Voltage supply for a control unit of an emergency lighting circuit

0001<b>The present invention relates to circuits for operating light-emitting diodes.</b>

0002For example, it is known from the prior art that the patent application <patcit id="pcit0001" dnum="US6502044B1"><text>US6502044 B1</text></patcit> discloses a circuit for self-diagnostic purposes of emergency lighting units, which comprises a battery to supply a lamp in emergency operation. Furthermore, the above-mentioned application also describes a microprocessor that is used for self-testing and diagnostic purposes.

0003The <nplcit id="ncit0001" npl-type="s"><text>IEEE document by Alonso et al. "A Microcontroller-Based Emergency Ballast for Fluorescent Lamps</text></nplcit>"describes a lamp control, which includes a battery charger, a lamp ballast for a fluorescent lamp and a microcontroller. The aim of this" paper "is to present a circuit design that ensures performance and durability with the smallest possible dimensions and light weight. However, the power supply of the microcontroller is independent designed by the power supply of the emergency lighting.

0004<patcit id="pcit0002" dnum="EP1202428A1"><text>EP1202428 A1</text></patcit> describes an emergency lighting unit that includes a battery to power a lamp in emergency operation. The lamp is primarily a gas discharge lamp. The supply current supplied by the battery is kept constant by a non-monolithic control circuit.

0005Even if in the present description the invention is described with reference to emergency lighting devices with LEDs, it should be understood that the invention relates generally to LED operating circuits.

0006Accordingly, emergency lighting devices have as their central element an energy storage unit, in particular a battery or an accumulator, which is charged by the general mains supply voltage during normal or charging operation. For this purpose, a charging circuit is provided which is connected on the input side to the mains supply voltage and, during the charging operation, continuously supplies the energy storage unit with energy, which stores it. Only in the event of an emergency occurring, which is usually automatically recognized by such devices by monitoring the mains supply voltage, is it switched to an emergency lighting mode in which the light source is activated and operated, for which purpose - if necessary - the one provided by the energy storage unit Energy is used. Since the storage capacity of the energy storage unit is of course limited, light sources are preferably used which consume relatively little energy. Accordingly, such emergency lighting devices are preferably equipped with gas discharge lamps, in particular fluorescent tubes. However, light sources in the form of light-emitting semiconductors, in particular LEDs, are also being used increasingly, since these light sources also have a high degree of efficiency and can accordingly be used in an energy-saving manner.

0007The present invention is based on the object of specifying a circuit with an improved voltage supply to a control unit of an emergency lighting device.

0008A first aspect of the invention relates to an emergency lighting device, comprising:<ul id="ul0001" list-style="dash" compact="compact"><li>a battery,</li><li>a charging circuit for the battery supplied with mains voltage,</li><li>a driver circuit for a light source in the form of a switching regulator with a switch which is controlled on the basis of a digital integrated control unit, the voltage supply to the control unit being based on the battery,</li></ul>comprising a battery monitoring circuit which is independent of the control unit and which switches the control unit off when the battery voltage drops below a predetermined value.

0009The control unit is preferably only reactivated when the battery voltage again exceeds the predetermined threshold value.

0010The device can have a DC / DC converter for supplying power to the control unit starting from the battery.

0011The step-up ratio of the DC / DC converter can be adjusted.

0012The step-up ratio can be increased as the battery voltage drops.

0013Another aspect of the invention relates to an emergency lighting device, comprising:<ul id="ul0002" list-style="dash" compact="compact"><li>a battery,</li><li>a charging circuit for the battery supplied with mains voltage,</li><li>a driver circuit for a light source in the form of a switching regulator with a switch which is controlled on the basis of a digital integrated control unit, comprising a battery monitoring circuit connected to the control unit,</li></ul>where, when the battery voltage drops below a predetermined value, all circuit parts with the exception of the control unit are switched off and the control unit goes into a stand-by state.

0014The control unit can be provided with a digital bus interface and can monitor and process incoming digital signals in the standby state.

0015The invention will be explained in more detail below with reference to the accompanying drawing. Show it:<ul id="ul0003" list-style="none"><li><figref idref="f0001">Fig. 1</figref> schematically the circuit diagram of a first example of an emergency lighting device;</li><li><figref idref="f0001">Fig. 2</figref> a second example of an emergency lighting device;</li><li><figref idref="f0001">Fig. 3</figref> a graphic for determining the mains supply voltage on the basis of operating parameters measured on the output side of the charging circuit,</li><li><figref idref="f0001">Fig. 4</figref> another graphic for determining the secondary power of the charging circuit, which is taken into account for the indirect determination of the light source power,</li><li><figref idref="f0002">Fig. 5</figref> an illustration of a battery circuit,</li><li><figref idref="f0003">Fig. 6</figref> an LED driver circuit,</li><li><figref idref="f0004">Fig. 7</figref> the control of the driver switch and the resulting diode current, and</li><li><figref idref="f0005">Fig. 8</figref> Details of the power supply of a microcontroller.</li></ul>

0016This in <figref idref="f0001">Fig. 1</figref> Emergency light device, shown in a simplified manner and generally provided with the reference number 1, is provided in the example shown for operating an LED as an emergency light source. The emergency lighting device 1 is connected on the input side to a power supply network, which has a power supply voltage U_in provides and has as essential components a control unit 2, a charging circuit 3, an energy storage unit 4 in the form of a battery or an accumulator and a driver circuit 5.

0017The charging circuit 3 is formed in the first example shown by a so-called flyback converter, which on the one hand has a transformer T with a primary winding n1 and a secondary winding n2 and on the other hand a controllable switch S1. In a known manner, a corresponding alternating opening and closing of the switch S1, the from the mains supply voltage U_in The energy made available is transferred to the secondary side of the flyback converter 3 and used to charge the energy storage unit 4. The energy transfer takes place in the open state of the switch S1, with a diode D on the output side of the flyback converter 3₁ is provided. Such flyback circuits are widely used in such emergency lighting devices due to their simple structure and reliable function.

0018The controllable switch S1 is activated by the control unit 2 of the emergency lighting device, the activation being carried out, in particular, in an electrically isolated manner via an optocoupler 6. The control unit 2 controls the switch S1 alternately, the so-called duty cycle D1 for the switching operation of the switch S1 being calculated as follows:<maths id="math0001"><math display="block"><mi>D</mi><mo>⁢</mo><mn>1</mn><mo>=</mo><msub><mi>t</mi><mrow><mi mathvariant="italic">on</mi><mo>⁢</mo><mn>1</mn></mrow></msub><mo>/</mo><mfenced><mi>T</mi><mo>-</mo><msub><mi>t</mi><mrow><mi mathvariant="italic">on</mi><mo>⁢</mo><mn>1</mn></mrow></msub></mfenced></math><img file="EP2249459B1_D0001.tif" /></maths>t_on1 corresponds to the switch-on time, while T denotes the total duration of a complete switching cycle for switch S1.

0019It is also conceivable that the control of S1 is a 'cantilever' and that the natural frequency of the 'cantilever' is influenced by the optocoupler 6 via the duty cycle.

0020During a charging operation of the emergency lighting device 1, usually only the charging circuit 3 is active in order to charge the battery 4 permanently. Only in the event that there is an emergency which is caused in particular by deviations in the mains supply voltage U_in is characterized by predetermined target values, an emergency lighting operation is initiated in which the driver circuit 5 is used to control the LED. For this purpose, the driver circuit 5 designed as a switching regulator has a further controllable switch S2, an inductance L and a diode D.₂ on. By alternately actuating the switch S2 by the control unit 2, a current is made available to the LED, via which the LED is operated. The duty cycle with which the switch S2 is controlled by the control unit 2 can be varied in order to set the level of the current supplied to the LED and thus the power at which the LED is operated. In this way it can be ensured in a very elegant way that the LED is operated with constant brightness despite everything, even when the battery power fluctuates.

0021A function of the emergency lighting device 1 thus consists in evaluating the mains supply voltage U_in to recognize whether there is an emergency, in order to initiate emergency lighting operation if necessary. Up to now it was known to directly determine the value of the input voltage U_in to determine for the charging circuit 3, which, however, is associated with disadvantages for the reasons mentioned above.

0022A direct measurement of the mains supply voltage U_in to be dispensed with. Instead, it is planned to determine these indirectly. In particular, it is provided that only the sizes of operating parameters of the emergency lighting device 1 are measured on the secondary side of the charging circuit 3.

0023According to the first example <figref idref="f0001">Fig. 1</figref> the voltage U present on the secondary side of the flyback converter or flyback converter 3_flb2 measured, for which in particular no electrical isolation is required, since this variable is at the same reference potential as the control unit 2, which evaluates the measured value. Now is the level of this secondary voltage U_flb2 known, it can be based on the level of the input voltage U_in be inferred. When switch S1 of flyback converter 3 is switched on, there is a connection between input voltage U_in and secondary voltage U_flb2, which is dependent in particular on the winding ratio between the two windings n1 and n2 of the transformer T and on the duty cycle of the switch D1. This relationship between the easy-to-measure secondary voltage U_flb2 and the input voltage U to be monitored_in is now stored in the form of a table of values ​​in the control unit 2, so that after measuring the secondary voltage U_flb2 in a simple manner the level of the input voltage U_in can determine without having to measure them directly. Now the control unit 2 determines that the input voltage U_in is outside certain target value ranges, this indicates an emergency state, which in turn will cause the control unit 2 to initiate an emergency operation.

0024The solution described thus enables very simple but effective monitoring of the status of the general power supply. There is also a particular advantage of the in<figref idref="f0001">Fig. 1</figref> example shown in that the level of the input voltage U_in regardless of whether the emergency light LED is switched on or not. The diode D₁ causes a separation between the secondary voltage U by its blocking effect_flb2 and battery voltage U_Bat, so that the activity of the driver circuit 3 is not based on the above-described process of determining the input voltage U_in affects.

0025A second, somewhat more general example of an emergency lighting device is in <figref idref="f0001">Fig. 2</figref> shown. The structure of this corresponds essentially to that in<figref idref="f0001">Fig. 1</figref> shown emergency lighting device 1, but now the charging circuit 3 is not formed by a flyback converter but generally by a circuit arrangement which has a potential separation and a switch S1 controlled by the control unit 2.

0026In this more general form, there is not necessarily a known relationship between the input voltage U_in and the voltage on the output side of the charging circuit 3. Nevertheless, the level of the input voltage U_in can be determined indirectly, but now other farm sizes are measured.

0027On the one hand, this is the battery voltage U_asked as well as the battery current I_asked. Both variables can in turn be determined relatively easily, that is to say without galvanic isolation, since they, like the control unit 2, which evaluates these measured variables, are at the same reference potential.

0028On the basis of these two measured variables U_asked and I_asked and the known duty cycle D1 for the switch S1 of the charging circuit 3 can then the input voltage U_in can be determined, since this is related to the three known quantities. This illustrates the graphic of<figref idref="f0001">Fig. 3</figref>which different characteristics of the secondary power P_flb2 shows the charging circuit 3 as a function of the duty cycle D1 for the switch S1. These characteristic curves are determined, for example, during the manufacture of the emergency lighting device and are in turn stored in the form of a table in the control unit 2. It can be seen that these characteristics in particular also depend on the input voltage U_in are dependent. So now the duty cycle D1 and the secondary power P_flb2 the charging circuit 3 known, as in the example of <figref idref="f0001">Fig. 1</figref> to the level of the input voltage U_in be closed back.

0029In the example shown by <figref idref="f0001">Fig. 3</figref> it is thus determined with which characteristic curve the known combination of duty cycle D1 and secondary power P_flb2 the charging circuit 3 which, when the driver circuit 5 is deactivated, the product of the battery voltage U_asked and battery power I_asked corresponds, agrees. In the measured values ​​shown, for example, this value lies on the characteristic curve for an input voltage U_in of 220 volts, which corresponds to a proper condition of the general mains supply. However, the determined value would lie on a characteristic curve, for example an input voltage U_in corresponds to 140 volts or 280 volts, this would in turn be interpreted by the control unit 2 in such a way that there is an error in the mains supply and accordingly initiate an emergency.

0030In both of the examples shown, the input voltage U can be measured reliably and without the need for a direct measurement_in determine whether the mains supply is OK or not. A limitation in the example of<figref idref="f0001">Fig. 2</figref> consists, however, in that the described determination of the input voltage U_in only when switched off

0031Driver circuit 5 is possible. In the example of<figref idref="f0001">Fig. 1</figref> however, as already mentioned, this restriction does not exist. Basically, however, the control unit 2 will activate the driver circuit 5 and accordingly switch the LED on if a fault is detected.

0032After the driver circuit 5 has been activated, the switch S2 can then be controlled at high frequency in the manner described above in order to operate the LED with a desired power. In this case, to ensure that the power of the LED is constant, it would be necessary to use both the LED voltage U_led as well as the LED current I_led to know to enable regulation.

0033Referring to <figref idref="f0002">Figure 5</figref> Details of the connection of the battery 4 will now be explained.

0034As in <figref idref="f0002">Figure 5</figref> can be seen, an element 12 is connected in series with the battery 4, which can have the function of a linear regulator and / or a switch. For example, this element 12 can be a transistor. A measuring resistor (shunt) 16 is also connected in series with the battery 4, so that the voltage drop across the shunt 16 is representative of the battery current.

0035The measurement signal tapped at the shunt 16 is fed to a current detection unit 13, which is preferably constructed as a discrete circuit and can have a comparator 14. The comparator 14 is only one example of how an offset to the measurement signal can be applied by the shunt 16. The application of the offset serves to be able to evaluate signals with different polarities that represent the battery current in a simplified manner by selecting the offset in such a way that the signal levels are shifted in such a way that both signal polarities now have the same polarity and therefore different amplitudes. Thus, both the battery charging current and the battery discharge current, which are known to have different polarities, can be measured in a relatively simple manner, for example by the control circuit 2. The control circuit 2 is thus preferably supplied with a measurement signal 15 with a uniform polarity.

0036If the transistor 12 is in the form of a linear regulator, the battery discharge current and / or the battery charging current can be regulated to a predetermined target value by means of the current detection unit 13 and the control of the linear regulator 12. In a simplified manner, this regulation can of course also be implemented as a protective circuit, so that the switch 12 is opened in the event of an excessively high battery charging current or battery discharge current in order to protect the battery 4.

0037In the event of a very long failure of the mains voltage, deep discharge of the battery 4 can occur. If the battery 4 is deeply discharged, it has a voltage of, for example, 1.3 volts, which is below the permissible voltage of, for example, 1.5 volts. Even if there is now a proper mains voltage on the charging circuit 3 on the input side and the charging circuit 3 is operated properly in the manner described above, the deeply discharged battery 4 will pull the secondary side of the charging circuit to an inadmissibly low value.

0038According to the invention, this deep discharge can be detected by detecting the voltage of the battery 4. When such a deep discharge is detected, the switch 12 is preferably operated in a clocked manner. In this case, the switch 12 is preferably closed only for a relatively short period of time, during which the battery 4 is charged. Thereafter, the switch 12 is opened again for a longer period of time, so that the battery 4 is disconnected from the secondary side of the charging circuit 3 and the charging circuit 3 can again provide the proper voltage of, for example, 1.5 volts on the secondary side. This means that there is a proper tension ratio on the secondary side for a much longer period of time.

0039There is therefore a pulsed charge of the deeply discharged battery.

0040While the switch 12 is open, the mains voltage that is present again supplies the connected LED driver circuit and LEDs correctly by means of the charging circuit. On the other hand, if the switch is briefly closed, the battery is gently recharged. For example, the pulse duty factor for the switch 12 can be chosen such that it is only closed for 10% of the total time period and correspondingly open for 90%, so that the battery can recover in this 90% time period.

0041When deep discharge of the battery is detected by means of the detection of the battery voltage, switch 12 can be automatically switched over in this pulsed operation. The battery voltage is preferably monitored via a discrete circuit and thus independently of the microcontroller 2 (see<figref idref="f0001">Figures 1, 2</figref>), which may not be fully functional if the battery voltage is too low.

0042For the rest, the current detection circuit 13 with the comparator 14 is preferably selected as a discrete circuit and thus independently of the microcontroller 2 and its proper operation.

0043<figref idref="f0003">Figure 6</figref> shows further details of an LED driver circuit 5. All features of the battery circuit of <figref idref="f0002">Figure 5</figref> can also be used with the circuit from <figref idref="f0003">Figure 6</figref> combine.

0044The basic principle of the driver circuit 5, as already described at the outset, is a switching regulator in which the coil L is magnetized when the switch S2 is closed and the magnetic energy is discharged when the switch S2 is reopened via a current path which in series comprises a Zener diode, at least one LED and has an ohmic resistor 17.

0045As is known, it is relatively easy to measure the current flow through the diodes when switch S2 is closed.

0046In contrast, it is somewhat more complex to measure the current flow through the LEDs when switch S2 is open.

0047For current detection, the resistor 17 is connected in series with the LEDs, this resistor 17 representing an example of a current detection means.

0048A current detection unit 18 can have a comparator 19, for example.

0049A current detection signal, ie a signal representing the current through the LEDs, is supplied to the control circuit 2. The control circuit 2 can configure the switching behavior of the switch S2, inter alia, as a function of the current detection signal.

0050A capacitor 30 is connected in parallel to the at least one LED. The capacitor 30 smoothes the LED current by storing energy and keeping the voltage across the at least one LED substantially constant. The capacitor 30 is dimensioned such that the regulation of the LED voltage is still possible.

0051<figref idref="f0004">Figure 7</figref> shows schematically the activation of the switch S2, that is to say, in the event that this switch is designed as a FET transistor, the activation of the gate of this transistor in conjunction with the resulting current through the LEDs. During the switch-on period of switch S2, the current through the LEDs increases in each case. During the switch-off period, it drops again, driven by the magnetic energy of the coil L. This results in a zigzag shape of the current through the LED around a fixed DC value. The current detection should now be able to determine, in particular, the mean value of this current over time, so that, for example, the power of the LEDs can be selected either to a constant value or to a freely selectable value (dimming).

0052It is provided that the control circuit 2, which specifies and thus knows the switching behavior of the switch S2, determines the current value at a first measuring point A and at a second measuring point B, in order to in turn form the mean value therefrom. The mean value of the LED current is then obtained from the mean value of the current value at the time of measurement A and time B of measurement.

0053The measuring point A is chosen such that it lies in the area of ​​the end of the switch S2 being switched on, while the time B is selected such that it lies in the area of ​​the end of the switch S2 being switched off.

0054A measurement of a plurality of minimum and maximum values ​​is preferably carried out for one measurement cycle. Inaccuracies regarding the time of measurement can thus be determined. For this measuring principle, the measuring times are synchronized with the timing of switch S2. However, this synchronization can lead to inaccuracies which are caused, for example, by the delays of A / D converters. These inaccuracies are averaged out by recording several values.

0055Thus, the temporal mean value of the current through the LED can be detected, and the control unit 2 can, depending on this, switch frequency and / or duty cycle ratio of the switch S2. For dimming the LED, the control unit 2 can, for example, superimpose a low-frequency modulation, for example in the form of a PWM modulation, on the high-frequency clocking of the switch S2. This represents an alternative or additional possibility for dimming, since, as stated above, dimming can also take place by changing the high-frequency clocking of switch S2 itself.

0056In the meantime, the dimming of the LED, ie the setting of the LED current to a defined, selectable value, can also take place depending on other variables. For example, the dimming can be dependent on a detected battery discharge current (see circuit of<figref idref="f0002">Figure 5</figref>) in order to extend the battery life.

0057The driver circuit 5 can also dim the LED if the mean value of the mains voltage which supplies the charging circuit 3 drops below a predetermined value. As a further assistive measure for extending the battery life, the flyback converter in the charging circuit 3 can be set to a pulsed operation, so that the flyback converter supplies the driver circuit 5 in a manner that assists the weakening battery voltage.

0058In order to detect a discharge of the battery voltage to battery 4, the battery voltage can be recorded (see again <figref idref="f0002">Figure 5</figref>).

0059In addition to the LED current detection, an LED voltage detection can also be carried out.

0060A soft start function can be provided, in which the control circuit 2 first allows a relatively low test current to flow through the LED (s) when it is started up. The test measurement is intended to determine electrical parameters in order to find out whether one or which LED is connected to the driver circuit 5. For example, it can be determined which LED voltage results from the specified test current. This allows conclusions to be drawn about the type (color) and number of connected LEDs, in order to regulate the steady-state operating current based on the determined LED type.

0061In principle, it can be provided that the control unit 2 increases the LED current during start-up by appropriately activating the switch S2 of the driver circuit 5, starting from a low value towards the steady-state operating value for the LED current.

0062If the application of the test current results in the absence or malfunction of an LED, the application of the test voltage can be repeated automatically at predetermined intervals until a correct LED is detected.

0063If the application of the test voltage results in the absence or malfunction of an LED, the application of the test voltage can be repeated automatically at predetermined intervals until a correct LED is detected.

0064As said, the current through the LED can either be achieved by superimposing a low-frequency modulation on the high-frequency clocking of the switch S2 and / or by changing (duty cycle, frequency) the high-frequency clocking itself.

0065Referring to <figref idref="f0005">Figure 8</figref> Aspects of the invention with respect to the voltage supply V_s and the correct function of the control unit 2, which, as is known, can be designed as a digital integrated circuit, in particular a software-controlled circuit such as a microcontroller. If the microcontroller 2 does not function properly, it can happen that the microcontroller 2 switches the switch S2 of the LED driver circuit 5 permanently into the closed (conductive) state.

0066A discrete error detection circuit 20 is now provided, which detects this state. For example, this discrete error detection circuit 20 can have a capacitor which integrates the gate drive signal of the switch S2 and thus detects the state of the permanent switching on of the LED driver switch S2. If the switch-on signal is applied permanently (logically high), the voltage at the capacitor of the error detection circuit 20 rises, which could discharge, for example, via an ohmic resistor in the case of a logic 'low' signal. The error detection circuit 20 is connected to the microcontroller 2, ie with a permanently present switch-on signal at the LED driver circuit S2, a signal builds up on the capacitor of the error detection circuit 20, which signal is led to a pin of the microcontroller 2 provided for this purpose. If this signal is applied to the pin of the microcontroller 2, it is provided that the current operating data of the microcontroller 2 are stored in a memory 21. The microcontroller 2 then carries out a restart after a reset.

0067A comparable process can be triggered if a detection circuit for the battery voltage 4 determines that the supply voltage of the microcontroller 2, for example, drops significantly below the setpoint of 1.5 volts. In this case too, the operating data from the microcontroller 2 are stored in the memory 21. The microcontroller 2 is of course only restarted when a sufficient supply voltage for the microcontroller 2 has been detected.

0068According to a further aspect of the invention, the supply voltage Vs for the microcontroller 2 is provided by means of a switching power supply (DC / DC converter) 22. The DC / DC converter 22 can be designed in the form of a so-called step-up converter (buck / boost converter) in such a way that the step-up ratio is increased when the battery voltage is low (for example in the event of a deep discharge), and thus a sufficient supply voltage for the microcontroller 2 is still available is provided.

0069According to a further aspect of the invention, the microcontroller 2, if the battery voltage also drops after a prolonged failure of the mains supply voltage, switches off all circuit parts that are not purely necessary to supply the microcontroller 2. The microcontroller 2 puts itself in a standby state. In this standby state, the interface of the microcontroller 2 continues to function, which in particular monitors incoming bus signals from the bus line 11 and thus from an external control unit 10. The microcontroller 2 can thus process an incoming digital signal (for example a DALI signal) or comparable signals even in this standby state.

0070Otherwise, the operational amplifier 14 of the current control 13 of the battery 4 can also be deactivated when the battery is very weak.