Whether the LED is driven via a buck, boost, buck/boost or linear regulator, the most common thread connecting each driver circuit is the need to control the output of the light. Today, there are only a few applications that require simple functions to turn on and off, and most need to fine tune the luminosity from 0 to 100%. Currently, for photometric control, the two main solutions are linearly adjusting the LED current (analog dimming) or switching the drive current from 0 to the target current value at high frequencies that are undetectable to the naked eye (digital adjustment) Light). Using pulse width modulation (PWM) to set the cycle and duty cycle may be the easiest way to implement digital dimming because the same technique can be used to control most switching converters.
PWM dimming can be used to match accurate color light
In general, analog dimming is easier to implement because the output current of the LED driver changes proportionally to the control voltage, and analog dimming does not cause additional electromagnetic compatibility (EMC) / electromagnetic interference (EMI) potential frequencies. problem. However, most of the reasons for PWM dimming are based on the basic characteristics of LEDs, ie the displacement of the emitted light is proportional to the average drive current (Figure 1). For monochromatic LEDs, the wavelength of the main light wave changes, and in the case of white LEDs, the relative color temperature (CCT) changes. For people's naked eyes, it is difficult to detect the change of nanometer wavelength in red, green or blue LEDs, especially when the intensity of light is also changing, but the color temperature change of white light is easier to detect. Most white LEDs contain a wafer that emits blue-spectrum photons that emit photons in various visible ranges after striking the phosphor coating. At lower currents, phosphorescence dominates and deflects light toward yellow; at higher currents, the LED emits more blue light, causing the light to deflect toward blue and also produce a higher CCT. For applications that use more than one white LED, the CCT difference between two adjacent LEDs can be significant and visually unpleasant, and this concept can further extend the source of multiple monochromatic LED rays. Once more than one light source is present, any CCT difference that occurs between them can be dazzling.
Figure 1 LED driver and waveform using PWM dimming
The LED manufacturer specifies the magnitude of the drive current in the current characteristics table of its product, which only guarantees the product's dominant wavelength or CCT under these specific current conditions. The advantage of PWM dimming is that there is no need to consider the intensity of the light, and it also ensures that the LED emits the color desired by the designer. This precise control is especially important for red, green, and blue (RGB) applications because these applications mix light of different colors to produce white light.
From the perspective of driver ICs, analog dimming faces serious challenges in output current accuracy. Almost all LED drivers add some form of serial resistor to the output to detect current, and the selected current sense voltage VSNS creates a coordination effect that allows the circuit to maintain a high signal-to-noise ratio (SNR). While maintaining low power consumption, the error caused by the tolerance, offset, and delay in the driver is relatively fixed. To reduce the output current in a closed loop system, the VSNS must be lowered, but the accuracy of the output current will decrease until the absolute value of VSNS equals the error voltage. Finally, the output current becomes uncontrollable. The target output current will not be determined or guaranteed. In general, in addition to improving the accuracy of PWM dimming, the linear control of low-order light output is also stronger than analog dimming.
Dimming frequency is inversely proportional to contrast
For PWM dimming signals, each LED has a defined response time. Figure 2 shows three different delays. The higher the delay, the lower the contrast that can be achieved (a measure of light intensity control). .
Figure 2 dimming delay
The amount of time tD in Fig. 2 represents the propagation delay from the rise of the logic signal VDIM to the start of the increase of the output current of the LED driver, and the amount of time tSU represents the time required for the output current to be converted from 0 to the target current, as for the time. The amount tSD represents the time required for the output current to transition from the target current back to zero. In most cases, the lower the dimming frequency fDIM, the higher the contrast, because these fixed delays only occupy a small portion of the dimming period TDIM. The lower limit of the dimming frequency fDIM is about 120 Hz. If it is lower than this frequency, the eyes can no longer mix the pulses into a continuous continuous light. The upper limit depends on the minimum contrast requirement, and the contrast is generally expressed as the reciprocal of the minimum on time.
CR=1 / tON-MIN: 1
tON-MIN =tD+tSU
Applications such as machine vision identification and industrial inspection often require higher PWM dimming frequencies, primarily because high-speed cameras and sensors respond much faster than human eyes. In such applications, the purpose of high-speed on and off of the LED source is not to reduce the average light output, but to synchronize the light output with the capture time of the sensor or camera.
Dimming with a switching regulator
In order to achieve switching hundreds of times per second or even thousands of times, switching regulator-based LED drivers require special design considerations. Regulators designed for standard power supplies typically have a "boot" or turn-off pin for logic PWM signals, but the associated delay tD is quite long because the silicon design emphasizes response time. Maintain low shutdown current inside. However, the switching regulator dedicated to driving the LEDs is just the opposite. It keeps the internal control circuit active when the "start" pin logic is low to minimize tD, and when the LED is turned off, Will face the trouble of large operating current.
When using PWM to achieve light control optimization, keep the Slew-up and Slew-down delays to a minimum, not only for optimal contrast, but also to reduce LED spend. The time required from 0 to the target. (Under this condition, the main wavelength or CCT is not guaranteed to be the same as the target value.) The standard switching regulator here will have a soft start, usually with a soft shutdown, and the dedicated LED driver will be in its control. All work is performed to reduce these slew rates. To reduce tSU and tSD, it is necessary to start with the design of the silicon chip and the topology used by the switching regulator.
The buck regulator with faster slew rate performs better than all other switching topologies in two places. First, the buck regulator is the only switch that delivers power to the output when the control switch is activated. Converter, this feature makes the control loop of a buck regulator in voltage mode or current mode PWM (not to be confused with PWM dimming) faster than a boost regulator or other buck/boost topology. In addition, the power transfer during the start of the control switch can be easily changed to hysteresis control, making it even faster than the optimal voltage mode or current mode controlled loop. Second, the inductor of the buck regulator is connected to the output during the entire switching cycle, which ensures continuity of the output current, which means that no output capacitor is needed. With fewer output capacitors, the buck regulator becomes a true high-impedance current source that can quickly convert the output voltage. While Cuk and Zeta converters offer continuous output inductors, they are not the best choice because of their slower control loops and lower efficiency.
PWM is faster than the "start" pin
Even a pure hysteresis buck regulator without an output capacitor is not sufficient for some PWM dimming systems. These applications require higher PWM dimming frequency, high contrast, which requires faster turnaround. Rate and a more short delay. When used in conjunction with mechanical vision and industrial inspection systems, some systems that require high performance, including liquid crystal (LCD) panels and backlighting systems for single-shot projectors, in some cases, the PWM dimming frequency must be adjusted. Up to the 25 kHz or higher band outside the audible band, as the overall dimming period has been shortened to within a few microseconds, including the conduction delay, the sum of the rise and fall times of the LED current must be shortened to within nanoseconds.
Starting with a fast buck regulator without an output capacitor, the delay in the output current turn-on and turn-off is the conduction delay from the integrated circuit itself and the physical characteristics of the output inductor. To achieve true high-speed PWM dimming, both delays must be skipped (By Pass). The best way to achieve this is to use a power switch in parallel with the LED (Figure 3). When the LED is turned off, the drive current is shunted through the switch, acting like a typical N-type metal oxide semiconductor field effect transistor (N-MOSFET), where the integrated circuit will continue to operate and the inductor current will continue to flow. The biggest disadvantage of this method is that when the LED is turned off, power is wasted even if the output voltage drops during the same period as the current sense voltage.
Figure 3 shunt FET circuit and its waveform
Dimming with a shunt field effect transistor (FET) results in a sharp shift in the output voltage, which causes the integrated circuit's control loop to respond in an attempt to maintain output current stability. Just as with logic pin dimming, the faster the control loop, the better the response, and the hysteresis controlled buck regulator provides the best response.
Fast PWM dimming with boost and buck/boost
Neither the boost regulator nor any type of buck/boost topology is suitable for PWM dimming. At the beginning of the design, it will be found that both will exhibit a Right-half Plane Zero limit in continuous conduction mode (CCM), which will not achieve the high control loop frequency required by the frequency regulator. Wide requirements. In addition, the time-domain effect of the right half-plane zero also makes it difficult for the system to hysteresis to control the boost or buck/boost circuit; another factor that complicates the situation is that the boost regulator cannot tolerate the output. The voltage drops below the input voltage, which can cause a short circuit at the input, making parallel FET dimming impossible. In addition, in various buck/boost topology techniques, parallel FET dimming is still difficult or extremely difficult to use, mainly because it requires output capacitors (SEPIC, buck/boost and flyback), or Uncontrollable input inductor currents (Cuk and Zeta) occur when the output is shorted.
If a fast PWM dimming is really needed, the best solution is to use a two-stage system with a buck regulator as the second stage LED driver stage. However, if size space and cost are not allowed, the best choice for the second best is the serial switch in Figure 4.
Figure 4: Boost regulator with serial dimmer switch
Although the LED current can be turned off in an instant, the response of the system must be carefully considered. This open circuit can be seen as a fast extreme off-load transient, which also interrupts the feedback loop and causes the regulator's output voltage to rise endlessly. Therefore, it is necessary to add a clamp circuit to the output and / or error amplifier to prevent damage caused by the overload voltage, but because these clamp circuits are difficult to implement by external circuits, that is, serial FET dimming must be combined with dedicated Boost and buck/boost LED driver ICs are available.
To effectively control the LED light source, care must be taken at the beginning of the design process. The more precise the light source, the greater the chance of PWM dimming, and the system designer must carefully consider the topology of the LED driver. Buck regulators have many advantages for PWM dimming, and designers must carefully consider the input voltage and the placement of the LEDs. If the dimming frequency is higher, the slew rate is faster, so it is easier to switch to a buck regulator at the beginning of the design process. (Edit: Looking up at the stars)
We manufacture and supply flexi-rigid PCBs, Flexible Circuit boards and Polyimide Flexible Circuit Assembly . Flexible Circuits are constructed by sequential build up of flexible component layers, such as base polyimide, copper, cover-lay, and stiffeners bonded by an adhesive layer (epoxy or acrylic)
Polyimide Flexible Circuit Assembly
FPC Flexible PCB Assembly,PI Flexible PCB Assembly,Polymide FPC Assembly
Hemeixin Electronics Co.,Ltd. , http://www.rigidflex-pcb.com