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.

LMK04133: clipped sinewave clock input and outputs

Part Number: LMK04133
Other Parts Discussed in Thread: LMK04208, LMK03318

Hello CTS team,

my customer is currently considering our LMK04133 or LMK04208, asked if these devices support clipped sinewave clock input and outputs?

- Input reference clock: 10MHz, Clipped sinewave (0.8-1.5vpp) or LVCMOS

- Output clocks: 10MHz CMOS - 1 output, 48MHz Clipped sinewave (0.8-1.5vpp) - 2 outputs

Thanks.

  • Hi Dima,

    While both of these devices support clipped sine wave input, neither of these devices produces clipped sinewave outputs. LMK04208 has a multi-format output buffer which supports LVPECL, LVDS, or LVCMOS outputs. LMK04133 supports the same formats, but only on specific outputs.

    Why is clipped sinewave required? Could an alternative high-speed I/O standard such as LVPECL or LVDS instead be used?

    Regards,

    Derek Payne

  • Customer - In the QCA chip used defined 48MHz crystal oscillator and  measured sinewave/clipped sinewave waveform using scope. Not defined other waveforms as LVPECL, LVDS, or LVCMOS.

    is there a TI device that supports clipped sinewave input and output?

    requirements are:

    input - 10MHz

    outputs:  x1 10MHz

                   x2 48MHz

    low jitter,   below 200fs rms jitter

  • Dima,

    Again, any of these devices will support clipped sinewave input. None of them generate clipped sinewave output.

    Normally, clipped sinewave is not a strict requirement; as long as the input signal swing is large enough, and the input slew rate is high enough, any clocking format can be sufficient.

    What is the integration bandwidth for the jitter requirement? Is this required for both 48MHz clocks and 10MHz clocks? Or just the 48MHz clock?

    LMK03318 with AC-CML or AC-LVPECL output format seems to achieve the required performance, provided the input reference is clean enough:

    Regards,

    Derek Payne