Method and device for precise current feedback control in an electronic device

Abstract

Device for precise current feedback control in an electronic device, wherein the device comprises: - an operational amplifier (205) with a first input, a second input and an output, - a reference potential generator (401) coupled to one of these first and second inputs, - a current feedback designed to generate a measured potential, indicating an output current of the electronic device, at the other of these first and second inputs, and - a control loop (206) coupled to the output of the operational amplifier (205), wherein the control loop is configured to control the generation of the output current in the electronic device; wherein the control loop (206) is configured to selectively switch the electronic device into a mode with an output current of zero (302) and into at least one further mode with a non-zero output current (301), characterized in that the control loop (206) is designed for this purpose: - to couple a calibration potential optionally with the other of the first and second inputs during the mode with a zero output current (302), - to measure (304) the response of the operational amplifier (205) to the reception of the calibration potential at the other of the first and second inputs and to store it as a correction factor (305), and - to use such a stored correction factor to correct the way in which the control loop (206) controls the generation of the output current during at least one of these other modes.

Description

Field of invention

[0001] The present invention relates to the technical field of electronic devices in which current feedback control is used to control an output current. In particular, the invention relates to improving the accuracy of such current feedback control despite suboptimal performance of the components used. Background of the invention

[0002] Many electronic devices use current feedback control to ensure that the magnitude of their output current is as it should be. An LED driver can be considered an example of such an electronic device, its purpose being to generate an output current that illuminates one or more solid-state semiconductor light sources to a desired brightness. However, under certain assumptions, the following description can be generalized to any electronic device that performs current feedback control.

[0003] US 2010 / 0289424 A1 relates to a method for voltage matching with dimming control for LED driver applications. The method includes providing an input voltage, generating an output voltage to drive a plurality of LED channels, determining a lowest voltage from the LED channels, generating a comparator voltage by comparing the lowest voltage with a feedback reference voltage, generating a summed voltage to stabilize the output voltage, and generating a PWM voltage to control the output voltage.

[0004] WO 2019 / 233996 A1 discloses a method for calibrating a current measuring device of a control device for a light source. The current measuring device comprises measuring instruments which are coupled to the light source during operation of the control device. The method comprises automatically setting a predetermined current by the measuring instruments during operation of the control device, automatically detecting an electrical quantity at the measuring instruments while the predetermined current is set, and automatically determining a correction value for the current measuring device as a function of the detected electrical quantity.

[0005] DE 102 36 872 A1 discloses a switch arrangement for a controllable power supply. The switch arrangement has an input for connecting a DC voltage and an output for the controlled supply of a load. Furthermore, the switch arrangement has a DC controller with capacitors and / or inductors and means for controlling the current at the output of the circuit arrangement, operating on the principle of pulse modulation. Additionally, the circuit arrangement has means for preventing the inductors and / or capacitors of the DC controller from discharging during pulse pauses of the controlled current.

[0006] DE 20 2017 104 810 U1 relates to a converter circuit. The converter circuit comprises a converter circuit with a clocked switch for generating an output current, a control circuit designed to control the switch for regulating the output current by means of a control loop, and measuring means for determining an output voltage and / or the output current. The control circuit is designed to change at least one parameter of the control loop when a change in the output voltage or output current is detected.

[0007] Fig. Figure 1 shows an example of current feedback control in an LED driver. The LED driver has two power stages: the PFC (Power Factor Correction) stage 101, which converts an input voltage VIN into an intermediate voltage VBUS, and the output stage 102, which generates an output voltage VOUT. Both stages 101 and 102 typically contain a corresponding switching converter, such as a buck converter, boost converter, flyback converter, or other converter. The light sources 103 are connected to receive an output current IOUT from the output stage 102. A current-sense resistor 104, which has a low resistance, is located along the output current loop. A comparator 105 compares the potential difference across the current-sense resistor 104 with a reference potential VREF. The output signal of the comparator 105 goes to a controller 106, which controls the output stage 102 accordingly.In this exemplary embodiment, the controller 106 is also responsible for generating the reference potential VREF, although this is not necessarily the case, and the reference potential VREF can also come from another source.

[0008] The basic idea of ​​the current feedback control scheme in Fig. One approach is to rely on the comparator 105 comparing the potentials of its two inputs and monitoring its output. By appropriately adjusting the reference potential VREF, the controller effectively establishes a control loop in which the desired magnitude of the output current of the output stage 102—and consequently a suitable voltage drop across the current-sensing resistor 104—causes the comparator 105 to generate a desired output signal. If the output stage 102 is of the type that produces a smooth, continuous output current, the magnitude of this current should assume a stable value that makes the two input potentials of the comparator 105 equal. If the output stage 102 produces a sawtooth-like output current, the switching output of the comparator 105 can be used to generate the appropriate switching signals for the output stage 102.

[0009] Problems can arise if the comparator 105 does not operate accurately. A major cause of inaccuracies is the so-called offset voltage of the comparator 105. The actual operating principle of the comparator 105 may be such that it does not respond to identical input potentials, but rather to input potentials that differ by an offset, producing a zero output signal. Due to random variations in component properties, the offset voltage can be either positive or negative, and its magnitude can be difficult or even impossible to predict. Even worse, the sign and magnitude of the offset voltage can change depending on the temperature and / or aging of the components. For example, LED drivers must operate in ambient temperatures ranging from freezing temperatures to hot summer days.Furthermore, they heat up during operation, so the actual temperature of the components can be anything between freezing cold and a defined safe upper limit, such as +85 degrees Celsius. Another cause of inaccuracies is the quiescent current of comparator 105, which causes similar errors during operation as the offset voltage. Summary

[0010] Accordingly, one objective is to present a circuit and a method for reducing any adverse effects caused by the inherent inaccuracy in the operation of a comparator in a current feedback control loop. Another objective is to present such a circuit and method capable of accommodating dynamically changing operating conditions. Furthermore, it is also an objective to achieve these advantageous goals without resorting to the use of complex electronic circuits and expensive components.

[0011] These and other advantageous objectives are achieved by a device having the features of claim 1 and a method having the features of claim 9. The electronic device is equipped with a calibration circuit with which it can independently calibrate the operation of a comparator in its current feedback control loop during periods in which no actual output current needs to be generated.

[0012] According to one embodiment, a device for precise current feedback control is provided in an electronic device. The device comprises: an operational amplifier with a first input, a second input, and an output; a reference potential generator coupled to one of these first and second inputs; a current feedback loop configured to generate a measured potential, indicating an output current of the electronic device, at the other of these first and second inputs; and a control loop coupled to the output of the operational amplifier. The control loop is configured to control the generation of the output current in the electronic device and is also configured to selectively switch the electronic device into a mode with a zero output current and into at least one other mode with a non-zero output current.The control loop is also configured to selectively couple a calibration potential to the other of the first and second inputs during the zero output current mode, to measure and store the operational amplifier's response to receiving the calibration potential at the other of the first and second inputs as a correction factor, and to use such a stored correction factor to correct the way in which the control loop controls the generation of the output current during at least one of these other modes.

[0013] According to one embodiment, the control loop is configured to control the generation of the reference potential. This offers at least the advantage that the control loop can use the controllable reference potential as a means of setting a desired output current level and also as a means of determining the correction factor values.

[0014] According to one embodiment, the electronic device has a control input for receiving one or more control signals indicating a desired output current value. The control loop can then be configured to control the generation of the reference potential based on one or more such received control signals. This offers at least the advantage that the controllable reference potential can be used to implement changes in the output current decided elsewhere.

[0015] According to one embodiment, the electronic device includes a temperature sensor configured to provide temperature readings to the control loop. The control loop can then be configured to record mappings of temperature readings provided by the temperature sensor to corresponding stored correction factors. The control loop can be configured to use an actually measured temperature value and these recorded mappings as part of this use of a stored correction factor to correct the way in which the control loop manages the generation of the output current during at least one of the other modes. This has at least the advantage that the calibration of the current feedback loop can be temperature-specific.

[0016] According to one embodiment, the electronic device includes a calibration potential generator for generating the calibration potential. The current feedback and the calibration potential generator can then share at least some components. This offers the advantage of reducing the total number of components required for all desired functionalities of the device.

[0017] According to one embodiment, the current feedback loop includes a current-sense resistor on an output current path of the electronic device, as well as a level-shifting and filtering circuit configured to generate the measured potential as a level-shifted and filtered version of a potential difference across the current-sense resistor. The calibration potential generator may include a controllable switch, controlled by the control loop, between a source potential and an intermediate point in the level-shifting and filtering circuit. This offers at least the advantage that a relatively simple circuit can be used to perform the calibration of the current feedback loop.

[0018] According to one embodiment, the electronic device is a lighting driver that supplies the output current to one or more solid-state semiconductor light sources. This offers at least the advantage that the idea of ​​the invention can be applied to very precise control of lighting under various conditions.

[0019] According to one embodiment, the electronic device has at least one switching current converter for generating the output current. The control loop can then be configured to control the generation of the output current by controlling the generation of switching pulses for at least one power switch in the at least one switching current converter. This offers at least the advantage that known and robust technology can be used for this purpose.

[0020] A second aspect involves providing a method for precise current feedback control in an electronic device. This method includes: - Providing a reference potential at an input of an operational amplifier, - during a mode with a zero output current of the electronic device, coupling a calibration potential to another input of the operational amplifier, - Measuring the operational amplifier's response to receiving the calibration potential at the other input and storing this response as a correction factor, and - Using such a stored correction factor to correct the generation of an output current of the electronic device during at least one other mode with a non-zero output current, in which a measured potential indicating an output current of the electronic device is coupled to the other of the first and second inputs of the operational amplifier.

[0021] According to one embodiment, the method involves changing the reference potential during at least one other mode with a non-zero output current in order to influence the magnitude of the electronic device's output current. This offers at least the advantage that a control circuit in the device can use the controllable reference potential as a means of setting a desired output current level and also as a means of determining correction factor values.

[0022] According to one embodiment, the method includes the following steps: - Measuring temperature values ​​both during the mode with a zero output current and during at least one other mode with a non-zero output current, - during the mode with a zero output current, recording mappings of temperature readings to corresponding stored correction factors, and - During at least one other mode with a non-zero output current, using an actually measured temperature value and the recorded mappings as part of using a stored correction factor to correct the way the output current is generated during at least one other mode. This has at least the advantage that the calibration of the current feedback loop can be temperature-specific.

[0023] According to a third aspect, the use of a method of the type described above is provided for in order to perform precise current feedback control in a lighting driver. Brief description of the drawings

[0024] The accompanying drawings, which are provided for a further understanding of the invention and form part of this description, illustrate embodiments of the invention and, together with the description, help to explain the principles of the invention. The drawings illustrate: Fig. 1. A principle of current feedback control according to the state of the art, Fig. 2 a principle of precise current feedback control, Fig. 3. A principle of precise current feedback control in the form of a state diagram, Fig. 4 a principle of precise current feedback control, Fig. 5 a circuit capable of performing current feedback control, Fig. 6 a circuit capable of performing precise current feedback control, Fig. 7 examples of dependencies between duty cycle and output current. Detailed description

[0025] Fig. Figure 2 schematically depicts an electronic device equipped with a device for precise current feedback control. The electronic device is designed to generate an output current IOUT and deliver it to a load 203. One or more power stages may be involved in generating the output current. Fig. Figure 2 represents an output stage 202 as an example; the output stage 202 can, for example, have one or more controllable switching converters.

[0026] The device for precise current feedback control in the electronic device according to Fig. 2 features an operational amplifier 205 with two inputs. These can be referred to below as the first input and the second input, although there is no need to specify a particular order of the inputs. A reference potential generator is connected to one of the inputs to provide the operational amplifier 205 with a reference potential VREF. In the embodiment according to Fig. 2, which describes a general principle of accurate current feedback control, it is not necessary to define the reference potential generator more precisely.

[0027] Current feedback is designed such that another, measured potential is generated at the other input of the operational amplifier 205. The measured potential indicates the output current IOUT of the electronic device. In particular, the measured potential can indicate the magnitude (also called amplitude) of the output current IOUT, although it is possible that it also indicates other properties. In the electronic device according to Fig. 2. A current-sensing resistor 204 is present on the output current path. The left end of the current-sensing resistor 204 is connected to ground, and thus in Fig. 2 the measured potential at the right end of the current measuring resistor 204 is essentially equal to the voltage drop across the current measuring resistor 204.

[0028] This output current value can represent an average or filtered output current, or a different processed value instead of the unprocessed, instantaneous output current. This possibility is in Fig. Figure 2 shows the filter block 208. It is possible that the current feedback generates two or more measured potentials, each indicating a different measured property of the output current. For example, there may be a filtered potential indicating a general amplitude of the output current and an unfiltered potential indicating an instantaneous magnitude of the output current.

[0029] A control loop 206 is coupled to the output of the operational amplifier 205. Furthermore, the control loop 206 is configured to control the generation of the output current of the electronic device. This latter task of the control loop 206 is described in Fig. 2 schematically represented as an arrow from block 206 to block 202. One aspect of such control is that the control loop 206 is designed to selectively switch the electronic device into a mode with a zero output current and into at least one other mode with a non-zero output current.

[0030] Assuming, for example, that the load 203 consists of light sources, the mode with a zero output current is a mode in which the light sources emit no light. Other modes, that is, modes with a non-zero output current, are modes in which the light sources emit at least some amount of light. In a simple lighting device, there can be exactly two modes: one with the light off and one with the light on. In a more versatile, dimmable lighting device, the mode with a non-zero output current can refer to any amount of output current, that is, to any selected brightness of the emitted light.For multi-colored light sources, such as LEDs with different color temperatures, the modes with a non-zero output current can also include modes in which selected currents are caused to flow through the different light sources to emit multi-colored light and / or composite white light with different color temperatures.

[0031] The zero-output mode referred to here should not be confused with the brief, high-frequency off-periods that occur when output stage 202 applies pulse-width modulation to cause the light sources to emit what, due to the inherently integrative nature of the human visual system, appears as continuous light at less than full brightness. When the electronic device is set to zero-output mode, it remains in this state for periods that even the human sensory system interprets as a switched-off load. Conversely, if light sources are considered the load, then the zero-output mode is one that a human observer would interpret as a switched-off light.Since the integration time of the human visual system is approximately on the order of 25 milliseconds, it can be defined that the zero-output current mode lasts at least 100 milliseconds, preferably more than one second, and even more preferably more than several seconds. In many cases, the zero-output current mode can last for a considerably longer period, such as several hours or even days, weeks, or months.

[0032] In contrast to the state-of-the-art circuit in Fig. 1 is in Fig. 2. The control loop 206 is configured to selectively couple a calibration potential VCAL to the other input of the operational amplifier 205, to which the measured potential is otherwise coupled. The control loop 206 is configured to do this during the mode with a zero output current. In this way, the control loop 206 is configured to measure the response of the operational amplifier 205 to the reception of the calibration potential at the other input and to store it as a correction factor. The possibility of such selective coupling of the calibration potential VCAL is described in Fig. 2 schematically shown with a controllable On / On switch 207.

[0033] One reason for measuring and storing such correction factors is that the control loop 206 can later use such a stored correction factor to correct the way in which it controls the generation of the output current. Such use of one or more correction factors occurs during at least one of the other modes, that is, the modes with a non-zero output current.

[0034] During the zero-output-current mode, it is not necessary to generate the measured potential indicating the output current, since the control loop itself initiates this mode and thus knows that the output current is zero anyway. Therefore, the control loop 206 can use at least part of this zero-current period for the so-called calibration of the current feedback loop. By appropriately choosing the reference potential VCAL, particularly with respect to the reference potential VREF, the control loop 206 can determine the extent to which the offset voltage (and quiescent current) of the operational amplifier 205 causes errors in the current feedback loop.

[0035] The schematic representation in Fig. Section 2 makes no statement regarding the generation of the reference potential VREF. It is possible, but not essential, that the control loop 206 be configured to control the generation of the reference potential VREF. Such control may be useful for the normal operation of the electronic device. As a non-restrictive example, the electronic device may have a control input for receiving one or more control signals indicating a desired output current magnitude. If the control loop 206 is configured to control the generation of the reference potential VREF based on one or more such received control signals, it provides a relatively simple and intuitive way to act in accordance with the control signals.Since the task of current feedback control is to determine when the measured potential equals the reference potential (or, in some cases, to measure the instantaneous difference between the measured potential and the reference potential) and to react accordingly in generating the output current, a change in the reference potential directly leads to a change in the desired output current.

[0036] In Fig. 2. The electronic device also includes a temperature sensor 209, which is configured to provide temperature readings to the control loop 206. If the control loop 206 remains informed (at least approximately) about the temperature, this information can be used in various ways, such as limiting or stopping the operation of the electronic device by the control loop 206 if the measured temperature is outside (or threatens to exceed) any predetermined safety limits. The temperature sensor 209 can be an integral part of the control loop 206 or an external component connected to a suitable input of the control loop 206.

[0037] Advantageously, but not necessarily, the control loop 206 is configured to record the mapping of temperature measurements provided by the temperature sensor 209 to corresponding stored correction factors. In other words, the control loop 206 can know which temperature value was supplied by the temperature sensor 209 when a specific correction factor was measured and stored. This operating principle has its advantages, since the operational amplifier 205 can exhibit a different offset voltage (and / or a different quiescent current) at different temperatures. By recording these mappings, the control loop 206 determines which measured and stored correction factor corresponds to which temperature.Or, more generally, the control loop 206 gains knowledge of how the error-causing properties of the operational amplifier 205 behave depending on the temperature and how they could be counteracted at each measured temperature.

[0038] Therefore, the control loop 206 can be configured to use an actually measured temperature value and these recorded mappings as part of its previously explained task, namely, using a stored correction factor to correct the way it controls the generation of the output current. This is of course true during at least one of those other modes when the electronic device generates a non-zero output current.

[0039] One noteworthy detail is that the temperature sensor 209 does not need to be perfectly accurate, as long as any assumed errors in the temperature measurement do not vary arbitrarily from one measurement to the next, but rather occur deterministically and systematically. This is because, when the recorded mappings are used later, it can be assumed that even a faulty temperature measurement will only be similarly erroneous to the measurement taken at the time the mapping was recorded, so that the resulting correction of the output current will still function as intended.

[0040] Fig. Figure 3 presents an example of precise current feedback control in the form of a state diagram. State 301 on the left represents the so-called normal operation of an electronic device, corresponding to a mode with a non-zero output current. A transition to state 302 in the middle signifies the initiation of a mode with a zero output current. This can occur, for example, in response to receiving a corresponding command, such as a standby command. Additionally or alternatively, other factors can trigger the transition from state 301 to state 302, such as adherence to a predefined operating schedule.

[0041] The calibration loop through states 302, 303, 304, 305 and back to 302 represents the selective coupling (state 303) of a calibration potential VCAL with one of the inputs of an operational amplifier, the measurement (state 304) of the operational amplifier's response to the reception of the calibration potential at that input, and the storage (state 305) of the received response as a correction factor. Fig. 3. The trigger factor that starts the loop through these states is referred to as a necessity. This could mean, for example, that a current correction factor is unavailable and therefore there is a need for one. Other possible trigger factors include, but are not limited to, the mere fact that sufficient time is available to perform a calibration, the occurrence of a trigger based on a timer, or any other deterministic way in which the program executed by the control loop makes it operate.In this sense, even a seemingly random decision of the control loop can be described as deterministic, since the control loop, as a programmable device, will never operate in a truly random manner, but will, for example, use a deterministically operating random number generator to create pseudorandom triggers for initiating certain actions.

[0042] Although this in Fig. Unless otherwise shown in Figure 3, performing the calibration loop also includes providing a reference potential at an input of the operational amplifier whose operation is to be calibrated.

[0043] When a temperature sensor is coupled to the control loop, as for example in Fig. 2. In the description above, a temperature reading supplied to the control loop by the temperature sensor can act as a trigger factor. For example, when the transition from state 301 to state 302 occurs, or when the measured temperature changes during a mode with a zero output current, the control loop may determine that the current temperature value is one for which there is no (or at least no recently created) mapping to a correction factor. In other words, the calibration loop may not have been executed, at least not recently, at this current temperature.

[0044] Between states 303 and 304; 304 and 305; and 305 and 302, there are in Fig. No specific trigger factors are shown. The transitions between these states can occur routinely, for example, as part of the process by which the control loop executes its program. The transition from state 302 to state 301 can occur in response to a command requiring the resumption of normal operation of the electronic device and a mode with a non-zero output current. Similar to the reverse transition, other trigger factors are possible, such as adherence to a predefined schedule.

[0045] If the calibration loop has been completed at least once through states 302, 303, 304, 305 and back to 302, at least one stored correction factor is available to the control loop. This means that, as illustrated by substate 306, the control loop can use such a stored correction factor to correct the way it controls the generation of the output current during at least one mode with a non-zero output current.

[0046] Fig. Figure 4 presents an example of an electronic device equipped with a precise current feedback control device. The electronic device according to Fig. 4 is the one in Fig. 2 partially similar, so that the same parts and sizes are represented with the same reference symbols.

[0047] In Fig. 4 The electronic device has a controllable reference potential generator 401 and a control coupling from the control loop 206 to it. An output of the controllable reference potential generator 401 is coupled to an input of the operational amplifier 205. In addition, the electronic device has according to Fig. 4. A control input 403 is provided for receiving one or more control signals indicating a desired output current value. The control loop 206 is configured to control the generation of the reference potential in the reference potential generator 401 based on one or more such received control signals. The control loop 206 can also be configured to perform other types of control operations based on control signals received via the control input 403.

[0048] The current feedback – designed to generate a measured potential indicating an output current of the electronic device – and a calibration potential generator – for selectively generating a calibration potential – are in Fig. 4 is not shown separately, but as a common block 402. This underscores the fact that the current feedback and the calibration potential generator may share at least some components. An example of this is given below in connection with the Fig. 5 and Fig. 6 shown.

[0049] Fig. Figure 5 shows a simplified circuit diagram of an example of an electronic device equipped with current feedback control. The upper part of Fig. Figure 5 shows a simple buck converter where a power switch SW1 controls the current flow from an input voltage source Vin to an inductor L1 and then to a load. The return path of the output current passes through a current-sensing resistor RSENSE. As required for the buck converter topology, there is another switch SW2 that optionally allows current flow in a loop containing the inductor L1. The additional switch SW2 could be replaced by the so-called freewheeling diode, which is often shown as part of the buck converter topology. An output capacitor Cout acts as a filter, smoothing out fluctuations in the output voltage of the electronic device. A driver circuit DRV controls the operation of switches SW1 and SW2.

[0050] Components in the lower right part of the Fig. 5 refer to the current feedback control. An operational amplifier OA has a first input, a second input, and an output. A reference potential generator is coupled to one of the inputs. In Fig. The reference potential generator includes capacitor C1, voltage divider resistors R1 and R2, and the associated couplings. One of the outputs of the microcontroller control loop is located in... Fig. The component labeled PWM (pulse-width modulation) 5 is connected to ground via the voltage divider R1, R2, and a midpoint of the voltage divider R1, R2 is coupled to the upper node of capacitor C1. In this way, the microcontroller (µC) can influence the voltage across capacitor C1. The microcontroller can, for example, output a pulse-width modulated (PWM) signal via the PWM output. The duty cycle of such a PWM signal essentially defines the voltage across capacitor C1, which—due to the connections to the ground potential—is equivalent to the generated reference potential. In such a case, the microcontroller can easily change the value of the reference potential by altering the duty cycle of the PWM signal.

[0051] Similar to the above regarding Fig. The principle shown in point 1 is the current feedback to a point between the load and the current-sensing resistor RSENSE. Fig. 5 is coupled. The coupling from this point goes via resistors Rfb1 and Rfb2 to the other input of the operational amplifier OA. A capacitor Cfb1 is connected to ground from the point between resistors Rfb1 and Rfb2. In this way, the current feedback is designed so that a measured potential is generated at the other input of the operational amplifier OA, indicating an output current of the electronic device. The effect of capacitor Cfb1 is to act as a filter, so that the measured potential indicates an average value of the output current. The in Fig. The current feedback shown in Figure 5 also includes a direct coupling between the load-side end of the current-sensing resistor RSENSE and an input AD2 of the control loop µC, so that the latter is given the opportunity to remain informed about the instantaneous value of the output current.

[0052] The resistor Rfb3 and the capacitors Cfb2 and Cfb3 form the resistive-capacitive feedback between the output and the inverting input of the operational amplifier OA.

[0053] Another input, AD3, of the microcontroller (µC) control circuit is coupled to the output of the operational amplifier OA and configured to control the generation of the electronic device's output current. For this purpose, a coupling between an output of the µC control circuit and a control input of the driver circuit DRV in the buck converter is provided. The µC control circuit can use this coupling to influence how the driver circuit DRV generates the switching pulses for switches SW1 and SW2. The µC control circuit can selectively put the electronic device into a zero-output-current mode by causing the driver circuit DRV to stop all switching pulses to the power switch SW1.Similarly, the microcontroller (µC) can put the electronic device into a mode with a non-zero output current by instructing the driver circuit (DRV) to generate synchronized switching pulses to switches SW1 and SW2 at a desired frequency and duty cycle. It should be noted again that one possible way to implement PWM dimming is to generate the synchronized switching pulses to switches SW1 and SW2 at a high frequency, for example, several tens or hundreds of kilohertz, and to repeatedly activate and deactivate the DRV driver circuit as a whole at a low frequency, for example, on the order of one or at most a few kilohertz.Such repeated deactivation periods of the driver circuit for PWM dimming are not equivalent to any mode with a zero output current, as they occur at a frequency which - although significantly below the frequency of the switching pulses to switches SW1 and SW2 - is nevertheless high enough to hopefully leave the respective short periods without light unnoticed by the human visual system.

[0054] For comparison, shows Fig. 6 A simplified circuit diagram of an example of an electronic device equipped with a precise current feedback control device. The upper part of Fig. Figure 6 shows the same elementary downconverter as the Fig. 5. Components in the lower right part of Fig. 6 belong to the device for precise current feedback control and contain important additions to the basic solution in Fig. 5. The coupling from the point between the load and the current-sensing resistor RSENSE proceeds via resistors R6, Rfb1, and Rfb2 to the other input of operational amplifier OA. The supply voltage Vcc is coupled to the point between resistors R6 and Rfb1 via two paths: via resistor R5 and via the collector-emitter path of an NPN transistor Q1 and resistor R4. The point between the emitter of transistor Q1 and resistor R4 is coupled to another input, AD1, of the microcontroller. A connection from the output of operational amplifier OA exists via resistor R3 to the base of transistor Q1. The base of transistor Q1 is also coupled via the pin labeled CAL to an internal open-collector switch of the microcontroller.

[0055] Resistors R5, R6, Rfb1, and Rfb2, as well as capacitor Cfb1, can be characterized as a level-shifting and filtering circuit designed to generate the measured potential indicating the output current as a level-shifted and filtered version of a potential difference across the current-sensing resistor RSENSE. The calibration potential generator of the electronic device according to Fig. 6 features a controllable switch, implemented as transistor Q1. It is controlled by the control loop µC and is coupled between the source potential Vcc and an intermediate point in the level shifting and filtering circuit.

[0056] In accordance with the above with reference to the Fig. 2, Fig. 3 and Fig. The principle described in section 4 is the control loop µC according to Fig. 6. The control loop is configured to selectively couple a calibration potential to the upper (inverting) input of the operational amplifier OA. For the reasons already explained, it is advantageous to cause the control loop µC to do this in a mode with a zero output current of the electronic device. Furthermore, the control loop is configured to measure the response of the operational amplifier OA to receiving the calibration potential at the other input and to store it as a correction factor. The control loop µC is configured to use such a stored correction factor to correct the way it controls the generation of the output current during at least one of the other (non-zero output current) modes.

[0057] An example of such a use of stored correction factors can be given by first considering some mathematical relationships between the essential quantities involved. In general, the identifier V can be err for the combined voltage error caused by the offset voltage and quiescent current of the operational amplifier OA, and it can be assumed that the control loop µC forms the reference potential across the capacitor C1 by applying a pulse-width modulated signal with duty cycle D between ground potential (0 V) and V to the voltage divider R1, R2 cc spends. In Fig. 5 is the average size I LED of the output current ILED=1Rsense⋅(D⋅Vcc⋅R2R1+R2−Verr).

[0058] It can be assumed that the maximum output current I max at a duty cycle of 100%, or D max= 1, occurs. Using this relationship, equation (1) can be written as follows. ILED=D⋅Imax−VerrRsense.

[0059] Next can Fig. 6 is considered and a situation is assumed in which the control loop µC of the circuit according to Fig. 6. The electronic device has been switched to a mode with zero output current and the internal open-collector switch at the CAL pin is open. The latter means that transistor Q1 is conducting and—as defined by the operational amplifier OA—supplies a current through resistor R4, which puts the operational amplifier OA into a balanced state. If the control circuit µC simultaneously has a duty cycle D cal When used to generate the reference voltage across capacitor C1, a potential V is applied to input AD1, which is coupled between the emitter of Q1 and resistor R4. AD1 observe.

[0060] According to a first embodiment, to calibrate for the voltage error in the operational amplifier OA, the control loop aims to adjust the potential V AD1 to make it assume a predetermined constant value, such as 1 V. This is done by varying D. cal and determine at which duty cycle V AD1 assumes the specified constant value. For the voltage error V err The following can be written Verr=R2R1+R2⋅Vcc⋅Dcal−(R6+Rsense)⋅(VAD1⋅R5+Vcc⋅R4)R4⋅R5+(R4+R5)⋅(R6+Rsense).

[0061] On the other hand, in normal operation the output current of the circuit is Fig. 6 ILED=1Rsense⋅(D⋅Vcc⋅R2R1+R2−Verr−Rsense+R6R5+R6+Rsense⋅Vcc).

[0062] The formula for the output current I LED , in which the error caused by the operational amplifier OA (by the offset voltage and the quiescent current) is compensated, reads ILED=1Rsense⋅[(D−Dcal)⋅Vcc⋅R2R1+R2 +(R6+Rsense)⋅(VAD1⋅R5+Vcc⋅R4)R4⋅R5+(R4+R5)⋅(R6+Rsense) −R6+RsenseR5+R6+Rsense⋅Vcc].

[0063] Solving equation (5) for D yields a formula for the duty cycle, which must be used to determine the error-corrected value of I. LED to obtain. Thus, by increasing the value of D cal stored as a correction factor by which V AD1 Once the specified constant value has been achieved, it is possible to calculate any value of the duty cycle D that leads to error-corrected output current values, as long as the conditions - such as the temperature - that were present during the determination and storage of the D are met. cal Values ​​prevailed, remain the same.

[0064] Fig. Figure 7 shows some examples of dependencies between the duty cycle D (in Fig. 7 expressed as a percentage) and the actual, error-corrected output current I LED They might have been found. The curves in Fig. The curves were calculated assuming that the maximum output current of the electrical device is 1 A and the resistance of the current-sense resistor Rsense is 200 mΩ. Other assumed component values ​​are R4 = 62 kΩ, R1 = 2.7 kΩ, R2 = 174 Ω, R5 = 330 kΩ, and R6 = 100 Ω. Curve 701 represents the assumption that no error is caused by the operational amplifier OA. Curve 702 assumes that the offset voltage and the quiescent current of the operational amplifier together produce a voltage error V. err of +1 mV, and curve 703 assumes a voltage error V errfrom -1 mV. The intersection points enclosed by a circle show that, in order to obtain an error-corrected output current of 10 mA, a duty cycle of 1.0%, 1.5%, or 2.0% must be used if the current conditions are such that they result in a voltage error V. err cause +1 mV, 0 mV or -1 mV.

[0065] According to a second embodiment, to calibrate for the voltage error in the operational amplifier OA, the control loop takes on the roles of V. AD1 and D cal swap them. In other words, instead of D cal to vary so that V AD1 If a predetermined constant value is assumed for calibration, the control loop can always use a predetermined constant duty cycle D. cal use and the value of V AD1 measure, which occurs under the prevailing conditions. Since the expression for D, obtained by solving equation (5), has a dependence on V AD1Including, the principle remains the same: the stored values ​​of V AD1 These serve as correction factors that can be used to calculate, in any case, the value of the duty cycle D, which corresponds to the error-corrected value of the output current I. LED leads.

[0066] It is obvious to a person skilled in the art that the basic idea of ​​the present invention can be implemented in various ways as technology advances. The invention and its embodiments are therefore not limited to the examples described above, but can vary within the scope of the claims.

Claims

[1] Device for precise current feedback control in an electronic device, wherein the device comprises: - an operational amplifier (205) with a first input, a second input and an output, - a reference potential generator (401) coupled to one of these first and second inputs, - a current feedback designed to generate a measured potential, indicating an output current of the electronic device, at the other of these first and second inputs, and - a control loop (206) coupled to the output of the operational amplifier (205), wherein the control loop is configured to control the generation of the output current in the electronic device; wherein the control loop (206) is configured to selectively place the electronic device into a mode with an output current of zero (302) and into at least one other mode with a non-zero output current (301), characterized by , that the control loop (206) is designed for this purpose: - to couple a calibration potential optionally with the other of the first and second inputs during the mode with a zero output current (302), - to measure (304) the response of the operational amplifier (205) to the reception of the calibration potential at the other of the first and second inputs and to store it as a correction factor (305), and - to use such a stored correction factor to correct the way in which the control loop (206) controls the generation of the output current during at least one of these other modes. [2] Device according to claim 1, wherein the control loop (206) is configured to control the generation of the reference potential. [3] Device according to claim 2, wherein - the electronic device has a control input (403) for receiving one or more control signals indicating a desired output current magnitude, and - the control loop (206) is designed to control the generation of the reference potential on the basis of one or more such received control signals. [4] Device according to any one of the preceding claims, wherein: - the electronic device has a temperature sensor (209) designed to provide temperature readings for the control loop (206), - the control loop (206) is designed to record assignments of temperature measurements provided by the temperature sensor (209) to corresponding stored correction factors, - the control loop (206) is configured to use an actually measured temperature value and these recorded mappings as part of this use of a stored correction factor to correct the way in which the control loop (206) controls the generation of the output current during at least one of those other modes (301). [5] Device according to any one of the preceding claims, wherein: - the electronic device has a calibration potential generator (Q1, µC) for generating the calibration potential, and - the current feedback and the calibration potential generator (Q1, µC) have at least some common components (402). [6] Device according to claim 5, wherein: - the current feedback includes a current-sensing resistor (204) on an output current path of the electronic device and a level-shifting and filtering circuit (208; R5, R6, Rfb1, Rfb2, Cfb1) configured to generate the measured potential as a level-shifted and filtered version of a potential difference across the current-sensing resistor (204), and - the calibration potential generator (Q1, µC) has a controllable switch (Q1) which is controlled by the control loop (206) between a source potential and an intermediate point in the level shifting and filtering circuit (208; R5, R6, Rfb1, Rfb2, Cfb1). [7] Device according to any one of the preceding claims, wherein: - the electronic device is a lighting driver that supplies the output current to one or more solid-state semiconductor light sources. [8] Device according to claim 7, wherein: - the electronic device has at least one switching converter for generating the output current, and - the control loop is designed to control the generation of the output current by controlling the generation of switching pulses for at least one power switch in the at least one switching converter. [9] Method for precise current feedback control in an electronic device, the method comprising: - Providing a reference potential at an input of an operational amplifier (205), - during a mode with a zero output current (302) of the electronic device, coupling (303) of a calibration potential with another input of the operational amplifier (205), - Measuring (304) a response of the operational amplifier (205) to the reception of the calibration potential at the other input and storing (305) this response as a correction factor, and - Using such a stored correction factor to correct (306) the generation of an output current of the electronic device, while in at least one other mode with a non-zero output current (301) in which a measured potential indicating an output current of the electronic device is coupled to the other of the first and second inputs of the operational amplifier (205). [10] The method of claim 9, wherein the method comprises: - Changing the reference potential during at least one other mode with a non-zero output current (301) to influence the magnitude of the output current of the electronic device. [11] Method according to any one of claims 9 or 10, wherein the method comprises: - Measuring temperature values ​​both during the mode with a zero output current (302) and during at least one other mode with a non-zero output current (301), - during the mode with a zero output current (302), recording mappings of temperature readings to corresponding stored correction factors, and - during at least one other mode with a non-zero output current (301), utilizing an actual measured temperature value and the recorded mappings as part of using a stored correction factor to correct the way in which the output current is generated during at least one other mode (301). [12] Use of a method according to any one of claims 9 to 11 to perform precise current feedback control in a lighting driver.

Images