Voltage converter and method
By introducing variable threshold and correction loops into the PFM type switch mode converter, the threshold value is adjusted to ensure that the current is close to zero at the end of the cycle, solving the problem of current non-zero, improving the voltage regulation accuracy and power equalization, and enhancing the Load power supply stability.
Patent Information
- Application Number
- CN202110874272.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2021-07-29
- Filing Date
- 2021-07-30
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2041-07-30
AI Technical Summary
At the end of the operation cycle of the existing PFM type switch mode converter, the current cannot always be accurately reduced to zero, resulting in inaccurate voltage Vout regulation and unbalanced power transfer, affecting the stability of the load power supply.
By introducing a variable threshold voltage into the converter, the threshold is adjusted in each operation cycle according to the slope and symbol of the current to ensure that the current is close to zero at the end of the cycle, and the threshold is adjusted by the correction loop so that the current is in the next cycle Converge to zero faster.
It effectively solves the problem that the current is not zero at the end of the cycle, improves the regulation accuracy of voltage Vout and the power transfer equalization, and enhances the stability of the load power supply.
Smart Images

Figure CN114070060B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims the benefit of French Patent Application No. 2008096, filed on July 30, 2020, which is incorporated herein by reference. Technical Field
[0003] The present disclosure relates generally to electronic circuits and more particularly to a switch mode power supply type DC / DC voltage converter that converts a DC supply voltage into a DC output voltage, such as a step-down DC / DC voltage converter in which the DC output voltage has a value lower than the value of the DC supply voltage. Background Art
[0004] In a switch-mode power converter, a direct current (DC) voltage used to power the converter is chopped by switching of switches to achieve a power storage phase of an assembly including inductive and capacitive elements, and a discharge phase of the power stored in the assembly to discharge to a load connected to the converter output.
[0005] In a pulse frequency modulation (PFM) type switch mode converter, each operation cycle of the converter includes a power storage phase in the components, followed by a power delivery phase to the load connected to the converter. During the power storage phase, the current flowing through the inductive element increases. During the power delivery phase, the current flowing through the inductive element decreases. For each operation cycle, it is desired that the current flowing through the inductive element is zero at the beginning of the power storage phase and at the end of the power delivery phase.
[0006] Known switched-mode converters, in particular PFM-type switched-mode converters, have various disadvantages. Summary of the invention
[0007] There is a need to overcome all or some of the disadvantages of known switch-mode converters, particularly PFM-type switch-mode converters.
[0008] One embodiment overcomes all or some of the disadvantages of known switch-mode converters, particularly PFM-type switch-mode converters.
[0009] An embodiment of a voltage converter includes: a first transistor connected between a first node of the converter and a second node configured to receive a power supply voltage; a second transistor connected between the first node and a third node configured to receive a reference potential; a first circuit configured to control the first transistor and the second transistor; and a comparator configured to compare a first voltage with a threshold, the first voltage being equal to a first rising slope during a first time period and being equal to a second falling slope during a second time period, and the threshold having a first value during the first time period and a second value during the second time period, the first value and the second value being variable values.
[0010] An embodiment of an operating method for a voltage converter, the voltage converter comprising: a first transistor connected between a first node of the converter and a second node configured to receive a power supply voltage; a second transistor connected between the first node and a third node configured to receive a reference potential; and a first circuit configured to control the first transistor and the second transistor; the method comprising: during each operating cycle, a first time period in which a comparator compares a first voltage with a threshold, the first voltage is equal to a first rising slope during the first time period, and the threshold has a first value during the first time period; and a second time period in which the comparator compares the first voltage with the threshold, the first voltage is equal to a second falling slope during the second time period, and the threshold has a second value during the second time period, and the first value and the second value of the threshold are variable values.
[0011] According to one embodiment, the converter comprises a second circuit configured to generate a threshold value.
[0012] According to one embodiment, the converter comprises a third circuit configured to compare the current in the first node with a value zero, the second circuit configured to modify the value of the threshold during an operation cycle if the current in the first node was different from zero at the end of a previous cycle.
[0013] According to one embodiment, the first value of the threshold is equal to the reference voltage plus a third value, and the second value of the threshold is equal to the reference value minus the third value, the third value being a variable value.
[0014] According to one embodiment, the third value varies between a plurality of predetermined values.
[0015] According to one embodiment, predetermined values of the third value are consecutively separated by the same fourth constant value.
[0016] According to one embodiment, the converter includes a fourth circuit configured to determine whether the power supply voltage is within a range of a value of the voltage at the first node minus 200 mV to a value of the voltage at the first node plus 200 mV, and the second circuit is configured to maintain the first value and the second value of the threshold for the next cycle if the power supply voltage is within the range during the cycle, or to make the first value and the second value of the threshold have predetermined values during the next cycle.
[0017] According to one embodiment, the converter includes a fifth circuit configured to determine whether a value of a current in the first node is less than zero or greater than zero.
[0018] According to one embodiment, the converter includes a sixth circuit configured to determine whether a slope of a current in the first node during a first time period in a cycle is greater or less than a slope during a second time period in the cycle.
[0019] According to one embodiment, the second circuit is configured to change the first value and the second value of the threshold value such that:
[0020] a) if the slope of the current in the first node during the first time period is greater than the slope of the current in the first node during the second time period, and if the current in the first node is less than zero, reducing the distance between the first value and the second value of the threshold;
[0021] b) if the slope of the current in the first node during the first time period is greater than the slope of the current in the first node during the second time period, and if the current in the first node is greater than zero, increasing the distance between the first value and the second value of the threshold;
[0022] c) if the slope of the current in the first node during the first time period is less than the slope of the current in the first node during the second time period, and if the current in the first node is less than zero, increasing the distance between the first value and the second value of the threshold; and
[0023] d) if the slope of the current in the first node during the first time period is less than the slope of the current in the first node during the second time period and if the current in the first node is greater than zero, reducing the distance between the first value and the second value of the threshold.
[0024] According to one embodiment, the third value increases in cases a) and d), and decreases in cases b) and c).
[0025] According to one embodiment, the third circuit is configured to compare the current in the first node with zero, and the third circuit includes an input coupled to the first node and an input coupled to a control voltage application node of the second transistor.
[0026] According to one embodiment, the third circuit is coupled to the application node of the control voltage through a circuit that applies a delay to the control signal. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The foregoing features and advantages as well as other features and advantages will be described in the following description of specific embodiments given by way of example and not limitation, with reference to the accompanying drawings, in which:
[0028] Figure 1 schematically illustrates an embodiment of a DC / DC voltage converter;
[0029] Figure 2 Shows the diagram Figure 1 Timing diagram of the converter's operation example;
[0030] Figure 3 Shows the diagram Figure 1 Other timing diagrams of the desired or theoretical operation and actual or practical operation of the converter;
[0031] Figure 4 An embodiment of a DC / DC voltage converter is shown;
[0032] Figure 5 Shown for illustration Figure 4 Timing diagram of the converter's operation;
[0033] Figure 6 An example of the variation of the variable DV is shown;
[0034] Figure 7 Another embodiment of a converter is shown; and
[0035] Figure 8 Show Figure 4 or Figure 7 An example of an implementation of a portion of a converter. DETAILED DESCRIPTION
[0036] In the various drawings, the same features are represented by the same reference numerals. In particular, common structural and / or functional features in various embodiments may have the same reference numerals, and may be provided with the same structure, size, and material properties.
[0037] For clarity, only the steps and elements that are useful for understanding the embodiments described herein are shown and described in detail.
[0038] Unless otherwise specified, when two elements are referred to as being connected together, it means a direct connection without any intermediate elements except conductors, and when two elements are referred to as being coupled together, it means the two elements may be connected or may be coupled through one or more other elements.
[0039] In the following disclosure, unless otherwise indicated, when referring to absolute position qualifiers, such as terms "front", "back", "top", "bottom", "left", "right", etc., or when referring to relative position qualifiers, such as terms "above", "below", "higher", "lower", etc., or when referring to orientation qualifiers, such as "horizontal", "vertical", etc., they all refer to the orientation shown in the figure.
[0040] Unless otherwise indicated, the expressions "about," "approximately," "substantially," and "on the order of" mean within 10%, and preferably within 5%.
[0041] In the following description, when referring to the voltage of a node, it is considered to mean the voltage between the node and a reference potential (usually ground) unless otherwise specified.
[0042] Figure 1 An example of a voltage converter 1 of the type to which the described embodiment applies is schematically shown. In this example, the converter 1 is a DC / DC converter which converts a DC supply voltage into a DC output voltage.
[0043] The converter 1 is configured to deliver a DC output voltage Vout. The converter comprises an output node 2 at which the voltage Vout is available.
[0044] The converter 1 is powered by a DC supply voltage Vbat.The converter 1 is then connected between a first conductive rail or node 3 set to a voltage Vbat and a second conductive rail or node 5 set to a reference potential GND.
[0045] The converter 1 is configured to deliver a voltage Vout at a value equal to a setpoint value. To this end, the converter 1 receives on an input node 7 a DC setpoint voltage Vref referenced to a potential GND having a value representative of, preferably equal to, the setpoint value of the voltage Vout.
[0046] In this example, the voltages Vout, Vbat, and Vref are positive.
[0047] In this example, the converter 1 is a buck converter, ie the set point value of the voltage Vout is smaller than the value of the voltage Vbat. In other words, the value of the voltage Vout is smaller than the value of the voltage Vbat.
[0048] The converter 1 comprises a first MOS (“Metal Oxide Semiconductor”) transistor 9, preferably a PMOS transistor (P-channel MOS transistor). As a variant, the transistor 9 can also be an NMOS transistor associated with a “bootstrap” system. The MOS transistor 9 is connected between the rail 3 and an internal node 11. In other words, a first conduction terminal (for example the source) of the transistor 9 is connected to the rail 3 and a second conduction terminal (for example the drain) of the transistor 9 is connected to the node 11.
[0049] The converter 1 further comprises a second MOS transistor 13, preferably an NMOS transistor (N-channel MOS transistor). The transistor 13 is connected between the node 11 and the rail 5. In other words, a first conduction terminal (e.g. source) of the transistor 13 is connected to the rail 5 and a second conduction terminal (e.g. drain) of the transistor 13 is connected to the node 11. As a variant, the NMOS transistor can be replaced by a diode or a Schottky diode.
[0050] Thus, transistors 9 and 13 are connected in series between rails 3 and 5 and are connected to one another at the level of internal node 11 .
[0051] The converter 1 comprises an inductive element or inductor 15. The inductor 15 is connected between the node 11 and the node 2.
[0052] The converter 1 comprises a control circuit 17. The circuit 17 is configured to implement or control an operation cycle of the converter 1 to regulate the voltage Vout such that the value of the voltage Vout is equal to a set point value Vref.
[0053] To this end, the circuit 17 comprises:
[0054] - a terminal 171 coupled to (preferably connected to) the node 7;
[0055] - a terminal 172 coupled to (preferably connected to) node 2;
[0056] - a terminal 173 coupled to (preferably connected to) the rail 3;
[0057] - a terminal 174 coupled to (preferably connected to) the rail 5;
[0058] a terminal 175 coupled to (preferably connected to) a control terminal or gate of transistor 9; and
[0059] —A terminal 177 coupled to (preferably connected to) a control terminal or gate of transistor 13 .
[0060] The converter 1 comprises an output capacitor 16 connected between the node 2 and the rail 5. As an example, the capacitance is of the order of 2.2 μF to 20 μF, or even more. Such an output capacitor acts as a filter. In other words, the converter output capacitor makes it possible to smooth the current present at the node 2 and to store the power supplied to the node 2 by the converter.
[0061] In operation, a load is connected between the node 2 and the rail 5 to be powered with the voltage Vout. The load comprises, for example, an input capacitor between the node 2 and the rail 5.
[0062] In this example, the converter 1 is configured to operate in pulse frequency modulation (discontinuous on mode). Then, the circuit 17 is configured to start the operation cycle of the converter 1 when the value of the voltage Vout is less than the set point value Vref and the two transistors 9 and 13 are in the off state. More specifically, at the beginning of each operation cycle, the circuit 17 is configured so that the transistor 9 is set to the on state and the transistor 13 is kept in the off state. Then, during the first time period TPon, power is stored in the inductor 15 and the capacitor 16, for example, when the transistor 9 is kept in the on state by the circuit 17, the power of each operation cycle is constant, and then the current IL flows through the inductor 15. At the end of the time period TPon, the circuit 17 is configured so that the transistor 9 is set to the off state and the transistor 13 is set to the on state. Then, in the second time period TNon, the power is transmitted back to the load connected to the output of the converter through the inductor 15 and the capacitor 16, for example, when the transistor 13 is kept in the on state by the circuit 17, the power of each operation cycle is constant, and the current IL in the inductor gradually decreases. At the end of the time period TNon, the circuit 17 is configured such that the transistor 13 is set to the off state.
[0063] The time period TNon is determined so that the time when the circuit 17 sets the transistor 13 to the off state corresponds to the time when the current IL flowing through the inductor 15 becomes empty. However, in practice, as will be described in further detail in the rest of this disclosure, this is not always correct, thus causing problems.
[0064] Figure 2 Show instructions Figure 1 Timing diagram of an example of desired operation of the converter.
[0065] Figure 2 The upper timing diagram shows the change of voltage Vout with time t, in volts V. Figure 2 The timing diagram at the bottom shows the corresponding change of the current IL flowing through the inductor 15 with time t.
[0066] At time t0 , transistors 9 and 13 are in the off state, current IL is null, and the value of voltage Vout is greater than its set point value, in this example the value of voltage Vref.
[0067] Between time t0 and a subsequent time t2 , the voltage Vout decreases, for example due to the fact that a load connected to the converter 1 consumes current and discharges the output capacitor.
[0068] At time t1 between times t0 and t2, the voltage Vout becomes smaller than its set point value Vref. This is detected by the circuit 17 of the converter 1, which then causes the transistor 9 to be set to an on state. The transistor 9 is turned on at time t2.
[0069] Therefore, starting from time t2, one terminal of the inductor 15 is connected to the node 2 and one terminal is coupled to the rail 3 through the transistor 9. The current IL flowing through the inductor 15 increases.
[0070] As a result, from time t2, current IL is delivered to node 2 and charges capacitor 16 between node 2 and rail 5. Voltage Vout increases and becomes greater than its set point value Vref.
[0071] At the next time t3, t3=t2+TPon, the circuit 17 sets the transistor 13 to the on state and the transistor 9 to the off state. At the time t3, the current in the inductor has a maximum value ILp.
[0072] Therefore, starting from time t3, one terminal of the inductor 15 is connected to the node 2 and one terminal is coupled to the rail 5 via the transistor 13. The current IL flowing through the inductor 15 decreases.
[0073] Although the current IL decreases starting from time t3 , if the current drawn by the load is less than the current IL supplied to node 2 , the capacitor between node 2 and rail 5 remains charged, and the voltage Vout keeps increasing.
[0074] At the next time t4, t4 = t3 + TNon, the circuit 17 sets the transistor 13 to the off state. It is considered here that the converter 1 operates as it should and the current IL is then empty at time t4. However, in practice this is not always true.
[0075] Starting from time t4 , current IL is nulled and voltage Vout decreases, similar to what would have happened at time t0 .
[0076] Although not shown herein, when the value of potential Vout falls back below its set point value Vref at a time after time t4 , circuit 17 implements a new operating cycle, such as described with respect to successive times t2 , t3 , and t4 .
[0077] Figure 3 Show instructions Figure 1 Another timing diagram of the converter operation. More specifically, timing diagram A ( Figure 3 The upper part) shows an ideal or theoretical example of the change of current IL, and the timing diagram B ( Figure 3 The two timing diagrams A and B show an example of operation in which the voltage Vout at the end of each operation cycle of the converter 1 is less than the voltage Vref for a plurality of consecutive operation cycles.
[0078] At time t30, although Figure 3 It is not shown in FIG. 1 , but the voltage Vout is less than the voltage Vref. The operation cycle starts with switching to the on state of the transistor 9. As a result, the current IL increases until the next time t31, t31 = t30 + TPon.
[0079] At time t31, the current IL reaches the maximum value ILp. In addition, the transistors 9 and 13 are switched to the off state and the on state, respectively. As a result, the current decreases until the next time t32, t32 = t31 + TNon.
[0080] In this example of ideal operation, the transistor 13 is switched to the off state at time t32, and the current IL becomes empty at time t32.
[0081] At time t32, voltage Vout is less than voltage Vref, and transistor 9 is switched to the on state, which marks the start of a new operation cycle. Then, current IL increases until the next time t33, t33=t32+TPon.
[0082] At time t33, the current IL reaches the value ILp. Furthermore, the transistors 9 and 13 are switched to the off state and the on state, respectively. As a result, the current decreases until the next time t34, t34=t33+TNon.
[0083] In this example of ideal operation, the transistor 13 is switched to the off state at time t34 , and the current IL becomes empty at time t34 .
[0084] At time t34, the voltage Vout is less than the voltage Vref, and the transistor 13 is switched to the on state, which marks the start of a new operation cycle. The current IL increases until the next time t35, t35=t34+TPon.
[0085] At time t35, the current IL reaches the value ILp. Furthermore, the transistors 9 and 13 are switched to the off state and the on state, respectively. As a result, the current decreases until the next time t36, t36=t35+TNon.
[0086] In this example of ideal operation, the transistor 13 is switched to the off state at time t36, and the current IL becomes empty at time t36.
[0087] At time t36, the voltage Vout is less than the voltage Vref, and a new operation cycle begins.
[0088] In the theoretical operation example shown by timing diagram A, at the end of each operation cycle, the switching of transistor 13 to the off state occurs when current IL becomes zero. Therefore, when one operation cycle is followed by a new operation cycle, in this new operation cycle, current IL increases from the zero value.
[0089] Timing diagram B shows a corresponding example of real operation of the converter 1. In this example of real operation, the actual situation is considered that the transistor 13 is not switched to the off state immediately after the time period TNon has passed since the last switching to the on state.
[0090] At time t40, the voltage Vout is less than the voltage Vref, and the operation cycle starts with switching to the on state of the transistor 9. As a result, the current IL increases until the next time t41, t41=t40+TPon.
[0091] At time t41, the current IL reaches a maximum value ILp. In addition, transistors 9 and 13 are switched to an off state and an on state, respectively. As a result, the current decreases until the next time t42, t42=t41+TNon. The current becomes empty at time t42. However, the switching of transistor 13 to an off state is only effective at time t43 after time t42. Therefore, between time t42 and t43, the current IL is negative and gradually decreases. In other words, before time t42, the current flows from node 11 to node 2 via inductor 15, becomes empty at time t42, and then flows from node 2 to node 11 via inductor 15 after time t42.
[0092] At time t43, voltage Vout is less than voltage Vref, and transistor 9 is switched to the on state at time t43, which marks the start of a new operation cycle. Then, current IL increases until the next time t44, t44=t43+TPon.
[0093] At time t44, since the time period TPon is constant in each cycle, the current IL reaches a value ILp' that is less than the maximum value ILp. In addition, the transistors 9 and 13 are switched to the off state and the on state, respectively. As a result, the current IL decreases until the next time t46, t46=t44+TNon, and at time t45 between time t44 and t46, the current IL becomes empty. In addition, the switching of the transistor 13 to the off state is only effective at time t47 after time t46. Therefore, between time t45 and t47, the current IL is negative and drops to a value lower (or higher in absolute value) than the value reached at time t43.
[0094] At time t47, voltage Vout is less than voltage Vref, and transistor 9 is switched to the on state at time t47, which marks the start of a new operation cycle. Then, current IL increases until the next time t48, t48=t47+TPon.
[0095] At time t48, the current IL reaches a value ILp" that is less than the value ILp'. In addition, the transistors 9 and 13 are switched to the off state and the on state, respectively. As a result, the current IL decreases until the next time t50, t50=t48+TNon, and at time t49 between time t48 and t50, the current IL becomes zero. In addition, the switching of the transistor 13 to the off state is effective only at time t51 after time t50. Therefore, between time t49 and t51, the current IL is negative and drops to a value lower (or higher in absolute value) than the value reached at time t47.
[0096] Due to the fact that in each operating cycle shown by the timing diagram B, the maximum value reached by the current IL (at times t41, t44 and t48) is getting lower and lower, the converter 1 does not provide enough power to the node 2 to regulate the voltage Vout at the value Vref, and the voltage Vout is, for example, getting lower and lower, which may cause problems. In addition, in each operating cycle shown by the timing diagram B, the negative value reached by the current IL (at times t43, t47 and t51) is getting lower and lower (or the absolute value is getting larger and larger), whereby the converter 1 samples more and more power from the node 2, which is not desirable. In fact, the value of the current supplied to the load, in particular the current peak value, gradually decreases from one cycle to another, which has a negative impact on the load power supply. In addition, although in theory the maximum value of the current IL can be reduced infinitely, in practice, in some cases, the transistor 13 may have been destroyed or damaged before this by the negative value of the current IL that the transistor 13 cannot connect between its conductive terminals.
[0097] Already combined Figure 3 The timing diagram B describes an actual operation example in which the transistor 13 is switched to the off state after the time when the current IL becomes zero.
[0098] In another practical operating example, not shown, in each operating cycle of a plurality of consecutive cycles implemented immediately after each other, the transistor 13 is switched to the off state while the current IL is non-empty and still positive. In this case, the current IL increases from an increasingly higher value in each operating cycle, whereby the current IL reaches an increasingly higher maximum value, and the operating cycle ends with an increasingly higher non-zero positive value of the current IL. Compared to the operation described with respect to the timing diagram B, this operation is less intrusive, because after a plurality of operating cycles, the voltage Vout will have recovered to its set point value Vref. Therefore, the next operating cycle will not be executed immediately, which will leave time for the current IL to become empty.
[0099] These two situations (i.e. Figure 3 The situation described in timing diagram B of FIG. 1 and other practical situations described above) are generally caused at least in part by the operating time of the components, such as the response or propagation time of the comparator.
[0100] Figure 4 One embodiment of a DC / DC voltage converter is shown. Figure 4 The converter includes about Figure 1 Describe the components and circuitry in more detail 17.
[0101] Circuit 17 includes circuit 300, such as a state machine. State machine 300 is configured to provide signals for controlling transistors 9 and 13 to respective terminals 175 and 177 to implement the following description of Figures 5 to 8 The described operation. To determine the control signals for transistors 9 and 13, the state machine 300 receives a number of signals. The circuit 300 is powered with a voltage Vbat and is connected between rails 3 and 5.
[0102] Circuit 17 includes a comparator 302 configured to pass a signal START on its output, which is representative of a comparison of the value of voltage Vout with its set point value. The output of comparator 302 is coupled to (e.g., connected to) circuit 300. When the value of voltage Vout is less than its set point value, signal START is in a first state, e.g., a high state, and when the value of voltage Vout is greater than its set point value, signal START is in a second state, e.g., a low state. Comparator 302 includes a first input, e.g., an inverting (-) input, configured to receive a voltage having a value representative of voltage Vout, and a second input, e.g., a non-inverting (+) input, configured to receive a voltage having a set point value representative of voltage Vout.
[0103] In this example where the value of voltage Vref is equal to the set point value of voltage Vout, comparator 302 is configured to compare voltage Vref with voltage Vout, a first input of comparator 302 is connected to node 2, and a second input of comparator 302 is connected to terminal 171 of circuit 17.
[0104] In this embodiment, the duration Tpon is determined by comparing the voltage ramp RP with the reference voltage V1. Therefore, the circuit 17 includes: a ramp generator 304, which is configured to deliver a potential ramp RP; a comparator 306, which is configured to compare the ramp RP with the voltage V1; and a circuit 307, which is configured to generate the voltage V1. The voltage RP is referenced to the reference voltage GND.
[0105] The generator 304 is powered by the potential Vbat and is connected between rails 3 and 5, in order to avoid overloading the drawing. Figure 4 These connections are not shown.
[0106] The generator 304 is controlled by the circuit 300 via the signal cmdP. More specifically, when the signal START is in the first state, the circuit 300 sets the transistor 9 to the on state and, at the same time, starts the voltage ramp RP via the signal cmdP. As an example, the signal cmdP is determined from the control signal supplied to the transistor 9 by the circuit 300, or is even identical to the control signal supplied to the transistor 9 by the circuit 300.
[0107] Each voltage ramp RP is, for example, a ramp that increases starting from the reference potential GND.
[0108] The ramp RP is transmitted to a first input, such as a non-inverting (+) input, of the comparator 306. A second input, such as an inverting (-) input, of the comparator 306 receives the voltage V1 and is coupled (preferably connected) to the output of the generator 307. The output of the comparator 306 transmits the signal COMP to the circuit 300.
[0109] exist Figure 4 , the voltage ramp RP is transmitted to the comparator 306 via the selection circuit 308 or multiplexer, which comprises two inputs, one output and one control terminal. The control terminal of the circuit 308 receives the signal sel from the circuit 300. Depending on the state of the signal sel, the circuit 308 transmits at its output the signal present at one or the other of its inputs. Thus, the output signal VRAMP of the circuit 308 follows the variations of the signal present at one or the other of the inputs of the circuit 308.
[0110] More specifically, when the circuit 300 sets the transistor 9 to the on state and the voltage ramp RP starts, the circuit 300 puts the signal sel in the first state so that the comparator 306 receives the voltage ramp RP.
[0111] Therefore, as long as the signal sel remains in the first state, the signal COMP is in the first state, for example, in the low state as long as the voltage RP is less than the voltage V1. Once the voltage RP becomes greater than the voltage V1, the signal COMP switches to the second state, for example, the high state.
[0112] The switching of signal COMP from its first state to its second state marks the end of time period TPon. Circuit 300 then controls transistor 9 to be in the off state and controls transistor 13 to be in the on state. Preferably, circuit 300 simultaneously causes the end of voltage ramp RP via signal cmdP.
[0113] The duration TNon is for example determined by comparing the voltage ramp RN with the voltage V2 generated by the circuit 307. Therefore, the circuit 17 comprises a ramp generator 310 configured to deliver the voltage ramp RN.
[0114] The generator 310 is powered by the potential Vbat and is connected between rails 3 and 5, in order to avoid overloading the drawing. Figure 4 These connections are not shown in FIG. 3. Generator 310 is controlled by circuit 300 via signal cmdN.
[0115] Voltages V1 and V2 are generated from voltage Vref by circuit 307. Voltages V1 and V2 are preferably different. Therefore, circuit 307 receives voltage Vref as input and outputs voltages V1 and V2. Voltage V2 is greater than voltage V1 and greater than voltage Vref. Voltage V1 is less than voltage V2 and less than voltage Vref. For example, voltages V1 and V2 are substantially equal to Vref-DV and Vref+DV, respectively, where DV is a variable.
[0116] exist Figure 4 In the example of , circuit 307 includes two outputs, which are coupled (preferably connected) to two inputs of selection circuit 309. Selection circuit 309 includes an output coupled (preferably connected) to one input (here, the inverting input (-)) of comparator 306. Selection circuit 309 also includes a control input that receives a control signal. When the control signal takes a first value, comparator 306 receives voltage V1, and when the control signal takes a second value, comparator 306 receives voltage V2. Figure 4 In the example of , the selection circuit receives the control signal sel, which is the same control signal as the selection circuit 308 .
[0117] At the end of the duration TPon, when the circuit 300 sets the transistor 9 to the off state and the transistor 13 to the on state, the circuit 300 also starts the voltage ramp RN via the signal cmdN. As an example, the signal cmdN is determined from the control signal provided by the circuit 300 to the transistor 13, or is even identical to the control signal delivered by the circuit 300 to the transistor 13.
[0118] Each voltage ramp RN is, for example, a voltage ramp that drops from a power supply voltage Vbat. The ramps RN and RP have slopes of relative sign. Therefore, one slope RP is increasing here, while the other slope RN is decreasing here. The slope RP, i.e. the increasing slope, is compared with the voltage V1, and the decreasing slope is compared with the voltage V2, which is greater than the voltage V1. The ramps RP and RN have slopes of equal absolute value.
[0119] The comparison of the ramp RN with the voltage V2 is implemented by the comparator 306. Then, the ramp RN is transmitted to the second input of the circuit 308. In addition, the circuit 300 is configured to switch the signal sel to the second state, while setting the transistor 9 to the off state and the transistor 13 to the on state, so as to transmit the ramp RN to the comparator 306.
[0120] Therefore, when each ramp RN drops from the voltage Vbat, as long as the signal sel remains in the second state, the signal COMP is in the second state, for example, it is in the high state as long as the voltage RN is greater than the voltage V2, and switches to the first state, for example, the low state, as long as the voltage RN becomes less than the voltage V2.
[0121] The switching of the signal COMP from its second state to its first state marks the end of the time period TNon. The circuit 300 then causes the transistor 13 to be set to the off state. Preferably, the circuit 300 simultaneously causes the end of the voltage ramp RN via the signal cmdN.
[0122] In an alternative embodiment not shown, the comparison of the voltage ramp RP with the voltage V1 is implemented by the comparator 306, and the comparison of the voltage ramp RN with the voltage V2 is implemented by an additional comparator provided by the circuit 17. This additional comparator then receives the voltage ramp RN on a first input (e.g. non-inverting (+)) and the voltage V2 on a second input (e.g. inverting (-)), the output of the additional comparator delivering to the circuit 300 a signal representing the comparison of the potential ramp RN with the voltage V2. In this variant, the circuits 308 and 309 are omitted, and the circuit 300 does not generate the signal sel. A person skilled in the art will be able to make the following description with respect to Figure 4 The above description of the converter is adapted to this alternative embodiment.
[0123] In another variant not shown, the ramp RP is a voltage ramp falling from the voltage Vbat, and / or the voltage ramp RN is a voltage ramp rising from the voltage GND. Figure 4 The converter is adapted to this situation.
[0124] The inventors have determined that the problem with timing diagram B, namely the fact that current IL has a value different from zero at the end of the cycle, is Figure 4 In the case of , it is mainly caused by the delay caused by the comparator(s) 306. The value of this delay TCOMP is known and is preferably substantially constant.
[0125] Figure 4 An embodiment uses a correction loop in which the values of the voltages or thresholds V1 and V2 are adjusted in each cycle when the current IL differs from zero. Preferably, the thresholds V1 and V2 are modified only once in each cycle. The voltages V1 and V2 are adjusted so that, at the end of the cycle, the current IL converges from one cycle to another towards the value zero. More specifically, the thresholds V1 and V2 are adjusted to reach values that enable a current substantially equal to zero to be obtained at the end of the cycle.
[0126] In order to regulate the voltages V1 and V2, the circuit 307 receives a control signal d generated by the circuit or circuit assembly 20. The receipt of the signal d results in a switching of the variable DV and thus in a switching of the values of the voltages V1 and V2.
[0127] According to one embodiment, the variable DV can take a predetermined number of values, preferably a fixed number of values. Therefore, the circuit 307 is configured to provide the voltages V1 and V2 from the variable DV in the predetermined number of values. For example, the variable DV can take one of 32 values, for example 32 values different from 0. For example, the different values of the variable DV are consecutive values, which are separated by the same constant, for example the constant is equal to 16mV.
[0128] Figure 5 A timing diagram illustrating the operating cycle of the graph converter is shown. Figure 5 An output voltage VRAMP of the comparator 308 , a current IL , and an output signal COMP of the comparator 306 are shown.
[0129] The operating cycle of the converter includes a single ramp RP generated by the generator 304 and a single ramp RN generated by the generator 310. More specifically, between time t60 and time t61, the voltage VRAMP adopts the value of the ramp RP, corresponding to the magnetization step, and between time t61 and time t62, the voltage VRAMP adopts the value of the ramp RN, corresponding to the demagnetization step. Thus, the operating cycle of the converter corresponds to the period between time t60 and time t62.
[0130] At time t60, voltage VRAMP increases from a value preferably equal to 0 as ramp RP changes. Thus, at time t60, voltage VRAMP is less than value Vref and thresholds V1 and V2. In particular, voltage VRAMP is less than threshold value delivered at output of circuit 309, namely threshold value V1. Thus, signal COMP has a first value, e.g. Figure 5 In the case of , the low value corresponds to the binary value 0.
[0131] At time t63 between time t60 and time t61, voltage VRAMP reaches threshold value V1. At time t61, after response time TCOMP of the comparator, output signal COMP of comparator 306 takes a second value, for example, a high value corresponding to binary value 1, indicating the fact that voltage VRAMP is greater than threshold value V1. Time t61 occurs after duration TCOMP has passed since time t63, i.e., time t61=t63+TCOMP. Therefore, between time t63 and t61, voltage VRAMP still follows the change of ramp RP and keeps increasing, exceeding value Vref.
[0132] Between times t60 and t61, multiplexer 308 (mux1) outputs signal RP generated by generator 304. Control signal sel of multiplexer 308 generated by circuit 300 takes a value that instructs the multiplexer to output the signal originating from generator 304. This value of signal sel also corresponds to the value that instructs circuit 309 to output threshold value V1.
[0133] At time t61, the voltage VRAMP becomes equal to the ramp RN, for example by modifying the value of the control signal sel of the multiplexer 308. Therefore, the voltage VRAMP takes a high value, higher than the value Vref and the value of the threshold V2, for example the value Vbat, and decreases as the ramp RN changes. Therefore, the value of the voltage VRAMP remains greater than the value Vref. In addition, at time t61, the output of the circuit 309 passes the threshold V2.
[0134] At time t64, voltage VRAMP reaches value V2. Output signal COMP of comparator 306 takes a first value, which indicates the fact that voltage VRAMP is less than output value (here threshold value V2) of circuit 309 at time t62 after response time TCOMP. Time t62 occurs after duration TCOMP has passed from time t64, i.e., time t62=t64+TCOMP. Therefore, between time t64 and t62, voltage VRAMP still follows the change of ramp RN and keeps increasing, exceeding value Vref.
[0135] Between times t61 and t62, multiplexer 308 (mux1) outputs signal RN generated by generator 310. Control signal sel of multiplexer 308 generated by circuit 300 takes a value that instructs the multiplexer to output a signal originating from generator 310. This value of signal sel also corresponds to a value that instructs circuit 309 to output threshold value V2.
[0136] Between times t60 and t61, transistor 9 is turned on and transistor 13 is turned off, causing current IL to increase, preferably from a current substantially equal to zero. Between times t61 and t62, transistor 9 is turned off and transistor 13 is turned off, causing current IL to decrease from the current peak value obtained at the end of the magnetization step. Figure 3 As mentioned above, the ideal current value IL at the end of the cycle is zero. Figure 5 In the example of , the value of the current IL at the end of the cycle is different from zero. More specifically, the value of the current at time t62 is less than zero. This means that the current IL did not increase long enough during magnetization, in other words, this means that the threshold V1 is too low and / or the current IL decreased for too long during demagnetization, in other words, this means that the threshold V2 is too low. Therefore, the thresholds V1 and V2 are adjusted in the next cycle.
[0137] Reference again Figure 4 , the circuit 20 includes a circuit 200 configured to determine whether the current IL is equal to zero at the end of each cycle. In other words, when the output of the comparator 306 determines that the voltage VRAMP has reached the value of the threshold V2, the circuit 200 compares the value of the current IL with the value zero. Preferably, the circuit 200 determines whether the current is greater than zero or less than zero. For example, the circuit 200 includes two outputs, which are coupled (preferably connected) to the circuit 204 of the circuit 20, and the circuit 204 is configured to generate a control signal d. The circuit includes, for example, an output that conveys information that the current is greater than zero, and an output that conveys information that the current is less than zero. If the current alternates between being greater than zero and less than zero, the circuit determines that the current is sufficiently close to zero.
[0138] An example of implementation of the circuit 200 is described in patent application EP2819288.
[0139] Circuit 200 includes a first input for receiving current IL. The first input is, for example, coupled (preferably connected) to node 11. Circuit 200 includes a second input for determining when circuit 200 compares the value of current IL with the value zero. The second input is preferably coupled to node 177, i.e., a node for receiving the control signal of transistor 13. Thus, circuit 200 determines the end of the cycle and, in turn, determines the time to compare current IL with the value zero when the control signal of transistor 13 commands transistor 13 to transition from the on state to the off state.
[0140] The second input of the circuit 200 is preferably coupled to the node 177 via a circuit 202, which is configured to delay the signal for a time period Δ. The duration Δ is determined to ensure that the value of the current IL compared with the value zero effectively corresponds to the value of the current at the end of the cycle. Thus, the value of the duration Δ is at least equal to (preferably substantially equal to) the time between the initial time when the control signal on the node 177 causes the state change and the time when the current is compared with the value zero at the initial time.
[0141] As a variant, the circuit 200 may comprise a single output. The output of the circuit 200 is then coupled (preferably connected) to a circuit 204 of the circuit 20, the circuit 204 being configured to generate the control signal d.
[0142] Circuit 200 delivers one or more signals from its (one or more) outputs, the one or more signals indicating whether the current IL is substantially equal to zero or different from zero. Based on the signal, circuit 204 determines whether thresholds V1 and V2 should be adjusted. Preferably, circuit 200 outputs information about whether the current is less than zero or greater than zero.
[0143] Circuit 204 provides a signal d to change the thresholds V1 and V2, as follows: Figure 5 as described.
[0144] The circuit 20 preferably includes a circuit 206 configured to compare the voltage Vbat with twice the voltage Vout. The comparison result corresponds to the comparison result between the value Vbat-Vout and the value Vout, in other words, to the comparison result between the value (Vin-Vout) / L and Vout / L, L being the value of the inductance L. In other words, the circuit 206 is configured to compare the slope of the current IL during magnetization and during demagnetization. Thereafter, when referring to the comparison of the slope, the comparison is performed without taking into account the sign of the slope. In other words, the comparison is performed based on the absolute value of the slope. In the following description, the term "slope" refers to the absolute value slope value.
[0145] Circuit 206 is coupled (preferably connected) to an input of node 11 for delivering voltage Vout. Circuit 206 is also coupled (preferably connected) to an input of rail 3. The link between circuit 206 and rail 3 is not Figure 4 Circuit 206 is coupled (preferably connected) to the output of circuit 204 .
[0146] If the value Vbat is close enough to the value 2*Vout, for example, they are equal to each other within a range of 200 mV, in other words, the voltage Vbat is within a range of 2*Vout-Vm to 2*Vout+Vm, Vm being equal to 200 mV, for example, this means that the slopes of the current IL during magnetization and demagnetization are substantially equal in their sign. Therefore, the modification of the threshold value will not allow the current IL to converge to zero at the end of the cycle. Therefore, the output of the circuit 206 is coupled (preferably connected) to the input of the circuit 204. The output of the circuit 206 delivers a signal indicating whether the value of the voltage Vbat is within the above range.
[0147] Preferably, if the signal provided by circuit 206 to circuit 204 indicates that voltage Vbat is within the above range, the control signal d generated by circuit 204 indicates to circuit 307 that thresholds V1 and V2 should take predetermined values, respectively, regardless of the value of current IL. For example, in a given cycle, the predetermined value is the value that the threshold had in the previous cycle. According to another example, the predetermined value is a fixed value, such as the lowest possible value of DV.
[0148] Figure 6 An example of the change of the variable DV is shown. Figure 5 In the example of , the variable DV can take 11 different values DV1, DV2, DV3, DV4, DV5, DV6, DV7, DV8, DV9, DV10 and DV11. The value DV1 corresponds, for example, to the value zero, so the variable DV can take ten values different from zero.
[0149] The variable DV successively assumes all possible values until it reaches the value DV and thus the thresholds V1 and V2 , making it possible to have a current IL substantially equal to zero at the end of the cycle.
[0150] exist Figure 6 In the example of , during the first cycle corresponding to the first phase P1 , the variable DV has the value DV6 . Thus, the thresholds V1 and V2 have the values Vref-DV6 and Vref+DV6 , respectively. If the current IL is different from zero at the end of the cycle, the variable DV takes the next value, which is preferably separated from DV6 by the value of the above constant. Figure 6 In the example of , the next value is the value DV7, which is equal to the value DV6 plus the value of the constant. Thus, during the second cycle corresponding to the second phase P2, the variable DV has the value DV7.
[0151] In each cycle, if the current at the end of the cycle is different from zero, i.e. actually greater than zero or less than zero, the value DV is modified, which corresponds to another phase. According to one embodiment, the value DV is modified in a first direction if the current is less than zero, and in a second direction opposite to the first direction if the current is greater than zero. For example, if the current is less than zero, the value DV takes a smaller value, corresponding to a lower phase, and if the current is greater than zero, the value DV takes a larger value. If the value DV reaches a maximum value (here DV11, corresponding to phase P3) or a minimum value (here DV1), the direction of modification of the value DV is for example reversed. The minimum value corresponds, for example, to the variable DV being equal to zero, i.e. to the thresholds V1 and V2 being equal to Vref.
[0152] For example, the end-of-cycle current is considered to be substantially equal to zero when it alternates between being below zero and being above zero from one cycle to another. For a value DV, this corresponds to alternating between two values of DV. Figure 5 , the value DV alternates between the values DV3 and DV4, which indicates that the end-of-cycle current is substantially equal to zero.
[0153] As a variant, the variable DV may be initially equal to zero, ie the thresholds V1 and V2 may be initially equal to the voltage Vref.
[0154] Figure 7 Another embodiment of a converter is shown. The converter includes Figure 4 Elements common to the embodiments. Only the elements that differ between the embodiments will be described below.
[0155] Figure 7 Examples and Figure 4 The embodiment of the present invention differs in that, when the signal COMP takes a second value (here a low value), the variation of the voltages V1 and V2 during a given cycle depends on the value of the current IL. More specifically, the variation depends on the sign of the current IL and the slope of the current IL during the magnetization period, i.e., when the voltage VRAMP follows the variation of the rising ramp (RP), and during the demagnetization period, i.e., when the voltage VRAMP follows the variation of the falling ramp (RN).
[0156] The slope of the current IL during magnetization is equal to (Vbat-Vout) / L. The slope of the current IL during demagnetization is equal to -Vout / L. Thus, when the value (Vbat-Vout) / L is greater than the value Vout / L, i.e., the voltage Vbat is greater than the voltage 2*Vout, it means that the effect of a given duration during magnetization is greater than the effect during demagnetization. In other words, the increase in the current IL during the time period T during magnetization is greater than the decrease in the current IL during the same time period T during demagnetization.
[0157] Likewise, when the value (Vbat-Vout) / L is less than the value Vout / L, i.e., the voltage Vbat is greater than the voltage 2*Vout, it means that the effect of a given duration during demagnetization is greater than that during magnetization. In other words, the increase in current IL during the time period T during magnetization is less than the decrease in current IL during the same time period T during demagnetization.
[0158] The reduction in threshold V1 means that voltage VRAMP reaches threshold V1 faster as ramp RP changes. Therefore, current IL increases during a shorter period of time and reaches a lower value.
[0159] The increase in threshold value V1 means that voltage VRAMP reaches threshold value V1 at a later time as ramp RP changes. Therefore, current IL increases during a longer period of time and reaches a higher value.
[0160] Similarly, an increase in threshold V2 means that voltage VRAMP reaches threshold V2 faster as ramp RN changes. Therefore, current IL decreases during a shorter period of time and reaches a higher value.
[0161] The reduction in threshold value V2 means that voltage VRAMP reaches threshold value V2 at a later time as ramp RN changes. Therefore, current IL decreases during a longer period of time and reaches a lower value.
[0162] In the case where voltage V1=Vref-DV and voltage V2=Vref+DV, the change of voltages V1 and V2 is achieved by changing the variable DV. Therefore, threshold V1 changes by the same value as threshold V2 in a relative manner. In other words, if threshold V1 increases by a certain value, threshold V2 decreases by the same value, and vice versa.
[0163] In the first case, consider that the variable DV is increased between a first cycle and a second cycle following the first cycle, i.e. the thresholds V1 and V2 are further from each other in the second cycle than in the first cycle. The voltage VRAMP reaches the threshold V1 faster during magnetization of the second cycle than during magnetization of the first cycle, and reaches the voltage V2 faster during demagnetization of the second cycle than during demagnetization of the first cycle. Therefore, the value of the current IL increases less during magnetization of the second cycle than during magnetization of the first cycle, and decreases less during demagnetization of the second cycle than during demagnetization of the first cycle. The ramps RP and RN have the same slope, and the magnetization and demagnetization are shortened by the same period Dur between the first cycle and the second cycle. This change has an effect in particular on the value of the current IL at the end of the cycle.
[0164] If the slope of the current during magnetization is greater than the slope of the current during demagnetization, the difference between the increase of the current IL during the first cycle magnetization and the increase of the current IL during the second cycle magnetization is greater than the difference between the decrease of the current IL during the first cycle demagnetization and the decrease of the current IL during the second cycle demagnetization. Therefore, the value of the current IL at the end of the second cycle is less than the end value of the first cycle, which corresponds to the lower value of the variable DV.
[0165] In other words, if the slope of the current during magnetization is greater than the slope of the current during demagnetization, in other words, if the value (Vbat-Vout) / L is greater than the value Vout / L, and the value of the current IL at the end of the cycle is greater than 0, the circuit 307 is configured to increase the variable DV in the next cycle, that is, the distance between the voltages V1 and V2 increases.
[0166] Similarly, in the first case, if the slope of the current during demagnetization is greater than the slope of the current during magnetization, the difference between the increase in current IL during the first cycle magnetization and the increase in current IL during the second cycle magnetization is smaller than the difference between the decrease in current IL during the first cycle demagnetization and the decrease in current IL during the second cycle demagnetization. Therefore, the value of current IL at the end of the second cycle is greater than the end value of the first cycle, which corresponds to the smaller value of variable DV.
[0167] In other words, if the slope of the current during demagnetization is greater than the slope of the current during magnetization, in other words, if the value (Vbat-Vout) / L is greater than the value Vout / L, and the value of the current IL at the end of the cycle is less than 0, the circuit 307 is configured to increase the variable DV in the next cycle, that is, the distance between the voltages V1 and V2 increases.
[0168] In the second case, it is considered that the variable DV decreases between the second cycle after the first cycle, i.e. the thresholds V1 and V2 are less distant from each other during the second cycle than during the first cycle. The voltage VRAMP reaches the threshold V1 more slowly during the magnetization of the second cycle than during the magnetization of the first cycle, and reaches the voltage V2 more slowly during the demagnetization of the second cycle than during the demagnetization of the first cycle. Therefore, the value of the current IL increases more during the magnetization of the second cycle than during the magnetization of the first cycle, and decreases more during the demagnetization of the second cycle than during the demagnetization of the first cycle. The ramps RP and RN have the same slope, and the magnetization and demagnetization extend for the same duration between the first cycle and the second cycle. This change has an influence in particular on the value of the current IL at the end of the cycle.
[0169] If the slope of the current during magnetization is greater than the slope of the current during demagnetization, the difference between the increase of the current IL during the first cycle magnetization and the increase of the current IL during the second cycle magnetization is greater than the difference between the decrease of the current IL during the first cycle demagnetization and the decrease of the current IL during the second cycle demagnetization. Therefore, the value of the current IL at the end of the second cycle is greater than the value at the end of the first cycle, which corresponds to the smaller value of the variable DV.
[0170] In other words, if the slope of the current during magnetization is greater than the slope of the current during demagnetization, in other words, if the value (Vbat-Vout) / L is greater than the value Vout / L, and the value of the current IL at the end of the cycle is less than 0, the circuit 307 is configured to reduce the variable DV in the next cycle, that is, the distance between the voltages V1 and V2 decreases.
[0171] Similarly, in the second case, if the slope of the current during demagnetization is greater than the slope of the current during magnetization, the difference between the increase in current IL during the first cycle magnetization and the increase in current IL during the second cycle magnetization is less than the difference between the decrease in current IL during the first cycle demagnetization and the decrease in current IL during the second cycle demagnetization. Therefore, the value of current IL at the end of the second cycle is less than the value at the end of the first cycle, which corresponds to a lower value of variable DV.
[0172] In other words, if the slope of the current during magnetization is greater than the slope of the current during demagnetization, in other words, if the value (Vbat-Vout) / L is greater than the value Vout / L, and the value of the current IL at the end of the cycle is less than 0, the circuit 307 is configured to reduce the variable DV in the next cycle, that is, the distance between the voltages V1 and V2 decreases.
[0173] exist Figure 7 In the embodiment, Figure 4 The circuit 20 is replaced by a circuit 20', which is configured to provide a signal d for controlling the circuit 307. Figure 7 As described, circuit 307 adjusts thresholds V1 and V2 according to signal d. Thus, signal d is generated by circuit 20' to cause an increase in the value of DV or a decrease in the value of DV according to the value of current IL at the end of the cycle and a comparison of the slope of current IL during magnetization and demagnetization.
[0174] Circuit 20 includes circuit 200' configured to determine whether current IL is less than, greater than, or equal to a value of zero. Figure 4 Similar to the circuit 200 of FIG. 1 , the circuit 200 ′ comprises a first input coupled (preferably connected) to the node 11 for delivering the voltage Vout, and a second input capable of determining the end of the operation cycle. Figure 4As shown in the embodiment of FIG. 1 , the second input is coupled to node 177 via circuit 202 .
[0175] exist Figure 7 In the example of , the circuit 200 ' includes two outputs. One output transmits a signal EAR indicating that the current IL was positive at the end of the previous cycle. For example, the signal EAR takes a first value to indicate that the current was positive at the end of the cycle, and takes a second value to indicate that the current was not positive at the end of the cycle. The other output transmits a signal LAT indicating that the current IL of the previous cycle was negative. For example, the signal LAT takes a first value to indicate that the current was negative at the end of the cycle, and takes a second value to indicate that the current was not negative at the end of the cycle. If the signals EAR and LAT remain at the second value, it means that the current IL is equal to zero at the end of the cycle.
[0176] The circuit 20' also includes a circuit 206', which is configured to compare the voltage Vbat with twice the voltage Vout. The comparison result corresponds to the comparison result between the value Vbat-Vout and the value Vout, in other words, to the comparison result between the value (Vin-Vout) / L and Vout / L. In other words, the circuit 206 is configured to compare the slope of the current IL during magnetization and during demagnetization.
[0177] Circuit 206' is coupled (preferably connected) to an input of node 11 that delivers voltage Vout. Circuit 206' is also coupled (preferably connected) to an input of Rail 3. The link between circuit 206' and Rail 3 is not Figure 7 Shown in.
[0178] and Figure 4 Similar to the circuit 206 in , the circuit 206 ′ delivers information according to which the value Vbat and the value 2*Vout are sufficiently close, for example equal to each other within 200 mV.
[0179] The circuit 206' is also configured to deliver information about whether the slope of the current IL during magnetization is greater or less than the slope of the current IL during demagnetization. In other words, the circuit 206' is configured to deliver information about whether Vbat is less than or greater than 2*Vout.
[0180] The circuit 20' further comprises a circuit 204', which is configured to generate a signal d. The circuit 204' is coupled (preferably connected) at its input to the outputs of the circuits 200' and 206'. As previously described, the circuit 204' thus generates the signal d based on the following information:
[0181] - information generated by the circuit 200 ′ indicating the sign of the current IL at the end of the cycle;
[0182] - information generated by the circuit 206 ′ indicating whether the slope of the current IL during magnetization is greater or less than the slope of the current IL during demagnetization; and
[0183] - Information generated by circuit 206' indicating whether the values of Vbat and 2*Vout are sufficiently close, for example equal to each other within 200 mV.
[0184] In other words, according to one embodiment, if:
[0185] - the voltage Vbat is in the previously described value range, for example between 2*Vout-200mV and 2*Vout+200mV, the thresholds V1 and V2 are set to predetermined values independent of the values of the thresholds in the previous cycle, or in another example, the values in the previous cycle are maintained;
[0186] The voltage Vbat is in the previously described range of values, for example between 2*Vout-200mV and 2*Vout+200mV, and the voltage Vbat is greater than the voltage 2*Vout, in other words, in this example, if the voltage Vbat is greater than 2*Vout+200mV:
[0187] a) if the current IL is less than zero, then the value DV is reduced, in other words, the threshold V1 is increased and the threshold V2 is decreased; and
[0188] b) if the current IL is greater than zero, then the value DV is increased, in other words, the threshold V1 is decreased and the threshold V2 is increased;
[0189] The voltage Vbat is in the previously described range of values, for example between 2*Vout-200mV and 2*Vout+200mV, and the voltage Vbat is less than the voltage 2*Vout, in other words, in this example, if the voltage Vbat is less than 2*Vout-200mV:
[0190] a) if the current IL is greater than zero, then the value DV is reduced, in other words, the threshold V1 is increased and the threshold V2 is decreased; and
[0191] b) If the current IL is less than zero, the value DV is increased, in other words, the threshold V1 is decreased and the threshold V2 is increased.
[0192] As a variant, thresholds V1 and V2 can be modified independently of each other. In this case, if the current IL is less than zero at the end of a cycle, threshold V1 is reduced or threshold V2 is increased in the next cycle. If the current IL is less than zero at the end of a cycle, threshold V1 is increased or threshold V2 is reduced in the next cycle.
[0193] Figure 7The advantage of the illustrated embodiment is that it is able to converge faster towards the value DV, so that at the end of the cycle there is an empty current IL.
[0194] Figure 8 Show Figure 4 or Figure 7 More specifically, Figure 8 An implementation example of the circuit 307 and the circuit 309 is shown.
[0195] Circuit 307 includes current source 50, resistor 52, resistor 54, and current source 56, which are sequentially coupled in series between node 3 and node 5. Therefore, source 50 is coupled (preferably connected) between node 3 and a terminal of resistor 52. Therefore, source 50 and resistor 52 are coupled (preferably connected) to node 62. The other terminal of resistor 52 is coupled (preferably connected) to center node 53. One terminal of resistor 54 is coupled (preferably connected) to center node 53. The other terminal of resistor is coupled (preferably connected) to one terminal of current source 56. Therefore, source 56 and resistor 54 are coupled (preferably connected) to node 64. The other terminal of current source 56 is coupled (preferably connected) to node 5.
[0196] Circuit 307 receives voltage Vref at input 58. Circuit 307 also receives voltage Vref at input 58. Figure 8 The two nodes in represent the signal d received at the input 60 .
[0197] A node 62 between source 50 and resistor 52 forms the output of circuit 307 and is coupled (preferably connected) to the input of selection circuit 309. The voltage at node 62 is equal to voltage Vref+DV, ie, voltage V2.
[0198] Similarly, node 64 between source 56 and resistor 54 forms the output of circuit 307 and is coupled (preferably connected) to the input of selection circuit 309. The voltage at node 64 is equal to voltage Vref-DV, ie, voltage V1.
[0199] Current sources 50 and 56 are controlled by signal d. Thus, signal d is generated to configure the sources and regulate the voltages on nodes 62 and 64.
[0200] Various embodiments and variations have been described. Those skilled in the art will appreciate that certain features of these various embodiments and variations may be combined, and those skilled in the art will recognize other variations.
[0201] Finally, based on the functional indications given above, the actual implementation of the described embodiments and variants is within the capabilities of a person skilled in the art.
Claims
1. A voltage converter, comprising: a first transistor connected between the first node of the converter and a second node configured to receive a supply voltage; a second transistor connected between the first node and a third node configured to receive a reference potential; a first circuit configured to control the first transistor and the second transistor; as well as a comparator coupled to the first circuit and configured to compare a first voltage with a threshold, the first voltage being equal to a first rising slope during a first time period and being equal to a second falling slope during a second time period, the threshold having a first value during the first time period and the threshold having a second value during the second time period, the first value and the second value being variable values; wherein the first time period is determined by comparing the first voltage equal to the first rising slope and the threshold having the first value; and The second time period is determined by comparing a second voltage equal to the second decreasing slope and the threshold having the second value. 2 . The converter of claim 1 , further comprising a second circuit configured to generate the threshold.
3. The converter according to claim 2, wherein: The converter comprises a third circuit configured to compare a value of a current in the first node to a value of zero; as well as The second circuit is configured to modify the value of the threshold during an operating cycle if the value of the current in the first node was different from zero at the end of a previous cycle. 4 . The converter according to claim 1 , wherein a first value of the threshold is equal to a reference voltage plus a third value, and a second value of the threshold is equal to the reference voltage minus the third value, the third value being a variable value. The converter of claim 4 , wherein the third value varies between a plurality of predetermined values.
6. The converter according to claim 5, wherein the predetermined values of the third value are consecutively separated by the same fourth constant value.
7. The converter according to claim 3, wherein: The converter further includes a fourth circuit configured to determine whether the power supply voltage is within a range of a value of the voltage on the first node minus 200 mV to a value of the voltage on the first node plus 200 mV; as well as The second circuit is configured to maintain the first value of the threshold and the second value of the threshold for a next cycle when the power supply voltage is within the range during a cycle, or to make the first value of the threshold and the second value of the threshold have predetermined values during the next cycle. 8 . The converter of claim 3 , further comprising a fifth circuit configured to determine whether a value of the current in the first node is less than zero or greater than zero.
9. The converter of claim 8, further comprising a sixth circuit configured to determine whether a slope of the current in the first node during the first time period of a cycle is greater or less than the slope during the second time period of a cycle.
10. The converter of claim 9, wherein the second circuit is configured to change the first value of the threshold and the second value of the threshold so that: a) if the slope of the current in the first node during the first time period is greater than the slope of the current in the first node during the second time period, and if the current in the first node is less than zero, the distance between the first value of the threshold and the second value of the threshold is reduced; b) if the slope of the current in the first node during the first time period is greater than the slope of the current in the first node during the second time period, and if the current in the first node is greater than zero, the distance between the first value of the threshold and the second value of the threshold is increased; c) if the slope of the current in the first node during the first time period is less than the slope of the current in the first node during the second time period, and if the current in the first node is less than zero, the distance between the first value of the threshold and the second value of the threshold is increased; as well as d) if the slope of the current in the first node during the first time period is less than the slope of the current in the first node during the second time period, and if the current in the first node is greater than zero, the distance between the first value of the threshold and the second value of the threshold is reduced.
11. A converter according to claim 10, wherein the first value of the threshold is equal to the reference voltage plus a third value, and the second value of the threshold is equal to the reference voltage minus the third value, the third value is a variable value, and wherein in a) and d), the third value is increased, and in b) and c), the third value is decreased.
12. The converter of claim 3, wherein the third circuit is configured to compare the current in the first node with zero, and the third circuit comprises an input coupled to the first node and an input coupled to a node applying a control voltage to the second transistor.
13. The converter according to claim 12, wherein: The third circuit is coupled to the node to which the control voltage is applied through a circuit that applies a delay to a control signal.
14. A method for operating a voltage converter, the voltage converter comprising: a first transistor connected between the first node of the converter and a second node configured to receive a supply voltage; a second transistor connected between the first node and a third node configured to receive a reference potential; and a first circuit configured to control the first transistor and the second transistor; the method comprising, during each operation cycle: During a first time period, comparing a first voltage with a threshold value by a comparator, the first voltage being equal to a first rising slope during the first time period, the threshold value having a first value during the first time period; and During a second time period, the first voltage is compared with the threshold value by the comparator, the first voltage is equal to a second falling slope during the second time period, the threshold value has a second value during the second time period, and the first value of the threshold value and the second value of the threshold value are variable values; wherein the first time period is determined by comparing the first voltage equal to the first rising slope and the threshold having the first value; and The second time period is determined by comparing a second voltage equal to the second decreasing slope and the threshold having the second value. The method of claim 14 , further comprising generating the threshold value by a second circuit.
16. The method according to claim 15, further comprising: comparing, by a third circuit, the value of the current in the first node to a value of zero; as well as The value of the threshold is modified during an operating cycle by the second circuit in response to the value of the current in the first node being different from zero at the end of a previous cycle. 17 . The method of claim 14 , wherein a first value of the threshold is equal to a reference voltage plus a third value, and a second value of the threshold is equal to the reference voltage minus the third value, the third value being a variable value. The method of claim 17 , wherein the third value varies between a plurality of predetermined values. The method according to claim 18 , wherein the predetermined values of the third value are consecutively separated by the same fourth constant value.
20. The method of claim 16, wherein: determining, by a fourth circuit, whether the power supply voltage is within a range from a value of the voltage on the first node minus 200 mV to a value of the voltage on the first node plus 200 mV; as well as In response to the power supply voltage being within the range during a cycle, the first value of the threshold and the second value of the threshold are maintained by the second circuit for a next cycle, or predetermined values are used by the second circuit for the first value of the threshold and the second value of the threshold during the next cycle.
21. The method of claim 16, further comprising determining, by a fifth circuit, whether a value of the current in the first node is less than zero or greater than zero.
22. The method of claim 21, further comprising determining, by a sixth circuit, whether a slope of the current in the first node during the first time period of a cycle is greater or less than the slope during the second time period of a cycle.
23. The method according to claim 22, further comprising: In response to the slope of the current in the first node during the first time period being greater than the slope of the current in the first node during the second time period, and the current in the first node being less than zero, the first value of the threshold and the second value of the threshold are changed by the second circuit so that the distance between the first value of the threshold and the second value of the threshold is reduced.
24. The method of claim 22, further comprising: In response to the slope of the current in the first node during the first time period being greater than the slope of the current in the first node during the second time period, and the current in the first node being greater than zero, the first value of the threshold and the second value of the threshold are changed by the second circuit so that the distance between the first value of the threshold and the second value of the threshold is increased.
25. The method of claim 22, further comprising: In response to the slope of the current in the first node during the first time period being less than the slope of the current in the first node during the second time period, and the current in the first node being less than zero, the first value of the threshold and the second value of the threshold are changed by the second circuit so that the distance between the first value of the threshold and the second value of the threshold is increased.
26. The method of claim 22, further comprising: In response to the slope of the current in the first node during the first time period being less than the slope of the current in the first node during the second time period, and the current in the first node being greater than zero, the first value of the threshold and the second value of the threshold are changed by the second circuit so that the distance between the first value of the threshold and the second value of the threshold is reduced.
