DE102021005158A1 - LED power supply - Google Patents

Abstract

An LED DC power supply is described, comprising a plurality of power channels which are configured to supply power from the same power source to LEDs of different colors; a microcontroller configured to drive power channels to result in a color temperature according to a power channel excitation ratio; and an overload protection circuit. In this context, it is proposed that the overload protection circuit be designed to detect a current fed to the LEDs as a whole, a power fed to the LEDs as a whole and/or a resistance of connected LEDs; and the microcontroller is configured, in response to an under-total resistance, over-current, and/or over-total power, to drive the power channels to alternately power LEDS of different colors while maintaining a color-determining excitation ratio, wherein between the power channels is alternated so rapidly that alternation is visually imperceptible.

Description

The present invention relates to what is claimed in the preamble and accordingly relates to the power supply of LEDs.

Light-emitting diodes are increasingly being used for lighting purposes. In this case, the light-emitting diodes are often supplied with power from a DC voltage power supply unit with a given output power, which not only allows the user energy-saving lighting, but also offers a wide range of setting options. As a rule, a user will generally want light that is perceived as white or colored light. White light is obtained by closely arranging and energizing light-emitting diodes of different colors, for example, light-emitting diodes emitting red, green and blue light. The light temperature, i.e. the "color" of the emitted light, can be adjusted by feeding power to the red, green and blue light-emitting diodes in different ratios depending on the colour. On the other hand, assuming linear behavior of excitation power, light emission and brightness perception, the overall brightness of the emitted light can be changed by increasing the power fed to the diodes to the same extent for all light-emitting diodes; for example, doubling the excitation power leads to lighting that is perceived as being twice as bright. (For the sake of accuracy, it should be mentioned that the assumption of a strictly linear relationship between excitation power, light emission and brightness perception is a simplification made for easier explanation, but in practice a different relationship can occur, which may need to be taken into account by taking suitable measures.)

Thus, the brightness can be adjusted, i.e. dimmed, by changing the light-emitting diode supply in the same way for all diodes. Incidentally, it is also possible to provide a so-called white-light LED, in which light that is perceived as “white” is emitted with a fixed predetermined color temperature of e.g. 5500° Kelvin, with a change in the excitation power fed to this so-called white-light LED only leading to a change in brightness leads. (It should be mentioned that even such an LED does not usually produce the exact spectrum of a blackbody radiator at the specified temperature. Nevertheless, for reasons of better comprehensibility, reference is made to a “color temperature” here.)

It is also possible to change the color intensity, i.e. to set whether the emitted light only deviates very slightly from a given color temperature or whether this hue deviation leads to a coloration that is perceived as very intensive or "pop". With a combination of the four LEDs RGB+W, this intensity can be changed by exciting the white light LED particularly strongly for only a low color intensity, while the RGB color diodes are excited overall more weakly in the ratio required for the desired color temperature; conversely, high color intensity is achieved by excitation of the white light LED particularly weakly and particularly strongly exciting the remaining RGB diodes in the same ratio as before. However, the color intensity can also be changed to the same extent if there is only a three-diode combination, i.e. only RGB diodes without a fourth channel. In such a case, the color intensity can be varied by using a varying part of the maximum brightness that can be achieved with the RGB diodes to jointly generate a light that is perceived as neutral white.

Precisely where it is felt to be particularly advantageous and desirable to be able to make such adjustments to brightness, color temperature and intensity, it is often desired that a specific characteristic actually remains guaranteed.

A large number of circuits have therefore already been proposed which deal with the power supply of LEDs.

From the DE 10 2008 0 29816 A1 a circuit for dimming a lamp and an associated method is known. It is said that the color temperature adjustment of LEDs is difficult for customers due to the need to adjust three individual controllers, especially to choose the color temperature depending on the brightness of the lamp. The specification aims to provide a lamp whose color temperature shifts toward warmer color temperatures with a reduction in power, similar to an incandescent lamp.

From the DE 20 2020 101 445 U1 a system for controlling LED lighting is known, which comprises a dimmable LED current source in which any dimming method is implemented and a unit for automatically adjusting the color temperature of the LED lighting is provided, which has a color temperature control and driver circuit , a program and data storage circuit, an automatic programming unit and a clock circuit, wherein the automatic color temperature adjustment unit interacts with the LED power source and the program and data storage circuit comprises an information database with color temperature data by year history from different geographic regions.

From the EP 2 592 903 A2 a lighting system for controlling a plurality of light sources is known, comprising at least a first and second light source, the second light source having a higher color temperature and luminous efficiency than the second light source. A controller is intended to avoid an uncomfortable feeling of turning on and off due to color mixing operations.

From the U.S. 8,736,183 B2 a method is known in which a dimmer changes an output current to change the power of an LED in such a way that a stable light is obtained, but allegedly avoids color temperature deviations and flickering.

From the U.S. 10,009,974 B2 an arrangement is known in which the LED power supply has first and second FET drivers and a microcontroller provides PWM signals for driving the same. Among other things, it should be achieved that a light temperature change is achieved when dimming as with a light bulb.

From the US 10 159 131 B1 a dimming arrangement for LEDs is known which is intended to retain its brightness when the color temperature changes.

From the US 2015/0145431 A1 a power supply for LED-based light modules is known, which are connected in parallel to a power supply. For dimming, a desired fixed light intensity can be specified once, for example to achieve a fixed brightness below the maximum possible brightness each time it is switched on; a currently desired light intensity can be preset, if necessary, for example by setting on a switch or a remote control, or a gradually changing light intensity can be set, for example to accompany a gentle awakening.

The U.S. 2019/00 290 931 A1 deals with an LED driving circuit with memory function for brightness and color adjustment. A circuit is shown in which first and second groups of LEDs are supplied with DC voltage and an electronic switching element such as a mosfet is associated with the first and second groups of LEDs towards the common ground. The same is controlled by a microcontroller unit.

Irrespective of the known arrangements, in the private sector there is still the problem that settings should not only be possible in a simple manner, but that there is often a mismatch between the LEDs and the power supply unit, for example because a purchased, initially sufficient power unit has been used for some time additional LEDs purchased for this purpose should also be supplied and is therefore overloaded per se. It should be mentioned that an overload can have various effects. If there is only a slight overload, i.e. only slightly too low impedance in the connected LEDs, it may be that only the supply voltage collapses without significant damage to components; If the impedance falls significantly short of the intended impedance or if the power pack is massively overloaded, damage to the connected components or the power pack itself can occur.

The object of the present invention is to provide something new for commercial use.

The solution to this problem is claimed in an independent form. Some of the preferred embodiments are found in the dependent claims.

Thus, according to a first basic idea, an LED DC voltage supply comprising a plurality of power channels which are designed to feed power from the same power source to LEDs of different colors; a microcontroller configured to drive power channels to result in a color temperature according to a power channel excitation ratio; and an overload protection circuit is proposed, which is further configured in such a way that the overcurrent protection circuit detects a current fed to the LEDs in total, a power fed to the LEDs in total and/or a resistance; wherein the microcontroller is configured, in response to unacceptable resistance, excess current, and/or excess power, to drive the power channels to alternately power LEDS of different colors while maintaining a color-defining excitation ratio, wherein between the power channels is alternated so rapidly that the alternation is visually imperceptible.

A plurality of LED power channels is generally used where a specific color impression is to be set by mixing LED colors. In the present application, a color temperature is used in this regard, even if a spectral total emission results from the excitation of several different LEDs that are arranged close together and therefore cannot be visually separated from one another due to insufficient viewing angle differences, which results from the emission of one on a specific temperature located blackbody differs. as white For the sake of completeness, brightness and color saturation are mentioned as other relevant quantities for the description of light emission. The brightness is determined from the total emission intensity, whereas the saturation is determined from the proportion of colored light to "pure white" light. Moreover, for the sake of accuracy, it should be noted that the brightness perceived at a given power input also depends on the color temperature of the light emitted, due to the spectral sensitivity of the eye.

In simplified terms, it can be assumed for the present description that when the same power is fed to all three LEDs of an LED group each comprising a red, a green and a blue LED, the impression of a purely white emitted light results. A fully saturated color with a given color temperature can be obtained simply by feeding power to only two of three colored diodes, where the desired color temperature determines which single diode or which two diodes should be energized at all and what power(s) these Diode receives or these two diodes receive. A light emission with a color that is not fully saturated is obtained if, in addition to this power part fed to the one or two selected colored LEDs in a given ratio, another power part, again the same for all three light-emitting diodes, is fed, which, in addition to the colored emission, has a pure gives a white impression, or if additional power is fed to a white LED. The total brightness can - at least in the simplified case in which no different sensitivities etc. have to be taken into account, which would be easily possible through calibration - be changed by correspondingly changing the absolute total power. In a corresponding manner, a setting for the given color temperature, brightness and saturation can also be made for groups with four LEDs, i.e. RBG+W.

It is possible that in a real implementation, the power to be fed to the LEDs to achieve a certain brightness, saturation and color temperature is determined taking into account the spectral sensitivity of the eye and non-linear light emission of the LEDs, which may not be identical from color to color. This can be accomplished through look-up tables and the like.

Strip-shaped carriers are often provided, along which a large number of groups of three LEDs mounted close together in the colors red, green and blue or groups of four LEDs mounted close together in the colors red, green, blue and white are arranged. Different color impressions can be achieved by specifying the power fed to the respective colors of a group. It is possible to power all groups equally, so that, for example, all blue LEDs in each group are fed slightly less power than the red and green LEDs on average over time; In such a case, apart from small production-related variations, e.g. the sensitivity of individual LEDs, the same color impression and the same brightness will also be retained in each group. Alternatively, it is possible to supply certain areas along the strip in different ways, for example to make selected areas shine brighter or in different colors. The respective strips can be connected to and driven by a power supply. It should be mentioned that there are other possibilities besides the power supply of LEDs on strips. It is thus possible to connect several LED lighting elements to the same power supply, even if the corresponding LEDs are not arranged on strip-shaped carriers.

Nevertheless, a number of things are explained below using the example of LED strips, since such a structural design is particularly relevant in the home and, moreover, mismatches often occur in the home, for example when longer carrier strips that may be chained together are fed from a power supply that is undersized as a result. The invention is also particularly advantageous in such an environment because possible mismatches are encountered particularly often here. In the case of commercial installations, in particular in the case of newly planned installations, it is more likely to be assumed that there will be no mismatches.

Up to the limit of the maximum possible power demand, that is, as a rule, the power limit of the DC voltage supply, it is still possible to ensure an optimal supply of power to the consumers without the mismatch being particularly noticeable to the user. In many cases, the LED arrays will be dimmable and will be operated in dimmed mode by users. A color temperature set in the dimmed state can be maintained regardless of an increase in the required power or the desired brightness without color changes occurring. This is advantageous in that the eye perceives differences in brightness essentially logarithmically (and therefore not necessarily linearly, as initially assumed), i.e. only very large differences in brightness are perceived, whereas small changes in color temperature are already noticeable. Since it may be necessary for the LED Ver Supply less energy has to be expended overall, to which the logarithmic character of the light intensity contributes, it is possible to save energy with the invention.

At the same time, the arrangement according to the invention prevents the LED groups or elements of the voltage supply from being damaged.

The alternation is visually imperceptible where the LEDs of different colors that are spatially close together are alternately energized quickly enough in succession and with sufficient frequency. It is known from video technology that the human eye gets the impression of continuous movement from around 24 frames per second and accordingly a change between the channels will typically happen so often that each color channel to be excited is excited at least 24 times per second . In principle, it would be possible for the use of luminescent materials to give the impression that the LEDs would shine together and continuously even where there is even less frequent switching between different color channels. As a rule, however, it is preferable to switch between the color channels even more frequently. In particular, this allows pulsed excitation and, where continuous or almost continuous power is not desired on all color channels, the thus possible excitation pauses to be evenly distributed, which can reduce the impression of a slightly flickering illumination; this can be advantageous not only for the human eye, but also where video recordings or photographs are to be made, on which stripe artifacts can otherwise be observed.

It should be noted that where AC power is fed to the LEDs, or DC power with (then preferably: periodically) varying voltage is provided, rapid alternation between the power channels while maintaining a color-determining excitation ratio would also be possible, if necessary. In such a case, however, in order to maintain a constant color temperature, it is necessary to take these fluctuating voltages into account when applying power to the individual LED groups, which requires additional effort in terms of the firmware and software of the microcontroller and also makes it sensible to record an instantaneous voltage . Even if this means greater structural complexity and insofar as the implementation of the invention with DC voltage power supplies is considered to be of much greater economic importance, it should be pointed out that the applicant reserves the right to also claim the invention when the LEDs are supplied with periodically or aperiodically fluctuating voltages . As mentioned, a current voltage can then be recorded and the excitation of different color channels required according to a current voltage can be timed according to the excitation voltage-luminosity behavior of the different channels so that within a time that is sufficiently short compared to the inertia of the eye, such as a thirtieth of a second Impression of a constant color temperature is maintained by controlled excitation.

In a particularly preferred variant, the supply voltage is kept constant and the luminous intensity is changed by completely switching the supply voltage on and off at the respective LEDs. This can happen periodically or cyclically; corresponding pulse width modulation methods are known per se. It would be possible for the microcontroller to generate its own control pulses for each LED color of an LED group, with which each LED color is controlled "on" or "off" independently of others. When the switches are implemented by transistors, no complex driver circuits are required per se, so that a comparatively simple circuit is sufficient when generating its own color control pulses. However, it is advantageous to avoid simultaneous switching of different LED colors within an LED group, since this can lead to undesirable current peaks precisely where large LED groups are present, i.e. where high power or currents may have to be switched. Here it is advantageous to avoid such spikes associated with switching processes. In addition, when generating individual control pulses, care must be taken to ensure that these do not coincide at some point due to various effects such as the gradual thermal runaway of phase-locked loops and the like. It is therefore proposed that at least when channels are not to be excited simultaneously, a turn-off edge of a first channel to be operated is detected and a second channel is only switched to active upon detection of this turn-off edge. Depending on the edge steepness, it may even be possible to implement a small time delay after detection of the switch-off edge before the second channel is activated. This falling edge detection can be used to safely prevent multiple channels from receiving power simultaneously in an undesired manner; this would also be undesirable in the case of currents that are only high for a short time, because it endangers the current switching stages. The advantageous edge control can also be implemented with a few logic gates, so that neither the structural complexity nor the energy required to operate the circuits are particularly large.

In addition, not only is a simple and fail-safe implementation achieved, but a high level of insensitivity to unstable clock generators (jitter) is also achieved. A particularly simple edge control of all colors can be implemented if the switch-on pulses run through all colors in the same order; then, by simply counting along each LED color, it can be seen whether the LED is currently supposed to be energized. It should be noted, however, that full performance may not be obtained with certain very strongly toned colors unless measures are taken, for example to allow two channels to be excited simultaneously and a third channel to be excited separately; this is always useful when 2, but not 3 color channels can be controlled simultaneously, i.e. there is a comparatively low overload, for example because a user has only connected a few LED groups too many to an existing power supply unit.

Since particularly strong, "pure" colors are rarely used and, if they are nevertheless desired, they are rarely used with particularly high intensity, especially in the home, a corresponding circuit for such a mixed parallel-simultaneous control can definitely make sense. Nevertheless, it should be mentioned that simple logic circuits are sufficient to enable simultaneous excitation of two channels and separate excitation of a third channel, if necessary.

Incidentally, the DC voltage supply is usually designed in such a way that if the DC voltage supply is correctly adjusted, i.e. when the DC voltage supply is used with an LED strip provided for this purpose, the maximum power is output by simultaneously outputting power to LEDs of all colors. This enables flicker-free operation with a desired high brightness of a desired white light.

However, there is a deviation from such simultaneous excitation as soon as critical loads result from simultaneous excitation of more than two instead of all three or more than three out of four LED colors.

In addition, it should be mentioned that it is easily possible to operate a DC voltage supply in such a way that even when a load that is not larger than the intended load is connected, the LEDs in different color channels are switched cyclically. However, in order to still enable operation with mismatched LED arrangements with such a DC voltage supply, which basically circulates between color channels, by distributing the available output current or the available output power to different channels, the use of addressable groups would be necessary, so that not all LEDs of the same color are switched on simultaneously then draw excessive current due to the mismatch.

It is therefore advantageous if, in the case of an LED DC voltage supply, the microcontroller is designed to control power channels in such a way that when a total current that is still permissible or a power that is still permissible is called up from the same power source, power is at least partially fed simultaneously to LEDs of different colors.

It should be mentioned in this respect that it is possible and preferred if the microcontroller is designed to do this, depending on a currently specified desired color temperature and depending on the size of the deviation of the detected current fed to the LEDs as a whole, the power fed to the LEDs as a whole and/or the total resistance of connected LEDs to the respective permissible maximum values to control power channels in such a way that power is fed to them partially simultaneously and partially alternately in the event that the permissible excess of color channels is only slight. That this is always advantageous when not all LEDs of all three (or four) colors can be excited simultaneously because this leads to an excessive current or an excessive power demand, but without exceeding a permissible maximum load at least two of three ( or three out of four or two out of four) different colors has been explained.

In a preferred variant, the overload protection circuit will be active during the entire operating time and will prevent excessive power or currents from being drawn from the DC voltage power pack. This makes sense because short circuits can occur as a result of defects or through carelessness on the part of users and such short circuits should be counteracted.

It is particularly advantageous if the LED DC voltage supply is designed in such a way that power is routed from the DC voltage power source to the power channels via a two-pole connection, with the overload protection circuit being designed to transmit an instantaneous total power in the first line, one to the LEDs Totally fed current, a total power fed to the LEDs and/or a resistance of connected, in particular a total of connected LEDs, and the microcontroller is designed to record a through switching the power to the LEDs of different colors towards the second line. There is then a common switch in a line in series and there are several parallel switches for each LED color in the current paths leading through the LEDs to the opposite pole. A two-pole cable is therefore sufficient and particularly advantageous where longer cables are required overall, for example in corridors and the like.

The power application of certain LEDs or certain LED groups can then be controlled, for example, by modulating corresponding control signals on the two lines, i.e. a broadcast-like transmission of control commands and their demodulation by local circuits specifically addressed with the broadcast.

It is possible to first measure the total current through a piece of line, for example with a shunt resistor; behind the shunt resistor, which is used to determine the total power or the total current, an overcurrent switch such as a transistor, e.g. a P-FET transistor, can be arranged to interrupt the current flow if necessary. This is particularly advantageous because it means that all elements can be switched off simultaneously and quickly in the event of a failure.

A line can then be routed to all LEDs. The color-dependent power control can take place in the paths to the second line, with either each color group being assigned a common second line in which a switch controlling the power application is arranged, or it is the same color for each color LED or each local plurality of LEDs a switch dedicated to the color is arranged, which controls the current flow through the color LED(s) accordingly. It is again possible to design these dedicated switches as transistors, for example as N-FET transistors.

Alternatively, dedicated channels can be controlled with the microcontroller itself.

It can be seen from the above that the arrangement is designed in a preferred structural embodiment such that a switch-off circuit which acts jointly on all channels is provided in a first line, and the overload protection circuit is designed to activate it directly. Such an arrangement is also advantageous because overcurrents or excessive power drawn can be detected easily, for example with a comparator, and a signal can be generated without great effort with which the power supply to the LEDs is direct, i.e. without additional intervention of the microcontroller, can be interrupted. This achieves a particularly fast shutdown in the event of a failure, because there is no need to jump to the interrupt routines of a microcontroller control program and then wait for the interrupt to be processed. Since the conduction state of a switch, such as a P FET transistor, can often be changed directly with the signal levels and currents available from conventional comparator components, the structural outlay is also extremely low. It is possible to provide a holding element in order to hold an off state at least for a specific time after the overload protection circuit has responded for the first time. Such a holding element can be reset, for example, in response to a microcontroller signal and/or reset to restart after a previous switch-off.

It should be pointed out that, if necessary, the microcontroller can also switch off all color channels as a further security measure; It is also possible to carry out a check cycle after the overload circuit has responded in order to determine whether there is an error (only) on individual channels. For this purpose, if necessary, a pulsed, preferably channel-by-channel, test power supply can take place, as in a switch-on cycle. If an error is determined on individual channels, these can be excluded from further power application. This at least allows emergency lighting at unaffected locations and/or with a color temperature that is not desired in this way.

If an apparent or actual short circuit or overload has been detected, all channels can be tested briefly one after the other. If necessary, emergency lighting can then be implemented through at least the still functioning channels.

It should be noted that it is not mandatory to detect a mismatch of a multi-color LED array to be powered, such as an LED strip that is too long, only and only then when too large a current is already being drawn, potentially for components could be harmful. Rather, it is advantageous to carry out one or more resistance measurements as part of an initialization when starting up the DC voltage power supply and use them to determine, for example, whether the connected LEDs overall have too low an impedance to be supplied with high power together. In such a case it is possible to have a lower rate Apply test voltage than required for operation and/or feed individual excitation pulses to the color channels, which are so short that damage to particularly sensitive components such as power-switching transistors is not yet to be feared. An overall overload that may be expected can be easily detected precisely through individual-channel-based initialization.

A total power output or a total current fed to the LEDs is regarded as too high if certain threshold values are exceeded as a result. In order to determine this, the voltage drop across a shunt resistor can be recorded, for example; If an excessive total power output is to be avoided, the potential present between the DC voltage poles can also be taken into account, either by means of actual measurements or by assuming that certain standard values are complied with. A simple comparison of the voltage drop across the shunt resistor against a permissible voltage drop is often sufficient, which is advantageous because such a simple comparison can be made with a simple comparator and does not even require digital-to-analog conversion; the comparator output can then be fed directly to the microcontroller in the typical form thanks to suitable signal levels in order to trigger an alternating feeding of power into the color channels.

The ability to alternately feed power from the same power source to LEDs of different colors can be achieved in a number of ways. Thus, all LEDs of the same color can each be assigned a common transistor that controls their power supply. In such a case, the same color temperature is obtained with all LED groups. Alternatively, it is possible to operate the LEDs differently in groups; an example is the lighting of a very long hallway, in which it is desired to illuminate certain areas close to room doors with higher brightness, whereas less brightness is required further away from these areas. Such control can be achieved by assigning an address to individual groups of LEDs, such as the groups near a room door, and sending control commands from the microcontroller to a local LED controller that can be reached at this address. In this way, it is possible to react dynamically to necessary local needs such as higher brightness or a different light colour.

It is considered advantageous, but not absolutely necessary, if the microcontroller has at least one additional sensor input and is designed to change, preferably locally, a light intensity or a light temperature in response to sensor signals received there.

A single motion sensor or possibly a number of motion sensors can be used to locally increase the brightness of a hotel guest walking down a particularly long, fundamentally dimly lit hotel corridor. It would also be possible with a contact sensor, in a larger cabinet arrangement with interior lighting, to illuminate more brightly those areas to which a user has just opened a door. Even with such a dynamic increase in brightness, the color temperature can be maintained without any problems, so that the dynamic increase in brightness is perceived as particularly pleasant and natural. If necessary, where a lot of power is already being used, the brightness can be slightly reduced in other places, which can easily be done by changing the on/off duty cycle. For example, during a period of less than 1/30 sec., i.e. a period that can no longer be resolved by the eye, it is very easy to cycle through all power channels very often, for example more than 10, 20 or 50 times each 1/ 30 sec. It should be mentioned that edge control is particularly advantageous for such rapid changes.

A dimmed illumination can then be obtained by providing excitation pauses in which none of the power channels are excited, in addition to going through power channel cycles with the channels being excited. If such excitation pauses are distributed sufficiently evenly, no flickering can be noticed. An increase in the illumination intensity can then be achieved in a simple manner by reducing the number of (preferably evenly distributed and therefore short) excitation pauses at those locations at which the illumination intensity is to be increased. The energizing pauses can be split, eg after energizing each LED color to pause before energizing the next LED color, or alternatively a single pause can be inserted after cycling through the LED energizing phases of all colors. Even with a single pause after running through the LED excitation phases of all colors, flickering cannot be observed if the phases and the pause are cycled through sufficiently quickly, ie the individual phases and pauses are sufficiently short. It should be noted that in embodiments that do not include a single pause after cycling through the LED energization phases of all colors is inserted, but rather a plurality of pauses is inserted, the pause durations do not necessarily have to be the same. For example, after the excitation of blue there is no need to insert a pause that is exactly as long as the pause inserted after the excitation of green. It should be noted that pauses of the same length simplify the control in terms of construction, but that pauses of different lengths can be advantageous because it makes it possible to control a specific point in the space of the HSV diagram that can be reached with a given LED arrangement (Hue-Saturation-Value, ie hue, saturation and lightness) partly simplified. With RGB/+W arrangements, the duty cycle can result from the HSV diagram and with several individual colors the duty cycle can vary according to the set intensity.

It is preferred if the LED DC power supply is designed to separately drive multiple groups of LEDs of all colors. This allows a supply to be guaranteed even if simultaneous excitation of all LEDs would lead to the total currents or the total power being very significantly exceeded because the total impedance is very low. By sufficiently fast cycling through all groups, the impression of flicker-free lighting can be guaranteed for a long time.

It should be noted that the overload protection circuit only needs to measure the real part of the resistance when measuring the resistance of connected LEDs, but that a complete impedance measurement including an imaginary part is also possible.

• 1 eine LED-Gleichspannungsversorgung der vorliegenden Erfindung;
• 2 den Stromfluss durch LEDs unterschiedlicher Farbe, welche einen für die Versorgung aus der Gleichspannungsversorgung vorgesehenen Widerstand aufweisen, und zwar
  1. a- bei weißem Licht und maximaler Erregung;
  2. b- bei weißem Licht und gedimmter Erregung;
  3. c- bei rotgefärbtem Licht und maximaler Erregung;
  4. d- bei rotgefärbten, gedimmtem Licht;
• 3 den Stromfluss durch LEDs unterschiedlicher Farbe, wenn der Gesamtwiderstand der (parallel) angeschlossenen LEDs niedriger ist als für die Versorgung aus der Gleichspannungsversorgung vorgesehen, und zwar
  1. a- bei weißem Licht und maximaler Erregung;
  2. b- bei weißem Licht und gedimmter Erregung mit mehreren kurzen, aber untereinander identisch langen Pausen;
  3. c- bei weißem Licht und auf gleiche Helligkeit wie in 3b gedimmter Erregung, jedoch mit einer einzigen längeren Pause anstelle von mehreren kürzeren, aber untereinander identisch langen Pausen.

The invention is explained below only by way of example and without restricting the generality with reference to the drawing. This is represented by:

• 1 an LED DC power supply of the present invention;
• 2 the current flow through LEDs of different colors, which have a resistor intended for supply from the DC voltage supply, viz
  1. a- with white light and maximum excitement;
  2. b- with white light and dimmed excitation;
  3. c- at red colored light and maximum excitation;
  4. d- for red-colored, dim light;
• 3 the current flow through LEDs of different colors when the total resistance of the connected (parallel) LEDs is lower than intended for the supply from the DC voltage supply, viz
  1. a- with white light and maximum excitement;
  2. b- with white light and dimmed excitement with several short, but mutually identically long pauses;
  3. c- with white light and at the same brightness as in 3b dimmed arousal, but with a single longer pause instead of several shorter but equally long pauses.

After 1 an LED DC power supply 1, generally designated 1, has a plurality of power channels 2R, 2G, 2B adapted to supply power from the same power source 6 to LEDs 3R, 3G, 3B of different colors; The LED DC voltage supply 1 also has a microcontroller 4 which is designed to control the power channels 2R, 2G, 2B in such a way that a color temperature results according to a power channel excitation ratio, as still with reference to FIG 2 and 3 will be explained. An overload protection circuit 5 is also provided. In this case, the overload protection circuit 5 is designed to detect a total current fed to the LEDs 3R, 3G, 3B, a total power fed to the LEDs and/or a resistance of connected LEDs. The microcontroller 4 is designed to drive the power channels 2R, 2G, 2B in response to a too low total resistance, too high current and/or too high total power, which are detected by the overload protection circuit and signaled to the microcontroller 4 such that the LEDs 3R, 3G, 3B of different colors are alternately supplied with power while maintaining a color-determining excitation ratio, alternating between the power channels so rapidly that the alternating is not visually perceptible.

The LED DC voltage supply has a supply plus pole and a supply minus pole, with a shunt resistor 5A and a power-switching transistor 5E terminating the current flow in the event of an overload being provided in the supply line from the plus pole to the LEDs. So the current for all colored LEDs that receive power from power source 6 flows through this shunt resistor. The LEDs are connected in such a way that the anodes of all LEDs are connected in parallel to the same line behind the power-switching transistor 5E.

In the present exemplary embodiment, which is particularly simple for reasons of easier understanding, the LEDs 3R, 3G, 3B of different colors include red, green and blue LEDs. Around In order to control red, green and blue LEDs in such a way that a color temperature results according to a specific power channel excitation ratio, the microcontroller has 3 control outputs 4R, 4G, 4B, which can be accessed via corresponding lines 8R, 8G, 8B - and, if necessary, via respective driver circuits (not shown) - are led to respective power switching transistors 7R, 7G, 7B. These power switching transistors 7R, 7G, 7B are each arranged in series with their associated LEDs, ie power switching transistor 7R in series with the red LED 3R; the power switching transistor 7G in series with the green LED 3G; and the power switching transistor 7B in series with the blue LED 3B. The currents that may be switched through by the power transistors and that flow through the LEDs are collected on the line connected to the minus pole.

In the exemplary embodiment shown, each LED has its own power-switching transistor 7R, 7G, 7B.

In practice, however, it is preferred that the power for multiple red LEDs 3R be switched via one and the same power switching transistor 7R; power for multiple green LEDs 3G can be switched through one and the same power switching transistor 7G and power for multiple blue LEDs 3B can be switched through one and the same power switching transistor 7B. Therefore, in a typical implementation, LEDs are arranged in groups of 3 LEDs each (one red, one green and one blue) and then several such groups are distributed along a strip-shaped carrier and connected in parallel to each other. Where this is the case and only one power switch per color is to be used, the power switching transistors 7R, 7G, 7B can be connected directly to the DC power supply 1, i. H. in the same housing, and there is then provided a first dedicated lead leading to the cathodes of all red LEDs, a second dedicated lead leading to the cathodes of all green LEDs and a third dedicated lead leading to the cathodes of all blue LEDs LEDs leads. Within the DC voltage supply 1, these 3 dedicated lines are then connected to their respective power-switching transistors, which in turn are all directly connected to a line leading to the pole, here the negative pole, of the DC voltage source 6.

The light-emitting surfaces of the red, green, and blue LEDs in a group are so close together that they cannot be visually separated; Simultaneous excitation of the 3 LEDs gives the impression of a uniformly illuminated surface, with the color with which the surface is illuminated being determined by the intensities with which each individual LED is illuminated.

The DC voltage source 6 is designed to feed current to a specific number of LEDs simultaneously. If the number of connected LEDs is greater than that specified by the design of the DC voltage source 6, then the DC voltage source 6 will be overloaded if an attempt is nevertheless made to simultaneously feed current to this excessive number of LEDs.

The connections of the DC voltage supply are now designed in such a way that any number of LEDs can be electrically connected to the DC voltage supply. This is the case, for example, where a large number of RGB LED groups are arranged equidistantly along a carrier strip and red, green and blue LEDs are each electrically connected in parallel to three dedicated power supply lines and terminals or always the same socket/plug combinations for connection of the LED strips are used. If too long a leadframe is then connected to the DC power supply, there will be too many red LEDs to be powered electrically in parallel, as well as too many green LEDs and too many blue LEDs. While it is possible, if necessary, to supply LED carrier strips up to three times the originally intended length with alternating feeding of power to LEDs of only one color at a time, without excessively large currents, for which the DC voltage source 6 is no longer designed, being called up this is no longer possible if all LEDs of different colors are to be supplied with power at the same time.

The maximum load monitor 5 is arranged to counteract an overload. The maximum load monitoring includes a 5A shunt resistor in the supply line, which leads from the positive pole of the DC voltage source to the anodes of the LEDs. The voltage drop across the shunt resistor 5A is fed to a comparator 5B and compared there with a maximum permissible voltage drop. The analog current is thus recorded in a simple manner without prior digitization, i.e. an analog measurement is carried out, so to speak. The comparator 5B outputs a first binary level, e.g. 5V, when the voltage drop across the shunt resistor 5A is not greater than the maximum permissible voltage drop, whereas the comparator 5B outputs a second level, e.g. 0V, when the voltage drop across the shunt resistor the maximum allowable span voltage drop has exceeded. The comparator 5B has a holding element in order to hold the second signal level even after the maximum permissible voltage drop across the shunt resistor has been briefly exceeded, until a reset signal is received from the microcontroller 4 via a line 5D.

The output of the comparator 5B is first fed to a driver circuit 5C, with which a power transistor 5E is switched on as long as the first level is signaled by the comparator and, in the exemplary embodiment shown, on a further line 5G by the microcontroller via a corresponding digital signal that power can in principle be output to the LEDs) approximately after an initialization has been completed), whereas the power transistor 5E is switched off as soon as the comparator outputs the second level or the microcontroller outputs no enable signal on line 5G. The switching process by the comparator 5B, with which a change is made from the first level to the second level indicating a critical state, takes place so quickly that no damage occurs to any of the circuit elements due to overloading until the power transistor is switched off. However, it should also be mentioned that the arrangement with the second line is optional, i.e. there is no need for an additional signal line from the microcontroller to the driver circuit.

The output of the comparator 5B is also fed to an input of the microcontroller 4, so that the microcontroller has information about whether or not a permissible maximum load is currently being exceeded. A memory is assigned to the microcontroller, in which memory a control program is stored for the microcontroller, the execution of which leads to operation of the arrangement according to the invention.

• Zunächst wird an die Gleichspannungsquelle ein LED-Streifen angeschlossen und die Anordnung angeschaltet. Der LED-Streifen möge erläuterungshalber eine zulässige Anzahl LEDs tragen, die noch simultan von der Gleichspannungsquelle mit Leistung versorgt werden kann und keinen Defekt aufweisen.

The arrangement is used as follows:

• First, an LED strip is connected to the DC voltage source and the arrangement is switched on. For the sake of explanation, let the LED strip carry a permissible number of LEDs which can still be supplied with power simultaneously from the DC voltage source and which are not defective.

An initialization cycle is then carried out first, in which the comparator first receives a reset signal via line 5D, so that the driver circuit 5C switches the power transistor 5E on. Thereafter, a first short power pulse is fed to the red LEDs, and only to these, in that the power transistor 7R is switched on by the microcontroller 4 via line 2R.

It is then checked in the microcontroller 4 whether the comparator 5B has detected an excessively large voltage drop across the shunt resistor 5A. Since this is not the case under the conditions described (defect-free, sufficiently short LED strip), it is determined that there is no error and a second short power pulse is then fed to the green LEDs by the power transistor 7 G via line 2 G from the microcontroller 4 is turned on.

Again it is checked whether an excessively large voltage drop was detected and again it is determined that this was not the case. A third test cycle is performed with the blue LEDs, again with no error detected.

It is therefore initially certain that the LED strip is so short that the colors can be supplied with power individually.

It is then checked whether two LED colors can be supplied with power simultaneously without an overload occurring.

To do this, all possible combinations of two are checked in pairs one after the other, i.e. red and green LEDs; red and blue LEDs; blue and green LEDs. The test is carried out in that the corresponding power transistors are switched on simultaneously with the application of a short test pulse and, as before, it is checked whether an overcurrent case occurs. It should be pointed out that such a test of two-way combinations makes sense insofar as it is possible that certain LEDs of individual colors draw a particularly large amount of current and as a result the corresponding LEDs of the certain color cannot be energized together with LEDs of a different color. Under the conditions assumed here, the simultaneous excitation of two colors is not critical.

Finally, a short power pulse is fed simultaneously to all three colors and checked again whether this leads to the maximum load being exceeded. Again, under the conditions assumed here, no overload is observed.

The initialization cycle described requires only a short time. The excitation pulses applied for each test are so short that even in the event of failure the most sensitive components in the circuit are not destroyed; at the same time, the excitation pulses are so long that in the event of failure or when connecting a mismatched LED Strip with too many LEDs, the overload condition is reliably detected with the comparator and a corresponding output signal is generated by the comparator. It should be noted in this regard that if a color in the single excitation mode is already determined to be in an overload condition, checks are no longer made to determine whether an overload condition also occurs when excitation in pairs with this color switched on; on the contrary, it is clear that a color for which an overload case already occurs in the single excitation mode must not be excited together with other colors. In the same way, it is clear that where an overload is already observed in at least one pairwise combination, there is no need to check for overload-free simultaneous excitability of all three colors.

It should also be pointed out that, if necessary, lower voltages could also be applied to the individual color LEDs or to the combination of compartment LEDs for the test case, but that this requires the provision of different voltage levels from the DC voltage source, which is usually due to the additional circuit complexity required for this is undesirable.

After completing the test cycle, the LED DC power supply that is turned on is ready to power the LEDs. A white illumination with maximum intensity is first called up via a remote control and a suitable interface of a remote control signal receiver to the microcontroller 4 (neither shown).

White illumination is obtained when all LEDs are energized with equal power; maximum intensity is obtained when all LEDs continuously receive an excitation signal. Accordingly, the microcontroller emits a signal on all 3 lines 8R, 8G, 8B, which switches on the power transistors 7R, 7G, 7B assigned to the colors red, green and blue, so that current flows through all 3 LEDs. As previously clarified, no overload occurs with this simultaneous excitation of all LEDs. The voltage drop across the shunt resistor 5A is so low that the voltage comparator 5B does not respond and the power switching transistor 5E in the connecting line between the positive pole and the LED anode, which is assigned to all LEDs, can remain conductive.

If the user then gives the instruction via the remote control to dim the light, i.e. to reduce the intensity, this is achieved by no longer continuously feeding power to the LEDs. This is caused by the fact that the LEDs are only operated in pulses, as in 2 B shown.

If the user subsequently commands illumination with a high-intensity reddish tint, as in 2c shown, the red LED is again continuously energized, while the blue and green LEDs are pulsed. Power may be fed to all LEDs at the same time. The exact shade of red desired can be set by energizing the green and blue LEDs for different times. An admixture of slightly more blue light is obtained in that the excitation phases for the blue LEDs are longer than the excitation phases for the green LEDs. Conversely, more green light can be mixed in by making the excitation phases of the green LEDs longer than those of the blue LEDs.

The intensity of the red color can be adjusted by changing the pause times for green and blue LEDs; shortening the pause times leads to a less intense staining.

If it is then desired that the brightness is reduced while the coloring remains the same, i.e. the color temperature and color intensity remain the same, the ratio of on/off times can be reduced in the same way for all LED colors. This is consistent reddish color in the case of a low intensity desired by the user 2-D shown. It should be noted that there are still times when all three LEDs are energized simultaneously. This is not mandatory, but it is particularly easy to implement in terms of circuitry.

It is now assumed that the user lengthens the existing LED strip by connecting a further section during ongoing operation, i.e. without immediately running through an initialization cycle again, supposedly by a factor of 2.5. Assume that at 2.5 times the length, while LED colors can still be energized individually, it is no longer possible to energize LEDs of one color simultaneously with LEDs of any other color.

Immediately after connecting the longer section, the overload protection determines, in view of the previous excitation pattern with simultaneous excitation of all colors, that the voltage drop across the shunt resistor is greater than permissible, i.e. that the currents are too high to ensure permanent, safe operation make possible. Accordingly, the comparator signal will change from the first to the second level, turning the power transistor 5E non-conductive and at the same time the microcontroller 4 signals that an overload case has occurred.

The microcontroller 4 will then end the generation of excitation pulses for the power transistors 7R, 7G and 7B, then reset the comparator 5B and, in a test cycle as in the initialization of the system after switching on, first check whether all LED colors may still be excited individually. For this purpose, a short, testing excitation pulse is first output only to excite the red LEDs for transistor 7R; the comparator 5B will not respond in this case, since despite the excessive length that has resulted from the use of the additional LED strip, there is still no case of overload with single excitation. Equal results are obtained using short excitation pulses to excite only the green LEDs and then only the blue LEDs.

It is therefore clear that operation per se is still possible. It is then checked whether 2 colors can still be operated simultaneously. In view of the length of the total connected LED strips selected in the example, which clearly exceeds the limit values, this is no longer the case. This is determined by the fact that initially when a test pulse simultaneously exciting red and green LEDs is applied to the associated power transistors 7R and 7G, an overload situation occurs and is signaled to the microcontroller 4 via line 5D, whereupon the latter resets the comparator; that further, even if the excitation pulses are provided for the simultaneous excitation of the red and blue LEDs, an overload situation occurs, which in turn is signaled to the microcontroller 4 via line 5D and triggers a reset of the comparator; and that finally, even if power is applied to green and blue LEDs simultaneously for testing purposes, an overload occurs and is signaled to the microcontroller, whereupon the latter resets comparator 5B again.

It was thus possible to determine on the microcontroller 4 that the extended LED strip arrangement can only be operated if each color channel is excited individually. It should be noted that in order to achieve a given color, the LEDs of different primary colors must each be operated in a pulsed manner, and thus the overall arrangement will shine less brightly than in cases in which, thanks to correct adaptation of the LED strip to the available power supply, a permanent excitation of the LED is possible.

It is now assumed that the user initially wants to obtain the maximum brightness that is still possible with a white color. Clearly, this requires energizing the LEDs of all colors in sequence.

Accordingly, a short pulse is first provided by the microprocessor, with which the power transistor 7R is switched on. After the termination of this pulse, the LED 3G is energized for the same time, and after the termination of the energization of the light emitting diode 3G, the blue LED 3B is energized.

It is clear that, on the one hand, pauses between arousal phases should be particularly short in order to ensure optimal utilization of the available power. It is also clear that a situation should be avoided in which more than one LED color is unintentionally excited.

In principle, this can be achieved in different ways. In this way, the microcontroller could repeatedly generate individual excitation pulses and directly control the corresponding power transistors 7R, 7G and 7B accordingly. However, since this means a comparatively high level of control expenditure, it is possible—without requiring greater expenditure on the microcontroller 4—to cycle through the individual colors only slowly, which is disadvantageous because the color change may then be more noticeable.

It would also be possible to increase a basic clock by frequency multiplication and then determine the on/off phases by counting the frequency multiplied clock signals. In order to avoid permanent spikes, however, this requires that the individual colors do not drift apart and, moreover, care is taken that LEDs of a first color are only switched on when the LEDs of the previously excited color are switched off.

The present invention achieves synchronization of the color sequence without great effort in such a way that LEDs of a color to be subsequently excited are only switched on after the LEDs of the previously excited color are switched off. Structurally, this can be implemented without problems and with extremely little effort by waiting before switching on the excitation power for LEDs of a further color until a falling edge is detected in the excitation power for the LEDs of the previously excited color. Such a detection is possible with simple logic gates.

It is also possible to store a pulse count value once for each color channel, which indicates how many (possibly frequency-multiplied) clock pulses an LED of a given color is excited until arousal is terminated. Such a pulse counter value must be stored only once in a register or the like for each LED color as long as the illumination requested by the user does not change, ie neither color temperature nor intensity nor color shade are changed; after that, the microcontroller is no longer needed to cycle through the LED sequence. Rather, upon detection of the falling edge of the previous excitation pulse, the next color can be switched to active, i.e. the LEDs of the next color are supplied with power, and at the same time a counter is started from 0, which counts clock pulses upwards until the count value corresponds to the one for the currently excited color corresponds to the stored counter value. In order to switch the respective LED active, a corresponding signal can be sent to the power transistors 7 or the driver circuit provided for them, similar to the transistor 5E controlling the total power receiving a signal from the comparator 5B allowing or preventing the current conduction.

Thereafter, the falling edge can be detected for the next LED to be excited and the process repeated for the next LED color to be excited.

It can be seen from the above that with such a procedure it is possible to cycle through the color sequence again and again without the intervention of the microcontroller 4 and that a color shade and a color temperature can also be selected in the process.

In the arrangement described, however, it is not yet possible to interpose the pause times required to reduce the maximum light intensity that is still possible, during which no LEDs are energized at all. However, this can easily be changed by additionally providing a register in which the duration of an excitation pause to be provided between two successive excitations to reduce the intensity is stored. Upon detection of a falling edge, an excitation pause can be provided by counting clock pulses without any LED being excited until the count value corresponds to the value stored in the respective register and only then is the excitation of the following LED started. It is even sufficient to provide a single pause register if, as in 3b shown, between each two excitation pulses always equally long pauses <p1>=<p2>=<p3> should be inserted or if only a single pause of the total length <p1>+<p2>+<p3> is inserted after all color groups have run through were, as in 3c shown. It should be noted that an implementation such as that in 3c is shown, is particularly simple in terms of hardware and, with sufficiently fast cycling through all colors, no flickering or the like is to be expected even if a single pause follows the excitation of all colors. It should also be mentioned that when the entire off-duration is divided into several breaks, these breaks may not all have to be of the same length. For example, spectral sensitivity differences of the human eye and different efficiencies of LED light emission can be taken into account from the outset by using different lighting durations from LED color to LED color to achieve a pure white maximum intensity. If only the light intensity is reduced without changing the desired color, an on/off duty cycle can then be calculated in each case, which is obtained for all Led colors by multiplying by the same factor, but given the differences from Led color to Led colour Lighting durations with pure white light lead to different lengths of lighting breaks for the different colors.

The cycling between the individual colors can also then take place very quickly without great effort, in particular without great effort with regard to the microcontroller after the adjustment has been made.

It should be mentioned that the procedure described also enables a very rapid, dynamic change in the intensity and/or color of the light generated with an overall LED arrangement or the light generated in different groups in each case with different groups; In the preferred embodiment with edge detection, it is particularly advantageous that the microcontroller is essentially only required to feed on/off counter values into corresponding (local) registers until the operating parameters change, and no bandwidth-intensive control signals are required.

The dynamic change in light thus preferably only requires a change in the respective locally stored register values, which also takes place quickly. In particular, there is no need for an interruption in the lighting if the stored register values can remain valid until they change thanks to edge detection, in which case, in particular, the count values stored in the individual registers can be changed successively. This possibility of rapid changes makes the invention particularly suitable for reacting dynamically to changed requirements, for example to respond to sensor signals such as motion detectors, buttons, door contacts, temperature sensors, etc. to react.

It should also be pointed out that, if necessary - instead of explicitly testing groups for overload as described - the impedance or a characteristic value of the connected LEDs can be determined by color and/or group by color and then it can be determined, for example, by calculation which LED (groups ) can be operated simultaneously. It is clear that channel groups can be preselected more precisely for alternating operation, in particular when the load is known, and power can be distributed accordingly; it can then be better guaranteed, for example, that changes in intensity and/or color are only permitted to the extent that no overload occurs.

In order to allow a precise measurement and thus an exact calculation of the permissible intensity and color saturation settings, the current flowing through the LEDs can also be digitized if necessary and the digitized current values can be used to determine permissible operating parameters instead of just carrying out a simple analog measurement using a comparator comparison . In 1 a data line 5h is shown for this option, with which the corresponding digitized current values can be transmitted to the microcontroller. It should be mentioned that this is not mandatory but optional.

If necessary, the impedance can also be determined with one or more low measuring voltages, which are fed to the respective LED or LED groups to be tested, by determining whether a specific current is exceeded in each case or not.

Since it is possible to record a total current centrally, such a check for excessive currents can be carried out with little effort, because the circuit required for this is only required once for sequential testing of all LEDs one after the other or certain LED groups one after the other.

By using one or the comparator, the currents occurring across the shunt resistor when the measuring voltage is applied are detected or it is detected whether a specific threshold is not exceeded. Since the measuring voltage only has to be provided with very low power, this is simple in terms of circuitry.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 102008029816 A1 [0007]
• DE 202020101445 U1 [0008]
• EP 2592903 A2 [0009]
• US 8736183 B2 [0010]
• US10009974B2 [0011]
• US10159131B1 [0012]
• US20150145431A1 [0013]
• US 201900290931 A1 [0014]

Claims (10)

A DC LED power supply comprising a plurality of power channels configured to supply power from the same power source to LEDs of different colors; a microcontroller configured to drive power channels to result in a color temperature according to a power channel excitation ratio; and an overload protection circuit, characterized in that the overload protection circuit is designed to detect a total current supplied to the LEDs, a total power supplied to the LEDs and/or a resistance of connected LEDs; and the microcontroller is configured, in response to an under-total resistance, over-current, and/or over-total power, to drive the power channels to alternately power LEDS of different colors while maintaining a color-determining excitation ratio, wherein between the power channels is alternated so rapidly that alternation is visually imperceptible. LED DC voltage supply according to the preceding claim, characterized in that the microcontroller is designed to control power channels in such a way that, in the case of a permissible total current, they feed power from the same power source at least partially simultaneously to LEDs of different colors. LED DC voltage supply according to the preceding claim, characterized in that power is fed from the DC voltage power source into the power channels via a two-pole connection, the overload protection circuit being designed to determine an instantaneous total power in the first line and the microcontroller being designed for this purpose to control a switching of the power to the LEDs of different colors in current paths leading to the second line. LED DC voltage supply according to the preceding claim, characterized in that a switch-off circuit which acts jointly on all channels is provided in the first line, and a comparator of the overload protection circuit is designed to drive this directly. LED DC voltage supply according to one of the preceding claims, characterized in that the microcontroller is designed to change the duration of a switch-on. LED DC voltage supply according to one of the preceding claims, characterized in that a power channel switching control is provided, which detects the switching off of the power in a previously energized power channel and in response to a switch-off edge in a previously energized power channel another channel switching power feeding, preferably after a brief all-off period and/or dividing an all-off time. LED DC voltage supply according to one of the preceding claims, characterized in that the microcontroller is designed to drive several groups of LEDs of all colors differently. LED DC voltage supply according to the preceding claim, characterized in that the microcontroller has at least one additional sensor input and is designed to change, preferably locally, a light intensity or a light temperature in response to sensor signals received there. LED DC voltage supply according to one of the preceding claims, characterized in that the microcontroller is designed to, depending on a currently specified desired color temperature and depending on the size of the deviation of the recorded current fed to the LEDs as a whole, the current fed to the LEDs as a whole Power and / or the resistance of all connected LEDs to the maximum permissible value to drive power channels in such a way that it is partially simultaneously and partially alternately fed in the case of only a small permissible excess of color channels. LED DC voltage supply according to one of the preceding claims, characterized in that the power output is pulse width modulated.

Images