This invention relates to a DC-to-DC converter.

The main application of this invention is in the field of electronics and, in particular, in the manufacture of electrical energy conversion systems for the automotive sector, more specifically in the design and manufacture of charging systems for electric batteries.

In the field of electric vehicles, in fact, there is a battery charger system configured to recharge two different battery packs, a high-voltage one, designed to power the vehicle traction unit, and a low-voltage one, designed to power the auxiliary and safety loads.

For this reason, once the alternating current has been converted into direct current, a DC-to-DC converter is required to reduce the level of the charging voltage.

This device also comes into play when energy is transferred directly from the high-voltage battery to low-voltage loads.

Modern DC-to-DC converters convert the voltage level by first carrying out a conversion from a DC voltage level to a high-frequency alternating voltage (AC) by means of an input circuit feeding a transformer in order to change the voltage level and create galvanic separation, then the output circuit rectifies the AC voltage in order to generate a DC voltage of a different level from the input DC voltage.

In addition, modern DC-to-DC converters perform the conversion using switching techniques, which significantly increase (compared to linear voltage regulators) the efficiency and reduce the space occupied by using smaller electronic components with fewer thermal dissipation issues.

Switching-type DC-to-DC converters (known as 'switched-mode DC-to-DC converters') temporarily store electrical energy in a magnetic (inductor, transformer) or electrical storage component (a capacitor) and then release the stored electrical energy at a different voltage level.

To perform efficient switching, electronic devices are used that switch with fast rise and fall times, such as power FET transistors. Typically, the converter structure comprises a high-voltage side comprising a first and second leg placed in parallel with each other and each having two MOSFETs connected in series, with the intermediate node between the two MOSFETs of the second leg connected to one end of a primary transformer winding.

There are also two blocking diodes placed in series on a third leg parallel to the first and second leg, with a connection node between the diodes connected to the other end of the transformer.

A resonant inductor is connected in series between the two connection nodes.

The low-voltage circuit, downstream of the transformer, consists mainly of two synchronous rectifiers (two MOSFETs), an output inductor and an output capacitor.

By appropriately switching the MOSFETs on or off, it is therefore possible to reduce the voltage from the input to the output, also reducing the switching losses; this, at low loads, is mainly made possible by the presence of the resonant inductor, which allows energy to be stored and used to charge/discharge the voltage of the nodes during switching to allow what is commonly known as zero voltage switching, i.e. switching that reduces the losses due to the switches switching on as much as possible.

Disadvantageously, however, the presence of the inductor and the third leg on which the blocking diodes are placed makes the component expensive and bulky, which runs totally counter to the current requirements of the automotive world, which is active in trying to limit the size and cost of the entire battery pack charging system as much as possible.

The purpose of this invention is, therefore, to provide a DC-to-DC converter that overcomes the above-mentioned drawbacks of the prior art. In particular, one purpose of this invention is to provide a structurally simple and compact DC-to-DC converter which is capable of maintaining high efficiency even at low loads.

Said purposes are achieved with a DC-to-DC converter having the characteristics of one or more of the following claims.

In particular, the DC-to-DC converter comprises an input circuit extending between two input terminals, an output circuit extending between two output terminals and a transformer stage operatively interposed between the input circuit and the output circuit and provided with a primary winding connected to the input circuit and a secondary winding connected to the output circuit.

The input circuit preferably comprises a first and a second leg parallel to each other and both provided with two switches electrically placed in series between which a first node or a second connection node is interposed, respectively.

The primary winding runs between a first and a second terminal connected to the first and second connection node respectively.

The output circuit preferably comprises a synchronous rectifier having at least two switches connected to the terminals of the secondary winding.

An output inductor operatively arranged in series with said synchronous rectifier is also, preferably, provided.

The DC-to-DC converter further comprises a control unit associated with the switches of the first and second legs of the input circuit and the synchronous rectifier of the output circuit.

This control unit is preferably configured to turn said switches on and/or off to define a sequence of electrical configurations in which the primary winding is charged or discharged to induce a voltage on the secondary winding and a corresponding current flowing in said secondary winding that is rectified by being directed into said output inductor.

Note that the sequence of configurations is performed over a predetermined switching period, which is inversely proportional to the switching frequency of the converter.

According to one aspect of this invention, the control unit is configured to detect, either directly or indirectly, one or more first quantities correlated to a value of current flowing in the primary winding and to modulate a duration of said switching period as a function of a value of said first quantity detected.

It should be noted that the expression "current flowing in the primary winding" preferably defines a quantity determined by the sum of a first contribution and a second contribution, in which:

- the first contribution corresponds to a magnetizing current, i.e., a current due to the non-ideal nature of the ferromagnetic core that, having a non-zero reluctance, requires a magnetizing current for the circulation of the magnetic flux];

- the second contribution corresponds to the current that would flow in the primary winding inside an ideal transformer.

Furthermore, the control unit is preferably configured to increase said duration of the switching period so as to:

- increase a peak value of the magnetizing current and/or;

- decrease a minimum current flowing in the output inductor, bringing said minimum value of the current flowing in the output inductor below zero.

Advantageously, in fact, the Applicant has found that the duration of the switching period has a significant impact on the value of the current flowing within the primary winding; by suitably modulating this parameter, it is possible to raise this value even at low loads, guaranteeing sufficient energy in the input circuit to eliminate the switching losses of the switches (preferably MOSFETs) under any output load condition.

In other words, thanks to the control of the switching frequency by the control unit, Zero Voltage Switching (ZVS) in the primary conversion bridge can be realized even in the absence of the resonant coil commonly used in converters of the prior art, simplifying the device and reducing both its size and cost.

Furthermore, an important aspect in terms of performance, the absence of the resonant inductor reduces the ohmic losses at high loads, which makes the subject of the invention highly advantageous in the whole field of application of the system.

Note that these "one or more first quantities" correspond to voltage and/or current values present in the input and output circuits, all of which impact the primary current value.

Secondary values are related to the value of the load at the output terminals, while primary values are inherent to the power supply.

The dependent claims, incorporated herein by reference, correspond to different embodiments of the invention.

Further features and advantages of this invention will become clearer from the indicative, and therefore non-limiting, description of a preferred, but not exclusive, DC-to-DC converter embodiment, as illustrated in the accompanying drawings:

- Figure 1 shows a circuit diagram of a DC-to-DC converter according to this invention;

- Figure 2 schematically shows the variation of the characteristic curves relating to the currents varying with the switching period;

- Figure 3 shows the contributions that determine the value of the primary current in the different configurations of the DC-DC converter according to this invention;

- Figure 4 shows a theoretical modeling of the converter in Figure 1 .

With reference to the appended figures, a DC-to-DC converter according to this invention is generically identified with the numerical reference 1 .

The converter 1 is of the "buck" type, i.e., a converter in which the power transfer occurs between a high-voltage input port and a low-voltage output port.

As mentioned above, in fact, the preferred application of the converter 1 that is the subject of the invention is in the automotive field and, in particular, in the transfer of low-voltage power to the LV battery pack, to auxiliary loads or emergency loads of the vehicle, all of which are supplied at a voltage significantly lower than the charging voltage of the traction battery pack (HV battery).

Examples of auxiliary loads and emergency loads are ABS, air conditioning, power steering, headlights, and passenger compartment lights.

In this regard, the converter 1 comprises an input circuit 2 running between two input terminals 2a, 2b and an output circuit 3 running between two output terminals 3a, 3b.

A transformer stage 4 is operationally interposed between the input circuit 2 and the output circuit 3.

The transformer stage 4 is provided with a primary winding 5 connected to the input circuit 2 and a secondary winding 6 connected to the output circuit 3, suitably dimensioned so that the coil ratio between the two windings results in a predetermined voltage drop.

For example, in the preferred embodiment, the coil ratio is between 1 :30 and 1 :15.

In the preferred embodiment, the input circuit 2 comprises a first leg 7 and a second leg 8 parallel to each other.

The first leg 7 is preferably provided with two switches 7a, 7b electrically placed in series with each other; in particular, the first leg 7 comprises a first switch 7a and a second switch 7b.

A first connection node 7c is interposed between the two switches 7a, 7b.

The second leg 8 preferably comprises two switches 8a, 8b placed electrically in series with each other; in particular, the second leg 8 comprises a first 8a and a second switch 8b.

A second connection node 8c is interposed between the two switches 8a, 8b.

In the preferred embodiment, the switches 7a, 7b, 8a, 8b are defined by respective transistors, preferably MOSFETs.

In the illustrated embodiment, the input circuit 2 also comprises an additional leg 12, parallel to the other two and provided with a capacitor 12a; this capacitor is a so-called 'snubber' because its function is to limit overvoltages to the fields of the input circuit 2. The capacitor in series with the transformer preferably has the function of preventing the circulation of direct current in the primary of the transformer. This series capacitor is not relevant to the operation of the converter but can also be removed if the control strategy is changed slightly from voltage mode to current mode.

The output circuit 3 instead comprises a synchronous rectifier 9 associated with the secondary winding 6 and an output inductor 10 operatively arranged in series with said synchronous rectifier 9.

In particular, the synchronous rectifier 9 preferably comprises two switches 9a, 9b connected to the terminals of the secondary winding 6.

In the embodiment illustrated, the secondary circuit further comprises an "active snubber" useful for reducing overvoltages in the switches 9a and 9b.

In this case too, the switches 9a, 9b are preferably defined by respective transistors, more preferably MOSFET ones.

Alternatively, the synchronous rectifier 9 could also comprise a topology without a central socket, replaced by an H-bridge topology analogous to that of the input circuit 2.

The converter 1 further comprises a control unit 11 associated with the switches 7a, 7b, 8a, 8b, 9a, 9b of the first 7 and second leg 8 of the input circuit 2 and the synchronous rectifier 9 of the output circuit 3.

The control unit 11 , preferably defined by a processor or micro-processor, is configured to turn the switches 7a, 7b, 8a, 8b, 9a, 9b on and/or off to define a sequence of electrical configurations in which the primary winding 5 is charged or discharged to induce a voltage on the secondary winding 6 and a corresponding current flowing in said secondary winding 6 that is rectified by being directed into said output inductor 10.

The sequence of configurations is performed over a predetermined switching period Ts, which is inversely proportional to the switching frequency fs. The switching frequency fs, therefore, refers to the frequency with which the control unit completes the sequence of configurations.

The switching period Ts can therefore be defined as the time interval between two successive instants at which converter 1 assumes the same operating configuration.

The control unit 11 is preferably configured to drive the switches 7a, 7b, 8a, 8b, 9a, 9b in a succession of four different operating configurations interspersed with respective dead time configurations.

The expression 'dead time configuration' refers to a configuration assumed by the converter 1 , and thus by the switches 7a, 7b, 8a, 8b, 9a, 9b, in the time interval (dead time) between two successive operating configurations (i.e. winding charge/discharge).

The control unit 1 is preferably configured to drive the switches 7a, 7b, 8a, 8b, 9a, 9b, cyclically, in a succession of at least four operating configurations.

More precisely, the operating configurations are at least a first, a second, a third and a fourth operating configuration.

In the first operating configuration, the first switch 7a of the first leg 7 and the second switch 8b of the second leg 8 of the input circuit 2 are on, while the second switch 7b of the first leg 7 and the first switch 8a of the second leg 8 of the input circuit 2 are off. In addition, the first switch 9a of the synchronous rectifier 9 is preferably on and the second switch 9b of the synchronous rectifier 9 is off.

In this configuration, there is a transfer of energy as the diagonal of the input circuit 2 is active and thus the input voltage is applied to the transformer. The current in the output inductor 10 and in the primary winding 5 of the transformer stage 4 is increasing. The magnetizing current has a negative slope and is decreasing.

In the second operating configuration, the first switch 7a of the first leg 7 and the first switch 8a of the second leg 8 of the input circuit 2 are on, while the second switch 7b of the first leg 7 and the second switch 8b of the second leg 8 of the input circuit 2 are off. In addition, both switches 9a, 9b of the synchronous rectifier 9 are preferably on.

In this configuration, there is a recirculating current phase as the voltage applied to the transformer stage 4 is zero. The secondary winding 6 of said transformer stage 4 is also short-circuited as the switches 9a and 9b are both on. The current flowing in the primary winding 5 remains almost constant, while the current flowing in the secondary winding 6 has a negative slope and is decreasing. The magnetizing current remains basically constant.

Note that the terms "positive" and "negative" are reported in this text in line with what is conventionally illustrated in Figure 4: when a quantity is considered positive, it means that it has the direction indicated by the arrow, while if it is opposite it is considered negative.

In the third operating configuration, the second switch 7b of the first leg 7 and the first switch 8a of the second leg 8 of the input circuit 2 are on, while the first switch 7a of the first leg 7 and the second switch 8b of the second leg 8 of the input circuit 2 are off. In addition, the first switch 9a of the synchronous rectifier 9 is preferably off and the second switch 9b of the synchronous rectifier 9 is on.

In this configuration, the course of the currents follows that of the first operating configuration. However, in this configuration the magnetizing current has a positive slope and is increasing up to a peak value Im.

In the fourth operating configuration, the second switch 7b of the first leg 7 and the second switch 8b of the second leg 8 of the input circuit 2 are on, while the first switch 7a of the first leg 7 and the first switch 8a of the second leg 8 of the input circuit 2 are off. In addition, both switches 9a, 9b of the synchronous rectifier 9 are preferably on.

In this configuration, the course of the currents follows that of the second operating configuration. The magnetizing current also tends to remain constant, around its peak value Im. In addition, the control unit 11 is preferably configured to drive the switches 7a, 7b, 8a, 8b, 9a, 9b in at least four dead time configurations, each interposed between two successive operating configurations.

A first dead time configuration provides for the first switch 7a of the first leg 7 of the input circuit 2 to be on, with the second switch 7b of the first leg 7 and both switches 8a, 8b of the second leg 8 of the input circuit 2 being off. In addition, the first switch 9a of the synchronous rectifier 9 is preferably on and the second switch 9b of the synchronous rectifier 9 is off.

In this configuration, the current of the primary winding 5, which is at its peak, simultaneously discharges the parasitic capacitance of the switch (MOSFET) 8a and charges the parasitic capacitance of the switch 8b and then flows (once this voltage swing phase is completed) to the body diode of the switch 8a and creates a lossless transition (ZVS). A second dead time configuration provides for the first switch 8a of the second leg 8 of the input circuit 2 to be on, with both switches 7a, 7b of the first leg 7 and the second switch 8b of the second leg 8 of the input circuit 2 being off. In addition, the first switch 9a of the synchronous rectifier 9 is preferably off and the second switch 9b of the synchronous rectifier 9 is on.

In this configuration, the current from the primary winding 5 simultaneously discharges the parasitic capacitance of the switch 7b and charges the parasitic capacitance of the switch 7a and then flows (once this voltage swing phase is complete) to the body diode of the switch 7b and creates a ZVS transition.

A third dead time configuration provides for the second switch 7b of the first leg 7 of the input circuit 2 to be on, with the first switch 7a of the first leg 7 and both switches 8a, 8b of the second leg 8 of the input circuit 2 being off. In addition, the first switch 9a of the synchronous rectifier 9 is off and the second switch 9b of the synchronous rectifier 9 is on.

In this configuration, the current of the primary winding 5, which is at its peak (in absolute value), simultaneously discharges the parasitic capacitance of the switch 8b and charges the parasitic capacitance of the switch 8a and then flows (once this voltage swing phase is complete) to the body diode of the switch 8b and creates a ZVS transition.

A fourth dead time configuration provides that the second switch 8b of the second leg 8 of the input circuit 2 is on, with the first switch 8a of the second leg 8 and both switches 7a, 7b of the first leg 7 of the input circuit 2 being off. In addition, the first switch 9a of the synchronous rectifier 9 is preferably on and a second switch 9b of the synchronous rectifier 9 is off.

In this configuration, the current from the primary winding 5 simultaneously discharges the parasitic capacitance of the switch 7a and charges the capacitance of the switch 7b, and then flows (once this voltage swing phase is complete) to the body diode of the switch 7a and creates a ZVS transition.

In the preferred embodiment, the control unit 11 is configured to set the first dead time configuration between the first and second operating configurations, the second dead time configuration between the second and third operating configurations, the third dead time configuration between the third and fourth operating configurations, and the fourth dead time configuration between the fourth and first operating configurations.

Thus, losses can advantageously be reduced while maintaining a safety margin that avoids the simultaneous turning on of both switches of the same leg.

According to one aspect of this invention, the control unit 11 is configured to detect, directly or indirectly, one or more first quantities correlated to a value of current flowing in the primary winding.

It should be noted that these "one or more first quantities" correspond to voltage and/or current values present in the input and output circuits, all of which impact the value of the primary current.

These first quantities are preferably defined by one or more of the following:

- Input voltage in the input circuit 2; - Output voltage in the output circuit 3;

- Output current flowing in the output circuit 3 (i.e., the current flowing in the output inductor 10);

- Voltage in the connection node of the first 7c or the second leg 8c, at least during the switching of the switches from a deactivated configuration to an active configuration (e.g. during idle times).

Other relevant parameters are specific to the device, such as the value of the inductances and the coil ratio at the transformer stage 4, but as they remain unchanged they are not measured during operation and their value is established at the design stage.

According to the invention, the control unit 11 is configured to modulate the duration of said switching period Ts according to the value of said one or more first quantities.

More specifically, given a predetermined sampling time, the control unit 11 is configured to determine the duration of the switching period at the current sampling instant as a function of the value of the first quantities detected at a preceding, preferably immediately preceding, sampling instant.

In a preferred embodiment, the control unit 11 is configured to modulate the duration of said switching period Ts as a function of the values of the first quantities detected, in each cycle, during at least one dead time, preferably at least two, plus preferably at least in the second and third dead times.

In addition, the control unit 11 is preferably configured to define, depending on the value of one or more of said first quantities detected, a current reference representing the minimum current value on the primary winding 5 necessary to switch on the switches 7a, 7b, 8a, 8b of the first 7 and second leg 8 of the input circuit 2 while minimizing or eliminating losses.

In other words, the control unit 11 defines a value of current flowing on the primary during the minimum switching necessary to obtain a condition of ZVS (Zero Voltage Switching) on the input circuit 2 according to the value of the first quantity.

As the load is reduced, in fact, the Applicant has noted that the current value on the primary winding tends to reduce below values that prevent the ZVS condition from being obtained; the control unit 11 is therefore configured to determine the minimum primary current value, i.e. the current reference, compatible with the ZVS condition.

The control unit 11 is thus configured to modulate the duration of the switching period Ts (and thus the switching frequency fs) so as to obtain, on the primary winding 5, a current value at least equal to said current reference.

Advantageously, in this way, i.e., by direct control of the switching period, it is possible to achieve loss-free switching even in the absence of a resonant inductor on the input circuit, thereby achieving the main purposes of the invention.

In a first embodiment, the control unit 11 is configured to modulate the duration of said switching period inversely proportional to the value of said first quantity, thus increasing the switching period (and consequently reducing the frequency) as the load decreases.

The control unit 11 is preferably configured to increase the duration of the switching period as the value of current flowing in the primary winding decreases when driving the converter in at least one of said dead time configurations, preferably in at least two of said dead time configurations, more preferably in the second and fourth dead time configurations.

In certain embodiments, the control unit 11 is configured to identify a low load condition and activate the switching period modulation only in said condition.

The control unit is preferably configured to detect a value of a second quantity representing a load applied to the output terminals, identify a low load condition if the value of the second quantity is less than a predetermined threshold value and activate a modulation mode of said switching period when the low load condition is identified.

The expression "quantity representing a load applied to the output terminals" preferably identifies a quantity or parameter proportional to the power absorbed by the low voltage battery or by the auxiliary or emergency powered loads.

This first quantity is preferably defined precisely by the power absorbed and/or a current flowing at a given voltage.

In the embodiment illustrated, therefore, the converter 1 includes a sensor for measuring the output current.

It should be noted that this second quantity may be partly or wholly analogous to one or more of the first quantities.

In other embodiments, the control unit 11 is configured to modulate the switching period only when the converter is in the dead time configurations and, preferably, in the dead time configurations of the "leading" leg, which in the embodiment illustrated is the first leg 7; in this embodiment, the dead time configurations in which the modulation of the switching period is particularly important are the second and fourth legs.

It should also be noted that, during said switching period, the current flowing within the output inductor 10 has a variable value between a maximum and a minimum value.

The control unit 11 , at least in the low load condition, is also configured, should it be necessary, to increase the duration of the switching period Ts by bringing said minimum value of the current flowing in the output inductor 10 below zero.

Advantageously, the Applicant has in fact understood how, since the minimum value of the current flowing in the output inductor is a subtrahend in the equation determining the value of the primary current, the attainment of negative minimum values contributes to increasing the value of the current flowing in the primary, which is a decisive parameter in enabling a loss-free switching of the switches (ZVS).

In this respect, it is preferably noted that the control unit 11 is configured to drive the two switches 9a, 9b of the synchronous rectifier 9 in opening and closing synchronously with the sign of the voltage of the secondary winding 6.

More preferably, the control unit 11 is configured to maintain each of the switches 9a, 9b of the synchronous rectifier 9 in the on condition even when the value of the current flowing in the secondary winding 6 falls below zero.

In other words, the control unit 1 is configured to modify the commonly known synchronous rectification, in which the switches are to be turned off (i.e. opened) for negative currents, by keeping the switches 9a, 9b closed even in this condition.

Advantageously, this allows the contribution of the current flowing in the output inductor to be made an addend in determining the value of the primary current.

The first leg 7, in fact, is the branch in which the ZVS transition does not occur by default.

For this reason, for the ZVS transition to take place, it has been estimated that during the second and fourth configurations of the dead time, the current flowing in the primary winding 5 is large enough to move the switching node from one rail to another before the next operating condition (third and first) begins.

The amount of energy required is related to the drain source capacitance of the switches 7a, 7b (MOSFET), but also to the general capacitance of the node Cnode_7tot between the switches 7a, 7b. This capacity is, for example, the capacity between the drain electrode of the switch 7b and the dissipative housing of the power module in which the switches 7a, 7b are located, commonly connected to the vehicle frame:

During the dead time, the primary current is almost constant. In the second and fourth dead time configurations, the minimum amount of current required to achieve a ZVS transition is

It should be noted that the primary current value li-iv minzvs is a known value, since Cnode7_tot is a design parameter and the dead time is a quantity set by the control unit 11 .

Thanks to the solution in accordance with this invention it is possible to ensure that the primary current IHV during the second and fourth dead time configurations remains at least equal to the minimum value to ensure a ZVS transition.

Referring to Figure 3, for example, a model of a transformer is illustrated in which the real transformer is modeled as an ideal transformer together with a magnetizing inductance separated by a leakage inductance.

It should be noted that, in reality, only the primary current IHV can be concretely measured, while its contributors related to the magnetizing inductance Im and transformer IT cannot be measured and are instrumental in describing the technical effect obtained by this invention.

During the fourth operating condition, the primary current IHV is positive. The voltage of the transformer is almost zero because it is short-circuited on both the primary and secondary sides.

It should be remembered that the terms "positive" and "negative" are included in this text in line with what is conventionally illustrated in figure 4: when a quantity is considered positive, it means that it has the direction indicated by the arrow, while if it is opposite it is considered negative.

As previously noted, during the fourth dead time configuration, the second switch 9b of the synchronous rectifier 9 is deactivated (i.e. turned off) and the transformer is no longer shorted to the secondary. Consequentially:

With II equal to the current flowing in the output inductor 10. Furthermore, according to Kirchoff's law:

Therefore, the difference between Im and II during the dead time must be greater than lHv_minzvs.

Observing the graphs shown in Figure 3, it can be seen how, at the end of the fourth operating configuration, the magnetizing current Im is at its maximum l _m , while the current IL is at its minimum lunin.

Consequently, during the fourth configuration of the dead time we have:

Therefore, if the ZVS transition is reached, it requires:

To reach the required value for Invminzvs it is therefore necessary to:

- increase the value for the peak of the magnetization current Im;

- decrease the minimum current Lmin flowing in the output inductor 10.

In this regard, preferably the first and second switches 9a, 9b of the synchronous rectifier 9 are switches, for which reason if the conditions require it, the current II can also assume negative values, becoming a positive contribution for Invminzvs.

In other words, by bringing the current II to negative values it is possible to achieve a ZVS transition on the switches of the primary circuit, in particular of the first leg 7.

In more detail, the peak of the magnetizing current Im is: where D is the ratio between the current rise time II and half of the switching period

It is usually derived from the "phase shift" (PS) control variable: PS

D

360

The peak-to-peak coil current AIL is equal to:

Thus, the minimum current lunin flowing in the output inductor 10 can be found as:

Considering that the minimum primary current required during the dead time iHVminzvs is a known value, it follows that:

From this equation it emerges that the only parameter that can be modified to reach the target value of IHVminzvs is the duration of the switching period Ts, since:

- louT,Vnv,VLv are conditions imposed from the outside;

- D is modified to achieve the DCDC power target required by the vehicle;

- k,L,L_m are fixed design parameters of the converter.

Thus, at low load, the converter can enter a frequency modulation mode to always obtain ZVS transition on the first leg 7.

For example, the right frequency for each operating point can be found by means of external sensors/detectors that can

- measure the voltage of the node during the dead time, or

- measure the value of the primary current IHV during the dead time, or

- measure the efficiency of the converter and, or

- measure in any other way an amount directly related to a ZVS transition of the first leg 7.

Alternatively, the correct switching frequency could be determined by calculating it directly from the exploded formula obtained for iHVminzvs. Thus, for all operating points of the converter, the switching period that allows the first leg 7 to operate in a fully ZVS condition would be:

In this case, no resonant energy sensing units or other sensors would be needed to achieve a ZVS transition.

The invention achieves its intended purpose and achieves important advantages.

In fact, the identification of the switching period (or switching frequency) as the determining parameter for obtaining a ZVS condition significantly increases the efficiency of the device while allowing a reduction in cost and size.

Furthermore, the adoption of synchronous rectification that keeps the switches on even for negative currents maximizes this effect, optimizing the device.