the same processes occur as in Figure 11. On occurrence of an MP or MR pulse, there is entry into phase 810 in which, with the aid of the C-flip-flop, the comparison is effected and, since MS is larger than HM, the C-flip-flop 30 is set, as may be seen in the sixteenth line. The value in the store 5 is transferred at time H = "L" into the store 8, the store 5 being at the end of this cycle cleared by the writing-in of zeros, whereas the store 8 now contains the value of the store 5. In contradistinction to the example of Figure 11, at the end of the second cycle the C-flip-flop is not reset, due to the condition "C = TO.T1.H.MR.B10, since the pulse MR is still present. Instead, due to the condition "Y = My.010, there is transfer to phase 811, so that in the following cycle a "1" is added to the store 5 by means of the C-flip-flop 30. As shown in the eighteenth line, at the end of the first cycle, a pulse IAK is supplied to the flip-flop 55, whereby at instant in time TO.T1H the electromechanical counter 20 is actuated and stepped-on by one step. This counter reproduces, as is known, the number of accumulations. At the end of the second cycle, the pulse IAK disappears. The flip-flop 55, however, remains set until the number set in counter 58 is reached and thus there is sufficient time for the electromechanical counter to operate. Only then is the flip-flop 55 reset by a reset pulse. This is shown in the twentieth line of Figures 13 and 14. After the pulse has disappeared, the W-flip-flop 31 is reset in accordance with the condition "W = By.811, the result of this being that due to the condition "X = W.811 there is transfer into the phase 801 in which the value in the store 5 is compared with the value in the store 8, this taking place in fact at time H. As is shown in the nineteenth line 19 of Figure 13, there occurs at HC a signal IKM producing the result that the cumulative counter 19 is stepped-on by one step. Due to the condition "X = H.C.001, at the positive edge TO,T1,H the C-flip-flop 30 is set and subsequently the W-flip-flop 31 is set. Due to the setting of the C-flip-flop 30, a further "1" is then added in the phase 811 to the value in the store 5, so that the value in the store 5 is then 2. Subsequently, once again there is transfer into the phase li)01, where the value in the store 5 is again compared with the value in the store 8. If the value in the store 5 is smaller than the value in the store 8, there is again change-ovcr into phase 811, as previously described, and a further "1" is added thereto with the aid of the C-flip-flop 30. This process is repeated until the value in the store 5 exceeds the value in the store 8 by one. This can be seen in Figure 14. After the signal BS has disappeared, there is once again transfer into phase 801. Due to the condition "C = MSv > HM,,.H000=0l1, the C-flip-flop is first of all set and at HC = "L" zeros are written into the two stores 5 and 8 and thus the two stores are cleared. Subsequently, there is transfer into the phase 800 in accordance with the condition "Y = H.C801. Under the condition "C = TO.Tl.H.C001, the C-flip-flop is reset.

The above described electronic maximum recording attachment for a pulse generating meter (i) has virtually no elements which are subject to wear, (ii) is reliable in operation, (iii) will cope with relatively long-lasting breakdown of the supply voltage, and (iv) has extremely high accuracy.

WHAT WE CLAIM IS: 1. A device for determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which device comprises: a) a first electronic store for recording the number of pulses in each metering period; b) a second electronic store for recording the maximum number of pulses recorded by the first store over the preceding metering periods; c) comparison means for comparing the value recorded in the second store at the end of each metering period; and d) transfer means (i) for transferring the value in the first store to the second store and subsequently clearing the first store, in the event of the value in the first store exceeding the value in the second store at the end of each metering period, and (ii) for simply clearing the first store, in the event of the value in the first store not exceeding the value in the second store.

2. A device according to claim 1, wherein a first counter is provided for recording the value in the second store at the end of a read-off period comprising a plurality of said metering periods.

3. A device according to claim 2, wherein the first counter is a cumulative counter which is adapted to add the value in the second store to the sum of values supplied to the counter at the end of the preceding read-off periods.

4. A device according to claim 3, wherein a second counter is provided for recording the number of values which have been supplied to the cumulative counter at the end of the read-off periods.

5. A device according to any preceding claim, wherein a control system comprising two flip-flops is provided for controlling the state of the device in accordance with four phases.

6. A device according to claim 5 when appending directly or indirectly to claim 2, wherein the control system is adapted to control the device in accordance with the following four phases: (i) a first phase in which pulses are recorded in the first store; (ii) a second phase in which the value in the first store is compared with the value in the second store and the first store is cleared and, if necessary, the value in the first store is transferred to the second store; (iii) a third phase in which pulses are added to the first store; and (iv) a fourth phase in which the value in the first store is compared with the value in the second store each time a pulse is added to the first store, in which, provided that the value in the first store does not exceed the value in the second store, a corresponding pulse is supplied to the first counter, and in which, when the value in the first store exceeds the value in the second store, the two stores are cleared and the device is returned to the first phase.

7. A device according to claim 4 or claim 5 or 6 when appended directly or indirectly to claim 4, wherein the first and second counters are electromechanical counters.

8. A device according to any preceding claim, wherein the first and second stores are shift registered adapted to cycle the stored data serially and in the form of tetrads comprising four bits each of which is supplied to a respective one of four shift register inputs disposed in parallel.

9. A device according to claim 8, wherein timing means are provided for cycling the data at a frequency which is high in relation to the frequency of the pulses recorded by the first store.

10. A device according to claim 9, wherein the timing means comprises an oscillator and at least one bistable switch connected sequentially of the oscillator.

11. A device according to claim 8, 9 or 10, wherein the first and second stores are each adapted to record a value in the form of a plurality of tetrads, each of which represents a decimal numeral.

12. A device according to any preceding claim, wherein the first and second stores are connected in parallel.

13. A device according to any preceding claim, wherein an adder is provided for adding a one to the value in the first store on receipt of a pulse.

**WARNING** end of DESC field may overlap start of CLMS **.

Claims

**WARNING** start of CLMS field may overlap end of DESC **. "C = MSv > HM,,.H000=0l1, the C-flip-flop is first of all set and at HC = "L" zeros are written into the two stores 5 and 8 and thus the two stores are cleared. Subsequently, there is transfer into the phase 800 in accordance with the condition "Y = H.C801. Under the condition "C = TO.Tl.H.C001, the C-flip-flop is reset. The above described electronic maximum recording attachment for a pulse generating meter (i) has virtually no elements which are subject to wear, (ii) is reliable in operation, (iii) will cope with relatively long-lasting breakdown of the supply voltage, and (iv) has extremely high accuracy. WHAT WE CLAIM IS:

1. A device for determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which device comprises: a) a first electronic store for recording the number of pulses in each metering period; b) a second electronic store for recording the maximum number of pulses recorded by the first store over the preceding metering periods; c) comparison means for comparing the value recorded in the second store at the end of each metering period; and d) transfer means (i) for transferring the value in the first store to the second store and subsequently clearing the first store, in the event of the value in the first store exceeding the value in the second store at the end of each metering period, and (ii) for simply clearing the first store, in the event of the value in the first store not exceeding the value in the second store.

14. A device according to claim 13, wherein first inputs of the adder are connected to

outputs of the first store and second inputs of the adder are held at earth potentional, and a first flip-flop is provided for adding a one to the value in the first store on receipt of a pulse by supplying a pulse to a carry input of the adder.

15. A device according to claim 14, wherein a second flip-flop is associated with the first flip-flop so that, in the event of a relatively long pulse being supplied to the first flip-flop, only a single one is added to the value in the first store.

16. A device according to claim 14 or 15 when appended directly or indirectly to claim 11, the arrangement being such that, in the event of a one being added to a value in the first store having a nine in the lowest tetrad, zeros are written into that tetrad and a one is added to the next highest tetrad.

17. A device according to any preceding claim, wherein the comparison means is a comparator.

18. A device according to any preceding claim, wherein the transfer means comprises two multiplexers connected to the two stores and arranged to be triggered by the comparison means.

19. A device according to any preceding claim, wherein a display is provided for indicating the values in the first and second stores by way of a multiplexer controllable by a switch.

20. A device according to claim 4 or any one of claims 5 to 19 when appended directly or indirectly to claim 4, wherein the first and second counters are roller counters adapted to be triggered (one after the other) by way of a time-period monitoring system.

21. A device according to claim 20, wherein the first and second counters are adapted to be operated in acknowledgement signal (Hand Shake) mode.

22. A device according to any preceding claim, wherein a battery is provided to supply current to the stores and timing means (if provided) via a unidirectional voltage transducer in the event of main voltage failure.

23. A device according to any preceding claim, wherein means are provided for ensuring that, in the event of mains voltage failure, the values in the first and second stores are compared by the comparison means as at the end of a metering period.

24. A device according to claim 22 when appended directly or indirectly to claim 10, wherein the bistable switch is adapted to cut off the timing pulses from the oscillator before the battery takes over current supply.

25. A device according to any preceding claim, wherein further means are provided to ensure that a value representing the number of pulses supplied to the first store prior to the end of a metering period is held in the first store during a mains voltage interruption and supply of pulses to the first store is automatically continued on resumption of the mains voltage.

26. A device according to claim 25, wherein said further means is a latch comprising two differentiating elements, the latch being adapted to be cleared on application of the mains voltage.

27. A device according to claim 19 or any one of claims 20 to 26 when appended directly or indirectly to claim 19, wherein the display is an LED display.

28. A device according to claim 27 when appended directly or indirectly to claim 2, the arrangement being such that the display is switched off at the end of each read-off period.

29. A device according to claim 28, the arrangement being such that the display is cleared at the end of each metering period.

30. A device according to claim 2 of or any one of claims 3 to 29 when appended directly or indirectly to claim 2, the arrangement being such that the value in the first store is compared with the value in the second store at the end of each read-off period.

31. A device according to claim 2 or any one of claims 3 to 30 when appended directly or indirectly to claim 2, the arrangement being such that, should the end of the lead-off period take place during means voltage failure, recording of the value in the second store takes place on resumption of the mains voltage.

32. A device according to any preceding claim, wherein means are provided for normalising pulses prior to there being supplied to the first store.

33. A device for determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which device is substantially as hereinbefore described with reference to, and/or as illustrated in, the accompanying drawings.

34. In combination, a pulse generating electricity meter and a device according to any preceding claim.

35. A method of determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which method comprises: (1) recording the number of pulses in each metering period in a first store; (2), at the end of each metering period, comparing the value recorded in the first store with a value recorded in a second store representing the maximum number of pulses recorded by the first store in the preceding metering periods; (3), if the value in the first store exceeds the value in the second store, transferring the value in the first store to the second store and clearing the first store; and (4), if the value recorded in the first store does not exceed the value recorded in the second store, simply clearing the first store.

36. A method of determining the maximum number of pulses recorded in a predetermined metering period over a plurality of such metering periods, which method is substantially as hereinbefore described with reference to the accompanying drawings.