1, chip fever
This is mainly for high-voltage driver chip with built-in power modulator. If the chip current consumption of 2mA, 300V voltage added to the chip above the chip's power consumption of 0.6W, of course, will cause the chip fever. The maximum current of the driver chip comes from the consumption of the driving power MOS tube. The simple calculation formula is I = cvf (Considering the resistance effect of charging, the actual I = 2cvf, where c is the cgs capacitance of the power MOS tube, v is the power tube conduction Gate voltage, so in order to reduce the chip's power consumption, we must think of ways to reduce c, v and f. If c, v and f can not be changed, then please find ways to divide the power consumption of the chip off-chip devices, Introduce additional power consumption. And then simpler, is to consider better cooling it.
2, the power tube fever
On this issue, but also seen someone posted in the power grid forum posted. Power tube power consumption is divided into two parts, switching losses and conduction losses. It should be noted that in most cases, especially for LED electric drive applications, the switching loss is much larger than the conduction loss. Switching losses and power tube cgd and cgs and the chip's drive capability and operating frequency, so to solve the power tube heat can be resolved from the following aspects: A, can not be based on the size of on-resistance MOS transistor selection, because The smaller the internal resistance, the larger the cgs and cgd capacitors.
Such as 1N60 cgs is about 250pF, 2N60 cgs is about 350pF, 5N60 cgs is about 1200pF, the difference is too big, choose the power tube, enough to use it. B, the rest is the frequency and chip driver capabilities, and here only talk about the impact of frequency. Frequency and conduction loss is also proportional, so the power tube fever, the first thing to think about is not a bit high frequency selection. Find ways to reduce the frequency of it! Note, however, that when the frequency is reduced, the peak current must be larger or the inductance larger in order to obtain the same load capacity, which may cause the inductor to enter the saturation region. If the inductor saturation current is large enough, consider changing the CCM (continuous current mode) to DCM (discontinuous current mode), which requires an additional load capacitance.
3, the working frequency down
This is also the user in the debugging process more common phenomenon, down mainly caused by two aspects. The ratio of input voltage and load voltage is small, the system interference. For the former, be careful not to set the load voltage too high, although the load voltage is high, the efficiency will be high. For the latter, you can try the following aspects: a, the minimum current set point again; b, wiring clean, especially the sense of this critical path; c, the inductor selection of small points or use closed magnetic circuit inductance ; D, plus RC low-pass filter it, the impact is a bit bad, C's consistency is not good, the deviation is a bit large, but for the lighting should be enough. In any case there is no benefit down-conversion, only the downside, so be sure to solve.
4, inductance or transformer choice
Finally talked about the key, and I have not got started, can only talk nonsense saturation. Many users respond to the same drive circuit, with a production of the inductor is no problem, with the inductor current b smaller. In this case, look at the inductor current waveform. Some engineers did not notice this phenomenon, directly adjust the sense resistor or the operating frequency to achieve the required current, doing so may seriously affect the life of the LED. Therefore, before the design, reasonable calculation is necessary, if the theoretical calculation parameters and debug parameters a bit far, we must consider whether the frequency down and transformer saturation. When the transformer is saturated, L will be smaller, leading to a sharp increase in peak current caused by the transmission delay, then the LED peak current also increases. Under the premise of constant average current, we can only watch the light fades.
5, LED current size
We all know LEDripple too big, LED life will be affected, how much has not seen which experts have said. LED factory before asked this data, they say that within 30% are acceptable, but then have not been verified. Advice or try to control the dots. If the cooling solution is not good, LED must derating use. Also hope that experts can give a specific indicator, or otherwise affect the promotion of LED.