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.

Value of resistor for signal attenuation between Crystek VCXO and LMK04828 OSCin

Other Parts Discussed in Thread: LMK04828

Hi,

In the ADS42JBX9 EVM, there is a network of 2 resistors between the Crystek VCXO Y2 and the LMK04828:

  • R138: 100 ohm, for signal attenuation.
  • R44: 49.9 ohm for impedance matching.

My question is, why does R138 be equal to 100 ohm exactly?

  • For Crystek VCXO, I don't know its output impedance. (If someone tell me this value, this will be helpful).
  • For LMK04828, I know that for OSCin buffer: Vinl = 1.6V, Vinh = 2.0V and it is self-biased around VCM=1.039V.

Thank you.

Best Regards,

Mourad Ghorbel.

  • R138 is to attenuate the 3.3V LVCMOS output swing from CVHD-950 VCXO down to about 1.1 Vpp, which is roughly in the center of the input voltage spec for OSCIN (0.2V <= V_OSCin <= 2.4Vpp).  R138 can be a different value, as long as the input voltage swing is within the V_OSCin spec.

    Regards,
    Alan

  • Hi Mourad,

    The CHDV-950-100 MHz Crystek VCXO has 3.3V CMOS output. I am not exactly sure what the impedance of their output driver is (~10 ohms) but the 100 ohms series resistor should limit the output swing to 50/(100+50+Ro)*3.3V where Ro is the output impedance of the CMOS driver. The OSCin specifications are shown on Page 12 of the datasheet. The single-ended AC coupled coupled input to OSCin is speced at 2.4V (max). It is recommended to AC couple the reference clock into OSCin. The DC bias is set internally.

    Regards

    Arvind Sridhar

  • Hi,

    Thanks for your explanations.

    In the LMK04828.ibs:

    [Model] OSCIN
    Model_type Input
    Vinl = 1.6
    Vinh = 2.0

    The OSCin input has an internal bias voltage of about 1 V (by simulation). If I have attenuated the voltage swing to 1.1V, then when  it is AC coupled, the voltage will oscillate between 0.45 and 1.55. 

    Thus, after the rising edge, the signal does not reache the Vinh.

    This is my problem.

  • Hi,

    Are Vinl and Vinh, found in LMK04828.ibs, specifications without any utility?

  • Yes, the Vinl and Vinh specs in the ibis model do not apply.  For AC coupled input, you just need to meet the single-ended input votlage swing requirement.

    Regards,
    Alan