EP2324565A1

EP2324565A1 - Converter device and interruption-free power supply including such a device

Info

Publication number: EP2324565A1
Application number: EP09737010A
Authority: EP
Inventors: Corentin Rizet; Alain Lacarnoy; Jean-Paul Ferrieux
Original assignee: MGE UPS Systems SAS
Current assignee: Schneider Electric IT France SAS
Priority date: 2008-09-12
Filing date: 2009-08-10
Publication date: 2011-05-25
Also published as: FR2936113A1; CN102150352B; BRPI0918171A2; US20110133554A1; CN102150352A; FR2936113B1; US8384246B2; WO2010029222A1; BRPI0918171B1

Abstract

The invention relates to a converter device that comprises a power supply input (121, 122), a rectifier means (D1, D4), a switching means (T2, T3), a control means (301) and a switching assistance circuit (231, 232), wherein said switching assistance circuit includes an inductive means, a means for bypassing an input current (IE), and a power storage means (137, 138). The device of the invention is characterised in that the inductive means essentially comprises a transformer (TP, TN) directly connected to the power supply input (121, 122) and including reversely wound windings, and in that the bypass means includes an auxiliary switching means (TX2, TX3) directly connected between said inductive means and a reference voltage or an output line (115, 117) in order to establish an input current bypass on said inductive means before the main priming P. The invention also relates to an interruption-free power supply including the above-described converter device.

Description

CONVERTER DEVICE AND POWER SUPPLY WITHOUT INTERRUPTION EQUIPPED WITH SUCH A DEVICE

TECHNICAL FIELD OF THE INVENTION

The invention relates to the field of converters such as rectifiers, for example those used in uninterrupted power supplies, in particular in uninterrupted power supplies of high power, that is to say whose power is generally between about 100 and 500 kVA.

The invention more particularly relates to a unidirectional converter device intended to supply a substantially continuous output voltage on an output line, said device being equipped with at least one switching unit comprising:

a supply input to which an input voltage, which is generally variable, is applied,

rectifying means connected to said supply input for supplying the output voltage,

switching means connected to said supply input for obtaining a main priming or a main blocking of an input current so that, during the main blocking, said input current is diverted towards said rectifying means,

means for controlling said switching means, and

a switching aid circuit arranged between the supply input and the output line for establishing, before the main boot, a switching voltage substantially equal to zero, said switching assistance circuit comprising inductive means, means for deriving an input current, and energy accumulation means connected in parallel to the switching means for establishing a resonance of said current in the inductive means before the main ignition. STATE OF THE ART

Uninterruptible power supplies are commonly developed to improve their performance and to reduce the noise generated by often low switching frequencies of the order of a few thousand hertz. In this context, it has been shown that it is interesting to use uninterrupted power supplies with topologies on several levels, generally three levels, using more efficient components to improve the problems mentioned above.

With reference to FIG. 1, such an uninterruptible power supply 11 comprises a network input 12 to which an electrical supply network is connected and making it possible to apply to said uninterruptible power supply 11 a variable input voltage that is most often alternative. The uninterruptible power supply also comprises a network output 13 to which loads are connected and making it possible to supply a so-called emergency power supply, that is to say a power supply for which the voltage and the frequency are controlled. The uninterruptible power supply 11 comprises a rectifier or an AC / DC converter 15 connected to the network input 12, lines 16, 17 of substantially continuous voltages, and a voltage reference 18 connected at the output of the rectifier. The uninterruptible power supply 11 also comprises a DC / DC converter 19 comprising electrical energy storage means 20, said converter and said storage means being connected to the lines 16, 17 of substantially continuous voltage. The uninterruptible power supply 11 further comprises decoupling capacitors 21, 22 connected between the voltage reference 18 and the lines 16, 17 of substantially continuous voltage, and a reversible inverter DC / AC 23 connected between said lines 16, 17 and the network output 13.

The rectifier 15 of the uninterruptible power supply 11 shown in FIG. 2 comprises six switching circuits 31 to 36. More precisely, the rectifier 15 comprises two switching circuits for each of the three phases, one dedicated to positive half-waves and one to the other. another dedicated to negative alternations. Furthermore, the rectifier 15 is of unidirectional type, that is to say that it is not reversible and only allows AC / DC conversion. To perform this AC / DC conversion, the rectifier 15 comprises transistors 41 to 46 and diodes 51 to 56. As can be seen in FIGS. 1 and 2, the uninterruptible power supply 11 has a three-level topology, that is to say that the rectifier 15 provides a substantially continuous voltage on three levels, namely a positive level on the line 16, a negative level on the line 17 and a zero level on the voltage reference 18. The positive and negative levels generally have the same electrical potential in absolute value substantially equal to half the voltage VDC between the lines 16 and 17.

When using the uninterruptible power supply 11 shown in FIG. 1 and in particular the rectifier device 15 in its AC / DC converter function, the switching speeds of the transistors 41 to 46 and the high currents flowing thereto impose very important structural constraints. Moreover, the switching losses in these active power electronics components limit the increase of the switching frequency.

One solution to overcome these disadvantages is to use switching assistance circuits in each switching circuit to obtain smooth switching, that is to say to reduce switching losses and to control current variations. . In the rectifying device 111 partially shown in FIG. 3, such switching assistance circuits are used. In the device shown in FIG. 3, only two switching circuits associated with one of the three phases are shown. It should be noted that the rectifier circuit 111 is not reversible and can therefore only be used to perform an AC / DC type conversion.

More specifically, with reference to FIG. 3, the rectifier device 111 comprises a voltage source 112 delivering an alternating voltage, a first switching circuit 113 making it possible to supply an output line 115 with a substantially constant voltage having a positive value and a voltage. second switching circuit 116 for providing on an output line 117 a substantially constant voltage having a negative value. A first branch of the rectifier device 111 comprising a diode DP makes it possible to supply the first switching circuit 113 for the positive half-cycles of the input voltage. In the same way, a second branch of the rectifier device 111 comprising a diode DN makes it possible to feed the second switching circuit 116 for the negative half-waves of the input voltage. Between the source of tension and the two aforementioned branches, an inductor 118 makes it possible to make an impedance matching on the scale of the cutting period. Each switching circuit 113, 116 includes a power input 121, 122 into which an input current IE is injected. Rectifying means, in this case diodes D1, D4 connected to their respective supply inputs 121, 122 make it possible to supply an output voltage VS by successively switching from a blocked state to a on state. Each switching circuit 113, 116 comprises switching means, in this case main transistors T2, T3 connected to their respective supply inputs 121, 122 and making it possible to obtain a change of state, in this case a primary priming or main blocking of the input current. During the main priming, the input current IE goes into the main transistor. During the main blocking, this input current IE is deflected towards the rectifying means. The diode DP, the rectifying means D1 and the switching means T2 form a topology often described as elevator structure, or in English "boost". It is the same for the diode DN, the rectifying means D4 and the switching means T3. Thus, the topology shown in Figure 3 is often referred to in English as "double boost". However, the invention can also be applied to a topology often referred to as a step-down structure.

As can be seen in FIG. 3, the rectifier device is equipped with switching assistance circuits 131, 132, the circuit 131 being arranged between the supply input 121 and the output line 115, the circuit 132 being it is arranged between the supply input 122 and the output line 117. These switching assistance circuits have the main function of reducing the switching losses in the power transistors T2 and T3 by limiting or even reducing canceling, current or voltage in said transistors T2 and T3 during state changes. In particular, these switching assistance circuits make it possible to obtain a main ignition of the switching means T2 and T3 under zero voltage. This switching mode is often referred to in English as "Zero Voltage Switching", or abbreviated as "ZVS". The switching aid circuits 131, 132 comprise inductive means respectively referenced 133, 134 and respectively connected to the supply inputs 121, 122. In the prior art, the inductive means generally comprise at least one inductor which is often directly connected to the power input. The switching assistance circuits 131, 132 also comprise means for deriving the input current IE, respectively referenced 135, 136 and connected to said inductive means for establishing, before the main priming, a derivation of the input current in said inductive means. The switching aid circuits 131, 132 furthermore comprise energy storage means referenced respectively 137, 138 connected in parallel on the switching means for establishing a resonance of the input current IE in the inductive means before the main boot. More specifically, these energy accumulation means 137 comprise a capacitor CR1 connected in parallel with the diode D1 and a capacitor CR2 connected in parallel with the transistor T2. In the same way, the energy accumulation means 138 comprise a capacitor CR4 connected in parallel with the diode D4 and a capacitor CR3 connected in parallel with the transistor T3.

The rectifying device shown in FIG. 3 operates in the following manner. Before initiating the switching means T2, T3, the input current IE is deflected via the means of derivation 135, 136. The intensity of the current IRP, IRN flowing in the inductive means 133, 134 increases in at the same time as the current flowing in the rectifying means D1, D4 decreases. When the current IRP, IRN in the inductive means 133, 134 reaches the value of the input current IE, the rectifying means D1, D4 are blocked. A resonance phase of the current is thus obtained between the inductive means 133, 134 and the energy accumulation means 137, 138. This resonance phase makes it possible to obtain the cancellation of the voltage V2, V3 across the means. switching T2, T3. It is then possible to prime these switching means T2, T3 with a switching voltage substantially equal to zero. During all this phase, magnetization is created in the inductive means, that is to say that the value of the magnetic field increases.

The switching assistance circuits of the converter devices of the prior art generally do not allow to obtain a complete demagnetization of the inductive means before the main blocking of the switching means. In addition, they comprise electronic power components, in particular transistors, the caliber and the amount of energy dissipated are not optimized.

STATEMENT OF THE INVENTION The aim of the invention is to remedy the drawbacks of prior art converter devices by proposing a unidirectional converter device intended to supply a substantially continuous output voltage on an output line, said device being equipped with at least one switching unit comprising :

a supply input to which an input voltage is applied,

switching means connected directly to said supply input to obtain a primary ignition or a main blocking of an input current so that, during the main blocking, said input current is diverted towards said rectifying means,

means for controlling said switching means, and

a switching aid circuit arranged between the supply input and the output line for establishing, before the main boot, a switching voltage substantially equal to zero, said switching assistance circuit comprising inductive means, means for deriving an input current for establishing a bypass of the input current on said inductive means before the main ignition, and energy accumulation means connected in parallel with said switching means to establish a resonance of said current in said inductive means before the main priming.

The converter device according to the invention is characterized in that the inductive means essentially consist of a transformer directly connected to the supply input and comprising windings wound in reverse, and in that the branching means comprise means for auxiliary switching directly connected between said inductive means and a voltage reference or between said inductive means and the output line.

Preferably, the transformer comprises: a first winding connected between the supply input and the derivation means, and

a second winding magnetically coupled to the first winding and connected between said supply input and the output line or between the supply input and the voltage reference.

Preferably, the transformer has a transformation ratio of less than unity.

Advantageously, a first non-return diode is connected between the first winding and the output line or between the first winding and the voltage reference. Preferably, a second non-return diode is connected between the second winding and the output line or between the second winding and a voltage reference.

Advantageously, the auxiliary switching means essentially consist of an auxiliary transistor connected directly between the first winding and the voltage reference or between the first winding and the output line, said auxiliary transistor enabling, during the main blocking and at the moment of the firing said auxiliary transistor, providing on the transformer windings a voltage having a value depending on the output voltage.

According to one embodiment, the control means comprise a delay module designed to force a delayed main priming after a duration greater than a predetermined duration. Preferably, the control means are applied to the auxiliary switching means and comprise a module designed to initiate the bypass of the current for a duration greater than the predetermined duration.

Preferably, the rectifying means comprise a diode comprising a current input, said input being connected to the supply input.

Advantageously, the energy storage means comprise a first capacitor connected in parallel with the rectifying means and a second capacitor connected in parallel with the switching means.

The invention also relates to an uninterruptible power supply comprising a power input to which a variable input voltage is applied, a rectifier connected to said input, at least one substantially continuous voltage line connected at the output of the rectifier, an inverter connected to said voltage line and having an output for providing a variable output voltage, characterized in that the rectifier is a converter device according to one of the preceding claims and provides a substantially continuous output voltage on said line.

BRIEF DESCRIPTION OF THE FIGURES

Other advantages and features will emerge more clearly from the following description of particular embodiments of the invention, given by way of non-limiting examples, and shown in the accompanying figures.

FIG. 1 represents an uninterrupted power supply according to the prior art.

FIG. 2 represents the rectifier of the uninterrupted power supply represented in FIG.

FIG. 3 partially shows a rectifying device with a switching aid circuit according to the prior art.

Figure 4 partially shows a converter device according to a first embodiment of the invention.

FIG. 5 diagrammatically represents the control means of a converter device.

Figures 6A-6M are timing diagrams illustrating the operation of the converter device shown in Figure 4 during most of the alternation.

Figures 7A to 7K are timing diagrams illustrating the operation of the converter device shown in Figure 4 at the beginning and end of alternation.

Figure 8 partially shows a step-down device according to a second embodiment of the invention.

Figure 9 shows an uninterruptible power supply according to the invention. DETAILED DESCRIPTION OF AN EMBODIMENT

The converter device 211 partially shown in FIG. 4 is a rectifier device comprising elements already described previously and indicated by the same reference numerals. As for FIG. 3, only the two switching circuits associated with one of the three phases have been represented. The converter device 211 comprises a voltage source 112 delivering an alternating voltage VE and an input current IE. As in the case shown in Figure 3, a first switching circuit 213 provides on the output line 115 a substantially constant voltage having a positive value. Likewise, a second switching circuit 216 makes it possible to supply an output line 117 with a substantially constant voltage having a negative value. As in FIG. 3, these switching circuits are of the lift type. Each switching circuit 213, 216 comprises a supply input 121, 122 on which the input voltage VE is applied and into which the input current IE is injected. The input voltage VE is variable, generally alternating and often sinusoidal. The diode DP, the rectifying means D1 and the switching means T2 form a first elevator-type structure. It is the same for the diode DN, the rectifying means D4 and the switching means T3 which form a second elevator-type structure. Each main transistor T2, T3 of the switching means generally comprises a diode D2, D3 connected in parallel and oriented in the opposite direction with respect to the current direction in the transistor. The converter device 211 is equipped with switching assistance circuits 231, 232, the circuit 231 being disposed between the supply input 121 and the output line 115, the circuit 232 being arranged between the supply input 122 and the exit line 117.

In the embodiment shown in FIG. 4, the components referenced DP and DN are diodes. In other embodiments, these components may be thyristors.

With reference to FIG. 4, each switching assistance circuit 231, 232 comprises inductive means consisting essentially of a transformer TP, TN. Each transformer TP, TN is directly connected to the supply input 121, 122 of the switching circuit considered. In other words, the two windings of the transformer are directly connected to the power input. Since the inductive means of each switching aid circuit essentially consist of a transformer, and the latter is directly connected to the supply input 121, 122, the topology of the converter device 211 and its circuits switching assistance 231, 232 is simplified.

Each switching assistance circuit 231, 232 shown in FIG. 4 also comprises means for deriving the input current IE comprising auxiliary switching means, in this case an auxiliary transistor TX2, TX3. Each auxiliary transistor is connected to the transformer TP, TN to establish, before the main boot, a derivation of the input current IE in said transformer. More specifically, each auxiliary transistor TX2, TX3 is directly connected between the transformer TP, TN and the voltage reference. By directly connected means that the connection means between the auxiliary transistor and the voltage reference, and between the same auxiliary transistor and the transformer, are essentially constituted by electrical conductors and / or equivalent resistances of these conductors.

Each switching assistance circuit 231, 232 shown in FIG. 4 further comprises energy storage means 137, 138 connected in parallel with the switching means, that is to say on each transistor T2. , T3, and the rectifying means, that is to say the diodes D1, D4. More precisely, the energy accumulation means 137 comprise a capacitor CR1 connected in parallel with the diode D1 and a capacitor CR2 connected in parallel with the main transistor T2. In the same way, the energy accumulation means 138 comprise a capacitor CR4 connected in parallel with the diode D4 and a capacitor CR3 connected in parallel with the transistor T3. These accumulation means make it possible, among other things, to establish, before the main ignition, a resonance of the current in the transformers TP, TN.

As shown in FIG. 4, the transformer TP, TN of each switching circuit 231, 232 comprises a first winding 251, 252 connected between the supply input 121, 122 and the auxiliary switching means TX2, TX3. This transformer TP, TN also comprises a second winding 253, 254 magnetically coupled to the first winding 251, 252 and connected between this same input 121, 122 and the output line 115, 117, more precisely between the supply input 121, 122 and the diode DA1, DA4. Moreover, the second winding 253, 254 is wound in inverse relation to the first winding 251, 252.

This configuration of the transformer TP, TN allows, when the auxiliary transistors TX2, TX3 are initiated, to deflect more current in each of the windings of the transformer TP, TN. Indeed, thanks to the inverted winding of the windings and to the connection of the contiguous ends of said windings to the power input, the input current IE is deflected to be shared in each of the windings. Thus the IRP input current, IRN is amplified by mutual induction. This allows a reduction of the current rating of the auxiliary transistor TX1, TX2. After the blocking of the diode D1, D4, the voltage V2, V3 at the terminals of the main transistor T2, T3 decreases to a value substantially equal to zero, and the corresponding diode D2, D3 becomes conducting, which makes it possible to start said main transistor at zero voltage.

This configuration of the transformer TP, TN allows, in addition, once the main transistor T2, T3 is initiated, to demagnetize said transformer, that is to say that no current no longer circulates in the windings of the transformer. This avoids an accumulation of energy in the transformer which would eventually destroy the converter device. This demagnetization is made possible by the diode DX1, DX4, which makes it possible to apply the VS output voltage in reverse on the winding 251, 252, when the auxiliary transistor TX2, TX3 is off and when said diode becomes conducting.

The transformer TP, TN generally has magnetic leaks on each of the windings that can not usually be neglected. It is thus possible to define an equivalent inductance created by the leaks and to link this inductance to an equivalent resonance inductance. This resonance inductance determines the rise slope of the current in the windings of the transformer. Advantageously, the transformer TP, TN comprises an electrically insulating material separating the windings. A choice of the thickness of this insulating material makes it possible, among other things, to adjust the leakage inductance of the transformer and therefore the rise slope of the current.

As can be seen in FIG. 4, a first diode DX1, DX4 is connected between the first winding 251, 252 and the output line 115, 117. When the transistor auxiliary TX2, TX3 is blocked, this diode allows the passage of the current in the first winding 251, 252 in one direction. This diode also limits the voltage across the auxiliary transistor TX2, TX3. A second diode DA1, D A4 is connected between the second winding 253, 254 and the output line 115, 117. This diode allows the passage of the current in one direction in this second winding. The presence of these diodes DA1, DA4 prevents any reversible operation of the switching aid circuits of the converter 211, and allows the demagnetization of the transformer TP, TN. This unidirectional operation is advantageous since it limits the operating time of the switching circuit 231, 232, and therefore limits the losses in said circuit.

The main transistors T2, T3 of the switching means can be used in dual thyristor mode, that is to say that the priming is done naturally. In general, the main ignition occurs naturally when the switching voltage V2, V3 of the switching means becomes substantially zero and the diode D2, D3 becomes conducting. However, in the case where the intensity of the input current IE on the supply input 121, 122 is too low, that is to say for an amplitude of the voltage VE less than about 10% of its maximum value, which generally corresponds to the beginning or the end of the alternation of said voltage VE, the output voltage does not have time to reach the value of the target voltage VS line and the natural start of main transistors is not possible. Indeed, in this case the capacitors of the energy storage means do not have time to load and it is difficult to obtain a resonance of the current entering the inductive means.

To remedy this drawback, the control means 301, shown in FIG. 5, comprise a delay module 315 designed to force a delayed main priming after a duration greater than a predetermined duration TMAX. This forced operating mode is mainly implemented at the beginning and at the end of the alternation of the voltage VE, when the value of the input current IE is not sufficient to charge the capacitors of the accumulator means. 'energy. The control means 301 are shown in FIG. 5 only for the switching unit 213. Equivalent means for the switching unit 216 as well as the switching units of the other phases can be used but have not been represented. . More specifically, as shown in FIG. 5, the control means 301 comprise a referenced module 311 for generating a first control signal 302 with pulse width modulation, abbreviated PWM and PWM. This first control signal is determined from the measurements of the output voltage VS, the input voltage VE ₅ and the input current IE. A module 316 makes it possible to prime the auxiliary transistor TX2 for a duration TMAX '. This duration runs from the rising edge of the first control signal 302. During normal operation and during this period TMAX ', the auxiliary transistor TX2 can thus be initiated, which makes it possible to cancel the voltage V2 to prime the main transistor. T2. For this, the control means comprise a comparator 312 for detecting the zero crossing of the voltage V2 across the main transistor T2. The output of this comparator is connected to an input of a first Boolean operator of the logical "AND" type referenced 313. Another input of this operator is connected to the output of the module 311 carrying the first modulated width modulation control signal. Thus, the zero crossing of the voltage V2 and the simultaneous presence of an active pulse width modulation signal enable the output of this Boolean operator 313 to be activated. This output of the operator 313 is connected to a second Boolean operator of "OR" type referenced 314 whose output is connected to the control input of the main transistor T2. Thus, in normal operation, when the output of the "AND" operator 313 is activated, the output of the operator 314 is also activated, which makes it possible to control the priming of the main transistor T2 at the moment when the voltage V2 passes by zero.

In forced operation, that is to say at the beginning or end of an alternation of the input voltage VE, the value of the input current IE is not sufficient to cancel the voltage V2 across the terminals of the main transistor T2. The output of the "AND" operator 313 therefore remains inactive. Thus, to prime the main transistor T2 forcibly, the aforementioned delay module 315 is used. At the input of this module 315 is connected the output of the carrier module 311 of the first control signal 302, the output of this module 315 being for its part connected to an input of the logical "OR" operator 314. Thus, in forced operation, the transistor T2 is automatically initiated after a predetermined time TMAX. The choice of the duty cycle used in the modulation module 311 is generally made taking into account the demagnetization time of the transformer TP, TN, which is generally of the order of half the boot time. This makes it possible to avoid saturation of these transformers.

With reference to the timing diagrams of FIGS. 6A to 6M, the operation of the switching circuit 213 of the converter device of FIG. 4 is described below, in the case where the intensity of the input current IE on the power input of the converter device is sufficient to obtain natural priming of the main transistors. In other words, the operation described below excludes to some extent the beginning and the end of P alternating voltage VE.

Initially, the main transistor T2 is in a primed or on state, which is indicated by the presence of a bold line in FIG. 6A. The auxiliary transistor TX2 is in turn in a locked state, which is indicated by the absence of bold lines in Figure 6B. As can be seen in FIG. 6F, the diode D1 is blocked. The transistor T2 sees a current IT2, shown in FIG. 6E, substantially equal to the input current IE. The voltage V2 across the terminals of the transistor T2 shown in FIG. 6D is therefore substantially equal to zero. The diode DA1 does not see any current as shown in Figure 6G and is in the off state. The voltage VDA1 at its terminals shown in FIG. 6H is therefore substantially equal to the value of the output voltage VS on the output line 115.

At time t1, the transistor T2 is off (FIG. 6A), the input current IE is derived in the storage means 137, which makes it possible to reduce the losses of said transistor. As can be seen in FIG. 6D, the voltage V2 across the main transistor T2 begins to increase gradually by charging the capacitor CR2 connected in parallel. The diode DA1 is always in the off state, and the voltage VDA1 at these terminals begins to decrease (FIG. 6H) until reaching a zero value. At the same time, as can be seen in FIG. 6K, the voltage VTX2 across the auxiliary transistor TX2 increases to the value of the output voltage VS. At time t2, the voltage V2 across the main transistor T2 reaches the value of the output voltage VS (FIG. 6D), and the diode D1 starts conducting a current ID1 whose value is substantially equal to the value of the current IE input shown in Figure 6F.

At time t3, the auxiliary transistor TX2 is started (FIG. 6B), which will cause a decrease in the current ID1 in the diode D1 (FIG. 6F) which is deflected towards said auxiliary transistor TX2 which has become on. As can be seen in FIG. 61, the auxiliary transistor TX2 thus sees a current ITX2 which increases progressively. Thus, the PIR current entering the transformer TP, shown in Fig. 6C, will increase as the current ID1 decreases. After initiation of the diode DA1, this current IRP results from the sum of the current ITX2 in the first winding 251 of the transformer TP (FIG. 6J) and the current IDA1 in the second winding 253 of this same transformer TP (FIG. 6G). As soon as the diode DA1 is primed, the output voltage VS is applied to the two windings 251, 253 of the transformer TP. Because of the magnetic leakage of this transformer, the winding 251 will be subjected to a voltage VSEC at its terminals, shown in Figure 6M, substantially equal to the output voltage VS. Since the transformation ratio of the transformer TP is very close to the unit, the current ITX2 in the winding 251 shown in FIG. 6J and the current IDAl in the winding 253 shown in FIG. 6G are substantially equal to half of the value of the PIR current entering the transformer TP, that is to say equal to half of the input current IE.

At time t4, no current flows in the diode D1, which causes its blocking (Figure 6F). The voltage V2 at the terminals of the transistor T2 (FIG. 6D) therefore begins to decrease. At the same time, as can be seen in FIGS. 6C, 6G, and 6J, the IRP current at the input of the transformer TP as well as the currents IDA1 and ITX2 in each winding will continue to increase by mutual induction. In this way, the IRP current entering the transformer will go into resonance. Indeed, at time t4, the capacitor CR1 which is discharged will charge as the voltage V2 across the main transistor T2 decreases to zero. At the same time, the capacitor CR2 which is initially charged will start to discharge. Between times t4 and t5, when the voltage V2 across the main transistor T2 is substantially equal to half of the output voltage VS, the current PIR entering the transformer TP will reach a resonance peak (FIGS. 6C and 6D) . During this period of time, the voltage VSEC across the winding 251 of the transformer TP will decrease (Figure 6M) and the voltage VPRI across the winding 253 of the same transformer will increase (Figure 61). In other words, the output voltage VS will simultaneously switch from the winding 251 to the winding 253.

At time t5, while the voltage V2 across the main transistor T2 is canceled (FIG. 6D), a small current will flow in the diode D2 mounted in inverse parallel to the transistor T2. This is visible in FIG. 6E representing the current IT2 flowing in the module constituted by the main transistor T2 and the diode D2. The main transistor T2 is initiated between the time t5 and the time t6, with a voltage V2 at these terminals which is therefore substantially equal to zero (Figure 6D). Thus, the energy dissipated during this initiation is minimized.

At time t6, the current IT2 in the main transistor T2 increases progressively (FIG. 6E) at the same time the intensity of the currents ITX2 and IDAl respectively in the first and second windings 251, 253 decrease (FIGS. 6J, 6G).

At time t7, no current no longer circulates in the diode DA1 and in the second winding 253 of the transformer TP (FIG. 6G), which causes the blocking of said diode. A low intensity current IMAG shown in FIG. 6J, due to the magnetization of the transformer TP, continues to flow in the transistor TX2 as well as in the first winding of said transformer. Between the time t7 and the time t8, the voltages VPRI, VSEC across the windings 253, 251 of the transformer TP being substantially equal to zero (Figures 61, 6M), the value of this current IMAG remains substantially constant.

At time t8, the transistor TX2 is controlled in a blocked state (FIG. 6B) and the diode DX1 makes it possible to completely evacuate the IMAG magnetization current flowing in the first winding 251, as represented in FIG. 6L. Thus a complete demagnetization of the transformer TP occurs before the main blocking of the main transistor T2. As can be seen in FIG. 6K, the value of the voltage at the terminals of the transistor TX2 is substantially equal to the output voltage VS. As is visible in Figure 6H, the voltage across the diode DAl is, in turn, substantially equal to twice the value of the output voltage VS. Thus, during the demagnetization of the transformer TP, the voltage VTX2 across the auxiliary transistor TX2 is twice as low as the voltage VDA1 across the diode DA1. It is therefore the diode DA1 which cash a large demagnetization voltage in place of the auxiliary transistor TX2, which allows to choose a TX2 transistor of lower caliber, therefore less expensive and operating with a lower energy consumption.

At time t9, the transformer TP is completely demagnetized, that is to say that the average value of the voltage at its terminals is zero. As a result, the IMAG current becomes zero, and the diode DX1 is blocked (FIG. 6L). Thus we find the initial situation preceding the time tl.

With reference to FIGS. 7A to 7K, the operation of the switching circuit 213 of the converter device of FIG. 4 is described below, in the particular case where the intensity of the input current IE on the power input is insufficient to obtain natural priming of the main transistors. The operation described below is therefore generally applicable at the beginning and at the end of the alternation of the voltage VE.

Initially, the transistor T2 is initiated or passing, as can be seen in FIG. 7A, and conducts a current IT2 represented in FIG. 7E whose value is substantially equal to the input current IE. As can be seen in FIGS. 7D and 7F, the value of the voltage V2 across the main transistor T2 is almost zero and the diode D1 is in a blocked state.

At time t1, the main transistor T2 goes from the primed state to the off state (FIG. 7A), and the input current IE is derived in the storage means 137. As can be seen in FIG. 7D, the Voltage V2 across the main transistor T2 begins to increase gradually by charging the capacitor CR2 connected in parallel. Since the intensity of the input current IE is too low, the voltage V2 at the terminals of the transistor T2 increases very slowly and fails to reach the value of the output voltage VS. As a result, the diode D1 can not be initiated and therefore does not conduct (FIG. 7F). At time t2, the auxiliary transistor TX2 is started (FIG. 7B). As can be seen in FIG. 71, the auxiliary transistor TX2 thus sees a current ITX2 which increases progressively. In the same way, the IRP current entering the transformer TP (FIG. 7C) and the current IDAl in the diode DA1 (FIG. 7G) increase.

The IRP current entering the transformer TP will then enter a resonance phase. Indeed, the capacitor CR1 which is initially discharged will charge as the voltage V2 across the main transistor T2 decreases to zero. At the same time, the capacitor CR2 which is initially charged will start to discharge. The IRP current entering the transformer TP will then reach a peak resonance (Figure 7C), which will continue to decline. As can be seen in FIGS. 7C, 7D, 7G, 7H, 71 and 7K, the resonance phase results in oscillations without the voltage V2 across the main transistor T2 being able to cancel out. The transistor T2 can not be initiated because the output of the logical Boolean operator "AND" 313 of the control means 301 remains in an inactive state.

At time t3, after the lapse of time TMAX defined by the delay module 316 of the control means 301, the main transistor T2 is automatically started (FIG. 7A). At the same time, the voltage V2 across the main transistor T2 is suddenly reduced to zero (FIG. 7D), which generates a peak current in the main transistor T2 (FIG. 7E). The IRP current decreases (Figure 7C) and the DA1 diode is in a blocked state (Figure 7G). Only one IMAG magnetizing current flows in the TX2 transistor (Figure 71).

At time t4, after a time TMAX 'defined by the module 315 of the control means 301, the auxiliary transistor TX2 is blocked (FIG. 7B). The diode DX1 makes it possible to obtain complete demagnetization of the transformer TP at time t5 (FIGS. 7H, 71 and 7J).

At time t5, the transformer TP is completely demagnetized. As a result, the IMAG current becomes zero, and the diode DX1 is blocked (FIG. 7J). Thus we find the initial situation preceding the time tl.

The converter device shown in FIG. 4, the operation of which has been previously described, corresponds to an elevating chopper type circuit. The device converter according to the invention can also be adapted to a downhooking type of mounting, as in the embodiment shown in FIG. 8.

The converter device represented partially in FIG. 8 comprises elements already described previously with reference to FIG. 4. For simplicity, only the switching circuit for the treatment of positive halfwaves and associated with one of the three phases has been represented. The converter device represented in FIG. 8 comprises a voltage source 402 delivering an alternating input voltage VE and an input current IE, and switching means 404. The switching circuit represented makes it possible to supply, on the line of output 415, a voltage VS substantially constant and having a positive value. The output of the converter circuit is represented as a DC voltage source VS. An inductor 403 serves to match the impedance between the two voltage sources. The switching circuit shown comprises a supply input 405 on which the input voltage VE is applied and into which the input current IE is injected. The rectifying means D1 and the main transistor T2 form a step-down type structure. The converter device is equipped with a switching aid circuit arranged between the supply input 405 and the output line 415 comprising inductive means 412 consisting essentially of a transformer TP directly connected to the power input. 405. The switching assistance circuit also comprises input current deriving means 411, in this case an auxiliary transistor TX2 connected to the transformer TP, TN to establish, before the main ignition, a current bypass. IE input into said transformer. More specifically, the auxiliary switch TX2 is connected between the first winding 451 and the output line 415. The switching assistance circuit further comprises energy storage means 413 connected in parallel to the means. switching 404.

As shown in FIG. 8, the transformer TP comprises a first winding 451 connected between the supply input 405 and the TX2 branching means, and a second winding 453 wound in reverse, magnetically coupled to the first winding and connected between said supply input 405 and the output line and a voltage reference. A first non-return diode DX1 is connected between the first winding 451 and the voltage reference. A second anti-return diode DA1 is connected between the second winding 453 and the voltage reference. Operation of this step-down arrangement is essentially the same as that of the elevator assembly previously described.

The converter devices described above, and in particular the rectifier devices, can be used in an uninterruptible power supply 501 such as that represented in FIG. 9. This uninterruptible power supply comprises a power supply input 502 on which a voltage of variable input of a first three-phase network. The uninterruptible power supply comprises a rectifier 503 of the type described above, said rectifier being connected between, on one side, the supply input 502 and, on the other side, two output lines 504 or bus of substantially continuous voltage. The uninterruptible power supply comprises an inverter 506 connected between the output lines 504 and an output 507 for supplying a safe three-phase AC voltage to a load 508. The DC voltage bus 504 is also connected to a battery 509 via a DC / DC converter 510.

As can be seen in FIG. 9, static contactors 511 and 512 make it possible to select between the supply input 502 of the first three-phase network and a supply input 513 of a second also three-phase network. Thus, it is possible to feed the load via the first secure network by the uninterrupted power supply 501, and if necessary switch to the second network.

Claims

A unidirectional converter device for providing a substantially continuous output voltage (VS) on an output line (115, 117), said device being provided with at least one switching unit (213, 216) comprising: - an input power supply (121, 122) to which an input voltage is applied

(VE),

Rectifying means (D1, D4) connected to said supply input for supplying the output voltage (VS),

switching means (T2, T3) connected directly to said supply input to obtain a main priming or a main blocking of an input current so that, during the main blocking, said input current is deflected to said rectifying means,

Control means (301) for said switching means, and

- a switching aid circuit (231, 232) arranged between the supply input (121, 122) and the output line (115, 117) to establish, before the main boot, a switching voltage (V2, V3) substantially equal to zero, said switching assistance circuit comprising inductive means, input current deriving means (IE) for establishing a bypass of the input current on said inductive means before the main priming, and energy storage means (137, 138) connected in parallel to said switching means for resonating said current in said inductive means prior to the main priming, characterized in that said means inductive are essentially constituted by a transformer (TP, TN) directly connected to the supply input (121, 122) and comprising reverse wound windings, and in that said branching means comprise switching means auxilia ires (TX2, TX3) directly connected between said inductive means and a voltage reference or between said inductive means and the output line.

2. Device according to claim 1, characterized in that the transformer (TP, TN) comprises: a first winding (251, 252) connected between the supply input (121, 122) and the bypass means (TX2, TX3), and

a second winding (253, 254) magnetically coupled to the first winding and connected between said power input (121, 122) and the output line (115, 117) or between the power input (121, 122) and the voltage reference.

3. Device according to one of claims 1 or 2, characterized in that the transformer has a transformation ratio of less than unity.

4. Device according to one of claims 2 or 3, characterized in that a first non-return diode (DX1, DX4) is connected between the first winding (251, 252) and the output line (115, 117) or between the first winding (251, 252) and the voltage reference.

5. Device according to claim 4, characterized in that a second anti-return diode (DA1, DA4) is connected between the second winding (253, 254) and the output line (115, 117) or between the second winding (253, 254) and a voltage reference.

6. Device according to one of claims 2 to 5, characterized in that the auxiliary switching means essentially consist of an auxiliary transistor

(TX2, TX3) connected directly between the first winding (251, 252) and the voltage reference or between the first winding (251, 252) and the output line (115, 117), said auxiliary transistor enabling, during the blocking main and at the time of booting said auxiliary transistor (TX2, TX3), to provide on the transformer windings a voltage having a value depending on the output voltage (VS).

7. Device according to one of claims 1 to 6, characterized in that the control means (301) comprise a delay module (315) designed to force a delayed main priming after a duration greater than a predetermined time (TMAX) .

8. Device according to claim 7, characterized in that the control means (301) are applied to the auxiliary switching means (TX2, TX3) and comprise a module (316) adapted to initiate current bypass for a duration greater than the predetermined duration (TMAX ').

9. Device according to one of claims 1 to 8, characterized in that the rectifying means comprise a diode (D1, D4) comprising a current input, said input being connected to the supply input (121, 122). .

10. Device according to one of claims 1 to 9, characterized in that the energy storage means (137, 138) comprise a first capacitor (CR1, CR4) connected in parallel with the rectifying means and a second capacitor (CR2, CR3) connected in parallel with the switching means.

An uninterruptible power supply (301) comprising a supply input (302) to which a variable input voltage is applied, a rectifier (303) connected to said input, at least one substantially continuous voltage line connected to the output of the input rectifier, an inverter (306) connected to said voltage line and having an output (307) for providing a variable output voltage, characterized in that the rectifier is a converter device according to one of the preceding claims and provides a voltage substantially continuous output on said line.