Device and method for operating a discharge lamp, in particular for projection purposes

Abstract

Device for operating a discharge lamp, in particular for projection purposes, comprising: - at least one AC-operable discharge lamp with a first electrode and a second electrode and - a control device which is designed to provide at least two current waveforms (W1, W2) for controlling the discharge lamp, wherein the respective current waveforms (W1, W2) are characterized by a respective envelope curve (I L ) with respect to a lamp current to be controlled through the discharge lamp and by a respective polarity curve (K1, K2) with respect to the flow direction of the lamp current to be controlled, - a measuring device for determining a measured value (X) correlated with a state variable of the discharge lamp, which is suitable for enabling a conclusion to be drawn about a state of the electrode geometry of the first electrode and / or the second electrode, characterized by - an evaluation device for determining an evaluation characteristic (Y1, Y2) which characterizes the influence of one of the at least two current waveforms (W1, W2) on the electrode geometry of the first electrode and / or second electrode when driving the discharge lamp with one of the at least two current waveforms (W1, W2), - wherein the control device is designed to store the evaluation characteristic (Y1, Y2) determined for one of the at least two current waveforms (W1, W2) for the associated current waveform of the at least two current waveforms (W1, W2), and - in a first operating interval (22a), to select one of the at least two current waveforms (W1, W2) with the aid of the evaluation characteristic (Y1, Y2) stored for this purpose for control in a second operating interval (22b) following the first operating interval.

Description

[0001] The invention relates to a device for operating a discharge lamp, in particular for projection purposes, according to the preamble of patent claim 1. Furthermore, the invention relates to a method for operating a discharge lamp, in particular for projection purposes, according to the preamble of patent claim 17.

[0002] Gas discharge lamps for use in video projection applications are typically designed for AC operation, so-called short-arc lamps. These provide an arc length of approximately 1 millimeter to achieve high optical image quality, which places particular demands on the stability of the electrode geometry. Ultra-high pressure mercury lamps are particularly used for this purpose. Depending on the manufacturer, such lamps are referred to as P-VIP lamps or UHP lamps, for example. Due to the high current density, the electrode tips of such a lamp are subject to high thermal stress during operation, which leads to changes in shape, i.e. a change in the electrode geometry, specifically burn-back of the electrodes and / or a change in the tip length or tip diameter or tip shape and / or tip wandering.The resulting changes in arc length and position significantly influence both the lamp voltage and the effective luminous efficacy in the projection device as well as the general aging behavior of the discharge lamp, for example devitrification or blackening.

[0003] Common approaches to achieving a long discharge lamp service life and the most consistent light generation quality possible, for example with regard to arc stability, aim to select the most favorable operating mode over the service life, especially with regard to a commutation frequency that changes the polarity of a discharge current flowing through the discharge lamp, alternating operating intervals with different frequencies, and introducing so-called DC phases, whereby, by omitting scheduled commutations in the lamp current, increased melting of the existing electrode tip occurs at the electrode of the gas discharge lamp, which is currently acting as the anode. Such interventions can be made depending on the measured lamp voltage.The voltage dependence is determined at the time of manufacture of the projection device and cannot be changed during the lamp burning time.

[0004] In this context, EP 2 168 408 B1 discloses a method for controlling a gas discharge lamp, wherein an operating frequency of the lamp is switched from a first fixed frequency value to a second fixed frequency value as a function of a fixed voltage value, whereby a peak builds up on an electrode of the lamp in a first operating mode, and the peak on the electrode is at least partially melted back in a second operating mode.

[0005] From DE 10 2009 006 338 A1 a method for operating a gas discharge lamp is known, in which a DC voltage phase is repeatedly applied at a predetermined time interval and depending on fixed voltage limit values.

[0006] From US 2011 / 0 204 811 A1 a discharge lamp operating method is known, wherein two operating modes are used successively in controlling the discharge lamp, and a division of the operating cycle between the modes is correlated by a ratio between a measured operating voltage and a desired operating voltage.

[0007] Generally, very low frequencies in the range 0 Hertz (DC) to 30 Hertz cause the electrode tip to melt and shorten, resulting in an increase in the lamp voltage. Medium frequencies in the range 30 Hertz to 120 Hertz generally lead to electrode tip growth. High frequencies in the range above 120 Hertz lead to slow electrode tip growth, which is, however, overlaid by material removal, usually on the sides of the electrode tips. The frequency limits can only be roughly defined and depend, among other things, on the electrode design, the operating current of the discharge lamp, and the aging state of the discharge lamp.

[0008] Although the basic mechanisms of electrode tip influence are known, they depend on many different factors, primarily the length and diameter of the electrode tip, the condition of the electrode head, and the discharge current. Thus, it is not possible to predictably assume the same current waveform to be optimal for achieving a long service life for every lamp type, especially since lamps also change in unpredictable ways with increasing service life.

[0009] It is an object of the present invention to provide a device and a method which operates an AC-operable discharge lamp in such a way that the lifetime performance is improved in a particularly simple manner.

[0010] This object is achieved by a device having the features of patent claim 1 and by a method having the features of patent claim 17. Advantageous developments of the present invention are the subject of the dependent claims.

[0011] The invention is based on a device for operating a discharge lamp, in particular for projection purposes, comprising at least one AC-operable discharge lamp having a first electrode and a second electrode, and a control device designed to provide at least two current waveforms for controlling the discharge lamp, wherein the respective current waveform can be described by a respective envelope curve with respect to a lamp current to be controlled through the discharge lamp and by a respective polarity curve with respect to the flow direction of the lamp current to be controlled. The device further comprises a measuring device for determining a measured value correlated with a state variable of the discharge lamp, which is suitable for enabling a conclusion to be drawn about a state of the electrode geometry of the first electrode and / or the second electrode.

[0012] According to the invention, the device is further developed by an evaluation device for determining an evaluation characteristic which characterizes the influence of one of the at least two current waveforms on the electrode geometry of the first electrode and / or second electrode when driving the discharge lamp with one of the at least two current waveforms, wherein the control device is designed to store the evaluation characteristic determined for one of the at least two current waveforms for the associated current waveform of the at least two current waveforms and to select one of the at least two current waveforms in a first operating interval with the aid of the evaluation characteristic stored for this purpose for driving in a second operating interval following the first operating interval.

[0013] The invention is based on the finding that current waveforms provided in a control device, which is part of the device for operating a discharge lamp, can be evaluated (classified) with regard to their influence on the electrode geometry of the discharge lamp, thus providing a way to specifically influence the state of the electrode geometry of the discharge lamp. Depending on the current flowing through this electrode, a local electrode tip, as already described above, can form on an electrode head of an AC-operated discharge lamp at the point of origin of the arc, which extends between the two electrodes during gas discharge.A change in the electrode geometry is therefore understood to mean the burnback of an existing electrode tip or the growth of an electrode tip on the respective electrode, i.e., on the first electrode and / or the second electrode. Operating the discharge lamp with shaped electrode tips in a specific configuration ensures stable and uniform operation of the discharge arc. The present invention enables a more reliable maintenance of a suitable electrode geometry.

[0014] The invention thus makes it possible to continuously determine during the service life which operating mode or current waveform can achieve optimum performance at the current time in order to be able to make a targeted change to the operating mode or current waveform if necessary.

[0015] In this way, a continuous learning process essentially takes place, with the device using the evaluation parameters to determine whether a currently executing current waveform continues to lead to optimal electrode behavior or whether there are other current waveforms that lead to better behavior at that time. This takes into account the realization that one and the same current waveform can change the type and strength of its influence on the electrode over the course of its service life and, for example, a current waveform leading to burn-back can lead to coalescence at a later point in time, and vice versa. In this sense, it can also be referred to as a self-learning controller. The invention is therefore also suitable for use with several different envelope curves that can be activated as required. Of course, a single envelope curve can also be used to generate all current waveforms.

[0016] The invention is proposed for use in devices for operating discharge lamps, in particular in the form of high-pressure or very high-pressure discharge lamps, for example for effect lighting applications or headlight applications for vehicles or for general lighting applications, such as in interior lighting.

[0017] Preferably, the envelope curve can be identical for each of the at least two current waveforms. A sectionally constant envelope curve can be described, for example, by an envelope vector. Likewise, a polarity curve of the discharge current can be described by a commutation scheme, for example, in the form of a commutation vector.

[0018] The measured value can in particular be a voltage value which is determined from the voltage applied across the discharge lamp. The voltage across the discharge lamp is usually referred to as the burning voltage. An increase in the burning voltage thus correlates with the melting of at least one electrode tip on the first or second electrode, which results in an effective lengthening of the electrode spacing and thus the length of the discharge arc. In a corresponding manner, a reduction in the burning voltage correlates with tip growth on at least one of the two electrode surfaces, which reduces the distance between the two electrodes, which is effective for the arc length.The influence on the electrode geometry can now be characterized using an evaluation index, for example which of the current waveforms causes tip growth and which of the current waveforms causes melting of the electrode tips.

[0019] According to an advantageous development, the control device is designed to select one of the at least two current waveforms for control in the second operating interval depending on a deviation of the measured value from a predeterminable target value. The predeterminable target value is, for example, a value at which a stable arc is achieved. It can be provided that the predeterminable target value is adjusted over the course of the operating time based on evaluations of the measured value. Preferably, the control device can be designed to select the corresponding current waveform in such a way that, in the event of a greater deviation of the measured value from the predeterminable target value, a current waveform with a stronger influence on the design of the electrode tips is selected, so that an accelerated approach to the target value can be achieved.

[0020] In an advantageous embodiment, the control device, together with the measuring device and the evaluation device, forms a closed control loop for regulating the measured value to the target value. This achieves automatic stabilization of the arc of the discharge lamp and allows the discharge lamp to operate in a routine mode.

[0021] According to an advantageous development, the control device is designed to specify the target value as a function of predefinable operating parameters of the discharge lamp. This can be done, for example, as a function of an operating current or an operating power of the discharge lamp. In particular, in dimmed operation of the discharge lamp, in which the operating power is reduced compared to the nominal power, provision can be made to adjust the target value accordingly. Furthermore, provision can be made for the previous operating time of the lamp, in other words the lamp age, to be taken into account when specifying the target value. For example, provision can be made to determine the target value from a characteristic curve.

[0022] Alternatively, the target value can be determined by analyzing the measured value with regard to its operating range, which can be influenced by setting different current waveforms. For example, the stability of the discharge arc can be taken into account here. For a measured value determined by the operating voltage of the discharge lamp, a target value of 70 volts can be specified for a brand-new discharge lamp. Over the course of this lamp's operating life, the effective distance between the two electrodes will increase due to continuous wear of the electrodes, causing the operating voltage to systematically increase. This increase can be determined, for example, based on a model or on the evaluation of the measured value.For example, a sudden reduction in the average lamp voltage detected after a device restart can be used to infer that the discharge lamp has been replaced. As a result, the target value can be reset to the initial value (default value).

[0023] According to a further advantageous embodiment, the device according to the invention comprises a device for determining the temporal change of the measured value, wherein the evaluation device is designed to determine the evaluation index as a function of the temporal change of the measured value. This also allows, for example, a rate of change to be determined as an evaluation criterion.

[0024] According to an advantageous development, the device for determining the temporal change of the measured value is designed to apply a nonlinear time scale when recording and evaluating the measured value. A preferred method for determining the temporal change consists in determining how long it takes for the measured value to change by a predefined increment in a positive or negative direction. The determination of this time period, which is required for this change by the predefined increment, can be provided by a counter, in particular a counter with a logarithmic time weighting.Thus, the measurement, preferably a voltage measurement, starts at a specific time, then continuously determines whether a specific step size is exceeded, whereby the counter then restarts again and again so that it can be continuously recorded whether the rate of change changes, in particular whether the sign changes.

[0025] According to a further advantageous embodiment, the evaluation index is correlated with the magnitude of the change in the state of the electrode geometry. This allows current waveforms to be directly selected based on their suitability for influencing the electrode geometry state.

[0026] According to a further advantageous embodiment, at least the respective polarity curve of the at least two current waveforms is stored in a table in the control device, together with the respectively associated evaluation index. Entries in the table, comprising at least the respective polarity curve and the respectively associated evaluation index, are preferably arranged sorted by the evaluation index. This results in particularly simple access for selecting a suitable current waveform in order to achieve a specific influence on the electrode formation. For example, it can be provided to store current waveforms that cause extreme tip growth on the electrodes in one edge area of ​​the table, and to store current waveforms that cause extreme melting of electrode tips at another end of the table.In a middle area of ​​the table, current waveforms can be arranged which do not or only insignificantly influence the shape of the electrode tips.

[0027] According to an advantageous development, the control device is designed to redetermine all evaluation parameters stored in the table and store them in the table, in particular to regenerate the table, as a function of a predeterminable update signal. Provision can be made to store the table contents in a non-volatile memory. The update signal can, for example, be designed such that it triggers a redetermination of the evaluation parameters of the current waveforms stored in the table at specific time intervals. It can also be provided that a re-evaluation of the current waveforms stored in the table takes place each time the device is switched on again after an interruption in lamp operation.

[0028] The so-called commutation schemes, i.e. the arrangement of the polarity changes (zero crossings) of the respective polarity curve for each of the current waveforms, are preferably specified during the device's manufacturing process, for example during programming of the control device, which may be in the form of an electronic ballast (EB). During operation, the evaluation parameters are continuously updated. This can take place, for example, by updating at regular intervals during the operation of a mode, preferably between 0.1 seconds and 30 minutes. The length of this interval is particularly preferably selected depending on the currently prevailing rate of change of the measured value correlated with a state variable of the discharge lamp. This means that for high rates of change, a shorter interval is selected, and vice versa.The regular intervals can be adjusted depending on the rate of change, in particular reduced if the rate of change increases and / or increased if the rate of change decreases. Particularly preferably, a possible range of change for the regular intervals is between 10 seconds, in particular 30 seconds, and 3 minutes, in particular 10 minutes; i.e., preferred ranges are optionally 10 seconds to 10 minutes, 10 seconds to 3 minutes, 30 seconds to 3 minutes, and 30 seconds to 10 minutes.

[0029] Furthermore, it may also be advantageous to activate currently suboptimal operating modes for an additional test phase in order to update their evaluation. The table can contain only the currently determined values ​​of the evaluation indicators and / or a progression of the evaluation indicators over time with corresponding historical values ​​(history) for volatile and / or non-volatile storage. For example, it can be provided that a spontaneous significant change in the evaluation indicator triggers a re-evaluation of all current waveforms.

[0030] According to a further advantageous embodiment, the control device is designed to store a validity index, in particular in the form of an age indication of the evaluation index, in addition to the evaluation index, in order to select the current waveform for the second operating interval additionally depending on the validity index. This results in the advantage that all current waveforms remain up-to-date, so that the most suitable current waveform can be selected for a desired electrode influence. Various scenarios are conceivable for this. On the one hand, an age indication can be provided that characterizes a current waveform that has not been used for a long time and for which, accordingly, no current evaluation indexes are available. Likewise, the exclusion of certain current waveforms in special operating modes, for example when dimming the discharge lamp, can be provided.Furthermore, a current waveform can be set as a default current waveform, which is activated, for example, under certain operating conditions.

[0031] According to a further advantageous embodiment, the control device is designed to include past values ​​of this evaluation index when determining the evaluation index in order to reduce the influence of a noisy measured value. In this case, it can be provided that the old, already existing value is included in the new value, or that a new entry follows and one or more old values ​​are retained as history, for example in the manner of a ring buffer. To include the old value or values ​​in the new values, it can be provided, for example, to form a moving average, to record the median over, for example, 20 values, or to add the new value and the old value to one another in a predeterminable ratio. For example, 80 percent of the old value and 20 percent of the new value can be added together, thereby achieving the effect of a low-pass filter.This eliminates high-frequency interference, such as glitches and similar phenomena. This, in particular, reduces the memory requirements for storing the corresponding values.

[0032] According to a further advantageous embodiment, the control device is designed to determine a new value for the respectively associated evaluation index for each of the at least two current waveforms after the device has been initially put into operation and / or restarted after a previous shutdown, in particular after the discharge lamp in the device has been replaced. Provision can be made for the control device to recognize that the lamp has been replaced when a significant jump in the lamp's operating voltage occurs. Furthermore, a predeterminable initialization method can be provided, for example, taking into account evaluation indexes achieved with the old lamp that has been replaced. Irrespective of a currently existing evaluation table, provision can be made for all available current waveforms to be cyclically tested and the table with the evaluation indexes to be rebuilt.

[0033] According to a further advantageous embodiment, the at least two current waveforms can each be represented by multiplying an unsigned envelope curve as a function of time by at least one first square-wave signal of constant magnitude and at least one second square-wave signal of constant magnitude, wherein the first square-wave signal has a first fundamental frequency and the second square-wave signal has a second fundamental frequency different from the first fundamental frequency. In the simplest case, this means that the square-wave signal has only a single square-wave fundamental frequency, i.e., the periodicity of this signal results from the fact that the respective square-wave signal, shifted by two zero crossings, comes to lie congruently on itself.When shifted by a distance of two zero crossings, at a duty cycle of 50 percent, i.e. by half a period, the rectangular signal is inverted, i.e., has the opposite sign. However, the rectangular signals can also have a superimposed periodicity, for example by the edges of the polarity change from -1 to +1 or from +1 to -1 being shifted relative to a period length predetermined by the respective rectangular fundamental frequency, for example for synchronization with segment transitions of a color wheel, which is commonly used in DLP projectors. For example, a rectangular signal can arise whose (exact) periodicity corresponds to a third or a quarter of the corresponding fundamental frequency.

[0034] According to an advantageous development, the respective functional curve of the at least two current waveforms can be represented by a corresponding definition of a first temporal offset of the first square-wave signal to the envelope curve and by a corresponding definition of a second temporal offset of the second square-wave signal to the envelope curve.

[0035] According to a particularly preferred development, a parameter set describing the at least one first square-wave signal and the at least one second square-wave signal as well as their respective temporal relationship to the envelope curve is stored in the control device, wherein the control device is designed to generate a commutation scheme characterizing the respective polarity curve as a function of the stored parameter sets. An envelope curve can, for example, be predetermined in a customer-specific manner in a device usable in a DLP projector, in particular as a function of a color wheel used in this projection technology, wherein a reference point is defined in the temporal progression of this envelope. A time offset to the first square-wave signal and a time offset to the second square-wave signal are specified as a function of this defined reference point.Together with the first fundamental frequency and the second fundamental frequency, the respective current waveform can be defined. If more than two square wave signals are intended to synthesize the current waveform, several parameter pairs must be added accordingly. Typically, a reference point on the envelope is defined at a transition from a high current to a low current.

[0036] According to a further advantageous embodiment, the control device is designed to provide a predeterminable bandwidth of the evaluation parameters of the at least two current waveforms and, in the event of an alignment of the evaluation parameters of the at least two current waveforms, to generate a further current waveform with a polarity curve that is not present in any of the at least two current waveforms previously provided. In this case, it can be provided to generate a new current waveform on the basis of existing parameter sets for generating current waveforms, for example using the first and second fundamental frequencies while simultaneously varying at least one time offset. Alternatively, it can be provided that at least one of the two fundamental frequencies is adjusted.

[0037] Another alternative is to replace an existing current waveform with a new one instead of inserting another current waveform, especially in cases where multiple current waveforms have approximately the same evaluation parameters. The same principle can also be applied when the specified bandwidth is exceeded, for example, when a current waveform exerts an extremely strong influence on the electrode tip design, for example, through rapid tip growth, which shifts the discharge lamp's operating voltage into an excessively low range, or the total melting of the electrode tips, which can have negative effects on arc stability.

[0038] Preferred ranges for the first frequency are in the range from 5 Hertz to 500 Hertz, in ascending order the lower limit is increasingly preferred to be at least 15 Hertz, 40 Hertz, 50 Hertz, 60 Hertz, and independently of this the upper limit is increasingly preferred to be 500 Hertz, 350 Hertz, 250 Hertz, 180 Hertz. The preferred range for the second frequency can be in the range from 0 Hertz to 500 Hertz. In ascending order the lower limit of the preferred range for the second frequency is increasingly preferred to be at least 0.1 Hertz, 1 Hertz, 5 Hertz, 10 Hertz, and independently of this the upper limit of the preferred range for the second frequency is increasingly preferred to be 500 Hertz, 350 Hertz, 250 Hertz, 180 Hertz, 150 Hertz, 120 Hertz, 90 Hertz. In particular, very small frequencies in the range close to 0 Hertz, but still different from 0 Hertz, can be used.Preferably, a fixed relationship between the two frequencies is chosen, for example, the first frequency can be twice, four times or six times the second frequency.

[0039] The invention further relates to a method for operating a discharge lamp, which comprises a first electrode and a second electrode, with alternating current, in particular for projection purposes, by providing at least two current waveforms for controlling the discharge lamp, wherein the respective current waveform can be described by a respective envelope curve with respect to a lamp current to be controlled through the discharge lamp and by a respective polarity curve with respect to the flow direction of the lamp current to be controlled. The method further comprises determining a measured value correlated with a state variable of the discharge lamp, which is suitable for enabling a conclusion to be drawn about a state of the electrode geometry of the first electrode and / or the second electrode.

[0040] According to the invention, the method is further developed by determining an evaluation index which is correlated with a change in the electrode geometry of the first electrode and / or second electrode when driving the discharge lamp with one of the at least two current waveforms, storing the evaluation index determined for one of the at least two current waveforms for the associated current waveform of the at least two current waveforms, and selecting one of the at least two current waveforms in a first operating interval with the aid of the evaluation index stored for this purpose for driving in a second operating interval following the first operating interval.

[0041] The advantages and features, as well as embodiments, described for the device according to the invention apply equally to corresponding methods and vice versa. Consequently, corresponding method features can be provided for device features and vice versa.

[0042] The features and feature combinations mentioned above in the description, as well as the features and feature combinations mentioned below in the description of the figures and / or shown alone in the figures, can be used not only in the respective combinations specified, but also in other combinations or on their own, without departing from the scope of the invention. Thus, embodiments are also to be considered disclosed by the invention that are not explicitly shown or explained in the figures, but which emerge and can be produced through separate combinations of features from the explained embodiments.

[0043] Further advantages and features will become apparent from the following description of exemplary embodiments, taking into account the accompanying figures. In the figures, identical reference numerals denote identical features and functions.

[0044] They show: Fig.1 shows a simplified schematic representation of a preferred embodiment of a method according to the invention, Fig. 2 shows a simplified schematic representation of a time course of a measured value correlated to a state variable of the discharge lamp, Fig. 3a-d show a simplified schematic representation of a first method for determining a current waveform, and Fig. 4a-d show a simplified schematic representation of a second method for determining a current waveform.

[0045] A projection device according to the present invention comprises a control device for controlling the discharge lamp. The control device is designed to provide at least two current wave functions for controlling the discharge lamp. A current wave function is composed of the envelope curve, also referred to below as envelope information (for example, correlated with a current amplitude), and the polarity information (commutation scheme). Different commutation schemes can be combined with the same envelope information. The envelope information is described, for example, by an intensity vector. A commutation scheme is defined either by a commutation vector, by a single square-wave signal, or by an alternative description. At least two different commutation schemes are provided for use.

[0046] A measuring device of the projection device serves to determine a measured value correlated with a state variable of the discharge lamp, wherein a conclusion can be drawn from the measured value about the electrode burn-back and / or the growth of electrode tips on the first and / or second electrode. Since the distance between the electrodes, which determines the length of the arc burning between the two electrode tips of the discharge lamp, cannot be determined directly by the control device of the projection device, for example an electronic ballast, the lamp voltage across the discharge lamp can be used as a measured value for an indirect measurement of the distance between the first electrode and the second electrode. The lamp voltage includes the so-called cathode drop, the anode voltage and the voltage drop in the actual discharge arc.To a good approximation, cathode drop and anode voltage do not depend on the electrode spacing at the same current; the voltage drop in the discharge arc is, to a good approximation, proportional to the spacing. The lamp voltage can therefore be used as a control variable for the electrode spacing. The growth of spikes on the electrodes causes a reduction in the voltage, while a melting of an increase causes a decrease. The rate of change of the voltage, i.e., the difference between an initial value and a final value of the voltage within a predefined measuring interval divided by the width of the predefined measuring interval, is a measure of the effectiveness of the individual operating modes with regard to specifically influencing the electrode tips.

[0047] A preferred embodiment of a method according to the invention for implementation on a device for operating a discharge lamp is described in Fig.1. Using this method, a closed control loop is implemented for controlling the distance between the first electrode and the second electrode. The method starts at a step S0, starting with a running index i=1. If the device has already been in operation and is switched on again after a break in operation, the running index i can expediently assume the value it had at the end of the previous operating period. In a step S1, the projection device is provided with a commutation scheme K i operated. In a step S2, a measurement of a measured variable X is performed. The measurement of the measured variable X can be performed continuously or at discrete, predefined times. Such measured value acquisition can be performed, for example, by a microprocessor / microcontroller with an integrated analog-to-digital converter.

[0048] In the following, the measured quantity X is represented by a lamp voltage (burning voltage) U as an example.

[0049] To eliminate interference caused, for example, by background noise in the voltage signal, voltage changes due to changes in the starting points of the discharge arc or changes in the lamp current due to the necessary power control, filtering of the measurement signal can be provided.

[0050] Based on the obtained measured values, a rate of change dX / dt is determined in a third step S3. For example, a predeterminable change level dX can be specified, and a counter can be used to determine how long it takes from a predeterminable starting time until a change in the measured value X by the change level dX has occurred. The change can occur in both directions, both positive and negative, i.e., an increase in the measured value X or a decrease in the measured value X.

[0051] A counter with nonlinear time scale evaluation can be used to evaluate the rate of change dX / dt. Logarithmic scaling is particularly suitable for this purpose.

[0052] Depending on the determined rate of change dX / dt, during operation with the commutation scheme K i based on the temporal change of the measured value X an evaluation value Y i which characterizes the intensity of the burn-back or growth of electrode tips on the first electrode and / or the second electrode when operating with this commutation scheme. In determining this evaluation value Y i Further factors can be taken into account, for example, whether this commutation scheme can achieve a continuous curve without jumps and / or a monotonic curve without inflection points. The determined evaluation value Y i is used together with the commutation scheme Ki stored in a rating table Tab. The rating table Tab can be arranged so that the individual entries correspond to the rating values ​​Y i Furthermore, the evaluation table Tab can also have an entry for each element, which indicates when the corresponding commutation scheme K i was last rated, i.e. information about the age of the rating value Y i contains.

[0053] In a fifth step S5, the calculation of a measured value deviation ΔX is carried out from the difference between the measured value X, which is correlated with the electrode distance, and a target value X Zdetermined. A target / actual comparison of the electrode spacing of the discharge lamp is therefore based on an indirect evaluation of a measured value correlated with the electrode spacing, which can be determined during operation of the projection device. A determination can therefore be made on the basis of a model, a so-called observer. The model parameters can be statistically determined by previously carried out systematic measurements on a type of discharge lamp intended for use in the device, for example using the arc projection method, in which the relationship between the voltage applied to the lamp and the arc length of the discharge arc between the first electrode and the second electrode of the discharge lamp is determined, taking into account the discharge current flowing through the discharge lamp, which current can also be determined in the control device.To determine the electrode distance X or its deviation from a predeterminable target value, the voltage at the discharge lamp and the current flowing through the discharge lamp or the voltage at the discharge lamp and the power supplied to the discharge lamp can be used as a basis.

[0054] Depending on the determined deviation of the electrode gap from the specified target value X Z In a sixth step S6, a new commutation scheme K is selected i depending on at least the measured value deviation ΔX and the evaluation values ​​Y1, Y2, to Y available in the evaluation table Tab N . In addition, for the determination of the new commutation scheme K i the measured value X itself, which is correlated with the electrode distance, and / or a time t are taken into account. This results in a new commutation scheme K ifor example i=f(t,X,ΔX,Y1,Y2,...,Y N ).

[0055] After selecting the new commutation scheme K i the return to the first step S1 now takes place with a commutation scheme K i with a correspondingly updated index i.

[0056] Instead of the commutation schemes K i (i=1... N) can of course be a complete current waveform W i (i=1... M) can be stored in the table Tab, in particular several sets of current waveforms, which are then optionally activated for selection. In this case, the evaluation table Tab would contain the commutation schemes K1, K2, ... K N by the current waveforms W1, W2, ... W M Here, N can be equal to M, for example when using the same envelope curve I L for all current waveforms. When combining all commutation schemes K i(i=1... N) with all stored envelopes I L results for the number of resulting current waveforms W M M = N * L

[0057] Fig. Figure 2 shows the functionality of the method using the example of a simplified curve 21, consisting of individual curve sections 21a, 21b, 21c, 21d, 21e, which represents the course of the measured value X over time t. In the Fig. In the diagram shown in Figure 2, time t is plotted on the abscissa with three highlighted points in time: a first switching point T1, a second switching point T2, and a third switching point T3. The voltage U across the discharge lamp is plotted on the ordinate, which, representing the measured value X, is correlated with the distance between the electrodes of the discharge lamp. For simplicity, it is assumed that the voltage U corresponds to the measured value X (U ≙ X).

[0058] A target value Z as well as a lower limit UG and an upper limit OG are entered on the ordinate. The target value Z lies between the lower limit UG and the upper limit OG. In a first time period 22a, which extends up to the first switching time T1, the course of the voltage U follows the first curve section 21a with a negative gradient. Starting from an initial voltage which is above the upper limit OG, the electrode distance is initially reduced towards the target value Z with a gradient of -0.5 in arbitrary units. As the first time period 22a progresses, the target value Z is undershot and finally the lower limit UG is reached at the first switching time T1.

[0059] At the first switching time T1, a switch is made from the first commutation scheme K1 to the second commutation scheme K2, resulting in a slope of a second curve section 21b in a second time period 22b of +2 in arbitrary units. As a result, after the previous growth of peaks on the electrodes of the discharge lamp, thereby reducing the voltage U between the two electrodes, the peaks are melted back, causing the voltage U across the lamp to continuously increase again in the second time period 22b between the first switching time T1 and the second switching time T2, with the voltage U reaching the value of the upper limit OG at the second switching time T2.

[0060] At the second switching time T2, the system now switches back to the first commutation scheme K1, which again results in a gradient of -0.5 in arbitrary units for a third curve section 21c, which follows the second switching time T2.

[0061] During operation with the first commutation scheme K1, a change in the direction of voltage change occurs. This can be induced by a change in the electrode geometry. Accordingly, the third curve section 21c is followed by a fourth curve section 21d, which is characterized by an increase in voltage. At a third switching time T3, the change in the direction of voltage change for K1 is detected. A fourth time period 22d following the third time period 22c ends accordingly at the third switching time T3. In a subsequent fifth time period 22e, a change to the commutation scheme K4 occurs, which leads to a voltage drop with a gradient of -0.06 in arbitrary units. The corresponding fifth curve section is designated 21e.

[0062] In summary, the following is the Fig.2 graphically represented course of the curve 21, consisting of the curve sections 21a, 21b, 21c, 21d, 21e, presented in compact and tabular form, where each table row represents the state of the table Tab (for N = 4) at the times T1, T2 and T3 respectively and the actions resulting at these times: Time t Y1(K1) Y2(K2) Y3(K3) Y4(K4) action T1 -0,5 +2 +0,1 -0,06 K1->K2 OG reached, change to voltage increase T2 -0,5 +2 +0,1 -0,06 K2->K1 OG reached, change to voltage reduction T3 +0,03 +2 +0,1 -0,06 K1->K4 Change of Voltage change direction for K1 detected, change to the next commutation scheme with voltage reduction (K4)

[0063] This is a condensed representation of the Tab table at different points in time. Additionally, the "Action" column indicates what happened at each point in time.

[0064] Thus, the control device provides a control loop with the aid of which the state of the electrode geometry of the first electrode and / or the second electrode can be kept within a predeterminable range.

[0065] An alternative method for describing a current waveform, in particular for synthesizing a current waveform, is presented below. The control device is designed to generate current wave functions for controlling a discharge lamp. A current wave function is composed of the envelope information (current amplitude) and the polarity information (commutation scheme). The envelope information can be described by an intensity vector or, as shown below, by an envelope curve I. L . A commutation scheme K N is obtained by multiplying several square wave signals R j with different frequencies. The square wave signals are described by their frequencies f j , the phase position φ1 of the first square wave to the envelope and the phase positions φ j the remaining square wave signals to the first square wave signal.

[0066] The synthesis of a current waveform for the special case of a projection device in the form of, for example, a 3-LCD projection device (Liquid Crystal Display) or a 3-chip DLP (Digital Light Processing) is described below with reference to the Fig. 3a, Fig. 3b, Fig. 3c, Fig. 3d. The envelope information is displayed according to the Fig. 3a by a constant envelope curve I L over time t. Unit 1 is chosen arbitrarily and can represent 100 percent operation, which, for example, represents operation of the discharge lamp at its rated power. Fig. 3b shows the time course of a first square wave signal R1. Fig.Figure 3c shows a time characteristic of a second square-wave signal R2. Both square-wave signals, the first square-wave signal R1 and the second square-wave signal R2, are mean-free and each have a duty cycle of 50 percent. Furthermore, both square-wave signals each have only a single square-wave frequency, meaning that each of the two square-wave signals R1 and R2 changes sign exactly after half a period. Thus, the interval between two zero crossings of the respective square-wave signals R1 and R2 is always constant at half the period, meaning half the reciprocal of a first frequency f1 or a second frequency f2. The two frequencies f1 and f2 of the square-wave signals R1 and R2 are freely selectable. A preferred range for the first frequency f1 extends from 5 Hertz to 500 Hertz, and a preferred range for the second frequency f2 extends from 0 Hertz to 500 Hertz. The product of the envelope curve I L, the first square wave R1 and the second square wave R2 is in the Fig. 3d. The signal generated in this way, which is a current waveform W M can, despite its seemingly complex character, be described by a simple set of coefficients, namely the frequencies f1, f2 of the underlying square wave signals R1, R2 as well as the corresponding previously mentioned phase positions in relation to the envelope curve I L , which also results in the mutual phase position of the two square wave signals R1 and R2.

[0067] In the example shown, the first square-wave signal R1 has a first frequency f1 = 130 Hertz, and the second square-wave signal R2 has a second frequency f2 = 60 Hertz. This results in a frequency ratio of f2 to f1 of approximately 0.46. The phase shift, or mutual temporal offset, between the two square-wave signals is approximately 1.28 milliseconds. Since the envelope is constant over time, the temporal offset from the envelope can, in principle, be chosen arbitrarily.

[0068] A current waveform suitable for a 1-chip DLP (Digital Light Processing) projector, for example, must meet additional requirements than the current waveform shown above. Precise synchronization with the commonly used color wheel is required. This color wheel rotates in the projector's beam path and features multiple color segments assigned to different colors, each of which can have a different length. This means that the individual color sectors can each encompass a predefined angular portion of the total 360°. This allows different colors of a video image to be projected one after the other. Modern DLP projectors also have a setting option that allows the lamp current level to be individually configured in the individual color segments.This can, for example, result in higher brightness through an increase in the white segment or better color reproduction through appropriate matching of the individual colors. Typically, a customer, for example a manufacturer, who uses the device according to the invention in a projection device can define a plurality of, for example, three to seven current curves (envelope information regarding the current amplitude) for different sets of brightnesses in the individual segments and store them in a non-volatile memory, usually an EEPROM, of the projection device.

[0069] For practical purposes, commutation is performed at the transitions between the individual color segments of the color wheel, at the so-called spokes. The brightness drop caused by the current commutation is then imperceptible to the user, since the mixed color created at the spokes is either masked out by the projector anyway or used, for example, to increase the white light component.

[0070] Fig. 4a shows, according to a further embodiment, the envelope curve I L , which represents the amplitude curve of the desired brightness in the individual segments of the color wheel. The envelope curve I is shown. Lusing the example of a color wheel with six segments, whereby, assuming a constant rotation speed of the color wheel, i.e. a constant number of revolutions, the segment duration or the respective angular components are formed according to the following table: White (White) W 70° Cyan blue (Cyan) C 30° Blue (Blue) B 90° Green (Green) G 30° Red (Red) R 65° Yellow (Yellow) Y 75°.

[0071] The vertical dashed lines in the Fig. 4a, Fig. 4b, Fig. 4c, Fig. 4d mark the segment boundaries.

[0072] Fig.Figure 4b shows a first square-wave signal R1, in which the period duration of the individual pulses is slightly different. Pulses denote a respective section of the first square-wave signal R1, which extends between a first zero crossing (e.g., G -> R) and a third zero crossing (e.g., Y -> W), wherein a polarity change of the first square-wave signal R1 occurs at a second zero crossing (e.g., C -> B), which is the only zero crossing between the first zero crossing and the second zero crossing. This means, in contrast to the previous embodiment according to the Fig. 3b, the duty cycle is not uniformly 50 percent, but the zero crossings of the Fig.4b, taking into account a segmentation predetermined by the color wheel of the DLP projector, is essentially compressed into a predefined shape with a finite number of predetermined zero crossing points. In other words, the respective rectangle width, both in the negative direction (-1) and in the positive direction (1), is rounded up or down depending on a non-equidistant rasterization.

[0073] The same applies to the second square wave R2 according to the Fig. 4c. The period duration of the individual pulses for this second square-wave signal R2 is also slightly different. One revolution of the color wheel is thus correlated with the passage of the segments white W, cyan C, blue B, green G, red R, and yellow Y. This means that the corresponding time duration is linked to one complete rotation of the color wheel (360°) via the rotational speed of the DLP color wheel.

[0074] For the product of envelope curve I L , first square wave signal R1 and second square wave signal R2, the curve of the current waveform W M according to the representation of the Fig. 4d.

[0075] As already explained above, the first square-wave signal R1 and the second square-wave signal R2 according to this embodiment which can be used for DLP projection applications also have the same mean-value characteristics.

[0076] Due to the restriction of the possible zero crossings to specified commutation points, the instantaneous values ​​of the frequencies of the square-wave signals vary in this case around an average value f1 or f2. For the Fig. 4b and Fig.In the example shown in Figure 4c, these average values ​​are f1 = 45 Hertz and f2 = 25.7 Hertz. The ratio of the two frequencies is 0.57. The time offset of the square-wave signals in this example is 0 milliseconds between the envelope curve I L and the first square wave signal R1 and approximately 3.2 milliseconds between the first square wave signal R1 and the second square wave signal R2.

[0077] The same applies to the second square wave signal R2.

[0078] When applying the proposed method for synthesizing commutation schemes K i , in particular by integrating predeterminable intensity vectors, which provide a sectionally constant intensity profile of the envelope curve I L can be used in a particularly simple way to create a current waveform W Mwhich, with a suitable choice of frequencies and phase positions, also meets the requirement for a mean-free current waveform.

[0079] This synthesis method for current waveforms can be used independently for any form of AC-operated discharge lamps, regardless of the use proposed here with the present invention, and is therefore also suitable, for example, as an alternative to a method for varying the commutation vector, which characterizes a corresponding current waveform, as proposed in the subsequently published application with the official file number 10 2014 220 275.2.

[0080] The exemplary embodiments serve merely to illustrate the invention and are not limiting. In particular, the number of rectangular functions used for synthesizing the current waveforms and the number of current waveforms available for implementing the invention can be varied as desired. Likewise, there are no restrictions on how the corresponding parameter sets or signals are stored or stored in a corresponding device according to the invention.

[0081] Thus, it has been shown above how the operation of AC-operated ultra-high pressure discharge lamps with two electrodes, especially lamps for projection devices, can be optimized. List of reference symbols S0 Start S1 first step S2 second step S3 third step S4 fourth step S5 fifth step S6 sixth step Tab Rating Table i Index W1 first current waveform W2 second current waveform K1 first commutation scheme K2 second commutation scheme K N Nth commutation scheme Y1 first valuation key figure Y2 second valuation indicator Y N Nth evaluation indicator I L Envelope curve X measured value dX / dt rate of change dX change level t time ΔX measured value deviation X Z Target value U burning voltage OG upper limit Z target value UG lower limit T1 first switching time T2 second switching time T3 third switching point 21a first curve section 21b second curve section 21c third curve section 21d fourth curve section 21e fifth curve section 22a first period 22b second period 22c third period 22d fourth period 22e fifth period I Envelope curve R1 first square wave R2 second square wave W M Current waveform, general

Claims

[1] Device for operating a discharge lamp, in particular for projection purposes, comprising: - at least one AC-operable discharge lamp with a first electrode and a second electrode and - a control device which is designed to provide at least two current waveforms (W1, W2) for controlling the discharge lamp, wherein the respective current waveforms (W1, W2) are characterized by a respective envelope curve (I L ) with respect to a lamp current to be controlled through the discharge lamp and by a respective polarity curve (K1, K2) with respect to the flow direction of the lamp current to be controlled, - a measuring device for determining a measured value (X) correlated with a state variable of the discharge lamp, which is suitable for enabling a conclusion to be drawn about a state of the electrode geometry of the first electrode and / or the second electrode, characterized by - an evaluation device for determining an evaluation characteristic (Y1, Y2) which characterizes the influence of one of the at least two current waveforms (W1, W2) on the electrode geometry of the first electrode and / or second electrode when driving the discharge lamp with one of the at least two current waveforms (W1, W2), - wherein the control device is designed to store the evaluation characteristic (Y1, Y2) determined for one of the at least two current waveforms (W1, W2) for the associated current waveform of the at least two current waveforms (W1, W2), and - in a first operating interval (22a), to select one of the at least two current waveforms (W1, W2) with the aid of the evaluation characteristic (Y1, Y2) stored for this purpose for control in a second operating interval (22b) following the first operating interval. [2] Device according to claim 1, characterized byin that the control device is designed to carry out the selection of one of the at least two current waveforms (W1, W2) for the control in the second operating interval as a function of a deviation (ΔX) of the measured value (X) from a predeterminable target value (Xz). [3] Device according to claim 2, characterized by that the control device together with the measuring device and the evaluation device forms a closed control loop for controlling the measured value (X) to the target value (Xz). [4] Device according to claim 2 or 3, characterized by that the control device is designed to specify the target value (Xz) as a function of predeterminable operating parameters of the discharge lamp. [5] Device according to one of the preceding claims, characterized by - a device for determining the temporal change of the measured value (X), wherein the evaluation device is designed to determine the evaluation index (Y1, Y2) as a function of the temporal change of the measured value. [6] Device according to claim 5, characterized by that the device for determining the temporal change of the measured value (X) is designed to apply a non-linear time scale when recording and evaluating the measured value. [7] Device according to one of the preceding claims, characterized by that the evaluation index (Y1, Y2) is correlated with the strength of the change in the state of the electrode geometry. [8] Device according to one of the preceding claims, characterized bythat in the control device at least the respective polarity curve (K1, K2) of the at least two current waveforms (W1, W2) together with the respectively associated evaluation index (Y1, Y2) is stored in a table (Tab), wherein preferably entries in the table comprising at least the respective polarity curve and the respectively associated evaluation index are arranged sorted according to the evaluation index. [9] Device according to claim 8, characterized by that the control device is designed to redetermine all evaluation indicators stored in the table as a function of a predeterminable update signal and to store them in the table, in particular to regenerate the table. [10] Device according to one of the preceding claims, characterized byin that the control device is designed to store a validity index, in particular in the form of an age indication of the evaluation index, in addition to the evaluation index (Y1, Y2), and to carry out the selection of the current waveform (W1, W2) for the second operating interval additionally as a function of the validity index. [11] Device according to one of the preceding claims, characterized by that the control device is designed to include past values ​​of this evaluation key figure when determining the evaluation key figure (Y1, Y2) in order to reduce the influence of a measured value (X) with interference. [12] Device according to one of the preceding claims, characterized bythat the control device is designed to determine a new value for the respectively associated evaluation characteristic (Y1, Y2) for each of the at least two current waveforms (W1, W2) after a first commissioning of the device and / or after a restart after a previous shutdown, in particular after a replacement of the discharge lamp in the device. [13] Device according to one of the preceding claims, characterized by that the at least two current waveforms (W1, W2) are each generated by multiplying an unsigned envelope curve (I L ) can be represented as a function of time with at least one first rectangular signal (R1) of constant magnitude and at least one second rectangular signal (R2) of constant magnitude, wherein the first rectangular signal has a first fundamental frequency (f1) and the second rectangular signal (R2) has a second fundamental frequency (f2) different from the first fundamental frequency. [14] Device according to claim 13, characterized by that the respective function curve of the at least two current waveforms (W1, W2) is determined by a corresponding determination of a first time offset of the first square wave signal (R1) to the envelope curve (I L ) and by a corresponding determination of a second time offset of the second square-wave signal (R2) to the envelope curve (I L ) can be represented. [15] Device according to claim 13 or 14, characterized by that the at least one first square-wave signal (R1) and the at least one second square-wave signal (R2) as well as their respective temporal relationship to the envelope curve (I L ) are stored in the control device, wherein the control device is designed to generate a commutation scheme (K1, K2) characterizing the respective polarity curve as a function of the stored parameter sets. [16] Device according to one of the preceding claims, characterized by in that the control device is designed to provide a predeterminable bandwidth of the evaluation parameters (Y1, Y2) of the at least two current waveforms (W1, W2) and, in the event of an adjustment of the evaluation parameters of the at least two current waveforms, to generate a further current waveform with a polarity curve which is not present in any of the at least two current waveforms (W1, W2) provided so far. [17] Method for operating a discharge lamp comprising a first electrode and a second electrode with alternating current, in particular for projection purposes, by: - providing at least two current waveforms (W1, W2) for controlling the discharge lamp, wherein the respective current waveforms (W1, W2) are characterized by a respective envelope curve (I L) with respect to a lamp current to be controlled through the discharge lamp and by a respective polarity curve (K1, K2) with respect to the flow direction of the lamp current to be controlled, - determining a measured value (X) correlated with a state variable of the discharge lamp, which is suitable for allowing a conclusion to be drawn about a state of the electrode geometry of the first electrode and / or the second electrode, characterized by - determining an evaluation characteristic (Y1, Y2) which characterizes the influence of one of the at least two current waveforms (W1, W2) on the electrode geometry of the first electrode and / or second electrode when driving the discharge lamp with one of the at least two current waveforms (W1, W2), - storing the evaluation index (Y1, Y2) determined for one of the at least two current waveforms (W1, W2) for the associated current waveform of the at least two current waveforms (W1, W2), and - selecting one of the at least two current waveforms (W1, W2) in a first operating interval (22a) using the evaluation characteristic (Y1, Y2) stored for this purpose for the control in a second operating interval (22b) following the first operating interval.

Images