NO140695B - DEVICE FOR DISTRIBUTION OF ELECTRIC ENERGY BETWEEN SEVERAL CONSUMER UNITS - Google Patents

Description

The present invention relates to control of the power supply

to groups of electrical devices, or groups of power-consuming elements in a single device, in the cases mentioned consumer units are operated so that their average power consumption is much lower than the maximum consumption.

In a conventional electrical installation where a number of appliances receive power from a common source, and all appliances can be switched on at the same time, it will usually be necessary to dimension the relevant supply lines and power source for maximum power load, which corresponds to the situation that all consumption units are in operation at the same time. In the case of large electrical installations, the consumer tariffs set by the power plants will often, and sometimes to a significant extent, depend on the installation's maximum power output. There is thus a clear desire to reduce this maximum power draw wherever possible.

It is a characteristic of many types of electrical equipment, and particularly such equipment that includes electrical heating elements, that the maximum power output is much greater than the normal, average power consumption. Such equipment is usually thermostatically controlled and under normal operating conditions the switch-off periods for the thermo-state will be much longer than the switch-on periods. The supply lines in an installation that includes several units of such equipment must, however, be dimensioned for the case that all thermostats are switched on at the same time. Similar considerations apply to a single device that contains several independently controlled heating elements, such as e.g. an electric stove. It has been proposed to arrange equipment for regulating the maximum power supply, by arranging the whole so that an increasing number of load units are switched off by a stabilizer as the units' combined power demand increases, thereby limiting the total consumption to a predetermined level, as the order in which the different load units are switched off is determined in accordance with a predetermined priority plan. It has also been proposed that load units with equal priority in such an arrangement are separately connected to the stabilizer through a commutation device, so that their position in the priority list is periodically exchanged, and in the long run power will be supplied to such load units with equal priority in accordance with the mean priority level for the connection between the switching device and the stabilizer.

A disadvantage of the hitherto proposed facilities of this nature is that the various load units are completely disconnected from the power source when the load equalization takes place, and likewise the necessary switching, although they keep the power consumption below a maximum level, will produce transient load variations, which will not only be able to cause radio interference , but also cause unwanted fluctuations in the mains voltage, especially if the loads that are switched in and out are large.

In power regulation devices for alternating current, it is well known to use thyristors and other similar semiconductor elements, e.g. triac elements. Such regulatory systems are of two main types. The first, which is a phase regulation type, is based on variation of the time in each, or every second, half period of the alternating current when the control direction is switched on. This type of regulator entails serious radio interference and power factor problems and is not favorable for installations with significant power consumption. The alternative interval control system, in which the control function is exercised by varying the rate and length of the control device's switch-on intervals over a whole number of half periods, overcomes the problems of the phase control method, but causes a load variation for the power source. If the duration of the mentioned intervals is relatively short, then the frequency of these fluctuations will be particularly disturbing.

On this background, the purpose of the present invention is to provide an apparatus for limiting the maximum power output for a plant or an apparatus which comprises a number of individually controllable electrical consumer units or groups of such units, and which does not entail the disadvantages mentioned above.

The invention thus relates to an apparatus for distributing electrical energy between several consumer units with each associated connecting device, e.g. a thermostat or manually operated switch device, the position of which indicates whether there is a need for the supply of energy to the relevant consumer unit, said consumer unit series with each gate-controlled switching element, e.g. a thyristor, is connected to a common alternating current source and the apparatus comprises a pulse generator arranged to produce a continuous series of control pulses which are timed to substantially coincide with each respective zero point crossing of current emitted from the alternating current source as well as pulse control means arranged and arranged for depending on the pulses in the continuous pulse series to emit output pulses to the one input of a number of gate devices, which are connected on their output side to the gate electrode for each of the aforementioned gate-controlled switching elements.

Equipment of this kind is known in principle from US patent document no.

3,521,077 . and 3. 529 .173 . on this background of known technology, the apparatus according to the invention has as a distinctive feature that the second input on each of said port devices is connected to the pulse generator to receive the continuous series of control pulses, and said coupling means are connected to the pulse control means, which are arranged and arranged to sensing the position of each individual switching device as well as only emitting output pulses to the port devices corresponding to consumer units whose switching devices indicate a need for energy supply, in such a way that these selected port devices receive

said output pulses in essentially uninterrupted cyclic order and each time for a period of time corresponding to a whole number of half-periods of the current from the alternating current source.

The term "consumer unit" is intended to include any electrical appliance or individually controllable element of such an appliance, and also any combination of appliances and/or elements with a significant power consumption in the connected state.

The switching means for the consumer units can be manually or automatically controlled switches, thermostats or any type of converter capable of emitting a control signal, and can either be utilized to interrupt the power supply to its assigned unit to thus produce a condition that can be sensed by the pulse control organs or preferably for direct influence on these organs.

The pulse control devices are preferably time-controlled through their functional sequence of clock pulses produced by dividing the frequency of the alternating current source.

With energy distribution according to the invention, it is never taken out

more power than what corresponds to the power consumption of a single consumer unit. None of the consumer units are prioritized in any way, and the units that need power must at all times share the available energy from the grid within a certain period of time.

Such a favorable working function produced by the simple means indicated above has not been achieved with previously known devices of the present kind.

The device of the invention will now be described in more detail with the help of design examples and with reference to the attached drawings, on which: Fig. 1 is a simplified block diagram of parts of an electrical home installation, which includes a number of electric magazine heaters.

Fig. 2 is a simplified block diagram showing a hot plate

system for an electric stove.

Fig. 3 is a simplified block diagram for an industrial plant comprising four equal heat loads, and Figs. 4, 5 and 6 are connection diagrams for different device designs according to the invention. Fig. 1, 2 and 3 show typical applications for the device of the invention, which in these figures is indicated by reference number 5 (fig. 1), 25 (fig. 2) or 31 and 33 (fig. 3) and has certain specified working functions. How these work functions are achieved will be described in more detail with reference to fig. 4, 5 and 6.

In fig. 1 shows an electrical home installation comprising four electric magazine ovens 1-4, which e.g. have a power consumption of 3 kW each. These heaters are connected by a fuse and distribution box 5, which is connected via a cable 6 to instruments and a time switch 7. These are in turn connected via a further cable with main fuses 8, to which the rest of the installation 9 is also connected. The fuses 8 are connected via a cable 10 to the network 11. In a conventional installation, the time switch 7 will switch on all four heaters 1-4 at the same time, and as these usually initially have a sufficiently low temperature for the built-in thermostats in the heaters to switch all ovens on, it will be necessary for the cable 6 to be dimensioned for power supply to a 12 kw load, at the same time that the contacts in the time switch 7 must be dimensioned accordingly, while the fuses 8 and the cable 10 must be dimensioned as much above 12 kv/ as corresponds to the additional load 9 in the installation. If many private homes in the vicinity have the same form of heating, it will also be necessary to increase the dimensioning of the supply network 11, and the sudden increase in load, which occurs at certain times of the day when all or most heaters are switched on at the same time, will also cause problems when maintaining the mains voltage within established limits.

Assuming normal setting of the heater's built-in thermostats, these heaters individually will typically not require more than a fraction of the total on-time allotted by timer 7 to reach a temperature that causes the furnace thermostats to disengage their associated heating element, and then require only short periods of power supply to keep the elements evenly at the desired temperatures set in the thermostats, during the rest of the switch-on period. A curve for the power consumption as a function of time would show an initial period of 12 kw power consumption, which would gradually decrease to 0 after a few hours, apart from occasional periods of power consumption of usually no more than 3 kw, to maintain a constant temperature.

It would obviously be desirable to equalize the aforementioned power consumption

by spreading this to a greater extent over a longer period of time while reducing the maximum effect. This can be achieved by placing the present device according to the invention in the distribution box 5, as this device in operation will provide power to each heater in turn as long as its assigned thermostat keeps it connected in the supply circuit. By arranging the whole thing so that the appliance's programming is controlled in accordance with which heaters are switched off, either by their own thermostats or in some other way, it can be ensured that an increased proportion of the supplied energy is distributed to the remaining heaters when one of the stoves is switched off , so that the load on the supply network remains constant.

Fig. 2 shows utilization of the present invention in connection with an electric stove with heating plates 21, 22, 23 and 24, two of these plates e.g. is 2 kw and the other two 1.5 kw. Each heating plate will normally be equipped with an independent control device, such as e.g. a so-called "Simmerstat", which will be effective for in-

and switching off the hot plate in question, the ratio between the switch-on and switch-off periods being regulated by setting the control device to give different desired hot plate temperatures.

It will be clear that with such a system it will be possible for all heating plates to draw current at the same time, even if their regulation setting is set quite low, so that the instantaneous power consumption can vary quickly between 0 and 7 kw in an unsatisfactory manner . If, however, the mutually independent control devices are connected together with apparatus 25 according to the invention, and which is arranged to, in operation, emit power to each of the connected heating plates in turn in a continuous connection interval, while the individual control devices 26 for the different heating plates affects the pulse control devices in the device 25 with the aim of regulating the total switch-on time for each heating plate within the said switch-on sequence, the above-mentioned problem can be solved. The total power consumption for the cooker cannot exceed the power requirement for the maximally dimensioned hotplate, which will to some extent reduce the operating flexibility of the cooker, but in many cases this disadvantage will be offset by the advantages of significantly reduced maximum power consumption and reduced power variations, as well as more appropriately sized supply lines 27. Although the rated effects for the heating plates differ from each other, the difference is nevertheless so small that strong fluctuations in the power supply will be avoided.

Fig. 3 shows the invention utilized in an industrial plant with e.g. four melting containers 30, each with a heating element of 20 kw. In a conventional installation, the supply lines must be dimensioned on the basis of the assumption that the heating elements for all four containers can be switched on at the same time, which will correspond to a maximum power consumption of 80 kw, although the average power consumption will only correspond to a small fraction of this value.

Each of the containers is connected to a power source 35 through a regulating device 33 according to the invention, and if pulse control devices are affected by signals that are fed back along, feedback loops 32 from the thermostats 31 in the respective containers 30. The regulating device 33 switches in order and out those containers whose assigned thermostat has closed contacts, indicating that the container is below a certain desired temperature. Containers whose thermostats have open contacts are excluded from the connection sequence.

In fig. 4 shows an embodiment of the device 33 (see fig. 3)

according to the invention. Load units Li, L2, L3 and L4 (e.g. containers 30 in Fig. 3) are connected between the lines ACL

and ACN from an alternating current source through triac elements Tl, T2,

T3 and T4. As shown, each load unit is assigned a manually adjustable switch and/or thermostatically regulated contacts, represented by switches Sl, S2, S3 and S4. These switches close or break the circuits independently of the switch functions performed in the triac elements and the loads, although the switch arrangement shown in fig. 3 can also be used, as will be explained below. The switches are connected via the input terminals IA, 2A, 3A and 4A to pulse control devices, which comprise data selectors Hl, H2, H3 and H4 assigned to each of the triac elements, respectively Tl, T2, T3 and T4. The outputs IB, 2B, 3B and 4B from data selectors are used for controlling the ports Ml, M2, M3 and M4, each of which receives a square wave with mains frequency. This square wave is produced by a pulse generator, which is generally denoted by reference number Y and comprises a bistable multivibrator formed by cross-connected gates A, which is driven by clipped pulses derived from the mains voltage by means of a transformer TX, resistors R5 and R6 and zener diodes Zl and Z2.

The output signals from gates Ml - M4 are applied to the gate electrodes

for the triac elements Tl - T4 through buffer amplifiers and/or pulse transformers XI - X4, which set the amplitude, output impedance and waveform of the output signals from the gates, to produce alternately positive and negative pulses at the beginning of each half-period of the alternating current from the mains, these pulses are used for ignition of the triac elements. Introduction of transformers or circuit isolators Xl - X4 is of significant importance, if the rest of the apparatus is to be electrically isolated from the circuits comprising the triac elements Tl - T4, as is often desirable for safety reasons.

In an alternative arrangement, the circuits XI - X4 are equipped with zero voltage switches of the type well known in integrated circuit form, a typical example of which is the CA30 59 circuit marketed by RCA. As these integrated circuits have built-in connections for producing suitably timed and shaped pulses on the basis of supplied mains voltage, the connection to the pulse generator can be omitted, and likewise the ports M, as the output lines IB, 2B, 3B and 4B are connected directly to a suitable TTl switch input on said integrated circuit which forms a zero voltage switch. To maintain correct polarities, wires IB, 2B, 3B and 4B should be connected to inverting outputs on selectors H1 - H4.

It will thus be understood that the output signals from the data selectors are used to control pulses from the generator Y to their respective assigned triac elements, thereby closing the circuits of the associated loads Li - L4.

The data selectors are programmed using the connection circuits

which connect the selectors to the switches Sl - S4, which when closed form connections between the ground rail of a direct current source (not shown) for the control circuits and one end of the respective associated resistors RI - R4, which at their other ends are connected to the positive of the direct current source power rail. A logic 0 signal is then applied to the selector inputs which are connected to any such short-circuited resistor. In the exemplary embodiment discussed here, all of these inputs are negative logic inputs. A control input s on each selector Hl - H4 is connected to its own resistor, respectively RI, R2, R3 and R4, and thus ensures that a data selector is only activated when its thus assigned switch is closed. The switches Sl - S4 are also connected in different combinations to the programming inputs a, b and c on each selector.

In accordance with which input or combination of inputs a, b and c is brought to logic 0 value on a selector, the logic state of one of the eight inputs shown in fig. 4 on the upper side of each selector Hl - H4, be connected to the assigned port Ml - M4. Of these inputs, one on each selector is permanently impressed with a fixed potential, while the rest of the inputs are connected, in a different arrangement for each selector, to different of the shown nine outputs ti - t9 from a pulse-generating timing control circuit, which includes counters which are generally denoted by X and Z and transmits suitable series of control pulses for selection of the selectors Hl - H4 and thus for selective supply to the gates Ml - M4.

The pulse generating timing circuit comprises firstly a counter X for dividing by 3 and which comprises two JK bistable multivibrators D and a NOG gate N, together with NOR gates E which emit three sets of output pulses with a ratio of 1:2 pulse length and space, and which are thus mutually 120° out of phase with each other, to the lines t3, t4 and t5. This circuit further comprises a counter Z for division by 4, and which comprises a first JK multivibrator Fl which emits output pulses with a ratio of 1:1 between pulse length and space and in mutually opposite phase to the lines ti and t2. One of these pulse trains is also emitted to a JK multivibrator F2. This multivibrator F2 emits in cooperation with gates G four pulse trains with a ratio of 1:3 between pulse duration and space to the lines t6-t9, so that each successive pulse train is 90° out of phase with the previous train. The pulse-generating timing control circuit is clocked by pulses derived from the output of the pulse generator Y after passage through a clock generator W, which in the present example is formed by two cascaded counters B and C for division by 16. However, the division ratio will be subject to appropriate selection depending on the desired cycle length for the pulse generator and timing circuit Z, and which in turn depends on the nature of the loads LI - L4. When these loads are made up of heating elements with a large thermal inertia, a long duty cycle is preferable, while loads with little inertia require a short cycle length, if uniform working function is to be achieved. It is not strictly necessary that the clock pulses be derived by dividing the mains frequency, and any suitable square wave generator will be able to serve as a clock pulse generator. However, it would be desirable for the clock pulses to be timed to begin and end in a specific relationship to the mains phase, thereby helping to avoid uncontrolled reactions in the logic control circuits.

The wire connections ti - t9 to the data inputs on the selectors

Hl - H4 are arranged in accordance with the programming of the inputs a, b and c on the selectors by means of the switches Sl - S4, that when only one switch is closed the assigned port of the ports Ml -

M4 be fed with pulses continuously. When two switches are closed, the input signals from lines ti and t2 will be applied to the assigned two ports of ports Ml - M4 in such a way that these ports are pulsed alternatively, each for half the time. When three of the switches are closed, the input signals from the lines t3, t4 and t5 are likewise applied to the assigned three of the ports Ml - M4, so that these ports are pulsed in sequence, each for 1/3 of the time. When all four switches are closed, the input signals from the line t6 - t9 will finally be applied to all four ports Ml - M4, so that these ports are pulsed in sequence, each for 1/4 of the time cycle. The present selection of loads LI - L4 assigned to ports M1 - M4 will thus each be energized in turn at appropriate time intervals, corresponding to a whole number of alternating current half periods. If the state of the switches S1 - S4 is changed, the selectors H1 - H4 will immediately be reprogrammed, and this new program will effectively direct the next successive pulse from pulse generators Y to the relevant triac element in accordance with the new program. Each of these pulses, which indicate the beginning of each half-period for the alternating current from the mains, will thus always be directed to one or other of the loads Li - L4.

The design of the circuits used has not been described in detail. However, the circuits are shown on the drawings as TTL integrated circuits of the standard 74 series, the designation of the integrated circuits used in most cases being indicated on the drawings together with the correct number of contact pins for the dual line versions of these circuits. Ports A, N and M are realized using type 7400

NOG Quadruple ports and ports E and G using 740 2 NELLER Quadruple ports. It will be obvious to those skilled in the art how the circuits shown can be expanded to be able to serve a greater number of loads, or how alternative logic circuits can be used or adapted for use in integrated circuits specially designed for the purpose.

It may sometimes be desirable or, when already existing installations are to be modified, more convenient to place the switches Sl - S4 in series with the loads, as shown in fig. 3. In this case, however, equipment must be provided for sensing each triac circuit to check whether the circuit is broken, as well as to apply appropriate logic signals to input terminals IA, 2A, 3A and 4A. Since, however, arcing between the contacts due to mechanical breaking usually results in a disconnection of a load circuit not becoming effective before the relevant half-period of the alternating current has ended, disconnection of a load during power supply will be detected too late for the logic control circuits can be reprogrammed in time to direct the next successive ignition pulse to another triac element. The previously described arrangement with the switches Sl - S4 in separate circuits which directly control the logic circuits, will thus be preferable.

Fig. 5 and 6 show two alternative designs of the device 33 in Fig. 3, however, the circuit which is controlled by the output lines IB - 4B is omitted, as it is of the same nature and can be subject to the same modifications as the corresponding circuit in fig. 4.

In fig. 5, pulses from a clock generator W of the same type as the generator W in fig. 4 for clock control of a two-stage Johnson counter, which comprises two appropriately connected D-type multivibrators N with set and reset inputs, respectively S and R. These inputs are used for setting the counter to any desired counting step in its count cycle, depending on the detection of a particular step in the count cycle by means of a system of OR gates P and Q. This system is programmable both with respect to the detected count step (by means of gates P) and the count step set by by means of inputs S and R in response to the detection of the former step (by means of gates Q). This programming is consistent with the logic level applied to inputs IA, 2A, 3A and 4A depending on the state of switches S1, S2, S3 and S4 (see above). The output signals from the outputs Q and Q for the counting steps are decoded using additional NOR gates T and applied to the outputs IB, 2B, 3B

and 4B, which controls the control circuits' firing pulses to the triac elements

T1, T2, T3 and T4, as in the previously described embodiment. The decoding gates T are pulse fed from the input lines through inverters I, thereby avoiding any random error pulses that may appear on an output line when its associated switch Sl -

S4 is open.

As will be clear from a study of the circuit diagram, the logic control circuits are designed so that the counter can only remain in counting stages that produce logic one signals over output lines corresponding to input lines with a logic O input signal (that is, inputs assigned switches Sl - S4 which are closed). When the counter is clocked, it will produce trains of ignition pulses which are successively transmitted to the triac elements Tl - T4 which are assigned to switches Sl - 4 which are closed.

Fig. 6 shows an alternative arrangement, whereby the clock pulse generator W is used for clock control of a ring counter with four D-type multivibrators Dl, D2, D3 and D4, one for each load to be controlled, and which are interconnected to form a shift register. The output lines IB, 2B, 3B and 4B are connected to the Q outputs of the bistable multivibrators Dl - D4, and the register is clocked by pulses from the clock pulse generator W, which is of the same type as the previously described generators W. Thus, when a single bi-signal is introduced into the register and is cycled by the clock pulses, each output line will in turn be driven high, to drive a train of ignition pulses to its associated triac element Tl - T4. The input lines IA, IB, 1C and ID are, however, used for controlling a gate system which programs the ring counter to skip the multivibrators D1 - D4 which are assigned to input lines IA - 4A with open switches Sl - S4. The multivibrators are thus connected into the register by connecting the Q output of each multivibrator with the D input of the following multivibrator through a NOG gate Jl - J4, while the programming inputs are connected through inverters Kl - K4. When these inputs assume a low level, they provide a pulse supply to the NOG gates El - E4, which thereby transfer the D input level of the associated multivibrator Dl - D4 to the gate Jl - J4 for the on-

following multivibrator. The assigned inverted level on

input IA - 4A is also applied over a NELLER gate NI - N4

to the reset input R of a multivibrator to be skipped, so that the assigned output IB - 4B remains at the logic 0 level.

The whole thing must be arranged so that the register is started upon introduction

of a single binary digit, while preventing the register from operating in the wrong mode. For this purpose, the Q output signals from the multivibrators Dl - 4 are supplied to a decoder which is generally denoted by S, and which primarily has the task of detecting a state in the ring counter where all Q outputs are at a low level, as it responds to this condition by supplying a digit to the multivibrator D1 through a further input for the gate J1. However, the decoder also detects a state in which more than one of the Q outputs is at a high level, as it reacts to this state by resetting all multivibrators. The decoder comprises a binary to decimal decoder unit DC, where outputs corresponding to the digits 0, 1, 2, 4 and 8 are connected to a NOG port and from there through an inverter to NELLER ports NI - N4. A 0 output is also connected to port Jl. An O output signal from the decoder corresponds to a count state in which all Q outputs are low, and the absence of an output signal corresponding to 0, 1, 2, 4 or 8 corresponds to a counter state in which more than one of the Q outputs is at a high level.

It will be understood that this embodiment can easily be extended to include apparatus for regulating any desired number of loads.

It will also be understood that the embodiments described above only constitute exemplary embodiments. Logical systems can thus be used which make it possible for groups of loads to be energized at the same time, provided that the relevant groups in all cases have approximately the same total power consumption. Likewise, the control device can be designed so that the energization periods for different loads can be varied in relation to each other, or so that selected loads are energized more than once during each individual switching cycle. One way in which this can be easily achieved is to use any of the above-described designs with parallel connection of inputs and outputs.

times for the operation of a single load.

Claims (4)

1. Apparatus for distribution of electrical energy between several consumer units (LI - L4, fig. 4) with each associated connecting device (S1 - S4), e.g. a thermostat or manually operated switch device, the position of which indicates whether the there is a need for the supply of energy to the relevant consumer unit, the aforementioned consumer units in series with each gate-controlled switching element (Tl - T4), e.g. a thyristor, is connected to a common alternating current source and the device comprises a pulse generator (Y) arranged to produce a continuous series of control pulses which are timed to essentially coincide with each respective zero-point crossing of current emitted from the alternating current source as well as pulse control means (Hl - H4, W, X, Z) arranged and arranged for, depending on the pulses in the continuous pulse train, to emit output pulses to one input of a number of gate devices (Ml - M4), which are connected on their output side to the gate electrode for each of the aforementioned gate-controlled switching elements, characterized in that the second input on each of said gate devices (Ml - M4) is connected to the pulse generator (Y) to receive the continuous series of control pulses, and said coupling means (S1 - S4) are connected to the pulse control means (Hl - H4, W, X, Z), which is arranged and arranged to sense the position of each individual coupling device and only to emit output pulses to the gate devices which responding consumer units (Li - L4) whose switching means (Sl - S4) indicate a need for energy supply, in such a way that these selected gate devices receive said output pulses in essentially uninterrupted cyclic order and each time for a period of time corresponding to a whole number of half-periods of the current from the AC source. 2. Apparatus as stated in claim 1, characterized in that the pulse control devices comprise data selectors (Hl - H4) which are assigned to each load (LI - L4) and have inputs connected to pulse counters (X, Z), which in turn are connected to a frequency divider (W). 3. Apparatus as stated in claim 1, characterized in that said pulse control means comprise a two-stage Johnson counter (N, fig. 5) of a type known in and of itself and with inputs which are connected via ports (P) to each respective switching means (IA - 4A). 4. Apparatus as specified in claim 1, characterized in that said pulse control means comprise a ring counter of a known type and with a number of steps (Dl - D4), the input signal to each step being derived from the output side of gates (El - E4) whose inputs are connected to the preceding counter step and an assigned switching means (IA - 4A, Fig. 6).