Explain the whole process of semiconductor madman making LED fluorescent lamp power supply

Introduction to non-isolated buck power supply design method

The non-isolated step-down power supply is a commonly used power supply structure, accounting for almost 90% of the fluorescent lamp power supply. Many people think that the non-isolated power supply is only a kind of buck type. Whenever it is said that it is not isolated, it thinks of the buck type, and it is thought that the lamp is not safe (refer to the power supply damage). In fact, the buck type is not only one type, there are two basic structures, namely boost, and buck-boost, ie BOOSTANDBUCK-BOOST, even if the latter two power supplies are damaged. Will not affect the benefits of LED. The buck power supply also has its advantages. It is suitable for 220, but it is not suitable for 110. Because the voltage of 110V is low, the drop is lower, so the output current is large, the voltage is low, and the efficiency is not too high. The buck type 220V AC is about 300V after rectification and filtering. After the step-down circuit, the voltage is generally reduced to about 150V DC, so that high-voltage and small-current output can be realized, and the efficiency can be made higher. Generally, MOS is used as a switch tube to make power of this specification. My experience is that it can be as close as 90%, and it is difficult to go up. The reason is very simple, the chip generally has a self-loss of 0.5W to 1W, and the fluorescent tube power supply is only about 10W. So it is impossible to go up again. Now that power efficiency is a virtual thing, many people are blowing, and the actual situation is not at all.

Some people often say that 3W's power efficiency is 85 percent, and it is still isolated. Tell everyone that even in the frequency hopping mode, the no-load power consumption is the smallest, it is 0.3W, and what is the output 3W low voltage, which can reach 85%, in fact, 70% is very good, anyway now Many people brag about not drafting drafts, can fool the layman, but now do not know much about LED power.

I said that to be efficient, we must first do non-isolated, and then output specifications and high voltage and small current, which can save the conduction loss of power components, so the main loss of such LED power supply, one is the chip itself. Loss, this loss generally has a fraction of a few W to a W, and one is the switching loss. Using MOS as the switching transistor can significantly reduce this loss, and the loss of the triode is much larger. So try not to use a triode. There is also a small power supply, it is best not to save too much, do not use RCC, because the general manufacturers of RCC circuits simply do not do quality, in fact, the chip is also cheap, ordinary switching power supply chip, integrated MOS tube, up to two yuan Money, there is no need to save a little bit, RCC only saves material costs, in fact, the cost of processing and repairing is higher, and it is not worth the loss.

The basic structure of the buck power supply is to string the inductor and load into the high voltage of 300V. When the switch is switched, the load is lower than 300V. There are many specific circuits and many online. I will not draw a picture. Now 9910, there are generally constant current ICs on the market that are basically implemented with this circuit. But this kind of circuit is when the switch tube breaks down, the whole LED light board is finished, which should be regarded as the worst place. Because when the switch tube breaks down, the entire 300V voltage is applied to the lamp board. The lamp board can only withstand more than one hundred volts, and now it is three hundred volts. This happens. The LED must be burned. Therefore, many people say that non-isolation is not safe. In fact, it means bucking, just because the vast majority of non-isolated is buck, so non-isolated damage must be considered bad. In fact, the other two basic non-isolated Structure, power supply damage, will not affect the LED.

The buck power supply should be designed as a high voltage and small current, and the efficiency can be high. Explain, why? Because of the high voltage and small current, the pulse width of the switching tube current can be made larger, so that the peak current is smaller, and the loss of the inductance is also smaller. It can be known through the circuit structure that the circuit is inconvenient to draw, and it is difficult to be specific. Let's go on again. Just sum up, the advantage of the step-down power supply is that it is suitable for 220 high-voltage input, so that the voltage stress of the power device is small, suitable for high-current output, such as 100MA current, which is easier and more efficient than the latter two methods. Be high. The efficiency is relatively high, the loss on the inductor is small, but the loss on the switch tube is larger, because all the power through the load must be transmitted through the switch tube, but the output power is only partially passed through the inductor, such as 300V input, 120V output. The step-down power supply, only the 180V part has to pass through the inductor, and the 120V part is directly turned into the load, so the inductance loss is relatively small, but the output power is all converted by the switch.

Decompose two constant current control methods

The following is to say, two switching modes of constant current control mode, resulting in two practices. These two approaches, whether they are principles, device applications, or performance differences, are quite large.

First of all, the principle. The first one is represented by the current constant current LED dedicated IC, mainly like the 9910 series, AMC7150. All the brands that currently use the LED constant current drive IC are basically this kind, and call him constant current IC type. But I think this kind of constant current IC is doing constant current, but the effect is not so good. The control principle is relatively simple. It is to set a current threshold in the primary circuit of the power supply. When the primary MOS is turned on, the current of the inductor rises linearly. When it rises to a certain value, When this threshold is reached, the current is turned off and the next cycle is triggered by the trigger circuit. In fact, this constant current should be a current limit. We know that when the inductance is different, the shape of the primary current is different. Although there are the same peaks, the current average is different. Therefore, when such a power source is generally mass-produced, the consistency of the constant current size is not well controlled. There is also a characteristic of such a power supply, generally the output current is trapezoidal, that is, the wave current, the output is generally not electrolytic smooth, which is also a problem, if the current peak is too large, it will affect the LED. This is basically the case if the output stage of the power supply does not have a power source that is electrolyzed to smooth the current. That is to determine whether this is the control method, it depends on whether the output is electrolytically filtered. This constant current I used to call it a pseudo constant current, because its essence is a current limiting, not the constant current value obtained by the op amp comparison.

The second constant current mode should be called switching power supply. This control method is similar to the constant voltage control method of the switching power supply. Everyone knows to use TL431 for constant voltage because it has a 2.5 volt reference inside and then uses a resistor divider. When the output voltage is a little higher, or lower, a comparison voltage is generated. After amplification, the PWM signal is controlled, so this control method can control the voltage very accurately. This kind of control requires a benchmark and an op amp. If the benchmark is accurate and the op amp is large enough, then it is accurate. Similarly, to do constant current, you need a constant current reference, an op amp, with resistance overcurrent detection, as a signal, and then use this signal to amplify, to control the PWM, but unfortunately it is not very good to find a very accurate reference signal, Commonly used are triodes, this is the benchmark temperature drift, and the diode can be used as a reference for the conduction value of about 1V. This can also be used, but it is not high. The best is to use the op amp plus TL431 as the benchmark, but The circuit is complicated. However, the constant current power supply with such constant current accuracy is still much better controlled. The constant current controlled by this mode must be electrolytically filtered, so the output power is smooth DC, not pulsating. If it is pulsed, it cannot be sampled. So to determine which one is only to see if its output has electrolysis.