• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 595 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.

DS90UB948-Q1: DS90UB948 : CMLOUT waveform cannot be measured correctly

?? ??
Prodigy 150 points

Part Number: DS90UB948-Q1
Other Parts Discussed in Thread: USB2ANY

Tool/software:

Hi,

I am measuring the CMLOUT output waveform of the DS90UB948.
However, the amplitude is small and the waveform is distorted, so I cannot measure the expected waveform.
Could you please give me some advice on how to obtain the correct waveform?

The communication itself is normal, and the signal is input to RIN0±.
The register settings are as follows using USB2ANY.

- 0x52h: 0x80 (RIN0)
- 0x56h: 0x08
- 0x57h: 0x02 (RIN0)
- 0x63h: 0x00

I have attached the waveform when RIN0 is selected.

  • 0
    TI__Mastermind 42015 points

    Hello,

    Thank you for the oscilloscope screenshot and the register settings. 

    ?? ?? said:
    - 0x52h: 0x80 (RIN0)

    Register 0x52 needs to be set to 0x00 for RIN0.

    Best,

    Jack

  • 0 in reply to Jack Scherlag
    Prodigy 150 points

    Hi,

    Thank you for your feedback. I mistakenly wrote it as 0x52h:0x80. In reality, it is 0x52h:0x00.

    Regarding the default value, just to be sure, I tried writing 0x00, but it did not work."

  • 0 in reply to ?? ??
    TI__Mastermind 42015 points

    Hello,

    For the CMLOUT measurement, are you measuring on a TI EVM or a custom board? And what is the forward channel rate you are expecting to see on the CMLOUT?

    The CMLOUT measurement needs to be viewed as an eye diagram to appropriately evaluate the quality of the signal. Without knowing the background of this system, I cannot say for certain if this is a CMLOUT issue or a channel with too much attenuation.

    What is the SER and the cable length?

    Best,

    Jack

  • 0 in reply to Jack Scherlag
    Prodigy 150 points

    Hello,

    Thank you for your inquiry.

    The CMLOUT measurements are conducted using a custom board. Here are the specifications:

    • Forward Channel Rate: 2.0545 Gbps
    • SER: DS90UB927Q
    • Cable Length: 2.5 m

    The deserializer is implemented in the monitor, and the video is displaying correctly.

    Additionally, we are using the Tektronix TDP7700 probe and the MSO64 oscilloscope. For comparison, we measured the USB differential signals using this probe and a differential signal probe, and we observed similar waveforms with an amplitude of approximately 500 mV.

    Best regards,

  • +1 in reply to ?? ??
    TI__Mastermind 42015 points

    Hello,

    Thank you for the additional information. CMLOUT performance can vary across systems and device. This is the why the datasheet only lists typical values for CMLOUT performance. 

    From the waveform you provided above, I am seeing a differential amplitude near 150mV which is lower than the typical 300mV. Another tool that can be used to verify the link stability is MAP (Margin Analysis Program). MAP will sweep the various strobe and EQ settings in the 948 and give you an indication of your link margin.

    Below is an app note for MAP if you are interested. Please note that MAP requires local I2C access to the deserializer.

    MAP App Note: Link

    Best,

    Jack

  • 0 in reply to Jack Scherlag
    Prodigy 150 points

    Hello,

    Thank you for your advice. Although it differs from the MAP method, the issue has been resolved.

    The reason for the inaccurate measurements was that the offset voltage setting for the waveform was not configured on the oscilloscope side. By adjusting the offset in the probe settings, I was able to obtain a p-p 500mV waveform.

    I will utilize the MAP method as a signal quality evaluation technique moving forward. Thank you very much.