EP1114505A1

EP1114505A1 - Switched-mode power supply

Info

Publication number: EP1114505A1
Application number: EP98950054A
Authority: EP
Inventors: Lothar Heinemann; Jochen Mast; Franz-Jozef Moers
Original assignee: DaimlerChrysler AG
Current assignee: Mercedes Benz Group AG
Priority date: 1998-09-18
Filing date: 1998-09-18
Publication date: 2001-07-11
Also published as: JP2002526021A

Abstract

The invention relates to a switched-mode power supply having a transformer (TR1) that is connected on the primary side to a circuit generating a voltage with zero passage per period and to the series connection of a rectifier diode (D3) having an output filter inductor (LF1) on the secondary side, wherein the parallel connection of a switch (T2) having an inverse diode is mounted between the connecting point of the rectifier diode with the output filter inductor and the other terminal on the secondary side and an output capacitor (Ca1) is provided between the output terminals of the switched-mode power supply (2a, 2b). A pulse width modulator (9) controls the switch (T2) depending on output direct current (Ua1) and synchronizes it with the zero passage of the voltage on the primary side, wherein a synchronizer (8) receives the alternating current applied on the transformer (TR1) on the secondary side and forms corresponding synchronization signals for the pulse width modulator and wherein an inverter (10) receives the output direct current (Ua1) and inversely feeds it to the pulse width modulator.

Description

Switching power supply

description

The invention relates to a switching power supply according to the preamble of claim 1. The switching power supply according to the invention, which separates very high potentials, can be used, for example, for IGBT gate drives of a traction-compatible IGBT converter for supplying a DC auxiliary power supply bus.

Switched-mode power supplies are usually dimensioned in such a way that a small volume, high reliability, good electromagnetic compatibility, high efficiency, high dynamics and the lowest possible price result. Furthermore, high insulation requirements are often to be observed, which complicate the technically good and at the same time inexpensive design of the power supply.

With R. Jovanovic, R. Farrington, FC Lee; Constant-Frequency Zero-Voltage-Switched Multi-Resonant Converters, IEEE Power Electronics Specialists Conference 1990, pages 197 - 205, is achieved by replacing the secondary-side freewheeling diode of a zero-voltage switched multi-resonance converter (ZVS-MRC = Zero Voltage Switched Multi-Resonant Converter) with a field effect transistor Operation of the converter possible at constant frequency. The field effect transistor is switched on without loss if the parasitic inverse diode is conductive during the natural freewheeling phase. The output voltage is stabilized by a variably kept on time of the secondary field effect transistor. However, the functionality and implementation of the regulation is not disclosed. However, the current and voltage curves shown indicate that the control signals for the primary and secondary switches are generated synchronized on the primary side using the same pulse voltage source. With this type of control, however, as with conventional switching power supplies, additional potential isolation for the control signals is necessary, for example by means of an optocoupler or by means of a pulse transformer.

The use of an optocoupler is to be regarded as disadvantageous due to the low reliability, the unfavorable drift behavior (poor long-term and temperature stability) as well as the high price and large volume, in particular with very high insulation voltages - more than 10 kV. For similar reasons, a pulse transformer can no longer be used efficiently with very high insulation requirements.

The invention has for its object to provide a switching power supply of the type mentioned, which meets very high insulation requirements - 10 kV and more - that generates the desired voltages with high precision and that is compact and simple.

This object is achieved in connection with the features of the preamble according to the invention by the features specified in the characterizing part of claim 1.

The advantages that can be achieved with the invention are, in particular, that the number of relatively expensive, electrically isolating components is minimized in the proposed switched-mode power supply, because only a power-transmitting transformer is required as an electrically isolating component, while an optocoupler or pulse transformer for feedback of the controlled variable (generally one of the Output voltages) from the secondary to the primary side is not necessary. Overall, an inexpensive, reliable switching power supply that is suitable for very high insulation requirements is created. The high level of reliability can be explained by the small number of components in the power path and the simple, but effective electronics. Pulse-shaped component loads are avoided by soft current and voltage profiles. This also results in good electromagnetic compatibility of the switching power supply. The cost-effectiveness is explained by the fact that the components of the power section are optimally electrical and be used thermally. The secondary switch switches without loss. The switching power supply has a very high efficiency.

In the case of the proposed PWM control, it is completely irrelevant at what frequency the clocking on the primary side and how the control signals for the primary-side switch are formed (for example resonant mode controller, PWM controller or self-oscillating arrangement). The proposed type of synchronization enables the power supply unit to be operated on the primary side with a variable switching frequency (frequency-modulated), which enables an additional degree of freedom in dimensioning - in particular with a wide input voltage range. At the same time, it is always ensured that the PWM control, which is synchronized on the secondary side, operates exactly frequency-synchronously with the control circuit of the primary-side switch. This has the following main advantages:

With a self-controlling design of the primary side (ensuring zero voltage switching), relatively large tolerances of the components of the resonance circuit and the oscillator circuit, which are caused, for example, by the manufacturing process, signs of aging and temperature dependence, can be automatically compensated for. The operating frequency adjustment of the primary side caused by self-control is automatically taken over by the secondary side.

To achieve a wide input voltage range, a change in the switching frequency depending on the input voltage can be implemented in a simple manner. For example, the frequency is increased as the input voltage increases, which has an extremely favorable effect on the efficiency of the power supply unit at high input voltages. Such behavior is only possible with the use of the proposed control concept.

The invention is explained below with reference to the embodiments shown in the drawing. Show it:

1 shows the basic design of the switching power supply, 2 shows the time course of quantities of interest,

3, 4 variants for generating more than one DC output voltage.

1 shows the basic design of the switching power supply. It is a zero voltage switched multi-resonance converter based on a flux converter. A transformer TR1 can be seen, which is connected on the primary side to the input terminals 1a, 1b of the switching power supply via a series inductance LR and the parallel connection of a switch T1, an inverse diode D1 and a series capacitor Cs. The transformer TR1 is the only link between the primary and the secondary side of the switching power supply. The input terminals 1a, 1b are connected to an input capacitor Ci. The DC input voltage Ui is present between these input terminals.

The transformer T1 is connected to a capacitor Cp on the secondary side. The secondary AC voltage applied to the capacitor is designated Usek. The series connection of a rectifier diode D3 and an output filter inductance LF1 lies between the positive output terminal 2a of the switching power supply and the secondary winding of the transformer. The negative output terminal 2b of the switching power supply is connected directly to the secondary winding of the transformer TR1. An output capacitor Ca1 is connected in parallel with the output terminals 2a, 2b. The output DC voltage Ua1 is present between the output terminals 2a, 2b.

The parallel connection of a switch T2 and an inverse diode D2 lies between the connection point of the rectifier diode D3 with the output filter inductance LF1 and the negative terminal on the secondary side. The switch T2 is used in conjunction with a synchronized PWM control on the secondary side (PWM = pulse width modulation) for controlling the DC output voltage Ua1. Three function groups are used for this purpose, namely a synchronization 8, a pulse width modulator 9 and an inverter 10. The synchronization 8 receives the secondary AC voltage Usek via its input terminals 3a, 3b and outputs corresponding synchronization signals to the input terminals 4a, 4b of the pulse width modulator 9. The inverter 10 receives the DC output voltage Ua1 via its input terminals 6a, 6b and for this purpose forms inverse signals which are fed to the input terminals 5a, 5b of the pulse width modulator 9. The pulse width modulator 9 controls the switch T2 via its output terminals 7a, 7b and, for this purpose, emits corresponding control signals UGS.

The secondary-side PWM control must work exactly frequency-synchronously to the control circuit of the primary-side switch T1. For this purpose, the synchronization wins 8 synchronization signals, which determine the correct point in time for the start of each new PWM switching period. The zero crossing of the alternating voltage Usek on the secondary side very simply marks such a point in time and is therefore detected and used to generate a short synchronization pulse.

Since the control sense of the proposed solution shown in FIG. 1 is opposite to that of a standard PWM, an additional inversion point must be inserted into the control loop. However, the inverter 10 detecting the DC output voltage Ua1 fulfills another important task. Since the P component of the controlled system is too large, the control amplifier must have a P component smaller than one to guarantee stability. This is not possible with a non-inverting basic circuit of the error amplifier of a PWM. In addition, the DC conditions must be taken into account, which are created by the internal determination of the controller reference voltage in the PWM chip. By inserting the inverter 10, it is advantageously possible to dimension the actual PID control amplifier conventionally.

As already mentioned above, the secondary PWM is controlled by the pulses obtained in the synchronization. It is important that the frequency of the pulse width modulator 9 in free-running operation is approximately 10% to 20% below the switching frequency of the switch T1 on the primary side. Then there is a perfect course of the ramp voltage causing the synchronization. As can easily be seen, the proposed purely secondary control system advantageously avoids the feedback of the controlled variable from the secondary to the primary side.

Fig. 2 shows the time course of quantities of interest, namely the secondary AC voltage Usek and the control signal UGS for the switch T1. As can be seen, the switch T2 is closed in the period from t1 to t3. The secondary AC voltage Usek has a zero crossing at time t2. Since the switch T1 is closed until the time t3, the alternating voltage Usek has the value zero in the period between t2 and t3. The positive half-wave of the AC voltage Usek begins with a delay at time t3. FIG. 2 also explains the control sense of the circuit according to FIG. 1. The longer the switch T1 remains closed due to the control signals UGS of the pulse width modulator 9, the lower the output voltage Ua1 formed from Usek.

FIGS. 3 and 4 show variants of how they are used to generate more than one DC output voltage. In the circuit according to FIG. 3, an additional transformer TR2 with its primary winding is arranged between the connection point of the capacitor Cp with the rectifier diode D3 and the negative terminal. The secondary winding of this transformer TR3 is connected on the one hand via a series circuit consisting of a rectifier diode D4 and an output filter inductance LF2 and on the other hand directly to further output terminals 11a, 11b of the switching power supply. An output capacitor Ca2 is connected in parallel to these output terminals. A diode D5 is connected between the connection point of the rectifier diode D4 with the output filter inductance LF2 and the negative terminal. The DC output voltage Ua2 is present between the output terminals 11a, 11b. Depending on the transformation ratio of the transformer TR2, the DC output voltage Ua2 is greater or less than the DC output voltage Ual

Of course, in a variation of the circuit according to FIG. 3, it is possible to replace the diode D5 by the parallel connection of a switch with an inverse diode. In such a variant, the DC output voltage Ua2 can be set more precisely. This switch can also be activated by the pulse width modulator 9. be controlled. Alternatively, it is also possible to provide additional synchronization / pulse width modulator / inverter function groups for controlling this switch.

3 shows a circuit for generating two different DC output voltages. Additional additional DC output voltages can be formed by connecting additional transformers in the same way. Furthermore, it is also possible to generate negative DC output voltages by inserting the rectifier diode D4 with reversed polarity.

The circuit shown in FIG. 4 is used to generate a positive DC output voltage Ua1 and a negative DC output voltage -Ua1 of the same amplitude. The negative DC output voltage -Ua1 is present between further output terminals 12a, 12b of the switching power supply, the output terminal 12a being connected via a series connection of a rectifier diode D7 and an output filter inductance LF3 at the connection point of the rectifier diode D3 to the capacitor Cp, and the output terminals 12b directly to the negative output terminal 2b is connected. An output capacitor Ca3 is connected in parallel to the output terminals 12a, 12b. The parallel connection of a switch T3 with an inverse diode D6 is arranged between the connection point of the rectifier diode D7 with the output filter inductance LF3 and the negative output terminal 12b. The switch T3 can in turn be controlled by the pulse width modulator 9. Alternatively, it is also possible to provide additional synchronization / pulse width modulator / inverter function groups for controlling the switch T3.

Of course, with a variation of the circuit according to FIG. 4, it is possible to replace the parallel circuit T3 / D6 with a diode (see also FIG. 3) if high precision is not required when setting the output DC voltage -Ua1.

In general, the control of the negative output is only necessary if high demands are made on the output even when this output is idling. Output voltage accuracy can be set. Otherwise the tension of the negative Output automatically stabilized by the controller of the positive output. This circuit expansion is based on the fact that the voltage-time area applied to the transformer secondary winding is always equal to zero within one switching period.

If several DC output voltages are required, at least one of which is negative, circuits can be implemented which are based on combinations of the configurations shown in FIGS. 3 and 4.

In general, with regard to the circuits discussed above, which are used to generate more than one DC output voltage (positive or negative), it should be noted that no additional secondary windings or winding taps are advantageously required for implementation. This saves a complicated and expensive construction of the transformer and facilitates electrical isolation. Only simple circuit modifications are required.

Claims

claims

1. Switching power supply with a transformer (TR1), which is connected on the primary side to a circuit that generates a voltage with a zero value crossing per period and the secondary side is connected to the series circuit of a rectifier diode (D3) with an output filter inductance (LF1), between which Connection point of the rectifier diode with the output filter inductance and the further secondary-side terminal, the parallel connection of a switch (T2) with an inverse diode (D2) and an output capacitor (Ca1) between the output terminals of the switching power supply (2a), characterized in that a pulse width modulator (9) controls the switch (T2) as a function of the output DC voltage (Ua1) and synchronized to the zero-value crossings of the primary-side voltage, a synchronization (8) receiving the secondary-side AC voltage present at the transformer (TR1) and corresponding synchronization signals for it r forms the pulse width modulator and an inverter (10) receives the DC output voltage (Ua1) and feeds the pulse width modulator inversely.

2. Switched-mode power supply according to Claim 1, characterized in that the transformer (TR1) is connected on the secondary side to a further series circuit of a further rectifier diode (D7) with a further output filter inductance (LF3), with between the connection point of the further rectifier diode with the further output filter inductance and the an additional inverse diode (D2) is arranged on the secondary terminal and a further output capacitor (Ca3) is provided between the further output terminals of the switching power supply (12a, 12b) and the further rectifier diode (D7) is arranged with the opposite polarity to the rectifier diode (D3).

3. Switching power supply according to claim 1 or 2, characterized in that the transformer (TR1) is connected on the secondary side to the primary winding of a further transformer (TR2), the secondary winding of which is connected to a further series Circuit of a further rectifier diode (D4) is connected to a further output filter inductance (LF2), an inverse diode (D5) being arranged between the connection point of the further rectifier diode with the further output filter inductance and the further secondary-side terminal of the further transformer and between the further output terminals of the switching power supply ( 1 a, 11 b) a further output capacitor (Ca2) are provided.

4. Switching power supply according to one of claims 1 to 3, characterized in that the further inverse diode (D5, D6) is a further switch (T3) in parallel, which is also controlled by the pulse width modulator (9).

5. Switching power supply according to one of claims 1 to 3, characterized in that the further inverse diode (D5, D6) is a further switch (T3) in parallel, for the control of additional function groups synchronization / pulse width modulator / inverter are provided.