Circuits for transforming DC voltages using autotransformations

(19)

REPUBLIC

AUSTRIA

Patent Office (10) number: AT 409 568 B ₍₁₂₎ PATENT WRITING (21) Application number: A 1346/97 (51) Int. CI.⁷: H02M 3/155 (22) Filing date: 08/12/1997 G05F1 / 613 (42) Start of patent term: 01/15/2002 (45) Date of issue: 09/25/2002 (56)

Citations:

(73)

Patent holder:

EP568123A1 EP508595A2 EP571232A2

HIMMELSTOSS FELIX DIPL.ING. DR.TECHN. A-2351 WR. NEUDORF, LOWER AUSTRIA (AT).

(72) Inventor:

HIMMELSTOSS FELIX DIL.ING. DR.TECHN. WR. NEUDORF, LOWER AUSTRIA (AT).

(54) CIRCUITS FOR THE CONVERSION OF DC VOLTAGES WITH THE HELP OF SAVING TRANSFORMERS

QQ oo <0 in σ>

ol · <

(57) Converter circuits (e.g. Fig. 1) for converting DC voltages (U1) into DC voltages (U2) with the help of an active (S) and a passive (D) semiconductor switch, two capacitors (C1.C2), two inductors (L1 , L2), one of which (X) has a tap (b). This results in other voltage transformation relationships which, in certain applications, lead to more favorable duty cycles for the active semiconductor switch and more favorable component loads (with regard to the required current-carrying capacity or reverse voltage of the semiconductor components) and thus enable the selection of cheaper components. The active switch (S) can be expanded with relief networks or with the help of quasi-resonance structures and similar softswitching structures to reduce switching losses. The use of the converter structures in PFCs (Power Factor Corrector) is possible with an upstream rectifier on the single-phase network.

<img file="AT409568B_D0001.tif" />

DVR 0078018

AT 409 568 B

The invention relates to converter circuits (Fig. 1 -Fig.22) for converting DC voltages into DC voltages with the help of two semiconductor switches (an active such as bipolar transistor, MOSFET, IGBT, GTO, MCT, SIT and a passive, ie a diode), one or more capacitors, one or more inductors, one of which is designed as a coupled coil in the form of an autotransformer.

The idea of ​​using tapped coils in switched-mode power supplies can already be found in RDMiddlebrook: A Continuous Model for the Tapped-Inductor Boost Converter, IEEE Power Electronics Specialists Conference, 1975 Record, pp.63-79 and further applied to the three classic DC / DC converter structures in M.Rico, J.Uceda, J.Sebastian, &amp; F.AIdana: Static and Dynamic Modeling of Tapped-Inductor DC-to-DC Converters, IEEE Power Electronics Specialists Conference, 1987 Record, pp.281-288 and in J.Sebastian, J.Uceda, MAPerez, M.Rico, &amp; F.AIdana: A Very Simple Method to Obtain One Additional Fully Regulated Output in Zero-Current-Switched Quasiresonant Converters, IEEE Power Electronics Specialists Conference, 1990 Record, pp.536542. Based on this, this method was used in the implementation of a step-up converter for a strongly fluctuating input voltage and, at the same time, linked to the use of pulse widths and frequency settings for regulation. This is described in detail in the publication LLErhartt &amp; FAHimmelstoss: A Simple Boost-Converter with High Step-Up Ratio, in Proceedings of Power Convertion and Intelligent Motion PCIM'95, pp. 433-441. The analysis of this circuit showed advantages in certain operational applications due to better utilization of the semiconductor components used.

In the context of this patent, the use of autotransformers in other switched-mode power supply structures than the basic structures (step-up converter, step-down converter or inverse converter) is proposed. These are structures that were published as part of an investigation into possible higher-order topologies (FAHimmelstoss: Fourth Order DC-DC Converters with Limited Duty Cycle Range, Proceedings of the International Telecommunication Energy Conference, INTELEC'93, Vol. 1, pp .358-364 and FAHimmelstoss: Analysis and Comparison of Half-Bridge Bidirectional DC-DC Converters, IEEE Power Electronics Specialists Conference, 1994 Record, pp.922-928), as well as the structures known under the name SEPIC and ZETA converters (K.-H. Liu, &amp; FCLee: Zero-Voltage Switching Technique in DC / DC Converters, IEEE Power Electronics Specialists Conference, 1986 Record, pp.58-70). In any DC-DC converter structure that uses or can be fitted with a transformer, it can be replaced by an autotransformer. The circuits are shown in Figures Fig. 1-Fig. 22 as an example with MOSFETs. This results in other voltage transformation relationships which, in certain applications, lead to more favorable duty cycles for the semiconductor switches and to more favorable component loads (with regard to the required current-carrying capacity or Reverse voltage of the semiconductor components) and thus enable the selection of cheaper components.

Depending on the application, the input DC voltage can be supplied by a battery, solar cells, fuel cells, or by rectification from the single or multi-phase network, or by rectifying the output voltage of alternators or three-phase generators and subsequent, possibly only coarse filtering.

The input DC voltage can also only be obtained by switching a single-phase rectifier between the single-phase network and the converter structure in question. This makes it possible to use the structure obtained in this way as a PFC (Power Factor Corrector) and, with suitable control of the active switch, draw a current from the network with a dominant network fundamental.

What is particularly new about the circuits in question is the fact that they have never been published. The fact that a coil of a known DC / DC converter topology in the form of an autotransformer or two coils wound on a core, of which the end of one is then directly connected to the beginning of the second (assuming the same winding sense) through the external circuit , is realized, there are voltage transformation relationships that are not only a function of the duty cycle (and, in the case of discontinuous operation, of the load), but also of the turns ratio; This creates a further degree of freedom in the dimensioning of the circuits, which in certain applications leads to more favorable pulse duty factors for the semiconductor switches and thus to more favorable ones

AT 409 568 B

Component loads (with regard to the required current-carrying capacity or blocking capacity of the semiconductor components) and thus enables the selection of cheaper components.

A further aspect of the circuits is the possibility of expanding both the active switch and the passive switch by designing it as a current-bidirectional switch with simultaneous opposite control so that the converter is converted into a bidirectional one, thus increasing the dynamics.

In the references, only step-up converters are used; the possibility of using them in higher-order converters (in the present case, fourth-order converters are shown) is not indicated anywhere. This makes these circuits more flexible and therefore more advantageous for practical use.

EP 508 595 A2 (VTL Corporation) uses a boost converter with a tapped coil as the step-up inductance. The partial winding, which is in series with the transistor, is used in cooperation with a capacitor, which is parallel to the series circuit, consisting of this partial winding with the active switch, to relieve the active switch, which is connected to a ZVS (zero voltage switch). The aim is not to change the voltage transmission ratio. It should also be noted that with ZVSs the output voltage is not controlled via the pulse duty factor, but by changing the frequency. This is necessary because fixed switch-on times are used to ensure switching in the current zero passage. Due to the series inductance, switching on is relieved; at the same time, after commutation of the current from the diode, the oscillation between series inductance and resonance capacitor starts, forcing the current in the active switch to zero again and thus forcing the switch-off time so that it can be switched off without loss. However, this means that the degree of freedom of the tapped coil with regard to the transmission ratio cannot be used. The partial winding in series with the active switch must be dimensioned according to the oscillation duration of the resonance circuit or according to the desired amplitude of the resonance oscillation (this must be greater than the maximum load current in order to achieve ZCS).

EP 571 232 A2 (VTL Corporation) represents a further development. Here, too, the focus is on low-loss switching. A saturation choke is connected between the high-side pole of the switch and the partial winding, the switch-on losses are reduced somewhat.

EP 568 123 A1 (Philips) represents a combination of buck converter with series regulator, the series regulator being implemented by the transistor of the buck converter. Topologically there is no direct connection to the circuits shown here, since the two partial windings are not connected together at one point.

As fourth-order converter circuits for converting direct voltages (unipolar voltage) ((/ 0 into a direct voltage (U₂) (DC / DC converter) with the help of an active semiconductor switch (S), implemented with a bipolar transistor, MOSFET, IGBT, GTO, MCT, SIT (h) or similar, on whose control electrode pulses with a definable duty cycle and frequency are applied and with a passive switch (diode) (D), those are designated that have four storage elements. These are two capacitors and two inductors each. A capacitor will always be parallel to the output voltage. (Third-order converters would do without this capacity, but then cannot be used to generate a constant output voltage.) Functional pulse-width-modulated fourth-order converters serve as the starting point for generating the new topologies.

The process of obtaining principally functioning topologies with an autotransformer or with a tapped coil is as follows: If more than one other component comes together at a node with a connection of an inductance of a fourth-order converter, this node is split (not useful when the output capacitor and load are connected in parallel) and between these two new nodes the Circuit laid the new partial winding of the autotransformer; the other components are now distributed over the two sub-nodes, with at least one component always having to be connected. In the subclaims, meaningful converter structures are formed or the expansion of the circuits to bidirectionality or to soft switching is expressed. It should be noted, however, that not all possible converter structures are shown here. Furthermore, nothing speaks against the fact that both inductors are designed as autotransformers and thus further combinations can be achieved.

AT 409 568 B

The relationship between output voltage and input voltage as a function of the duty cycle d and the number of turns of the autotransformer is given as an example for the circuit in FIG

U, (W_{1 +} / V₂) - (1-c0 'the autotransformer has the total number of turns N1 + N2, the sub-coils the number of turns N1 or N2.

It should also be mentioned that the active switch can be expanded by means of relief networks (for example as in FIG. 23) or with the aid of quasi-resonance structures and similar soft-switching structures in order to reduce the switching losses. An overview with extensive references can be found in the article Soft-Switching Techniques in PWM Converters, G.Hua &amp; FCLee, IEEE Transactions on Industrial Electronics, Vol 42, Dec. 1995, 595-603.

Figures 1 to 22 show embodiments of the invention. Figure 23 shows an example of a switching relief of the active switch.

Figure 1 shows a converter circuit in which the input voltage (U1) is connected to the series circuit of inductance (L1) and active switch (S), parallel to the active switch (S) is the series circuit of a capacitor (C1) with the coil with tap ( X) is connected and the anode of the diode (D) is connected to the tap and the cathode to a capacitor (C2), the second terminal of which is connected to the negative pole of the input voltage, whereby the output voltage (U2) is tapped at the capacitor.

Figure 2 shows a converter circuit in which the input voltage (U1) is connected to the series circuit of the active switch (S), the second partial winding of the coil (L12) with tap (X), and that parallel to the coil with tap is a series circuit consisting of a Capacitor (C1) and a diode (D), the anode of which is connected to the negative pole of the input voltage, is connected, and that a low-pass filter in parallel with the diode (D) formed by the series connection of an inductance (L2) and a capacitor (C2), the output voltage (U2) being tapped at the capacitor (C2).

FIG. 7 shows a converter circuit in which the input voltage (U1) is applied to the series circuit of the passive switch (D), capacitor (C1) and the coil with tap (X), between the cathode of the diode (D) and the output terminal (3 ) an inductance (L1) is switched, the active switch is between the input terminal (1) and the tap of the coil with tap (X), the reference point for input (2) and output (4) is the same and a capacitor (C2) is connected in parallel to output (3, 4) with the voltage U2.

FIG. 9 shows a converter circuit in which the input voltage (U1) is connected to the series connection of inductance (L1), capacitor (C1) and the coil with tap (X), the diode (D) between the tap of the coil with tap (X) ) and the output terminal (3), the active switch is connected between the connection point of inductance (L1) and capacitance (C1) and output terminal (3), the reference point for input (2) and output (4) is the same and a capacitor (C2) is connected in parallel to output (3, 4) with the voltage U2.

FIG. 11 shows a converter circuit in which the (negative) input voltage (U1) is connected to the series circuit of inductance (L1), capacitor (C1) and the coil with tap (X), the active switch (S) between the tap of the coil is connected to the tap (X) and the input terminal (1), furthermore the diode (D) is connected between the connection point of inductance (L1) and capacitor (C1) and output terminal (3), the input terminal (2) and the output terminal (4) are directly connected, a capacitor (C2) is connected in parallel to the output terminals (3, 4), from which the output voltage (U2) is tapped.

Figure 13 shows a converter circuit in which the input voltage (U1) is on the series circuit of inductance (L1), capacitor (C1) and the coil with tap (X), the active switch (S) between the input terminal (1) and the Tapping of the coil is connected to tapping (X), furthermore the diode (D) is connected between output terminal (3) and the connection point of inductance (L1) and capacitor (C1), the input terminal (2) and the output terminal (4) are connected directly are connected, A capacitor (C2), from which the output voltage (U2) is tapped, is connected in parallel to the output terminals (3, 4).

AT 409 568 B

Figure 15 shows a converter circuit in which the output voltage (U2) is connected to the series connection of inductance (L1), capacitor (C1) and the coil with tap (X), the passive switch (D) between the tap of the coil with tap ( X) and the output terminal (3) is switched, furthermore the active switch (S) is switched between the connection point of inductance (L1) and capacitor (C1) and input terminal (1), the input terminal (2) and the output terminal (4) are directly connected, a capacitor (C2) is connected in parallel to the output terminals (3, 4), from which the output voltage (U2) is tapped.

FIG. 17 shows a converter circuit in which the input voltage (U1) is applied to the series circuit of passive switch (D), capacitor (C1) and the coil with tap (X), between the cathode of the diode (D) and the output terminal (3 ) an inductance (L1) is connected, the active switch is between the tap of the coil with tap (X) and the output terminal (3), the reference point for input and output is the same and capacitor C2 is connected in parallel to the output with voltage U2.

Figure 19 shows a converter circuit in which the input voltage (U1) is connected to the series connection of the active switch (S), capacitor (C1) and the coil with tap (X), an inductance between the active switch (S) and output terminal (3) (L1) is switched, the passive switch is between the output terminal (3) and the tap, the reference point for input and output is the same and the capacitor C2 is connected in parallel to the output with the voltage U2.

FIG. 21 shows a converter circuit in which the input voltage (U1) is connected to the series connection of inductance (L1), capacitor (C1) and the coil with tap (X), the diode (D) between the tap of the coil with tap (X) ) and the input terminal (1), the active switch is connected between the connection point of inductance (L1) and capacitance (C1) and output terminal (3), the reference point for input and output is the same and capacitor C2 is connected in parallel to the output with voltage U2.

In Figs. 5, 6, 8, 10, 12, 14, 16, 18, 20, 22 are converter circuits in which the connections to the coil with tap (X) (a) and (b) are interchanged.

In FIG. 3 the method is shown on a normal second-order buck converter and in FIG. 4 the connections to the coil with tap (X) (a) and (b) are interchanged.

In Fig. 23 a simple relief network for reducing the switching losses on the active switch is shown.

List of reference symbols

U1 input voltage

U2 output voltage

L inductance, coil

L1 inductance, coil

L2 inductance, coil

X Coil with tap a Terminal of the coil with tap X b Terminal of the coil with tap X c Terminal of the coil with tap X

L21 Partial winding of the coil with tap X

L22 Partial winding of the coil with tap X

L11 Partial winding of the coil with tap X

L12 Partial winding of the coil with tap X

N1 number of turns of a partial winding of the coil with tap X N2 number of turns of a partial winding of the coil with tap XC capacitor

C1 capacitor

C2 capacitor

S active switch

D passive switch (diode)

RL load resistance

Terminal for input voltage

AT 409 568 B

Terminal for input voltage

Terminal for output voltage

Terminal for output voltage

Rs resistance in the relief network

Ls inductance in the relief network Cs capacitor in the relief network