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.

TLK2711-SP: How much UI is when the bit error rate is 10^-10 and 10^-8?

Part Number: TLK2711-SP

Hi team

First of all, please let me make sure my understanding as below.

According to the link as below, we can see "The TLK2711 as specified will achieve a BER of 1E-12 or better. This means there could be a single bit error in 10^12 bits." Does this mean that the jitter can be 0.4UI when BER is 1E-12?

https://e2e.ti.com/support/interface-group/interface/f/interface-forum/944890/tlk2711-sp-microsemi-fpga-serdes-communication-with-tlk2711-sp-at-2gbps?tisearch=e2e-sitesearch&keymatch=TLK2711

If my understanding is correct, could you estimate the value of UI when the bit error rate is 10^-10 and 10^-8?

Regards,

Noriyuki Takahashi

  • Related to: According to the link as below, we can see "The TLK2711 as specified will achieve a BER of 1E-12 or better. This means there could be a single bit error in 10^12 bits." Does this mean that the jitter can be 0.4UI when BER is 1E-12?

    • I believe you are referencing the input jitter tolerance per datasheet of 0.4UI. Yes, this jitter value assumes BER 1E-12

    Related to: If my understanding is correct, could you estimate the value of UI when the bit error rate is 10^-10 and 10^-8?

    • Total jitter (for a given probability level = deterministic jitter (DJ) + sigma_RJ*N where N is  the probability factor
    • Deterministic jitter will not change with probability level
    • The customer would need to measure the random jitter sigma value (sigma_RJ) for their system then they can multiply it by the N value corresponding to the target BER level

    Thanks,

    Rodrigo Natal

  • Rodrigo,

    Could you tell how do you get 0.4UI when you assumes BER 1E-12?

    Sorry, I can't understand fully, but I think it would be helpful if you show the way to get 0.4UI when you assumes BER 1E-12.

    And could you tell the value of UI when you assume EBER 10^-10 and 10^-8 in TI case? They want the value of UI when you assume EBER 10^-10 and 10^-8 in TI case.

    Regards,

    Noriyuki Takahashi

  • Hi,

    I already covered it in my previous post: 

    • Total jitter (for a given probability level) = deterministic jitter (DJ) + sigma_RJ*N where N is  the probability factor
    • For your spurce in question you would need to perform a jitter measurement of DJ and RJ_sigma using the scope method (or other method) then perform the calculation. Most new scopes like Keyshight DCA-X do this for you
    • You may refer to the application note from Anritsu per link below to familiarize yourself with jitter methodology 

    Thanks,

    Rodrigo Natal

  • Rodrigo,

    Do this mean your team also meausred DJ and RJ_sigma to get 0.4UI and performed the calculation? 

    If this is yes, could you provide any information about the measurment environment? This is required to judge whether their measurement result is reasonable or not.

    Regards,

    Noriyuki Takahashi

  • Hi Noriyuki,

    Yes, you would need to measure DJ and RJ to perform the calculation. The customer simply needs to measure their system DJ and RJ and ensure the TJ is within 0.4UI; it should not be that complicated.

    Thanks,

    Rodrigo Natal

  • Rodrigo,

    Thanks and sorry for confusing. I wanted to confirm datasheet's test condition to calulate 0.4UI. Do you have any information?

    Regards,

    Noriyuki Takahashi

  • I do not have the exact jitter decomposition, sorry.

    Thanks,

    Rodrigo