EP2425684B1 - Power-controlled operating circuit for a lighting means and method for operating the same - Google Patents

Description

The invention relates to a power-controlled operating circuit for a lighting means having means for determining a power-related actual value associated with the circuit-specific power loss, with a controller to which a power actual value and a power setpoint are supplied and which generates a control difference.

The invention further relates to a method for operating a power-controlled operating circuit for a light source, wherein a power-related with the circuit-specific power loss actual power value is determined by means of one or more the performance yielding parameter and compared with a power setpoint, and wherein with a by the comparison obtained control difference which is controlled a size that determines the power supplied to the bulbs.

Power-controlled operating circuits of the aforementioned type are widely used in electronic ballasts for lighting. For power control, the actual power value associated with the circuit-specific power loss is compared with a predetermined setpoint power. The resulting control difference is used as a control value for the inverter frequency.

Due to unavoidable manufacturing tolerances of the components, the circuit-specific power dissipation varies from device to device. Without countermeasure this has the consequence that the light output of the light sources operated with such power-controlled operating circuits is different despite equal power setpoints.

It is known for a certain type of ballasts to first measure the lamp current at each device after its completion in the production facility and to use a matched resistance corresponding to the measured value in the lamp circuit. In this way it is achieved that all devices after leaving the production site under otherwise identical conditions also have a same lamp current, which ensures that even the light bulbs operated with it give the same light output under the same conditions.

Furthermore, it is known for another type of ballasts also to measure immediately after completion in the production facility, the difference between a predetermined setpoint power and the measured actual value power and to digitize the measured value. The digitized measured value is then stored by means of an external programming device in the ASIC, with which the operating circuit is at least partially realized.

The patent document JP 2005 14 9975 A discloses a method according to the preamble of independent claims 1 and 12. However, the known adjustment measures described above are time consuming and therefore limit productivity limits. This in particular also This is because the adjustment takes place before the operating circuits are inserted into a housing. In this case, individual adapters are required, which represent an additional cost factor.

The invention is therefore based on the object to outsource the elimination of the influence of the circuit-specific power loss of a power-controlled Betriebsschaltüng from the production process and to automate as much as possible.

The object is solved by the features of the independent claims. The dependent claims further form the idea of the invention in a particularly advantageous manner.

The invention further relates to an integrated control circuit, in particular ASIC, microcontroller or hybrid version thereof, which is designed to perform a method according to any one of the preceding claims, and a control device for lighting means, comprising such a control circuit.

In other words, the solution according to the invention for both the operating circuit and for the method is that each operating circuit or each ballast, by incorporating such an operating circuit or such a method, is first implemented in use on site after being switched on. Must go through routine. In this case, the circuit-specific power loss is first determined in a first phase. In a second phase then the control parameters, ie the measured actual value power or the predetermined setpoint power or the control difference determined from the comparison is corrected or modified in such a way that the power loss no longer has any influence on the control result. This is possible because the power loss is either added to the setpoint power or subtracted from the actual power or added to the control difference. The power loss is thus calculated from the calculation process for the control difference. The light output of the light source therefore always corresponds to the setpoint power. This is - with otherwise the same preconditions - the light output of bulbs that are operated with such corrected power-controlled operating circuits, always the same.

The measurement of the power loss during the pre-routine is carried out according to a first possibility with the regular operating parameters for the light source before it absorbs useful power for light emission in a time-starting process. It is known that certain light sources, in particular gas discharge lamps after switching on the operating circuit ignite only with a certain time delay before they absorb useful power for light emission. If one measures the actual powers within this phase, then the measurement result represents the power loss of the operating circuit.

A second possibility is that at least one operating parameter for the lighting means is selected such that the lighting means can not absorb any useful power for emitting light. Such an operating parameter can be, for example, a Be inverter frequency. This can either be so much lower than the resonant frequency of the resonant circuit or so much higher than the latter can be chosen so that the operating voltage for the lamp is not sufficient for the lamp can absorb useful power for light emission. In the case of a gas discharge lamp, this means that the operating voltage is below the ignition voltage. In the case of a light emitting diode there is no recording of useful power, provided that the operating voltage is less than the breakdown voltage of the light emitting diode.

A third possibility may be to replace the bulbs with a known substitution resistance. In this case, if the operating circuit includes an inverter, the inverter frequency may assume a value at which normally ignition of the gas discharge lamp or breakthrough of the LED would occur. The measured power loss is then composed of the power loss of the operating circuit and the power loss of the substitution resistor. Now, if the operating voltage across the substitution resistance or the current through the substitution resistance is additionally measured, then one can calculate the power loss of the substitution resistance, since its resistance value is known. In order to determine the power loss of the operating circuit, then the calculated power loss of the substitution resistance must be subtracted from the measured power loss.

The method of determining the power loss of the operating circuit using a substitution resistor has the advantage that the choice the inverter frequency during the pre-routine phase, in which the power loss of the operating circuit to be measured, is not subject to any restriction. This is important because the power dissipation of the operating circuit in this example is frequency dependent.
This means that the correction values for the control parameters must also be frequency-dependent if the desired independence from the power loss is to apply for each operating frequency.

A first approximation of the desired goal is possible in that the measurement of the power loss in the pre-routine phase is carried out at a fixed frequency, which is chosen so that the operated with the operating circuit bulbs does not absorb useful power for light emission.

A further approximation is possible by measuring the power loss at a plurality of such frequencies, all of which are still in the frequency range at which the luminous means does not yet absorb any useful power for the light emission. Due to the plurality of measured values, an extrapolation can then take place into those frequency ranges in which the luminous means would normally receive useful power for emitting light. The measured values and the extrapolation values can be recorded in a table, which is then queried in the control process to correct the relevant control parameter.

With the use of the substitution resistance, it is then possible, the frequency dependence of the power loss over the entire frequency range of interest in exact Way to determine as a continuous function. The function values of this function are then stored as well as the individual values or the extrapolation values and can be queried to correct the relevant control parameter.

The correction of the control parameters then takes place in an operating phase following the pre-routine phase, during which the light-emitting means absorbs useful power for light emission.

It should be noted that in order to avoid repetition, the features of the claims should include the full disclosure of the description.

Figur 1

eine erste schematisierte Ausführungsform der leistungsgeregelten Betriebsschaltung für eine Gasentladungslampe, bei der der Leistungs-Sollwert korrigiert und die Verlustleistung nur bei einer Frequenz gemessen und gespeichert wird;

Figur 2a

eine zweite schematisierte Ausführungsform der leistungsgeregelten Betriebsschaltung für eine Leuchtdiode, bei der der Leistungs-Istwert korrigiert und die Verlustleistung bei mehreren Frequenzen gemessen und dann extrapoliert und gespeichert wird;

Figur 2b

eine Modifikation der Ausführungsform von Figur 2a,

Figur 3

eine dritte schematisierte Ausführungsform der leistungsgeregelten Betriebsschaltung, wiederum für eine Gasentladungslampe, bei der die Regeldifferenz korrigiert und die Verlustleistung als Funktion von der Frequenz unter Verwendung eines Substitutionswiderstandes gemessen und gespeichert wird, und

Fig. 4

zeigt Beispiel für die Anwendung der Erfindung auf einen Gleichspannungs-Wandler (hier ein Tiefsetzsteller bzw. Buck-Konverter) dargestellt.

Embodiments of the invention will be described below with reference to the drawings. Showing:

FIG. 1

a first schematic embodiment of the power-controlled operating circuit for a gas discharge lamp, in which the power setpoint is corrected and the power loss measured and stored only at one frequency;

FIG. 2a

a second schematic embodiment of the power-controlled operating circuit for a light-emitting diode, in which the actual power value is corrected and the power loss is measured at several frequencies and then extrapolated and stored;

FIG. 2b

a modification of the embodiment of FIG. 2a .

FIG. 3

a third schematic embodiment of the power-controlled operating circuit, again for a gas discharge lamp in which the control difference is corrected and the power loss is measured and stored as a function of the frequency using a substitution resistor, and

Fig. 4

shows an example of the application of the invention to a DC-DC converter (here a buck converter or Buck converter) shown.

Those circuit parts and reference numerals in the Figures 2 and 3 that is opposite the FIG. 1 new or different are marked by bold lines.

• Frequenz
• Tastverhältnis, und/oder
• Totzeiten.

The invention will be explained below with reference to an operating circuit with an inverter with a connected resonant circuit. However, the invention can be applied to all operating circuits in which, for example, the timing of one or more switches is a manipulated variable for the power supplied to the lamps. To change the performance, for example, one or more of the following parameters can be changed:

• frequency
• Duty cycle, and / or
• Dead.

Other power controls, for example, a linear regulator are conceivable in principle, in which so no switch is clocked.

According to the invention, therefore, the operating device may have a DC-DC converter and / or an inverter (DC / AC converter). A combination of a DC-DC converter with the following inverter is possible.

In general, the bulbs can be operated with AC or DC voltage within the scope of the invention.

According to the invention, therefore, a DC-DC converter or any other switching controller topology can be used.

There must be no resonant behavior in the output circuit (it can also be a DC converter such as a buck or buck-boost applied, with PWM (ie the change of the duty ratio) or PFM (pulse-frequency modulation, ie frequency and Duty cycle is changed) dimmed or the output power can be adjusted.

FIG. 1 shows an operating circuit 1 for a gas discharge lamp LP. To the operating circuit according to this embodiment includes an inverter formed by a half-bridge, which consists of a series connection of two push-pull connected electronic switches S1, S2 and a shunt resistor R1. This series circuit is powered by a DC voltage, which is characterized by a positive pole + and ground is marked. The DC voltage is normally generated from the AC mains by rectification and smoothing. With the inverter S1 S2, a series resonant circuit is coupled, which is formed by an inductance L and a resonance capacitor C1. The series resonant circuit is located between the connection point of the two switches S1, S2 and ground. The voltage drop across the resonant capacitor C1 is supplied via a coupling capacitor C2 to a gas discharge lamp LP. To the gas discharge lamp LP is a series circuit of two resistors R2, R3 connected in parallel, whose task later in conjunction with FIG. 3 is described.

The two switches S1, S2 of the inverter are controlled by a variable oscillator with a switching frequency f _s , such that one switch is open and the other is closed. In this case, a voltage dependent on the switching frequency f _s arises across the resonance capacitor C1. This can reach well over 1000 volts in the vicinity of the resonant frequency depending on the circuit losses of the devices for the operation of a gas discharge lamp. At such voltages, the gas discharge lamp LP ignites and absorbs useful energy for emitting light. For light-emitting diodes, which come as an alternative for a gas discharge lamp in question, the operating voltage is usually considerably lower.

By changing the switching frequency f _s , the gas discharge lamp LP can be dimmed. Dimming is carried out with simultaneous power control. For example. via a bus 7 the operating circuit is next to a signal to and turn off a power setpoint P _{soll is} supplied. The power setpoint P _soll is normally compared in a controller 3 with a measured actual power value P _ist . The actual power value P _ist is obtained in this example by means of the voltage drop across the shunt resistor R1 as a power-reproducing parameter (indirectly). _To from the comparison of power set value P _and the power value P, the controller 3 provides a control difference P _diff, which is supplied to the variable oscillator 2 as a control value for the switching frequency of f _s.

Quite generally, provision is made in the invention for the power control to be a performance-determining variable, which may be, for example, the timing of one or more switches (in this example frequency of the clocking of the switches of the inverter) of a DC-DC converter or an inverter.

From the voltage drop across the shunt resistor R1, however, the power actual value can not be determined indirectly because this measured value is still associated with a contribution of the power loss of the operating circuit. This is disadvantageous in that the power loss from operating circuit to operating circuit, for example, is different due to the different component tolerances. The aim of the invention is therefore to eliminate or compensate for the power loss from the control process, so that the output from the gas discharge lamp LP light line always the predetermined power setpoint P _soll corresponds, regardless of the individual operating circuit, or the ballast , in which this operating circuit is used.

In order to achieve this goal, in the embodiment of the operation circuit after FIG. 1 the actual power loss P _v measured and used to correct the power setpoint by this by the power loss P _{v is} increased. The corrected power setpoint P _{soll (korr)} is equal to the sum of the predetermined setpoint P _soll and the measured power loss P _v .

The actual power loss P _v is at the in FIG. 1 shown embodiment, at a switching frequency f _s measured by a processor 6 for a Vorab routine the variable oscillator 2 for the switching frequency f _s . This frequency is a fixed frequency and is chosen such that the voltage dropping thereby at the resonance capacitor C1 is not or no longer sufficient to cause the gas discharge lamp LP to absorb useful energy for emitting light. If the switching frequency f _s lies in the inductive range of the resonance curve, this means that the gas discharge lamp does not yet ignite at this frequency. If the switching frequency f _{s is} in the capacitive range of the resonance curve, this means that the gas discharge lamp LP - after it was in operation - no longer emits light. Due to the fact that the gas discharge lamp LP is inoperative, the voltage drop across the shunt resistor R1 then gives the power loss of the operating circuit 1. The measured power loss P _v is fed to a memory 5.

In an operating phase subsequent to the pre-routine, the processor 6 causes the oscillator 2 to change the switching frequency f _s so that the Gas discharge lamp LP ignites or ignites again. As a result, a voltage drops across the shunt resistor R1, which represents the sum of the actual power loss P _{v of} the operating circuit 1 and the power consumed by the gas discharge lamp LP for the light emission. This voltage _is fed to the controller as a power actual value P _ist . At the same time, the power setpoint in block 4 is corrected by increasing it by the stored actual power dissipation P _v . Accordingly, a corrected setpoint value P _{soll (korr) is} supplied to the controller as setpoint, which _is compared with the actual value P _ist . Since the measured in the operating phase actual power value P _is , as previously mentioned, even increased by the power loss P _v , the control difference P _{diff is} loss-power neutral. The oscillator 2, therefore, a control value is supplied to the - by out of calculated power loss P _v - in fact equal to the difference between the desired power value P _soll and the actual power value P _is is.

The embodiment of the power-controlled operating circuit according to FIG. 2a differs from the one after FIG. 1 in that here not the power setpoint, but the actual power value is corrected. The same components or function blocks have the same reference numerals.

Also in FIG. 2a is the power loss P _v by means of the voltage drop across the shunt resistor R1 in a. Pre-routine phase measured and stored in a memory 15. Since the power loss P _V, however, is frequency-dependent, here are several power loss values at different frequencies measured and stored in the memory 15. All frequencies are however - as in the case of FIG. 1 - Selected so that the light source, which is here a light emitting diode LD, which is connected in series with a series resistor R14, still no useful energy for light emission receives. In other words, this means that the operating voltage at the light-emitting diode LP is still below the breakdown voltage of the light-emitting diode. Because of frequency dependence, additional loss power values are extrapolated from the measured power loss values for the frequency range at which the light-emitting diode LD would emit light. The measured values and the extrapolation values are stored as a table in the memory 15.

The measured in the operating phase actual power value P _is , as in connection with FIG. 1 has been described, still with the power loss P _v afflicted, ie excessive, in this case a block 14 is supplied for correction. The determined corrected actual power value P _{ist (korr)} is therefore the measured actual power value P _ist reduced by the power loss P _v .

The corrected actual power value P _{ist (korr)} is supplied to the controller simultaneously with the power setpoint P _soll . The controller 13 forms from this the control difference, which is also adjusted in this case of the power loss P _v .

As in the modification according to FIG. 2b is shown optionally at the output of the inverter (preferably in front of the capacitor parallel to the LED) a rectifier present (diode DG) or the LED (s) are connected in anti-parallel. In all examples, especially in the Example of Fig. 2a , The inductance L may be designed as a transformer or in the output circuit, an additional transformer may be present, so that a potential separation can be achieved.

FIG. 2b shows an example of the operation of lighting means DC voltage, which is here generated by rectifying an AC voltage of an inverter, but can also be generated by a DC-DC converter.

The third embodiment of the power-controlled operating circuit according to FIG. 3 has three special features.

The first special feature is that the control difference is corrected here as a control parameter. The second peculiarity is that in this case not - as in FIG. 2a - Measured several individual values of the power loss at different frequencies in the pre-routine phase, extrapolated and stored, but it is measured power loss over a whole range of interest frequency as a function and stored in a memory 25. The frequency range should in particular encompass all those frequencies which are controlled during the power regulation of the luminous means that here again is a gas discharge lamp LP. To ensure this, the operating circuit designated here by the reference numeral 21 has as a third feature a substitution resistor RS, which is connected to the operating circuit 21 by means of a switch S3 during the pre-routine phase instead of the gas discharge lamp LP.

In the pre-routine phase, the function memory 25 is now supplied not only with the power loss P _v as a voltage drop across the shunt resistor R 1, but also with a voltage drop across the resistor R 3 of the voltage divider R 2 / R 3 , The voltage drop across the resistor R3 is a measure of the voltage drop across the substitution resistor RS whose resistance value is known. Accordingly, it is also possible to calculate the power loss P _RS , which is absorbed by the substitution resistor RS. It is understood that the power loss P _v measured as a voltage drop across the shunt resistor R 1 must be reduced by the power loss P _RS .

The function memory 25 now supplies the correction block 23 with the power loss P _v as a function of the switching frequency f _s and the power loss P _RS recorded by the substitution resistor _RS . The block 23 forms from a control value P _{diff (corr)} which is adjusted by both the power loss P _v of the operation circuit as well as the substitution caused by the resistance RS power loss P _RS.

The switching of the switch S3 from the gas discharge lamp LP to the substitution resistor RS is performed by the processor 6. The processor 6 thus ensures that the operating phase precedes a pre-routine phase in which the power loss P _{v is} determined and stored. The stored values can then be used in the subsequent operating phase to correct a control parameter, in order in this way ensure that the control is independent of the circuit-specific power loss and therefore the light output emitted by the light source always corresponds to the specified power setpoint.

The substitution resistance RS can also be easily formed by a bridge (that is, a 0 ohm resistor), which bridges the light source in the pre-phase (ie during scanning), ie short-circuiting.

The switch S3 can also be arranged externally or the switchover or bridging can also take place externally (ie the user connects a reference load or a substitution resistance in the pre-stage instead of the light source). In this case, the values of the useful power of the luminous means or of the substitution resistor RS caused by the light emission PRS can also be determined by an external circuit (ie external connection of the voltage divider R2 / R3 and the series resistor R14) and via an existing control line Operating circuit 1 (in particular the function memory 25) are supplied. This offers the advantage that the circuit parts required for the measurement in the pre-phase need not be present in the operating circuit 1 itself, but only have to be connected to the operating circuit 1 for the measurement in the pre-phase. Thus, the measurement in the pre-phase can be done as a kind of calibration measurement, for example during the manufacture of the operating circuit 1 or during the first start-up or even installation of the operating circuit 1 and only for the implementation of the measurement in the pre-phase required circuit parts can be arranged in a kind of programming device, wherein the programming device can be used for a plurality of operating circuits according to the invention. The control line present in the operating circuit 1 may be a programming input or a digital interface for receiving control commands (in particular dimming commands).

As a further supplement, it can be stated that, with knowledge of the recorded power loss characteristic (power loss depending on the lamp power or depending on the timing of one or more switches), the losses actually occurring or the losses which have occurred over a certain (operating) period are also exceeded Control system or a reader can be reported back or read. This feedback can also be linked back to the current or emitted over a certain (operating) period lamp power via a control system or a read-out or read back. It can therefore be a monitoring of the efficiency of the operating circuit 1 (efficiency monitoring).

The measurement (scanning of the losses) can also be repeated at regular intervals, if necessary, an error message can be issued (eg. By signal via bus or optical). The losses can also be integrated over predefined periods.

The measurement in the measurement phase (ie the scanning) can be performed by a control command or similar. be initiated by the user or a center, for example, practice pure Control command. In this case, for example, the detection of a substitution resistance as a load (instead of the luminous means) can also be detected by a load detection, and thus the measurement in the preliminary phase can be initiated with the aid of load detection.

In Fig. 4 is an example of the application of a DC-DC converter (here a buck converter or Buck converter) shown. There is only one switch S2, which is controlled by a change in the duty cycle (ie PWM). Due to the high-frequency activation of the switch S2, energy is stored in the inductance L, wherein after the opening of the switch S2, this energy is discharged via a demagnetization of the inductance L in the freewheeling path formed by the light-emitting diode LD and the freewheeling diode DF. During the switch-on phase of the switch S2, the inductance L is magnetized, this magnetizing current also flows through the light-emitting diode LD, while the freewheeling diode DF is blocked.

Via the voltage divider R2 / R3, the voltage across the light emitting diode LD and the series resistor R14 can be determined by the light emitting diode LD, whereby the power absorbed by the light emitting diode LD can be determined (at least in the measuring phase, these components similar to Fig. 3 described specifically for the measuring phase with the operating circuit 1 can be connected).

The power consumed by the operating circuit 1 can, for example, by a current monitoring in the Supply of the operating circuit 1 (for example via a current measurement by means of differential measurement, current sensor such as current transformer or potential offset stage or by a current measurement between ground and the feedback of the operating circuit 1). Knowing the supply voltage can be concluded on the recorded power.

The capacitor C1 acts in this example as a smoothing capacitor (parallel to the light emitting diode LD).

The light-emitting diode LD can, as in the embodiment of the Fig. 3 for the measurement of the losses in the measurement phase, for example a pre-phase (ie the scanning) by a substitution resistor RS (this can also be simply a bridge (ie a 0 ohm resistor)) bridged or replaced.

Through a switch S3 (not shown here), the switching or bridging can also take place externally (i.e., the user connects a reference load or a substitution resistance in the measuring phase instead of the luminous means).

Thus, even in this example, in a measuring phase, for example, a pre-phase, the losses (ie, the power loss) of the operating circuit 1 can be determined. In this case, it is also possible to carry out a determination of the power loss of the operating circuit 1 at different duty cycles via the range of the possible change in the duty cycle (ie PWM) used for the brightness control (ie dimming) and the determined values can be stored in a table become. From these values an extrapolation for further pairs of values can take place again. Thus, it is possible to measure, extrapolate and store a plurality of individual values of the power loss at different duty cycles in the pre-advance phase (similar to the example of FIG Fig. 3 Alternatively, the power loss over a whole interesting range of the duty cycle can be measured as a function and stored in a memory 25.

Claims (17)

1. Method for operating a power-controlled operating circuit (1, 21) for a lighting means, wherein a parameter, subject to the circuit-specific power dissipation (Pv) and specifying the power, is measured for controlling the power, with said parameter specifying the actual value of the power and being compared with a target value of the power,
   and wherein, dependent on a control deviation obtained by the comparison, a power-determining variable, in particular the clocking of one or several switches, is adjusted,
   and in an operating phase - when the lighting means absorbs effective power for light emission - at least one of the control parameters, such as actual value of the power, target value of the power, control deviation, is corrected in such a way that the light power emitted by the lighting means corresponds better to the target value of the power;
   characterized in that the actual power dissipation (Pv) of the operating circuit is determined and memorized during a measuring phase in which the lighting means does not absorb effective power for light emission.
2. Method according to claim 1, wherein the operating circuit (1, 21) is an inverter (S1, S2, R1), fed by a DC voltage source, with a downstream resonant circuit (L, C1), wherein the lighting means is coupled with the resonant circuit (LP, LD) in such a way that it is supplied with an operating voltage determined by the inverter frequency and the resonance curve.
3. Method according to claim 2, characterized in that the measuring of the power dissipation takes place during the measuring phase with at least two different frequencies (f_S1, f_sn), and in that a chart is generated, based on the measured values by means of extrapolation, showing the power dissipation (P_V1, P_Vn) depending on the frequencies (f_S1, f_sn), by means of which the power dissipation (P_V1, P_Vn) relevant for the respective inverter frequency (f_S1, f_sn) is provided during the operating phase for correcting the measured actual value of the power (P_ist).
4. Method according to claim 2, characterized in that the measuring of the power dissipation takes place during the measuring phase via a relevant frequency range, and in that a chart is generated, based on the measured values, showing the power dissipation (Pv) depending on the frequency (f_S), by means of which the power dissipation (P_V1, P_Vn) relevant for the respective inverter frequency (f_S) is provided during the operating phase for correcting the measured actual value of the power (P_ist).
5. Method according to one of the preceding claims, characterized in that the power dissipation (Pv) of the operating circuit (1, 31) is measured during the measuring phase

   (a) with the regular operating parameters for the lighting means (LP, LD) before it absorbs effective power for light emission in a temporal starting process,
   or

   (b) with at least one operating parameter (operating voltage, operating current or inverter frequency) for the lighting means, which is selected in such a way that the lighting means (LP, LD) does not yet absorb any effective power for light emission,
   or

   (c) with a known substitution resistance (RS), replacing the lighting means (LP, LD), by whose substitution power (P_RS) the measured power dissipation (Pv) has to be reduced.
6. Method according to one of the preceding claims, characterized in that the correction of a control parameter takes place during the operating phase in such a way that the power dissipation (Pv) is either added to the target value of the power (P_soll) or subtracted from the actual value of the power (P_ist) or added to the control deviation (P_diff).
7. Method according to one of the preceding claims, characterized in that the parameter specifying the power is returned from the region of the lighting means and/or from the region of an upstream supply circuit.
8. Method according to one of the preceding claims, where the lighting means are operated with DC voltage, for example rectified AC voltage, or with AC voltage.
9. Method according to one of the preceding claims, characterized in that the lighting means are a gas-discharge lamp (LP) or one or several LED/s (LD).
10. Integrated control circuit, in particular ASIC, microcontroller or hybrid version thereof, being designed to carry out a method according to one of the preceding claims.
11. Operating device for lighting means, having a control circuit according to claim 10.
12. Power-controlled operating circuit (1, 21) for a lighting means (LP, LD), with a control unit (3, 13, 23) to which an actual value of the power as well as a target value of the power are fed, out of which it generates a control deviation, and a means for correcting at least one of the control parameters during an operating phase in which the lighting means absorbs effective power for light emission in such a way that the light power emitted by the lighting means corresponds better to the target value of the power, characterized,
    by means (R1, R3) for measuring and memorizing the actual power dissipation (Pv) of the operating circuit (1, 31) during a measuring phase during which the lighting means (LP, LD) does not absorb any effective power for light emission.
13. Power-controlled operating circuit according to claim 12, characterized by having a DC converter and/or DC/AC converter, preferably clocked by means of one or several switches, wherein the clocking of the switch/es is the power-determining control variable of the power control.
14. Power-controlled operating circuit according to claim 12 or 13, with an inverter (S1, S2, R1) fed by a DC voltage source, with a resonant circuit (L, C1) downstream from the inverter (S1, S2, R1), wherein the lighting means (LP, LD) is coupled with the resonant circuit (L, C1) in such a way that it is supplied with an operating voltage determined by the inverter frequency (f_s) and the resonance curve, with means for determining an actual value of the power (P_ist) subject to the circuit-specific power dissipation (Pv).
15. Power-controlled operating circuit according to one of the claims 12 to 14, characterized in that the power dissipation (P_v) of the operating circuit (1, 21) is measured during the measuring phase

    (a) with the regular operating parameters for the lighting means (LP, LD) before it absorbs effective power for light emission in a temporal starting process,
    or

    (b) with at least one operating parameter for the lighting means (LP, LD), which is selected in such a way that the lighting means does not yet absorb any effective power for light emission,
    or

    (c) with a known substitution resistance (RS) replacing the lighting means (LP, LD) by whose substitution power (P_RS) the measured power dissipation (P_V) has to be reduced.
16. Power-controlled operating circuit according to one of the claims 12 to 15, characterized in that the means (4, 14, 23, 34) for correcting at least one control parameter during the operating phase are in such a way that the power dissipation (Pv) is either added to the target value of the power (P_soll) or subtracted from the actual value of the power (P_ist) or added to the control deviation (P_diff).
17. Power-controlled operating circuit according to one of the claims 12 to 16, characterized in that the means (R1, R3) for measuring and memorizing the actual power dissipation (P_v) of the operating circuit (1, 31) are formed by a processor (6) during a measuring phase.

Images