DE102016205529A1 - Method for measuring light - Google Patents

Abstract

The present invention relates to a method for measuring a light emitted by a lighting unit (1) with a plurality of light sources (2), using a sensor (6) having a sensitive sensor surface (7), wherein the light sources (2) cover the sensor surface (7 At the same time, at least one of the light sources (2) is operated with an output power P which varies with a variation frequency such that a radiation power measured with the sensor (6) varies with the variation frequency over time, whereby a fraction over the variation frequency the radiation power measured with the sensor (6) is assigned to the at least one light source (2).

Description

Technical area

The present invention relates to a method for measuring a light emitted by a lighting unit with a plurality of light sources.

State of the art

The lighting unit has a plurality of light sources, which emit light in operation, for example, at a respective light emitting surface. In this case, the light sources are preferably arranged relative to one another in such a way that a surface light source results, that is, the light emission surfaces together form a flat overall light emission surface. The total light emission surface can then be associated, for example, with an imaging optic which guides the light emitted in each case at the different light emission surfaces into different spatial directions. A challenge may, for example, to the effect that the light sources z. B. different production due to their respective light emission, which then over the total light emitting surface an inhomogeneous brightness distribution or in the just mentioned example, an inhomogeneous light distribution in the far field can result.

Presentation of the invention

The present invention is based on the technical problem of specifying an advantageous method for measuring a lighting unit with a plurality of light sources.

• – die Lichtquellen die Sensorfläche gleichzeitig bestrahlen;
• – mit dem Sensor zeitaufgelöst eine Strahlungsleistung gemessen wird;
• – zumindest zeitweilig mindestens eine der Lichtquellen mit einer Ausgangsleistung betrieben wird, die mit einer Variationsfrequenz derart variiert, dass auch die mit dem Sensor zeitaufgelöst gemessene Strahlungsleistung mit der Variationsfrequenz variiert;

und wobei ferner:

• – über die Variationsfrequenz ein Anteil der mit dem Sensor gemessenen Strahlungsleistung der zumindest einen Lichtquelle zugeordnet wird.

According to the invention, this object is achieved by a method for measuring a light emitted by a lighting unit having a plurality of light sources, using a sensor having a sensitive sensor surface, wherein during a measurement:

• - the light sources simultaneously irradiate the sensor surface;
• - Time-resolved with the sensor, a radiation power is measured;
• At least temporarily, at least one of the light sources is operated with an output power which varies with a variation frequency such that the radiation power measured with the sensor in a time-resolved manner varies with the variation frequency;

and further wherein:

• - A proportion of the radiation power measured by the sensor is associated with the at least one light source via the variation frequency.

Preferred embodiments are to be found in the dependent claims and the entire disclosure, wherein the presentation does not always distinguish in detail between device and method or use aspects; In any case, implicitly, the disclosure is to be read with regard to all categories of claims, in particular always to a lighting device that is set up for a corresponding method.

After measuring the illumination unit, the emission of the individual light sources can then, for example, be adjusted in dependence on the measurement result in such a way that a particularly homogeneous brightness profile results, see below in detail. A basic idea with respect to the measurement is to measure several light sources with a common sensor, which may mean, for example, compared to the provision of a separate sensor per light source a reduced effort, also in cost perspective. The light sources are also measured simultaneously, which, for example, the time required can be reduced in comparison to a sequential measurement; In addition, the lighting unit does not have to be "scanned" with the sensor, so the sensor does not necessarily have to be moved from light source to light source. However, a disadvantage results in so far as an assignment of the measured radiation power is lost, so to speak a spatial resolution, from which the light sources which proportion originates.

According to the invention, therefore, the output power of the at least one light source is periodically varied, so that in the time-resolved measured radiation power, the proportion of the first light source is encoded in some respects with the variation frequency. Although the spatial resolution is therefore initially lost, an individualization is then possible again via the variation frequency, ie via the frequency space. A measurement signal which yields the time-resolved measurement can be converted, for example, with a Fourier transformation into a frequency-resolved evaluation signal, which results in the proportion attributable to the variation frequency.

The "sensor" may, for example, be a single photodiode or also an array thereof, for example also a CCD or CMOS image sensor. In any case, the "radiation power" is thus detected and evaluated indirectly; for example, it is thus also possible to evaluate another photometric quantity which permits at least one qualitative, preferably quantitative, conclusion on the radiation power, for example the luminous flux. In any case, light that falls on the "sensitive sensor surface" is partly converted into the measurement signal.

In general, a respective pulsed operation may be preferred for the light sources (see below in FIG Detail). Even against this background, the "simultaneous irradiation" of the sensor surface during a measurement at any rate refers to a consideration in the time integral, and even from an integral width of not more than 50 ms, 40 ms, 30 ms, 20 ms, 15 ms, 10 ms, or 5 ms, wherein possible lower limits (independently thereof) may be, for example, at least 1 ns, 10 ns, 100 ns, 1 μs, 10 μs, 100 μs or 1 ms (in each case in the order of naming increasingly preferred ). In general, in the case of a pulsed operation, it would also be conceivable for the pulses of the individual light sources to follow one another sequentially in a staggered manner, that is to say that only one pulse is present at a particular time. Preferably, the pulses of at least some of the light sources overlap, ie light from a plurality of light sources also falls on the sensor surface at a particular point in time (not only integrated).

The light sources are preferably LEDs, ie light emitting diodes. Depending on the light source, the "LED" can also be composed of a plurality of individual LEDs, which then preferably differ in the color of their respective emitted light, but in general also an LED with several single LEDs of the same color is possible. In general, single LEDs are also conceivable which emit infrared light, but "light" is generally readable for visible light (380 nm-780 nm). In the case of a plurality of individual LEDs, their respective emitted light (eg blue, red, green and / or yellow light, but also white light) then results in a mixture of preferably white light. However, the LED or its individual LED can also emit original white light, for example in the case of a so-called conversion LED, in which the LED chip is provided with a wavelength-converting phosphor.

In general, a full conversion is also possible with a conversion LED, ie a complete conversion of the radiation originally emitted by the LED chip; However, preference is given to a partial conversion in which the originally emitted radiation is only proportionally converted and an unconverted part forms the light emitted by the LED or the individual LED with the conversion light emitted by the phosphor in mixture. In general, "LED" can be read both on the LED chip itself and on a packaged LED chip. The LEDs can be used as inorganic light emitting diodes, z. B. based on InGaN or AlInGaP, be provided, but also as organic LEDs (OLEDs, for example, polymer OLEDs).

Alternatively or in addition to the LEDs could or could, for example, (so-called) also be provided a so-called remote phosphor arrangement (s). In this case, a phosphor element is irradiated with a pump radiation source of high power density, for example a LASER, which is arranged above a fluid volume (optionally also via further materials, such as a sapphire carrier), preferably a gas, in particular air, at a distance therefrom the conversion light emitted by the phosphor element then used (partial or full conversion, see above). In the case of a lighting unit with at least one LED, this could generally also be used as a sensor (that is to say in the non-light-emitting state due to the photoelectric effect as a photodiode) and thus the remaining light sources could be measured, for example via a back reflection on an optical system or also specifically a reference surface. However, the sensor is preferably a separate component to the light sources.

The "plurality" means at least 2, in this order increasingly preferably at least 3, 4, 5, 6, 7, 8, 9 or 10, with possible upper limits (independent thereof), for example, at most 100,000 (one hundred thousand), 50,000, 10,000, 5,000, 1,000, 500, 100, 50, and 25 respectively (at least preferred in the order of entry). The "plurality of light sources", which are measured in parallel according to the invention, do not necessarily have to be all the LEDs of the lighting unit. Thus, for example, the light sources (of a larger illumination unit) can be measured in groups of a plurality in each case in parallel, but the individual groups in succession (sequentially). In general, the light emitting surfaces can also be arranged only in a row next to each other, preferred is a matrix-like arrangement, each having a plurality (at least 2, 3, 4, 5, 6, 8, 9 and 10) rows and columns, with possible upper limits of the respective Number of rows or columns, for example, can be at most 100, 80, 60 and 50, respectively.

It is expressly referred to the explanations at the outset, each light source (preferably each LED) preferably has a light emitting surface, which more preferably represents a coherent surface per light source and / or is flat (both is generally not mandatory). Relative to one another, these light sources are then preferably arranged such that the light emission surfaces extend parallel to one another (preferably in parallel planes), preferably in their entirety (preferably in a common plane) Give an offset between edge-side and center-arranged light-emitting surfaces). Downstream of the light emission areas, the light preferably passes through an illumination optical system such that light originating from the different light radiation areas passes into different spatial directions.

In a preferred embodiment, the at least one light source with a pulsed Operated output power and is a fundamental frequency of this pulsed operation equal to the variation frequency, based on which the light source is individualized. In the "pulsed operation", the respective light source is operated discontinuously such that after a respective pulse the output power drops to a minimum value which, for example, is at most 20%, 10% or 5% of the mean pulse output power of the previous pulse, preferably equal to zero (preferably a completely discontinuous pulsing). The frequency at which the pulses follow each other is called the "fundamental frequency". The fundamental frequency of a light source does not have to be constant over the entire operation of the illumination unit, but at least temporarily during the time-resolved measurement.

In general, in the context of this disclosure, a period duration, and thus indirectly the frequency, is taken from half-increased edge to half-raised edge (of the subsequent pulse). A "pulse width" results between half-increased and half-dropped surface of the pulse, preferably as half-width; Accordingly, a "pause duration" between the half-fallen edge of the one and the half-increased edge of the subsequent pulse is taken. In general, however, other pulse width definitions would be possible, such. B. the respective increase or decrease to the 1 / e-th or 1 / e ^2- th value of the pulse maximum value (peak value), but is just half the value.

In a preferred embodiment, each of the plurality of light sources is operated with a pulsed output power, preferably all light sources of the illumination unit. In this case, a pulsed operation is preferably not limited to the period of the measurement, but the light sources are then also pulsed independently thereof in normal operation. In a preferred embodiment of the "individualization via the fundamental frequency" each of the plurality of light sources is operated with its own fundamental frequency, which differs from the other fundamental frequencies. In the frequency domain, each light source can then be assigned a respective proportion of the radiation power via its respective fundamental frequency.

In the embodiments just described, the individualization took place via the fundamental frequency of the pulsed operation; In contrast, embodiments are described below in which, although the respective light source is also pulsed, the light sources preferably do not differ from each other in the fundamental frequency of their pulsed operation. For individualization, pulsed operation is then modulated with a respective variation frequency.

In a preferred embodiment, therefore, a different frequency of variation is modulated with the fundamental frequency. In this case, the variation frequency may be greater or smaller than the fundamental frequency, wherein the change of the output power with the variation frequency in relation to the pulses of the pulsed operation is rather small, see below in detail.

In a preferred embodiment, the variation is modulated by varying the output power with the variation frequency during at least some, preferably during all pulses (at least during the measurement) within a respective pulse. Thus, during the pulse, the output power is varied by a mean pulse output power, which is averaged over the pulse duration. The period of this variation may be equal to the pulse duration or make up only a fraction of it, so that then in a particular pulse then several periods can be accommodated. In general, the pulses are preferably rectangular pulses, each of which (on consideration of the pulses alone) have a plateau with a constant output power, at least in the context of what is technically possible also with analog circuits.

In a preferred embodiment, the variation with the variation frequency is rather small, the output power varied with the variation frequency deviates by at most 25%, preferably at most 15%, particularly preferably at most 10%, from the average pulse output power of the respective pulse during a respective pulse (possible lower limits may, for example, be at least 1% and 3%, respectively, and be of interest regardless of the upper limits). This refers to the minima and maxima of the output power which is varied with the variation frequency and which therefore should at no time be greater or less than a corresponding percentage value (25%, 15% or 10%) of the mean pulse output power as a result of the variation with the variation frequency ,

In a preferred embodiment, the fundamental frequency is modulated on the variation by distinguishing at least some of the pulses in their respective mean pulse output power. Over time, the mean pulse output power varies with the variation frequency.

In a preferred embodiment, the variation of the mean pulse output power is again rather small, the average pulse output power of the modulated pulses differs by not more than 25%, more, and particularly preferably not more than 15% or 10%, from one average in-output power taken across all pulses of pulsed operation (possible lower limits may be at least 1% and 3%, respectively) also be of interest regardless of the upper limits). The average "on-output power" is taken over all the pulses of the respective light source during the measurement, ignoring the pauses between them.

In a preferred embodiment, the variation frequency is modulated onto pulsed operation by having at least some of the pulses differ in their respective pulse width. Over time, the pulse width varies with the variation frequency. This variation can, for example, also be combined with an adaptation of the respective average pulse output power such that the energy of the pulses is kept constant. In a pulse with reduced pulse width so the average pulse output power is increased, and vice versa.

In a preferred embodiment, the variation of the pulse width is rather small, because the respective pulse width differs due to the variation by at most 25%, further and most preferably at most 15% or 10%, of a mean pulse width taken over all pulses of the pulsed operation during the measurement from. Possible lower limits may be (independently of) at least 1% or 3%, for example.

In a preferred embodiment, each of the plurality of light sources is pulsed and each light source is modulated at a respective frequency of variation. In this case, each variation frequency differs from the variation frequencies of the remaining light sources (at least in the case of the considered plurality of light sources). The fundamental frequencies are preferably identical; In general, however, a combination of "individualization via the fundamental frequency" with the "modulated variation frequency" would also be possible. The reference to "a" respective fundamental frequency / variation frequency in a particular light source also generally does not imply that the respective light source need only be encoded with a single frequency; "One" and "one" are generally read as indefinite articles. However, with regard to a technically most simple implementation, it may be preferred that each light source is coded with a single variation frequency.

In a preferred embodiment, the fundamental frequencies are or are generally at least 30 Hz, 60 Hz, 100 Hz, 150 Hz, 200 Hz, 250 Hz, 300 Hz, 350 Hz and 400 Hz, respectively (independently) possible upper limits, for example. can be at most 1 GHz, 1 MHz, 100 kHz, 10 kHz or 1 kHz (in each case in the order of naming increasingly preferred). The variation frequency or frequencies may lie in the same ranges, wherein in the case of the modulation the variation frequency preferably deviates from the fundamental frequency by at least 5%, 10%, 20% or 30% (in terms of magnitude) (possible upper limits may, for example, be at most 1 × 10 ⁵ %, 1 × 10 ⁴ % and 1 × 10 ³ %). With the present subject, a frequency encoding, so at least a periodic variation, claimed; Deviating from this (not in accordance with the main claim), coding could also take place via an aperiodic, but preferably deterministic, variation, for example in the manner of a Fibonacci series.

By means of the at least indirectly detected radiation power, for example, a respective color locus can also be determined for a respective light source (the assignment can in turn be given via the frequency coding) and preferably also adapted (color locus shift). Depending on the light source, in particular in LEDs, in particular in conversion LEDs, the color locus can be set, at least within certain limits, via the mean pulse output power or via the associated temperature increase. In this case, for example, the average total output power can also be kept constant, for example by reducing the duty cycle when increasing the average pulse output power and increasing it when lowering.

"Output power" in the context of this disclosure generally refers to the power with which the respective light source emits. Preferably, the output power is adjusted via the current (analog and / or digital), so a current control is made. A pulse-operated light source are thus impressed, for example, current pulses; a modulated variation frequency is then realized, for example, with a correspondingly modulated current. In general terms, "output power" is always also to be read as "current", ie, there is also, for example, a mean pulse current (averaged per current pulse), a mean on-current (averaged over all current pulses, ignoring the pulse pauses ) and a mean total current (averaged over the current pulses and the pulse pauses). The color location shift just discussed can be achieved in particular by a current adjustment (adaptation of the current level), preferably in conjunction with a PWM adaptation, and is also referred to as "DC / PWM adaptation".

In a preferred embodiment, a color location is determined for the at least one light source from its assigned proportion of the radiation power, that is to say for the part of the light emitted by the at least one light source. In general, only deviations from a desired spectral distribution could be determined, which need not necessarily lead to a different color location (light of different spectral distribution, for example. Have the same color location, but different color rendering or color contrast values). In the present case, however, especially deviations that actually require a different color location are of interest.

Preferably, a respective color location is determined for all of the plurality of light sources. The color locus (s) thus determined are respectively assigned via the frequency coding. The color location of the light emitted by a light source can change, for example, over its lifetime, for example due to degradation processes in the phosphor of a conversion LED. By means of the color location measurement, for example, the deviation from a standard value can be checked per light source and / or a relative deviation of the color loci or spectral shifts from each other can be checked, so that, for example, an adjustment for compensation is carried out for particularly high outliers (color locus shift).

In a preferred embodiment, the color locus is determined for the at least one light source (preferably for all) by arranging a light filter in front of the sensor surface of the sensor, that is to say the light sources irradiate the sensor surface through the light filter. Depending on the wavelength, the light filter allows only part of the light to pass through to the sensor surface, so the spectral distribution downstream of the light filter is different from the preceding one. If a corresponding measurement is then made with a plurality (see definition at the beginning) of the light filter, it is used, for example, for different colors, ie, for example. Places in a CIE standard color chart (CIE standard color system of 1931), the respective radiant power measured; In this case, a respective portion of a respective light source can be assigned in a frequency-coded manner (the at least one or preferably in each case all light sources).

The measurement with the different light filters can be done in parallel, so that then a corresponding plurality of sensors would be provided for a time-resolved measurement of the radiant power, each of which is preceded by another light filter (different colors). On the other hand, it is also possible that the measurements are made sequentially with a single sensor and the respective upstream filter is replaced, for example, with a filter wheel. In general, the "light filter" may, for example, also be a dichroic filter, which therefore reflects part of the light and thus alters the spectral composition of the transmitted light. However, the light filter can also be provided as a so-called color filter (the spectral composition can be changed, for example, by absorption in the filter).

The invention also relates to a method for adjusting the light emitted by a lighting unit with a plurality of light sources, wherein the light is measured in parallel in a manner disclosed herein.

In a preferred embodiment, the mean total output power is then changed for at least one of the light sources as a function of this, that is, adjusted. The average total output power in the preferred pulsed operation is generally set preferably via the duty cycle, ie pulse width modulated, preferably exclusively above it.

The adaptation of the average total output power "as a function of" the associated radiation power means, for example, that the assigned component is compared with a standard value; if it is greater than the standard value, the average total output power is reduced, whereas in the opposite case it is increased. The standard value may, for example, be a quantity measured real on a reference illumination unit, but it may also be a calculated value. It is not absolutely necessary to compare the fraction measured per se according to the invention, but this measured variable can also be converted, for example, into a photometric quantity and / or to another reference surface (standard surface), for example in the preferred use in a motor vehicle. Headlight on a reference surface in a plane perpendicular to the direction of travel; the conversion can be done by means of a mathematical transformation matrix.

When adjusting the mean total output power "as a function of the associated radiation power, however, the latter need not necessarily be compared with a fixed standard value, but can, for example, also serve as a reference for the (other) light source (s) (respectively) assigned radiation power so that, for example, a certain relative luminance profile is set. This can be z. B. be as homogeneous as possible, so that the luminance by, for example, not more than 10% or 5% varies (with possible lower limits for example, 0.1% or 1%). However, it is also possible to deliberately set a certain non-homogeneous course, precisely via the relative comparison, or else via a respective comparison with fixed standard values.

In a preferred embodiment, the light emitted by the at least one light source is changed as a function of a color location measurement (see above) in order to shift the color location. For example, "color locus shift" in the CIE standard color diagram may mean an offset of the amount to at least 0.001, 0.005, 0.01, 0.02, 0.03, 0.04, and 0.05, respectively, with possible upper limits (independently thereof). may be at most 0.15, 0.1, 0.08, 0.07 and 0.06, respectively (in the Order of naming increasingly preferred). The color location shift is therefore preferably rather small; it preferably serves to adapt white light, ie it moves in the CIE standard color diagram around the Planck curve.

In a preferred embodiment, the radiant power is measured (once again) following the color locus shift, wherein the sensor surface is preferably irradiated unfiltered. Depending on the radiation power then assigned to the at least one light source, its average total output power is adjusted. It is preferred to proceed accordingly for all of the plurality of light sources, that is, the brightness is readjusted after a color location determination and, if appropriate, shift. This procedure can be repeated in an iterative procedure until the desired result is achieved.

In a preferred embodiment, the spectral composition is changed in the Farbortverschiebung by the average pulse output power is changed in the pulsed operation of a conversion LED (see above). In any case, the color location depends on the average pulse output power within certain limits and is shifted by its adaptation. In order to keep, for example, the average output power constant and / or subsequently make an adjustment of the brightness (previous paragraph), then the duty cycle of the pulsed operation can be changed. This is preferably done at a constant average pulse output power, whereby the (previously set) color locus remains unchanged. The conversion LED is preferably an InGaN LED with a YAG: Ce phosphor.

In a preferred application in the automotive environment, a brightness adjustment of individual light sources of a lighting unit, ie z. B. a headlamp module with matrix-shaped for selective supply different spatial directions LEDs arranged, for example, be advantageous insofar as due to the environmental conditions to which the motor vehicle may be exposed, degradation processes and consequently can set deviations.

In general, due to aging processes, there may be an increasing deviation in the emission properties of the light sources. Deviations can occur not only on a long-term scale over the lifetime, but also within a respective operating interval, for example. Due to different temperature dependencies in conjunction with a generally elevated operating temperature. In the preferred pulsed operation also undesirable beats may occur, for example, in interaction with certain vibration frequencies of the vehicle (here, in addition, data from acceleration and gyro sensors of the vehicle can be evaluated).

An adjustment by measuring and adjusting the light source (s) (see above) in the case of the motor vehicle, for example, once, for example in the context of a final inspection (before delivery), carried out and / or repeated, for example. As part of a normal inspection in a workshop. In both cases, the sensor can then be part of an external measuring unit that is just kept in the automobile plant and / or the workshop. On the other hand, however, the sensor can also be integrated into the vehicle so that, for example, a camera or 3D camera otherwise provided for detecting the oncoming / preceding traffic can also be used as a sensor (temporarily). For this purpose, the vehicle can, for example, in front of a reference surface, such as a white wall, placed and the measurement be made.

But it is also an adjustment continuously while driving possible, in which case preferably a separate sensor is provided (ie preferably not an already provided camera is shared). The output power of the light sources can also be adjusted concretely depending on the reflective properties of the illuminated street (or the surrounding environment), so that, for example, in the case of disproportionate reflections, such as due to wet, the output power can be reduced. Thus, different road surfaces can be illuminated in such a way that, despite different reflection properties, a homogeneous light image for the driver results.

The invention also relates to a lighting device with a lighting unit (cf the above disclosure) and a sensor, wherein the two are arranged relative to one another such that from each of the plurality of light sources a portion of the respectively emitted light falls on the sensor surface (not necessarily of all light sources of the lighting unit, see above). In this case, the lighting device is set up to carry out a method disclosed herein, ie it has, for example, a control and evaluation unit (as an integral component or in separate units), the former controlling the light sources and the latter converting a measurement signal of the time-resolved measurement into an evaluation signal ; Control and evaluation unit are preferably coupled such that a just described adaptation of the average total output power takes place.

The invention also relates to the use of such a lighting device for lighting, especially in the automotive sector, preferably in a car headlight. It is expressly also referred to the above statements ("adaptation in the context of the final inspection" etc.), which should be disclosed in particular with regard to such use. The lighting is preferably adaptive, for example. Depending on the preceding / oncoming traffic; Here, "adaptive" means that individual light sources and thus spatial directions are completely switched off, depending on the situation. On the other hand, with the above-described adaptation, the total output power in the on state of the respective light source (when the assigned spatial direction is supplied with light) is adjusted.

Brief description of the drawings

In the following, the invention will be explained in more detail with reference to exemplary embodiments, wherein the individual features within the scope of the independent claims may also be of interest independently of one another and continue to be distinguished not always in detail between the different categories of claims.

In detail shows

1 a lighting device according to the invention with a plurality of frequency-coded light sources;

2 a first possibility for modulating a variation frequency to a pulsed light source;

3 a second possibility for modulating a variation frequency to a pulsed light source;

4 a third possibility for modulating a variation frequency to a pulsed light source;

5 a possibility for adjusting the color locus of the light emitted by a pulsed light source.

Preferred embodiments of the invention

1 shows a lighting device according to the invention with a lighting unit 1 that have multiple light sources 2 in the present case, conversion LEDs for emitting white light at a respective light emitting surface 3 , The light sources 2 are present on a common substrate 4 assembled.

The light emission is in each case only schematically with a respective main beam 5 which is perpendicular to the respective light emitting surface 3 stands, indicated, in realiter, the light emission takes place each Lambertsch. Accordingly, the respective emitted radiation beams overlap then also at a certain distance to the light emitting surfaces 3 , On a sensor 6 , in particular its sensitive sensor surface 7 , therefore falls from each of the light sources 2 share some light. In one with the sensor 6 detected radiation power is thus initially a spatial resolution, which of the light sources 2 which contribution to the detected radiation power is lost.

Therefore, according to the invention, each of the light sources 2 coded with a respective variation frequency, which is schematically indicated by the sine waves of different frequency.

Next to each of the light sources 2 is in each case over the time t on the x-axis normalized to a maximum value P _max output power P of the respective light source 2 plotted on the y-axis. 1 thus illustrates, in addition to the basic structure, a first possibility for implementing the frequency coding, namely via a respective one per light source 2 different fundamental frequency.

The light sources 2 namely operated with a pulsed output power P and thereby differ in their respective fundamental frequency of the pulsed operation. Will that be with the sensor 6 Time-resolved measured radiation power as a measuring signal then Fourier transformed results in a frequency-resolved evaluation signal. It is then every fundamental frequency and thus every light source 2 assigned a share of the measured radiation power.

The 2 to 4 now illustrate further possibilities for frequency coding, in each case only on the basis of the output power P of one of the light sources (the remaining light sources would then be encoded in an analogous manner, but with a different frequency). Also in this case, the light source is operated pulsed, however, the variation frequency of the fundamental frequency is additionally modulated; The light sources can thus be operated with mutually identical fundamental frequencies.

In the variant according to 2 is within a respective pulse (the pulsed operation), the output power P is varied by a mean pulse output power P _mp , with the variation frequency between a maximum value P _max and a minimum value P _min . The maximum and minimum values deviate by about 5% from the average pulse output power P _mp . For each of the light sources, a separate variation frequency can thus be stored within the pulses, which in turn allows an assignment of the simultaneously measured components. Shown is a modulated sinusoid, which can be constructed in realiter example. Of several stages (digitized).

In the variant according to 3 if the pulses are rectangular, the output power P during a respective pulse is therefore constant within the framework of what is technically possible. In this case, however, the variation frequency is coded by a variation of the mean pulse output power P _mp from pulse to pulse. The mean pulse output power P _mp varies by an average on-output power P _mOn taken over all pulses. The maximum deviation of the mean pulse output power P _mp from the average on-output power P _mOn is around 5%.

In the variant according to 4 the average pulse output power P _mp is constant over the pulses and also the output power P is constant within a respective pulse. In this case, the variation frequency is modulated over the pulse width, which initially increases in the pulse sequence shown and then decreases again. The pause duration varies accordingly so that the average output power is kept constant. For each of the light sources, a different variation frequency of the pulse width is now stored, which in turn allows individualization.

5 shows in one 1 analog representation a pulsed light source 2 , in turn, a conversion LED on a substrate 4 is arranged. For the sake of clarity, in this case only one light source 2 an overall matrix-shaped arrangement shown. In the graph to the left, in this case, the operating current I (magnitude analogous to the output P of the previous figures) is plotted over time t. If the average pulse current I _{mp is} changed, this also shifts the color locus of (by way of example with the main beam 5 ) illustrated light of the light source 2 ,

Concretely, in this case, the conversion LED is made of an InGaN chip emitting blue primary light, and on the chip is disposed an yttrium aluminum garnet phosphor (YAG: Ce) which proportionally converts the blue primary light into yellow conversion light; Conversion light and primary light in mixture form the illumination light (partial conversion).

By changing the average pulse current I _mp , the spectral composition of the illumination light is changed and shifts the color location (see also "DC PWM adjustment"). This is in 5 schematically illustrated with the double arrow in a CIE standard color chart. The illumination light is essentially white light, the color locus shift thus varies slightly to the Planck curve 50 added.

By means of the above-described frequency coding, a respective color locus can be determined for each of the light sources of the illumination unit, and a correction can then be made by a color locus shift, for example in the case of the deviation from a predefined fixed norm value and / or from a specific relative deviation.

For this purpose, the average pulse current I _{mp is} adjusted (in general, more complex changes in the current profile would also be conceivable, which could even leave the average pulse current unchanged). In order to ensure that the brightness or brightness curve does not change, the average total output power is then adjusted after the duty cycle. In order to achieve a desired average output power, the ratio of pulse duration to period duration is thus increased or reduced (for an increase or decrease in the average output power, ie the average current).

LIST OF REFERENCE NUMBERS

1

lighting unit

2

light sources

3

light emission

4

substratum

5

main beam

 6

sensor

 7

sensor surface

50

Planck curve

 P

output

P _mp

Mean pulse output power

P _mon

An output power

P _min

Minimum value output power

P _max

Maximum value output power

Claims (21)

Method for measuring one of a lighting unit ( 1 ) with a plurality of light sources ( 2 ) emitted light, using a sensor ( 6 ) with a sensitive sensor surface ( 7 ), during a measurement: - the light sources ( 2 ) the sensor surface ( 7 ) at the same time; - with the sensor ( 6 ) Time resolved, a radiation power is measured; At least temporarily at least one of the light sources ( 2 ) is operated with an output power P, which varies with a variation frequency such that even with the sensor ( 6 ) time-resolved measured radiation power with the variation frequency varies; and further wherein: - About the variation frequency, a proportion of the sensor ( 6 ) measured radiation power of the at least one light source ( 2 ). Method according to Claim 1, in which the at least one light source ( 2 ) is operated at a pulsed output power P and a fundamental frequency of the pulsed operation is equal to the variation frequency. Method according to Claim 2, in which each of the light sources ( 2 ) is operated with a pulsed output power P, each light source ( 2 ) having a respective fundamental frequency which is different from the fundamental frequencies of the remaining light sources ( 2 ) is different. Method according to one of the preceding claims, in which the at least one light source ( 2 ) is operated at a pulsed output power P, wherein a fundamental frequency of the pulsed operation is different from the variation frequency and the pulsed operation modulates the variation with the variation frequency. The method of claim 4, wherein the variation is modulated by varying, during at least some pulses of the pulsed operation, the output power P at the variation frequency by an average pulse output power P _mp of each of the at least some pulses. Method according to Claim 5, in which the output power P varied during the at least some pulses differs in each case by at most 25% from the average pulse output power P _{mp of} the respective pulse. The method of claim 4, wherein the variation is modulated by having at least some of the pulses of the pulsed operation differ in their respective average pulse output power P _mp taken per pulse, the average pulse output power P _{mp of} the at least some pulses being equal to Variation frequency varies. A method according to claim 7, wherein the average pulse output power P _{mp of} the at least some pulses differs by no more than 25% from a mean _input power P _mOn taken over all pulses of the pulsed operation. The method of claim 4, wherein the variation is modulated by having at least some of the pulses of the pulsed operation differ in their respective pulse width taken per pulse, the pulse width of the at least some pulses varying with the variation frequency. The method of claim 9, wherein the pulse width of the at least some pulses each deviates by at most 25% from a mean pulse width taken over all pulses of the pulsed operation. Method according to one of Claims 4 to 11, in which each of the light sources ( 2 ) is operated with a pulsed output power P at a respective fundamental frequency and the pulsed operation of each light source ( 2 ) in each case a variation is modulated with a respective variation frequency which differs from the variation frequencies of the other light sources ( 2 ) is different. Method according to one of claims 2 to 11, wherein the fundamental frequency is at least 30 Hz and / or the variation frequency is at least 30 Hz. Method according to one of the preceding claims, in which for the at least one light source ( 2 ) with the associated portion of the radiation power for one of the at least one light source ( 2 ) emitted part of the light determines a color location and thus the at least one light source ( 2 ). The method of claim 13, wherein the color location is determined by using the sensor ( 6 ) the radiation power is measured according to one of claims 1 to 12 and the light sources ( 2 ) while the sensor surface ( 7 ) through a light filter, the wavelength dependent only a portion of the light to the sensor surface ( 7 ) lets happen. Method for adjusting one of a lighting unit ( 1 ) with a plurality of light sources emitted light, wherein the light is measured in a method according to any one of the preceding claims. A method of adjusting according to claim 15, wherein a mean total output power of said at least one light source ( 2 ) is then changed in dependence on the assigned proportion of the radiation power. A method of adjusting according to claim 15 or 16 in connection with a measuring method according to claim 13 or 14, in which a spectral composition of the at least one light source ( 2 ) emitted light for the purpose of a color locus shift as a function of the at least one light source ( 2 ) assigned color location is changed. A method of adjusting according to claim 17 when appended to claim 16, wherein Following the color locus shift, the radiant power of one of claims 1 to 12 is measured accordingly and the average total output power of the at least one light source ( 2 ) is then changed in dependence on the proportion of the radiation power assigned on the basis of this measurement. A method of adjusting according to claim 17 or 18, wherein the at least one light source ( 2 ) is operated with a pulsed output power P and the spectral composition of the at least one light source ( 2 ) emitted light is changed by a change in the average pulse output power P _mp . Lighting device with a lighting unit ( 1 ) with a plurality of light sources ( 2 ) and a sensor ( 6 ) with a sensitive sensor surface ( 7 ), whereby the sensor ( 6 ) with the sensor surface ( 7 ) and the light sources ( 2 ) are arranged relative to one another such that each of the light sources ( 2 ) a part of one of them in operation respectively emitted light on the sensor surface ( 7 ), and wherein the lighting device is further adapted to perform a method according to any one of the preceding steps. Use of a lighting device according to claim 20 for lighting, in particular for automotive lighting, in particular in a motor vehicle headlight.

Images