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.

LMK1C1104: Rise and fall time

Part Number: LMK1C1104

Hi Team,

During the test, the customer found that the rise and fall time of the 125MHz clock output by the LMK1C1104 is long, about 1ns. Could you please help support the following requests and questions:

1. From the DS, when the clock is 156.25MHz and CL=5pF, the maximum rise and fall time is 0.8ns. Is this time related to clock frequency? Could you please test the time under the condition of 125MHz, CL=6pF with EVM and share your test data?

2. From the PCB, the distance between the buffer and the CPU is not far (but there are two vias). From the schematic, there is a 33.2Ω resistor on the trace. Is this resistor the reason for the extended rise/fall times? Is it recommended to use 0 ohm resistors?

*EC1_GTX_CLK125 is collect to CPU input pin directly.

*R9966 pin2 & CPU pinAC3 have 2vias, and the EC1_GTX_CLK125 trace on inner layer18

3. The customer's index is less than or equal to 0.8ns. Is this device a good choice? Is there any other recommendation?

Regards,
Hailiang

  • Hi Hailiang,

    I have forwarded this thread to an expert and we will get back to you as soon as possible.

    Thanks,

    Evan Su

  • Hi Hailiang,

    This seems like signal integrity issue on the board due to impedance mismatch. 

    For LMK1C1104, 0.8 ns is defined for 5pF loading at 156.25MHz. Clock frequency is one difference between but can you confirm the pF is with trace capacitance + load?

    We can remove the series resistor because driver output impedance is 50 ohm. It could help with reflections and performance. 

    Let me know if you are still seeing the same issue, then I will try to see your specific case.  

    Best,

    Asim

  • Hi Asim,

    The load capacitance is known to be no greater than 6pF, but the trace capacitance is unknown. If you can share the test data at 125MHz, 6pF, it will help to confirm whether this device is a good choice.
    Also, should I remove the resistor or change it to 50 ohms?

    Regards,
    Hailiang

  • Hi Hailing,

    You can replace it to 0 ohm. I will look for the data and get back to you.

    Best,

    Asim

  • Hi Asim,

    We tried 0 ohm and 50 ohm. Improvements are very limited.

    The customer found that the rise/fall time of the clock signal on the input side of the LM is shorter than that on the output side.

    If you can share the test data at 125MHz, 6pF, it will help to confirm whether this device is a good choice.

    When can I except to get the test date? It is urgent for this project. Could you pls provide it asap? Thanks.

    Regards,

    Hailiang

  • Hi Hailiang,

    I have shared data on your email. I am closing this thread since I have moved this conversation to email.

    Best,

    Asim