Abstract
A self-healing overtemp circuit is described and illustrated comprising a temperature sensing circuit, a voltage sensing circuit, and optionally, a current sensing circuit. A lower cost, simplified alternative overtemp circuit is also discussed. The self-healing overtemp circuit is designed to ramp down power in an LED lighting system (or other electrical circuit) in response to a sensed or impending thermal runaway (and optionally, overcurrent) event. Said thermal runaway and overcurrent events may be a result of failure of one or more components (e.g., driver, active cooling means) of the lighting system. The self-healing overtemp circuit further comprises means of restoring power to said LEDs in a manner that avoids (i) a perceivably bright flash of light or (ii) increased risk of component failure.
Claims
1. A method of reducing and reestablishing power to an array of LED lighting fixtures powered by one or more drivers comprising: a. detecting a first temperature at one LED lighting fixture of the array of LED lighting fixtures; b. reducing power to the array of LED lighting fixtures upon detection of said first temperature; c. reestablishing power to the array of LED lighting fixtures upon detection of a second temperature, wherein said second temperature is lower than said first temperature.
2. The method of claim 1 wherein said first temperature is selectable, and wherein the selection of the first temperature is based, at least in part, on one or more of: a. ambient temperature; b. characteristics of the LEDs; and c. characteristics of the LED drivers.
3. The method of claim 1 wherein the step of reducing power to the array of LED lighting fixtures comprises dimming of the LED drivers.
4. The method of claim 1 wherein the step of reducing power to the array of LED lighting fixtures comprises reducing duty cycle of the LED drivers.
5. The method of claim 1 further comprising a bi-metallic switch having a closed state and an open state, and wherein the step of detecting a first temperature comprises detecting a change in the state of the bi-metallic switch.
6. The method of claim 5 wherein the step of detecting a second temperature comprises detecting a change in the state of the bi-metallic switch.
7. The method of claim 6 wherein each LED lighting fixture of the array of LED lighting fixtures is associated with a bi-metallic switch, and wherein the step of detecting a first temperature by detecting a change in the state of the bi-metallic switch comprises detecting a change in the state of any of the bi-metallic switches associated with the array of LED lighting fixtures.
8. The method of claim 7 wherein the step of detecting a second temperature by detecting a closing of the bi-metallic switch comprises detecting a closing of any of the bi-metallic switches associated with the array of LED lighting fixtures.
9. An apparatus for reducing and reestablishing power to an array of LED lighting fixtures powered by one or more drivers comprising: a. a bi-metallic strip which opens at a first temperature and closes at a second temperature; b. a detection circuit to detect whether the bi-metallic strip is open or closed; c. a control circuit to effectuate power reduction in said one or more drivers; and d. a latching circuit to effectuate power reestablishment according to one or more predetermined conditions.
10. The apparatus of claim 9 further comprising additional bi-metallic strips.
11. The apparatus of claim 10 wherein the bi-metallic strips are daisy chained across the array of LED lighting fixtures.
12. The apparatus of claim 11 wherein the detection circuit detects whether any of the bi-metallic strips is open or closed.
13. The apparatus of claim 9 wherein the detection circuit is electrically isolated from the control circuit.
14. The apparatus of claim 9 wherein the one or more predetermined conditions comprises: a. latching configuration; and b. last driver setting.
15. The apparatus of claim 9 further comprising a driver controller, and wherein power reduction comprises dimming of the LED drivers via said controller.
16. The apparatus of claim 9 further comprising a driver controller, and wherein power reduction comprises reduction of a duty cycle of the LED drivers via said controller.
17. A method of cycling power to an array of LED lighting fixtures powered by one or more drivers comprising: a. detecting a first temperature at one LED lighting fixture of the array of LED lighting fixtures; b. reducing power to the array of LED lighting fixtures upon detection of said first temperature; c. reestablishing power to the array of LED lighting fixtures upon detection of a second temperature, wherein said second temperature is lower than said first temperature.
18. The method of claim 17 wherein the step of reestablishing power to the array of LED lighting fixtures comprises resetting a power input to the one or more drivers.
19. The method of claim 18 wherein the step of reestablishing power to the array of LED lighting fixtures further comprises returning to a previous driver power setting.
20. The method of claim 18 wherein the step of reestablishing power to the array of LED lighting fixtures further comprises repeating steps a.-c.
Description
III. BRIEF DESCRIPTION OF THE DRAWINGS
(1) From time-to-time in this description reference will be taken to the drawings which are identified by figure number and are summarized below.
(2) FIG. 1 illustrates a typical prior art sports lighting application including generic LED lighting fixtures.
(3) FIG. 2 illustrates the generic lighting system of FIG. 1 modified according to at least some aspects of the present invention.
(4) FIG. 3 illustrates one possible method of practicing the invention according to a first embodiment.
(5) FIG. 4 illustrates, in wiring diagram form, a self-healing overtemp circuit for use in a lighting system such as that illustrated in FIG. 2 in accordance with a method such as that illustrated in FIG. 3. Electrical symbols denoting resistors having 0Ω illustrates jumpers for single-sided boards and electrical symbols denoting “TP” illustrate test points used in troubleshooting.
(6) FIG. 5 illustrates the power supply circuit for the self-healing overtemp circuit according to a first embodiment.
(7) FIG. 6 illustrates the voltage sensing circuit of the self-healing overtemp circuit according to a first embodiment.
(8) FIG. 7 illustrates the temperature sensing circuit of the self-healing overtemp circuit according to a first embodiment.
(9) FIG. 8 illustrates the current sensing circuit of the self-healing overtemp circuit according to a first embodiment.
(10) FIG. 9 illustrates the switching circuit of the self-healing overtemp circuit according to a first embodiment.
(11) FIG. 10 illustrates the controller circuit of the self-healing overtemp circuit according to a first embodiment.
(12) FIGS. 11-13 illustrates, in wiring diagram form, an alternative self-healing overtemp circuit for use in a lighting system such as that illustrated in FIG. 2 in accordance with a method such as that illustrated in FIG. 3. Electrical symbols denoting resistors having 0Ω illustrate (i) jumpers for single-sided boards or (ii) indication of an alternative circuit arrangement on a common board, and electrical symbols denoting “TP” illustrate test points used in troubleshooting. FIG. 11 illustrates the two-part temperature sensing circuit, FIG. 12 illustrates a detailed look of the power supplies of FIG. 11, and FIG. 13 illustrates a detailed look of the control circuit of FIG. 11.
(13) FIG. 14 illustrates the generic lighting system of FIG. 1 modified according to an alternative embodiment of the present invention.
(14) FIGS. 15A and B illustrate generically daisy chaining of lighting fixtures according to an alternative embodiment of the present invention. FIG. 15A illustrates a front view of a generic layout of wiring within pole 10 and associated cross-arm. FIG. 15B illustrates generically daisy chaining in wire diagram form.
IV. DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS
A. Overview
(15) To further an understanding of the present invention, specific exemplary embodiments according to the present invention will be described in detail. Frequent mention will be made in this description to the drawings. Reference numbers will be used to indicate certain parts in the drawings. Unless otherwise stated, the same reference numbers will be used to indicate the same parts throughout the drawings. Likewise, frequent mention will be made to various components and circuits. Reference numbers or letters will be used to indicate certain portions of circuits. For simplicity, all circuit portions are referred to herein as “circuits”, regardless of whether said referenced portion comprises a complete circuit in and of itself. Further, reference is made herein to “user(s)”, “designers”, and/or “operators”. It is to be understood that these terms are used for convenience, and in no way place limitations on who may practice the invention, or benefit from aspects thereof. Further, it should be noted that the terms “luminaire”, “fixture”, and/or “lighting fixture” are used interchangeably herein; these terms are used for convenience, and in no way place limitations on load characteristics, circuit design, or other aspects of the present invention. Lastly, communications between electrical components are referred to herein as “pulses”, “communications”, “signals”, and/or “instructions”. It is be understood that communications could take a variety of forms and that the aforementioned terms—used interchangeably for convenience—in no way limits the invention to a particular communication mode, bandwidth, or the like.
(16) The exemplary embodiments envision an overtemp circuit formed from solid state components designed to disable or reduce power to a load when a threshold indicative of a thermal runaway event is exceeded, or in some cases approached, and then reestablish power at a desired level after the thermal runaway event has been resolved (and/or when temperature is below some threshold and the driver has been cycled). As envisioned, the overtemp circuit is formed from components which permit selectivity of the threshold based, at least in part, on characteristics of the load. In practice, the overtemp circuit could be applied to a number of electrical systems and many different kinds of loads; though the following discusses an LED lighting system, the present invention is not limited to such.
(17) FIG. 1 illustrates a generic LED lighting system in which a pole 10 elevates one or more LED lighting fixtures 1001, said fixtures oriented and aimed so to project light generally towards a target area 20 (e.g., a sports field); lighting fixtures could be designed in accordance with U.S. Pat. No. 8,789,967, or otherwise. Generally, power to the LEDs (i.e., the load) is supplied along power wiring from a distribution source 60 at some voltage and amperage regulated by national, state, or local codes (e.g., 480 VAC, 60 Hz). Said codes may also require equipment grounding, distribution panels, a main power disconnect, or the like; for brevity, no such functionality is illustrated. Power wiring generally enters hollow pole 10, then enclosure 40—where it is regulated by some combination of controllers 50 and drivers 30 (here, one driver per fixture, though this is by way of example and not by way of limitation)—before continuing along pole 10 to said fixtures 1001.
(18) Power from a distribution source 60 is often ill-suited for direct application to a load. Incoming power may be three-phase alternating current (AC) whereas an LED requires direct current (DC). The forward voltage drop across an LED (e.g., any of the XLamp series of LEDs available from Cree, Inc., North Carolina, USA) may only be on the order of 3V, but in large number (e.g., as may be required in the example of FIG. 1) may greatly exceed incoming power. Considerations such as these dictate the type and number of drivers. Likewise, in a typical lighting system such as that illustrated in FIG. 1 there may be a need to alter duty cycle (e.g., to manage temperature), adjust current (e.g., in response to some sensor input), or modify some other aspect of operation to effectuate a lighting change (e.g., provide dimming). Considerations such as these dictate the type and number of controllers. While a particular combination of controller and driver may be such as is described in U.S. patent application Ser. No. 14/831,986 incorporated by reference herein in its entirety, it is to be understood that it is impractical to discuss every configuration of power control in LED lighting systems, and that a thorough understanding of each is not necessary to practice the invention.
(19) Specialty lighting systems such as the sports lighting system of FIG. 1 typically rely on several hundreds of LEDs so to provide adequate illumination of a target area to one or more standards; often this requires running one or more long strings of LEDs in series for each associated driver (e.g., so to ensure all light sources within a single luminaire act in unison and emit uniformly). To accommodate a long string of series-connected LEDs (e.g., several hundreds with a forward voltage drop on the order of 3V each) given distribution power at 480V, 277V, or even 208V (depending on the geographic area, local codes, and the like), a boost-type driver is needed. It is assumed one of ordinary skill in the art of circuit design, and more specifically LED lighting circuit design and/or power supply design, is well aware of how boost-type converter circuits operate, and so for brevity any discussion of such is omitted.
(20) With respect to the aforementioned boost-type LED drivers, it is well known that when the load (e.g., lighting fixture 1001) is removed from the boost-type driver circuit as might be the standard protocol for a thermal runaway event—the MOSFET (or similar switching device) associated with the boost-type driver circuit is, in essence, switched off, thereby acting as a current source to an infinite load (an open circuit) until the output voltage reaches a maximum. What may not be well known is that when the load is returned and the circuit closed, there is an inrush of voltage as energy is supplied to the load; this may be harmful to the LEDs (i.e., the load). Initial experiments have shown that for boost-type drivers designed for a maximum 700V output, an inrush voltage of 800V is actually supplied under these conditions—almost instantaneously. Results have shown a best case scenario to be a disabling bright flash of light and a worst case scenario to be a damaging of parts (e.g., the LEDs themselves).
(21) Presented according to a first embodiment are apparatus, methods, and means for dissipating the excess charge that stores in an LED boost-type driver when the circuit is opened in response to a thermal runaway or overcurrent event; according to one example, by dimming down to 0% (i.e., reducing driver output to 0V or near 0V). Also presented herein are apparatus, methods, and means for closing the circuit and reestablishing power to the load once both (i) the thermal runaway or overcurrent event has been resolved and (ii) driver output will not damage the LEDs or other components of the lighting system. This is achieved, in one example, by a combination of sensing circuits, logic, and pulse generator. Communications between the drivers and controller at the base of the pole (see, e.g., reference nos. 30 and 50 of FIG. 1, respectively) and envisioned overtemp circuit components at the top of the pole rely, in one example, upon a combination of existing power wiring and the residual potential in the driver at 0% dimming; this ensures no extra wiring is needed to add the self-healing overtemp functionality to the lighting system, though of course, this is by way of example and not by way of limitation.
(22) A more specific exemplary embodiment, utilizing aspects of the generalized example described above, will now be described.
B. Exemplary Method and Apparatus Embodiment 1
(23) FIG. 2 illustrates the sports lighting system of FIG. 1 modified according to aspects of the present invention. As can be seen, the lighting system includes some kind of active (or passive) cooling—in this example, fans 200 which are indicated generically by blocks—located proximate the temperature-sensitive components of the lighting system (e.g., the LEDs). Active cooling could be powered from a battery or line power, could be in accordance with the aforementioned incorporated references (or otherwise), or could even cool other components of the lighting system (e.g., drivers); the exact configuration may vary as is needed for the thermal management demands of the load. Most pertinent to the present invention is simply that a failure or absence of cooling means will cause an undesirable rise in temperature of one or more components of the system (at least under some operating conditions).
(24) FIG. 2 also illustrates an exemplary self-healing overtemp circuit 1000, details of which are illustrated in FIGS. 4-10. As is illustrated in the figures, according to the present embodiment circuit 1000 is installed in the fixture itself; as envisioned, on the same board as the LEDs contained within the fixture. While this does permit highly accurate readings from temperature sensing circuit 1005 which can be used to determine LED junction temperature, it is to be understood some or all of the functionality of self-healing overtemp circuit 1000 could be installed elsewhere in the lighting system; additional options are later discussed.
(25) FIG. 3 illustrates one possible method of practicing the invention with respect to FIG. 2. According to method 500, a thermal runaway event occurs (step 501); overcurrent conditions are later discussed. Step 501 could comprise a failure of passive or active cooling component 200, some kind of weather event, structural failure of one or more components of the lighting system—any number of events or conditions which causes an undesirable temperature increase in the LEDs or other temperature-sensitive parts of the lighting system. A temperature sensing circuit detects an increase in value and in response to approaching or exceeding some limit a controller circuit sends a signal to a switching circuit to open the lighting circuit (step 502). In response to opening the lighting circuit there is a corresponding rise in the voltage output of the driver (step 503). A voltage sensing circuit detects the increase in voltage and in response to approaching or exceeding some limit, a communication circuit sends a signal to the driver to dim down to 0% (step 504). The output of the driver decreases (step 505). At some point the temperature of the system will decrease (step 506). It is preferred if a decrease in measured temperature is the result of the thermal runaway event being resolved, though the invention is not limited to such. Once both the temperature (as indicated by the aforementioned temperature sensing circuit) and the output voltage of the driver (as indicated by a high voltage temperature sensing circuit at the fixture) are below defined thresholds, the controller circuit sends a signal to the switching circuit to close the circuit (step 507). Simultaneously, or after the circuit is closed, the communication circuit sends a signal to the driver to re-establish power to the load. Power is reestablished by sending an instruction to the LED driver to increase dimming to some level (e.g., to 100%, to the dimming level set just before the thermal runaway event), which causes a commensurate increase in light output of the LEDs (step 508). It is of note that only some LED drivers are adapted to receive such instruction directly, whereas other designs of driver may require an associated controller.
(26) Specific details and functionality of self-healing overtemp circuit 1000 is illustrated in FIGS. 4-10. As envisioned, circuit 1000 is formed from solid state, non-terminal event components; this ensures a robust design which requires little to no intervention from a user. Further, and as will be discussed, a number of components of circuit 1000 are selectable either in terms of quantity or value. This selectivity permits a designer significant flexibility in developing the various thresholds and triggers mentioned above in the description of method 500. Additional features and options are later discussed.
(27) FIG. 4 illustrates (in wiring diagram form) self-healing overtemp circuit 1000. As envisioned, overtemp circuit 1000 is installed in fixture 1001—preferably on the same board as the LEDs to ensure more accurate temperature readings—though as will be discussed, the invention is not limited to such. After a thermal runaway event has occurred (step 501, FIG. 3) the temperature of the LEDs is expected to rise. This temperature rise is detected by a temperature sensing circuit 1005, shown in detail in FIG. 7. As can be seen from FIG. 7, temperature sensing circuit 1005 generally comprises an analog temperature sensor, amplifiers, and a number of standard capacitors and resistors. In practice, different configurations of lighting fixtures (i.e., loads) are thermally cycled and three values measured: temperatures at the LED solder point, temperature at the board where temperature sensing circuit 1005 resides, and input voltage. Knowing the solder point temperature, input voltage, and LED characteristics such as thermal resistance (commonly available from the LED manufacturer) one can readily calculate junction temperature of the LEDs using well known formulas. In this manner, a temperature at temperature sensing circuit 1005 (i.e., at the temperature sensor) is correlated to an LED junction temperature. Operationally, a lighting designer can select a high temperature reading which correlates to a high junction temperature to use as a trigger for opening the circuit, and a low temperature reading which correlates to a low junction temperature to use as a trigger for closing the circuit.
(28) Temperature readings from temperature sensing circuit 1005 are fed to controller circuit 1007 of FIG. 10. As can be seen from FIG. 10, controller circuit 1007 generally comprises a microprocessor and a number of standard capacitors and resistors. Controller circuit 1007 acts as the hub of self-healing overtemp circuit 1000—as can be ascertained from the number of inputs and outputs illustrated. Controller circuit 1007, and components of overtemp circuit 1000 on the whole, is powered by a power circuit 1002 (see FIG. 5). As can be seen from FIG. 5, power circuit 1002 comprises a depletion mode MOSFET, amplifier, transistor, fast switching diode, and a number of standard capacitors and resistors. Additional power supply options are later discussed.
(29) When temperature readings from temperature sensing circuit 1005 correlate to a thermal runaway condition (i.e., a high junction temperature) as determined by the settings of controller circuit 1007, controller circuit 1007 sends a signal to switching circuit 1008 of FIG. 9 which, in turn, opens the circuit (step 502, FIG. 3). In essence, the thermal overtemp circuit physically interrupts the path between the driver the fixture. In practice, a temperature reading of 50° C. at the temperature sensor of circuit 1005 might correlate to an LED junction temperature of 80° C. for a particular design of lighting fixture, and may necessitate circuit 1007 sending a control signal to circuit 1008 to open the circuit. As can be seen from FIG. 9, switching circuit 1008 generally comprises a number of MOSFETs (some in enhancement mode), transistors, and a number of standard capacitors and resistors.
(30) Once the circuit is open due to the thermal runaway event, voltage output of the boost-type LED driver climbs (step 503, FIG. 3). Voltage increase is measured by voltage sensing circuit 1004 of FIG. 6. As can be seen from FIG. 6, circuit 1004 generally comprises an amplifier and a number of standard capacitors and resistors. In practice, voltage sensing circuit 1004 is different than temperature sensing circuit 1005 insomuch that the signal is already existing and does not need to be generated (i.e., the output of the driver already exists), requiring only conditioning (see e.g., the string of resistors in FIG. 6); but voltage sensing circuit 1004 is the same as temperature sensing circuit 1005 insomuch that the input provided to controller circuit 1007 triggers an instruction based, at least in part, on characteristics of the load. For example, for some loads a voltage input into controller circuit 1007 correlating to 50V may indicate a deleterious condition, but if the load is different (e.g., number of LEDs in a series-connect string are varied), that same 50V may no longer indicate a deleterious condition.
(31) When voltage readings from voltage sensing circuit 1004 correlate to an excessively high driver output voltage due to open circuit as determined by the settings of controller circuit 1007, controller circuit 1007 sends a communication to the driver to dim to 0% (step 504, FIG. 3). This communication can be sent wirelessly; communication circuit 1006 (FIG. 4) could be completed by mating to a wireless controller board (e.g., Bluetooth Low Energy (BLE) board) at the top of the pole (possibly in the lighting fixture) which could send the communication wirelessly to a complementary Wi-Fi controller board at the driver (or controller for the driver) at the bottom of the pole. This wireless option may be preferable in situations where wireless communications are permitted and cost is not the prevailing concern; two other signal/communication options are later discussed.
(32) In practice, LED driver output voltage will likely decrease to below some threshold—which is defined by controller circuit 1007 (FIG. 10)—before the temperature of the LED fixture decreases below the temperature threshold defined by controller circuit 1007; this is indicated by steps 505 and 506 of method 500, respectively. Again, thresholds may be varied by a designer in accordance with preferences, characteristics of the load, or otherwise. Further, because upper and lower thresholds are defined at controller circuit 1007, the circuit is self healing; self-healing in the sense that no intervention from an operator is required to reestablish power to the lighting circuit after a fault related to temperature or voltage (step 507, FIG. 3). That being said, as previously discussed, certain types of drivers (e.g., boost-type drivers) are prone to supplying an inrush voltage far exceeding what is safe and/or required for a load when power is reestablished under the conditions described. Therefore, as part of ramping light back up (step 508, FIG. 3) power may need to first be cycled. This could happen automatically or semi-automatically (e.g., implemented as part of standard driver controller protocol) depending on the type of driver and associated controller. Also, as will be discussed, the nature of the communications means may impact how power is cycled pursuant to step 508.
(33) 1. Power Supply Options
(34) As previously stated, power for self-healing overtemp circuit 1000 is provided by power supply circuit 1002 of FIG. 5. Other power supply options are possible, and envisioned. For example, power for self-healing overtemp circuit 1000 may be pulled from line voltage; this eliminates the need for additional wiring in the interior of pole 10 (FIGS. 1 and 2). The interior space of a lighting pole is typically already filled with wires, wire harnesses, and the like—which may impact an airflow path for some active cooling techniques—and so minimizing the impact to the already crowded space within the interior of such a pole may yield multiple benefits. Of course, line voltage (e.g., 700-800 VDC post-driver on its way to the top of the pole) is ill-suited to the components of self-healing overtemp circuit 1000, and so some conditioning may be required.
(35) As a further alternative, power for self-healing overtemp circuit 1000 could be provided by a battery system; such a system might be similar to that in U.S. Pat. No. 8,946,991, or otherwise.
(36) 2. Signal/Communication Options
(37) As previously stated, communication from controller circuit 1007 of FIG. 10 so to facilitate method 500 of FIG. 3 is wireless; i.e., communicated to communication circuit 1006 (FIG. 4) which is operationally connected to a commercially available wireless controller (not illustrated), which communicates wirelessly with a complementary wireless controller (not illustrated) at the bottom of the pole. Another option is to use existing power wiring as the communications means.
(38) A wired configuration relying on powerline communications would require a nominal potential at the driver to carry a signal; therefore step 505 of FIG. 3 would be modified insomuch that the driver would not dim to 0% but to some higher percentage that was still below the excitation voltage of the LEDs (e.g., 1-5%). With a signal path in place, controller circuit 1007 could provide communication directly to the driver or controller associated with the driver along existing wiring. If the microprocessor of controller circuit 1007 did not have capability to send such a communication pulse, controller circuit 1007 could be modified to include a pulse generator circuit (e.g., including a 555 timer IC).
(39) Alternatively, self-healing overtemp circuit 1000 could exist as a standalone option; namely, with no communication means to reestablish power to the driver. In such a scenario a user would likely ascertain when a lighting fixture had likely cooled, and could manually flip a circuit breaker to reset the AC input to the driver (i.e., cycle power to the driver). The standalone option would not permit a slow ramp-down of light—the driver would immediately go to 0%—but a standalone version might be preferential for situations where an overcurrent condition is suspected, as it requires a manual override from a user who, presumably, would be equipped to troubleshoot driver failures. Alternatively, if self-healing overtemp circuit 1000 is in operative communication with remote control means for the lighting system, the remote control means may be used to cycle power. One possible example of remote control means having functionality for cycling power upon input from self-healing overtemp circuit 1000 may be as is described in U.S. Pat. No. 7,209,958 incorporated by reference herein in its entirety.
(40) 3. Overcurrent Protection
(41) Overcurrent is a situation where excessive current is provided to the load. In traditional electrical systems, an overcurrent condition is associated with a grounding fault or a short in the circuit. In the specialty LED lighting system of FIGS. 1 and 2, though, an overcurrent condition can be associated with failure of a particular component within the aforementioned driver. An overcurrent condition can arise when said component—an optocoupler—fails and the driver no longer has an accurate reading of output current. Results have shown that left unchecked, output current goes to the maximum rated for the driver, which in turn increases heat and damages parts (particularly the LEDs). If desired, self-healing overtemp circuit 1000 could include a current sensing circuit such that method 500 of FIG. 3 could also detect an overcurrent condition. As can be seen from FIG. 8, current sensing circuit 1009 generally comprises a current sensing resistor, amplifier, and a number of standard capacitors and resistors. In practice, current sensing circuit 1009 is similar to voltage sensing circuit 1004 insomuch that the signal is already existing and does not need to be generated (i.e., the output of the driver already exists), requiring only conditioning; and similar insomuch that that the input provided to controller circuit 1007 triggers an instruction based, at least in part, on characteristics of the load. In practice, method 500 would proceed similarly, but taking into account the additional current input to controller circuit 1007.
C. Exemplary Method and Apparatus Embodiment 2
(42) An alternative embodiment which is likewise formed from solid state, non-terminal event components, has a robust design, with no intervention required from a user—but which differs in a number of ways from Embodiment 1—is presently discussed.
(43) As previously stated, there are no high voltage (e.g., 1000V) solutions to thermistors or bi-metallic switches that would permit detection of a thermal runaway event and act to open a circuit, thereby terminating power to the LEDs. Recently, however, there has been an emergence of bi-metallic switches which can operate at a relatively high AC voltage (e.g., any of the Klixon® 204XX series fixed temperature thermostats available from Sensata Technologies, Attleboro, Mass., USA). While this is not an ideal solution (as LEDs such as those in FIGS. 1 and 2 operate on high voltage DC), it does present the potential for inclusion in an overtemp circuit according to aspects of the present invention.
(44) FIG. 11 illustrates an alternative overtemp circuit 2000 generally comprising a two-part detection/control circuit in combination with the aforementioned relatively high AC voltage bi-metallic strip 2010. In practice, the mains AC typically run to pole 10 (FIG. 14) is used as the input for a first circuit 2001 which—unlike Embodiment 1—exists in enclosure 40; as can be seen from FIG. 14, only bi-metallic switch 2010 exists at the top of the pole (thereby adding to the ease of daisy-chaining, which is later discussed). The electrically isolated region of circuit 2001 generally comprises surge protection and filtering means, a power supply 2002 (here rated for 1000V/1 Å, though this is by way of example and not by way of limitation), a series of capacitors and other standard devices configured to detect whether bi-metallic switch 2010 is open or closed (reference no. 2006), and the isolated portion of an opto-isolator 2007. Circuit 2001 detects opening of bi-metallic switch 2010, the characteristics of said switch (e.g., wire size, material composition) dictating the rated voltage, open temperature, and close temperature—as is well known in the art. In practice, it is anticipated an open temperature on the order of 110° C. and a close temperature on the order of 70° C. is adequate for a sports or wide area lighting systems such as that in FIG. 14; though this may differ depending on operating conditions and available cooling means.
(45) A second circuit 2009 of overtemp circuit 2000 also exists down at ground level near the drivers (e.g., in enclosure 40)—which aids in preventing signal loss and immunity to noise—and generally comprises a set-reset (SR) latching circuit 2004 (shown in greater detail in FIG. 13), a driver dimming control circuit 2005 (with optional power landing for a fan, if active cooling means are desired), and a power supply circuit 2003 (shown in greater detail in FIG. 12). Circuit 2009 instructs the LED driver(s) to dim to a preset value—which, like in Embodiment 1, can be 0% or some non-zero percentage lower than 100% —upon detection of a high voltage signal from circuit 2001. Options for restoring power are later discussed.
(46) While the present embodiment does not have the same degree of selectivity so to define conditions for a thermal runaway event as in Embodiment 1 (due to a lack of programmable thresholds), the present embodiment is a lower cost option and provides an added layer of driver control without physically interrupting the circuit (as is the case in Embodiment 1)—which should prevent issues relating to inrush current when power is restored. That being said, the present embodiment requires an additional wire pair up pole 10 (FIG. 14), which as previous stated, is an area already filled with wires and other devices which could impact airflow for some active cooling techniques. However, as a boon the present embodiment presents the opportunity to daisy chain each lighting fixture 1000 such that switches 2010 at each lighting fixture 1001 on a pole (i.e., an array) can collectively be controlled by a single driver command from circuit 2009—see FIGS. 15A and B. The benefit here is again in reduced cost and simplicity; namely, by controlling all fixtures on a pole uniformly (regardless of whether switch 2010A, 2010B, or some other switch opened), only one circuit 2009 is required, and all fixtures emit light uniformly (whether at full output or a dimmed state).
(47) Operation of overtemp circuit 2000 in a lighting system such as that in FIG. 14 according to method 500 of FIG. 3 would proceed in much of the same fashion as that in Embodiment 1. A thermal runaway event may occur (step 501), and if so, bi-metallic switch 2010 would open and be detected by circuit 2001 (step 502). Circuit 2009 would send a dimming command (as generated by circuit 2011, FIG. 13) to all drivers associated with the fixture or any fixture in the array (step 504), and driver output would decrease commensurately (step 505). Step 503 would be omitted because in the present embodiment the driver never sees an infinite load/open circuit.
(48) At some point the temperature of the fixture decreases (step 506) and the bi-metallic switch closes, which automatically restores the signal from circuit 2001 to circuit 2009 (step 507). Light output increases (step 508) in one of two ways; see FIG. 13. SR latch 2008 can be wired to latch in a fault state or not. If a fault occurs in the no-latch configuration (one in which resistors R9 and R10 are not placed and resistor R26 is placed) overtemp circuit 2000 is self-healing insomuch that power returns to the last driver setting. If a fault occurs in the latch configuration (one in which resistors R9 and R10 are placed and resistor R26 is not placed) power to the system must be cycled before normal operation can occur; however, it is important to note no devices must be physically replaced such as the e.g., aforementioned fuses or other terminal devices, to return the driver to the last setting. After cycling, circuit SR latch 2008 responds in the manner already described (i.e., in accordance with predetermined conditions). If a fault still exists, power is reduced as described in step 504. If a fault no longer exists, power returns to the last driver setting.
(49) 1. Power Supply Options
(50) As previously stated, power for self-healing overtemp circuit 2000 is provided by power supply circuit 2003 of FIG. 12, here including two linear voltage regulators—one primarily for generating dimming control voltage (U1) and one primarily for powering logic level components (U2)—though this is by way of example and not by way of limitation. Other power supply options are possible—see Embodiment 1.
(51) 2. Signal/Communication Options
(52) In line with the reduced cost/simplified circuit approach with the present embodiment, signal options are more or less limited to the wiring of SR latch 2008; though this is by way of example, and not by way of limitation.
(53) 3. Overcurrent Protection
(54) If desired, self-healing overtemp circuit 2000 could include a current sensing circuit such that method 500 of FIG. 3 could also detect an overcurrent condition; see Embodiment 1 for one example of a current sensing circuit (FIG. 8).
D. Options and Alternatives
(55) The invention may take many forms and embodiments. The foregoing examples are but a few of those. To give some sense of some options and alternatives, a few examples are given below.
(56) Exemplary embodiments have addressed electrical circuits in terms of a load comprising LEDs, a thermal runaway event indicative of failure of active or passing cooling, an overcurrent condition indicative of failure of a driver component, and triggering circuits based on (i) output voltage of a boost-type LED driver, (ii) output current of said driver, and (iii) LED temperature (for at least one embodiment). It is to be understood that there are a number of options and alternatives that could be explored without departing from at least some aspects according to the present invention. For example, while there is a benefit to measuring temperature of the LEDs—since efficacy is so closely tied to junction temperature in LEDs—temperature could be measured with respect to other parts of the system (e.g., the driver or controller). Temperature sensing circuit 1005/2001 could be installed nearly anywhere regardless of whether LED temperature or some other temperature was being measured. This is likewise true for voltage measurements. If a different power supply was used which exhibited different characteristics but still resulted in an undesirable effect, voltage may no longer be a relevant or convenient triggering metric. For example, assume a driver (if overheated) no longer provides a predictable dimming profile (i.e., a command from the controller no longer produces an expected output to the load). Fluctuations in output voltage may render the metric unreliable and so a self-healing overtemp circuit such as that described herein could rely upon photocell input for method 500. This approach could be extended to non-catastrophic or terminal events such as the thermal runaway or overcurrent events previously described. For example, aspects according to the present invention could be applied to normal driver operation so to effectuate normal dimming profiles, to run LEDs at a lower output when photocells indicate an abundance of ambient light, or to extend LED life by running LEDs at a lower output when they (for any reason) are “running too hot”.
(57) As a few additional examples of options and alternatives to those already described herein, the load could comprise other light sources (e.g., HID light sources) or non-light source loads; one possible example being a flow control system wherein a thermal runaway event results in an undesirable change in pressure or viscosity of a substance. Multiple temperature or voltage thresholds could be developed: to permit some low level light output once LEDs have cooled some (but before the thermal runaway is resolved), to permit a ramping down of light to make the change less abrupt, to more proactively identify an impending thermal runaway or overcurrent event and provide preemptive power reduction, etc. The switching circuit itself might differ—for example, using a parallel switching circuit arrangement so to permit a normally open bi-metallic switch—rather than the series circuit with a normally closed bi-metallic switch as is illustrated. Dimming could be effectuated by a reduction of driver output at full duty cycle, or by not modifying driver output and instead adjusting duty cycle, or some combination thereof. The controller could even be programmed to keep track of how many times a circuit is experiencing a fault (e.g., by tracking number of times the driver is dimmed). This may be useful in (i) determining number of cycles on parts (e.g., bi-metallic switches) so to determine when parts are reaching end of life, or (ii) troubleshooting possible nuisance tripping such as when a fault is not actually occurring due to a thermal runaway event but perhaps due to, e.g., an increase in ambient temperatures triggering a threshold.
