• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 62 subscribers
  • Views 6277 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Power management forum

yuan cheng
Prodigy 40 points
Part Number: LM5041A
Other Parts Discussed in Thread: LM5041,

Use LM5041 as a current-fed push-pull topology, use an external error amplifier, disable the internal error amplifier, and ground the FB pin of LM5041; in the initial stage of power-up, the output voltage of the error amplifier and the REF pin voltage rise at the same time, causing the REF pin to output The voltage is only about 1.86V, but cannot reach 5V, causing the entire chip to fail to work normally. May I ask what causes this? How to deal with it?

This is the voltage waveform of the REF pin and the COMP pin when the power is turned on. Yellow waveform: REF pin voltage waveform. Blue waveform: COMP pin voltage waveform. Purple waveform: The output waveform at pin 1 of the error amplifier. It can be seen that after power-on, the COMP voltage has already been established, while the REF voltage has been established relatively late. After C42 is increased to 1uF, it can start normally, but C42 cannot be designed to such a large capacity, which affects the control loop. You help me analyze what is the problem? I feel that when the LM5041 is powered on, there are certain requirements, but they are not written in the datasheet. In addition, after this circuit is powered on, short-circuit the COMP pin to ground for a short time, the REF pin voltage can immediately rise to 5V, and the chip can work normally.

Yellow waveform: +11V main power supply waveform. The UVLO threshold voltage of LM5041 is designed to be 12.3V. Blue waveform: REF pin voltage waveform. Purple waveform: COMP pin voltage waveform. The +11V power supply is a BUCK-type DC DC step-down power supply. It can be seen from the figure that the startup process is slightly slower. The power supply part of the circuit in the figure below, the power supply of the error amplifier comes from +6VA.

  • +1
    TI__Expert 7900 points

    Hello Yuan,

    Without having access to the internal schematics of the LM5041A and the help of an IC Design engineer it’s difficult to identify the root cause. I suspect the external op amp is turning on some internal junction(s), like a protection diode or parasitic PN junction, which causes the REF pin to get stuck at 1.86V.

    Anyway, how can we fix it? In the second picture I see the op amp output at 4.5V (purple) but the COMP pin (blue) is at only 2.1V. Therefore, the op amp is sourcing (4.5V – 2.1V) / 100 ohms = 24mA. This is quite a bit of current for the unbiased LM5041 to sink. I would increase the 100 ohm resistor to 1k (or 10k) to see if that resolves the issue.

    Also, the LM5041A data sheet recommends a 0.1uF capacitor to bypass the REF pin. Your schematic shows 1uF at the REF pin. Try reducing this to 0.1uF.

    Lastly, the SS pin has a 0.1uF capacitor on it. This delays the Enable signal to the VCC/VREF regulators for about 4.5ms (0.1uF*0.45V/10uA). Reduce the SS capacitor to 10nF, or less, so the VCC/VREF regulators will turn on sooner.

     

    If none of these methods resolve your issue then I would recommend keeping the op amp output low by either 1) delaying the U-CTRL and HV-2-OV signals, or 2) keeping the (+) pin to the op amp low via R41/D28, so the LM5041 has time to come out of UVLO.  Similar to the effect of changing C42 to 1uF.

    Good luck, and thank you for using TI’s controllers.

    Regards,

    Eric