• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 680 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.

LM5035: Half-bridge dcdc output power based on LM5035 is slightly higher and unstable

zoujiangyilang
Genius 4560 points
Part Number: LM5035
Other Parts Discussed in Thread: LM158

Hello, I simulated your LM5035 development board and made a 36V input 10V output dcdc power supply.
Tests are normal when three power supply prototypes are produced. The output is about 9.98V.
However, in the last three days, I repeatedly conducted high temperature tests and frequency response tests on these three prototypes.

Found that 3 prototypes are abnormal. The output voltage floats to about 10.3V to 11V, and the output voltage ripple is about 200mV, and it is disorderly. Normally, the output voltage ripple is about 80mV and is regular.
I measured the positive input of the secondary op amp LM158, the 2.5V reference, which is stable and accurate, about 2,495V. I measured the positive input of the secondary operational amplifier LM158, that is, the voltage across the voltage divider sampling resistor, which averaged about 2.5V, but was fluctuating and unstable.
I don't know if the operational amplifier LM158 is damaged after repeated high temperature tests and frequency response tests, resulting in 2.5V instability at the positive terminal and 10V output instability.

How can I troubleshoot this failure in the next step?

Thank you.

  • 0
    TI__Expert 3095 points
    Hi Zoujiangyilang ,
    Is the output voltage always high regardless of load?
    What is the current flowing into the COMP pin, is this stable and in the correct range? You should be able to observe this if you place a small resistor in series with the collector of the opto-transistor and observe the voltage drop across this resistor with a differential probe.
    Some waveforms of output voltage and COMP current may help us to assist you. Also a schematic always helps.
    Look for the COMP pin current having a repetitive signal of fixed frequency - suggesting a stability problem.
    Look for the COMP pin current being at one extreme end of its range - suggesting that the control loop is unable to achieve the power level required to regulate the output voltage.
    I hope this helps
    Joe Leisten
  • 0 in reply to Joe Leisten
    Genius 4560 points
    All three samples in my hand showed this phenomenon of output shift and ripple fluctuation.
    I replaced the optocoupler of the three samples, in which two samples returned to normal and the other sample failed.
    After I removed the soldered test leads from this sample + resistance, the third sample also returned to normal.
    I suspect that the optocoupler is damaged because the photocoupler was blown with a 200-degree hot air gun and the load step was observed.
    I think the hot air may have caused damage to the optocoupler.
    However, I cannot understand why the test line soldered to the sence+ resistor will also affect the output voltage. It is just a test line that is used to facilitate the injection of the CH1 signal when testing the Bode plot. I am very puzzled by this.
    Thank you
  • 0 in reply to Joe Leisten
    Genius 4560 points
    I have another question I want to ask.
    When I debug the half-bridge circuit based on LM5035, I found that the load jump amplitude is relatively large, about 450mV, and I want to reduce this jump amplitude to 350mV.
    What aspects should I start with to reduce the load jump amplitude?
  • 0 in reply to zoujiangyilang
    TI__Expert 3095 points
    Hi Zoujiangyilang ,
    I am not quite sure what you mean by load jump amplitude.
    Do you mean the output voltage deviation that occurs in response to a load current step?
    There are a number of ways to improve transient response. The best method to use depends upon the cause of the step. Please post a scope plot of your output voltage transient response waveform to help us understand the best approach to recommend.
    Please refer to this document.
    e2e.ti.com/.../how-to-determine-bandwidth-from-the-transient-response-measurement
    Thanks
    Joe Leisten