• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 102 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.

DS90UB941AS-Q1: How to adjust the drive strength of FPD-LINK of DS90UB941

?? ?
Prodigy 20 points
Part Number: DS90UB941AS-Q1

Tool/software:

Dear Expert

The head unit and the screen communication via FPD-LINK. The serializer is DS90UB941, and the deserializer is DS90UB948.

In the EMC RE test, the following frequency points exceeded the limit value.
400 MHz, 450 MHz, and 845 MHz.

Spectrum analyzer on the head unit, the exceeded noise from the FPD-LINK between DS90UB941 and DS90UB947.

Question: Is it possible to adjust the FPD-LINK drive strength through software? If so, which registers control this setting?

Thank you !

  • 0
    TI__Expert 3778 points

    Hi customer,

    Thank you for your questions.

    Some data that will help us better support you:

    • Please provide RE data.
    • What is the PCLK you are operating at?
    ?? ? said:
    Question: Is it possible to adjust the FPD-LINK drive strength through software? If so, which registers control this setting?

    In regards to your question, lowering the VOD will help improve RE performance. However, all of our device testing and validation is done on the default VOD setting.

    I would recommend the customer should perform their own validation and stress testing to ensure the lower VOD setting is robust for their application.

    Below is the procedure to modify VOD. (Default VOD =1.1Vppk, new setting=0.8Vppk):

    1. Write reg 0x40=0x14
    2. Write reg 0x41=0x58
    3. Write reg 0x42 = 0x02 (0.8Vppk)
      1. default value is 0x42=0x00 (1.1Vppk)

    Thanks,

    Ragav Subramanian

  • 0 in reply to Ragav Subramanian
    Prodigy 20 points

    Hi Ragav

    Thank you for your feedback.

    1. below table is the RE data

    2. The PCLK of screnn is 89MHz

    3. Here are two questions.

    (1) One is about your suggestion of register. I study the register of spec of DS90UB941. These registers don't seem to be related to VOD level?

    (2) Regarding the noise point at 445MHz, it is 5 times the PCLK frequency. Could PCLK be the noise source of the 445MHz noise point?

    By the way, when the DS90UB941 is turned off, the 445MHz and 400MHz noise points disappear.

    Thank you!

  • 0 in reply to ?? ?
    TI__Expert 3778 points

    Hi,

    Can you please share customer name and project for our records.

    ?? ? said:
    (2) Regarding the noise point at 445MHz, it is 5 times the PCLK frequency. Could PCLK be the noise source of the 445MHz noise point?

    Based on this data, the two failing frequencies are 400MHz and 445MHz. The 445MHz spike could be resonance of the PCLK at 5 times the nyquist as you mentioned. The 400MHz would not fall as a harmonic of the PCLK, which is interesting.

    ?? ? said:
    By the way, when the DS90UB941 is turned off, the 445MHz and 400MHz noise points disappear.

    This is a good test to verify source of emissions. Was the deserializer turned off as well?

    We need to pinpoint if RE is Serializer or Deserializer related and whether it is PCLK dependent.

    Two things I would try:

    1. Test the following test matrix below and see which device is the root cause of emission
      Serializer Deserializer Pass/Fail
      Powered off Powered off ???
      Powered on with PDB held low Powered on with PDB held low ???
      Powered on with FPD termination on Des side  Powered on with PDB held low ???
      Powered on with PDB held low Powered on with FPD termination on Ser side ???
    2. Have you tested the RE with different PCLK?
    ?? ? said:
    (1) One is about your suggestion of register. I study the register of spec of DS90UB941. These registers don't seem to be related to VOD level?

    The register to control VOD is on an indirect register page and the register is not made public on our datasheets. Register 0x41-0x42 control access to these indirect registers. Please follow the recipe provided earlier and verify the output change on oscilloscope.

    Thanks,

    Ragav Subramanian