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.

TLK2711EVM-CVAL: How to meet TLK jitter requirement?

Part Number: TLK2711EVM-CVAL

Hi,

   We are doing the interoperability test between TLK2711 and Microchip RTG4 FPGA devices. Here we aim to send the PRBS-7 pattern at data rate of 2.5 Gbps from TLK2711 TX to RTG4 RX and through far end loopback send it back to TLK2711 Rx. But PRBS_PASS signal is never high (always toggling). We also tried simplex tests (no loopback) as follows:

  1. Transmitted PRBS-7 pattern from TLK2711 Tx to RTG4 Rx successfully. CDR of RTG4 SERDES is locked with cumulative errors count zero.
  2. Transmitted PRBS-7 pattern from RTG4 Tx to TLK2711 Rx unsuccessfully. PRBS_PASS signal always toggling.

 

We also noticed that TLK2711 data sheet says that TXCLK jitter peak to peak is 40 ps but we are not able to meet that requirement. We are giving 125 Mhz clock from SI5338 clock generator board and when we tried to measure the jitter of this clock its PJrms value is 19.8 ps which if we multiply to Crest factor of 14 (as applicable for the bit rate of 1e-12 for clock signals) gives us peak to peak value of 278 ps (way higher than the data sheet specification of 40 ps). Please let us know how to meet TLK jitter requirement?

 

Also VICM (common mode voltage) of both TLK2711 and RTG4 devices are different. VICM for TLK2711 Rx, Tx is 1250 mV typical where as for RTG4 Tx and RTG4 Rx it is 20mV and 150 mV typical respectively.

 

What could be the possible causes of this behavior of not able to see PRBS_PASS signal status as high at the TLK2711 side? Please note that we tried tlk2711 internal loopback and it is consistently high.

 

Thanks

Deepak

  • Hi,

       Again today we have tried TLK2711 SMA loopback and observed the PRBS_PASS status still it is toggling continously. Then we removed the TLK TX side capaciotrs (C25,C24) and tried same thing still we observed PRBS_PASS is toggling continously. 

    Can someone help on this quickly? We are waiting for reply from last 2 days.

    Thanks

    Deepak

  • Hi,

    • Yes, your CLK reference jitter is way too high. Please try using a clock generator from TI's portfolio. Many low jitter options available. See link below.

    www.ti.com/.../products.html

    • Please confirm how the TLK pins below are configured for your problem case
      • LCKREFN, LOOPEN, PRBSEN, TESTEN

    Thanks,

    Rodrigo Natal

    HSSC Applications Engineer

  • Hi Rodrigo,

         Thanks for reply. We don't have TI clock generators but we have given clock from function generator, observed 90ps peak to peak periodic jitter with same results i.e PRBS_PASS status toggling continously.

    For TLK SMA loopback we have below J7 jumper settings :

    1. TESTEN - installed (closed)

    2.PRBS_EN - not installed (open)

    3.LCKREFN - not installed (open)

    4.ENABLE - not installed (open)

    5.LOOPEN - installed (closed)

    J31 - OPEN (not installed)

    Also, please can you recommend us TI clock generator that meets TI TLK2711 jitter specifications (40 ps peak to peak)?

    Thanks

    Deepak

  • Hi,

    • For case were you have RTG4 Tx to TLK Rx I believe PRBSEN needs to have jumper installed (Logic 0.) Please review the TI EVM user's guide below.

    www.ti.com/.../sglu001a.pdf

    For clock generator you may try part number below.

    www.ti.com/.../CDCM61001

    Thanks,

    Rodrigo

  • Hi Rodrigo,

          We believe PRBS_EN jumper should be always open for all the PRBS related tests. Even with RTG4 as Tx and TLK as Rx we need to PRBSEN in PRBS verification.

    Do you have any other suggestions?

    Thanks

    Deepak

  • Thanks. At this point, assuming your TLK device hardware configuration is following the TI datasheet and application note guidelines, I would speculate the problem is with the clock you are using. The fact that you are not even able to get the TLK serial loopback mode to run PRBS successfully points to it I think. I unfortunately do not have other suggestion to add. Please double check your setup for the serial loopback test relative to Figure 2-2 in the application note per link below.

    www.ti.com/.../sglu001a.pdf

    Cordially,

    Rodrigo