DE102020214265A1 - Device for low-common-mode conversion of alternating voltage - Google Patents

Abstract

The invention relates to a device for low-common-mode conversion of AC voltage comprising: • a voltage supply (E1 ... E4), the voltage supply during operation via a 1-phase 2-wire AC voltage network, a 1-phase 3-wire AC voltage network or a 3-phase multi-wire AC voltage network,• three first switching branches, with each first switching branch having a low-frequency clockable bidirectional switch (Sa, Sb, Sc),• whereby the outputs of the first switching branches can each be switched to a first conductor (m). ,• wherein each of the inputs of the voltage supply (E1 ... E4) is conducted by means of a first rectifier device (D1a, D2a, D3a) in a positive semi-oscillation to a second conductor (p) during operation,• wherein each of the inputs of the voltage supply ( E1 ... E4) by means of a second rectifier device (D1b, D2b, D3b) in a negative semi-oscillation to a third conductor(s) during operation, • wherein the device further has at least two second switching branches, wherein each second switching branch has a switch (S1, S2) that can be clocked at high frequency,• wherein a first of the second switching branches (S1) is connected between the first conductor (m) and the second conductor (p) is arranged, wherein a second of the second switching branches (S2) is arranged between the first conductor (m) and the third conductor (n),• wherein in the supply of the second conductor (p) to the first of the second switching branches (S1) exactly a first effective step-up choke (Lp) is arranged,• wherein exactly one second effective step-up choke (Ln) is arranged in the supply of the third conductor (n) to the second of the second switching branches (S2).

Description

The invention relates to a device for low-common-mode conversion of AC voltage.

background

DC voltages are required in many areas of technology. On the other hand, the transport of electrical energy is mainly realized with AC voltage networks.

However, a simple, passive diode rectification for converting the AC voltage into a DC voltage leads to significant repercussions on the supply network with sometimes considerable reactive power inputs and, in the worst case, can endanger the stability of the network.

For this reason, three-phase, grid-friendly pulse rectifiers (PFC: Power Factor Correction) were developed in the past, which are already used in a large number of applications, e.g. in the industrial and IT sectors (supplying server racks in data centers, of telecom equipment for mobile networks, etc.) as interface circuits to the AC power supply.

These known pulse rectifiers emulate the (almost) ideal behavior of an ohmic resistance load on the supply network side (input side/AC side). As a result, only minor mains reactions, i.e. minor current distortions and minor reactive currents, are caused. This corresponds to a high power factor.

On the output side, i.e. on the direct current side, these circuits supply a mostly regulated direct current (DC), which either supplies a direct current load directly or, as an intermediate circuit, feeds further direct current converter stages (DC-DC) or inverter stages.

Three-phase pulse rectifiers are typically used in the high power range of around 5...100 kW and beyond.

In addition to the applications mentioned above, they are currently also being used more and more frequently for high-performance battery charging of electric vehicles (xEV), both in terms of infrastructure in the area of fast charging stations (off-board) and on the vehicle side (on-board) for charging from the general alternating voltage network (AC). .

It is precisely these on-board chargers (OBL) that are currently pushing into the mass market in the 11 kW and 22 kW power classes, with material costs, size and weight, as well as efficiency and operating costs being the decisive criteria.

A characteristic of most previous three-phase, mains-friendly pulse rectifier circuits is the switching-frequency common-mode interference (common mode, abbreviation CM) generated as the dominant interference component of electromagnetic compatibility (abbreviation EMC).

Until now, the effects of common-mode interference could only be contained by using a bulky EMC filter in the pulse rectifier circuit and, as a rule, by transformer-based potential isolation in the DC-DC stages.

In addition to the volume and weight problem, which plays a major role in the field of vehicle technology in particular, the efficiency is also of great importance.

It is also advantageous to obtain a low-common-mode output voltage. As a rule, this can only be achieved with correspondingly large, often multi-stage filter circuits consisting of passive elements such as current-compensated chokes and capacitors, each with a considerable size.

A similar problem occurs with the circuits known as Minnesota rectifiers, e.g. from the article "A novel approach to minimize line-current harmonics in interfacing power electronics equipment with 3-phase utility systems" by the author N. Mohan, published in IEEE Trans. Power Delivery, Vol. 8, No. 3, pp. 1395-1401, Jul. 1993. These are comparatively complex and require large, passive filter components (L, C) as a so-called "current injection network". In addition, these circuits sometimes have significant residual line current distortion (THD).

Furthermore, the so-called Vienna rectifier is also known from the prior art, e.g. from the article "A novel three-phase three-switch three-level pwm rectifier" by the authors J. W. Kolar and F. C. Zach, published in Proc. 28th Power Conversion Conference, PCIM '94, Jun. 28-30, 1994, pp. 125-138. However, this is also complex and requires three high-frequency clocked semiconductor switches and three step-up chokes on the AC voltage side and, moreover, cannot be described as inherently low in common mode. In particular, this has switching-frequency (or: high-frequency) common-mode interference voltages at its output, which can only be sufficiently suppressed by input-side filter circuits made of large-volume passive components (chokes, capacitors).

The so-called Taipei Rectifier, see article "The TAIPEI Rectifier - A New Three-Phase Two-Switch ZVS PFC DCM Boost Rectifier" by the authors Y. Jang and MM Jovanovic, published in IEEE Trans. on Power Electronics, Vol. 28, No . 2, 2013, on the other hand, requires a total of five power chokes on the AC and DC side and, due to the principle, can only work in discontinuous step-up mode (DCM boost), which limits it to smaller transmittable powers.

The topology known as the Swiss Rectifier, which can be found, for example, in the article "SWISS rectifier - A Novel Three-Phase Buck-Type PFC Rectifier Topology for Electric Vehicle Battery Charging" by the authors T. B. Soeiro, T. Friedli, and J. W. Kolar, published in Proc . 27th Applied Power Electronics Conference and Exposition (APEC), 2012, has significant switching-frequency common-mode interference voltages (CM) and, on the other hand, the Swiss rectifier works in buck mode (buck). Both points lead to a significantly increased EMC filter requirement (CM component and DM component) on the mains side.

Proceeding from this, it is the object of the invention to specify a device for converting AC voltage with low common mode, which avoids one or more problems from the prior art.

The object is achieved by a device for low-common-mode conversion of AC voltage. The device has a voltage supply, with the voltage being supplied during operation via a 1-phase, 2-wire AC voltage network, a 1-phase, 3-wire AC voltage network or a 3-phase multi-wire AC voltage network. Furthermore, the device has three first switching branches, with each first switching branch having a bidirectional switch that can be clocked at a low frequency, with the outputs of the first switching branches each being able to be routed in a switchable manner to a first conductor, with each of the inputs of the voltage supply using a first rectifier device in a positive half-oscillation is routed to a second conductor during operation, each of the inputs of the voltage supply being routed to a third conductor during operation in a negative semi-oscillation by means of a second rectifier device. The device also has at least two second switching branches, with each second switching branch having a switch that can be clocked at high frequency, with a first of the second switching branches being arranged between the first conductor and the second conductor, with a second of the second switching branches being arranged between the first conductor and the third Head is arranged. Exactly one first effective step-up inductor is arranged in the feed line of the second conductor to the first of the second switching branches. Exactly one second effective step-up inductor is also arranged in the supply of the third conductor to the second of the second switching branches.

This means that a cost-effective, lightweight, energy-efficient and low-common-mode converter can be made available with minimal effort, namely only two high-frequency chokes.

Further advantageous configurations of the invention are the subject matter of the dependent claims, the description and the figures.

• 1 eine schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 2 eine weitere schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 3 eine schematische Darstellung eines Aspektes der Erfindung,
• 4a-4c unterschiedliche Netzanschlüsse an Ausführungsformen der Erfindung,
• 5a und 5b eine Auswahl von Schalterimplementierungen gemäß Ausführungsformen der Erfindung,
• 6 eine weitere schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 7 eine weitere schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 8 eine weitere schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 9 eine weitere schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 10 eine weitere schematische Darstellung von Elementen gemäß Ausführungsformen der Erfindung,
• 11 ein beispielhafter Verlauf von Eingangsspannungen und das Schaltverhalten gemäß Ausführungsformen der Erfindung, und
• 12 einen weiteren Aspekt gemäß Ausführungsformen der Erfindung.

Embodiments of the present invention are described by way of example with reference to the attached drawings, in which:

• 1 a schematic representation of elements according to embodiments of the invention,
• 2 a further schematic representation of elements according to embodiments of the invention,
• 3 a schematic representation of an aspect of the invention,
• 4a-4c different power connections on embodiments of the invention,
• 5a and 5b a selection of switch implementations according to embodiments of the invention,
• 6 a further schematic representation of elements according to embodiments of the invention,
• 7 a further schematic representation of elements according to embodiments of the invention,
• 8th a further schematic representation of elements according to embodiments of the invention,
• 9 a further schematic representation of elements according to embodiments of the invention,
• 10 a further schematic representation of elements according to embodiments of the invention,
• 11 an exemplary course of input voltages and the switching behavior according to embodiments of the invention, and
• 12 a further aspect according to embodiments of the invention.

Detailed presentation of the invention with reference to the drawings

In the following the invention will be illustrated in more detail with reference to the figures. It should be noted that different aspects are described, which can be used individually or in combination. That is, each aspect can be used with different embodiments of the invention unless explicitly presented as purely alternative.

Furthermore, for the sake of simplicity, only one entity will generally be referred to below. Unless explicitly noted, however, the invention can also have several of the entities concerned. In this respect, the use of the words “a”, “an” and “an” is only to be understood as an indication that at least one entity is used in a simple embodiment.

Insofar as methods are described below, the individual steps of a method can be arranged and/or combined in any order, unless something different is explicitly stated in the context. Furthermore, the processes can be combined with one another, unless expressly stated otherwise.

Specifications with numerical values are generally not to be understood as exact values, but also include a tolerance of +/- 1% to +/- 10%.

References to standards or specifications or norms are to be understood as a reference to standards or specifications or norms which are/were applicable at the time of filing and/or - insofar as priority is claimed - also at the time of priority filing. However, this is not to be understood as a general exclusion of applicability to subsequent or superseding standards or specifications or norms.

In 1 Schematically elements of the invention are shown according to different embodiments.

Very generally, a device according to the invention for converting AC voltage with low common mode has a voltage supply. Depending on the type of network used and the task, a different number of inputs can be provided.

For example, the provision of 3 inputs E1 ... E3 may be sufficient for a (European) three-phase network. These can, for example, as in 4a respectively. 4b shown, be guided to corresponding voltage leads E1 ... E3. In addition and optionally, a neutral conductor N can also be routed to another "voltage supply" E4, as in 4a shown. For an (American) biphase network, on the other hand, as in 4c shown, one phase can be applied as a voltage feed to an input - here E1 - and another phase (the opposite phase) can be applied to another voltage feed on another input - here E3 - while the one "voltage feed" - here E2 - is connected to the neutral conductor is connected.

A special switching position can be provided for certain network types, which of course can also be designed differently in hard-wired applications (for only one network type).

In embodiments of the invention, it can also be provided that there is only one voltage supply and one neutral conductor. This case can also occur, for example, if not all phases are present in a three-phase system.

That means, in general, the voltage can be supplied during operation via a 1-phase, 2-wire AC voltage network, a 1-phase, 3-wire AC voltage network or a 3-phase multi-wire AC voltage network.

The device also has three first switching branches, each first switching branch having a bidirectional switch Sa, Sb, Sc that can be clocked at a low frequency.

The precise design of the bidirectional switches Sa, Sb, Sc that can be clocked at low frequencies is irrelevant for understanding. In particular, however, the in 5a symbolically shown transistor switches or technologies are used. The three bidirectional power switches are only clocked at a low frequency, ie typically at 3 times the frequency of the AC voltage present in one phase, and therefore cause practically no switching losses. In particular, the bidirectional switches Sa, Sb, Sc of the first switching branches can be selected from a group comprising Si-based IGBT or GaN (gallium nitride)-based transistors, in particular monolithic bidirectional GaN transistors. A bidirectional switch Sa, Sb, Sc that can be clocked at a low frequency can be configured as in 5a shown, be constructed of several individual elements (eg IGBT and diodes) or monolithic.

The outputs of the first switching branches can each be switched to a first conductor m - dar put in the 1 , 2 , 6-10 - to be guided.

In the device according to the invention, each of the inputs of the voltage supply E1 Rectifier device D1b, D2b, D3b is performed in a negative semi-oscillation on a third conductor n in operation.

The device also has at least two second switching branches, each of the second switching branches having a high-frequency switch S1, S2, wherein a first of the second switching branches - e.g. S1 - is arranged between the first conductor m and the second conductor p, and wherein a second of the second switching branches - e.g. S2 - is arranged between the first conductor m and the third conductor n.

The precise design of the switches S1, S2 that can be clocked at high frequency is irrelevant for understanding. In particular, however, the in 5b symbolically shown transistor switches or technologies are used.

Exactly one first effective step-up inductor Lp is arranged in the feed line of the second conductor p to the first of the second switching branches S1. Exactly one second effective step-up inductor Ln is also arranged in the supply of the third conductor n to the second of the second switching branches S2.

An effective step-up choke can be constructed from a number of individual step-up chokes which are connected in parallel or in series in order to have specific electrical properties.

In the device according to the invention, the bidirectional switches Sa, Sb, Sc of the first switching branches can be controlled during operation in such a way that the smallest input phase voltage of the voltage supply can be switched to the first conductor m.

This results in a low-common-mode output voltage with the partial voltages uZK1 and uZK2, which does not contain any switching-frequency components and thus enables a simplified and smaller EMC filter. On the other hand, the downstream DC-DC stages in the application for onboard chargers can be constructed without a transformer (non-floating), which means that losses during operation and material costs as well as weight and size can be saved during construction. That is, improved energy efficiency and cost efficiency can be provided.

But also independent of the exemplary OBL application, the proposed three-phase PFC topology has some other advantageous properties for other areas of application. The devices according to the invention manage with only two high-frequency clocked circuit breakers S1, S2. These can be clocked with comparatively low clock rates of 100 kHz. That is, in the device according to the invention, no effective step-up choke is arranged in the first conductor m.

The devices according to the invention also manage with only two step-up chokes Lp, Ln, while most conventional circuits each require three (i.e. one high-frequency switch and one choke per mains phase).

The devices according to the invention also work in boost operation, which leads to continuous, non-pulsating input currents and thus also minimizes the push-pull requirements (differential mode, abbreviation DM) on the line-side EMC filter.

In general, the push-pull component of the EMC filter can advantageously also be shifted to the DC side of the circuit. For example, in 1 , - which differs, for example, from the filter function of the 2 differs - the two filter capacities CF1, CF2 fulfill this DM filter function. If necessary, filter chokes LF1/2, drawn in the series branches p and n, on the network side of CF1, CF2 can be used to increase the filter effect. In contrast, with conventional circuits on the AC side, three filter elements are usually required for DM suppression.

For this purpose, it can be provided in embodiments of the invention that a first inductance L1 is arranged downstream of the first rectifier device D1a, D2a, D3a in the second conductor p and a second inductance L2 is arranged downstream of the second rectifier device D1b, D2b, D3b in the third conductor n, a first filter capacitance CF1 being arranged downstream of the first inductance L1 towards the first conductor m and a second filter capacitance CF2 downstream of the second inductance L2 towards the first conductor m.

In embodiments of the invention, a (further) EMC filter arranged on the input side can also be present. However, this can be dimensioned significantly smaller compared to previous circuit arrangements. In particular, the EMC filter arranged on the input side can be dimensioned in such a way that it Essentially for common mode filtering (engl. Common Mode) is set up.

Furthermore, the proposed circuit can also work in discontinuous (DCM) or discontinuous limit operation (BM or CrCM), but discontinuous operation (CCM) is to be seen as the primarily advantageous mode of operation, which requires high transmission power (>10 kW ) with low conduction losses. However, if DCM or BM operation is used, the increased conduction losses in the higher power range can, as usual, be reduced by parallel connection (and phase-shifted control "interleaving") of several strands.

It should also be noted that the transformer symbols shown in the figures in the background of the DC-DC stages are to be regarded as optional. When using a conventional onboard charger, for example, transformers can be used. However, the device according to the invention can supply both DC-DC stages with a transformer and without a transformer, since the device has a low level of common mode. For this reason, in the 1 and 2 the DC-DE stages are represented by a dashed outline. Likewise, the representation of the high-voltage batteries and the transformers in the figures are only optional unless explicitly described as components of an embodiment.

Since the proposed circuit consists of only two strands in the high-frequency clocked part, the overall effort with power-scaled parallel circuitry is lower than with most conventional topologies (2N vs. 3N total strands, N: number of parallel strands).

Such a strand arrangement is in 3 shown. This means that the performance can be scaled accordingly by connecting such strands in parallel.

Advantageously, in the device according to the invention, the network phase with the minimum absolute value is switched through to the center point m, which also defines the center potential of the two intermediate circuit partial voltages uZK1, uZK2. A bidirectional semiconductor switch Sa, Sb, Sc is provided for each network phase for this connection.

In the devices according to the invention, the bidirectional switches Sa, Sb, Sc of the first switching branches can be controlled or switched during operation with up to three times the respective input network frequency, in particular twice the respective input network frequency.

This means that there are practically no relevant switching losses in these circuits Sa, Sb, Sc, so that they have little effect on the overall efficiency of the circuit - especially since it is possible to use power semiconductor types that are optimized for low on-state losses. However, this results in a high overall efficiency of the devices according to the invention.

The through-connection of the network phase with the minimum amount at point m, in addition to the EMC behavior, also benefits the realizable current flow in this phase, which, due to the principle, always has to go through the zero crossings. With some conventional three-phase PFC topologies (e.g. Vienna Rectifier), unwanted gap operation occurs in the area of the current zero crossings and thus current distortions (deviations from the sinusoidal shape). Since in the proposed device the bidirectional switch practically always connects the network phase with the minimum absolute value directly to the midpoint m, there is also no temporary discontinuous operation with associated current distortions.

The operating principle of the proposed PFC circuit provides independent control of the two currents ip and in in the step-up chokes Lp, Ln, the differential current (ip - in = im) is routed via the relevant bidirectional switch to or from the network phase with the minimum absolute value. With a corresponding setpoint specification for ip and in, a sinusoidal curve results for all three mains currents ia, ib, ic. ip corresponds to the current in the respective maximum mains phase (positive polarity), in corresponds analogously to the current in the respective minimum mains phase (negative polarity).

In 11 shows the typical course in relation to a three-phase European AC voltage network.

6 shows an advantageous implementation of the low-frequency clockable bidirectional switches Sa, Sb, Sc with IGBTs and silicon (Si) diodes. For example, each bidirectional switch Sa, Sb, Sc that can be clocked at a low frequency can have an IGBT and four diodes. This refinement with only one mit, that is to say only one standard power semiconductor (IGBT) that can be turned off, is particularly cost-effective and requires only little control effort.

However, there are still alternative implementations, which then usually have two IGBTs (eg also reverse-blocking RB-IGBTs arranged antiparallel) per bidirectional switch S1SA, S2SB, S3SC that can be clocked at a low frequency and therefore have fewer diodes. Such configurations allow the To further reduce conduction losses per bidirectional switch Sa, Sb, Sc that can be clocked at a low frequency.

IGBTs have comparatively low conduction losses, but higher switching losses (than, for example, MOSFETs).

Due to the only low-frequency switching of the bidirectional switches Sa, Sb, Sc, which can be clocked at low frequencies, the switching losses have practically no effect, while benefit can be gained from the low conduction losses. Furthermore, IGBTs are very robust semiconductor switches and are available in many versions at low cost. They are also ideally suited for production-related integration into a common power module (together with the surrounding diodes, which results in one power module per bridge branch, for example), which is an important argument for automotive (onboard charger) and industrial applications.

In addition, IGBTs are available in the various relevant voltage classes. This is worth mentioning, since the voltage load of the bidirectional switches, as well as the two high-frequency controlled switches S1, S2 in the proposed PFC topology, when operated on a typical European three-phase network (400 V AC), assumes an intermediate value in relation to standard voltage classes. In this application, types with a blocking voltage of 1000 V would be preferable, which is covered by IGBTs.

A further advantageous implementation of the bidirectional semiconductor switch can be seen in the use of monolithic, bidirectional GaN transistors. Such transistors based on GaN material can implement a low-frequency clockable bidirectional switch Sa, Sb, Sc directly as a single element. In addition to a further reduction in circuit complexity, this results in further reduced conduction losses and thus a further increase in overall efficiency.

Monolithic, bidirectional GaN transistors also have a comparatively weak switching behavior, which would not have an effect in the proposed application due to the low-frequency switching. Instead, types optimized for minimal conduction losses could be fully utilized.

As already mentioned at the beginning, any network / network type can be used with suitable wiring. In the following, the use with a single-phase network or the fallback solution of a single-phase operation will be briefly discussed.

The neutral conductor N can be connected by means of a further diode branch D4a, D4b. This is particularly advantageous when the invention is used as an onboard charger (abbreviated OBL), because in this way charging, which is important in an emergency, can also be ensured at a single-phase household socket. In this case, one of the mains phases a,b,c can work with N as a return conductor.

The bidirectional switch Sd drawn in this branch is only to be regarded as optional. If Sd were omitted or permanently switched off, both intermediate circuit capacitances CZK1, CZK2 would be charged simultaneously at all times, even in single-phase operation, which advantageously leads to a lower total capacitance requirement and smaller sizes for the intermediate circuit capacitors.

If Sd is used and switched on in single-phase operation, the operating behavior results in which the two intermediate circuit capacitances CZK1, CZK2 are charged one after the other in the mains half-period rhythm (at 50 Hz corresponding to 10 ms), which requires higher capacitance values with the same charging power.

In addition, it can be provided that a (small) capacitance (eg 3 μF) is arranged in the first conductor m. This also enables operation with a lower DC output voltage. This is an example in 12 shown. This capacity makes it possible to achieve a desired output voltage on the DC side of 2 * 400 V even when connected to a conventional (European) three-phase network with 400 V AC.

However, the device of the invention can also be operated in a split-phase network (see 4c ), as found in North America, for example. In the proposed device, this can advantageously be implemented directly by laying the neutral conductor N via the supply line to a mains phase connection terminal on the OBL side (here, for example, b/E2) and permanently connecting it directly to the circuit center point m via the existing bidirectional switch (here, Sb). can.

The center point m is therefore on the extremely quiet neutral conductor potential (0 V), which minimizes any low-frequency leakage currents from a high-voltage (abbreviated HV) battery side.

Due to the special network configuration (uL1,US and uL2,US phase-shifted by 180°), both intermediate circuit capacitances CZK1, CZK2 can also be charged simultaneously in this case, so that no increased capacity requirement arises. In particular, when operating on the North American split-phase grid, it can be provided that an HV battery (traction battery) with a charging voltage range of about 400-800 V can also be charged by the proposed device alone, ie without downstream DC-DC stages for reducing the intermediate circuit voltage uZK1/2, can be loaded.

If a uniform OBL device is to be retained for the European and North American grids, the existing non-floating DC-DC stages could be deactivated during operation on the split-phase grid, i.e. the associated step-down switches could be switched through permanently. This avoids switching losses there and thus further increases the overall efficiency.

In general, it should be noted that the proposed three-phase PFC topology, regardless of the application, is also of particular interest for use in North America, since it is universal here for the 120/240V split-phase network described, as well as for the existing 208V AC and 240V AC three-phase networks (with grounded star point) is to be used - with power semiconductors that can all be in the blocking voltage class of 600 V.

This means that in particular the high-frequency clocked switches S1, S2 could advantageously be implemented using efficient standard MOSFETs in Si technology or using highly efficient GaN transistors that are already available.

In addition to the implementations of the proposed PFC circuit shown and explained in more detail above, there are also other variants worth mentioning that are based on the basic principle described.

For example, in 7 a buck/boost variant (buck-boost) is shown using the example of a European three-phase network, which can also implement low output voltages as a single-stage circuit (ie with only two chokes and then four switches operated at high frequency).

Furthermore, in 8th an isolated single-stage variant based on flyback converters e.g. for single-stage OBC is shown. Output capacitances can be connected in parallel or in series (not shown). In 8th an isolated single-stage variant based on a boost full bridge, eg single-stage OBC, is shown. In 10 an isolated single-stage variant based on a phase controlled full bridge (PSFB), eg for single-stage OBC, is shown.

Without restricting the generality of the invention, variants with a bidirectional power flow can also be made possible, both for the non-floating buck/boost version (buck-boost) and for the non-floating versions.

To enable the bidirectional power flow, the rectifier diodes of the proposed PFC basic topology must each be provided with an anti-parallel power transistor (e.g. IGBT) with low-frequency control, regardless of the variant. In practice, for example, the future SmartGrid integration (vehicle-to-grid, vehicle-to-load) for OBL applications (power flow: charging station ↔ EV) could be implemented with little effort.

That is, by means of the invention, a particularly simple use of a device as an onboard charger for a vehicle is made available. In doing so, e.g. by using suitable mains cables, the connections to a suitable network (corresponding to 4a-4c ) will be realized.

Claims (12)

Device for converting AC voltage with low common mode, comprising: • a voltage supply (E1 ... E4), the voltage supply during operation being via a 1-phase 2-wire AC voltage network, a 1-phase 3-wire AC voltage network or a 3-phase multi-wire AC voltage network, • three first switching branches, each first switching branch having a low-frequency clockable bidirectional switch (Sa, Sb, Sc), • the outputs of the first switching branches can each be routed to a first conductor (m) in a switchable manner, • each of the inputs of the voltage supply (E1 ... E4) by means of a first rectifier device (D1a, D2a, D3a) in a positive half-oscillation to a second conductor (p) during operation, • each of the inputs of the voltage supply (E1 ... E4) by means of a second rectifier device (D1b, D2b, D3b) in a negative semi-oscillation on a third conductor (n) is performed during operation, • the devices g furthermore has at least two second switching branches, wherein each second switching branch has a switch (S1, S2) that can be clocked at high frequency, • wherein a first of the second switching branches (S1) is arranged between the first conductor (m) and the second conductor (p), wherein a second of the second switching branches (S2) is arranged between the first conductor (m) and the third conductor (n), • wherein exactly one first effective step-up choke (Lp) is arranged in the supply of the second conductor (p) to the first of the second switching branches (S1), • wherein in the supply of the third conductor (n) to the second of the second switching branches (S2 ) Exactly one second effective step-up inductor (Ln) is arranged. device after claim 1 , characterized in that no effective step-up choke is arranged in the first conductor. device after claim 1 or 2 , characterized in that the bidirectional switches (Sa, Sb, Sc) of the first switching branches are selected from a group comprising Si-based IGBT or GaN-based transistors, in particular monolithic bidirectional GaN transistors. Device according to one of the preceding claims, characterized in that the bidirectional switches (Sa, Sb, Sc) of the first switching branches can be controlled during operation in such a way that the smallest input phase voltage of the voltage supply can be switched to the first conductor (m) in terms of absolute value. Device according to one of the preceding claims, characterized in that the bidirectional switches (Sa, Sb, Sc) of the first switching branches can be controlled during operation with up to three times the respective input mains frequency. Device according to one of the preceding claims, characterized in that following the first rectifier device (D1a, D2a, D3a) in the second conductor (p) a first inductance (L1) and following the second rectifier device (D1b, D2b, D3b) in the third Conductor (n) a second inductor (L2) is arranged, following the first inductor (L1) to the first conductor (m) towards a first filter capacitance (CF1) and subsequently to the second inductor (L2) to the first conductor (m) towards a second filter capacitance (CF2) is arranged. Device according to one of the preceding claims, characterized in that the device also has an EMC filter on the input side. device after claim 7 , characterized in that the input-side EMC filter is set up essentially only for common-mode filtering. Device according to one of the preceding claims, characterized in that a capacitance is arranged in the first conductor (m). Use of a device according to one of the preceding claims as an onboard charger for a vehicle. Use of a device according to any one of the preceding Claims 1 until 9 with a 3-phase AC voltage as power supply. Use of a device according to any one of the preceding Claims 1 until 9 with a 1-phase AC voltage as power supply.

Images