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.

ADC12DJ5200RFEVM: How can it work correctly with Xilinx FPGA (VCU118)?

Part Number: ADC12DJ5200RFEVM

Hello!  

I'm doing a project where I need to connect a high speed ADC(ADC12DJ5200RFEVM) to an FPGA(VCU118). The system could be like this:

ADC -> (FMC) -> FPGA.

I need to transfer 12bit wide data sampled by ADC from channel A at 10Gsps from ADC to FPGA, so I choose to use JMOD19 mode.

ADC has link A and link B, and they correspond to BANK121(link A), BANK122(link A), BANK125(link B) and BANK126(link B) on VCU118. In addition, BANK 121 and BANK 122 in the SLR0 of VU9P, and BANK 125 and BANK 126 in the SLR1 of VU9P. The SLR is  SLR(Super Logic Region) in VU9P.

In each link, I use two jesd204 PHY cores and one jesd204 core, and the two PHY cores share a REFCLK (reference clock).

Results when debugging, the result received from link a is correct. On the contrary, the result received from link B is wrong.

I found that the alignment of lines is K28 5, there is no difference between the two, but in the ILAS stage, the data output by the two link PHY cores is different.

In the ILAS stage, the first multi frame should be an incremental counter. The PHY core output (py0_gt0_rxdata) of link a is correct and the PHY core output (py1_gt1_rxdata) of link B is wrong.

I don't know what causes the error in the data received by link B. Can you help me see what causes it?

Additionally, I have used JMODE0, JMODE3, but these JMODES also have encountered the above problems.

Thank you!

  • Hi,

    Kindly let me know if you have accounted for the fact that all the lanes on link B are inverted in terms of polarity.

    In addition, you can also simplify the design by using the following combinations:

    1> 1 8 lane PHY core connected to 1 8 lane JESD core (for each link)

    2> 1 16 lane PHY connected to 1 16 lane JESD core (if the Xilinx IP allows you to generate this). The timing closure run should be able to handle the SLR crossing.

    Regards,

    Ameet

  • Hi, Ameet

    Thanks for your reply, I have solved my problem. The cause of the problem is LINB's differential input of the PHY CORE is inverted.

    Regards,

    Jianguo