This disclosure relates generally to electronics and more particularly to driving light emitting elements.
The output color of a white Light Emitting Diode (LED) has some deficiencies in the form of reduced color in some parts of the visible spectrum. To correct for the white LED deficiencies a second “color adjust” (CA) LED string is used to fill in the spectrum in the areas where the white string is deficient. The combination of the white LED string and the CA LED string produce a pleasing white output. Due to increased demand for low cost solutions for various LED lighting applications, color correcting device drivers must now be designed with fewer or less expensive components.
A color correcting device driver is configured to vary the equivalent current into light emitting elements (e.g., LEDs) with the frequency of the AC input current (e.g., 120 Hz). In implementations that include a fly-back controller with a PFC controller on the primary side, the color correcting device driver performs the method of: 1) turning on the loads (e.g., white and CA strings of LEDs); 2) determining if the voltage supplied to the loads has dropped by a first threshold amount; 3) turning off the loads; and 4) determining if the voltage supplied to the loads has recovered by a second threshold amount (or waiting for a fixed amount of time). The method is then repeated.
In implementations that do not include a PFC controller on the primary side, the color correcting device driver can create a pulse width modulation (PWM) signal by detecting the starting point for a sine wave PWM approximation and starting the PWM approximation at the correct frequency. In some implementations, an inductor in series with the CA string is removed and the CA string is driven linearly.
Particular implementations of a color correcting device driver can provide several advantages, including but not limited to: 1) power factor correction; 2) high efficiency; 3) long product life time; 4) reduced size for capacitor used to compensate for current supplied by the PFC controller; 5) removal of the inductor that is connected in series with the CA string; and 6) removal of the recirculating diode that is connected in parallel with the CA string.
The details of one or more disclosed implementations are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages will become apparent from the description, the drawings and the claims.
The number of strings 116, as well as the number of elements in each string, may depend on the particular type of device and application. For example, the device driver technology described here can be used, for example, in backlighting and solid-state lighting applications. Examples of such applications include LCD TVs, PC monitors, specialty panels (e.g., in industrial, military, medical, or avionics applications) and general illumination for commercial, residential, industrial and government applications. The device driver technology described here can be used in other applications as well, including backlighting for various handheld devices. The device driver 100 can be implemented as an integrated circuit fabricated, for example, on a silicon or other semiconductor substrate.
An AC input voltage (e.g., sinusoidal voltage) is input to FWR 102, which provides a rectified AC voltage. PFC controller 104 is configured to convert the rectified AC voltage on the primary side of transformer 103 to a DC voltage (Vout) on the secondary side of transformer 103, for driving strings 116, 117. PFC controller 104, together with transistor 104 and sense resistor 105 assures that the current drawn by strings 116, 117 is in the correct phase with the AC input voltage waveform to obtain a power factor as close as possible to unity. By making the power factor as close to unity as possible the reactive power consumption of strings 116, 117 approaches zero, thus enabling the power company to efficiently deliver electrical power from the AC input voltage to strings 116, 117.
Capacitor 110 compensates for the current supplied by PFC controller 104 by holding a DC voltage within relatively small variations (low ripple) while the load current is approximately DC and the current into capacitor 110 is at twice the frequency of the AC input voltage. When the AC input voltage is zero, the current in secondary coil 103b goes to zero and capacitor 110 provides the current for strings 116, 117. To keep the DC ripple low, a large electrolytic capacitor often is used, which can be unreliable, costly and have a limited life span.
Resistors 108, 109 form a voltage divider network for dividing down Vout before it is input to the feedback (FB) node of device controller 111 and shunt regulator 107. Device controller 111 forces current out of the FB node to regulate the Dw node at a desired level (typically 1V). Shunt regulator 107 acts as a reference for the feedback loop and provides current to opto-coupler 106. Recirculating diode 118 (e.g., a Schottky diode) recirculates current from CA string 117 when the PWM on the gate of transistor 120 is turned off.
In the circuit configuration shown, white string 116 uses most of the power CA string 117 uses a smaller amount of power to fill in the color spectrum. For example, white string 116 may require approximately 40 volts and 350 mA (14 watts), while CA string 117 requires approximately 20V and 150 mA (3 watts).
Device controller 111 resides on the secondary side of transformer 103. Device controller 111 is coupled to the drain, gate and source terminals of transistor 112 through nodes Dw, Gw and Sw. Device controller 111 is further coupled to the drain and source terminals of transistor 120. Device controller 111 sets the voltage and current through white string 116 by commanding transistor 112 (e.g., MOSFET transistor) on and off using a PWM waveform (e.g., applied to the gate of transistor 112 through node Gw) with a suitable duty cycle. The current is set by an amplifier loop in device controller 111 (not shown) by controlling the voltage across sense resistor 115. The voltage across white string 116 is controlled by measuring the drain voltage (Dw) of white string 116 and feeding back a current into the feedback node (FB) such that the drive (transistor 112 and sensor resistor 115) has just enough headroom to supply the required continuous current to strings 116, 117.
Similarly, device controller 111 sets the voltage and current through CA string 117 by commanding transistor 120 (e.g., MOSFET transistor) on and off using a PWM waveform (e.g., applied to the gate of transistor 120 through node Gfb) having a suitable duty cycle. The current is set by an amplifier loop in device controller 111 (not shown) by controlling the voltage across sense resistor 121. The voltage across CA string 117 is controlled by measuring the drain voltage (Dw) of CA string 117 at node Dfb. Since CA string 117 has a lower voltage than white string 116, a floating buck configuration can be used to regulate the current in inductor 119 (L1) to regulate the current in CA string 117. Internal to device controller 111 is a look-up table for adjusting CA string 117 brightness as a function of temperature.
Circuit 100 provides power factor correction, high efficiency and a long product life, but also has deficiencies in that capacitor 110 is extremely large, both physically and in value. This adds cost and space to the design. The large capacitor 110 also has a shorter useful life span. Additionally, inductor 119 used in the floating buck is both large in value and physically large, adding cost to the design.
In circuit 200, shunt regulator 107 has been removed and opto-coupler 106 is coupled directly to FB drive node. The equivalent of a shunt regulator is internal to device controller 111. Inductor 119 and recirculating diode 118 have also been removed from circuit 200, as these parts are no longer needed in this circuit configuration.
Process 400 can continue by determining if a voltage supplied to the loads has dropped by a first threshold amount (404), such as 500 mV. The voltage can be measured from a resistor divider from the output (resistors 108, 109), by observation of the Dw node of device controller 111 or by observation of the Dfb node of device controller 111. This has the effect of determining how much ripple is allowed on capacitor 110.
Process 400 can continue by turning off the loads (406) and determining if the voltage supplied to the loads has recovered by a second threshold amount (408) (e.g., 500 mV). For example, a recovery time can be a fixed amount of time (e.g., about 1 μs). Process 400 then returns to step 402 and repeats.
To vary the ratio of the white string to CA string duty cycles, the average PWM over the frequency of the AC input (e.g., 120 Hz) can be determined. Once the ratio is determined, CA string 117 can be turned off for the rest of the duty cycle and only white string 116 is pulse width modulated.
With process 400, if the current into capacitor 110 is equal to the current out of capacitor 110, then the voltage on capacitor 110 is DC. If the voltage on capacitor 110 is DC, then capacitor 110 can have a very small capacitance value. Since the ripple on capacitor 110 is regulated, capacitor 110 is kept at the correct DC voltage (plus some ripple), and only a small capacitor 110 is required to maintain the desired voltage.
When controller 104 on the primary side of transformer 103 does not include PFC, a good PFC can be obtained by creating the PWM using an n-level PWM approximation of a sine wave and synchronizing the sine wave to the AC input waveform.
The start time and correct frequency for the current waveform can be determined by detecting zero crossings of the AC waveform or FWR waveform. The correct frequency can be determined by detecting two start times. Because a perfect power factor of one cannot be created, the AC waveform will be superimposed on the DC output at the secondary side. By monitoring the output voltage, we can determine the phase of the input and the correct phase to load the output. The output can be directly monitored through the FB pin. A comparator can detect the zero crossing. It may be desirable to AC couple the output to device controller 111 for a larger sense signal. Additionally, a low pass filter can be added to remove the switching and PWM noise to improve the signal-to-noise (SNR) ratio in the zero crossing detector. Alternatively, Dfb or Dw can be used to sense the output.
Typically, a non-PFC controller (e.g., a standard controller) requires a large hold capacitor on the primary side of transformer 103 to provide power when the AC voltage drops (in the valleys of the rectified AC voltage). Because circuit 200 draws current in the correct phase/frequency for good PFC, a hold capacitor on the primary side of transformer 103 is not necessary, although a small capacitor can be added for electromagnetic interference (EMI). Because the hold capacitor is very small, the secondary voltage will drop significantly under any load near the valleys of the AC input. This signal can be used to synchronize both the phase and the frequency of the LED loads.
Circuit 100 includes a floating buck topology as a power converter. Such a configuration includes inductor 119 and recirculating diode 118 (e.g., Schottky diode). Circuit 200 can be configured without the large inductor (L1) of circuit 100, which can be about 800 μH. Instead of using 20V and 150 mA LEDs for CA string 117, inductor 119 can be removed and lower current LEDs can be used for CA string 117. For example, white string 116 can be 40V and 350 mA (14 watts) and CA string 117 can be 20V and 15 mA (3 watts). Eleven 85 mA LEDs in CA string 117 in series requires about 36.7V but uses 40V. This uses a total of 17.4 mA for a loss of just 2.3%. Accordingly, with 10 or 11 diodes in CA string 117, the loss is so small that it can be more efficient than the floating buck configuration used in circuit 100. It is not necessary to use lower current LEDs in CA string 117 to get the higher efficiency if the number of series connected LEDs is set to the correct valued described above. If a higher current LED is used, the duty cycle can be reduced accordingly to get the correct average light output required. However, there is typically a cost savings associated with lower current LEDs.
While this document contains many specific implementation details, these should not be construed as limitations on the scope what may be claimed, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub combination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can, in some cases, be excised from the combination, and the claimed combination may be directed to a sub combination or variation of a sub combination.
This application is a continuation of and claims priority to U.S. application Ser. No. 13/291,943, filed on Nov. 8, 2011, which is hereby incorporated by reference in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
8054008 | Kimura | Nov 2011 | B2 |
20110128303 | Yonemaru et al. | Jun 2011 | A1 |
20120249005 | Jin et al. | Oct 2012 | A1 |
20130082621 | Kotowski et al. | Apr 2013 | A1 |
20130113381 | Cai et al. | May 2013 | A1 |
20130147358 | Kotowski et al. | Jun 2013 | A1 |
20130147658 | Burri et al. | Jun 2013 | A1 |
Number | Date | Country | |
---|---|---|---|
20140055047 A1 | Feb 2014 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 13291943 | Nov 2011 | US |
Child | 14071281 | US |