• State TI Thinks Resolved
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 67 views
  • Users 0 members are here
Support feedback
Options
Options
Related

UCD3138: PWM waveform cannot be expanded according to normal feedback logic

user6101889
Prodigy 40 points
Part Number: UCD3138

Tool/software:

Hi TI gays,

One of our mass-produced products using the UCD3138 experienced a problem with the PWM waveform not expanding properly during customer application, resulting in abnormal output voltage. After receiving the module in question, we were able to replicate the issue and further test the module. We found that the module's hardware circuitry was functioning properly; it was only the PWM waveform that failed to expand after the enable function was enabled that caused the abnormal output.

Do you have any suggestions or testing methods to determine the problem? Customer complaint: The product cannot blow down UCD3138 temporarily.

Thanks,

Ezra

  • 0
    TI__Genius 9680 points

    Hello Ezra,

    I will provide a response later today

    Regards,

    Jonathan Wong

  • 0 in reply to Jonathan Wong
    TI__Genius 9680 points

    Hello Ezra,

    Do you have any waveforms to help me understand the issue? Also, what is the expected PWM duty cycle vs. the actual PWM duty cycle?

    Regards,

    Jonathan Wong

  • 0 in reply to Jonathan Wong
    Prodigy 40 points

    Hi Jonathan

    Thank you for your replied.

    This is the waveform comparison between the faulty module and the normal module.

    This is the process of turning on the computer after I became paranoid.

    You can see that at the beginning PWM outputs according to the duty set by the bias, and then gradually changes to the minimum duty.

    Thanks,

    Ezra

  • 0 in reply to user6101889
    TI__Genius 9680 points

    Hello Ezra,

    The third waveform looks like something is wrong in the hardware. The waveform looks noisy and shaky, which indicates a weak signal. You have confirmed that the customer's hardware is correct and it is only the UCD that is causing the issue? How often do these faults occur?

    One thing you can do is make sure that the UCD3138 is receiving the right input information. You can use the Fusion GUI's memory debugger to find values such as "EADCRAWVALUE". Compare the bad UCD3138 reading vs. the good UCD3138 reading. An example of how to read the value is shown below.

    Have you submitted a customer return for the part?

    Regards,

    Jonathan Wong

  • 0 in reply to Jonathan Wong
    Prodigy 40 points

    Hi Jonathan,

    Thank you for your reply.

    We have confirmed that there are no issues with the power stage hardware and the circuit is normal, but the UCD PWM waveform is not displayed properly. This is the first time this problem has occurred on our client.

    II cannot get the stored information by reading the register. In our mass production mode, we have commented out the register reading function.

    We have not submitted this UCD yet. Our material engineer is communicating with the corresponding sales window about submitting the application.

    Thanks,

    Ezra

  • 0 in reply to user6101889
    TI__Genius 9680 points

    Hi Ezra,

    Are you able to read the UCD at all, such as reading the Device ID or reading the device in ROM mode? Without communicating to the device, it will be difficult to know if it is working properly.

    If this is the first time this has happened, then it may be a device failure. Whether failure is due to TI manufacturing or external damage is unknown without more information. Was this a new UCD device or has the UCD device been in the field for a while? What is the failure rate of the UCD?

    You can submit a request to return the part. However, we must be able to communicate to the UCD3138 in ROM mode for a successful test. Please make sure the UCD3138 is in ROM mode before you return the part.

    Regards,

    Jonathan Wong