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.

DS90CR288A: CMOS/TTL output setup hold time issue

Part Number: DS90CR288A
Other Parts Discussed in Thread: DS90CF386

Hi Team

I was exchanging with private messages, but I haven't responded even after a week, so I will post it again on E2E.
The contents of the previous E2E are as follows. I am disappointed with the slow response.

We received a single shot waveform from a customer. Check the waveform.
Put the waveform here again.

DS90CR288A does not keep the setup / hold time of LVCMOS output.
At DS90CR288A setup is out of specs. Please tell us your opinion.

Best Regards,
Ishiwata

  • Hello Ishiwata-san

    Here is the specification from CR288A Datasheet. Minimum Setup time is defined at 85MHz at the pins to be 3.5ns. In the plot you attached above it seems you are measuring setup time at 100MHz. Since we do not specify the setup time at 100MHz, my opinion is DS90CR288A setup specification is not violated. Can you please ask your customer to measure the setup time at 85MHz and ensure the measurement is made at the pins?

    Thanks

    Vijay

  • Hello Vijay-san

    PCLK = 85MHz (11.7nsec).
    It is not 100MHz.

    Best Regards,
    Ishiwata

  • Ishiwata-san

       Thank you for the clarification. Can you also clarify my other question. Was the measurement made at the device pins? If the measurement is made exactly on the device pins, then this particular part may have a marginality and you might need to remove the part and submit to TI Quality team for further analysis. But if the measurement is made after the signal went through a PCB trace or if the Clock line and Data line are loaded differently then you may need to review your board design

    Thanks

    Vijay

  • Hello Vijay-san

    Thank you for your reply.

    I will proceed with the request for analysis of DS90CR288A at E2E content as evidence.
    I understand that is DS90CR288A out in Setup / hold time?

    This spec-out has not occurred in certain products. The same spec-out can be confirmed on many devices.
    So, can you measure the setup / hold time of DS90CR288A once at TI?
    Measurement can be performed by replacing DS90CF386 of FVM3V8BT-85 EVM with DS90CR288A.
    (As you may know, DS90CF386 and DS90CR288A are pin compatible.)

    There is further confirmation. Jitter is very big. Can I check if the same jitter occurs in the above EVM?

    The customer is very in trouble. Furthermore, support is considerably delayed.
    I hope, you quickly support.

    Best Regards,
    Ishiwata

  • Ishiwata san

          DS90CR288A is NOT out in setup/hold time. We measure it on our tester before we ship the parts. We expect that the customer measure the setup/hold time at the pins of the device. If there is an issue with this particular device you have to go through TI quality. Yes, once you send the failing device , our Quality team automatically will open a case and run the measurement again at TI to see if that is still the case. If we see that the setup/hold matches our spec, we will return the device with NTF designation (No trouble found. Setup/hold time will probably get tested on our Tester. On an EVM we cannot measure this on the pins

    Who is the customer? Are they seeing a failure in the system?

    Thanks

    Vijay

  • Hello Vijay-san

    Thank you for your reply.
    You will ask the TI quality team.

    The customer was measuring at the pin of the device.
    The result is the previously sent waveform.

    Jitter of data signal of LVCMOS signal is large.
    Please tell us why.
    ・ Crosstalk noise or power supply noise?
    ・ DDJ or ISI?

    Are there other factors? advice please.

    Customer information is also sent in a private message.

    Best Regards,
    Ishiwata

  • Ishiwata-san

           Jitter can be due to Impedance mismatches or reflections (as I see kinks in the waveform), I do not see any power supply noise issues. Also please find out if the issue is on one part or on all parts and systems

    Thanks

    Vijay

  • Hello Vijay-san,

    Thank you for your reply.

    "Also please find out if the issue is on one part or on all parts and systems"

    Does this mean the following?
    Is there a problem other than DS90CR288A?

    Many DS90CR288A have the same problem (setup / hold sepc-out).

    Best Regards,
    Ishiwata

  • Ishiwata-san

    Is there a system level issue that is being seen? Data being sampled incorrectly or data being captured improperly or are they just seeing measurement issues? Can you let me know agter you submit the part to TI quality?

    Thanks

    Vijay

  • Ishiwata-san,

    Is there any update here to Vijay's question? 

    Thanks,

    Casey

  • Casey-san

    Sorry for not replying.There is no update now.

    As Vijay advises, we will ask the TI quality team for analysis.
    We are preparing in-house procedures and explaining to customers.

    Because it is a very rare case issue.
    Also, there are many DS90CR288A that out of spec the setup/hold.
    And, it took a lot of time to get support from the quality team.

    Thank you for your support.

    Best Regards,
    Ishiwata