• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 551 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LMK04828BEVM

user6360432
Intellectual 320 points

Other Parts Discussed in Thread: LMK04828

when I use DCLKout0 and SDCLKout1 as output, select SDCLKout1 as device clock in the TICS Pro, Tthe output wave shape of the two channel has 269 ps delay,why?

  • 0
    TI__Mastermind 18070 points

    Hello,

    My coworker will get back to you tomorrow.

    Regards,
    Hao

  • 0
    TI__Mastermind 28590 points

    Hello,

    I don't expect such a large skew variation at the device output.  Are you measuring using the LMK04828EVM at the SMA?

    Are you connecting exactly the same cable length from the SMA to scope?  Assuming cable with propagation velocity of 65%, a length error of 5.24 cm (length error = 0.65 * (3e10 cm / s) * 269 ps), would result in approximately 269 ps timing error.

    Do you have any skew turned on in the oscilloscope?

    If you swap the cables where they connect at the PCB, does the skew change to -269 ps?

    If you are using probes, if you probe the exact same point, is the skew = 0 ps?

    Please share the configuration file you had programmed.

    Thanks.

    73,
    Timothy

  • 0 in reply to Timothy T
    Intellectual 320 points

    hello,

    Timothy.

    I connect the LMK04828EVM to oscilloscope,but the junction of the input to oscilloscope is different, one of them has a female-to-female adapter more,it about 2mm.And when I change the divider to 3 ,4 and 5,the delay is 120ps, 260ps and 60ps different, why?

  • 0 in reply to user6360432
    TI__Mastermind 28590 points

    Hello,

    After changing the divide value, the phase will not be deterministic.  Press the SYNC dividers at the top of the toolbar to align.

    73,
    Timothy

  • 0 in reply to Timothy T
    Intellectual 320 points

    hello,

    Timothy. 

    I still have problem with it. when I use DCLKout0 and SDCLKout1 as output, select SDCLKout1 as device clock,press the SYNC dividers at the top of the toolbar after changing the value of divider, but the delay didn't change, and when I change the value of divider to 3, 4 and 5, the delay of the signal of the two output is 120ps,260ps,and 60ps differently,why?

    when I use SDCLKout1 as device clock, can I change the delay individually of SDCLKout1 for aligning with DCLKout0? How can I operate？

  • 0 in reply to user6360432
    TI__Mastermind 28590 points

    Hello there,

    You know, on second thought, SYNC should not have impacted the phase between DCLKout0 and SDCLKout0 with both outputs operating in device clock mode as they are sourced by the same divider and have no ability to make individual adjustments between them except by changing the DCLKout0_POL and SDCLKout1_POL bits.  Different output formats may have a small affect, but I wouldn't expect it to change for different divides.

    Can you share the saved .TCS file of your configuration?

    Again as above, I don't expect this to be an issue for your configuration, but please uncheck DDLY_PD for any outputs for which you would like deterministic phase.  Clocking SYNC dividers should uncheck this bit.

    73,
    Timothy

  • 0 in reply to Timothy T
    Intellectual 320 points

    Hello,

    Thank for anwering.

    Here are my .TCS files and different delay signal in the oscilloscope.3divider.tcs

  • 0 in reply to Timothy T
    Intellectual 320 points

    Hello,

    Thank for anwering.

    Here are my .TCS files and different delay signal in the oscilloscope.The connections of the two channels of the oscilloscope is different which may cause some delay but I think the delay  between DCLKout0 and SDCLKout0 should be invariant when I change the value of divider. I don't know why I got different delay.

    1667.3divider.tcs4divider.tcs5divider.tcs

  • 0 in reply to user6360432
    TI__Mastermind 28590 points

    Hello,

    I tested this today and did not measure such large differences.  All divide settings were within a 20 ps range.

    I presume you are using 50 ohm termination on the scope and have the unconnected output of the pair terminated with 50 ohms also?

    I would use cables of the exact same length.  I know the LMK04828 EVM has some impedance mismatches at the SMA connector output, the thin traces where the differential traces split and the thick pad at the SMA which I'm wondering if is increasing the measurement error due to the fact that your cables are different length.

     - Can you test with exactly the same length cable?

    73,
    Timothy

  • 0 in reply to Timothy T
    Intellectual 320 points

    Thanks very much, I'll test again with same length cable