Patent Application
20090200952
The present invention disclosed herein relates generally to energy saving electronic lighting devices and, more particularly, to methods and apparatus for dimming light sources in an efficient and effective manner.
In the field of electronic lighting ballasts, some light sources (e.g., gas discharge lamps, fluorescent lamps, etc.) generally present a negative resistance, which causes a power source to increase the amount of current provided to the light source. If not limited, the light source, or the power source, or both, would encounter a catastrophic failure. As a result, a ballast circuit is typically provided to limit the current that the power source provides to the light source.
In many applications, it is desirable to be able to dim the light source. In residential applications, dimmers are often used to create a desirable lighting atmosphere (“mood” lighting) and/or save energy. For example, in energy saving applications, dimming is involved in “daylight harvesting” to conserve energy usage. Essentially, lighting levels in a room are dimmed as daylight available from windows or skylights increases to create a constant light level. However, conventional dimmer circuits were initially designed for resistive loads, such as incandescent light bulbs. In addition, such dimmer circuits were designed to operated with loads greater than 40 watts. Using a conventional dimmer with a conventional ballast operating with a fluorescent light, can lead to problems, because fluorescent lights are not resistive loads, but reactive loads that are primarily capacitive in nature. Furthermore, many fluorescent lights, such as compact fluorescent lamps, are less than 40 watts. Thus, using a convention dimmer on a fluorescent light can lead to flicker or limited operability (dimming) of the light source. Thus, this can preclude use of a dimmer with a single conventional compact fluorescent light bulb in a lamp. As a result, conventional dimmers are typically not operated with light sources that require ballast circuits that limit the amount of current. Thus, there is a need for a two-wire dimmer that can work with fluorescent lights.
Methods and apparatus for dimming light sources are disclosed. A dimmer circuit includes a switch to selectively couple a first node to a second node. In particular, the first node receives a line voltage from a power source which is provided to the second node when the switch is biased ON. A biasing circuit is operable to actuate the switch after a delay during each half-cycle of the line voltage. Further, the delay of the biasing circuit is typically based on a setting provided by a user. A charge circuit provides energy to the switch for a period of time to maintain the actuation of the switch for a portion of the duration of the half-cycle. In particular, the charge circuit is operable to provide energy to the switch such that the switch remains biased in the event of an operating condition that, in some instances, would cause the switch to open prematurely. This operating condition is due to a ‘ringing current’ that can cause the switch to become prematurely unlatched, causing undesirable operation. Thus, the charge circuit ensures that once the switch is biased ON, it remains ON for a portion of the duration of the half cycle, specifically during the duration that the ringing current may occur.
The charge circuit generally comprises a circuit that generates a voltage from the line voltage. The voltage generated by the charge circuit is provided in part by energy stored in a capacitor, for example. However, if the voltage generated in the charge circuit exceeds a certain threshold, a further circuit is operable to remove excess voltage from the capacitor. Further, the charge circuit comprises a second switch that is implemented by a transistor in one embodiment to provide a current to the switch in response to the switch actuating.
a and 1b are respectively a block diagram of a prior art dimmer circuit connected to a ballast and an electrical model thereof.
a-6d depict embodiments of the present invention for dimming a compact fluorescent bulb.
a and 7b represent different amounts of energy provided to a CFL at different dimming levels.
Methods and apparatus for dimming light sources are described herein. In the described examples, a dimmer circuit allows an operator to control the intensity of the light emitted by a light source with little or no flickering of the light. In addition, the dimmer circuit can be used with ballasts for gas discharge lamps (e.g., fluorescent lamps, etc.) as well as traditional light sources (e.g., an incandescent lamp, etc.) and other types (e.g., LED, etc.). Thus, the dimming system can be used with various light sources.
a illustrates a block diagram of a lighting system 100 implementing a ballast with a conventional dimmer circuit. As will be described in detail below, the conventional lighting system 100 is generally not implemented with conventional gas discharge lamps because of substantial flickering that can occur due to the interaction between a dimmer circuit 120 and a ballast 115. In
In
During the operation of the dimmer circuit 115, capacitor 140 and the adjustable resistor 145 form a RC network that has a time constant, which generally modifies the voltage provided from the power source presented at node 135. If the voltage at node 135 does not exceed a threshold associated with diac 150 (e.g., ±30 volts, etc.), diac 150 will not apply a current to triac 130, which allows the triac to remain OFF, thus preventing the line voltage from being presented to ballast 115. On the other hand, once the voltage at node 135 exceeds the threshold, diac 150 turns ON and applies a current to the gate of triac 130 turning it ON, thereby allowing voltage from power source 105 to be presented to the ballast 115. In this configuration, triac 130 is latched ON and does not turn OFF until a predetermined condition occurs, which is typically when the current flowing across the terminals of triac 130 drops below its holding current. Thus, the resistance value of adjustable resistor 145, in conjunction with capacitor 140, determines the time delay at which the voltage at node 135 rises to turn diac 150 ON, thereby turning the triac ON, and consequently turning light source 125 ON. This process occurs during each half-cycle of the line voltage. If the light source is turned on early in the half-cycle, more energy is provided to the light source and hence more light is typically perceived to the human eye than when the light source is turned on later in the half cycle. However, in many prior art systems, due to the design of the ballast, the energy savings when dimmed is not always corresponding to the reduction in light. In other words, it is not always the case that when the light is reduced by 25% that the energy consumed is 25% less. This provides a false impression to the user that dimming the light source saves a commensurate amount of energy. For some applications, such as daylight harvesting (where lights are automatically dimmed when the ambient sunlight increases the light in a room), the lack of corresponding energy savings negates the potential benefits of a daylight harvesting system.
In a typical commercial or residential building, many light sources are typically required. As a result, a substantial amount of in-building electrical wire is required to electrically couple the light sources to their respective power source. Generally, the inside wiring itself has a small amount of parasitic inductance, which for some purposes can be estimated at 19 nH/inch. The sum of the inductances due to the in-building wire itself can cause a substantial amount of parasitic inductance to be present on the power wires coupled to the ballast circuits. While it can be generally assumed there is a certain amount of inductance present, the actual values present in a particular instance are usually not known, because the exact value is highly dependent on the particular building and other parameters which vary from installation to installation. Thus, parasitic inductance is usually present, but the degree to which it is present is not known. The presence of this inductance can cause undesirable effects with regard to operation of a conventional dimmer with CFL of other light sources.
Further, the conventional ballast 115 typically includes a large electrolytic capacitor (not shown) that stores energy therein. The combination of the capacitance in the light ballast and the parasitic inductance in the wire can produce an adverse effect on the operation of a conventional dimming circuit with respect to the ballast and light source resulting in a phenomenon known as “line current ringing.”
During the operation of the ballast 115, the capacitor therein is charged at the beginning of every half-cycle of line voltage until its voltage is substantially equal to the voltage provided via the power source 105. However, the inductance that is activated by the turning ON of the triac results in additional energy that is converted into current that is present when the triac 130 is turned ON. This additional current causes a ringing condition in the LC circuit voltage. As a result, during the operation of the lighting system 100, the current of triac 130 can briefly reverse direction (i.e., it becomes negative) or drop to zero, or nearly zero. Because of this condition, the “ringing current” causes the triac 130 to experience little or no current flowing across its terminals, and it unlatches and temporarily turns dimmer circuit 120 OFF. In other words, the parasitic inductance in the power lines and the capacitance in the ballast form an LC circuit that is known to cause a current ringing condition at the triac.
Once the triac 130 is turned off, the light source goes off and the capacitor 140 is recharged causing the diac 150 to be retriggered. Thus, the diac 150 presents a current to the gate of the triac, thereby turning ON the triac. The ballast then functions to turn on the light source. This results in the triac (and the light source) being rapidly turned ON and OFF for the portion of a half-cycle of line voltage during which the light source should be on.
In other words, dimmer circuit 120 cause the light to blink when used in conjunction with ballast 115. Generally, the ringing current occurs several times within the first 250 microseconds after initially actuating triac 130. This causes light source 125 to exhibit a flicker that much longer and that is perceivable to the human eye. As a result of the flicker, conventional dimmer circuits are not used with conventional ballast circuits using gas discharge type light sources as their operation is annoying to the user.
This problem can be illustrated in another way using
The conventional dimmer 180 is modeled as having a switch 182 (e.g., triac or other suitable device). There is essentially no capacitive or inductive load associated with modeling the dimmer. Finally, the infrastructure portion 160 of the building includes an AC power source 162 which can be modeled as a DC battery in this instance, since the series of very short time windows that are used to analyze the circuit makes the power source appear as a DC source. Hence, the AC power source is modeled as a DC power source and the triac is shown as a SCR, since only one current direction is involved for the narrow time window of a half cycle. The parasitic inductance of the household wiring is shown as an inductor 164.
Assuming that the dimmer 180 is OFF, then there is no current flowing through the circuit as the voltage rises at the rectified AC power source 162 (DC source). Once the dimmer 180 turns on, current flows through the circuit and the capacitor 172 is charged to the present (instantaneous) voltage. The current also creates energy stored in the inductor 164, which is the inductance in the house wiring. The combination of the voltage source and the energy stored in the inductor added together may cause the voltage across the dimmer to be actually greater than the line voltage, and this causes the dimmer to turn OFF. The voltage at the capacitor is then reduced due to the generation of light in the bulb (not shown) which is modeled as the resistor R 174, causing the voltage across the dimmer to decrease, and causing the cycle to begin over. Typically, this ON/OFF switching of the dimmer will occur several times during a half cycle.
The dimmer circuit modifies the half wave voltage presented to the ballast from the power source as shown in
As the voltage increases, there is a voltage at another point, a node, which also is increasing with respect to time, although at a different rate. The rate of increase at the node is determined by a RC circuit, not directly by the input voltage. Further, the rate of increase is settable by the user altering the “R” value of the R-C ladder circuit. This is accomplished a user-settable potentiometer. Thus, the time constant of the RC circuit determines the aforementioned t1 of
Once the diac is turned ON (also referred to herein as “activated”) at step 306, the diac causes a solid state switch to turn ON, which provides the incoming voltage to the ballast. However, the possibility of a ringing current due to line inductance in the household wiring may cause the solid state switch, which can be embodied in a SCR, to turn OFF (also referred to as de-activated). In summary, the presence of additional voltage due to the parasitic inductance can cause the solid state switch to briefly encounter a decrease of current below its holding current, effectively shutting off the solid state switch. To prevent the solid state switch from prematurely shutting off, a voltage from a charge circuit is provided in step 308 to the solid state switch to keep it in an ON condition.
However, the solid state switch must be kept ON for a short duration—only long enough to prevent the ringing current from inadvertently turning the solid state switch off. In any event, the solid state switch should not be kept ON by the charge circuit past the half cycle. Thus, in step 310, the energy from the charge circuit is dissipated shortly after activating the solid state switch ON, which allows the solid state switch to turn OFF when the voltage across its terminals is near zero at the end of the half cycle. In other words, the charge circuit keeps the solid state switch ON for a short while after it is initially turned ON, to prevent it from prematurely turning OFF. In some embodiments, exemplary process 300 biases the switch ON at step 308 for a period of time in the range of approximately 100 to 2000 microseconds. The time period for which the switch is biased ON depends on the point (relative to the incoming voltage waveform) when the switch is initially triggered ON. Once the charge circuit is deactivated, the charge circuit does not by itself cause the solid state switch to turn OFF, but merely allows the solid state switch to turn OFF when conditions are appropriate.
In step 312 when the voltage of the half cycle nears zero, and the current through the solid state switch is near zero, the solid state switch unlatches, and turns OFF, as is desired. Because the charge circuit is no longer preventing the switch from turning OFF, and the voltage across the solid state switch is zero, the solid state switch is able to turn OFF. Thus, process 300 unlatches the switch (i.e., opens) at the end of each half-cycle of line current. At the beginning of the next half cycle, the process then repeats at step 300.
In exemplary process 300, the operation of the charge circuit keeps the solid state switch in an ON condition regardless of the load current. Thus, in the event of an operating condition such as a ringing current, which would normally otherwise cause the SCR to experience substantially no current flowing from across its terminals and thereby turning it OFF, the gate of the SCR remains biased to keep the SCR latched ON. Further, the charge circuit is operable to allow the switch to shut OFF at the end of each half-cycle of the line voltage. Accordingly, a light source connected to such a dimmer that implements exemplary process 300 would experience no perceivable flickering during its operation and would be presented with the waveform 706 of
In the illustrated embodiment of
In the illustrated embodiment, transistor 442 is implemented by an N-channel metal oxide semiconductor field effect transistor (MOSFET), but transistor 442 can be implemented by any suitable solid state device (e.g., a switch, a bipolar transistor, a P-Channel MOSFET, an insulated gate bipolar transistor, HEXFET, triac, sensitive gate SCR, etc.). The drain of transistor 442 is connected to node 410 and its respective source is connected to the gate of a silicon controlled rectifier (SCR) 446. In the embodiment of
The operation of the dimmer circuit 424 will be explained in conjunction with a half-cycle of the line frequency of the power source 402. In particular, the diodes 406, 408, 414, and 416 allow a line voltage to be present to the dimmer circuit 424 via node 410. Initially, the only current flowing from node 410 to 412 is due to current flowing through the adjustable resistor 450 and capacitor 452, and current through capacitor 426 in series with winding 428. However, although capacitor 426 stores energy at a voltage, it is of a small enough value that it does not effect the RC time constant of potentiometer 450 and capacitor 452. The adjustable resistor 450 and capacitor 452 increase the voltage at node 448 at a rate that is determined by the resistance value of the adjustable resistor 450, which is typically selected by a user. After a delay based in part on the value of adjustable resistor 450, the voltage at node 448 exceeds a threshold voltage associated with diac 454. As a result, diac 454 enters what is commonly referred to as a “breakdown” mode and allows current to flow through its respective terminals. In response, current flows into the gate of SCR 446, which causes SCR 446 to latch ON and couple node 410 to node 412 via a low impedance path. SCR 446 is latched ON, thereby causing its respective gate to lose control over its operation. SCR 446 remains latched ON until it experiences an operating condition causing it to unlatch, which is typically when the current flowing through its respective terminals is below its holding current. In this embodiment, the components comprising diac 454, adjustable resistor 450, and capacitor 452 comprise the “bias circuit” as these component initially bias the SCR into an ON condition. Other components can be used to construct a biasing circuit.
There is a nominal amount of current required to run the ballast. When the SCR turns ON, there is an excessive amount of current that rings in flowing from node 410 to 412. By “ringing” this means that there is a current peak above the nominal amount of current (thereby adding to the nominal current) and an amount less than the nominal current amount (thereby subtracting from the nominal current). When current level subtracts from the nominal amount, this can cause the current through the SCR to drop below the holding current level, causing to turn off. The current added to the nominal amount is due to the parasitic inductance in the power line wiring, which is produced when the SCR turns on. Thus, a higher than normal current is provided to the ballast, which then reduces in level causing the current in the SCR drops to zero or near zero, resulting in the aforementioned ringing condition causing the SCR to turn OFF.
This undesirable condition is addressed in one embodiment by the charge circuit which comprises the component shown within the dotted line 499 in
When current begins to flow through SCR 446 (e.g., when SCR 446 is ON or activated), capacitor 426 discharges the energy stored therein as a current flowing through the primary winding 428, which induces a voltage in the secondary winding 432 that turns into a current causing the charging of capacitor 436. The transformer in this embodiment is a non-gapped, wound transformer, double E core, with a 4 to 1 turn ratio, and having a 10 micro-second hold up time at 50 volts. However, other configurations having similar functional properties can be used, as will be discussed below. In particular, primary and secondary windings 428 and 432 cause node 430 to have a voltage, but the voltage at node 430 is configured to not exceed the voltage at node 410 by means of zener diode 438. In this embodiment, the transformer can be viewed as a voltage transformer, where the voltage generated by the transformer is determined by the voltage associated with capacitor 426. As will be described in detail below, because node 410 is connected to power source 402, the voltage at node 430 is reduced because of the step-down of the transformer. This voltage at node 430 bias transistor 442 such that it supplied gate current to SCR 446 to prevent it from unlatching (i.e., turning OFF).
The amount of energy discharged by capacitor 426 depends on the amount of energy stored therein. Recall that the discharging of the capacitor is caused by the triggering of SCR 446, and thus the amount of energy stored in the capacitor is a function of when the SCR is triggered. Thus, the amount of energy stored (and discharged) depends on the relative time when the SCR 446 is triggered. For example, if the SCR is triggered shortly after the incoming line voltage increases above zero, (such as corresponding to time t1 in
In the illustrated embodiment, the voltage from the secondary winding 432 causes a charge to be stored in capacitor 436, thereby causing node 430 to have a voltage present. Further, diode 434 prevents the charge in capacitor 436 from discharging backing into the winding 432. However, if the voltage at node 430 exceeds a breakdown voltage associated with zener diode 438, zener diode 438 enters what is commonly referred to as the “avalanche breakdown mode” and allows current to flow from its cathode to its anode (i.e., into node 412). Once the voltage at node 430 does not exceed the breakdown voltage, the zener diode 438 recovers and prevents current from flowing into node 412. Stated differently, the zener diode 438 limits the voltage stored in the capacitor 436 so that its voltage does not exceed a predetermined threshold. While the zener diode could be omitted, it provides a safety mechanism to avoid damage to the FET 442.
Resistor 440 cause capacitor 436 to dissipate the energy stored therein at a predetermined time. Resistor 440 ensures that the energy in capacitor 436 will dissipate so capacitor 436 does not keep transistor 442 ON (and thereby keeping the SCR 446 ON) longer than desired. Resistor 444 is used as a current limiter if a bi-polar transistor is used and to prevent parasitic oscillation conditions if a MOSFET is used. The transistor 442 should only keep the SCR ON for a short duration so that the SCR is not turned OFF due to the ringing current, and certainly the SCR should not be kept ON past the duration of the half cycle. In particular, resistors 440 and 444 are configured to cause transistor 442 to have a gate-source voltage thereby turning ON and causing the gate of SCR 446 to have a gate-cathode current resulting from on the charge stored in capacitor 436. Stated differently, resistors 440 and 444 keep the gate of SCR 446 energized only for a period of time based on the amount of charge stored in capacitor 436. In the illustrated embodiment, zener diode 438, capacitor 436, and resistors 440 and 444 are configured to bias the gate of SCR 446 by way of transistor 442 for a period of time approximately in the range of 100 to 2000 microseconds. The duration of the biasing of SCR 446 by transistor 442 depends on the amount of energy stored in capacitor 436, which is charged from the energy stored in capacitor 426. Thus, the point in time relative to the input voltage waveform when the SCR is triggered impacts how long the SCR will be biased by the charge circuit. The biasing duration is also limited by the zener diode 438 and the resistor 440. Consequently, the charge circuit 499 biases SCR 446 for a short portion of each half-cycle of the line voltage and allows the SCR to unlatch itself at the end of each half-cycle. Although the biasing duration is variable, it is long enough (e.g., typically in the range of 100-2000 microseconds) to ensure that the SCR remains ON, but is not kept on past the end half cycle. The charge circuit provides a current through the gate to turn the SCR ON only when the SCR is OFF. That is, the charge circuit is configured to provide a biasing current to the SCR during the required time period when it is OFF, but no current is required if the SCR is latched ON. It is only when a ringing current condition exists that the SCR may become unlatched, and that is typically when the charging current provides current to turn the SCR back ON.
As described above, if driving a capacitive load such as an electronic ballast, the parasitic impedance in the wiring of a building may cause SCR 446 to experience a ringing current, which may cause the current flowing through the SCR 446 to be less than its holding current. In other words, SCR 446 may experience the operating condition that may cause it to unlatch prematurely. If so, then at the same time, current will begin to flow through adjustable resistor 450 and the capacitor 452, which will cause the diac 454 to retrigger. Thus, this will result in a flickering condition of the light source. However, as described above, capacitor 436 stores a charge in response to SCR 446 being turned ON, which causes the transistor 442 to have a gate-source voltage. As a result of the gate-source voltage of transistor 442, SCR 446 has a gate current due to the load current that was through the SCR and remains latched ON for substantially the same duration that transistor 442 is turned ON. That is, when SCR 446 is turned ON, it receives a current to prevent it from becoming unlatched as a result of the ringing current. As a result, the light source 420 does not flicker during the operation of each half-cycle of the line current.
In the embodiment of
The first terminal 504 of power source 502 is connected to a first node 512, which is further coupled to a second node 514 via a primary winding 516 and a capacitor 518. Node 512 is further coupled to a second node 520 via a secondary winding 522 and a diode 524. In particular, the cathode of diode 524 is connected to node 520 and its respective anode is connected to secondary winding 522. In addition, node 512 is coupled to node 526 via secondary winding 522 and a diode 527, which has its respective anode connected to node 526 and its cathode connected to secondary winding 522.
Node 520 is also coupled to node 512 via capacitor 528 and resistor 530, which are configured in parallel. Further, node 520 is also connected to the cathode of a zener diode 532, which is coupled to node 512 via its respective anode. Further still, node 520 is also coupled to the gate of a transistor 534 via a resistor 536. In the embodiment of
The drain of transistor 534 is coupled to node 514 via a diode 548. In particular, the anode of diode 548 is connected to node 514 and its respective cathode is connected to the drain of transistor 534. The source of transistor 534 is connected to the source of transistor 544, both of which have their respective sources that are further coupled to a node 550 via a diac 552. In addition, the sources of transistors 534 and 544 are connected to the gate of a triac 554. The drain of transistor 544 is coupled to node 514 via a diode 556. In particular, the cathode of diode 556 is connected to node 514 and its respective anode is connected to the drain of transistor 544.
In the illustrated embodiment of
In the illustrated embodiment of
If the current generated by the secondary winding 522 is negative, a current flows into node 526 and capacitor 538 stores the current as a voltage. However, zener diode 542 limits the voltage across capacitor 538. As a result of the voltage, the resistors 540 and 546 cause a predicated amount of current to flow into node 512. The resistors 540 and 546 are configured to limit the amount of current. As a result, a voltage is generated and causes transistor 544 to have a gate-source voltage, thereby turning ON transistor 544. However, because the resistors 540 and 546 limit the current, transistor 544 is turned ON for a period of time after SCR 554 latches ON. In some embodiments, the transistor 544 is operable for a range of approximately 100 to 1000 microseconds. As a result of turning ON transistor 544, triac 554 continues to have a gate current, thereby ensuring the triac 554 is latched for a period of time after turning ON.
On the other hand, if the current generated from secondary winding 522 is positive, a current flows into node 520 via diode 524. The current is stored as a charge in the capacitor 528 as a voltage; however, zener diode 532 limits the voltage stored therein. As a result of the voltage, the resistors 530 and 536 cause a predicable amount of current to flow into node 512. The resistors 530 and 536 are configured to limit the amount of current. In response to the current, a voltage is generated and causes transistor 534 to have a gate-source voltage, thereby turning it ON. However, because resistors 530 and 536 limit the current, transistor 534 is turned ON for a period of time once triac 554 is latched ON (e.g., typically between 100 microseconds-1000 microseconds). As a result of turning ON transistor 534, triac 554 continues to have a gate current thereby ensuring the triac 554 is latched for a period of time after turning ON.
In the embodiment of
In the described embodiments, a dimmer circuit is provided that is able to dim light sources operating with a ballast having a capacitive load without noticeable flicker. Further, the dimmer circuit is capable of operating with any type of light source (e.g., incandescent bulbs, gas discharge lamps, LEDs, etc.) over the wide range of light output (e.g., from 20% to 100%) and for a variety of power loads. The dimmer circuit can be easily implemented into existing manufacturing processes without substantial additional costs. In addition, the dimmer circuit is capable of handling lower current, approximately in the range of 10 to 20 milliamps, thereby allowing the ballast to function with a single dimmable CFL. As a result, the described embodiments above are capable of handling low power light sources such CFLs.
a is an illustration of another embodiment of the invention as used for dimming a conventional CFL, typically in the 10-40 watt range. The diagram illustrated in
In
As the current flows across the SCR, capacitor 606, which is a 0.1 μf capacitor, causes the charge to be transferred at a voltage into the transformer 608 and then into capacitor 612. This transformer in this embodiment can be viewed as functioning as a voltage transformer. The transformer can be made using a Ferrite Core No. 9478016002, using #27 wire, where the primary has 80 turns, and the secondary has 20 turns. The presence of current on the primary winding induces a current on the secondary windings, causing a voltage to appear at the cathode of diode 610 and the energy is stored in capacitor 612. Diode 610 is a conventional INF4004 diode and prevents any current from flowing back into the transformer.
The zener diode 614 is rated at 12 volts and 0.5 watt, and prevents the voltage at the cathode of diode 610 from exceeding 12 volts due to the release of energy from capacitor 612 through resistor 616. Resistor 616 has a 1K value and resistor 616, which is a value of 10K. The presence of the voltage at the resistors causes the transistor 618 to have a gate-source voltage, which turns the transistor 618 ON. The transistor in this embodiment is a FET IRFU420 from International Rectifier™. This transistor causes the SCR's gate to be energized, and keeps the SCR from turning off.
As noted previously, the DIAC turns “ON” the SCR at a delayed point relative to the start of each half cycle. The time at which this occurs is determined by the RC value of capacitor 626 and resistor 624. Since resistor 624 is a user-settable potentiometer, the time value varies based on the user's setting. The varying delay at which the DIAC turns the SCR ON determines the energy delivered to the CFL, and therefore determines the light produced.
a also includes a line filter 627, which may be present in a commercial embodiment of the invention. The line filter, embodied as an inductor, lowers the di/dt of the current thereby reducing the high frequency electrical noise being introduced back into the power lines. Because of the potential proliferation of dimmers, such noise limiting inductors (or other equivalent circuitry) are used to avoid introducing noise on the power line infrastructure, whether it be in the building where the dimmer is being used, or otherwise. Because the noise filter reduces the change in current (e.g., di/dt), it by itself can be used in some embodiment (as discussed below) to facilitate the current ringing problems.
b is another embodiment, which is similar in concept to
c is another embodiment, which is similar in concept to
d is another embodiment, which includes a subset of the components shown in embodiments illustrated in
As noted previously, the SCR 620 can be de-activated, or turned OFF, by the presence of a ringing current due to inductance in the power lines, which the full wave bridge processes into a ringing current present on node 603 when SCR 620 is turned ON. In the alternative embodiment shown in
The inductor in this embodiment preferably comprises #21 wire turned around a powered iron core, such as an E75-26 core available from Micrometals™. For applications supporting a low power load (e.g., less than 20 watts), the SCR could become unlatched, and cause flickering. This can be avoided by using a SCR with a lower holding current, such as those readily available having a 6-8 ma holding current.
As shown in
The RC value (based on the adjustable potentiometer) discussed previously defines the time value or delay at which the diac reaches the threshold voltage, and thus turns on the SCR. In
The above dimmer can be manufactured to be contained in a conventional dimmer switch housing with a shape and size allowing it to be installed in a conventional single gang work box (i.e., the box used in construction to contain electrical switches). This allows the dimmer switch to be retrofitted into existing residential or commercial applications, as well as for new construction. The embodiments of the dimmer disclosed herein can accommodate lighting load applications of 5-300 watts, and different component values may be scaled for higher (e.g., 300+) wattage applications. Such values for higher wattage applications can be readily determined by one skilled in the art. Such applications include dimming single fixture lighting sources, or a plurality of lighting sources controlled by a single dimmer. Further, the aforementioned dimmer can function with a variety of lighting technologies and provide flicker-free dimming over a wide range of luminescent output of the lamp. Further, multiple light sources can be dimmed using a single dimmer.
The dimmer described herein has an additional advantage of effecting a linear or approximately linear dimming response as the dimmer switch is operated, and can dim certain dimmable CFLs down to as low as 20% of maximum luminous intensity. Further, the dimmer can effectively dim lighting loads having a lower power load than prior art dimmers, which often do not function well with low wattage CFL bulbs. The present dimmer is particularly well suited for dimmable low wattage LED based lights. These features are improvements over many commercially marketed dimmer circuits for CFL bulbs. Further, the dimmer contains no programmed microprocessor. The advantages of this dimmer potentially lead to wider use of energy-saving CFL bulbs, and further save energy by allowing more CFL bulbs to be operated at reduced energy consumption.
In the above embodiment, a potentiometer in the disclosed circuit that determines the light output is operated by a user. The user varies the potentiometer setting to obtain the light output as desired. In other embodiments, the dimmer may be incorporated into a system where the value of the resistance is controlled automatically by an additional circuit or the time which the diac is turned on is otherwise determined.
The resistance value itself, or the RC time constant formed by the combination of the resistance and capacitance values, may be adjusted based on various conditions. For example, the dimmer circuit can be incorporated with a motion detector, which can be embodied in a night-time security light. Such devices commonly control outdoor lighting, and upon detecting motion, turn on a security light. Such a device may incorporate multiple light output levels. For example, at night and when motion is not detected, the lights can be dimmed to a certain level (e.g., 50%) to provide a low level security light. When the motion detector detects motion, the light is then turned on to full power, often for a limited time period after which no further motion is detected. After the limited time period expires, the light returns to the lower level. At dawn, the light is then completely turned off. In such embodiments, a control circuit would determine when the diac turns on to cause the partial light level in the absence of motion, and change the time when the diac turns the SCR on (or the time delay caused by the RC time constant) based on when motion is detected such that full light is produced. Such circuits are well known to those skilled in the art, and can be found, for example, in U.S. Pat. No. 7,164,238, the contents of which is incorporated by reference.
In other embodiments, the time delay and dimming level may be varied by other means. Continuing with the security example, a photocell could measure ambient darkness for controlling the security light. Such a circuit would cause the security light to be activated, initially at a very low light output level. As darkness increases, as detected by the photocell, a commensurate change in the time delay would occur so as to cause the light to gradually increase its output. This would avoid turning on the lamp at full power, when full power may not be initially required based on ambient conditions. In other embodiments, a microprocessor can be programmed to cause the light power to gradually increase over a set time period by changing the time delay according to its program.
Such arrangements are used for “daylight harvesting.” Daylight harvesting is another approach for conserving energy usage which involves dimming lights. One application is in an office environment, where it is common for offices on a given floor to have different directional exposures (e.g., south, north, east and west). If an office has an eastern exposure, light fixtures in that office may be coupled with a light sensor to detect ambient lighting conditions which are processed by a microcontroller. The controller then determines which lights to dim and when. The selective dimming of various lights can be used to “balance” or average the light in a work environment by dimming lights adjacent to a window having, for example, an eastern exposure in the early morning (when the morning light is brighter). Further, the present dimmer circuit can be used by a controller to change dimming levels at a slow pace such that the change in light output is hardly perceived by the occupants, and does not cause the light to blink. Thus, the dimmer circuit may be coupled with processors, timers, light detectors, motion detectors, and/or other circuitry in various ways to efficiently dim a lighting source as needed.
One embodiment of a light harvesting system is shown in
The second portion detects the ambient light conditions via a photo-resistor 812, which is placed to detect the desired ambient light conditions as appropriate. The photo-resistor 812 and a second resistor 818 form a voltage ladder, such that an input 814 measurement voltage changes according to the light conditions. The input is provided to a microprocessor or microcontroller which is able to convert the analog voltage reading to a digital value and process it according to its program. The controller 816 is programmed to effect the desired operation.
The controller 816, based on the input voltage reading 814 then adjusts an analog output 820. The output is provided to an operational amplifier 810 which drives a transistor 806. When the transistor 806 is turned ON, current flows from the LED 804 through the transistor 806 and is limited by resistor 808. Thus, based on the level of the current passing through the LED, the light level and therefore the resistance of the photo-resistor 802 can be changed by the controller 816. In this manner, the controller can be programmed to dim a light (or series of lights) controlled by the dimmer, based on detected ambient light conditions. Similarly, the controller could receive other inputs (such as motion detection, time of day, etc.) and use these inputs to alter the resistance of photo-resistor 802. Those skilled in the art will recognize that the microprocessor could utilize external A/D and/or D/A circuits. Similarly, those skilled in the art will readily recognize that the digital microcontroller 816 can be replaced with analog circuitry to control the brightness of LED 804.
The purpose of daylight harvesting is to save energy when ambient natural light conditions allow reduction of artificial light. The present dimmer effectively accomplishes this when operated with a light source using the ballast described in U.S. patent application Ser. No. 12/277,014, filed on Nov. 24, 2008. The use of such a ballast with the dimmer described herein allows a generally commensurate reduction in energy consumption when dimming. Thus, the combination realizes the benefits of daylight harvesting while maintaining a high efficiency and high power factor, without any flickering of light, even when dimmed to a very low level. Thus, this allows artificial lights to be dimmed when there is sufficient ambient light and increased gradually as ambient light increases. This combination allows energy savings to be realized.
This application is a continuation-in-part of U.S. application Ser. No. 12/205,564 filed on Sep. 5, 2008, which in turn claims the benefit under 35 U.S.C. §119(e) to U.S. Provisional Patent Application entitled “Two-Wire Dimmer Switch for Dimmable Fluorescent Lights” filed on Feb. 8, 2008, bearing Ser. No. 61/006,967, both of which are herein incorporated by reference for all that each teaches.
| Number | Date | Country | |
|---|---|---|---|
| 61006967 | Feb 2008 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12205564 | Sep 2008 | US |
| Child | 12353551 | US |
