WO2019063333A1

WO2019063333A1 - Lamp operating device with converter in dcm

Info

Publication number: WO2019063333A1
Application number: PCT/EP2018/075050
Authority: WO
Inventors: Ludwig Erasmus DE CLERCQ; Harald Netzer; Frank Lochmann; Clemens KUCERA
Original assignee: Tridonic Gmbh & Co Kg
Priority date: 2017-09-29
Filing date: 2018-09-17
Publication date: 2019-04-04
Also published as: DE102017221786A1; AT16163U1

Abstract

The invention relates to an operating device for the dimmable operation of illuminants, in particular of one or more LED(s), wherein the operating device has a control circuit, and also a converter circuit having an energy storage element and at least one switch driven on the basis of the control circuit; in this case, the control circuit is designed, by driving the switch, to operate the converter circuit optionally at least in the critical mode or in the mode with discontinuous current operation, wherein in discontinuous operation the restart switching time in discrete increments is put into the range of a rising zero crossing of the current through the storage element, wherein the control circuit is designed in this case, upon an incremental variation of the restart switching time, to carry out a regulation of a parameter influencing the illuminant power by means of direct or indirect variation of the switched-on time duration of the switch.

Description

Lamp operating device with converter in DCM

The present invention relates to a lamp operating device, which is designed in particular for the dimmable operation of light sources, such as LEDs. Of course, LEDs also mean organic LEDs (OLEDs).

More specifically, the invention relates to dimmable control gear for lighting devices that use an active clocked converter. In this case, a control circuit controls a switch of the clocked converter, such that in the switched (conductive) state of the switch, an energy storage element (such as an inductance) is charged, which energy storage element in the off state of the switch (non-conductive state of the switch) again discharges via the light path or preferably a further energy storage device (eg, capacitor) loads, which in turn feeds the LED track with a possibly rippled with DC voltage. Thus, ultimately, there is an increasing and decreasing current flow through the energy storage element (inductance).

Of course, if a relatively high average light output is desired, the current drop through the inductor is limited to a value greater than zero, and the switch is again rendered conductive before the current through the inductor has dropped to zero. Meanwhile, for example, if the light output is to be decreased for dimming, the restart threshold (or the corresponding time switch-on time) is correspondingly reduced (or the reconnection time extended), until finally the state is reached that the current through the inductance actually drops to zero may before the switch is again switched conductive and the current thus rises again.

This mode of operation (restarting on reaching the zero level) is typically referred to as "critical mode" (BCM, also abbreviated to "limit operating mode").

Of course, if now, starting from this critical mode, the light output is to be further reduced, a dead time must be introduced between the drop of the current through the inductor to zero and the reconnection of the switch. This operating mode is accordingly called "mode with a gaping current" (English, discontinuous mode, DCM, also: gaping operating mode, gaping mode). It is advantageous, however, not to perform the reclosing in the lapping current mode at arbitrary times, but only when the current that trails after the first zero crossing (due to resonance effects) and thus crosses the zero line several times through the inductor makes a rising zero crossing , Only upon reconnection of the switch in these time ranges of the positive zero crossing of the current through the inductance, a zero-voltage switching (ZVS) can be achieved.

A positive zero crossing of the current through the inductance is a current which at the same time shows a positive gradient at a time when the current has the value zero, ie a first derivative of the course of the current at the time of the zero crossing greater than zero (positive) is.

The time ranges of the positive zero crossings can be detected by measurement (for example of the current through the inductance), or can be predicted on the basis of the resonant frequency known from the dimensioning of the components. Thus, there is thus the problem that in the intermittent operation or the transition from the critical mode in the lopsided operation, the light output can not be changed continuously by continuously changing the reconnection time, but only in increments ("valley switching operation") between a or more positive zero crossings of the current through the inductance, which can lead to jumps in the light output, which can also be perceived optically as flickering during a dimming ramp.

The invention addresses this problem and provides a technique which, at dimming in the leaky mode and transitioning from the critical mode to the spanning mode, at least reduces the problem of jumps in the output power of the converter, even if reconnecting only at the times the positive zero crossings of the current through the inductance takes place. Repeated jumping of the converter between two operating points, wherein the two operating points are characterized by the respective discretely spaced reclosing times of the switch, in a steady state operation of the converter is to be prevented.

As a further aspect, the embodiment should be such that an ASIC is not necessarily required as a control device, but that a microcontroller can also initiate the corresponding activation of the switch. This object is solved by the features of the independent claims. The dependent claims further advantageously form the central idea of the invention.

A first aspect of the invention relates to an operating device for the dimmable operation of lighting devices, in particular one or more LED (s). It has a control circuit which has a clocked converter circuit with an energy storage element, in particular at least one inductance, and at least one switch, which is clocked starting from the control circuit. The converter circuit can be, for example, a buck converter or a boost converter.

The control circuit is designed to operate the converter circuit selectively, in particular by a signal specifying the power, by driving the switch, at least in the critical mode or in the mode with a gaping current. In a first dimming range, therefore, there is an operation in the latching mode, and in another separate dimming range, there is an operation in the critical mode. These two modes may be adjacent to each other, however, a hybrid mode may also be provided between them, in which the gaping or critical mode is temporally multiplexed. (Optionally, the critical mode can also be connected to a continuous mode).

In the latching operation, the control circuit sets the switch-on instant of the switch in discrete increments into one of the time ranges after the first zero crossing, in which the current during the discharge of the energy storage element performs a rising zero crossing. In this case, the control circuit is designed to regulate, with an incremental change in the reconnection time, a feedback variable influencing the luminous flux power by direct or indirect change in the switch-on duration of the switch.

The control circuit can be designed to

incrementally or step-by-step, with an incremental prolongation of the reconnection time, the switch-on time duration and then carry out a control with the control variable "switch-on time duration" and

- Incremental or abruptly reduce the switch-on time and then perform a control with the control variable "switch-on time" at an incremental reduction of the reclosing time.

The control circuit can be designed to carry out an incremental change of the reclosing time, if the control with the control variable "switch-on time" has predetermined minimum or maximum values of the duty cycle or of a variable influencing them, so to speak, this regulation by means of the change of the switch-on to their Limits are encountered. Accordingly, the control circuit may be designed so that when the switch-on time is maximum and at the same time a set long dead time is not sufficient to achieve a desired average current, make an incremental reduction of the dead time by selecting a time earlier Wiedererschaltzeitpunkts. The on-time is maximum when, for example, a maximum allowable peak current value I _pea k is achieved by the inductor current II (coil current II).

An incremental extension of the reconnection time is to be understood as a time shift of the reconnection time at a later time later by a predetermined period of time. The predetermined period of time (increment) may in particular comprise one or more periods of the resonance oscillation. The decrease in the reconnection timing is accordingly a time shift of the reconnection timing to a time earlier by a predetermined period of time.

The control circuit may be configured to specify the change in the switch-on period of the switch indirectly by specifying a switch-off threshold for the current through the switch or directly by specifying the switch-on period.

The control circuit may be configured to, in the event of a transition from the critical mode to the latching current mode, suddenly increase the value for the switch-on time duration of the switch. The converter circuit can be, for example, a boost, buck, buck / boost or flyback converter.

The operating device of a preferred embodiment, the control circuit configured to perform in the latching operation, a change in the reconnection time to a calculated different reclosing time different from a current reconnection time only if the new reclosing time of the switch from the current reconnection time by a time greater than deviates predetermined value.

This avoids undesirable jumping of the reclosing time between adjacent positive-gradient current passes through the energy storage element during the dead time. Such unwanted and possibly multiple repeated gating, especially between adjacent zero crossings of the current, occurs when a calculated duration of dead time for setting a required average load current is approximately midway to two adjacent positive gradient zero crossings the current through the energy storage element is located. Then the case occurs in that, depending on the currently detected zero crossing, a reclosing time of two adjacent zero crossings of the current through the energy storage element is selected. However, the detection of a zero crossing is subject to inaccuracies (English: "jitter") .The reconnection time can therefore change over and over again, possibly for each switch period t _pe riode If there is a determination of a peak current I _pea k, this temporal change of the reconnection time for could compensate for the same required average load current, ideally, so no deviation of the actual average load current from the required average load current is detected.

In reality, however, a non-ideal determination of the peak _{current Ipea} k in conjunction with a frequent change in the selection of the nearest zero crossing to the calculated reclosing time as a selected reclosing time will result in short-term fluctuations in the average load current. These short-term fluctuations of the average load current can be perceptible, for example, as flickering of a light source.

If the peak _current I _pea k is set via a switch-off threshold, then the switch-on time can be changed from period to period t _pe period, that is to say very quickly. The turn-off, however, because it is generated for example by a digital / analog converter and supplied to a comparator, not be changed arbitrarily fast. Although the turn-off threshold can be set to achieve the ideal value for _Ipea k, fluctuations in the average load current can occur because the turn-off threshold can not be varied as rapidly as it is for the reclosing time.

If a change in the reconnection time to a different calculated new reconnection time different from a current reconnection time only executed when the new reconnection time of the switch from the current reconnection time by a time greater than a predetermined value deviates, this hysteresis is a frequent change (engl .: toggle) of the reconnection time. A stability of the average load current is improved, so that, for example, a light output of the lamp can now be flicker-free.

Preferably, the predetermined value is a period of time that is longer than 50% of a period of a resonant oscillation of the current through the energy storage element during a dead time of the clocked converter. In particular, the predetermined value is a period of time greater than 55%, preferably a period of time greater than 62.5% of the period of a resonant oscillation of the current through the energy storage element during a dead time of the clocked converter. Other values for the predetermined value between 0.5 and 1.0 are also possible. Another Aspect of the invention relates to a method for dimmable operation of lamps, in particular one or more LED (s), using a control circuit having a clocked converter circuit with an energy storage element and at least one switch, which is driven by the control circuit. In this case, the converter circuit is operated in at least a partial area of the total dimming range in the mode with a gaping current. In latching operation, the reclosing time is placed in discrete increments in the range of a rising zero crossing of the current through the storage element. For setting a dimming value, a switch-off threshold for the current through the switch or a switch-on time of the switch predefined for the dimming value to be set is set, and a rising zero crossing is set as the actual switch-on time closest to the theoretical, calculated switch-on time which results from the switch-off threshold or the turn-on time period and the dimming value to be set, and - the switch-off threshold or the switch-on duration changed depending on the deviation of the actual switch-on time from the theoretical switch-on time.

The predefined switch-off threshold for the current through the switch or a switch-on period of the switch can depend not only on the dimming value to be set, but on at least one further parameter, such as the voltage across the lighting means.

The invention also relates to a control circuit, for example. ASIC or microcontroller, which is designed for such a method. A further aspect of the invention relates to a circuit for detecting the zero crossing of the current through the inductance of a buck converter, in particular a buck converter in an operating device of the type described above, having one with the potential at a connection point of the diode of the Buck converter and a switch of the Buck converter connected diode circuit which generates a preferably digital detection signal whose preferably logic level changes at a zero crossing of the current through the inductance.

This circuit may include a software or hardware block that prevents the level of the detection signal from changing after a first zero crossing at further zero crossings. This circuit may comprise a control circuit, preferably a microcontroller, to which the detection signal is supplied.

A defined level change of the detection signal can trigger a starting of a counter in the microcontroller. The detection signal starts the dead time Tdead. Another aspect of the invention relates to a circuit for detecting the lower reversal point of the current through the inductance of a buck converter, comprising an auxiliary winding magnetically coupled to the inductance of the buck converter which generates a signal which switches a transistor producing a detection signal when the current through the inductor reaches its lower reversal point. The detection signal thus triggers a storage of the counter readings of the lower reversal points (temporal position of the "valleys").

Yet another aspect of the invention relates to an operating device for lighting, comprising a clocked by a switch converter (clocked converter), in particular Buck converter, comprising a control circuit, the switch after a predetermined time t _on or upon reaching a switch-off of the rising Switch current again nonconductive switches,

wherein the control circuit is supplied with a feedback signal which reproduces the voltage via the light source supplied by the operating device, the control circuit correcting the set time period t _on or the switch-off threshold depending on the feedback signal.

Basically, the switch-off threshold is set as a function of a set average output current I _aV g_nom. In particular, the set peak _current I _pea k may consist of a part which is adaptively calculated on the basis of the selected reclosing time, and a corrective part, which is given for example by a regulator (I regulator), in order to achieve the average output current Lvg _j om to be set ,

In this case, the turn-off threshold is preferably set lower at lower voltage across the lighting means than at higher voltage across the lighting means.

In addition to the calculated peak _current I _pea k, which is corrected by a component determined by the regulator, a correction value (further correction value) is thus taken into account, which is determined on the basis of the voltage across the lighting means. This ensures that the set average output current is really reached, for example independently of one Slope of a current increase via the inductance. Thus, the correction value, which takes into account the voltage across the lighting means, prevents higher peak currents resulting from given turn-off current delays at the same turn-off threshold at lower output voltage than at high output voltage. This is because a steepness of a current increase through the switch also depends on a voltage across the lamps.

If the time duration t _on is specified directly, then it can be set independently of a voltage across the lighting means. In this case, a value for a switch-off delay can be taken into account, in particular subtracted. Further features, properties and advantages of the invention will now be explained with reference to embodiments and the accompanying figures of the drawings.

Fig. 1 shows a schematic view of a known buck converter for

Operation of an LED track.

Fig. 2 shows a current waveform as a function of switch state and

Conductivity of the diode in such a converter of FIG. 1.

Fig. 3 shows the application of the so-called. Valley switching in a selected

Dimming range.

Fig. 4 shows a block diagram for carrying out the invention.

Fig. 5 shows waveforms in the practice of the invention. Fig. 6 shows an implementation of the invention with two possible zero-crosses

Detection circuits.

FIGS. 7 to 10 show signal and current waveforms as well as a schematic circuit in FIG.

9 to compensate for the off delay in the peak current detection in clocked converters. 11 shows a time characteristic of a predetermined average load current Lvg _j om.

Fig. 12 shows, associated with the time course of the predetermined load current after

11, calculated values for a dead time tdead_do _m and set values for the dead time tdead.

Fig. 13 shows, associated with the time course of the predetermined load current after

Fig. 11, calculated values for a dead time tdead_dom and set values for the dead time tdead after a further improved design with a hysteresis.

Fig. 14 shows further time courses of the predetermined average load current I _; avg nom for predefined dimming curves.

FIG. 15 shows, for the time courses of the predetermined average load current according to FIG.

14 calculated values for the dead time tdead_do _m and set values for the dead time tdead with a hysteresis.

16 calculated values for the dead time tdead_dom and set values for the

Dead time tdead with a hysteresis for time-increasing and time-decreasing values of the given load current in a section.

FIG. 1 schematically shows a known buck converter 1 for operating a light path 6 as an example of a typical load of the Buck converter 1.

In this case, starting from a DC bus voltage, also referred to as VB _us , when conducting a semiconductor switch 5 (switch 5), an inductance 4 is charged, which in the switched-off state of the switch 5 via a capacitor CBUCK (or alternatively directly via a light source Load) discharges.

Starting from the capacitor CBUCK, therefore, a DC operating voltage VLED for a light path 6 is provided. In the on state of the switch 5, the voltage VLED increases by charging the capacitor, while it decreases in the off state of the switch 5. Thus, there is a varying course of the current ILED through the light path 6, wherein the human eye perceives only the time average with appropriate high-frequency control of the switch 5.

The switch 5 is controlled by a control circuit 2 via a signal input (control input 3). The control circuit may be an ASIC or preferably a microcontroller.

Fig. 2 shows corresponding signal, - voltage and current waveforms.

The signal HS is the level at the control input 3 of the switch 5. During a time t _on , this switch 5 is turned _on , while it is switched non-conducting in the periods and tdead.

In the period f, the diode 7 in Fig. 1 is conductive. Finally, the lowermost curve in FIG. 2 is the current course through the inductance 4, which is denoted by II. During the turn-on period t _{on of} the switch 5, this current increases, while in the turn-off period f it decreases until it makes a first (falling) zero crossing. Due to resonance effects, a vibration pattern then occurs during the time period tdead. A reconnection in this current gap according to the invention is always carried out to a temporal range (time) in which the current II through the inductor 4 performs a rising zero crossing.

The on-time t _on can be output by the control circuit 2 timed. The on-time t _on can be specified by the control circuit 2 by the switch 5 is switched non-conductive as soon as the rising switch _current reaches a predetermined upper cut-off _threshold I _pea k.

As already mentioned above, reconnecting only in discrete increments to the positive zero-crossing ranges of the current means that jumps in the light output can occur in intermittent operation, or at the transition from the critical mode to the mode with intermittent operation, since just no continuous Change of the reconnection time should be made and thus the supply of the light path 6 could not be continuous.

If, for example, the luminous flux power is to be reduced in intermittent operation, a chronologically higher reclosing time for the switch 5 must be triggered abruptly. According to the invention, the thus threatening power jump is now compensated for by maintaining a time- _on- control when the switch-off period is held fixed. That is to say, when selecting an increment of the positive zero crossings of the current II that differs from the current state through the inductance 4, it is estimated at the same time by the control circuit 2 how, in order to maintain a continuous power reduction, the t _on time first has to be extended abruptly to a new operating point ( as already mentioned, either by a real time specification or by increasing the switch-off threshold in the case of a hysteresis control), before a t _on- time control is then carried out around this estimated working time.

In contrast to the prior art, the on-time t _on not so held in lopsided operation, but designed adaptive and maintain a t _on- time control. In addition, this has the advantage, in terms of control technology, that a continuous _on-time regulation is maintained throughout, ie both in intermittent operation and in critical mode or possibly also in continuous operation mode. Conversely, if, for example, the light output is to be increased again to effect dimming, first the dead time tdead is reduced by a jump of the zero crossing increments, and at the same time the t _on- time is suddenly reduced to an estimated operating point to avoid light power jumps by then continuing the timing at this estimated operating point.

It is important to emphasize once again that the switch-on time t _{on of} the switch 5 can result either directly from a time- _on-time regulation, but also indirectly, namely in particular by predefining a switch-off threshold I _pea k for the switch current. The new operating point can thus be a new switch-on time or a new switch-off threshold. A jump in the dead time increment, that is to say the selected positive zero crossing during the time tdead, can also be triggered when the time control or the shutdown threshold regulation encounters an upper or lower predetermined limit value for the t _on time or the switch off threshold I _pea k.

If, for example, a new time average for the LED current is to be triggered when a new dimming value is specified, the sequence is as follows:

After specifying a change, in particular a jump for the setpoint value for the average LED current, the control circuit 2 first calculates how long the idle time tdead would have to be in the latching mode in order to achieve the average time current at a given switch-off threshold I _pe ak_nom , This calculated dead time tdeadjom will normally not fall on a zero crossing of the current with positive gradients, so that then on the one hand the nearest zero crossing with positive gradient is selected, but at the same time the difference between the calculated dead time tdead _j om and the nearest zero crossing is determined. Because of this known difference can then be set according to the new operating point of the cut-off threshold I _pea k or WZeitregelung accordingly, whereupon the WZeitregelung is continued at this operating point.

Alternatively, the new shutdown threshold calculation circuit tdead 2 may not use the difference between tdead _j om and the selected tdead, but may use the absolute value of the selected tdead based on the selected reclosing time.

The given switch-off threshold I _pe ak_nom also depends on the desired value of the mean output current Iavgjom (average current Lvgjom). With this predetermined switch-off threshold I _pe ak_nom, the dead time tdeadjom is then calculated. In order to prevent unnecessary jumping between different zero-crossings ("valleys"), which may possibly be visible as a jump in the luminous power, always taking into account a new average current Lvg _j om the previous value of the zero crossing is taken into account to jump between to prevent two different zero crossings with positive gradients by a kind of hysteresis control.

It is thus defined when specifying a new average value for the current Lvg _j om by calculation or by comparison with a table, etc., first, a new switch-off threshold I _pea k, following which then the dead time is calculated Tdead _j om in the continuous mode.

The conversion of a set average current value Lvg _j om to a switch-off threshold Ipeak or t _on- time can be multidimensional, such that the selection of the switch-off threshold I _pea k (or the direct t _on- time default) takes place taking into account further parameters (via the Average current value addition). These further parameters can be, for example: the LED voltage V _LED , since this has an influence on how sensitively the average value I _aV g of the current reacts to the change in the switch-off _threshold I _pea k or the t _on time.

Another aspect of the invention is that the determination of the switch-off threshold I _pea k or the t _on- time dependent on the dimming value to be achieved (expressed by the average current Lvg _j om) determines the frequency of the occurrence of the switch-on. Preferably, in a dive jump (ie jump of the average value of the current ILED according to Iavgjom), the switch-off threshold I _pea k is set so that as far as possible a constant frequency of the occurrence of the switch-on results, ie if possible the dead time tdead is kept constant.

However, it is possible that for different dimming ranges different specifications for the set switching frequency are set, so that in different dimming areas (ie different areas of the current I _LED ) the constant to be held frequency of occurrence of the switch-on of the switch 5 may be specified differently.

FIG. 3 shows that the valley switching according to the invention, ie the incremental jump between different increments at positive zero crossings of the current through the energy storage element (inductance 4), is preferably only in an upper dimming range, then at 100% nominal output of the illuminant path 6 is performed. In a dimming down, for example, below 10% dimming level, the dead time tdead in the lopsided operation is so large that the decaying reverberation of the current II through the inductance 4 is no longer present or no longer plays a role. Thus, there is a transition zone in which below the dimming range for the valley switching operation, the reclosing time t _on the switch 5 can be adjusted continuously. Likewise, the dead time tdead can be changed continuously for a change in the mean output current I _aV g.

If the dead time tdead is very large, the switching frequency of the switch 5 becomes correspondingly low. Losses due to the switching operations are correspondingly low. At a low dimming level, therefore, the dead time tdead can also be changed continuously accordingly.

However, this continuous increase in the dead time tdead in the lopsided mode encounters limits in the further reduction of the dimming level, for example below 1% nominal power, since a very long dead time tdead can possibly lead to optically visible effects. Therefore, a maximum dead time tdeadmax is provided. When this is achieved, the dead time tdead is kept at the maximum dead time tdeadmax, and further reduction of the lamp power can then be achieved, for example, only by other effects, such as an amplitude reduction.

Fig. 4 shows a schematic block diagram for carrying out the invention.

A block A designates a calculation block for calculating the nominal dead time tdead nom and the nominal shutdown threshold I _pe ak_nom.

In doing so, account is taken of this calculation:

the predetermined desired value Ia _V g_nom for the LED current ILED,

- The value of the DC supply voltage Vbus for the clocked converter 1 and

- The lighting voltage VLED.

From these input values, as well as a value of the inductance L _B UC _K , the block A in FIG. 4 calculates the nominal values for the dead time tdead nom and the switch-off threshold I _pe ak_nom (or the switch-on time t _on _nom) on the basis of a function or an adjustment table. Block B then serves to implement the dead time control on the basis of these calculated nominal values.

In a first block "select nearest neighbor", that "valley" (ie the positive zero crossing) is selected which comes closest to the nominal dead time tdead nom.

In this block B is stored in advance the temporal position of the "valleys", which is designated in Fig. 4 with "valley array". The "valley" that comes closest to the tdeadjom is taken as the "valley" to be set and accordingly the dead time is set with the value "new Tdead".

All calculations in block A are outside the actual control loop and are therefore non-time critical. They can thus easily be handled by a microcontroller. In block B, the actual current control "current regulator" is provided to which the setpoint Iav _g _nom and the currently measured actual value for the LED current IMEAS is supplied.

Furthermore, there is provided a dead time compensation unit "tdead compensator" which receives as input information the selected valley from the block "select nearest neighbor" as well as the time mean value of the current through the LED path Lvg _j om In order to prevent a jump in the luminous flux which results from the deviation of the valleys from the calculated nominal dead time tdead_nom, the output variable of the current regulator is changed by the dead time compensator tdeadcompensator for shifting the operating point, resulting in a new Switch-off threshold for the switch current "new I _pea k" results. These procedures will be shown again with reference to the waveforms of Fig. 5.

In this case, the scenario is assumed that at a time t1 a jump in the setpoint input i nom for the time average of the current Ia _V g_nom is received. Upon receipt of such a set point jump, a computation phase required to perform the computation illustrated in FIG. 4 follows. As can be seen in FIG. 5, the calculation of the nominal dead time tdeadnom necessary for the setting of the new nominal value I _n0 m takes place. Actually, however, the nearest valley "nearest neighbor" is set to the nominal dead time tdeadjom. This deviation of the nominal dead time tdead nom from the temporal position of the next valve results in a compensation value i _p k_com _P , which is used directly by the block "tdead compensator" in FIG. 4 for a sudden change in the operating point of the current regulation Adjustment of the operating point thus actually prevents the jumps in the luminous flux, since on the other hand the current regulator would also correct this control error (deviation of the actual dead time tdead from the nominal dead time tdead nom), but the duration of the excitation could lead to optically visible effects.

ZX Detection Circuits: Referring to Fig. 6, an application of the invention to the buck converter 1 will now be described. Two possible zero-cross detection circuits 8, 9 are explained, which may alternatively or simultaneously be present in a converter 1.

Shown in FIG. 6 is a buck converter 1 according to the present invention, the power path of which comprises a buck switch M1, a buck diode D1, an output filter capacitor C1 fed by the converter, which may be any low-pass filter, and the buck - Inductance Ll-A and a measuring resistor (shunt) Rl for measuring the buck current.

Diode-based circuit 8:

The circuit in Figure 6, on the one hand, a first detection circuit 8 and a second detection circuit 9, which may be present alternatively or simultaneously, and which are each adapted to generate a signal representing the time at which the inductor Ström II through the buck Inductance Ll-A crosses the zero line, which is referred to in English as 'zero crossing ZX'.

First, the first detection circuit 8 for the ZX event will be explained. This circuit has two diodes D20, D21, two ohmic resistors R20, R21 and a Zener diode Z20.

This circuit is used to generate a signal which detects the beginning of the dead time tdead, ie the first drop of the inductor current II to zero. During the off period of the switch Ml the current II decreases from a maximum value Ιρκ to zero. During this drop, the voltage at the midpoint (measured from the cathode of the diode D1 to ground potential) is at the logic low level, which means that the signal ZCD_l_filtered, which can be supplied to a pin of a microcontroller or ASIC, for example, during this time is also pulled low because the diode D20 is conductive.

When the current II has completed the first zero crossing, the current flow becomes negative, i. the current direction through the inductance Ll-A turns around and the voltage at the center increases. This means that the diode D20 blocks and thus the signal ZCD_l_filtered is pulled to the logic high potential, namely by the supplied low-voltage supply voltage VDD and the resistors R20, R21, this rising edge (from low to high) of the signal ZCD_l_filtered then starts in the control circuit (for example, a microcontroller) a counter for the dead time tdead.This counter runs until the desired dead time has expired and then the control circuit 2 via the signal to the control input 3 again switches the switch Ml (conductive).

In the implementation shown in Figure 6, during the run of the dead time counter, the control circuit 2 outputs a ZCD_Filter_l_out signal at a logic high (high), which in turn pulls the signal ZCD l filtered high via the diode D21 has the advantage that thereby re-starting the dead time counter is avoided, ie restarting at further rising edges of the signal ZCD_l_filtered be prevented or hidden.

However, it is also possible to achieve this suppression of further rising edges (each time in reverse current direction), for example by software solution in a microcontroller, so that then the output of the signal ZCD_Filter_l_out and the diode D21 would not be necessary.

Detection circuit 9 with auxiliary winding and transistor:

Now, the second detection circuit 9 for the ZX event will be explained.

This circuit is used to determine possible reclosing times for the switch M1 (in the event that the reclosing times are not preliminarily stored due to known component dimensions).

As is known, low minimum switching losses are achieved when the voltage across the switch M1 is low and as minimum as possible when it is switched on, which means that the center point voltage (cathode of the diode D1) has a maximum to this region. It is therefore a question of determining those times (counts of the dead-time counter) for which the mid-point voltage is as high as possible. The relationship between voltage across an inductor and current through the inductor is known to be given by: u _L (t) = L * -

This means that the voltage across inductor L1-A is positive when the slope of current II is positive and the voltage is negative when the slope of the current is negative. Due to the 90 ° phase shift between current and voltage at an inductance, a voltage maximum occurs at the positive going zero crossing of the current, or a voltage minimum at the negative going zero crossing of the current. A reversal of the voltage takes place in each case at the maximum and minimum of the current through the inductance. In order to achieve minimum switching losses, so therefore the switch Ml should be turned on again at the positive going zero crossing of the current II. A reversal of the voltage across the inductance Ll-A occurs at a current minimum before this zero-crossing with a positive gradient. This reversal of the voltage across the inductor is used in accordance with the invention to detect the optimal reclosing times. More specifically, the detected time in this approach according to the invention is half a half wave of the oscillation of the inductor current II after its first zero crossing before the optimal time for reconnection.

Meanwhile, in real circuits, there are turn-on delays between the output of a signal and the actual turn-on of the switch M1, so that when the turn-on operation in the switch M1 is triggered at the current minimum before the positive going zero crossing in practice, then the actual turn-on of the Switch Ml falls approximately to the optimal time, ie the time range of the positive going zero crossing of the inductor current II.

According to the invention, as can be seen, an auxiliary winding Ll-B is provided, which is coupled to the actual inductance Ll-A in the power path of the converter. If the voltage above LI is positive, diode D30 will turn off and switch Q30 will be on. The signal ZCD_1 which can be supplied to a pin of a control unit, in particular the control circuit 2, is thus logically low. If the voltage across LI is negative, diode D30 will conduct and switch Q30 will be off. Thus, the signal ZCD_1 is pulled up to the potential high (high) via the resistor R30, which corresponds to the potential of a supplied low-voltage power supply VDD. This means that the falling edges of the signal ZCD_1 represent the times at which the voltage across the inductance changes from negative to positive, which means that the current II has a minimum, followed by a positive zero crossing with a certain time delay of the inductance current takes place. For each of these falling edges of the signal ZCD_1, the current counter value of the dead time counter is stored in a memory "valley array".

The rule algorithm that calculates the target dead time now selects one of the values in this filed valley array. The control algorithm (block B) then selects, based on tdead nom, the valley which is closest to tdead nom. This tdead_vaiiey is then set ("New Tdead" in FIG. 4).

Typically, the calculated dead time will differ from the available Valley values, such that a new turn-off threshold I _pe ak_comp for the peak current is calculated (or a new tone time is calculated) to provide the desired average current Lvgjom with the selected dead time tdead_vaiiey to reach. In the illustrated implementation, measurement paths are provided, which can be defined as follows:

R40, R41 and C40: These are used for LED voltage measurement.

R50, C50: These are used for averaging the measuring voltage across the measuring resistor Rl, in order to determine the average current, which can then be fed via an AD converter of the control circuit.

R51: This signal is compared by means of a comparator with a set threshold. When the threshold is reached, the switch Ml is turned off. So this is one possible implementation for a threshold shutdown.

Compensation of the switch-off delay in peak current detection in clocked converters

As already described above, with clocked converters (clocked converters) with switch 5 activated on the primary side, there is the approach that the switch is switched to nonconducting when the current through the switched-on switch 5 reaches a predetermined switch-off _value I _pea k (peak current , Peak current) has risen. There is the problem that due to component-inherent turn-off delays, a delay arises between the actual achievement of the peak current value I _pea k and the time of turn-off.

These turn-off delays are caused for example by a so-called propagation delay of the comparator and the gate driver, as well as finite edge slopes and parasitic capacitances. The actually flowing peak current and the thus resulting time average is thus falsified, this deviation depending on the steepness of the current dl / dt.

For example, typical turn-off delays may be in the range of several hundred nanoseconds, which may, for example, cause the peak current at 1.2 amps to be turned off instead of a desired turn-off value of 1 amp at a steep current increase dl / dt of the switch current.

The present aspect of the invention now addresses this problem. This aspect of the invention can be combined with the aspects mentioned earlier in the present description, but of course independently apply to any other primary side actively clocked converters, for example Buck converters 1.

The steepness of the current increase through the primary-side switch 5 depends inter alia on the voltage V _LED across the load, for example an LED load. At low LED voltages V _LED , the current increase is very steep, resulting in a large overshoot of the peak current, conversely, a high LED voltage V _{LED leads to} a less steep increase in current when the primary side is on Switch of the clocked converter, which in turn leads to a lower overshoot of the peak current.

According to the invention, the switch-off threshold value is now set as a function of the detected output-side voltage, in particular a measured LED voltage. At a high detected LED voltage V _LED, the threshold value I _pea k is reduced relatively little (for example, from 1 ampere to 0.95 amperes), while it is significantly reduced at a lower detected LED voltage V _LED , for example, from a nominal value of 1 amp to 0.9 amperes, so that in fact results in the desired peak current of 1 ampere in both cases due to the off-delay. The requirements for the accuracy of keeping the peak current are particularly large when the peak current is calculated, for example, as described above. As explained in the present description above, there is the approach that fits a stepwise (incrementally) adjustable dead time tdead a necessary peak current I _pea k_com _{P is} calculated and set in steps to achieve a desired current average I _aV g. Any deviation in the real peak current from the calculated desired value I _pea k_com _P leads to an error in the mean value I _aV g. It is therefore important that the actual peak current in the converter agrees with the calculated (desired) peak current I _pea k_com _P.

With such a precalculation of a necessary peak current with abruptly changed dead time tdead (in the discontinuous mode of the converter), a compensation of the real switch-off delay is particularly important.

FIG. 7 shows how a different peak current can actually be set at two different operating points (slope of the current through the buck inductance when the switch 5 is conductive). As can be seen in FIG. 7, with a switch-off threshold value of 1A, depending on the operating point, a real peak current of 1.2A or 1.1A can be set, for example, even if the switch-off delay (which is dependent on the component) is constant.

According to the invention, the threshold values are now varied according to the operating point in order to compensate for the switch-off delay.

This is shown in FIG.

In a first case, the shutdown threshold is set to 0.9A and in the second case to 0.8A. Due to the switch-off delay, in both scenarios (with a different slope of the current through the inductance 4), the desired (for example, pre-calculated) peak current of 1A is obtained, and the switch-off delay is thus compensated in this regard.

In Fig. 9, a possible implementation of this compensation of this switch-off delay is shown.

An example is calculated peak current I _pea k_com _P , which is calculated, for example, from a desired dimming value (current average) and a selected dead time, said calculated peak current I _pea k_com _P by the output signal of a controller, such as a Integral controller, deltalpk is easily corrected. The sum of both paths then gives a desired peak current of, for example, 1A. Depending on the operating point, a correction value I _p k_overshoot is now deducted from this calculated desired target value. With small LED voltages with steep current increase (large dl / dt) is corrected with a relatively high value (for example, the value I _p k_overshoot be 0.2A). At high LED voltages V _LED with flat current increase (low dl / dt) is corrected accordingly less strong (then, for example, the value of I _p k_overshoot be 0.1A).

After subtracting this correction or compensation factor I _p k_overshoot, the digital peak current value is converted in a DA converter DAC into an analog threshold value. By means of an optional level adjustment and filtering (by means of the resistors R1, R2 and the capacitor C1 in FIG. 9), the threshold value is passed to a comparator, which compares the threshold value with the actual voltage drop at, for example, a shunt, the voltage drop the shunt reflects the current through the buck inductor. When the measured current (or the falling voltage) reaches or exceeds the threshold value, the comparator switches its output, which leads to switching off the Buck switch 5.

According to the invention, the factor "I _pe ak_overshoot" is determined as a function of the measured LED voltage V _LED , for example according to an analytical function or a stored table which reproduces the function, for example, according to FIG. 10. However, other possibilities can also be used Information regarding the slope of the current increase by the switch 5. In this case, the use of at least one of the value of the inductance 4, the input voltage Vbus or the output voltage V _LED and any combinations thereof.

The relationship between the correction value I _pea k_overshoot and the selected parameter is linear in an example of FIG. 10 (linear dependence of the voltage across the LED segment V _LED ). However, this dependence can not be linear.

It is also possible that the correction factor I _pea k_overshoot is calculated continuously or at regular intervals. During the switch-off delay tdoff, the buck current increases by a value I _p kov, so that, with a constant switch-off delay, the overshoot value I _p k_ov to be corrected can be calculated at any time according to the following formula:

doff

In this case, L denotes the value of the inductance 4 of the clocked converter. 1

If the determination of the phase on time does not happen indirectly via a comparator threshold, but directly by specifying the on period t _on, can of course easily the on period t _on to the off tdoff be reduced to compensate for the negative effect of the OFF tdoff. 11 shows a time characteristic of a predetermined load current I _aV g_nom. The abscissa represents the time t for a time segment of a dimming process of a light path 6, for example a slow dimming for 4 seconds for a transition from a dimming level (dimming value) 100% to a dimming level of 1% percent. In Fig. 11, only a portion of the dimming process is shown in the drawing. For the complete dimming process, the value reached on the y-axis at the end of the dimming process has fallen to exactly 1% of the initial value.

In the direction of the ordinate, the target brightness value in FIG. 11 is plotted in the form of a nominal average load current Lvg _j om. The dimming process for the light path 6 is characterized by a desired falling curve 10 of the nominal average load current Lvg _j om, the clocked converter 1 provides at its output as a nominal value Lvg _j om possible actual value of the average load current Lvg.

FIG. 12 shows values for the dead time tdead_do _m and set values for the dead time tdead assigned to the time profile 10 of the predetermined load current I _aV g_nom according to FIG. 11. The time t for the time interval of the dimming process according to FIG. 11 is plotted on the abscissa of FIG.

The calculated dead time tdead _j om shows a gradient 11 rising in the opposite direction to the falling curve 10 of the nominal average load current Iavg _j om. The clocked converter 1 generates the sinking load current I _aV g with an increase in the calculated dead time tdeadjom. Since a restart of the switch 5 is to take place only at discretely spaced time points, an increase of the actually set dead time tdead takes place only in discrete steps. This leads to the stepped rising curve 12 of the dead time tdead shown in FIG. 12 over the time t for the time segment of the dimming process of the light path 6 in FIG. 11.

It can be seen in FIG. 12 that always the, the calculated (nominal) dead time tdead _j om nearest zero crossing of the inductor current II is selected, and thus determines the current value for tdead. For all transitions from a lower value of the dead time tdead to a next higher value of the dead time tdead, FIG. 12 shows that the threshold lies at 50% of a period of the oscillation process of the clocked converter 1. In terms of time, this corresponds exactly to half the distance between two "valleys", ie two negative half-waves of the inductance current II or half a period of the resonance oscillation.

FIG. 12 also shows that in the case of three transitions 12.1 of the total of four transitions of the curve 12 shown, the dead time tdead enters a next higher value of the dead time tdead toggle occurs between two adjacent positive-gradient zero crossings.

This short-term change between adjacent discrete reconnection times is based on non-ideal determination of the peak current I _pea k in conjunction with a frequent change of the positive zero crossing of the inductance current II closest to the calculated reclosing time. Also, the detection of the valleys is subject to fluctuations. The detected possible reclosing times fluctuate. This results in short-term fluctuations in the average load current Lvg and thus ILED, which can be perceptible, for example, as flickering of the light output of the illuminant section 6. If a change in the reconnection time to a calculated new reconnection time other than a current reclosing time is performed only if the new reconnection time of the switch deviates from the current reconnection time by a time greater than a predetermined value, in particular greater than 50% of the period of the resonance oscillation a frequent short-term change of the reconnection time is prevented by this hysteresis. A stability of the average load current is improved, so that, for example, a light output of the light source can be flicker-free.

This is shown in FIG. 13. FIG. 13 shows values for a dead time tdead_d _om and set values for the dead time tdead, assigned to the time characteristic of the predetermined load current Lvg _j om according to FIG. 11, according to a further improved embodiment with a hysteresis.

The predetermined value is set to 10/16 and 62.5%, respectively, in the embodiment of FIG. This means that for the dimming process in the direction of a lower light output or a lower average load current I _aV g, one of the calculated dead time tdeadjom remains adjacent smaller dead time tdead until the calculated dead time tdead _j om is greater than 10/16 of the period the resonance oscillation deviates from the currently selected and set dead time tdead.

Fig. 13 shows in the illustrated step-shaped curve 13 of the dead time tdead reached by means of the introduced hysteresis suppression of kurzeitigen alternation at the transitions 12.1 of the course 12 of the dead time tdead to a next higher value of the dead time tdead in Fig. 12. Thus, for example, an improved since flicker-free light output of the light-emitting means path 6 is achieved by an increased temporal stability of the average load current I _aV g. The hysteresis introduced in the previously discussed embodiment of the clocked converter 1 further causes different operating points of the clocked converter 1 to result for a specific dimming value of the load current I _aV g_nom. The different operating points are each characterized by a set dead time tdead and a set peak current I _pea k. This will be further explained with reference to FIG. 16.

FIG. 14 shows two further time courses of the predetermined average load current Ia _V g_nom for the clocked converter 1 over the time t plotted in the direction of the abscissa. The predetermined average load current is given in representative digital values.

The first curve 14 represents a dimming process in the direction of a lower average load current I _aV g.

The second curve 15 represents a dimming process in the direction of a higher average load current I _aV g.

The first curve 14 of the predetermined average load current I _aV g_nom and the second curve 15 of the predetermined average load current I _aV g_nom intersect at an intersection point 16. Intersection point 16 is characterized by an assigned value of the predetermined average load current I _aV g_nom which is both part of the curve 14 as well as the course 15 of the dimming operations of the average load current I _aV g.

FIG. 15 shows values for the dead time tdead_dom and set values for the dead time tdead with a hysteresis for the time profiles 14, 15 of the predetermined load current according to FIG. 14.

The first curve 16 of the calculated average dead time tdeadjom represents the calculated dead time d nom for the second curve 15 of the predetermined average load current Iavg _j om for a dimming operation in the direction of a higher average load current I _aV g.

The curve 17 is the actually set dead time tdead associated with the first curve 16 of the calculated average dead time d nom. The dead time tdead decreases with increasing time t. Thus, the dimming value increases towards a higher light output of the light-emitting means path 6 due to an increasing average load current I _aV g.

The second curve 18 of the calculated average dead time tdead _j om represents the calculated dead time for the first curve 14 of the predetermined average load current Iavg _j om for a dimming operation in the direction of a reduced average load current I _aV g. This reduces the dimming value towards a weaker light output of the illuminant path 6 due to a decreasing average load current I _aV g.

The curve 19 is the actually set dead time tdead to the second curve 18 of the calculated average dead time tdead _j om corresponding to a dimming operation towards smaller dimming values.

FIG. 15 further shows the effect of hysteresis, for example with a hysteresis factor k with k = 5/8 = 0.625 = 62.5%. A change of the selected zero crossing of the inductance current II as a reclosing time of the switch 5 is made by the control circuit 2, if the time interval between the calculated theoretical reconnection time and the currently selected reclose time greater than the product of hysteresis factor k and the period of the resonant oscillation of the clocked converter 1 is.

The period of the resonant oscillation of the clocked converter 1 corresponds to the time interval between two in-phase current values of the inductor current II during the resonant oscillation during the dead time tdead.

In Fig. 15, the time interval of two in-phase current values of the inductor current II during resonant oscillation corresponds to a height of one stage of the curves 17, 19 of the set dead time tdead in the ordinate direction of Fig. 15. The height 20 in Fig. 15 is counter values , according to a normalized time.

A change of the selected reconnection time takes place in the exemplary embodiment selected in FIG. 15 with a hysteresis factor k of approximately 0.625. This can be seen from the fact that the change does not take place at half, ie 0.5 or 50% of the step height 20, which would correspond to the selection of the zero crossing with a positive gradient of the inductance current II temporally closest to a calculated switch-on time, but after approximately 0.625 or 62.5% corresponding to the first distance 21 at the curve 16 for decreasing dead time tdead or the distance 22 of the curve 18 for increasing dead time tdead.

It also becomes clear that there may possibly be different operating points of the clocked converter 1 for a value of the average load current Iavg _j om to be set. 16 shows calculated values for the dead time tdead_d _om and set values for the dead time tdead with a hysteresis for time-increasing and time-decreasing values of the predetermined load current in a section.

From FIG. 16, it becomes clear that for a given value of the average load current I _aV g_nom and a correspondingly set value of the average load current I _aV g, two different operating points exist due to the hysteresis of the clocked converter 1.

Due to the hysteresis set the different operating points, depending on whether the control circuit 2 from a lower or a higher dead time tdead nom or tdead to a current Zielimmwert, and thus to a current nominal average load current Iavg _j om, to a corresponding calculated current dead time tdead nom or tdead as a switch-on time for the switch 5 and calculated peak current I _pe ak_nom as Ausschaltschwelle for the switch 5 passes.

In one example, a predetermined average load current I _aV g with an average current of 100 mA is provided by the clocked converter 1. This can be done on the one hand in a first operating point, which is characterized by a dead time tdead with a duration of four periods of resonance oscillation. There is thus a reconnection of the switch 5 of the clocked converter 1 at a zero crossing of the inductor current Ibuck with a positive gradient of the course of the coil current II in the fourth period of the resonant oscillation after turning off the switch 5. Turning off the switch 5 takes place in the first operating point upon reaching a peak current I _pea k with a current of 300 mA.

A second operating point is characterized by a dead time tdead with a duration of five periods of the resonance oscillation. There is thus a reconnection of the switch 5 of the clocked converter 1 at a zero crossing of the inductor current II with a positive gradient of the course of the inductor current II in the fifth period of the resonant oscillation after turning off the switch 5. Turning off the switch 5 takes place in the second operating point upon reaching a peak current I _pea k with a current of 330 mA.

In the first operating point and in the second operating point, an average current intensity of the load current I _aV g corresponding to I _LED in the amount of 100 mA is achieved in the selected numerical example. FIG. 16 shows a value 26 of the dead time tdead of the rising dimming curve ""

27 of the average load current I _aV g and a corresponding falling curve 21 of the dead time tdead nom is reached. This results in a step-shaped curve 22 of the set dead time tdead.

For the dead time tdead for the first operating point, this results in a first dead time tdeadi 28 in FIG. 16, since the hysteresis for the illustrated curve does not result in a calculated dead time tdead_nom 26 which deviates from the already set dead time tdeadi by more than 62.5%.

The control circuit 2 sets the dead time with a hysteresis in FIG. 16. For example, the hysteresis may be 62.5% of the period of the resonant oscillation of the clocked converter 1.

The corresponding value 26 of the dead time tdead can alternatively be achieved via a falling dimming curve of the average load current I _av g and a correspondingly increasing curve 23 of the dead time tdead_nom. This results in a step-increasing course 24 of the dead time tdead.

For the dead time tdead for the second operating point, this results in a second dead time tdead2 29 in FIG. 16, since the hysteresis for the illustrated curve does not result in a calculated dead time tdead_nom 26 which differs from the already set dead time tdead by more than 62.5% , From FIG. 16 it thus becomes clear that for a calculated dead time tdead_nom 26 for execution with hysteresis two operating points exist, which are characterized by different first dead time tdeadi 28 and second dead time tdead2 29.

The control circuit 2 can determine the switch-off threshold for the peak current I _pea k corresponding to the different values for the dead time tdeadi 28, tdead2 29 to I _pea ki or I _pe ak2 to the same setpoint I _av g _nom with the current actual value of the load current I. _{reach aV} g.

The discussed execution of the hysteresis for the determination of a reconnection time of the switch 5 is carried out with the same value hysteresis factor k for a rising and a falling dimming curve of the load current Iavg.

Alternatively, hysteresis factors each having different values for increasing and decreasing dimming characteristics of the load current I _av g can be selected.

Claims

claims

1. operating device for the dimmable operation of light sources (1), in particular one or more LED (s),

- comprising a control circuit (2) having a clocked converter circuit (1) with an energy storage element (4) and at least one switch (5), which is clocked starting from the control circuit (2),

- wherein the control circuit (2) is adapted to operate by driving the switch, the converter circuit (1) optionally at least in a critical mode or in a mode with lückendem current,

- wherein the control circuit in the latching operation sets a reconnection time of the switch (5) in discrete increments in one of the temporal ranges after the first zero crossing, in which a current through the energy storage element (4) performs a rising zero crossing,

characterized,

the control circuit (2) is designed to regulate a feedback variable influencing the luminous flux power by an indirect or incremental change in a switch-on time duration of the switch (5) in the event of an incremental change in the restarting time.

2. Operating device according to claim 1,

in which the control circuit (2) is designed to

incrementally with an incremental prolongation of the reconnection time the on-time and to perform a control with the control value "on-time" and

- Incrementally reduce the on-time for an incremental decrease in the reclose time, and then perform a closed-loop control with the "on period" control.

3. Operating device according to one of the preceding claims,

in which the control circuit (2) is designed to

to carry out an incremental change of the reclosing time, if the control with the control variable "ON period" specifies predetermined minimum or maximum values of the Duty cycle or one of these influencing size result.

4. Operating device according to one of the preceding claims,

in which the control circuit (2) is designed to

the change in the switch-on period of the switch (5) indirectly by specifying a

Abschaltschwelle for the current through the switch (5) or directly specify by specifying the switch-on.

5. Operating device according to one of the preceding claims,

wherein the control circuit (2) is adapted to transition from the critical

Mode in the mode with lapping current operation, the value for the switch-on period of the switch (5) to increase suddenly.

6. Operating device according to one of the preceding claims, wherein

the control circuit (2) is designed in the intermittent operation

to perform a change in the reconnection timing to a calculated new reconnection time different from a current reconnection time only when the new reconnection time of the switch (5) deviates from the current reconnection time by a time greater than a predetermined value.

7. Operating device according to claim 6, wherein

the predetermined value is a period of time longer than 50% of a period of one

Resonant vibration of a current through the energy storage element (4) during a dead time of the clocked converter (1).

8. Operating device according to one of the preceding claims,

wherein the converter circuit is a boost, buck or flyback converter.

9. A circuit for detecting the zero crossing of the current through an inductance (4) of a Buck converter, in particular a Buck converter in an operating device according to one of

Claims 1 to 8,

comprising a diode connected to a potential at a connection point of a diode (7) of the Buck converter and a switch (5) of the Buck converter, which generates a preferably digital detection signal whose preferably logic level at a zero crossing of the current through the inductance (4) changes.

10. A circuit according to claim 9, comprising a software-moderate or hardware-trained block, which prevents the level of the detection signal after a first zero crossing at further zero crossings changed.

11. Operating device having a circuit according to one of claims 9 or 10 and a control circuit (2), preferably a microcontroller, to which the detection signal is supplied.

12. Operating device according to claim 11,

wherein a defined level change of the detection signal starts a counter in the

Microcontroller triggers.

13. A circuit for detecting the lower reversal point of the current through an inductance (4) of a Buck converter, in particular a Buck converter in an operating device according to one of claims 1 to 8, comprising

an auxiliary winding (Ll-b) magnetically coupled to the inductance (4) of the Buck converter and generating a signal which switches a transistor (Q30) producing a detection signal when the current through the inductance (4) reaches its lower reversal point.

14. Operating device having a circuit according to claim 13 and a control circuit (2), preferably a microcontroller, to which / the detection signal is supplied.

15. Operating device according to claim 14,

wherein a defined level change of the detection signal triggers storage of a current counter value of a counter in the microcontroller.

16. operating device for lighting means (6), operating device according to one of claims 1 to 8, 11, 12, 14 or 15 having a primary side by means of a switch (5) clocked converter (1), in particular Buck converter,

comprising a control circuit (2) which switches the switch (5) non-conducting again after expiration of a predetermined period of time t _on or upon reaching a switch-off threshold of the rising switch current, wherein the control circuit (2) is supplied with a feedback signal representing the voltage across the lamp (6) supplied by the operating device, the control circuit (2) setting the time duration t _on or the switch-off threshold as a function of the feedback signal.

17. operating device for lighting means according to claim 16,

in which the switch-off threshold is set lower at lower voltage across the lighting means (6) than at higher voltage across the lighting means (6).