WO1998043346A2

WO1998043346A2 - Apparatus for generating an analog ac signal from dc voltage

Info

Publication number: WO1998043346A2
Application number: PCT/FI1998/000243
Authority: WO
Inventors: Seppo Suoranta; Jyrki Valkeakari
Original assignee: Nokia Telecommunications Oy
Priority date: 1997-03-20
Filing date: 1998-03-19
Publication date: 1998-10-01
Also published as: AU6501398A; FI971179A0; WO1998043346A3; FI971179A

Abstract

The invention relates to an apparatus for generating an analog AC voltage signal from a DC voltage source. A high-frequency switched-mode power converter (5) bi-directionally transferring power generates from a DC voltage a low-frequency voltage in the form of a full-wave rectified sine wave or another full-wave rectified waveform. A full-bridge inverter (6) converts the voltage in the form of a full-wave rectified sine wave or another full-wave rectified wave form into a voltage in the form of a pure sine wave or another waveform. The full-bridge inverter (6) operates at the frequency of said sine wave or another waveform.

Description

APPARATUS FOR GENERATING AN ANALOG AC SIGNAL FROM DC VOLTAGE

The invention relates to an apparatus for generating an analog AC voltage signal from a DC voltage.

Generating AC voltage levels that are considerably higher than the operating voltage is problematic particularly in battery-driven or battery backed-up apparatuses whose operating voltage is, for example, within the range from 5 to 48 V. Such apparatuses include battery-driven or battery backed-up telephone systems and telephone sets where a ringing voltage of 25 Hz 75 VAC should be generated from the battery voltage. Another typical application is to generate a mains voltage of 230 VAC 50 Hz for example from a 12-V battery.

High AC voltages have been generated from a DC voltage by using a variety of solutions, such as power amplifiers, transformers, mechanical rotary generators, etc. A typical approach to generating AC voltages is a switched-mode technique.

FI 822762 discloses a pulse-width modulated power amplifier whose power stage includes a power transformer having four windings. A primary winding has a tap output connected to one terminal in a DC voltage source, whereas both ends of the primary winding are connected through a corresponding pair of switches to the other terminal in the DC voltage source. Both pairs of switches can conduct in both directions. The ends of a secondary winding are connected through a corresponding pair of switches to one output terminal of the power amplifier. The tap output of the secondary winding is connected to the other output terminal. The output voltage of the amplifier is compared with an input signal to be amplified, and an error signal thus formed controls a pulse-width modulator that, in turn, controls the pairs of switches. In this way, a high-level AC voltage having two sides (a positive and a negative half cycle) and whose waveform follows the input signal waveform is inverted directly from the battery voltage by means of high-frequency switches. A dis- advantage of this known circuit is the large number of high-frequency components required (at least four).

W093/19516 discloses a DC/DC/AC power supply wherein a pulse- width modulator controls an FET switch by supplying power to the primary winding of a power transformer. The secondary side comprises two synchro- nous FET rectifiers supplying negative and positive half cycles to a second transformer stage. The desired ringing voltage is obtained from the secondary side of the second transformer stage. In this circuit, a high-frequency AC voltage having two sides (a positive and a negative half cycle) is also inverted directly from the battery voltage by means of high-frequency switches. It is a disadvantage that at least three high-frequency switch components (FET) are required, two of which being controlled by means of pulse transformers for achieving a sufficient isolation and buffering. Another drawback is the number of required transformer stages increasing the size of the apparatus. The third problem is that reactive power within a reactive load must be converted into heat losses on the secondary side. This requires that cooling (cooling ele- ments) must be arranged in order to convey the heat off from the apparatus. The cooling elements need space thus increasing the size of the apparatus, which is contrary to the current tendency towards reducing the size of apparatuses.

EP 0489512 discloses a ringing signal generator wherein a pulse- width modulator controls a high-frequency switch on the primary side of a power transformer. The secondary side comprises a pair of rectifiers implemented by high-frequency switches (FET) and diodes. The pair of rectifiers passes a negative and a positive half cycle, alternately, to charge a capacitor at the output. A sinusoidal signal is thus obtained directly at the output. Reac- tive power within a reactive load must be converted into heat on the secondary side.

EP 375250 discloses an electronic ringing signal generator based on a bi-directional switched-mode circuit. A sinusoidal signal is obtained directly by means of the generator, the sinusoidal signal, however, appearing entirely within a positive voltage range (between earth and the operating voltage). An output signal includes therefore a DC voltage component that must be eliminated by means of a large series capacitor. However, the series capacitor prevents a DC loop circuit in a subscriber line from closing via the apparatus, whereby a separate DC circuit is needed for this purpose. GB 2187312 discloses a bridge circuit controlling the supply of current from a DC source to a load. A switching transistor that is connected in series with the primary winding of a power transformer is controlled by means of a high-frequency pulse-width modulated signal. A pulse-width modulator is in turn controlled by the difference between a signal that is full-wave rectified from a low-frequency sinusoidal signal and a full-wave rectified signal extracted from the secondary side. As a result, on the secondary side of the power transformer there is a low-frequency full-wave rectified sinusoidal signal across a capacitor. From the full-wave rectified sinusoidal signal is generated a pure sinusoidal signal across the load by means of a full-bridge inverter controlled by said low-frequency sinusoidal signal. This known solution en- ables the reduction in the number of high-frequency components. However, in the circuit according to GB 2187312, reactive power within a reactive load must also be converted into heat on the secondary side causing the above described cooling problems. The capacitor on the secondary side must therefore be carefully chosen taking into account the properties of the load at the desired waveform for enabling dissipation of energy in the load. However, for example a ringing voltage generator load in telephone exchanges varies greatly according to the number of telephone sets to which the ringing voltage is supplied at a given time. In addition, the ringing voltage generator load is highly capacitive, a substantial amount of energy thus returning from the load rather than dissipating in the load. In addition, at the output circuit of the ringing voltage generator, there occasionally flows a significant DC component either returning or forwarding energy depending on the instantaneous phase of the sine wave. It is therefore impossible to select a value for the capacitor on the secondary side of the circuit in accordance with GB 2187312 in such a way that a sufficiently pure sine wave would be obtained as the output voltage in actual load situations.

The object of the invention is to provide a new type of AC voltage generating apparatus for avoiding conversion of reactive power into heat on the secondary side, and wherein fewer high-frequency switches are required than in prior art solutions.

This is achieved by the apparatus of the invention for generating an analog AC voltage signal from a DC voltage source, which apparatus is characterized in that it comprises a high-frequency switched-mode power converter generating from a DC voltage a low-frequency voltage in the form of a full- wave rectified sine wave or another full-wave rectified waveform and transferring power bi-directionally between the primary side and the secondary side, and a full-bridge inverter converting the voltage in the form of a full-wave rectified sine wave or another full-wave rectified waveform into a voltage in the form of a pure sine wave or another pure waveform, said full-bridge inverter operating at the frequency of said sine wave or another waveform. The basic idea of the present invention is that a switched-mode power converter transferring energy bi-directionally and operating at a high frequency is used for generating at first from a DC voltage a voltage in the form of a full-wave rectified sine wave. In other words, both half cycles (a negative and a positive) of the voltage are as a successive sequence on the same side of a ground potential (for example on the positive operating voltage side). The voltage in a form of a full-wave rectified sine wave is inverted into a pure sine wave by means of a full-bridge inverter circuit operating at the frequency of the sine wave. The waveform of the generated AC voltage may be some other than a sine wave, whereby it can be referred to as a full-wave rectified AC voltage that is inverted by means of a full-bridge inverter into a pure AC voltage.

The invention allows the implementation of a switched-mode power converter by means of a very simple constructure comprising a minimum number of high-frequency switches, i.e. two switches. Only one of these, i.e. the switch on the output side, may have to be controlled by means of a pulse transformer. The same constructure and operation principle that provides the full-wave rectified AC voltage waveform also makes the switched-mode power converter transfer energy bi-directionally. As a result of this, the switched- mode power converter is able to transfer reactive power from the load back to the DC voltage supply, whereby the apparatus of the invention can be used for supplying reactive loads without a need for converting reactive power into heat losses. Because of this, separate cooling elements are not needed, whereby there is less need for space. The power transfer ratio of the apparatus of the invention is in practice about 80%, whereas the power transfer ratio in conventional solutions remains under 50%.

As the output of the switched-mode power converter is a full-wave rectified waveform, it does not include a significant DC voltage component. Thus, there is no need for a large series capacitor used in some of the con- ventional apparatuses.

One of the advantages of the invention is that the full-bridge inverter can operate at the same frequency as the AC voltage to be generated, for example at the frequency of the sine wave. In other words, instead of high- frequency switches, the full-bridge inverter can be implemented by means of low-frequency components and the number of high-frequency switches can be reduced to a minimum of two. Low-frequency switches cost less and their power consumption is lower than that of high-frequency switches. The switched-mode power converter preferably comprises a power transformer isolating a DC voltage side from an AC voltage output. In a preferred embodiment of the invention, a bi-directional switched-mode power converter is im- plemented by means of a flyback topology. Other structures may also be used as a bi-directional switched-mode power converter providing a galvanic isolation. For example an isolating half-bridge switched-mode circuit having controlled switches in parallel to second diodes acts in the reverse direction as a current-driven push-pull switched-mode circuit. If galvanic isolation is not nec- essarily needed, for example a pure half bridge circuit can be used as the bidirectional switched-mode power converter.

In the following, the invention will be described in more detail by means of the preferred embodiments with reference to the accompanying drawings, in which Figure 1 is a block diagram generally illustrating the ringing signal generator of the invention,

Figure 2 is a block and a wiring diagram presenting the ringing signal generator according to a preferred embodiment of the invention, and

Figure 3 is a wiring diagram for the full-bridge inverter 6 of Figure 4. In Figure 1 , an oscillator 1 generates a reference AC voltage having the desired frequency and the desired waveform. In the case of the example, the oscillator 1 generates a sinusoidal AC voltage whose frequency is 25 Hz. A peak value is, for example, about 4 V. The sinusoidal AC voltage is rectified by means of a rectifier circuit 2 in such a way that an AC voltage in the form of a full-wave rectified sine wave and comprising negative half cycles is supplied to the input of a differential amplifier 3. A positive feedback voltage proportional to the output voltage of the power converter 5 is supplied from a power converter 5 to the other input of the differential amplifier. This voltage type is an AC voltage in the form of a full-wave rectified sine wave and comprises positive half cycles. The differential amplifier 3 compares the feedback voltage proportional to the output voltage of the power converter with the reference voltage given by the oscillator 1 (via the rectifier 2) and generates a current or a voltage that is proportional to the difference between the two and controls the pulse width of a pulse-width modulator 4. The pulse-width modulated out- put signal of the pulse-width modulator 4 in turn controls switches on the primary and the secondary side of an isolating bi-directional power converter 5. An AC voltage in the form of a full-wave rectified sine wave is generated at the output of the power converter. Depending on the circuit design, this voltage is entirely on the positive or on the negative side of the ground potential. The full- wave rectified sine signal is applied to a full-bridge inverter or an inverter 6 operating at a sine wave frequency of 25 Hz. The full-bridge inverter 6 is preferably controlled by a 25-Hz signal generated by the oscillator 1.

It is to be noted that the above described principle of implementing the AC voltage generator of the invention is only one example of the potential implementations. It is essential to the basic idea of the invention that the bi- directional high-frequency power converter 5 is used for generating the AC voltage that is in the form of a full-wave rectified sine wave and is inverted into a pure sine wave by the full bridge circuit operating at the same frequency as the sine wave. In this case, high frequency refers to power converter switching frequencies higher than 10 kHz, preferably not lower than 100 kHz. The man- ner in which the bi-directional power converter 5 is controlled is not essential to the invention. The oscillator 1 , the rectifier 2, the differential amplifier 3 and the pulse-width modulator 4 may therefore be replaced with some other control solution obvious to those skilled in the art.

Figure 2 gives a more detailed description of the implementation of the ringing voltage generator of the type shown in Figure 1 and particularly of the implementation of a bi-directional isolated power converter. The actual input voltage from which the desired AC voltage is generated is 48 V. The operation of an oscillator 1 , a rectifier 2 and a differential amplifier 3 corresponds to that presented in Figure 1. Full-bridge inverters 6 substantially correspond to those presented in Figure 1. In the embodiment in Figure 2, circuits 1 , 2, 3 and 4 have a 12-V operating voltage having a common ground terminal (or a negative terminal) with a 48-V supply voltage. In addition, a voltage reference of 5 V DC is generated and used as virtual earth for circuits on the primary side, since the circuits have a unipolar 12-V operating voltage. The power converter presented in more detail in Figure 2 is based on a bi-directional switched-mode circuit of a flyback type. The power converter comprises a power transformer TR1 with a primary winding M1 , a secondary winding M2 and an auxiliary winding M3. A DC voltage (for example from a battery power source) whose nominal value is 48 V is connected across the secondary winding. A filter capacitor C2 is also connected across the primary winding M1. The lower end of the primary winding is connected to earth through a forward-biased diode, a field-effect transistor switch T1 and a resistor R1. The purpose of a diode D1 is to prevent the current in a reverse direction (a negative half cycle) from flowing through the transistor switch T1. A reverse-biased diode D2 is parallel to D1 , T1 and R1. A filter circuit composed of a capacitor C1 and a resistor R2 is parallel to the diode D2.

Correspondingly, the lower end of the secondary winding M2 is connected to earth through a series connection between a forward-biased diode D6, a field-effect transistor switch T2 and a resistor R7. The diode D6 prevents the current in a reverse direction (a negative half cycle) from flowing through the transistor switch T2. A series connection between reverse-biased diodes D4 and D5 is parallel to D6, T2 and R7. A filter circuit composed of a series connection between a capacitor C6 and a resistor R5 is parallel to the diode D4, and a filter circuit composed of a capacitor C7 and a resistor R6 is parallel to the diode D5. The upper end of the secondary winding is connected to earth through a capacitor C4 and to one output terminal 202 of the power converter. In addition, the upper end of the secondary winding is connected via a coil L1 to the other output terminal 201 of the power converter. A filter capacitor C5 is connected between the output terminals 201 and 202. The bi-directional flyback switched-mode power stage presented in

Figure 2 is able to transfer energy in both directions and it generates a positive voltage in the form of a rectified sine wave on the secondary side at the terminals 201 and 202. When energy is transferred from the voltage source on the primary side to the load on the secondary side, it always first changes into the energy of the magnetic field of the transformer TR1. In other words, when the transistor T1 conducts, the current starts to flow in the winding M1 , and energy is charged from the voltage source to the magnetic field of the transformer TR1. When the transistor T1 is driven to a non-conductive state, the current flow through the primary winding M1 is cut off, and energy in the magnetic field of the transformer TR1 is discharged to the secondary side, in other words, the current is transferred to the secondary winding M2 and flows through the diodes D4 and D5 on the secondary side to the capacitor C4 and charges the capacitor. The coil L1 and a capacitor C5 filter off the switching frequency (for example 100 kHz) of the transistor T1 from the voltage across the capacitor C, and, at the output 201 , 202 there is an AC voltage in the form of a full-wave rectified sine wave. Correspondingly, the transistor T2 is able to transfer energy from the secondary side to the primary side. In other words, as the current produced by the above described transfer of energy in a primary-secondary direction (that charges the capacitor C4) in the secondary winding has fallen to zero (all the energy is transferred) and the transistor switch T2 conducts, the current starts to flow back in the secondary winding M2 through the transistor switch T2, i.e. the current starts from the capacitor C4 back to the secondary winding M2, whereby energy is transferred from the capacitor back to the magnetic field of the power converter TR1. When the transistor switch T2 is driven to a non-conductive state, the current flow in the secondary winding M2 is cut off, and the energy in the magnetic field of the transformer is discharged to the primary side, in other words, the current is transferred to the primary winding M1 and flows through the diode D2 to the capacitor C2 and to the voltage source of 48 V DC. A pulse-width modulator 4 adjusts the primary winding M1 current by controlling the field-effect transistor T1. The pulse-width modulator 4 measures the current flowing through the primary circuit M1 and the transistor switch T1 by measuring the voltage acting across the resistor R1. This voltage is compared with an offset voltage received from the differential amplifier 3. The pulse-width modulator 4 switches the transistor T1 to a conductive state at a frequency of about 100 kHz. The pulse-width modulator drives the transistor switch T1 to a non-conductive state (off) always when the voltage measured across the resistor R1 has increased to the offset voltage given by the differential amplifier 3. The offset voltage is the difference between a negative recti- fied sine wave reference received from the rectifier 2 and a positive rectified voltage 200 proportional to the secondary voltage of the transformer TR1. The voltage 200 proportional to the secondary voltage is obtained by rectifying the voltage acting across the auxiliary winding M3 by means of a diode D3. The pulse width of the of control pulses supplied to the transistor T1 by means of the pulse-width modulator is thus proportional to the amplitude of a sine wave. An envelope acting across the capacitor C4 thus follows the shape of the full- wave rectified sine wave, the shape being obtained to the output 201 and 202 of the power converter when a high switching frequency is filtered off by means of the coil L1 and the capacitor C5. The pulse-width modulator 4 also controls the transistor switch T2 on the secondary side in such a way that the transistors T1 and T2 conduct by turns. If the pulse ratio of a control signal passing to the transistor T1 falls to zero and the transistor T1 is no longer controlled, the control signal is, however, further supplied to the transistor T2 in such a way that the operation of the transistor T2 enables ringing voltage energy to be transferred from the ca- pacitor C4 on the secondary side back to the capacitor C2 on the primary side and to the voltage source of 48 V DC, as it was described above.

In the following, the structure and operation of a full-bridge inverter 6 is described in more detail with reference to Figure 3. The full-bridge inverter 6 converts a voltage in the form of a rectified sine wave appearing at output terminals 201 and 202 of a power converter 5 into a pure 25-Hz sine wave.

In Figure 3, field-effect transistors T4, T5, T7 and T8 form a full bridge where field-effect transistors, always cross-connected, conduct simultaneously. In other words, when field-effect transistors T3 and T7 conduct, field- effect transistors T4 and T6 are in a non-conductive state. Correspondingly, when the field-effect transistors T4 and T6 conduct, the field-effect transistors T3 and T7 are in a non-conductive state. The full-bridge inverter is controlled by a 25-Hz signal 203 coming through an opto-isolator 25 from an oscillator 1 (Figure 2). The full bridge 6 is arranged to change its state always at the zero points of the output voltage. The lower field-effect transistors T4 and T7 of the full-bridge inverter are controlled directly by a push-pull principle by means of the 25-Hz signal 203 amplified by transistors T9 and T10. Controlling the upper field-effect switches T3 and T6 depends on the control of the corresponding lower field-effect transistors T4 and T7. For example, when the field-effect transistor T4 conducts, emitters of the source electrode of the field-effect tran- sistor T3 and those of the bipolar transistor T5 controlling the source electrode are in practice at the zero voltage on the secondary side of the power converter 5. In that case, the transistor T5 also receives through a base current diode D8 and conducts, which pulls down the gate voltage in the transistor T3, whereby the transistor T3 no longer conducts. At the same time, a capacitor C11 is charged through a diode D10 nearly to the operating voltage. When a 25-Hz sine signal 202 changes its state, the gate voltage in the field-effect transistor T4 is pulled down, whereby the base current in the field-effect transistor T5 is cut off. Consequently, the gate voltage in the field-effect transistor T3 rises through a resistor R8, and the field-effect transistor T3 starts to con- duct. The gate voltage in the field-effect transistor T3 remains high since the gate receives the voltage from the capacitor C11 floating with the field-effect transistor T3. When the AC voltage in the form of a rectified sine wave supplied from the power converter again reaches the zero point, the gate voltage in the field-effect transistor T4 and the base current in the transistor T5 again rise, and the above described cycle is repeated. The field-effect transistors T6 and T7 act in a corresponding manner. When the field-effect transistor T7 conducts, the source electrode of the field-effect transistor T6 and the emitter electrode of the bipolar transistor T8 controlling the source electrode are in practice at the zero voltage on the secondary side of the power converter 5. In that case, T8 also receives base cur- rent through the diode D9, pulling down the gate voltage in the field-effect transistor T6, whereby the field-effect transistor T6 no longer conducts. At the same time, a capacitor C12 is charged through a diode D11 nearly to the operating voltage. When the 25-Hz sine signal given by the opto-isolator 25 changes its state, the gate voltage in the transistor T7 is pulled down, causing the base current in the field-effect transistor T8 to be cut off. Consequently, the gate voltage in the field-effect transistor T6 rises through a resistor R15, and the field-effect transistor T6 starts to conduct. The gate voltage in the field- effect transistor T6 remains high since the voltage is received from the capacitor C12 floating with the field-effect transistor T6. When the AC voltage in the form of a rectified sine wave at the secondary of the power converter 5 again reaches the zero point, the gate voltage in the field-effect transistor T7 and the base current in the transistor T8 again rise, and the cycle is repeated.

The switching point between the transistors T3 and T4 is connected to an output terminal 204. The switching point between the transistors T6 and T7 is connected to an output terminal 205. The output terminals 204 and 205 are connected to earth by means of filter capacitors C8 and C9, respectively. Between the terminals 204 and 205 there is a sinusoidal AC voltage of 75 V.

The figure and the related description are only intended to illustrate the invention. The details of the invention can vary within the scope and spirit of the attached claims.

Claims

1. An apparatus for generating an analog AC voltage signal from a DC voltage source, characterized in that the apparatus comprises a high-frequency switched-mode power converter (5) generating from a DC voltage a low-frequency voltage in the form of a full-wave rectified sine wave or another full-wave rectified waveform and transferring power bi- directionally between the primary side and the secondary side, a full-bridge inverter (6) converting the voltage in the form of a full- wave rectified sine wave or another full-wave rectified waveform into a voltage in the form of a pure sine wave or another pure waveform, said full-bridge inverter (6) operating at the frequency of said sine wave or another waveform.

2. An apparatus as claimed in claim 1, characterized in that the switched-mode power converter (5) is arranged to galvanically isolate an input stage and an output stage.

3. An apparatus as claimed in claim 2, characterized in that the switched-mode power converter (5) is a switched-mode circuit of a flyback type or a galvanically isolating half-bridge switched-mode circuit.

4. An apparatus as claimed in claim 1, characterized in that the switched-mode power converter (5) is a half-bridge circuit.

5. An apparatus as claimed in claim 1, characterized in that the apparatus is a ringing voltage generator for telephone systems.

6. An apparatus as claimed in claim 1, characterized in that high-frequency is a frequency higher than 10 kHz.

7. An apparatus as claimed in any one of claims 1 to 6, c h a r a c - t e r i z e d in that the switched-mode power converter comprises a power transformer comprising a primary winding and a secondary winding, a first high-frequency switching means operationally connected in series with the primary winding, a second high-frequency switching means connected in series with the secondary winding, a control means for controlling the first and the second switching means alternately into a conductive and non-conductive state for transferring energy from the primary side to the secondary side and for returning reactive power from the secondary side to the primary side.

8. An apparatus as claimed in claim 7, characterized in that the control means is arranged to control the first switching means into a conductive state for switching on the current to flow through the primary winding and for charging energy from the DC voltage source to the power transformer and, after that, for controlling the first switching means into a non- conductive state in order to cut off the current flowing through the primary winding and in order to discharge the energy charged in the power transformer to the secondary side, and that the control means is arranged to control the second switching means into a conductive state for switching on the current to flow through the secondary winding and for charging reactive power from the secondary side to the power transformer and, after that, for controlling the second switching into a non-conductive state in order to cut off the current flowing through the secondary winding and in order to discharge the energy charged in the power transformer to the primary side.