Background of the Invention
This invention relates to a circuit arrangement for operating a high-pressure discharge lamp which supplies a luminous flux of at least 60% of the nominal luminous flux within 1 s after ignition. The circuit arrangement comprises
a static converter for generating a DC voltage,
a commutator provided with lamp connection terminals,
means A for current stabilization during stable lamp operation by controlling the static converter, and
means B for operating the lamp with an increased run-up current and an increased power as compared with nominal or normal rated lamp operation and on the basis of an instantaneous lamp voltage signal.
A circuit arrangement of the kind mentioned in the opening paragraph is known, inter alia, from EP-A-443795.
Important types of high-pressure discharge lamps are high-pressure sodium lamps, high-pressure mercury lamps and metal halide lamps. A comparatively high luminous flux immediately after switching-on of the lamp is required, especially when a high-pressure discharge lamp serves as a motorcar headlamp. Metal halide lamps whose discharge vessels contain, inter alia, Xe are particularly suitable for motorcar headlamp applications.
A disadvantage of the known arrangement is that the luminous flux of the lamp after ignition in the case of a cold start exhibits a considerable fluctuation during the run-up of the lamp to a stable operating condition. This fluctuation may amount to more than half of the nominal rated luminous flux under certain circumstances. The occurrence of such a major luminous flux fluctuation is unpleasant both for the user and for the other road users, and may even lead to hazardous traffic situations.
The nominal or rated lamp voltage often lies between 60 V and 110 V in practice. This depends to a high degree on the lamp manufacturing technology.
A reduction of the nominal rated lamp voltage, all other circumstances remaining the same, will lead to an increase in the nominal current, which results in a rise in ohmic losses.
The choice of a higher lamp voltage, all other circumstances remaining the same, in general leads to a greater electrode spacing. This is undesirable particularly for reflector lamps, as in the case of motorcar headlamp systems, because the formation of the illumination beam depends to a high degree on the dimension of the light source, i.e. the discharge path between the electrodes of the discharge lamp.
Summary of the Invention
It is an object of the invention to restrict the occurrence of a fluctuation in the lamp luminous flux to a considerable degree, while maintaining a lumen output of 60% of the nominal value one second after ignition of a cold lamp. The nominal value of a quantity means the rated value thereof.
A circuit arrangement according to the invention is for this purpose characterized in that the circuit arrangement comprises a transition means C such that the operation at the increased power comprises at least a period for transition to the power for normal lamp operation. During this period the power has a falling characteristic with a time constant that is independent of lamp voltage. It was surprisingly found to be possible to counteract luminous flux fluctuations to a considerable degree by having the final portion of the run-up take place by means of a signal having a falling characteristic with a time constant that is independent of lamp voltage. The choice of the moment may be determined, for example, empirically. Preferably, this moment is determined by the circumstance that the instantaneous lamp voltage reaches a preset limit value.
It was experimentally found that the achievement of a lamp voltage of approximately half the nominal rated value is a suitable limit value. It is preferable, accordingly, to cause the increased power to fall in accordance with the falling characteristic with a time constant independent of lamp voltage from the moment this level of lamp voltage is reached. Preferably, the means C comprise a timer circuit for realising a fixed time constant as the lamp voltage independent time constant. Further improvement is possible in that the means C are provided with two timer circuits with mutually differing fixed time constants.
Brief Description of the Drawings
These and other aspects of the invention will be explained in more detail with reference to the accompanying drawing in which:
FIG. 1 is a diagram of a circuit arrangement according to the invention,
FIG. 2a shows a portion of the circuit arrangement of FIG. 1 in more detail,
FIGS. 2b and 2c show modifications of a portion of the circuit arrangement of FIG. 2a,
FIG. 3 is a graph showing the lamp current and lamp voltage as a function of time for a lamp operated with the circuit arrangement of FIG. 1, and
FIG. 4 is a graph showing the luminous flux as a function of time accompanying the lamp current and lamp voltage of FIG. 3.
Description of the Preferred Embodiments
In FIG. 1, reference numerals 1 and 2 denote terminals for connecting the circuit arrangement to a supply source. In the embodiment shown, a battery 3 acts as the supply source. A static converter 4 is shown, which serves to generate a DC voltage which is subsequently applied to a commutator 5. The commutator 5 is provided with lamp connection terminals 6, 7 between which a high-pressure discharge lamp 8 is connected.
The circuit arrangement is also provided with means A, indicated with I in the Figure, for current stabilization during stable lamp operation by control of the static converter. II denotes means B for operating the lamp at an increased current and increased power on the basis of an instantaneous lamp voltage signal. Finally, III denotes means C which ensure that during the run-up period of the lamp the operation at increased power comprises at least a portion of the run-up period for connection to and transition to nominal operation. During this portion of the run-up period, the means C derives a signal with a falling characteristic and a time constant which is independent of lamp voltage.
A portion of the diagram comprising the means A, B and C of a practical embodiment is depicted in more detail in FIG. 2a. Vla and Ila denote signal connection points of the means A, further denoted I, which receive an instantaneous lamp voltage signal and an instantaneous lamp current signal, respectively. A signal representing an instantaneous lamp power is derived from these signals by means of a combination circuit 10 and subsequently compared with a reference signal V.sub.ref at 11, the outcome serving as a control signal S for controlling the static converter.
The signal representing the instantaneous lamp power may be formed in that the signals Vla and Ila are multiplied. This provides an accurate representation of the power, however, it does require a comparatively complicated circuit.
A fair approximation of the lamp power is obtained by summation of the signals Vla and Ila. Such an approach has been found to be very suitable in practice for lamp stabilization control and can be realised, for example, by means of a simple resistance network.
In order for the connected high-pressure discharge lamp to supply a luminous flux of at least 60% of the nominal luminous flux after 1 s, facilities are provided for operating the lamp temporarily with an increased current, called run-up current, and an increased power compared with nominal operation. For this purpose, the lamp voltage signal Vla in the embodiment described is also connected via an impedance 21 to a comparison circuit 22 in which the instantaneous lamp voltage signal is compared with a reference value Vla.sub.ref. The outcome of this comparison is subsequently subtracted from the signal representing the instantaneous lamp power formed by the combination circuit 10 by means of a summation circuit 23.
The apparatus II is also provided with means D for the automatic adaptation of the run-up of the lamp in the case of ignition after a short interruption. For this purpose, II is provided with a capacitor 24 which, together with the ohmic impedance 21, acts as a low-pass filter during the lamp run-up after ignition in the cold state. The voltage across the capacitor 24, accordingly, accurately follows the change in the lamp voltage signal V.sub.la during the run-up.
The capacitor 24 also forms part of a discharge circuit 200 in which the discharge time of the capacitor 24 is a measure of the degree of cooling-down of the connected lamp after the lamp supply has been disconnected.
When the lamp is restarted after a short interruption, on the one hand the voltage across the capacitor 24 will have fallen, while on the other hand the lamp voltage signal V.sub.la abruptly rises to a comparatively high value. As a result of the low-pass filter characteristic of the combination of ohmic impedance 21 and capacitor 24, the lamp voltage signal V.sub.la causes a comparatively large charging current to flow through the ohmic resistance 21 to the capacitor 24. This means that a reduced voltage as compared with the instantaneous value of the lamp voltage signal V.sub.la will be applied to the input of the comparison circuit 22. It is achieved by this that the run-up function is automatically corrected relative to a run-up from the cold state after re-ignition in the case of a short interruption of the voltage supply to the lamp. The signal coming from the comparison circuit 22 is subtracted from the power signal at the circuit 23, after which this power signal is compared with a reference value V.sub.ref.
According to the invention, means C, indicted with III in the drawing, are also provided such that the operation at the increased power comprises a portion of the run-up period for transition to nominal or normal operation. This portion has a falling characteristic signal with a fixed time constant i.e. a time constant that is independent of the lamp voltage. For this purpose, a timer circuit is included in the embodiment described, formed by resistors 30, 31 and a capacitor 32, which timer circuit is so connected in the circuit that the signal coming from the comparison circuit 22 is blocked the moment it has fallen to a preset threshold value. The timer circuit then supplies a signal which is a function of the discharging current of the relevant capacitor 32 only. It is achieved by this that the control signal applied to the converter (FIG. 1) has a falling characteristic with a fixed time constant. To ensure that the signal determined by the timer circuit is carried during the run-up of the lamp only, i.e. those conditions when considerable light fluctuations are in danger of occurring, charging of the relevant capacitor 32 takes place through the signal coming from the comparison circuit 22. To limit the voltage across capacitor 32, and thus the maximum duration of that portion of the run-up period which produces the signal with the falling characteristic having the fixed time constant, the capacitive means 32 are shunted by zener diode means 33.
In a modification of the embodiment described, a further improvement is realised in that III is provided with two timer circuits with mutually differing fixed time constants. This is shown in FIG. 2b. Compared with the timer circuit shown in FIG. 2a, the resistor 30 is here replaced by a resistor network formed by resistors 301 and 302 and a diode 303. The one timer circuit with the first fixed time constant is formed by the discharging circuit 32, 31,301,303. When the voltage at point 304 has fallen so far that the voltage across diode 303 is equal to the voltage drop caused by its internal impedance, the second timer circuit with the second fixed time constant becomes active, which timer circuit is formed by the discharging circuit 32, 31,301,302.
Another modification of the embodiment described is shown in FIG. 2c and comprises, instead of a zener diode 33, a transistor circuit 330 in series with a diode 331 for reducing losses in the circuit arrangement.
Time averaging also takes place in the comparison circuit 22 in the embodiment described for achieving stability of the control circuit. Time averaging also takes place in the comparison step at 11 for forming the control signal S for controlling the static converter.
It is favourable for discharge lamp life to operate the lamp at a negative polarity relative to ground. To form the signal representing the power, a polarity change of Vla is provided in this connection in the embodiment described.
A practical realisation of the circuit arrangement as described with reference to the drawing was used for operating a 35 W high-pressure discharge lamp which contains, in addition to Hg, also Xe and metal halides. The nominal lamp voltage is approximately 90 V and the nominal lamp current approximately 0.4 A. A 12 V battery served as the supply source.
FIG. 3 shows the run-up of the lamp voltage (curve 100) and lamp current (curve 101) upon ignition from the cold state.
FIG. 4 shows the measured luminous flux (curve 102) on the same time base as used in FIG. 3. For comparison, a curve 103 in FIG. 4 shows the measured luminous flux in the case in which the circuit arrangement is not provided with means C.
The transition to the falling characteristic with fixed time constant takes place at the moment X.
A comparison of the luminous flux indicated by curve 102 with that indicated by curve 103 makes it clear that after the moment X the luminous flux exhibits a much smaller fluctuation when the lamp is operated with the circuit arrangement according to the invention than when the lamp is operated with a circuit arrangement according to the present state of the art. Curves 200 and 201 in FIG. 4 indicate the maximum and minimum values, respectively, for the luminous flux as allowed by the SAE standard.