WO2015044846A2

WO2015044846A2 - Electronic half-bridge zeta converter and related method for operating an electronic half-bridge zeta converter and method for designing an electronic half- bridge zeta converter

Info

Publication number: WO2015044846A2
Application number: PCT/IB2014/064657
Authority: WO
Inventors: Daniele Luccato
Original assignee: Osram Gmbh; Osram S.P.A. - Societa' Riunite Osram Edison Clerici
Priority date: 2013-09-25
Filing date: 2014-09-19
Publication date: 2015-04-02
Also published as: EP3050400B1; CN105580497B; CN105580497A; WO2015044846A3; EP3050400A2; US20160211759A1; US9831784B2

Abstract

An electronic half-bridge ZETA converter comprises a transformer (T), wherein a half-bridge (S₁, S₂) is connected to the primary winding (T1) of transformer (T), and wherein a respective capacitance (C_A1, C_A2) and a respective diode (D_A1, D_A2) are associated with the half-bridge switches (S₁, S₂) · Moreover, the converter comprises a ZETA converter which is connected to secondary winding (T2) of transformer (T), so that the ZETA converter comprises a first inductance, which includes the magnetization inductance (L_M) of transformer (T), and a second inductance (L₀). Finally, the converter comprises a control unit (112) which drives the half-bridge switches (S₁, S₂) with four time intervals that are repeated periodically. Specifically, during the fourth time interval the first (S₁) and the second switch (S₂) are opened, so that the capacitance (C_A2) associated with said second switch (S₂) is charged and the capacitance (C_A1) associated with said first switch (S₁) is discharged, enabling a zero voltage switching. According to the present disclosure, the first (L_M) and said second inductance (L₀) are dimensioned such that said inductances (L_M, L₀₎ provide, during said fourth time interval, a current which contributes to the charging of capacitance (C_A2) associated with said second switch (S₂) and to the discharging of capacitance (C_A1) associated with the first switch (S₁).

Description

"Electronic half-bridge ZETA converter and related method for operating an electronic half-bridge ZETA converter and method for designing an electronic half- bridge ZETA converter"

Technical field

The present disclosure relates to electronic converters .

This disclosure was devised with specific attention paid to improve the driving of a half bridge ZETA converter.

Description of the related art

Electronic converters for light sources comprising, for example, at least one LED (Light Emitting Diode) or other solid-state lighting devices, may supply a direct current output. Such a current may be steady or else may vary in time, for example, in order to set the brightness of the light emitted by the light source (so-called dimming function) .

Figure 1 shows a possible lighting system, comprising an electronic converter 10 and a lighting module 20 comprising, for example, at least one LED L.

For example, Figure 2 shows an example of a lighting module 20 comprising, for example, a LED chain, i.e. a plurality of LEDs connected in series. For example, in Figure 2 four LEDs Li, L₂, L₃ and L are shown .

Electronic converter 10 usually comprises a control circuit 102 and a power circuit 12 (for example a switching supply AC/DC or DC/DC) , which receives as input a supply signal (for example from the mains) and supplies as output, through a power output 106, a direct current. Such current may be steady or may vary in time. For example, control circuit 102 may set, via a reference channel I_{re f} of power circuit 12, the current required by LED module 20.

For example, such a reference channel I_{re f} may be used to regulate the intensity of the light emitted by lighting module 20. In fact, generally, a regulation of the light intensity emitted by LED module 20 may be achieved by regulating the average current flowing through the lighting module, for example by setting a lower reference current I_{re f} or by switching off power circuit 12 through a Pulse Width Modulation (PWM) signal .

However, if module 20 is supplied with a regulated voltage, i.e. if converter 12 is a voltage generator, it is typically necessary to arrange a current regulator connected in series with light sources L, in order to limit the current. In this case, the dimming function may also be implemented via such a current regulator, for example:

a) either by selectively activating or deactivating such a current regulator through a driving signal, e.g. a PWM signal,

b) if an adjustable current regulator is used, by setting the reference current of such a current regulator .

Generally speaking, there are many types of electronic converters, which are divided mainly into isolated and non-isolated converters. For example, non^¬ isolated converters are "buck" , "boost" , "buck-boost" , "Cuk", "SEPIC" and "ZETA" converters, while isolated converters are "flyback" , "forward" , "Half-bridge" and "Full-bridge" converters. Such converter types are well known to the skilled in the art.

For example, Figure 3 shows a circuit diagram of a ZETA converter which operates as a DC/DC converter. The person skilled in the art will appreciate that an input AC current may be converted into a direct current via a rectifier, for example a diode-bridge rectifier, and possibly a filtering capacitor.

Substantially, a ZETA converter comprises an electronic switch S, a diode D, two inductors Lo and Li and two capacitors Co and Ci, and therefore represents a fourth-grade converter.

In the presently considered example, converter 12 receives as input, via two input terminals 110, a voltage V_in and supplies as output, via two output terminals 106, a regulated voltage V₀ or a regulated current i₀.

Specifically, the first terminal of input 110 is connected through switch S to a first terminal of inductor Li and the second terminal of input 110 is connected directly to the second terminal of inductor Li and represents ground GND.

In the considered example, the first terminal of inductor Li is connected through capacitor Ci to the cathode of diode D, and the second terminal of inductor Li is connected directly to the anode of diode D. The anode of diode D, i.e. the second terminal of inductor Li, is also directly connected to output 106, specifically to the second terminal of output 106.

The cathode of diode D is connected via the second inductor Lo to the first terminal of output 106. Finally, capacitor Co is connected in parallel with output 106, i.e. directly to the terminals of output 106.

In the considered example, a load R_L is connected to such an output 106, which for example may be the previously described lighting module 20.

As previously mentioned, the control may take place in current or in voltage. To this purpose a control unit 112 is typically used which drives switch S so that the output voltage V₀ or the output current i₀ is set to a desired value, such as for example the reference current I_ref- To this purpose it is possible to use, in a manner known per se, a sensor adapted to detect current i₀ or voltage V₀.

Referring to Figures 4a and 4b, a possible driving of such a ZETA converter will be described.

Specifically, as shown in figure 4a, during a first operation interval switch S is closed and diode D is OFF, i.e. diode D is inversely biased. In this case, inductor Li saves the energy received from the input and capacitor Ci supplies energy, through output switch L₀ and capacitor Co, to load R_L. In this condition, the currents flowing through inductances Li and Lo increase in a substantially linear way, while no current flows through diode D.

On the contrary, as shown in Figure 4b, during a second operating interval switch S is open and diode D is ON. In fact, diode D is directly biased, because the bias of the voltage across inductance Li is inverted. In this condition, the currents flowing through inductors Li and Lo decrease in a substantially linear way. Specifically, the energy saved in inductance Li is mainly transferred to capacitor Ci and the load receives energy mainly from inductor Lo. Therefore, the current flowing through diode D is:

Details about the operation of such a ZETA converter are described for example in the paper by Huai WEI, et al . , "Comparison of basic converter topologies for power factor correction" , IEEE Proceedings Southeastcon '98, p. 348 - 353, 24-26th April 1998, Orlando, FL, the content whereof is incorporated herein as a reference. Figure 5 shows an alternative embodiment of a ZETA converter, wherein the arrangement of capacitor Ci and of inductor Lo is different.

In the presently considered embodiment, the first terminal of inductor Li is connected directly to output 106, specifically to the first terminal of output 106. On the contrary, the second terminal of inductor Li is connected, through capacitor Ci, to the anode of diode D, wherein the cathode of diode D is connected directly to the first terminal of output 106, i.e. to the first terminal of inductor Li.

In the considered embodiment, the anode of diode D is connected through inductor Lo to the second terminal of output 106.

Finally, in this case, too, capacitor Co is connected in parallel to output 106, i.e. directly to the terminals of output 106.

Therefore, in the presently considered embodiment, the positions of capacitor Ci and of inductor Li have changed: in Figure 3 they were connected in series between the first terminal of inductor Li and the first output terminal, while capacitor Ci and inductor Lo are connected in series between the second terminal of inductor Li and the second output terminal in Figure 5.

However, the general operating principle is substantially unaltered.

Typically, both operating intervals are repeated periodically with a fixed frequency, wherein the energy transfer is controlled via a PWM signal, i.e. the duration of the first and of the second interval are variable, while the sum of the durations is constant.

The skilled in the art will appreciate that such a PWM driving and the control of the durations of the operating intervals are well known and can be implemented, for example, through a feedback of the output voltage or current through an error amplifier. For example, in the case of a control by current, the duration of the first interval is increased until the (average) output current reaches a predetermined threshold.

In the known state of the art it has been moreover proposed to drive such a ZETA converter with a sort of driving called "soft switching", wherein switch S is switched when the voltage across switch S is zero (zero voltage switching, ZVS) . As a matter of fact, this sort of driving may reduce the switching losses and the electromagnetic interference (EMI) .

For example, the paper by TSAI-FU, et al . , "Design optimization for asymmetrical ZVS-PWM zeta converter" , IEEE Transactions on Aerospace and Electronic Systems, Vol. 39, Iss. 2, p. 521 - 532, April 2003, the content of which is incorporated herein as a reference, describes the use of an active clamp to such a purpose.

Specifically, as shown in Figure 2 of said paper, a ZETA converter may be transformed into an isolated ZETA converter by replacing inductor Li with a transformer T comprising a primary winding Tl and a secondary winding T2. Specifically, transformer T may be modelled as an ideal transformer with a given turn ratio l:n, an inductance Lm connected in parallel with primary winding Tl, representing the magnetization inductance of transformer T, and an inductance Lr connected in series with primary winding Tl, modelling the leakage inductance. Moreover, switch S may also be arranged between the primary winding of the transformer and ground, which enables the use of an N-MOS transistor. Subsequently, an "active clamp" comprising an electronic switch (S2 in the paper) and a capacitor (capacitor Cc) is added to said isolated ZETA converter. According to the teaching of the above- mentioned paper, switches SI and S2 are then driven with 4 driving modes, which are shown in Figures 3 and 4 of said paper.

Essentially, the paper by WU TSAI-FU describes that such a converter is adapted to switch the switches SI and S2 at zero voltage and the diode at zero current. Specifically, in section "F. Selection of Resonant Inductor and Clamping Capacitor" there is stated that the zero voltage switching may be obtained if leakage inductor Lr of the transformer has a stored energy which is sufficient to discharge the capacitance of switches and other capacitances (see equation (35) of the paper) .

On the basis of such considerations, the paper indicates two relations (equations (37) and (39) of the paper) which enable to determine the minimum value for the leakage inductance of the transformer, as a function of the current flowing through switches SI and S2, the input voltage, the duty cycle and the value of capacitance Cr.

Moreover, the paper proposes to dimension the clamping capacitor (Cc in the paper) as a function (see equation (41) of the paper) of the leakage inductance, of the duty cycle and of the switching frequency, so that a half-period of the resonance frequency of the leakage inductance Lr and of the clamping capacitor is higher than the maximum time during which switch SI is open .

For example, in the article a leakage inductance is used which equals 20 μΗ and a clamping capacitor with 0.22 ]iF.

However, the inventor has observed that this kind of dimensioning of components has a number of drawbacks. In fact, the dimensioning of capacitor Cc causes the resonance current of leakage inductance Lr and of clamping capacitor Cc to be different from zero when switch SI is opened or when switch S2 is closed. Therefore, in this case, the diode of the ZETA converter does not switch at zero current, which is visible for example in Figure 14 of the paper. Moreover, as described at "C. Selection of power switches and diode", switches must have a low capacitance, because otherwise the zero voltage switching of switches might be lost, or the leakage inductance of the transformer should be very high. Therefore, a typical value for the capacitance of switches is 42 pF.

Object and Summary

The object of the invention is to overcome the drawbacks outlined above.

According to the invention, this object is achieved thanks to an electronic half-bridge ZETA converter having the features set forth in the claims that follow. The claims also concern a related method for operating an electronic half-bridge ZETA converter, and a corresponding method for designing an electronic half-bridge ZETA converter.

The claims are an integral part of the technical teaching provided herein with reference to the invention.

As previously mentioned, the present description concerns an electronic half-bridge ZETA converter supplying a power signal through an output.

In various embodiments, the electronic half-bridge ZETA converter comprises a transformer with a primary winding and a secondary winding, wherein each winding comprises at least a first and a second terminal. The converter comprises moreover a half-bridge, i.e. a first and a second electronic switch, which is configured to selectively connect the first terminal of the primary winding to a supply signal or to ground, wherein a respective capacitance and a respective diode are associated to each switch of the half-bridge. Finally, the converter also comprises, on the primary side, at least one capacitor which is connected between the second terminal of the primary winding and the supply signal and/or ground.

In various embodiments the converter comprises, on the secondary side of the transformer, a ZETA converter. Specifically, in various embodiments, such a ZETA converter comprises three branches which are connected in parallel, wherein:

a) the first branch comprises a first capacitor, which is connected in series with the secondary winding of the transformer, and therefore the first branch comprises the magnetization inductance of the transformer (and possible further inductors connected in parallel with the primary and/or the secondary winding of the transformer) and the first capacitor, b) a second branch comprising a diode, and

c) a third branch comprising a second capacitor, connected in series with a second inductance, wherein the output is connected in parallel with the second capacitor .

In various embodiments, the switching of the half- bridge switches is driven by a control unit. For example, in various embodiments, the control unit drives the first and the second electronic switches with four time intervals which are repeated periodically:

a) a first time interval, wherein the first switch is closed and the second switch is open, so that a current flow from the supply signal flows through the primary winding of the transformer and such a current flow increases; b) a subsequent second time interval, wherein the first switch is open and the second switch is open;

c) a subsequent third time interval, wherein the first switch is open and the second switch is closed; and

d) a subsequent fourth time interval, wherein the first switch is open and the second switch is open.

Specifically, in various embodiments, the capacitance associated to the first switch is charged and the capacitance associated to the second switch is discharged during the second interval, and vice versa the capacitance associated to the first switch is discharged and the capacitance associated to the second switch is charged during the fourth interval, which allows a switching of the switches at zero voltage.

Specifically, while in the paper by WU TSAI-FU the leakage inductance contributed to the discharging the capacitance associated to the first switch and to the charging the capacitance associated to the second switch during the fourth time interval, according to the present disclosure such a function is now performed via the inductances of the ZETA converter, i.e. the magnetization inductance of the transformer (and possible other inductances connected in parallel with the primary and/or the secondary windings of the transformer) , and the inductance of the third branch of the ZETA converter are dimensioned in such a way that such inductances supply, during the fourth time interval, a current which charges the capacitance associated with the second switch and discharges the capacitance associated to the first switch.

Thanks to such a sizing, the equivalent capacitance at the intermediate point of the half- bridge comprising the capacitances associated to the switches of the half-bridge may also range between 200 pF and 1 , 5 nF .

As a matter of fact, by knowing such an equivalent capacitance, inductances may be dimensioned as a function of the input voltage, of the duty cycle of the converter, i.e. the duration of the first time interval with respect to the duration of the switching period, of the output current supplied through the output, and the equivalent capacitance of the intermediate point of the half bridge.

In various embodiments, during the third time interval an oscillation of a resonant circuit occurs which comprises the capacitor (s) connected to the second terminal of the primary winding, the capacitor connected in series with the secondary winding of the transformer and the leakage inductance of the transformer. In various embodiments, such a resonant circuit is dimensioned so that:

a) during the third time interval there are present one or various complete half periods of the resonance of such a resonant circuit, and

b) the current flowing through the diode is zero at the end of the third time interval.

Therefore, thanks to such a sizing, also the diode of the ZETA converter may be switched at zero current.

Brief description of the annexed views

The invention will now be described, by way of non-limiting example only, with reference to the enclosed views, wherein:

- Figures 1 to 5 have already been described in the foregoing,

- Figure 6 shows an electronic half-bridge ZETA converter according to the present description; and

- Figures 7 to 12 show details of embodiments of a driving of the converter of Figure 6. Detailed description of embodiments

In the following description numerous specific details are given to provide a thorough understanding of embodiments. The embodiments can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, materials or operations are not shown or described in detail to avoid obscuring aspects of the embodiments.

Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner in one or more embodiments.

The headings provided herein are for convenience only and do not interpret the scope or meaning of the embodiments .

Substantially, also the electronic ZETA converter of the present description is driven via a half bridge, and therefore this topology is typically called a half- bridge ZETA converter.

In this topology, the converter comprises a transformer T comprising a primary winding Tl and a secondary winding T2. Specifically, transformer T may be modelled as an ideal transformer with a given turn ratio l:n, an inductor L_M which represents the magnetization inductance of transformer T and an inductor L_R which represents the leakage inductance of transformer T. The connection on the secondary side of the transformer corresponds to a typical ZETA converter, i.e. a diode, two capacitors and an inductor, because the second inductor may be represented by the magnetization inductance of the transformer.

On the contrary, on the primary side of the transformer no single switch is used, but a first terminal of the primary winding of the transformer is connected to a half bridge, i.e. at the intermediate point between two switches, and therefore such a half bridge is configured to selectively connect such first terminal of the primary winding of transformer to the input voltage or to ground. Moreover, the second terminal of the primary winding of the transformer is connected through at least a capacitor to the input voltage and/or to ground.

For example, Figure 6 shows a possible embodiment of an electronic half-bridge ZETA converter.

Also in this case, converter 12 receives as an input, via two input terminals 110, a supply signal, such as for instance a DC voltage V±_n, and supplies as an output, via two output terminals 106, a regulated voltage V₀ or a regulated current i₀.

In the presently considered embodiment, converter 12 comprises a half bridge, i.e. two electronic switches Si and S₂ which are connected in series between the two input terminals 110, wherein the switching of electronic switches Si and S₂ is driven via a control unit 112. For example, the control unit 112 may be an analogue and/or digital circuit, such as for example a micro-processor which is programmed via a software code.

For example, in the embodiment such electronic switches Si e S₂ are N-MOS transistors or n-channel MOSFETs (Metal-Oxide-Semiconductor Field-Effect

Transistors) .

In the presently considered embodiment, such switches Si and S₂ have respective capacitances C i, C_A2 and respective diodes D_Ai, D_A2 connected in parallel. For example, capacitances C_Ai and C ₂ may represent the intrinsic capacitances of a MOSFET and/or may be implemented with additional capacitors, which are connected in parallel with switches Si e S₂.

Specifically, according to the present disclosure, the capacitance of switches Si e S₂ may even be remarkably higher than the value 42 pF mentioned in the paper by WU TSAI-FU. For example, the sum of capacitances C_Ai and C ₂ of switches SI and S2 can even reach 1.5 nF. For example, in the case MOSFETs are used, the capacitances of switches Si and S₂ may reach 100-200 pF, but normally it is necessary to add parallel capacitances, in order to slow down the switching leading edges and to reduce EMIs, and therefore the total capacitance of the node between both switches Si e S₂ may reach 1.5 nF, preferably between 200 pF and 1 nF.

On the contrary, diodes D_Ai and D_A2 may represent the body diodes of a MOSFET and/or may be implemented through additional diodes.

For example, in the considered embodiment, a first terminal of primary winding Tl of transformer T is connected directly to the intermediate point between both electronic switches Si and S₂. On the contrary, the second terminal of primary winding Tl of transformer T is connected via a capacitor C_RP to the second input terminal, which represents a first ground GNDi . Therefore, switches Si e S₂ may be used to selectively connect the first terminal of primary winding Tl of transformer T to voltage Vi_n or to ground GNDi .

Generally, on the secondary side of the transformer, the ZETA converter comprises three branches which are connected in parallel:

- the first branch comprises the secondary winding of transformer T2 (and therefore magnetization inductance L_M) and a capacitor C_RS;

- the second branch comprises diode D; and

- the third branch comprises inductor Lo and a capacitor Co, wherein load R_L, i.e. output 106, is connected in parallel with capacitor Co.

In the presently considered embodiment, the connection on the secondary side of transformer T substantially corresponds to what is shown in Figure 5. However, the diagram in Figure 3 might be used as well.

Specifically, in the presently considered embodiment, the first terminal of secondary winding T2 of transformer T is connected directly to output 106, specifically to the first terminal of output 106. On the contrary, the second terminal of secondary winding T2 is connected through a capacitor C_RS (corresponding to capacitor CI in Figure 5) to the anode of diode D, wherein the cathode of diode D is connected directly to the first terminal of output 106.

In the presently considered embodiment, the anode of diode D is connected via output inductor Lo to the second output terminal 106, which represents a second ground GND₂ , which due to the isolating effect of transformer T is preferably different from ground GNDi and therefore is denoted with a different ground symbol .

Finally, also in this case, capacitor Co is connected in parallel to output 106, i.e. directly to the terminals of output 106. In the presently considered embodiment, the leakage inductance L_R of transformer T is shown on the secondary side of transformer T, and the magnetization inductance L_M is shown on the primary side of transformer T. Generally speaking, such inductances L_M and L_R may also comprise additional inductors, such as for example an inductor connected in parallel with the primary and/or the secondary winding of transformer T, and/or an inductor connected in series with the primary and/or secondary winding of transformer T.

Therefore, from the point of view of the circuits, the circuit shown in Figure 6 may be compared with the circuit shown in the paper by WU TSAI-FU. In fact, according to the connection of the second terminal of primary winding Tl to ground and not to voltage V±_n, switch Si anyway performs the function of controlling the energy flow from supply Vi_n through the primary side of transformer T. Moreover, the serial connection of capacitors C_RS and C_RP performs the function of the clamping capacitor Cc described in the paper by WU TSAI-FU.

However, while the zero voltage switching of half- bridge switches was obtained, in the paper by WU TSAI- FU, thanks to the dimensioning of clamping capacitance Cc and of the leakage inductance Lr, the inventor has observed that the zero voltage switching of half-bridge switches may be obtained thanks to an energy flow coming from the magnetization inductance L_M and the output inductance L₀.

A possible embodiment of the driving of switches

Si and S₂ of Figure 6 will be described in the following .

In the presently considered embodiment, it is assumed that the turn ratio of primary winding Tl is the same as the turn ratio of secondary winding T2, i.e. n = 1, which makes the mathematical processing simpler. However, the person skilled in the art will appreciate that the exemplary case may also be applied to the general case wherein the turn ratio differs from each other.

Figure 7 shows in this respect the wave forms of some signals during these driving modes. Specifically, Figures 7a-7f respectively show:

a) the driving signal V_G1 for switch Si and the driving signal V_G∑ for switch S₂,

b) the voltage V_Ds2 across switch S2,

c) the current I_DS2 which corresponds to the sum of the currents flowing through switch S₂, capacitance C_A2 and diode D_A2,

d) the current I_DSi corresponding to the sum of the currents flowing through switch Si, capacitance C_Ai and diode D_Ai,

e) the current I_D flowing through diode D, and f) the voltage V_CRS across capacitance C_RS .

Specifically, at a time tO, switch Si is closed and switch S₂ remains open. Moreover, diodes D_Ai, D_A2 and D are open during this operating mode (Ml) . The corresponding equivalent circuit of this driving mode is shown in Figure 8a.

Specifically, as it will become clear in the following, in the presently considered embodiment switch Si is closed at zero voltage (ZVS) .

Substantially, during this driving mode, a voltage corresponding to V_IN - V_CRP, i.e. the voltage across capacitor C_RP, is applied to primary winding Tl of transformer T, i.e.

Therefore, the same voltage is also applied to secondary switching T2 of transformer T, because n = 1: V_T2 = VTI (3) This voltage across secondary winding T2 and voltage V_CRS across capacitor C_RS are applied to inductor L₀ and to output 106, i.e. Co and R_L.

Typically, inductance L_R is much smaller than inductance L₀, i.e. L_R << L₀, and therefore the voltage across inductance L_R may be neglected. Therefore, the current I_Lo flowing through inductor Lo corresponds during this mode substantially to:

(V_T9 + V_RPQ - V )

I_L0(t) = I_LO(tO) + ^ °- (t - tO) (4) i. e., during this operating mode current I_Lo which flows through inductance Lo increases in a substantially linear way, starting from the value of current I_Lo(tO) at time tO, which is also visible in Figure 7d, showing current I_DSi flowing through switch Si.

At a time tl, switch Si is opened and switch S2 still remains open. Therefore, at time tl, current I_Lo flowing through inductor Lo has reached the following value :

(V_T9 + V_RPQ - V )

J_i0(tl) = J_iO(t0) + ^ °- (tl - tO) (5)

Moreover, diodes D_Ai, D_A2 and D are still open during this operating mode (M2) . The corresponding equivalent circuit diagram of this driving mode is shown in Figure 8b.

In the presently considered embodiment, current I_P flowing through the primary side of transformer T, i.e. current I_LM flowing through magnetization inductance and the current flowing through the ideal primary winding Tl (i.e., for example, the current across inductor Lo) , also flows through capacitances C_Ai and CA2 :

Ip = ILM + ILO (6) Therefore, during this operating mode, capacitor C_A2 is discharged, and voltage V_DS2 across switch S2 falls to zero thanks to that current. For example, as shown in Figure 7b, voltage V_DS2 across switch S₂ (and therefore also across capacitance C_A2) decreases.

At time t2 voltage V_DS2 across switch S₂ reaches zero. Therefore, in this operating mode (M3), at this point current I_P flowing through the primary side of transformer T only flows through diode D_A2 and the voltage V_DS2 across switch S₂ stays set to zero, which creates the condition for a zero voltage switching of switch S2. Moreover, during this operating mode diode D is closed and the current of the resonant circuit comprising inductance L_R and capacitors C_RS and C_RP connected in series, and also the current flowing through inductor Lo start flowing through diode D (see for example Figure 7e) :

The corresponding equivalent circuit diagram of this driving mode is shown in Figure 8c.

At time t3 switch S₂ is closed and switch Si stays open. Therefore, in this operating mode (M4) diodes DAI _A DA2 are open and diode D stays active. The corresponding equivalent circuit diagram of this driving mode is shown in Figure 8d.

Substantially, in this operating mode, driving signal V_G∑ closes switch S₂ and, thanks to the fact that voltage V_DS2 has been set to zero through diode D_A2, switching is performed at zero voltage.

Moreover, the resonant frequency of the current flowing through diode D substantially corresponds to: fr = ^= (8) wherein C_R corresponds to capacitance C_RP connected in series with capacitance C_RS . Substantially, capacitor C_RS is charged again during this mode, which is shown in Figure 7f, and the magnetization current in the magnetization inductance L_M increases.

Therefore, at time t4 the current flowing through diode D reaches zero and diode D is switched off, while switch S₂ stays closed, switch Si stays open, and in this operating mode (M5) , too, diodes D_Ai, D_A2 are open. The corresponding equivalent circuit of this driving mode is shown in Figure 8e.

Specifically, the inventor has observed that diode D opens when the current flowing through inductor Lo equals the resonance current of the resonant circuit, comprising inductance L_R and capacitors C_RS and C_RP .

Nevertheless, there may exist situations, typically at a low duty cycle, wherein the magnetizing current on the primary may still be negative when the resonance has ended.

This is shown for example in Figures 9a-9e, which respectively show:

b) the voltage V_Ds2 across switch S2,

c) the current I_DSi which corresponds to the sum of the currents which flow through switch Si, capacitance

C_AI and diode D_Ai,

d) the current I_DS2 which corresponds to the sum of the currents flowing through switch S₂, capacitance C_A2 and diode D_A2, and

e) the current I_D flowing through diode D.

Therefore, in this situation, current I_Lo which flows through inductor Lo will no longer flow through diode D, but it will start flowing through capacitor C_RS, which is shown in Figure 10. However, such a current will also impose a current on leakage inductance L_Ri bringing its voltage from negative to zero .

This is shown for example in Figures lla-llc, which respectively show:

a) the voltage V_LR across inductance L_R,

a) the voltage V_CRS across capacitor C_RS, and voltage V_CRP across capacitor C_RP, and

c) the current I_D which flows through diode D, the current I_Lo which flows through inductor Lo, and the current I_LR which flows across inductance L_R.

Specifically, at time t4 currents I_L0 and I_LR are equal and opposed, and therefore current I_D flowing through diode D is zero (see Figure 11c) and current I_L0 takes voltage V_LR across inductance L_R to zero (see Figure 11a), i.e.:

and, neglecting the voltage across leakage inductance L_R, voltage V_D across the diode is:

As a consequence, the current will no longer flow through diode D, but it will charge capacitor C_RP and discharge capacitor C_RS . However, at a time t4' the voltages across both capacitors are the same, i.e. voltage V_T1 across primary winding Tl is equal to the voltage across secondary winding T2, to the voltage across capacitor C_RP, and to the voltage across capacitor C_RS :

and therefore the voltage across diode D is zero:

V_D = 0

As a consequence, from time t4', current I_L0 may flow again across diode D, but now the circuit is the same as at time t3 and inductance L_R with capacitors C_RP and C_RS may start oscillating again, and therefore a new oscillation will start; i.e., during the third time interval there may be present one or several half periods of the resonance of such an oscillation.

At time t5, switch S₂ is opened, while switch Si remains open. Moreover, diodes D_Ai, D_A2 and D are open in this operating mode (M6) . The corresponding equivalent circuit diagram of this driving mode is shown in Figure 8f. In this context, the dimensioning of components and the switching period should ensure that at time t5 there is no more orcillation, as shown in Figures 9 and 11, and current I_P through the primary winding has become positive.

Therefore, during this first operating mode, current I_P on the primary side of transformer T discharges capacitance C_Ai and charges capacitance C_A2 until the voltage across switch S₂ reaches the input voltage Vi_n at a time t6, which is shown in Figures 7b- 7d.

In fact, at time t6, diode D_Ai is closed and voltage V_DS2 corresponds to input voltage V±_nr which generates the condition for a zero voltage switching for switch Si. Diodes D_A2 and D stay open during this operation mode (M7) . The corresponding equivalent circuit diagram of this operating mode is shown in Figure 8g.

Therefore, in various embodiments, the control unit drives the half-bridge switches with the following steps, which are periodically repeated:

- during a first time interval Atl switch Si is closed and switch S₂ is open (times tO-tl);

- during a second time interval At2 switch Si is open and switch S₂ is open (times tl-t3); wherein, during said second time interval At2 , capacitance C_Ai associated to switch S₂ is charged and capacitance C₂ associated to switch S₂ is discharged, creating the condition for a zero voltage switching of switch S₂; - during a third time interval At3 switch Si is open and switch S2 is closed (times t3-t5); wherein, during said third time interval At3 , the magnetization current of the magnetization inductance increases and diode D of the ZETA converter is switched off at zero current; and

- during a fourth time interval At switch Si is open and switch S2 is open (times t5-t0); wherein, during such a time interval AtA, capacitance C_A2 associated to switch S2 is charged and capacitance C_Ai associated to switch Si is discharged, creating the condition for a switching at zero voltage of switch Si.

Therefore, the duration of the first time interval Atl in which switch Si is closed may be controlled so that output voltage V₀ reaches a desired value, as shown for example in equation (5) . For example, to the purpose it is possible to use a feedback of the output voltage V₀ which increases or decreases such durations until the supply signal reaches a desired value.

In various embodiments, intervals At2 and At may be constant, and the duration of interval At3 may be varied to have a fixed switching frequency, i.e.

At3 = Ts - Atl - At2 - AtA (9) wherein Ts is the duration of a switching period which is constant.

Some general considerations about the dimensioning of the various components enabling a zero voltage switching of switches Si and S2 will be explained in the following.

Specifically, the inventor has observed that, thanks to the previously described driving arrangement, the switching of switch Si at time tl and the switching of switch S2 at time t3 are not very critical. However, there may occur situations wherein: a) the switching of switch Si at time tO does not take place at zero voltage, because voltage V_Ds2 across switch S2 has not reached the value of input current Vin, and/or

b) the switching of switch S2 at time t5 switches off diode D at non-zero current.

Figure 12 shows in this respect a simplified circuit diagram for the operating mode M6.

Specifically, in the considered embodiment, a capacitor C_eq is connected in parallel to switch S2. Moreover, a circuit LC comprising an inductance L_P and a capacitance C_RP is connected in parallel to switch S2.

In the presently considered embodiment:

a) the voltage V_CRP across capacitor C_RP is considered substantially constant;

b) the inductor L_P represents the magnetization inductance connected in parallel with inductor Lo (also in this instance neglecting leakage inductance L_R) ;

c) the capacitance C_eq represents capacitances C_Ai e C_A2, and possible other capacitances, wherein capacitance C_eq is typically much smaller than capacitance C_RP (C_eq << C_RP) .

As previously explained, when switch S2 is opened at time t5, voltage V_DS2 increases starting from 0 V.

If current I_P flowing through the primary side of transformer T equals zero at time t5, voltage V_DS2 starts oscillating with a frequency of:

The maximum value that voltage V_DS2 may reach in the operating mode M6 substantially corresponds to the peak-to-peak value of the oscillation, and the minimum value V_DS2,m_±n corresponds to

v D,S2,min (11) wherein D_m±_n represents the minimum duty cycle, i.e. the duty cycle for the smallest ratio between output voltage V₀ and input voltage V_in.

However, as such a duty cycle is typically lower than 0,5, voltage V_DS2 cannot reach voltage Vi_n at time t6, which is essential to enable a zero voltage switching of switch Si.

Therefore, according to the present description, the missing energy is supplied via inductor L_Pi which must store at least the following energy:

- (2 ^■ Vm. ^■ Dmm. )'²1J (12)

wherein current I_p corresponds

Vm. ^■ Dmi.nfL(>l - Dmi.n )'Ts1i /' 2 (¹

^JP = I (13)

2

For example, if the minimum duty cycle D_m±_n is higher than 0.5 {D_min > 0.5), current I_p may also be zero, i.e.

and, as a consequence, the value of inductance L_P may be calculated as:

V. D, D, K]/2

(15)

I

2

On the contrary, if the minimum duty cycle D_m±_n is lower than 0.5 {D_min < 0.5), current I_p must supply the missing energy. For example, this may be obtained by dimensioning L_P in the following way:

{2BC + A) - V(2BC + A)² - B²C²

(16)

2C²

wherein :

A c_eqW_in - (2 ^■ V_inD_min)²}

Therefore, by knowing the value of inductance L_p, it is possible to dimension inductances L_M and L₀.

For example, inductance Lo is typically sized according to the current ripple ΔΤ_ί0 which is desired at the output, as if it was a Buck converter.

Therefore, by knowing L_p and Lo, it is possible to solve the following equation in order to calculate the value of L_M:

l₌ (L_M * n² * L₀)

P (L_M + n² * L₀)

Finally, the resonance time Tr — n^L_RC_R limits the maximum duty cycle. Therefore, by reducing time Tr a higher duty cycle can be obtained, and the output voltage may vary more. For this reason, with the previously described sizing, such a time should be lower than the minimum time during which switch Si is off, i.e.:

Tr = < (1 - D_max)T₃ (17) wherein D_max is the maximum duty cycle.

However, by reducing duration Tr, the current peak in diode D increases. In this context, the inventor has observed that a reasonable value for duration Tr is:

1

Tr = - T (18) 4

wherein T_s is the switching period of the converter, i.e. the duration between two subsequent times tO in which switch Si is switched on.

Finally, in order to enable a switching of switch S₂ at time t5 by switching off diode D at zero current, the value of capacitance C_R may be calculated as:

Tr²

C_R ≤ —— (19) π L_R

For example, if input voltage Vi_n is 400 VDC, the output current is between 0 and 2A, and the switching frequency fs is between 20 and 100 kHz, the components may have the following values:

- a capacitance C_eq up to 1.5 nF;

- an inductance L_p between 200 μΗ and 3 mH.

Therefore, with the previously described dimensioning and considering a supply of 60 W, it is possible to use MOSFETs with a capacity of about 100 pF, wherein an additional capacitor with a capacity up to 400 pF may be connected in parallel with each switch. In fact, in this case, the equivalent capacity C_eq in the switching node may therefore be seen as a capacity of about 1000 pF.

Accordingly, by sizing the various components in a suitable way, the previously described driving method may therefore operate switches Si and S₂ of the half bridge at zero voltage. Such a zero voltage switching may also be obtained for a wide interval of possible duty cycles / output voltages. Moreover, also the diode of the ZETA converter is switched at zero current. As a consequence, the switches of the half bridge may have a reduced size, and also the electromagnetic interferences are reduced.

Generally, in this way, the converter may be operated with a fixed switching frequency, wherein the output inductance Lo may be operated both in the CCM mode and in the DCM mode .

Of course, without prejudice to the underlying principles of the invention, the details and the embodiments may vary, even appreciably, with respect to what has been described herein by way of non-limiting example only, without departing from the scope of the invention, said scope being defined by the annexed claims .

Claims

1. An electronic half-bridge ZETA converter for providing a power signal (V_or i_a) through an output (106), said electronic half-bridge ZETA converter comprising:

- a transformer (T) with a primary winding (Tl) and a secondary winding (T2), wherein each winding (Tl, T2) comprises at least a first and a second terminal,

- a half-bridge ( S i , S2 ) comprising a first ( S i ) and a second ( S2 ) electronic switch for selectively connecting said first terminal of said primary winding (Tl) to a supply signal ( Vin ) or a ground ( GNDi ) , wherein a respective capacitance (C _i, C_A2 ) and a respective diode (D_Ai, D_A2) are associated with said first ( S i ) and said second ( S2 ) electronic switch,

- at least one capacitor (C_RP) connected between said second terminal of said primary winding (Tl) and said supply signal {V_in) and/or said ground ( GNDi ) ,

- an electronic ZETA converter connected to said secondary winding (T2), wherein said ZETA converter comprises three branches which are connected in parallel, wherein:

a) the first branch comprises a first capacitor (C_RS) connected in series with the secondary winding (T2) of said transformer (T2), so that said first branch comprises a first inductance with includes the magnetization inductance (L_M) of said transformer (T) and possible other inductors connected in parallel with the primary (Tl) and/or secondary (T2) winding of said transformer (T) ,

b) the second branch comprises a diode (D) , and c) the third branch comprises a second capacitor (Co) connected in series with a second (L₀) , wherein said output (106) is connected in parallel with said second capacitor (Co) ; and - a control unit (112) configured for driving said first (Si) and said second (S2) electronic switch with the following time intervals which are repeated periodically :

a) a first time interval (Atl) , wherein said first switch (Si) is closed and said second switch (S2) is opened, so that a current flow from said supply signal (V±_n) through said primary winding of said transformer increases ;

b) a subsequent second time interval (At2) , wherein said first switch (Si) is opened and said second switch (S2) is opened;

c) a subsequent third time interval (At3) , wherein said first switch (Si) is opened and said second switch (S2) is closed; and

d) a subsequent fourth time interval (At4), wherein said first switch (Si) is opened and said second switch (S2) is opened, so that the capacitance (C_A2) associated with said second switch (S2) is charged and the capacitance (C_Ai) associated with said first switch (Si) is discharged,

characterized in that said first (L_M) and said second inductance (L₀) are dimensioned such that said first (L_M) and said second inductance (L_o) provide during said forth time interval (At4) a current which charges the capacitance (C_A2) associated with said second switch (S2) and discharges the capacitance (C_Ai) associated with said first switch (Si) .

2. The electronic converter according to Claim 1, wherein during said third time interval (At3) occurs an oscillation of a resonant circuit comprising said at least one capacitor (C_RP) connected between said second terminal of said first winding (Tl) and said supply signal (V±_n) and/or said ground (GNDi) , said first capacitor (C_RS) connected in series with the secondary winding (T2) of said transformer (T2) and the leakage inductance (L_R) of said transformer (T) , wherein said at least one capacitor (C_RP) , said first capacitor (C_RS) and said leakage inductance (L_R) are dimensioned such that:

a) during said third time interval (At3) occur one or more half-periods of the oscillation of said resonant circuit, and

b) the current which flows through the diode (D) is zero at the end of said third time interval (At3) .

3 . The electronic converter according to Claim 2, wherein the capacitance of said at least one capacitor (C_Rp) and the capacitance of said first capacitor (C_RS) are dimensioned such that:

where C_R is the capacitance of said at least one capacitor (C_RP) connected in series with said first capacitor (C_RS) ; Tr is the minimum duration of said third time interval (At3) , and L_R is the leakage inductance (L_R) of said transformer (T) .

4. The electronic converter according to any of the previous claims, wherein the equivalent capacitance (C_eq) at the intermediate point of said half-bridge (Si, S2) which comprises said capacitances (C_Ai, C_A2) associated with said first (Si) and said second (S2) electronic switch is between 200 pF and 1.5 nF.

5. The electronic converter according to any of the previous claims, wherein the sum of the durations of said first time interval (Atl) , said second time interval (At2) , said third time interval (At3) and said fourth time interval [At 4) , i.e. the switching period (T_s), is substantially constant.

6. The electronic converter according to any of the previous claims, wherein the durations of said second time interval (At2) and the duration of said fourth time interval (At4) are substantially constant.

7. The electronic converter according to any of the previous claims, wherein the values of said first (L_M) and said second inductance (L₀) are dimensioned as a function of:

a) the voltage of said supply signal {V_in) , b) the duration of said first time interval (Atl) with respect to the duration of a switching period (T_s), i.e. the duty cycle,

c) the output current (i₀) provided through said output (106), and

d) the equivalent capacitance (C_eq) at the intermediate point of said half-bridge (Si, S2) which comprises said capacitances (C_Ai, C_A2) associated with said first (Si) and said second (S2) electronic switch.

8. A method for operating an electronic converter according to any of Claims 1 to 7, comprising driving said first (Si) and said second (S2) electronic switch with the following time intervals which are repeated periodically :

a) a first time interval (Atl), wherein said first switch (Si) is closed and said second switch (S2) is opened, so that a current flow from said supply signal {Vi_n) through said primary winding of said transformer increases ;

wherein the method comprises:

- during said fourth time interval (At4), charging the capacitance (C_A2) associated with said second switch (S₂) and discharging the capacitance (C_Ai) associated with said first switch (Si) via a current which is provided by said first (L_M) and said second inductance (L₀) .

9 . A method for designing an electronic converter according to any of Claims 1 to 7, comprising:

- dimensioning said first (L_M) and said second inductance (L₀) such that said first (L_M) and said second inductance (L₀) provide during said forth time interval (At4) a current which charges the capacitance (C_A2) associated with said second switch (S2) and discharges the capacitance (C_Ai) associated with said first switch (Si) .

10. The method according to Claim 9, wherein during said third time interval (At3) occurs an oscillation of a resonant circuit comprising said at least one capacitor (C_RP) connected between said second terminal of said first winding (Tl) and said supply signal (V±_n) and/or said ground (GNDi) , said first capacitor (C_RS) connected in series with the secondary winding (T2) of said transformer (T2) and the leakage inductance (L_R) of said transformer (T) , and wherein the method comprises:

- dimensioning said said at least one capacitor

(C_Rp) , said first capacitor (C_RS) and said leakage inductance (L_R) are dimensioned such that:

a) during said third time interval (At3) occur one or more half-periods of the oscillation of said resonant circuit, and b) the current which flows through the diode (D)ro at the end of said third time interval (At3) .