EP1579562A1

EP1579562A1 - Power controller

Info

Publication number: EP1579562A1
Application number: EP03767849A
Authority: EP
Inventors: Martti Sairanen; Mikko KAIJÄRVI; Mikael Porthin; Raimo Riuttala
Original assignee: Schneider Electric Industries SAS
Current assignee: Schneider Electric Industries SAS
Priority date: 2002-12-31
Filing date: 2003-12-30
Publication date: 2005-09-28
Also published as: WO2004059825A1; AU2003292284A1; FI20022300A; FI20022300A0; FI114588B; AU2003292284B2; CN1732616A

Abstract

The invention relates to a method for controlling the alternatingcurrent power supplied to a load by controlling the phase angle of the alternating voltage and to the related power controller. The switching unit (2) of the power controller (1) consists of twoswitching elements (kl, k2) connected in succession mutually and relative to the load (L) and of reverse-current diodes (dl, d2) disposed in parallel with these. The switching elements are brought in turn into electrically conductive and non-conductive statesduring one half cycle of the alternating voltage cycle (VAC) to be supplied. In accordance with the invention, during the first half cycle (PA) after a delay determined on the zero position of the first load current (IL) the first switching element (kl) is brought into conductive state, while the second switching element (k2) is brought into non-conductive state, the load current (IL) being disposed to flow across the first switching element (kl) andthe second forward-current diode (d2) until the load current (IL ) is reversed at the second zero position of the current, the direction of the load current (IL) being reversed relative to the second diode (d2), thus preventing current flow to the load. After a delay determined on the second zero position (N2) of the load current (IL) during the second half cycle (PB), the second switching element (k2) is brought into conductive state while the first switching element (kl) is brought into nonconductive state, the load current (IL) being disposed to flow across the second switching element (k2) and the forward-current first diode (dl) until the direction of the load current (IL) is reversed again at the following zero position of the current, the direction of the load current (IL) being reversed relative to the first diode (dl), thus preventing current flow to the load until the operation steps described above are repeated during the following half cycle, which corresponds to the first half cycle.

Description

Power controller

The invention relates to a method as defined in the preamble of claim 1 for controlling the alternating current power supplied to a load by controlling the phase angle of the alternating voltage.

The invention also relates to a power controller as defined in the preamble of claim 3 for controlling the alternating current power supplied to a load by controlling the phase angle of the alternating voltage.

Usually alternating current power connected to a load is controlled by switching off the electric current across the load, so that the electric power supplied to the load is determined by the ratio between load current on and off periods and by the amplitude of the alternating voltage varying in the course of time. In most alternating current applications, the switching cycle of the load current occurs once during one half cycle of the alternating voltage. Such a device is often called a phase angle controller. Phase angle controllers can be used for controlling lighting fixtures, motors and heating elements, for instance.

All real loads are reactive - they comprise a resistive, a capacitive and an inductive part, respectively. If the capacitive and the inductive parts are very small compared to the resistive part, the load is called a resistive load, such as an incandescent lamp, for instance. A load is called capacitive if it comprises substantial capacitance in addition to the resistive part (e.g. an electronic power converter for low voltage halogen burners), and accordingly, loads comprising substantial inductance are called inductive (such as a motor or a power converter, for instance).

The current across capacitance C is generally represented by the equation:

i(t) = C du/dt, in which (1)

u is the voltage across capacitance C and t is time. A study of equation 1 shows that a capacitive load is advantageously controlled with a phase angle controller, which turns on the load current during zero position of line voltage. Thus the charging current i of capacitance C depending on the voltage conversion rate du/dt will remain low. Such a controlling manner is called trailing-edge control below. A device using trailing-edge control comprising FET, i.e. a channel transistor as the switch component, has been disclosed by WO patent application 9322826.

The voltage u across inductance L is generally represented by the equation:

u(t) = -L-di/dt, in which (2)

i is the current across inductance L and t is the time. A study of equation 2 shows that current conversions di/dt at a high conversion rate generate a high voltage u across inductance L. To avoid this voltage peak, inductive load is advantageously controlled by switching off the load current at zero position of the current across the load. Such control is called leading-edge control below, and a commonly known controller utilising a bi-directional thyristor, i.e. triak, for implementing this control has been disclosed i.a. by US patent specification 6175195.

The prior art techniques described above involve shortcomings and problems:

- the triak triggered in conductive state cannot be brought into non-conductive state before the current across the triak has dropped to zero. For short-circuit and overcurrent protection, the triak controller will thus require a fuse. Practical electric installation work has showed that replacement of an internal fuse in a device is a frequent reason for calling an electrician to the installation site. This means that a fuse in the short-circuit network causes the end user of the device extra costs that often exceed the purchase price of the device.

- triak has relatively high switching speed, and cannot be actively controlled. Due to electromagnetic compatibility (EMC) requirements, the known triak controller comprises an inductor, which limits the current increase rate to an acceptable level. This inductor causes power losses in the power circuit, resulting in higher temperature of the device. The inductor also incurs expenses, it has a large physical size requiring abundant space in a typical device, and it has been noted to disturb the end user because of its specific humming noise.

- the controller based on a FET semi-conductor mentioned in WO patent specification 9322826 does not necessarily comprise a fuse or an inducer, yet it is a controller implementing trailing-edge control. Consequently, it is not suitable as a power controller of inductive load in terms of equation 2. The controller described in the reference can be converted to leading-edge control, however, this requires exact measurements of the current across the controller, resulting in more complex circuit structure and hence in higher production costs of the device.

The purpose of the invention is to accomplish such a method and a device for controlling the alternating current power supplied to a load that resolve the problems mentioned above and that allow control of the alternating current power of resistive load and inductive load by using two power transistor switches connected in succession, especially a MOSFET switch.

The method of the invention for controlling the alternating current power supplied to a load by controlling the phase angle of alternating voltage is characterised by the features defined in claim 1. The power controller of the invention for implementing the method, again, is characterised by the features defined in claim 3. The dependent claims describe preferred embodiments of the invention.

The invention achieves several advantages:

- short-circuit protection can be provided in the device of the invention without a fuse. This allows for simpler device design and reduces the operating costs of the end user of the device.

- the device of the invention does not necessarily comprise an inducer. This allows a decrease of the physical size of the device, or accordingly, the provision of a device having higher power handling capacity in the same size range.

- compared to a power controller based on a triak switch, especially a dimmer, the device of the invention does not generate noise in the environment that would be audible to the human ear.

- owing to automatic commutation, the invention allows for the use of a transistor switch in a leading-edge controlled power controller, so that the load current is switched off precisely as the current has reached the zero level, without requiring exact measurements of the load current.

The invention is explained in further detail below with reference to the accompanying drawings, in which Figure 1 is a principle view of a power controller, which is serially connected with the load and is controlled in line with the method of the invention;

Figure 2 is a block scheme of the power controller of the invention;

Figure 3 shows a preferred embodiment example of the power controller of the invention, especially of its control unit;

Figure 4 shows curve shapes of voltage and current as a function of time at different points of the power controller of figure 1 with resistive load; and

Figure 5 shows curve shapes of voltage and current as a function of time at different points of the power controller of figure 1 with inductive load.

Figure 1 is a schematic view of a power controller 1, by means of which the method of the invention for controlling alternating current power supplied to a load can be accomplished. The power controller 1 comprises a switching unit 2 and the control unit 3 of this. The power controller 1 is serially connected with the load L. The alternating current source AC, such as an alternating current network, is connected across the serially connected power controller 1 and the load L.

The switching unit 2 consists of two switching elements connected in succession mutually and simultaneously relative to the load L, i.e. of a first and a second switching element kl, k2, and of reverse-current diodes arranged in parallel with these, i.e. a first and a second diode dl, d2. The control unit 3 is used for controlling the switching elements kl, k2 of the switching unit 2 in turn into electrically conductive state, i.e. off state, and non-conductive state, i.e. on state, respectively, during one half cycle of the alternating voltage serving as supply voltage.

Figure 4 illustrates curve shapes of voltage and current in connection with the power controller 1 of figure 1, with load L being resistive. The load L is then e.g. an incandescent lamp, whose luminosity is controlled by controlling the alternating current power supplied to it. Figure 4a shows the curve shape of the alternating voltage V_Ac supplied from alternating current source AC during several successive half cycles PA, PB, PA,... as a function of time t. Figures 4b and 4c show the states of switching elements kl and k2, respectively, with 0 denoting the on state of the switch. and 1 the off state of the switch. The curve shape of the voltage V_L across the load L is shown in figure 4d, and the curve shape of load current I_L in figure 4e.

In the method of the invention, during the first half cycle PA of the alternating voltage V_AC ^af^ter the delay t_a determined on the first zero position Nl of the load current (moment A, figure 4a), the first switching element kl is conducted into off state while the second switching element k2 is conducted into on state (moment B, figures 4b and 4c). Then the load current I is disposed to pass across the first switch element kl and the forward- current second diode d2 until the load current I_L is reversed at the second zero position N2 of the current (moment C, figure 4a), and then the load current I_L is reversed relative to the second diode d2, thus preventing current flow to the load L (moment C, figure 4e). During the second half cycle PB of the alternating voltage V_AC after the delay t_a determined on the second zero position N2 of the load current, the second switching element k2 is conducted into off state while the first switching element kl is conducted into on state (moment D, figures 4b and 4c). Then the load current I_L is disposed to pass across the second switch element k2 and the forward-current first diode dl until the load current is reversed again at the following zero position N3 of the current (moment E, figure 4a). With regard to the change of the load current I_L, this zero position N3 corresponds to the first zero position Nl, where the load current I_L is reversed relative to the first diode dl. The first diode dl thus prevents current flow to the load until, in the course of the third half cycle, which corresponds to the first half cycle PA, the operation steps described above are repeated. After the delay t_a determined on the zero position Nl, N2 of the current, the load voltage V_L of figure 4d is acting across the load L until the following zero position N2, N3 of the current. With resistive load L, the load current I_L follows the load voltage N_L in the same phase, as appears in figures 4d and 4e. The load voltage V_L follows the alternating voltage V_AC of the alternating current source AC over the periods B-C and D-E.

Based on the alternating voltage V_AC of the alternating current source AC, successive positive PA and negative PB half cycles are determined, and based on these, the control of the switching elements kl, k2 is carried out by means of control unit 3. In so doing, the forward directions of diodes dl, d2 are taken into account by switching unit 2. In the embodiment example of figure 1, diodes dl, d2 are switched to one of the anodes and between the switching elements kl, k2. They are directed forward and away from a common switching point. This switching point is also used as the virtual ground VM of control unit 3. In this case, switching elements kl, l 2 are controlled alternatingly with the first switching element kl conductive during the positive half cycle PA of the alternating voltage V_AC after a delay t_a and accordingly, the second switching element k2 conductive during the negative half cycle PB after a delay t_a.

Optionally, the pairs of switching element diodes kl, dl, and k2, d2, respectively, of the switching unit 2 in the embodiment example of figure 1 can be disposed to switch into reversed positions, with the second switching element k2 and the second diode d2 on the alternating current source AC side and the first switching element kl and the first diode dl on the load L side. The diodes dl, d2 are thus mutually connected with their cathodes connected to each other and f rther between switching elements kl, k2 in virtual ground VM. They are directed forward towards a common switching point, i.e. virtual ground VM. In this case, viewed from alternating current source AC, switching elements k2, kl are conducted with the first switching element l 2 conductive during the negative half cycle PB of the alternative voltage V_Ac after ^a delay t_a and accordingly, with the switching element kl conductive during the positive half cycle PA after a delay t_a from the zero position.

Figure 2 is a block diagram of the power controller 1 implementing the method of the invention. The alternating current power supplied from the alternating current source AC to the load L is controlled by controlling the phase angle of the alternating voltage V_AC- The power controller 1 comprises a switching unit 2 and a control unit 3. The switching unit 2 comprises two MOSFET power transistors 21, 22 connected in succession, which act as switching elements kl, k2. In this embodiment example, they are serially connected with their sources SI, S2 connected together and further to the virtual ground VM of the power controller 1. Then the drain Dl of the first MOSFET transistor 21 is the input terminal of switching unit 2 and it is connected to the alternating current source AC. The drain D2 of the second MOSFET transistor 22 is the output terminal of switching unit 2 and it is connected to the load L. Each MOSFET transistor includes, besides the actual transistor switch, which corresponds to the switching element kl, k2 of the power controller of figure 1, an additional parallel diode built in connection with the transistor, a "body diode" dl, d2, which is disposed in forward direction between the source SI, S2 and the drain Dl, D2. MOSFET transistors 21, 22 are controlled by means of control voltages supplied to their gate terminals Gl, G2 from control unit 3 in turn into electrically conductive, i.e. off state and non-conductive, i.e. on state, respectively, during one half cycle of the supply voltage i.e. alternating voltage V_AC-

Control unit 3 comprises a half cycle detector 4, a synchronisation unit 5, a switch control unit 6 and a setting unit 7. The half-cycle detector 4 detects the positive PA and negative PB half cycle of the alternati-ag voltage V_Ac to be supplied. The synchronisation unit 5 detects inversion of the load current I_L, i.e. the zero position Nl, N2, N3 of the current (cf. figures 4a and 5a). The setting unit 7 determines the set value of the delay t_a. The synchronisation unit 5 is connected with the setting unit 7. The synchronisation unit 5 informs the setting unit 7 about the zero positions Nl, N2, of the current I and this information is detennined as the starting moment of the delay t_a. The setting unit 7 is supplied with the set value of the delay t_a from the outside of the control unit 3, this set value determining the electric power to be supplied to the load L. The setting unit 7 is connected to the switch control unit 6, which is immediately informed by the setting unit 7 of the set delay t_a having been achieved. The half-cycle detector 4 is also connected to the switch control unit 6. The half-cycle detector 4 informs the switch control unit 6 of the positive PA and negative PB half cycle of the alternating voltage V_Ac- Using the switch control unit 6 and based on data supplied by the half-cycle detector 4 and the setting unit 7, both the MOSFET transistors 21, 22 acting as switching elements kl, k2 are controlled so as to get into reverse states simultaneously; as the one MOSFET transistor is brought into conductive state, i.e. off state, the other one is brought into non- conductive i.e. on state, as has been explained in principle in conjunction with figure 1 above.

A preferred control unit 3 of the power controller 1 of the invention is illustrated in figure 3. It has the following embodiment.

The half-cycle detector 4 is e.g. a level detector, which has been formed of resistor 40 and zener diode 41. The level detector is disposed to monitor the voltage between the input terminal of the power control unit 1 and thus of the switching unit 2 on the one hand, and the virtual ground VM on the other hand. As the instantaneous value of V_Ac ri^ses above the zener voltage Vz of the zener diode 41, a voltage pulse lasting substantially over one half cycle is obtained in the output of the half-cycle detector, the pulse being marked e.g. as state "1". This is then the positive half cycle PA of the voltage V_AC of the alternating current source AC. During the negative half cycle PB of the voltage V_Ac of the alternating current source AC, the voltage across the zener diode 41 is its forward threshold voltage V_D, which is similarly marked e.g. as state "0".

Synchronisation unit 5 is a detector with a transistor 53 accomplishing the open- collector output and connected to the first input of its setting unit 7 in non- conductive state during the delay t_a. As the load current I_L is switched off, almost all of the voltage V_AC of the alternating current source AC acts between the in and out gates of the power controller. Depending on the polarity of the alternating voltage, this voltage generates current either through diode 50 or 51 to the transistor 52 base, bringing transistor 52 into conductive state. When transistor 52 is in conductive state, current flow to the base of transistor 53 from pull-up resistor 54 is prevented, so that transistor 53 is in non- conductive state. When the load current I_L is switched on, the voltage difference between the in and out gates of the controller is close to zero, so that the base of transistor 52 is not supplied with current and it is in non- conductive state. Then current flows across the pull-up resistor 54 to the base of transistor 53 and transistor 53 is in conductive state. Since the load current I_L is always switched off at its zero position, this interruption point also indicates the zero position Nl , N2, N3.

The setting unit 7 includes an RC time constant circuit consisting of a variable resistor 71 and capacitor 72 and a comparator 73. When the load current I_L is switched on, the conductive open-collector output of the synchronisation unit 5 maintains the capacitor 72 nearly uncharged. Immediately after the load current I_t, has been switched off, the open-collector output of synchronisation unit 5 passes into non- conductive state and the charge of capacitor 72 starts increasing in the manner determined by the values of resistor 71 and capacitor 72. As the charge increases, the voltage across capacitor 72 increases accordingly. When the voltage across capacitor 72 reaches the reference voltage level of comparator 73, the set pulse occurs on the output side of comparator 73. The set pulse is led to the clock input CL of the D flip flop of switch control unit 6.

The switch control unit 6 is preferably a flip-flop, especially a D flip-flop. Its data input Dl is connected to the output of the half-cycle detector 4 and its clock input

CL is connected to the output of the setting unit 7. When the set pulse is supplied from the setting unit 7 to the clock input CL, the status information prevailing in the data input Dl passes as such to the output Q of the D flip-flop, and in reverse direction to the output Q . The outputs Q and Q of the D flip-flop are connected across resistors 8 and 9 to the corresponding gates Gl, G2 of MOSFET transistors 21 and 22.

The switching speeds of MOSFET transistors 21 and 22 are set by means of resistors 8 and 9.

The electronic units 4, 5, 6, 7 of control unit 3 require a power source of their own. The operating voltage V+ is accomplished e.g. by means of a power source lαiown by those skilled in the art, which is not shown in the drawings. Generally speaking, the operating voltage V+ is preferably generated by rectifying and filtering alternating voltage V_Ac-

The power controller 1 shown in figure 2 acts with resistive load L in the same manner as the basic circuit in figure 1, which was explained above with reference to figure 4. Then figures 4b, 4c show control voltages supplied to the gates Gl, G2 of the first and the second MOSFET transistor 21, 22 as a function of time t, in which e.g. state "1" denotes the appropriate positive control voltage and state "0" a voltage lower than this, such as virtual ground VM.

Figure 5 shows schematically the curve shapes of voltage and current in connection with the power controller of figure 1. In figure 5 load L is then assumedly inductive. Figure 5a shows the curve shape of the alternating voltage V_Ac supplied from alternating current source AC during several successive half cycles PA, PB, PA,... as a function of time t. Figures 5b and 5c, respectively, show the states of switching element kl and k2, respectively, with 0 denoting the on state of the switch and 1 the off state of the switch. These figures show on principle also the control voltages supplied to the gates Gl, G2 of the MOSFET transistors 21, 22 of figure 2. The curve shapes of voltage V and current I , respectively, acting across the load L, are shown in figures 5d and 5e.

The following is an explanation of the operation of the power controller 1 of figure 2 in the control of an inductive load, with reference to figures 5; 5 a, 5b, 5 c, 5d, 5e and figure 3. At instant A, the voltage of alternating current source AC is at zero position. At instant B, the current I flowing across the pair of MOFSET transistors 21, 22 stops passing from load L to alternating current source AC, because the diode dl of the first MOFSET transistor 21 is reversed relative to the voltage across it, in other words, automatic commutation occurs. This appears at zero position Nl of load current I_L. It should be noted that due to the inductive load, the load current I is no longer in the same phase as the load voltage V_L, but follows the voltage reversal with a delay. At instant B, the synchronisation unit 5 activates the setting unit 7 and the half-cycle detector 4 conducts the data input D of the D flip-flop acting as the switch control unit into state "1", this being the positive half cycle PA of the voltage VA_C of the alternating current source AC. At instant C, the setting unit 7 gives a set pulse to the clock input CL of the D flip-flop and the Q output of the D flip-flop passes into state "1", and accordingly, the reverse-current output Q passes into state "0". Now current I_L starts flowing to load L across the conductive channel of the first MOFSET transistor 21 and the body diode d2 of the second MOFSET transistor 22. At instant D, the voltage V_Ac of the alternating current source AC reaches the zero position. At instant E, the load current I_L passing across the first MOFSET transistor 21 stops, because the body diode d2 of the second MOSFET transistor 22 is reversed relative to the voltage across it, in other words, automatic commutation occurs. This appears at zero position N2 of load current I_I- Then synchronisation unit 5 activates the setting unit 7 and the half-cycle detector 4 conducts the switch control unit 6, i.e. the data input D of the D flip-flop into state "0", because this is the negative half cycle PB of the voltage V_AC of the alternating current source AC. At instant F, the setting unit 7 gives a set pulse to the clock input CL of the D flip-flop and the Q output of the D flip-flop passes into state "0" and the reverse Q output into state "1". Now current I_L starts flowing from load L across the conductive channel of the second MOFSET transistor 22 and the body diode dl of the first MOSFET transistor 21 to the alternating current source AC. At instant G, the voltage V_Ac of the alternating current source AC is at zero position and the cycle resumes from instant A, i.e. A = G. At the following instant H = B, the situation is back at zero position N3 = Nl of the current.

In the embodiment of the invention described above, the switch unit 2 has been formed by two MOSFET transistors 21, 22 connected successively. It is obvious to those skilled in the art that the MOSFET transistors can be replaced with other power transistors connected in succession, such as IGBT transistors or a bipolar power transistor, with diodes dl, d2 disposed in parallel with these. The invention is not restricted to the embodiment example above alone, many variations being conceivable without departing from the inventive idea defined in the claims.

Claims

1. A method for controlling alternating current power supplied to a load by controlling the phase angle of the alternating voltage, in which a switching unit (2) consists of two switching elements (kl ,k2) connected in succession mutually and relative to the load (L), i.e. a first and a second switching element and reverse- current diodes (dl, d2) disposed in parallel with these, the switching elements being controlled in turn into conductive and non- conductive state, respectively, during a half cycle of the alternating voltage (V_Ac) to be supplied, characterised in that, during the first half cycle (PA), after a delay (ta) determined on the zero position (Nl) of the first load current (I_I), the first switching element (kl) is brought into conductive state while the second switching element (k2) is brought into non- conductive state, the load current (I ) being disposed to flow across the first switching element (kl) and the second forward- current diode (d2) until the load current (I_L) is reversed at the second zero position (N2) of the current, the direction of the load current (I_L) being reversed relative to the second diode (d2), thus preventing current flow to the load, and after a delay (ta) determined on the second zero position (N2) of the load current (I ) during the second half cycle (PB), the second switching element (k2) is brought into conductive state while the first switching element (kl) is brought into non- conductive state, the load current (I ) being disposed to flow across the second switching element (k2) and the first forward- current diode (dl) until the direction of the load current (I_L) is reversed again at the following zero position (Nl, N3), the direction of the load current (I_L) being reversed relative to the first diode (dl), thus preventing current flow to the load until the operation steps described above are repeated during the following half cycle, which corresponds to the first half cycle (PA).

2. A method as defined in claim 1, characterised in that successive half cycles (PA, PB) of the alternating voltage (V_Ac) are monitored and the positive and the negative half cycles are determined on which basis the control of the switching elements (kl, k2) is realized, while taking account of the forward directions of the diodes (dl, d2).

3. A power controller (1), by means of which alternating current power supplied to a load (L) is controlled by controlling the phase angle of alternating voltage, the power controller including a switching unit (2) consisting of two switching elements (kl, k2) connected in succession mutually and relative to the load (L), i.e. a first and a second switching element and reverse-current diodes (dl, d2) disposed in parallel with these, and a control unit (3), by means of which the switching elements (kl, k2) are controlled in alternating conductive and non-conductive states during a half cycle of the alternating voltage (V_Ac) supplied, characterised in that the control unit (3) comprises:

- a half-cycle detector (4), which detects the positive (PA) and negative (PB) half cycle of the alternating voltage supplied;

- a switch control unit (6), which is informed of the positive and negative half cycle (PA, PB) of the alternating voltage (V_Ac) and which selects the switching element

(kl, k2) and controls it into conductive and non-conductive state, respectively;

- a synchronisation unit (5), which detects the moment the zero position of the load current (I_L) is exceeded, i.e. the zero position (Nl, N2); and

- a setting unit (7), which is informed by the synchronisation unit (5) about the zero positions (Nl, N2) of the load current (I ) in order to determine the initial moment of the delay (ta) and which sets the delay (ta) within the half cycle (PA, PB) at the end of which the current supply to the load (L) is started by informing the switch control unit (6) of the end of the delay, the switch control unit being disposed to connect either of the switching elements (kl, k2) into conductive state, depending on the prevailing half cycle (PA, PB) of the alternating voltage cycle, the load current (I_L) being disposed to flow across one diode and one switching element (kl, d2; k2, dl); and in which

current supply to the load is switched off in the power controller substantially at the zero position (Nl, N2) of the load current (I_L) as the current is reversed relative to the forward direction of the diode (d2; dl).

4. A power controller as defined in claim 3, characterised in that the switching unit (2) has been formed by two MOSFET transistors (21, 22), connected in succession.

5. A power controller as defined in claim 3, characterised in that the switching unit (2) has been formed of two power transistors connected in succession, such as an IGBT or bipolar transistor, with diodes (dl, d2) disposed in parallel with these.

6. A power controller as defined in claim 3, 4 or 5, characterised in that the switch control unit (6) is a D flip-flop, whose mutually reversed outputs (Q, Q) are connected to the control terminal of the first and the second switching element (kl, k2), respectively.

7. A power controller as defined in any of the preceding claims 3-6, characterised in that the half-cycle detector (4) is a level detector, which has been formed of a resistor (40) and a zener diode (41), the level detector being disposed to monitor the voltage between the input terminal (in) of the power control unit (1) and thus of the switching unit (2), on the one hand, and the virtual ground (VM), on the other hand.

8. A power controller as defined in any of the preceding claims 4-7, characterised in that the synchronisation unit (5) is a detector that detects the reversal of the direction of the load current (I_L) from positive to negative and vice versa and informs the setting unit (7) about the reversal.

9. A power controller as defined in any of the preceding claims 4-8, characterised in that the setting unit (7) is a timer circuit, which is activated by commutation information provided by the synchronisation unit and whose delay (ta) is set by appropriate external control.