Energy-saving LED lighting system based on EZDimming

introduction

Lighting designers must first consider the luminous intensity (measured in candela, cd) or luminous flux, and also consider the illuminance (measured in lux, lx) received by the illuminated object or illuminated plane. If we have a 10W LED desk lamp, the total luminous flux emitted is 600lm. If all of this 600lm is concentrated on a 1m2 desktop, then the illuminance of the desktop is 600lx.

Different lighting locations and different public places have different lighting requirements. For example, in store lighting, bright food can stimulate appetite; bright illumination of jewelry, clocks, clothing, etc. can stimulate the desire to buy; color printing, galleries, museums, clinics, etc. must have an illumination of 1000lx. The lighting of offices, classrooms, factories, production lines, etc. requires 300~800lx. Non-working places such as stations, airports, building halls, corridors, etc. can have 100~300lx. In parks, parking lots, and streets, it can be as low as 10~50lx. Use different light sources according to actual needs. Good lighting design not only saves energy and electricity, but also meets environmental protection requirements.

1 The role and significance of dimming

Lights are artificial lighting that provide sufficient illumination at night or in places where natural light is insufficient. However, lights consume electricity, and turning off lights when not needed can save energy. However, in some periods and situations, if the lights can be dimmed to a darker level (25%~50%), energy consumption can be significantly reduced (-50%~-75%).

According to incomplete statistics, lighting loads account for about 20% of the total electricity consumption in China, which means that lighting consumes about 20% of the power of the power generation system. Lighting products with dimmable functions are more energy-efficient than those without dimming functions. The function of dimmers is to adjust the illumination and luminous flux. Therefore, in lighting design, giving full play to the function of dimming can completely achieve the goal of substantial energy saving.

2 Existing dimming methods

2.1 Resistive Dimming

In the past, the most primitive resistor dimming was used, as shown in Figure 1, where a variable resistor was used to dim the lighting circuit. The application is very simple and does not cause interference, but its voltage division principle is to prevent the electric energy from being completely used in the lamp (electrical appliance), which is inefficient. When dimming the light, the dimming resistor generates a lot of heat due to excessive voltage division, resulting in energy waste and environmental degradation. Therefore, the thyristor dimming method was used instead.

Figure 1 Resistive dimming

2.2 Thyristor Dimming

As shown in Figure 2, the basic principle of thyristor control technology is: using RC phase shift to delay triggering, achieve chopping, and change the root mean square value of the voltage (effective value is Vrms). At the same time, the symmetry of the conduction voltage of the bidirectional thyristor (DIAC) is used to achieve symmetrical triggering of the thyristor (TRIAC). Therefore, changing the resistance value can obtain different conduction angles, and applying the phase cutting principle to reduce Vrms can reduce the power of ordinary loads (resistance loads).

Figure 2: Changing the resistance value can get different conduction angles

Therefore, the efficiency is high and the performance is stable. Figure 3 is a working principle diagram of the thyristor dimmer.

Figure 3: Working principle of thyristor dimming

The characteristics of thyristor technology in operation (IV characteristics) are shown in Figure 4. According to the IV characteristics, after TRIAC is turned on, it needs to maintain a minimum current (holding current Ihold) to keep it on, otherwise it will return to the cut-off state; different TRIACs have different Ihold values, usually between a few mA and tens of mA, and some even reach 50mA to stably conduct. Different brands and types of dimmers available on the market now include 100W, 200W, 600W~1000W. The greater the power, the greater the holding current required. If there is not enough Ihold to ensure a stable conduction angle of TRIAC, the output waveform will be uneven, resulting in additional jitter and spikes, affecting the output Vrms.

Figure 4 Characteristics of thyristor technology during operation (IV characteristics)

2.3 Common dimming methods for LED lighting

There are usually three methods for LED lighting dimming:

(1) Linear dimming;

(2) Pulse width (PWM) dimming;

(3) Thyristor dimming.

Among them, linear dimming is shown in Figure 5. By adjusting the current sensing voltage threshold, the size of the current flowing through the LED can be adjusted. In Figure 5, the current sensing voltage threshold of SSD1075 is 1.2V. We input a DC voltage of 0~1.2V to its LD pin to achieve linear regulation of the LED current. The actual measurement of the dimming curve of SSD1075 shows that when 0~1.2V is input to LD, the LED current is linearly regulated. The disadvantage of linear dimming is that an additional DC voltage source is required. In addition, since the current sensing voltage threshold is generally set low, its anti-interference ability is poor and flickering will occur.

Figure 5 Linear dimming

Pulse width PWM dimming is shown in Figure 6. By changing the duty cycle of the PWM input pulse signal to modulate the gate control signal of the LED driver chip to the power field effect tube, the current through the LED can be adjusted. The measured dimming curve of SSD1075 shows that when the duty cycle changes from 0 to 100%, the LED current is linearly adjusted. The disadvantage of this method is that it requires an additional PWM signal source, and the closer the pulse width modulation signal frequency is to the gate control signal frequency of the LED driver chip to the power FET, the worse its linearity.

Figure 6 Pulse width PWM dimming

The biggest problem with linear dimming and PWM dimming in home and commercial applications is not the trouble of adding load voltage divider circuits or generating control pulse signals, but because both methods require control circuits. Traditional lamps only have two electrical contacts. If control lines are required, replacement modification cannot be achieved, so LED lamps must be compatible with existing SCR dimming control technology.

Thyristor control is to extract the conduction angle information from the phase-cut power supply in the chip or application circuit, and adjust the driving current of the LED according to the signal control to achieve the dimming effect.

According to the thyristor dimming curve of SSD1075, it can be seen that the LED current and thyristor dimming are linearly adjusted, as shown in Figure 7.

Figure 7 thyristor dimming principle diagram and dimming curve

3 Problems and solutions of thyristor dimming in application

3.1 Thyristor dimming will cause 100Hz flicker

As shown in Figure 8, the power frequency of the power grid is 50Hz. After the phase-cut waveform is rectified, a 100Hz pulse signal can be obtained, which can be directly used to generate a dimming PWM signal. However, in actual applications, any slight voltage fluctuation or current change will affect the duty cycle of the pulse, which is equivalent to a 100Hz flicker. Therefore, the phase-cut waveform cannot be used directly for dimming. The modulation signal needs to be shifted to a higher frequency band so that the human eye cannot detect the flicker phenomenon. The specific implementation is to adjust the dimming signal to a higher frequency band through SSD1075, as shown in Figure 8.

Figure 8 100Hz flicker and its solution

3.2 Load issues of thyristor dimming

For the thyristor dimmer to work stably, a stable Ihold is required to ensure the dimmer works. However, most LED power supplies are nonlinear switching power supplies. In addition, the current LED is very energy-saving and the power of the whole lamp is not high, so the working state of TRIAC is unstable. Since the conduction of the thyristor requires a certain current, and the LED load is not linear, the required input current is variable, so the conduction state and trigger voltage of the thyristor will also change as shown in Figure 9. The solution is to add a virtual load before the SCR signal input terminal to generate a stable 10~30mA load for TRAIC. As can be seen from Figure 10, the waveform of the SCR has been significantly improved.

Figure 9 Changes in the conduction state and trigger voltage of the thyristor

Figure 10 Adding a dummy load significantly improves the waveform

3.3 Load overheating problem of thyristor dimming

As mentioned above, after adding the virtual load, the power consumption is calculated as 10~30mA, and the power consumption is P=2500.01~2500.03=2.5W~7.5W. It is conceivable that a 7.5W virtual load is not feasible to be placed in an E27 bulb, and the efficiency is also very low. Therefore, the virtual load current must be as small as possible (I10mA). In addition, the conduction angle signal must be accurately analyzed, so it is necessary to apply signal filtering technology to shape and filter the intercepted signal in order to improve the availability of the dimming signal.

3.4 Disadvantages of thyristor dimmer

Will silicon-controlled dimmers become the mainstream dimming method for LED lighting in the future? The answer is quite pessimistic. The main disadvantages of silicon-controlled dimmers are as follows.

(1) Power factor degradation

The principle of this dimming method is to use the silicon controlled tube, which is the main component, to intercept the conduction angle of the AC voltage to change the input power of the lamp. However, the shorter the conduction angle, the worse the power factor. When adjusted to 1/4 brightness, the power factor will be less than 0.25.

(2) Decreased efficiency

Because when dimming, insufficient load will cause the silicon controlled tube to shut down early, causing the adjustable resistor used to change the conduction angle of the silicon controlled tube to be confused with the conduction angle, further leading to the problem of unstable operation and flickering of LED lights. For this reason, it is usually solved by adding a bleeder resistor (bleeder dummy load). In this way, using bleeders to consume power to maintain the normal operation of the silicon controlled tube violates the original intention of dimming to save energy.

(3) Difficult to match

It is quite difficult to develop a dimmable LED lamp that is compatible with dozens of silicon-controlled dimmers on the market. For matching issues, please refer to NxpApplicationNoteAn10754.

(4) High installation cost and limited application market

In view of the above reasons, if LED lighting industry can create a dimming method that is close to zero cost, does not require wiring changes, does not require expensive remote control dimmers, and can maintain high efficiency and high power factor, and can be easily promoted and popularized, this will be very important for LED lighting to achieve energy saving and open up a broad application market.

4 Ideal driver IC for LED lighting dimming

For a traditional industry like lighting, the most feasible standard is the original standard; the easiest solution is not to change the status quo. So far, the more successful LED lighting products are very similar in shape and usage to traditional bulbs and tubes. Therefore, it can be inferred that the most feasible dimming method is a solution that does not formulate new standards, does not change the status quo, does not require changing wiring, and does not add any dimming devices to achieve the dimming function.

EZDimming is an ideal dimming method that builds the dimming function into the LED driver integrated circuit according to the basic concept of four-stage dimming. No other external parts are required, and high efficiency and high power factor can be maintained at each dimming level. Figure 11 shows the driver IC for LED lighting dimming.

Figure 11 High power factor, high efficiency, driver integrated circuit with dimming function

EZDimming detects the number of times the wallswitch switches to change the dimming level. It starts from full brightness and gradually decreases to six-point brightness, four-point brightness, and two-point brightness. And in any dimming stage, it can maintain a power factor above 0.90 and an efficiency above 80%. The switching is soft and smooth, and there is no adverse phenomenon of surge current or noise. The circuit is streamlined, and the hardware cost is almost zero. Compared with other existing dimming solutions, EZDimming not only has transcendent functional advantages, but also does not require changes to wiring or any hardware, and has the lowest installation cost. Just remove the traditional light bulbs, fluorescent tubes or energy-saving bulbs, and then install LED lights equipped with EZDimming, and you can immediately change a lighting fixture or system that originally had no dimming function into a lighting system with dimming function that is both energy-saving and environmentally friendly.