你好，在Direct PWM Dimming模式下，我想用DIM/PWM pin输入5V 1Hz的PWM信号，实现LED可视的闪烁功能，因此我在J1输出接了四串LED白灯。但是除了第一个周期的亮灯正常，之后每次亮灯前LED都会瞬间亮一下再正常亮，通过示波器看IMON引脚波形，我发现IMON在PWM上升时会有一个大脉冲，然后IMON输出为0，等待35ms左右有一个小脉冲，然后再过5ms左右输出正常波形，如图1所示。我尝试了1-4Hz的PWM信号，其中1-3Hz会出现上述现象，4Hz时虽然仍然有浪涌电流，但是不会出现IMON断续的情况，如图2所示。
我猜想是因为浪涌电流导致hiccup模式激活，因此我打算用示波器测量SS引脚的电压，但是当我把示波器探头接到TP2时，IMON输出变正常了，如图3所示。
因此我在Css两端并联了一个0.1uF的电容，再次观察实验现象。
当我PWM输入为5V 3Hz的PWM信号时，可以看出区别。
下面分别是5v 3Hz的PWM信号输入，测量PWM，IMON和SS，如图4所示。5v 3Hz的PWM信号输入，在Css两端并联了一个0.1uF的电容，测量PWM，IMON和SS，如图5所示。可以看出由于hiccup模式激活，Css电压降为0V，因此36ms后，由于软开关的存在，使得浪涌电流变小，电路不会进入hiccup模式，电路正常运行，之后重复上述结果。
请问有什么方法可以解决上述问题。
Hello, In Direct PWM Dimming mode, I want to use DIM/PWM pin to input 5V 1Hz PWM signal to realize visible flashing function of LED, so I connected four strings of white LED to J1 output. But except for the normal lighting of the first cycle, the LED will light up for a moment before lighting up again. Looking at the IMON pin waveform through the oscilloscope, I found that IMON will have a large pulse when PWM rises, and then IMON output is 0, wait for about 35ms to have a small pulse, and then output the normal waveform after about 5ms, as shown in Figure 1. I tried 1-4Hz PWM signals. Among them, the above phenomenon will occur at 1-3Hz. Although there is still surge current at 4Hz, IMON will not be intermittent, as shown in Figures 2. I guess it is because of the surge current that the hiccup mode is activated, so I plan to use an oscilloscope to measure the voltage of the SS pin, but when I connect the oscilloscope probe to TP2, the IMON output becomes normal, as shown in Figure 3. Therefore, I connected a 0.1uF capacitor in parallel across Css and observed the experimental phenomenon again. When I input 5V 3Hz PWM signal, the difference can be seen. Below are 5v 3Hz PWM signal input, PWM, IMON and SS are measured, as shown in Figure 4. 5v 3Hz PWM signal input, a 0.1uF capacitor is connected in parallel across Css, PWM, IMON and SS are measured, as shown in Figure 5. It can be seen that due to the activation of hiccup mode, the Css voltage drops to 0V, so after 36ms, due to the existence of soft switching, the surge current becomes smaller, the circuit will not enter the hiccup mode, the circuit runs normally, and then repeats the above results. Is there any way to solve the above problem?
Florens, Liang
Fig.1
Fig.2
Fig.3
Fig.4
Fig.5