• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 70 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.

UC3846: The application issues of chips

Tonyx.wu
Genius 3945 points
Part Number: UC3846

Tool/software:

This is an old project that has been around since 27 years ago without any design changes.  During the production of WSM-400T this time, there was a problem of the large current being too small.  The technical requirement was 400A, but the actual output was only 392A.  Out of 500 units produced, 200 were defective.



After research and development analysis, it has been confirmed that the problem is caused by the abnormality of UC3846.  Currently, the initial analysis shows that the output pulse width of UC3846 has decreased.  What is the problem and how should it be solved?


The A-B-A cross-validation has been conducted, and the poor follow IC.




Actual test result:

Bad COMP 3.1165V


Ok.  COMP 3.14



Ok.  CS+ 712.5mV

Bad CS+ are basically the same



Bad Vref 4.93V

Ok.  Vref 4.98V

It is suspected that the fluctuation range of Vref is too large, causing the output current to decrease




Modifying the four resistors 2R49 to 2R52 can increase the output.  However, the resistance values of these four resistors sometimes need to decrease and sometimes need to increase in order to make the output reach around 400A;  otherwise, it will reach around 390A.  Can this anomaly be determined that the fluctuation range of Vref is too large?  Could there be any other reasons and solutions

  • 0
    TI__Guru 73535 points

    Hi,

    If this is a 27-year project, and you confirmed the issue due to the IC duty cycle smaller, then you need to determine if the duty cycle still meets the datasheet min specs. If so then you need tune up your design as your design needs to cover the min specs.

    If you confirmed the IC does not meet the datasheet specs then you need to initiate a return process to report this issue.

  • 0 in reply to Hong Huang
    Genius 3945 points

    Sorry, this is a design from 2017




    Now it's the same design, but the duty cycle is sometimes large and sometimes small. The A-B-A test found that it follows the chip. Besides, the Vrefs of different chips are A bit different. I want to know if this duty cycle is affected by the Vref or what else you think it is?

  • 0 in reply to Tonyx.wu
    TI__Guru 73535 points

    Hi,

    The duty cycle varying usually means the feedback loop not stable. VREF noise can make this as well. You can measure the VREF and the bode plots to find out.