DC / DC converter for converting one input voltage into at least two desired output voltages
Patent Information
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- ELMOS SEMICON AG
- Filing Date
- 2024-03-13
- Publication Date
- 2026-05-13
AI Technical Summary
Existing DC-DC converters require multiple inductive energy storage devices to generate multiple output voltages, increasing hardware complexity, especially when the output voltage priorities are not predetermined.
A DC-DC converter design utilizing a single inductive energy storage device, with bidirectional high-side switching units and a control unit to manage energy distribution to multiple outputs based on decision criteria, preventing cross-currents and optimizing power loss.
Enables generation of multiple output voltages with reduced hardware complexity by using a single inductor, minimizing power loss, and efficiently managing energy distribution to meet varying output demands.
Description
[0001] The invention relates to a DC-DC converter for converting an input voltage into at least two target output voltages, which can assume different or the same values within a predefinable output voltage range. It is not specified at which of the output terminals of the DC-DC converter the higher target output voltage, and thus also the higher actual output voltage, is applied.
[0002] DC-DC converters in the form of boost converters or buck-boost converters are generally known. In both cases, an inductive energy storage device is charged and discharged in a clocked manner.
[0003] If you want to generate several different output voltages from a single input voltage using a DC-DC converter (SIMO - Single Input Multiple Output), you typically need multiple inductive energy storage devices. This increases the hardware complexity. This is especially true if, depending on the application, you don't want to be able to specify in advance which output should have the higher output voltage.
[0004] From JIN WEIJAN Et AL: "Single-Inductor Multiple-Output Inverter With Precise and Independent Output Voltage Regulation", IEEE TRANSACTIONS ON POWER ELECTRONICS, INSTITUTE OF ELECTRONICAL AND ELECTRONICS ENGINEERS, USA, Vol. 35, No. 10, March 10, 2020 (2020-03-10), pages 11222-11234, XP011796582, ISSN: 0885-8992, DOI: 10.1109 / TPEL.2020.2979893 [accessed 2020-06-29], a DC-DC converter with a single inductor (energy storage device) is known, which can supply energy to two outputs of the DC-DC converter.
[0005] The object of the invention is to provide a DC-DC converter for converting an input voltage into at least two desired output voltages, which has advantages over the prior art, particularly with regard to the provision of the inductive energy storage.
[0006] To solve this problem, a first embodiment of the invention proposes a DC-DC converter (in the form of a buck-boost converter) for converting an input voltage into at least two desired output voltages, which can be different or the same within a predefinable output voltage range, wherein the DC-DC converter is provided with an input half-bridge comprising an input voltage terminal for applying the input voltage, a reference potential terminal, and a series connection of an electronic high-side switch and an electronic low-side switch with a connection node between the two switches, wherein the low-side switch of the input half-bridge has a diode connected in parallel to the switch with a cathode connected to the reference potential terminal; an output half-bridge comprising at least a first and a second output terminal, from which at least a first and a second output voltage can be tapped, a reference potential terminal, an electronic low-side switch, and at least two electronic high-side switch units, each of which is connected between one of the output terminals and a common connection node to which the low-side switch is further connected; an inductor;which is arranged between the connection nodes of the input and output half-bridges and which can be charged with electrical energy during a charging phase of one of several successive clocked control cycles and discharged during a discharging phase of a control cycle to forward electrical energy to one of the output terminals, wherein each high-side switching unit has a first electronic switch and a second electronic switch, each with a diode connected in parallel to the switch, wherein the two switches of each high-side switching unit are connected in series with their diodes in anti-series orientation, and the switch of each high-side switching unit whose diode is reverse-biased with respect to a current flow from the connection node to the output terminal connected to the high-side switching unit in question,the first switch of the respective high-side switching unit and the other switch is the second switch of the respective high-side switching unit, and an evaluation and control unit which, based on the magnitude of the input voltage, controls the input voltage terminal and the magnitudes of the actual and target output voltages at the output terminals, controls the electronic switches of the input half-bridge and the output half-bridges, wherein the evaluation and control unit selects, based on a decision criterion that results from a potential undershooting of a target output voltage or electrical power to be provided at at least one of the output terminals by the actual output voltage or electrical power currently provided at that terminal, the high-side switching unit which is connected to that output terminal.to forward the electrical energy stored in the inductor during the discharge phase of a control cycle, wherein the evaluation and control unit, during the charging phase of a control cycle intended for storing electrical energy in the inductor, switches the low-side switch of the input half-bridge to blocking, switches the high-side switch of the input half-bridge to conducting, switches the low-side switch of the output half-bridges to conducting, and, if this has not already occurred at the end of the discharge phase of the previous control cycle, and in the event that one of the two actual output voltages is lower than the respective specified target output voltage, selects one of the two high-side switch units for the current control cycle based on the decision criterion, whereby this high-side switch unit is the selected high-side switch unit and the other high-side switch unit is the unselected high-side switch unit.switches the two switches of the unselected high-side switch unit to blocking or keeps them blocked, switches the first switch of the selected high-side switch unit from blocking to conducting or keeps this first switch conducting if it is already conducting, and switches the second switch of the selected high-side switch unit to blocking or keeps it blocking if it is already blocking, in a first transition phase following the charging phase keeps the first switch of the selected high-side switch unit conducting, switches the low-side switch of the output half-bridges and the high-side switch of the input half-bridge each from conducting to blocking, and switches the second switch of the selected high-side switch unit and the low-side switch of the input half-bridge from blocking to conducting,In the discharge phase of a control cycle following the first transition phase, the states of the electronic switches of the input half-bridge and the output half-bridges that were assumed at the end of the first transition phase are maintained, and in a second transition phase following the discharge phase, which is followed by the charging phase of the next control cycle, the second switch of the selected high-side switch unit and the low-side switch of the input half-bridge are switched from conducting to blocking, and the low-side switch of the output half-bridges and the high-side switch of the input half-bridge are switched from blocking to conducting, provided that this is to occur now and in the event that one of the two actual output voltages is lower than the respective specified target output voltage,For the next control cycle, one of the two high-side switch units is selected based on the decision criterion, and the two switches of the unselected high-side switch unit of the current control cycle are kept in the blocked position if, according to the decision criterion, the unselected high-side switch unit of the current control cycle should also not be selected in the next control cycle, and the selected high-side switch unit of the current control cycle should also be selected in the next control cycle, or the first switch of the high-side switch unit selected in the current control cycle is switched from conducting to blocking, and the first switch of the high-side switch unit not selected in the current control cycle is switched from blocking to conducting.if, according to the decision criterion, the selected high-side switch unit of the current control cycle should not be selected for the next control cycle, and the unselected high-side switch unit of the current control cycle should be selected.
[0007] The DC-DC converter according to the invention is characterized, among other things, by the fact that only a single inductive energy storage device (hereinafter referred to as an inductor) is required to provide several output voltages. The inductive energy storage device is typically a coil or a transformer, i.e., a device that stores inductive energy and releases it in the form of a discharge current that automatically (continues to) flow after disconnection from the supply voltage.
[0008] The multiple output half-bridges feature bidirectional high-side switching units, each connected between a different output terminal and a common connection node. A low-side switch, common to all output half-bridges, is located between the connection nodes and a reference potential. During the dead-time intervals in the switching phases or during commutation, one of the two electronic switches of the high-side switching unit connected to the output terminal being supplied with electrical charge is closed. This switch, which can also be called the select switch, is closed during the charging phase.On the input side, the input voltage connection is either coupled to the connection node of the output half-bridges via the inductor (boost converter) or there is an input half-bridge with high-side switch and low-side switch, whose common connection node is connected to the inductor (buck-boost converter).
[0009] As is typical for switching converters, the conversion of the input voltage to an output voltage occurs in clocked drive cycles. A drive cycle includes a charging phase in which the inductor is charged with electrical energy. In a subsequent initial transition phase, commutation takes place, i.e., the switching of the individual electronic switches, as is generally known. During a dead-time interval, the freewheeling diode of some of the electronic switches, typically implemented as power transistors, is used.
[0010] Since the output half-bridges are interconnected via the connection node, care must be taken to prevent cross-currents. These arise when different output voltages need to be provided at the output terminals. The required output voltages, and indeed whether any output voltages are required at all, can be determined by the connected loads, which communicate with the evaluation and control unit of the DC-DC converter, for example, according to a communication standard. This also allows the system to determine whether, and if so, which output terminal is not connected to a load.
[0011] According to the invention, the electronic switches of the high-side switch arrangements, which are connected in anti-series with respect to their freewheeling diodes, now fulfill two functions. For the sake of illustration, let us consider a DC-DC converter with two output terminals, where the second output is intended to provide a higher output voltage than the first. During the charging phase, it is therefore necessary to ensure that no cross-current flows from the second output to the first output. This is prevented by opening the switch of the high-side switch arrangement connected to the second output terminal whose freewheeling diode is reverse-biased with respect to the potential cross-current.During the transition phases, this switch is the one that, when the second output terminal is supplied with electrical charge, directs the (discharge) current flowing from the inductor to the output terminal via its freewheeling diode during the dead time intervals.
[0012] The other switch in each high-side switch assembly acts as a select switch, which is closed during the charging phase. Therefore, if, in the example above, the second output terminal is to be supplied with electrical energy during the discharge phase, this select switch is already closed. This results in the scenarios typically encountered in DC-DC converters with a single output half-bridge during the initial transition phase between charging and discharging. As is standard practice, after the charging phase is complete, the common low-side switch of the output half-bridges must first be opened. Then, after observing the dead time interval, the remaining open switch of the relevant high-side switch assembly can be closed.During this dead time interval, the current from the inductor flows through the freewheeling diode of said switch, accepting that there will be increased power loss in said switch for the duration of the dead time interval (incidentally, increased power loss also occurs in the case of a buck-boost converter in the low-side switch of the input half-bridge).
[0013] After the closing of said switch of the relevant high-side switch arrangement (and in the case of a buck-boost converter after closing the low-side switch of the input half-bridge), the current then flows through the respective closed switches, thus significantly reducing the power loss compared to its value during the dead-time interval.
[0014] The concept according to the invention thus offers the advantage of being able to provide several output voltages of different magnitudes using a single inductor. In each drive cycle, a new decision must be made as to which of the output terminals must be supplied with electrical energy during the transition phases and, in particular, during the dead-time intervals. Ideally, a DC-DC converter with, for example, two output terminals can be implemented in this way. It is also conceivable that a DC-DC converter with three or four (or even more) output terminals can be implemented, although it should be noted that the clock rate at which the DC-DC converter operates increases with the number of outputs it has.
[0015] According to a second embodiment of the invention, the above-mentioned problem is solved by a DC-DC converter (in the form of a boost converter) for converting an input voltage into at least two desired output voltages, which can be different or the same within a predefinable output voltage range, wherein the DC-DC converter is provided with an input voltage terminal for applying the input voltage, an output half-bridge comprising at least one first and one second output terminal from which at least one first and one second output voltage can be tapped, a reference potential terminal, an electronic low-side switch and at least two electronic high-side switch units, each of which is connected between another of the output terminals and a common connection node to which the low-side switch is further connected, an inductor arranged between the input voltage terminal and the connection node of the output half-bridges and which can be charged with electrical energy during a charging phase of one of several successive clocked drive cycles and discharged during a discharge phase of a drive cycle to transfer the electrical energy to one of the output terminals,wherein each high-side switch unit comprises a first electronic switch and a second electronic switch, each with a diode connected in parallel to the switch, wherein the two switches of each high-side switch unit are connected in series with their diodes in anti-series orientation, and wherein the switch of each high-side switch unit whose diode is reverse-biased with respect to a current flow from the connection node to the output terminal connected to the high-side switch unit in question is the first switch of the high-side switch unit in question, and the other switch is the second switch of the high-side switch unit in question, and an evaluation and control unit which, on the basis of the magnitude of the input voltage at the input voltage terminal and the magnitudes of the actual and target output voltages at the output terminals, controls the electronic switches of the output half-bridges,wherein the evaluation and control unit selects, based on a decision criterion derived from a potential shortfall in a target output voltage or electrical power to be provided at at least one of the output terminals (Vout1, Vout2) due to the actual output voltage or electrical power currently provided at that terminal, the high-side switching unit connected to the output terminal to which electrical energy stored in the inductor during the discharge phase of a control cycle is to be transferred, wherein the evaluation and control unit switches the low-side switch of the output half-bridges conducting during the charging phase of a control cycle and, if this has not already occurred at the end of the discharge phase of the previous control cycle and in the event that one of the two actual output voltages is lower than its respective target output voltage,For the current control cycle, one of the two high-side switch units is selected based on the decision criterion, whereby this high-side switch unit is the selected high-side switch unit and the other high-side switch unit is the unselected high-side switch unit, the two switches of the unselected high-side switch unit are switched off or kept switched off, the first switch of the selected high-side switch unit is switched from switched off to switched on or, if this first switch is already switched on, kept switched on, and the second switch of the selected high-side switch unit is switched off or, if this second switch is already switched off, kept switched off, and in a first transition phase following the charging phase, the two switches of the unselected high-side switch unit are kept switched off.keeps the first switch of the selected high-side switch unit in the conducting state, switches the low-side switch of the output half-bridges from conducting to blocking, and then, after waiting for the dead time interval, switches the second switch of the selected high-side switch unit from blocking to conducting, maintains the states of the electronic switches of the output half-bridges assumed at the end of the first transition phase in the discharge phase of a control cycle, and in a second transition phase following the discharge phase, which is followed by the charging phase of the next control cycle, switches the low-side switch of the output half-bridges from blocking to conducting, then, after waiting for the dead time interval, switches the second switch of the selected high-side switch unit from conducting to blocking, and ,provided that it is to be done now and in the event that one of the two actual output voltages is lower than the respective specified target output voltage, selects one of the two high-side switch units for the next control cycle based on the decision criterion and keeps the two switches of the unselected high-side switch unit of the current control cycle blocked, if, according to the decision criterion, the unselected high-side switch unit of the current control cycle is also not to be selected in the next control cycle and the selected high-side switch unit of the current control cycle is also to be selected in the next control cycle,or changes the first switch of the high-side switch unit selected in the current control cycle from conducting to blocking, and changes the first switch of the high-side switch unit not selected in the current control cycle from blocking to conducting, if, according to the decision criterion, the selected high-side switch unit of the current control cycle will not be selected for the next control cycle and the unselected high-side switch unit of the current control cycle will be selected.
[0016] In contrast to the DC-DC converter according to the first embodiment of the invention, the DC-DC converter according to the second embodiment of the invention does not have an input half-bridge. As already mentioned above, the inductor is directly connected to the input terminal, while the half-bridges at its outputs are configured as described above in connection with the buck-boost converter.
[0017] For the special features relating to the preparation of the switching phases, including the decision as to which output terminal must be supplied with electrical energy in the discharge phase following the charging phase after the transition phase, the following considerations apply to both variants according to the invention.
[0018] In both variants of the DC-DC converter according to the invention, the aforementioned decision as to which of the outputs should be supplied with energy stored in the inductor is made on the basis of at least one of the following decision criteria. According to these decision criteria, The high-side switching unit for the discharge phase of a control cycle is selected as the high-side switching unit connected to the output terminal where an actual output voltage is currently present that is lower than the target output voltage to be provided, whereby the selected high-side switching unit remains selected for a maximum of a predefinable number of control cycles, provided that the actual output voltage of the output terminal connected to the selected high-side control unit continues to be lower than the respective target output voltage to be provided, and / or for the discharge phase of a control cycle, when an actual output voltage is currently present at all output terminals that is lower than the respective target output voltage to be provided, the high-side switching unit connected to the output terminal where an actual output voltage is present is selected.the respective target output voltage falls below the required output voltage by the larger absolute difference (in % or absolute values), whereby the selected high-side switching unit remains selected for a maximum of a predefined number of control cycles, provided that the actual output voltage of the output terminal connected to the selected high-side switching unit continues to fall below the required output voltage by the largest absolute difference compared to the actual output voltages of the other output terminal(s), and / or for the discharge phase of a control cycle, when all output terminals currently have an actual output voltage that is lower than the respective target output voltage, the high-side switching unit connected to the output terminal with an actual voltage is selected.the reduction rate compared to its value in the previous control cycle is greater than the reduction rate of the actual output voltages at the other output terminals compared to their respective values in the previous control cycle, wherein the selected high-side switching unit remains selected for a maximum of a predefinable number of control cycles, and / or for the discharge phase of a control cycle, if an actual output voltage is currently present at all output terminals that is lower than the respective target output voltage to be provided, the high-side switching units are selected alternately or cyclically, or the respective selected high-side switching unit remains selected for a maximum of a predefinable number of control cycles before a high-side switching unit is selected that is connected to another output terminal whose actual output voltage is lower than the respective target output voltage to be provided.The aforementioned cases can be applied to the consideration of the respective actual electrical power supplied and the corresponding required electrical target power.
[0019] Alternatively, the system does not necessarily check which output currently has the largest difference between its actual output voltage and the required target output voltage, but only whether any of the output terminals have an actual output voltage lower than the respective target output voltage, and which output terminal was supplied with energy during the previous drive cycle. Thus, if, for example, all output terminals have an actual output voltage below their respective target output voltages when the DC-DC converter starts up, the output terminals are charged alternately or cyclically until one of them supplies the relevant target output voltage. This prevents an output terminal with high energy demand (because, for example, the target output voltage is currently quite high) from preventing the other output terminals from being charged, which could cause them to fall below their minimum or target value.
[0020] In a further advantageous embodiment of the invention, it can be provided that the evaluation and control unit has a comparator, in particular in the form of a Schmitt trigger, whose two inputs are connected to the two output terminals and which provides an output signal at its output that is representative of the decision criterion for which of the high-side switch units is to be selected.
[0021] In the aforementioned variants for the decision criterion, the maximum number of activation cycles for which each selected high-side switch unit remains evaluated is chosen depending on the application. Thus, it is possible that the high-side switch unit selected for the discharge phase of an activation cycle will not be retained for the next activation cycle, in which case the number would be zero. However, depending on the situation, it can be advantageous to retain a selected high-side switch unit for, for example, only a few subsequent activation cycles, such as one to three or one to five. In particular, the high-side switch unit connected to an output terminal that has a higher recharging energy requirement in the meantime can retain the selected high-side switch unit for one or more subsequent activation cycles.The discharge phases of these activation cycles remain unchanged. The number of such charging cycles ultimately also depends on how much energy is demanded from the other output terminals. If the loads connected to these other output terminals currently have a low energy demand, but the aforementioned output terminal has a high demand, a greater number of activation cycles can certainly be tolerated without changing the selected high-side switching unit than, for example, in cases where loads with a comparatively high current energy demand are connected to several output terminals.
[0022] Therefore, no absolute values can be given for the exact number of maximum predefinable control cycles for whose discharge phases a once selected high-side switch unit remains selected; at most, one can say that this number would be in a range of 2 to 10, in particular from 2 to 6 and preferably from 2 to 4.
[0023] The decision-making process for which output to supply with electrical energy is highly complex. In the case of applying the invention to the USB-C system, for example, a higher system level decides which output to supply with electrical energy. The connected devices, for instance, report their power requirements to the USB-C system, and the higher system level then continuously decides which output will be supplied with electrical energy in the next discharge cycle. The energy or power requirements of the devices can change during the time they are connected to the converter. For example, a device whose battery has only a small amount of energy remaining and therefore urgently needs to be recharged to a certain level may require more energy or power during this phase than afterward.The entire concept therefore involves, in a sense, an energy management system that controls the decision-making process and the decision criterion for which output should be supplied with energy.
[0024] In the event that none of the output terminals need to be supplied with electrical energy (because all output terminals deliver actual output voltages that are greater than the respective required target output voltage), the DC-DC converter according to the first variant of the invention can be operated in free-wheeling mode during the discharge phase, in which In the first transition phase, the first switches of the two high-side switch units are held in a conducting state, or in a conducting state, or in a blocking state, or in a blocking state, the second switches of the two high-side switch units are held in a blocking state, the low-side switch of the output half-bridges is held in a conducting state, and the high-side switch of the input half-bridge is switched from conducting to blocking, and then, after waiting for the dead time interval, the low-side switch of the input half-bridge is switched from blocking to conducting.During the discharge phase, the states of all switches of the input and output half-bridges that were in place at the end of the first transition phase are maintained, and in the second transition phase, the low-side switch of the input half-bridge is switched from conducting to blocking, and subsequently, after waiting for the dead time interval, the high-side switch of the input half-bridge is switched from blocking to conducting.
[0025] The DC-DC converter according to the invention, in both variants, can be equipped with a peak current control (PCC) system, which is advantageously implemented in combination with a slope compensation system. Both concepts are generally known in switching converters, which is why these systems will not be discussed further here.
[0026] In addition to these two controls, according to an advantageous embodiment of the invention, the current for the inductor can also be regulated depending on the extent to which the respective actual output voltages at the output terminals are below or above their respective target output voltages. The extent of the deviations (both above and below) can be represented by error signals of varying magnitudes, which, after summation and low-pass filtering, provide an additional signal. This signal, like the slope compensation signal, is subtracted from the specified maximum current, and the resulting signal is then used as the current value to which the inductor current is to be regulated. This requires a measuring device for detecting this current or its magnitude and a controller, for example, a finite-state machine.
[0027] In other words, it is advantageous if the evaluation and control unit has a measuring device for supplying a measurement signal representing the magnitude of the current through the inductor coil, wherein the evaluation and control unit terminates the charging phase of a control cycle when the measurement signal exceeds a threshold value representing a predefinable maximum current, wherein it can be provided in particular that the threshold value is variable depending on whether one or each of the current actual output voltages is greater or less than the respective target output voltage, and / or depending on the target value of a slope compensation control.
[0028] As briefly mentioned above, power transistors, such as MOS transistors with appropriately designed freewheeling diodes, are conveniently used as the high-side and low-side switches of the output half-bridges and, if present, the input half-bridge.
[0029] In principle, these MOS transistors can be configured as bidirectional switches with either a common-source or common-drain configuration in any high-side switch arrangement. The common-source arrangement is simpler to implement with regard to driving the gates of the MOS transistors, and therefore offers advantages over the common-drain arrangement in this respect.
[0030] As an alternative to implementing the electronic switches of the DC voltage converter according to the invention as MOS transistors, bipolar transistors or IGPTs are suitable, wherein a (freewheeling) diode is connected in parallel to the current paths of these transistors, which can in particular be designed as a Schottky diode.
[0031] The DC-DC converter according to the invention is particularly suitable for systems in which the input voltage provided to the DC-DC converter fluctuates within a predefinable voltage range. This is the case, for example, with a vehicle electrical system, which can fluctuate depending on the state of the vehicle battery. The required output voltages can be higher or lower than the currently available electrical system or input voltage, depending on the connected load and the level of the available input voltage. Therefore, it must be assumed that the DC-DC converter must be operable not only as a boost converter but also as a buck-boost converter.It should also be ensured that, during operation of the DC-DC converter, one of the outputs is not supplied with any further electrical charge due to safety measures. This is the case, for example, if a power bank with a lithium-ion battery or a device with such a battery is connected to one (or both) outputs. For safety reasons, such a battery must not be supplied with any further charging current once it reaches a certain voltage.
[0032] During operation of the DC-DC converter according to the invention, it may occur that one of the two outputs needs to be switched off. However, this is ultimately unproblematic, since in this case only the other output is supplied with electrical charge, and the DC-DC converter according to the invention then operates like a DC-DC converter with a (single) output.
[0033] The aforementioned protective function for switching off or deactivating one or both or all outputs when sufficient output voltage is provided for the connected consumer, and the partial or complete switching off of the DC-DC converter (e.g., by opening all switches for the duration of the switch-off), is controlled, for example, by the evaluation and control unit of the DC-DC converter according to the invention.
[0034] The invention is explained in more detail below with reference to an exemplary embodiment and the drawing. Specifically, the drawing shows: Fig. 1 shows the basic structure of a buck-boost converter, assuming that the output voltage at Vout2 is greater than at Vout1 and that Vout2 is to be supplied with energy from the inductor, and showing the switch positions for the charging phase. Figs. 2 to 6 show different switching states of the switches in the subsequent phases following the charging phase and during the dead time intervals for the Fig. 1 In the case shown, where output Vout2 is to be supplied with energy from the inductor during the discharge phase, Figs. 7 to 12 show the switch positions of the buck-boost converter with the boundary conditions according to Fig. 1 In the event that Vout1 is to be supplied with electrical energy from the inductor during the discharge phase, Figs. 13 to 16 show the switch positions of the buck-boost converter with the boundary conditions according to Fig. 1 For the case where none of the outputs need to be supplied with electrical energy from the inductor during the discharge phase (free wheeling), Figs. 17 to 22 show an embodiment of the DC-DC converter as a boost converter assuming the boundary conditions as defined above, where output Vout2 is to be supplied with electrical energy from the inductor, Figs. 23 to 28 show an embodiment of the DC-DC converter as a boost converter assuming the boundary conditions as defined above, where output Vout1 is to be supplied with electrical energy from the inductor, Fig. 29 shows a block diagram of the components of the evaluation and control unit for the switches, Fig. 30 shows an example of the time profiles of the actual output voltages and the associated time profile of the current through the inductor as well as the different duty cycles concerning the ratio of charging phase duration to discharge phase duration, and Fig.31 a tabular overview to explain the regulation according to the block diagram of the . Fig. 29 in the case of the DC voltage converter being implemented as a buck-boost converter.
[0035] Fig. 1 Figure 1 schematically shows the basic structure of a first embodiment of a DC-DC converter according to the invention, which in this case is configured as a buck-boost converter. The DC-DC converter GSW comprises, in a known manner, an input half-bridge EH with an electronic high-side switch M1 and an electronic low-side switch M2. Both switches each have a backgate diode BGD1 and BGD2, respectively, in a known manner. The backgate diodes BGD1 and BGD2 are connected in the same orientation, with the anode of the backgate diode BGD2 connected to the reference potential GND and its cathode connected to the junction VKEH of the input half-bridge.
[0036] The DC-DC converter GSW has at least two output half-bridges, AH1 and AH2, each comprising a high-side switch unit, HSE1 or HSE2 respectively, and a common electronic switch, the low-side switch M3. The low-side switch M3 is connected between the reference potential GND and a junction VKAH connecting the output half-bridges AH1 and AH2. The high-side switch units HSE1 and HSE2 are connected to each other at the junction VKAH.
[0037] Between the connecting nodes VKEH and VKAH of the input half-bridge EH and the two output half-bridges AH1 and AH2 there is an inductor IND, i.e. an inductive energy storage device, which in this embodiment is designed as a coil.
[0038] Each of the high-side switch units HSE1 and HSE2 comprises two electronic high-side switches M4a, M4b and M5a, M5b, respectively. These high-side switches are also each equipped with a backgate diode BGD4a, BGD4b and BGD5a, BGD5b, respectively. The two switches of each high-side switch unit HSE1 and HSE2 are connected such that their backgate diodes are connected in anti-series. In this embodiment, the anodes of the respective anti-series connected backgate diodes are connected together. When the electronic switches are implemented as MOS transistors, this results in a common-source circuit arrangement.
[0039] The input voltage is applied to the input voltage terminal Vin of the input half-bridge EH, while the output voltage terminals Vout1 and Vout2 are provided at the high-side switching units of the two output half-bridges AH1 and AH2. All of the aforementioned terminals of the DC-DC converter GSW can be equipped with buffer capacitors Pkin, Pkout1, and Pkout2.
[0040] The electronic switches of the DC-DC converter GSW are controlled by an evaluation and control unit (AAE), which is typically an ECU with, for example, a microcontroller and associated hardware such as I / O ports, CPU, ROM, and RAM. The evaluation and control unit AAE receives signals from the input and output voltage terminals Vin, Vout1, and Vout2 via the connecting lines VBVin, VBVout1, and VBVout2, and thus receives information about the current output voltages during operation of the DC-DC converter GSW.Via a communication bus KB between a bus communication interface BKSS of the control and evaluation unit AAE and the bus communication interfaces BKSSout1 and BKSSout2, consumers connected to the output terminals Vout1 and Vout2 send data to the control and evaluation unit AAE about the magnitude of the target output voltages that must be provided for the operation of the consumers.
[0041] The DC-DC converter GSW supplies at least two output terminals, Vout1 and Vout2, with output voltages of different magnitudes from a single energy source, namely the input voltage or the inductor IND. It is irrelevant which of the two output terminals, Vout1 or Vout2, requires the higher output voltage. In the example shown here... Fig. 1 Let us assume that the second output terminal Vout2 should have a target output voltage of, for example, 10 V, while the first output terminal Vout1 should have a target output voltage of 5 V.
[0042] As is typical for DC-DC converters, the GSW DC-DC converter also operates in a clocked mode. However, the special feature is that in each clock cycle, either one of the two output terminals, Vout1 or Vout2, is supplied with electrical energy. This results in the further special requirement that measures must be taken to prevent cross-currents between the two output terminals.In the case considered here, it is therefore necessary to prevent current from flowing from the second output terminal Vout2 via the second high-side switch unit HSE2, the connection node VKAH and the first high-side switch unit HSE1 to the first output terminal Vout1, in all three possible operating situations, namely that the first output terminal Vout1 must be supplied with electrical energy, that Vout2 must be supplied with electrical energy, or that neither of the two output terminals needs to be supplied with electrical energy (free wheeling).
[0043] In the Fign. 1 bis 6 The switching states of the individual electronic switches in the individual phases of a control cycle are shown for the case that the second output terminal Vout2 is to be supplied with energy because, for example, the actual output voltage at the first output terminal Vout1 is currently greater than the required target output voltage, but at the second output terminal Vout2 there is currently an actual output voltage that is less than the required target output voltage.
[0044] First, as is typical for a DC-DC converter, a charging phase takes place ( Fig. 1 The inductor IND is charged with electrical energy via the high-side switch M1 of the input half-bridge EH and the low-side switch M3 of the output half-bridges AH1 and AH2. These two switches are therefore closed, while the low-side switch M2 of the input half-bridge EH is open. During this phase, the charging current ALS flows, as shown in Fig. 1 As shown, the electronic switches of the two high-side switch units HSE1 and HSE2 are open on the output side. The backgate diode BGD5b of the high-side switch M5b prevents current flow from the second output terminal Vout2 to the first output terminal Vout1 (the output voltage at Vout2 is higher than that at Vout1, as assumed above).
[0045] During the charging phase, the system determines which output terminal should be supplied with electrical energy during the discharge phase of the current control cycle. As mentioned above, it is assumed that the second output terminal, Vout2, must be supplied with electrical energy. Therefore, the high-side switch M5a (select switch) of the second high-side switch unit HSE2 is closed during the charging phase. This prepares the second high-side switch unit HSE2 for the dead-time interval of the first switching phase following the charging phase. After the low-side switch M3 opens, the backgate diode BGD5b of the high-side switch M5b will then direct the inductor IND's current to the second output terminal, Vout2.
[0046] The situation in the first switching phase is in Fig. 2 As shown, the low-side switch M3 is opened first. Simultaneously, the high-side switch M1 of the input half-bridge is also opened. The current from the inductor IND now flows as discharge current ELS via the second high-side switch unit HSE2 to the second output terminal Vout2.
[0047] Provided in Fig. 2 and in the further figures where switches with states symbolized by dashed and solid lines are shown, the dashed state refers to the one that is left in the relevant phase of the control cycle in order to assume the switch state marked with a solid line.
[0048] According to the switching scenario typical for switching converters, which must be observed, short-circuit currents are prevented by closing the switches only after the switch to be opened is open ("break before make", i.e., first interrupt the connection before closing it again), the high-side switch M5b is initially still open during the dead time interval following the opening of the low-side switch M3 (see Fig. 2 ), which is why the current flows through the backgate diode BGD5b and is therefore limited in magnitude. This state should be kept as short as possible, as the power dissipation in the high-side switch M5b would otherwise increase significantly.
[0049] Therefore, during the dead time interval, the high-side switch M5b is closed (see Fig. 3 ), which initiates the discharge phase (see Fig. 3 ).
[0050] During the dead time interval (see Fig. 3 ) and thus, from the opening of the low-side switch M3, the discharge current ELS also flows through the backgate diode BGD2 of the low-side switch M2 of the input half-bridge EH. This switch is now also closed in the first dead-time interval (see Fig. 3 ), so that at the beginning of the discharge phase (see Fig. 3 ) is closed, which now reduces the power losses in the two switches M2 and M5b to a minimum due to the conductive state of both switches.
[0051] Towards the end of the in Fig. 4 The charging phase shown is followed by a switching phase, namely the second sales phase (see Fig. 5 Here too, care must be taken to ensure that the electrical connection between the second output terminal Vout2 and the reference potential GND, to which the low-side switch M2 of the input half-bridge EH is connected, is interrupted before the electrical connection between the input terminal Vin of the input half-bridge EH and the reference potential GND is closed via the electronic switches M1 and M3.
[0052] Therefore, first the high-side switch M5b of the second high-side switch unit HSE2 and the low-side switch M2 of the input half-bridge EH are opened, whereby the discharge current EAS continues to flow through the backgate diodes BGD5b and BGD2 of these two switches for a dead-time interval. Now the high-side switch M1 of the input half-bridge EH and the low-side switch M3 of the output half-bridges AH1 and AH2 are closed (see Fig. 6 ), so that the charging current ALS can now flow again from the input terminal Vin via the two aforementioned switches to the reference potential GND.
[0053] At the end of the second switching phase (see Fig. 6 It could already be investigated which of the two output terminals Vout1, Vout2 is to be supplied with electrical energy in the discharge phase of the next control cycle.
[0054] If, during the discharge phase of the next control cycle, the second output terminal Vout2 is again to be supplied with electrical energy, the high-side switch M5a of the second high-side switch unit HSE2 could remain closed. However, the control scenario can also be configured so that, towards the end or during the discharge phase, the select switch—in this case, the high-side switch M5a, which was closed during the charging phase—is always reopened. This is because the decision as to which of the two select switches M4a and M5a must be closed for an upcoming charging process of one of the two output terminals is always made during the charging phase of a control cycle.
[0055] In the case under consideration here, the Select switch M5a is opened again in the second switching phase, since it is assumed that it is already certain that the first output terminal Vout1 is to be supplied with electrical energy in the discharge phase of the next control cycle.
[0056] The situation in the charging phase of the next control cycle is in Fig. 7 shown. It differs from the situation according to Fig. 1 , when the Select switch M4a is closed, because in the dead time interval of the first switching phase and in the discharge phase of the next control cycle the current from the inductor IND should flow to the first output terminal Vout1.
[0057] The Fig. 8 shows the switching phase in which the high-side switch M1 of the input half-bridge EH and the low-side switch M3 of the output half-bridges are opened first, while the high-side switch M4b of the high-side switch unit HSE1 and the low-side switch M2 of the input half-bridge EH are still open ("break before make", i.e., first breaking the connection before closing it again). Fig. 8 This shows the situation during the first dead time interval. During this interval, the discharge current EAS flows again through the backgate diodes BGD2 and BGD4b of the low-side and high-side switches M2 and M4b, respectively, until these switches themselves are closed (see Fig. 9 ), which is synonymous with the start of the discharge phase (see Fig. 10 ).
[0058] To the discharge phase (see Fig. 11 The next switching phase then follows, in which the high-side switch M4b and the low-side switch M2 are preferably opened simultaneously, so that for the subsequent second dead-time interval the discharge current again flows through the backgate diodes of these two switches until the high-side switch M1 of the input half-bridge EH and the low-side switch M3 of the output half-bridges are closed. This results in the situation according to Fig. 12 This figure again indicates that the select switch M4a is now generally opened again, regardless of whether, in the discharge phase of the next control cycle, it will again be the first output terminal Vout1 that needs to be supplied with electrical energy. This makes it clear that it is advantageous for the decision as to which select switch must be closed and which must be open in the next control cycle to be made towards the end of the second switching phase.
[0059] Based on the Fign. 13 bis 16 We will briefly discuss the case where, during the discharge phase of a control cycle, neither output terminal Vout1 nor Vout2 needs to be supplied with electrical energy because the actual output voltages at both terminals are higher than the respective required target output voltage. This is known as "free wheeling".
[0060] During the charging phase of such a control cycle, the situation initially presents itself according to Fig. 13 One. Neither of the two Select switches M4a, M5a is closed during this charging phase.
[0061] Towards the end of the charging phase, i.e. in the subsequent first switching phase (see Fig. 14 The high-side switch M1 of the input half-bridge EH is now opened, while the low-side switch M3 of the output half-bridge remains closed. The inductor current continues to flow through the backgate diode BGD2 of the still-open low-side switch M2 of the input half-bridge EH, essentially in the circuit defined by the two low-side switches M2 and M3 and the inductor IND, as shown in Fig. 14 shown.
[0062] Subsequently, during the dead time interval (see Fig. 15 ) the lowside switch M2 of the input half-bridge EH is now closed, so that the energy stored in the inductor IND is virtually retained for the discharge phase.
[0063] This process will begin with the start of the next switching phase (see Fig. 16 ) terminates. First, the low-side switch M2 of the input half-bridge EH opens, so that the current of the inductor IND flows again through its backgate diode BGD2, until the high-side switch M1 of the input half-bridge EH closes again, so that the situation returns to the state described above. Fig. 13 adjusts.
[0064] In the Fign. 17 bis 28 The various switch positions of the output switches of an alternatively designed DC-DC converter GSW', configured as a boost converter, are shown. The boost converter does not have an input half-bridge; rather, the inductor IND is connected directly to the input terminal Vin.
[0065] The current of the inductor IND is therefore directed via the lowside switch M3 of the output half-bridges during the charging phase of a control cycle, while during the discharging phase of a control cycle it is diverted to one of the two output terminals Vout1, Vout2.
[0066] The Fign. 17 bis 22 This illustrates the scenario where the second output terminal Vout 2 is to be supplied with electrical energy during the discharge phase. Accordingly, during the charging phase (see Fig. 17 ) the Select switch M5a is closed. If, during the switching phase (see Fig. 18 ) to initially open the low-side switch M3, the current from the inductor IND flows through the closed select switch M5a and the backgate diode BGD5b of the high-side switch M5a of the selected high-side switch unit HSE2. At the end of the dead-time interval (see Fig. 19 ) the highside switch M5b of the selected highside switch unit HSE2 is now closed, so that from this point on and for the duration of the subsequent discharge phase the discharge current ELS flows through the closed switches of the selected highside switch unit HSE2 ( Fig. 20 ).
[0067] The second switching phase then follows the discharge phase (see Fig. 21 ), in which the highside switch M5b of the selected highside switch unit HSE2 is first opened, with the result that the discharge current ELS now flows through its backgate diode BGD5b.
[0068] The switch positions then follow this situation according to Fig. 22 by closing the lowside switch M3 of the output half-bridges, with the result that the (in this case) charging current ALS now flows through the closed lowside switch M3 and no current flows through the previously selected highside switch unit HSE2.
[0069] The Fign. 23 bis 28 show the corresponding situation in the case where, during the discharge phase of a control cycle of the DC voltage converter GSW', the first output terminal Vout1 is to be supplied with electrical energy.
[0070] During the charging phase ( Fig. 23 ) the Select switches are first opened or closed (switch M5a is now opened again if it has not already been opened at the end of the second switching phase, while the Select switch M4a is closed).
[0071] According to Fig. 24 In the first switching phase, the low-side switch M3 is opened, so that the current of the inductor IND flows as discharge current ELS through the closed select switch M4a and the backgate diode BGD4b of the still open high-side switch M4b of the selected high-side switch unit HSE1. This state is maintained for the dead-time interval at the end of which the high-side switch M4b of the selected high-side switch unit HSE1 is closed (see Fig. 25 ).
[0072] This results in the situation of the discharge phase of this control cycle (see Fig. 26 ).
[0073] In the Fign. 27 and 28The situations that arise in the second switching phase and the second dead-time interval are shown again. Before the low-side switch M3 can be closed, the high-side switch M4b of the selected high-side switch unit HSE1 must first be opened, with the result that the discharge current ELS flows through its backgate diode BGD4b. Only then can the low-side switch M3 be closed (see Fig. 28 ), so that charging current ALS now flows again from the input terminal Vin via the inductor IND and the closed lowside switch M3.
[0074] In Fig. 29 The diagram schematically shows, at block diagram level, which components the evaluation and control unit AAE can have.
[0075] The current IL in the inductor is measured using a suitable measuring device, such as a shunt SH. The output of a dedicated measuring amplifier MV is compared with the input of a comparator K. The signal for the currently valid limiting current Ilim of the inductor IND is present at the other input of the comparator K. This signal is calculated as the difference between the application-specific maximum inductor current Imax and a slope compensation value SC. Slope compensation increases the ramp rate of the inductor current IL by adding a parameter derived from the system clock as a ramp. This is achieved by increasing the current rise edge by slightly more than half the magnitude of the current fall edge. This measure, intended to prevent oscillation, is generally known and will not be discussed further here.
[0076] Furthermore, a signal Lfout is subtracted from the application-dependent maximum inductor current Imax. This difference results in the inductor limiting current Ilim at the output of the summing amplifier SU.
[0077] In this embodiment, the output signal of comparator K is fed to a finite-state machine (FSM). If the inductor current exceeds the limiting current Ilim during a charging phase, the DC-DC converter switches from the charging phase (taking into account the first switching phase with its dead-time interval) to the discharging phase. Its output signals then control the electronic switches as described previously. At the beginning of the next control cycle T, the converter switches again from the discharging phase (with the second transition phase and dead-time interval intervening) to the next charging phase.
[0078] The state machine FSM is controlled by a clock generator CG and receives information about the evaluation of the decision criterion via a port select module PSM, according to which it is decided which of the output terminals Vout1, Vout2 must be supplied with charge.
[0079] In Fig. 29 It is also shown what the basis for the filter signal Lfout is. Lfout is the output signal of an external control loop, which, as previously shown, is determined by... Fig. 29 described, enters into the inner control loop for determining the duty cycle, i.e., it co-determines the ratio of the duration of the charging phase P to the duration of the discharging phase S.
[0080] The operation of the outer control loop can be summarized as follows.
[0081] The degree of deviation of the actual output voltages V01 and V02 from their respective target output voltages V1T and V2T is evaluated.
[0082] These distortions form error signals E1 and E2, which are summed in the summing amplifier SUM, taking their signs into account. The output signal of this summing amplifier SUM is filtered in a loop filter LF, and its output signal forms the output signal Lfout. The loop filter LF is implemented, for example, as an integrator, i.e., as a second-order filter.
[0083] In Fig. 30 Assumed examples of the actual output voltages V01 and V02 are shown. Over five consecutive drive cycles T1 to T5, the behavior of the DC-DC converter GSW is observed and explained according to... Fig. 1 , which is implemented as a buck-boost converter. It is assumed that at the beginning of the charging phase P of the drive cycle T1, the actual output voltage V01 is lower than the target output voltage V1T and the actual output voltage V02 is higher than the target output voltage V2T, so that the decision is made to charge Vout1 in the subsequent charging phase.
[0084] At the beginning of the next control cycle T2, it is already known that the actual output voltage V02 is lower than the target output voltage V2T, while the actual output voltage V01 is higher than the target output voltage V1T. In other words, Vout2 is charged during the subsequent discharge phase of control cycle T2.
[0085] The situation is then reversed at the beginning of the subsequent control cycle T3, with the result that output Vout1 is now charged during its discharge phase. In this embodiment, the situation has not yet changed at the beginning of the next control cycle T4. Therefore, in this example, the voltage at output Vout2 remains higher than the target output voltage V2T, while the actual output voltage at output Vout1 remains lower than the target output voltage V1T.
[0086] Finally, in this example, it is assumed that at the beginning of the next control cycle T5, both actual output voltages are higher than their respective target output voltages. It is then decided that neither output terminal Vout1 nor Vout2 will be charged. Thus, during the actual discharge phase of this control cycle, a so-called free wheeling (FW) occurs.
[0087] In the example of the Fig. 30 Ultimately, two situations are shown that can lead to the decision to implement free wheeling. If the decision is made at the beginning of a control cycle, then in the example shown, free wheeling could also be implemented in the first control cycle T1, since both actual voltages are higher than the respective target voltages. However, one can also observe the behavior of the respective target output voltages during a charging phase P. Should the actual output voltage at one of the two output terminals then drop below the target output voltage, the decision can be made to charge the affected output terminal during the discharge phase S1, S2 instead of implementing free wheeling.
[0088] If this last-mentioned principle were applied consistently, then in the example case shown, the decision in the control cycle T5 would also be to charge the output terminal Vout2 in the discharge phase instead of to free wheeling.
[0089] In the example of the Fig. 30 The diagram also shows how the inductor current is regulated. The inductor current IL increases during each charging phase and decreases during each discharging phase, unless free wheeling occurs. The charging phase ends the moment the current IL equals the limiting current Ilim. This controls the duty cycle. During free wheeling, the limiting current Ilim can decrease.
[0090] In Fig. 31 The table shows how the Port Select Module (PSM) reacts depending on the configuration of the error signals E1 and E2. If V02 is below V2T and V01 is above V1T, the error signal E1 is greater than 0 and the error signal E2 is less than 0. Based on this, a decision is made, for example, to charge Vout2 in the following discharge phase.
[0091] Conversely, if V01 is below V1T and V02 is above V2T, the error signal E1 becomes less than 0 and the error signal E2 becomes greater than 0, resulting in output Vout1 being selected during the discharge phase of the current control cycle.
[0092] If both actual output voltages are lower than their respective target output voltages, i.e., both error signals E1 and E2 are less than 0, the choice of output during the discharge phase of the current drive cycle may depend on the magnitude of the deviations. Other decision criteria are also conceivable. For example, if the two actual output voltages remain lower than their respective target output voltages over several cycles, as will be the case when starting up the DC-DC converter, they could be charged alternately. This could also depend on the amount of electrical energy drawn from the two output terminals.
[0093] If both actual output voltages are greater than the respective target output voltages, then both error signals E1 and E2 are greater than 0, which means that in the discharge phase of the current control cycle neither of the two output terminals is supplied with electrical energy and thus free wheeling occurs, so that the electrical energy is held in the inductor IND, which in turn can lead to the inductor limiting current Ilim being reduced in the next control cycle. BEZUGSZEICHENLISTE
[0094] AAE Control and evaluation unit AH1 First output half-bridge AH2 Second output half-bridge ALS Charging current BGD1 Backgate diode of electronic switch M1 BGD2 Backgate diode of electronic switch M2 BGD3 Backgate diode of electronic switch M3 BGD4a Backgate diode of electronic switch M4a BGD4b Backgate diode of electronic switch M4b BGD5a Backgate diode of electronic switch M5a BGD5b Backgate diode of electronic switch M5b BKSS Bus communication interface of the evaluation and control unit BKSSout1 Bus communication interface of the first output BKSSout2 Bus communication interface of the second output CG Clock generator E1 Error signal E2 Error signal EHE Input half-bridge ELS Discharge current FSM Finite-state machine FW Free-wheeling GND Reference potential GSW DC-DC converter Buck-boost converter GSW' DC-DC converter as boost converter HSE1 first high-side switch unit HSE2 second high-side switch unitIInductor current Ilim Inductor limiting current Imax Maximum inductor current IND Inductor K Comparator KB Communication bus LFLoop filter Lfout Filter output signal M1 Electronic switch (high-side switch of the input half-bridge) M2 Electronic switch (low-side switch of the input half-bridge) M3 Electronic switch (low-side switch of the output half-bridges) M4a Electronic switch (high-side switch of the first output half-bridge) M4b Electronic switch (high-side switch of the first output half-bridge) M5a Electronic switch (high-side switch of the second output half-bridge) M5b Electronic switch (high-side switch of the second output half-bridge) MV Measuring amplifier P Charging phase Pkout1 Buffer capacitor at the first output terminal Pkout2 Buffer capacitor at the second output terminal Pkin Buffer capacitor at the input terminal PSM Port select module S1 Discharging phase S2 Discharging phase SC Slope compensation value SH shunt resistor, measuring instrument SU summator SU summator T control cycleV01 Actual output voltage at the first output terminal V02 Actual output voltage at the second output terminal V1T Target output voltage at the first output terminal V2T Target output voltage at the second output terminal Vout1 First output terminal Vout2 Second output terminal Vin Input terminal VBVin Connection line between control and evaluation unit and input terminal VBVout1 Connection line between control and evaluation unit and Vout1 VBVout2 Connection line between control and evaluation unit and Vout2 VKAH Connection node of the output half-bridge VKEH Connection node of the input half-bridge
Claims
1. A DC / DC converter for converting an input voltage into at least two desired output voltages, which can be different or the same within a predeterminable output voltage range, comprising - an input half-bridge (EH) having an input voltage terminal (Vin) for applying the input voltage and a reference potential terminal (GND) and a series circuit comprising an electronic high-side switch (M1) and an electronic low-side switch (M2) with a connection node between the two switches (M1, M2), wherein the low-side switch (M2) of the input half-bridge (EH) has a diode (BGD2) connected in parallel with the switch with a cathode connected to the reference potential terminal (GND), - at least two output half-bridges (AH1, AH2), each of which has an output terminal (Vout1, Vout2) at which a first or second output voltage can be tapped, a common connection node (VKAH), a common reference potential terminal (GND), a common low-side switch (M3) arranged between the connection node (VKAH) and the reference potential terminal (GND) and an electrical high-side switch unit (HSE1, HSE2) which is arranged between the common connection node (VKAH) and the respective output terminal (Vout1, Vout2), - an inductor (IND), which is arranged between the connection nodes (VKEH, VKAH) of the input and output half bridges (EH, AH1, AH2) and which can be charged with electrical energy in a charging phase of one of several successive clocked control cycles and can be discharged in a discharging phase of a control cycle for forwarding electrical energy to one of the output terminals (Vout1, Vout2), - wherein each high-side switch unit (HSE1, HSE2) has a first electronic switch (M4a, M5a) and a second electronic switch (M4b, M5b) each with a diode (BGD4a, BGD4b, BGD5a, BGD5b) in parallel with the switch, wherein the two switches (M4a, M5a; M4b, M5b) of each high-side switch unit (HSE1, HSE2) are connected in series with their diodes in antiserial alignment, and the switch of each high-side switch unit (HSE1, HSE2), the diode of which is located in a reverse direction relative to a current flow from the connection node (VKAH) to the output terminal (Vout1, Vout2) connected to the relevant high-side switch unit (HSE1, HSE2), is the first switch (M4a, M5a) of the respective high-side switch unit (HSE1, HSE2) and the respective other switch is the second switch (M4b, M5b) of the respective high-side switch unit (HSE1, HSE2), and - an evaluation and control unit (AAE) which is configured to control the electronic switches (M1, M2, M3, M4a, M4b, M5a, M5b) of the input half-bridge (EH) and the output half-bridges (AH1, AH2) on the basis of the magnitude of the input voltage at the input voltage terminal (Vin) and the magnitudes of the actual and the target output voltages at the output terminals (Vout1, Vout2), wherein the evaluation and control unit (AAE) is configured to select, on the basis of a decision criterion based on a potential undershoot of a target output voltage or target electrical power to be provided at at least one of the output terminals (Vout1, Vout2) by the actual output voltage currently applied thereto or the actual electrical power provided, the high-side switch unit (HSE1, HSE2) connected to the output terminal (Vout1, Vout2) to which electrical energy stored in the inductor (IND) is to be transferred during the discharging phase of a control cycle. - wherein the evaluation and control unit (AAE) is configured to - in the charging phase of a control cycle provided for storing the inductor (IND) with electrical energy - switch the low-side switch (M2) of the input half-bridge (EH) to be blocking, - switch the high-side switch (M1) of the input half-bridge (EH) to be conducting, - switch the low-side switch (M3) of the output half bridges (AH1, AH2) to be conducting, and - if not yet done at the end of the discharging phase of the previous control cycle and in the event that one of the two actual output voltages is smaller than the respective specified target output voltage, select one of the two high-side switch units (HSE1, HSE2) for the current control cycle using the decision criterion, wherein this high-side switch unit (HSE1, HSE2) is the selected high-side switch unit and the other high-side switch unit is the non-selected high-side switch unit, - switch the two switches (M4a, M4b, M5a, M5b) of the non-selected high-side switch unit to be blocking or keep them switched to be blocking, - transfer the first switch (M4a, M5a) of the selected high-side switch unit from switched to be blocking to switched to be conducting or to keep this first switch (M4a, M5a) switched to be conducting if it is already switched to be conducting, and - switch the second switch (M4b, M5b) of the selected high-side switch unit to be blocking or, if it is already switched to be blocking, keep it switched to be blocking, - in an initial transition phase following the charging phase - keep the first switch (M4a, M5a) of the selected high-side switch unit switched to be conducting, - transfer the low-side switch (M3) of the output half-bridges (AH1, AH2) and the high-side switch (M1) of the input half-bridge (EH) from switched to be conducting to switched to be blocking, and - transfer the second switch (M5a, M5b) of the selected high-side switch unit and the low-side switch (M2) of the input half-bridge (EH) from switched to be blocking to switched to be conducting, - in the discharging phase of a control cycle following the first transition phase - keep the states of the electronic switches (M1, M2, M3, M4a, M4b, M5a, M5b) of the input half-bridge (EH) and the output half-bridges (AH1, AH2) assumed at the end of the first transition phase, and - in a second transition phase following the discharging phase, which is followed by the charging phase of the next control cycle, - transfer the second switch (M5a, M5b) of the selected high-side switch unit and the low-side switch (M2) of the input half-bridge (EH) from switched to be conducting to switched to be blocking, - transfer the low-side switch (M3) of the output half-bridges (AH1, AH2) and the high-side switch (M1) of the input half-bridge (EH) from switched to be blocking to switched to be conducting, - if it is already to take place now and in the event that one of the two actual output voltages is smaller than the respective specified target output voltage, select one of the two high-side switch units (HSE1, HSE2) for the next control cycle on the basis of the decision criterion, and - keep the two switches of the non-selected high-side switch unit of the current control cycle switched to be blocking if the non-selected high-side switch unit of the current control cycle is not to be selected for the next control cycle in accordance with the decision criterion and the selected high-side switch unit of the current control cycle is also to be selected in the next control cycle, - or transfer the first switch (M4a, M5a) of the high-side switch unit selected in the current control cycle from switched to be conducting to switched to be blocking, and transfer the first switch (M4a, M5a) of the high-side switch unit not selected in the current control cycle from switched to be blocking to switched to be conducting, if the selected high-side switch unit of the current control cycle is not to be selected for the next control cycle in accordance with the decision criterion and the non-selected high-side switch unit of the current control cycle is to be selected, - wherein the evaluation and control unit (AAE) is configured to cause, when it has to select the high-side switch unit (HSE1, HSE2) on the basis of the decision criterion, that: - for the discharging phase of a control cycle, that high-side switch unit (HSE1, HSE2) is selected which is connected to that output terminal at which an actual output voltage is currently applied which is smaller than the target output voltage to be provided, wherein the selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles, provided that the actual output voltage of the output terminal (Vout1, Vout2) connected to the selected high-side control unit (HSE1, HSE2) is still smaller than the respective target output voltage to be provided, and / or - for the discharging phase of a control cycle, if an actual output voltage is currently applied to all output terminals (Vout1, Vout2) which is smaller than the respective target output voltage to be provided, the high-side switch unit is selected which is connected to the output terminal (Vout1, Vout2) at which an actual output voltage is applied which is smaller than the respective target output voltage to be provided by the greater difference in amount, wherein the selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles, provided that the actual output voltage of the output terminal (Vout1, Vout2) connected to the selected high-side switch unit (HSE1, HSE2) continues to fall below the target output voltage to be provided by the difference which is greater in amount compared to the undershoot of the actual output voltages of the other output terminal or terminals (Vout1, Vout2), and / or - for the discharging phase of a control cycle, if an actual output voltage is currently applied to all output terminals (Vout1, Vout2) which is smaller than the respective target output voltage to be provided, the high-side switch unit (HSE1, HSE2) is selected which is connected to the output terminal (Vout1, Vout2) at which an actual voltage is applied, a degree of reduction compared to its value in the previous control cycle is greater than the degree of reduction of the actual output voltages at the other output terminals (Vout1, Vout2) compared to their respective values in the previous control cycle, wherein the selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles, and / or - for the discharging phase of a control cycle, if an actual output voltage is currently applied to all output terminals (Vout1, Vout2) which is smaller than the respective target output voltage to be provided, the high-side switch units (HSE1, HSE2) are selected alternately or cyclically or the respectively selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles before a high-side switch unit (HSE1, HSE2) is selected which is connected to another output terminal (Vout1, Vout2), the actual output voltage of which is smaller than the respective target output voltage to be provided.
2. The DC / DC converter for converting an input voltage into at least two desired output voltages, which can be different or the same within a predeterminable output voltage range, comprising - an input voltage terminal (Vin) for applying the input voltage, - at least two output half-bridges (AH1, AH2), each of which has an output terminal (Vout1, Vout2) at which a first or second output voltage can be tapped, a common connection node (VKAH), a common reference potential terminal (GND), a common low-side switch (M3) arranged between the connection node (VKAH) and the reference potential terminal (GND) and an electrical high-side switch unit (HSE1, HSE2) which is arranged between the common connection node (VKAH) and the respective output terminal (Vout1, Vout2), - an inductor (IND), which is arranged between the input voltage terminal (Vin) and the connection node (VKAH) of the output half bridges (AH1, AH2) and which can be charged with electrical energy in a charging phase of one of several successive clocked control cycles and can be discharged in a discharging phase of a control cycle for forwarding the electrical energy to one of the output terminals (Vout1, Vout2), - wherein each high-side switch unit (HSE1, HSE2) has a first electronic switch (M4a, M5a) and a second electronic switch (M4b, M5b) each with a diode (BGD4a, BGD4b, BGD5a, BGD5b) in parallel with the switch, wherein the two switches (M4a, M5a; M4b, M5b) of each high-side switch unit (HSE1, HSE2) are connected in series with their diodes in antiserial alignment, and the switch of each high-side switch unit (HSE1, HSE2), the diode of which is located in a reverse direction relative to a current flow from the connection node (VKAH) to the output terminal (Vout1, Vout2) connected to the relevant high-side switch unit (HSE1, HSE2), is the first switch (M4a, M5a) of the respective high-side switch unit (HSE1, HSE2) and the respective other switch is the second switch (M4b, M5b) of the respective high-side switch unit (HSE1, HSE2), and - an evaluation and control unit (AAE) which is configured to control the electronic switches of the output half-bridges (AH1, AH2) on the basis of the magnitude of the input voltage at the input voltage terminal (Vin) and the magnitudes of the actual and the target output voltages at the output terminals (Vout1, Vout2), wherein the evaluation and control unit (AAE) is configured to select, on the basis of a decision criterion based on a potential undershoot of a target output voltage or target electrical power to be provided at at least one of the output terminals (Vout1, Vout2) by the actual output voltage currently applied thereto or the actual electrical power provided, the high-side switch unit (HSE1, HSE2) connected to the output terminal (Vout1, Vout2) to which electrical energy stored in the inductor (IND) is to be transferred during the discharging phase of a control cycle. - wherein the evaluation and control unit (AAE) is configured to - in the charging phase of a control cycle - switch the low-side switch (M3) of the output half bridges (AH1, AH2) to be conducting, and - if not yet done at the end of the discharging phase of the previous control cycle and in the event that one of the two actual output voltages is smaller than the respective target output voltage associated therewith, select one of the two high-side switch units (HSE1, HSE2) for the current control cycle using the decision criterion, wherein this high-side switch unit is the selected high-side switch unit and the other high-side switch unit is the non-selected high-side switch unit, - switch the two switches of the non-selected high-side switch unit to be blocking or keep them switched to be blocking, - transfer the first switch (M4a, M5a) of the selected high-side switch unit from switched to be blocking to switched to be conducting or to keep this first switch (M4a, M5a) switched to be conducting if it is already switched to be conducting, and - switch the second switch (M5a, M5b) of the selected high-side switch unit to be blocking or, if it is already switched to be blocking, keep it switched to be blocking, - in an initial transition phase following the charging phase - keep the two switches of the non-selected high-side switch unit switched to be blocking, - keep the first switch (M4a, M5a) of the selected high-side switch unit switched to be conducting, - transfer the low-side switch (M3) of the output half-bridges (AH1, AH2) from switched to be conducting to switched to be blocking, and - then, after waiting for the dead time interval, transfer the second switch (M5a, M5b) of the selected high-side switch unit from switched to be blocking to switched to be conducting, - in the discharging phase of a control cycle following the first transition phase - keep the states of the electronic switches of the output half-bridges (AH1, AH2) assumed at the end of the first transition phase, and - in a second transition phase following the discharging phase, which is followed by the charging phase of the next control cycle, - transfer the low-side switch (M3) of the output half-bridges (AH1, AH2) from switched to be blocking to switched to be conducting, - then, after waiting for the dead time interval, transfer the second switch (M5a, M5b) of the selected high-side switch unit from switched to be conducting to switched to be blocking, and - if it is already to take place now and in the event that one of the two actual output voltages is smaller than the respective specified target output voltage, select one of the two high-side switch units (HSE1, HSE2) for the next control cycle on the basis of the decision criterion, and - keep the two switches of the non-selected high-side switch unit of the current control cycle switched to be blocking if the non-selected high-side switch unit of the current control cycle is not to be selected for the next control cycle in accordance with the decision criterion and the selected high-side switch unit of the current control cycle is also to be selected in the next control cycle, - or transfer the first switch (M4a, M5a) of the high-side switch unit selected in the current control cycle from switched to be conducting to switched to be blocking, and transfer the first switch (M4a, M5a) of the high-side switch unit not selected in the current control cycle from switched to be blocking to switched to be conducting, if the selected high-side switch unit of the current control cycle will not be selected for the next control cycle in accordance with the decision criterion and the non-selected high-side switch unit of the current control cycle will be selected, - wherein the evaluation and control unit (AAE) is configured to cause, when it has to select the high-side switch unit (HSE1, HSE2) on the basis of the decision criterion, that: - for the discharging phase of a control cycle, that high-side switch unit (HSE1, HSE2) is selected which is connected to that output terminal at which an actual output voltage is currently applied which is smaller than the target output voltage to be provided, wherein the selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles, provided that the actual output voltage of the output terminal (Vout1, Vout2) connected to the selected high-side control unit (HSE1, HSE2) is still smaller than the respective target output voltage to be provided, and / or - for the discharging phase of a control cycle, if an actual output voltage is currently applied to all output terminals (Vout1, Vout2) which is smaller than the respective target output voltage to be provided, the high-side switch unit is selected which is connected to the output terminal (Vout1, Vout2) at which an actual output voltage is applied which is smaller than the respective target output voltage to be provided by the greater difference in amount, wherein the selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles, provided that the actual output voltage of the output terminal (Vout1, Vout2) connected to the selected high-side switch unit (HSE1, HSE2) continues to fall below the target output voltage to be provided by the difference which is greater in amount compared to the undershoot of the actual output voltages of the other output terminal or terminals (Vout1, Vout2), and / or - for the discharging phase of a control cycle, if an actual output voltage is currently applied to all output terminals (Vout1, Vout2) which is smaller than the respective target output voltage to be provided, the high-side switch unit (HSE1, HSE2) is selected which is connected to the output terminal (Vout1, Vout2) at which an actual voltage is applied, a degree of reduction compared to its value in the previous control cycle is greater than the degree of reduction of the actual output voltages at the other output terminals (Vout1, Vout2) compared to their respective values in the previous control cycle, wherein the selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles, and / or - for the discharging phase of a control cycle, if an actual output voltage is currently applied to all output terminals (Vout1, Vout2) which is smaller than the respective target output voltage to be provided, the high-side switch units (HSE1, HSE2) are selected alternately or cyclically or the respectively selected high-side switch unit (HSE1, HSE2) remains selected for a maximum of a predeterminable number of control cycles before a high-side switch unit (HSE1, HSE2) is selected which is connected to another output terminal (Vout1, Vout2), the actual output voltage of which is smaller than the respective target output voltage to be provided.
3. The DC / DC converter according to claim 1 or 2, comprising a communication bus (KB) to which the evaluation and control unit (AAE) is connected and via which, in the case of connection of loads to the first and second output terminals (Vout1, Vout2), the evaluation and control unit (AAE) is configured to receive from the respective load a voltage request signal representing the magnitude of the target output voltage to be provided.
4. The DC / DC converter according to any one of claims 1 to 3, wherein the evaluation and control unit (AAE) has a comparator, in particular in the form of a Schmitt trigger, the two inputs of which are connected to the two output terminals (Vout1, Vout2) and which supplies at its output an output signal which is representative of the decision criterion as to which of the high-side switch units (HSE1, HSE2) is to be selected.
5. The DC / DC converter according to claim 1 or claim 3 or claim 4, if referred back to claim 1, wherein the evaluation and control unit (AAE) is configured not to select any of the high-side switch units (HSE1, HSE2) of the output half-bridges (AH1, AH2) when the actual output voltages are greater than or equal to the respective target output voltages, wherein the evaluation and control unit (AAE) is configured to perform the following steps: - in the first transition phase - keep the first switches (M4a, M5a) of the two high-side switch units (HSE1, HSE2) conductively switching or switched to be conducting or blockingly switching or switched to be blocking, - keep the second switches (M4b, M5b) of the two high-side switch units (HSE1, HSE2) switched to be blocking, - keep the low-side switch (M3) of the output half bridges (AH1, AH2) switched to be conducting, and - transfer the high-side switch (M1) of the input half-bridge (EH) from switched to be conducting to switched to be blocking, and - after waiting for the dead time interval, transfer the low-side switch (M2) of the input half-bridge (EH) from switched to be blocking to switched to be conducting, - in the discharging phase - keep the states of all switches (M1, M2, M3, M4a, M4b, M5a, M5b) of the input half-bridge (EH) and the output half-bridges (AH1, AH2) assumed at the end of the first transition phase, and - in the second transition phase - transfer the low-side switch (M2) of the input half-bridge (EH) from switched to be conducting to switched to be blocking, and - after waiting for the dead time interval, transfer the high-side switch (M1) of the input half-bridge (EH) from switched to be blocking to switched to be conducting.
6. The DC / DC converter according to any one of claims 1 to 5, wherein the evaluation and control unit (AAE) comprises a measuring device (SH) for supplying a measuring signal representing the magnitude of the current through the inductor (IND) and that the evaluation and control unit (AAE) is configured to terminate the charging phase of a control cycle when the measuring signal exceeds a threshold value representing a predeterminable maximum current.
7. The DC / DC converter according to claim 6, wherein the threshold value can be changed depending on whether one or each of the current actual output voltages is greater or smaller than the respective target output voltage and / or depending on the target value of a slope compensation control (SC).
8. The DC / DC converter according to any one of claims 1 to 7, wherein the first and second switches (M4a, M4b, M5a, M5b) of each high-side switch unit (HSE1, HSE2) and, if present, the high-side switch (M1) and the low-side switch (M2) of the input half-bridge (EH) are each designed as MOS transistor (M4a, M5a; M4b, M5b) with free-wheeling diode.
9. The DC / DC converter according to claim 8, wherein the source terminals of the two MOS transistors of each high-side switch unit (HSE1, HSE2) are connected to each other and the drain terminal of one MOS transistor (M4b) of each high-side switch unit (HSE1, HSE2) is connected to the low-side switch (M3) of the output half-bridges (AH1, AH2) and the drain terminal of the other MOS transistor (M5b) of each high-side switch unit (HSE1, HSE2) is connected to the respective output terminal (Vout1, Vout2) of the output half-bridges (AH1, AH2).
10. The DC / DC converter according to any one of claims 1 to 8, wherein the first and second switches (M4a, M4b, M5a, M5b) of each high-side switch unit (HSE1, HSE2) are designed as bipolar transistors or IGBTs, with a diode, in particular a Schottky diode, in parallel with each of their current paths.