• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 21 subscribers
  • Views 452 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.

LMK1C1102: can't output becomes constant high at high frequency of 68MHz

Kelly Wang
Intellectual 550 points
Part Number: LMK1C1102


LMK1C1102 is used as the following way. Powered by 3.3V and gate is always enabled. When input frequency is 68MHz, output  becomes constant high with a low level of about 1V.  When input is 120Hz 50% square clock the output is normal. When input is 1KHz 50% square clock, it is good also.  When input is 2MHz 50% square clock, the output becomes triangle waveform with peak value of 2V, valley value of 1V.  The output load is open when perform  tests. 

         

Please help on this issue. 

Thank you. 

  • 0
    TI__Mastermind 34270 points

    Hi Kelly,

    What is the load impedance for LMK1C1102? This device is rated for up to 250MHz as long as the load capacitance on the outputs is kept relatively small (5pF is used in the datasheet for supply current tests). On the other hand, driving a higher capacitive load will eventually exceed the drive strength of the output.

    Regards,

    Derek Payne

  • 0
    Intellectual 550 points

    clock output drives CMOS image sensor clock input. Image sensor data sheet does not specify input capacitance. What is the maximum capacitance load of LMK1C1102 @68MHz?  Please advise on options that can drive bigger load that is pin to pin compatible with LMK1C1102.  Thank you very much. 

  • 0 in reply to Kelly Wang
    TI__Mastermind 34270 points

    Kelly,

    If the device can handle 5pF at 250MHz, I'd expect it to handle at least 20pF at 68MHz.

    How far is the LMK1C1102 from the device it is clocking?

    How is the clock signal being probed to observe this waveform degradation? What equipment are they using? At 68MHz, the signal is fast enough that the inductance of a standard oscilloscope probe could be impacting the signal integrity, and a pigtail grounding strategy with a small loop inductance may be required.

    Regards,

    Derek Payne

  • 0 in reply to Derek Payne
    Intellectual 550 points

    I got reply from image sensor technical support. Input capacitance of the two devices are 0.4657pF and 0.3978pF individually. Now large load caused deformation can be excluded. The trace length are 0.804 inch and 0.567inch. The trace are microstrip and strip line combination. Microstrip trace impedance is 78.51 ohms  while the strip line impedance is 15.1 ohm.  Impedance in layout is not controlled very well.  I am not sure how strict the 50 ohm impedance trace control should be controlled for LMK1C1102. 

    oscilloscope probe uses the short ground pin to avoid introducing extra interference. I probed  oscillator LVCMOS33 68MHz output using the same equipment and probe, it shows no problem. 

  • 0 in reply to Kelly Wang
    TI__Mastermind 34270 points

    Kelly,

    I don't think that <1 inch trace is going to matter in terms of load capacitance.

    Have you probed the supply during operation? I wonder if the VCC pin is actually at 3.3V. Maybe either the ferrite impedance or the supply connection is higher impedance than expected, so as the device attempts to switch at higher frequency the supply voltage at the pin is dropping and the device is unable to switch the output buffer. Check this next.

    Regards,

    Derek Payne

  • 0 in reply to Derek Payne
    Intellectual 550 points

    Derek,

    I measured the power supply of LMK1C1102, it is 3.28V with very low noise level.  

    Thank you

  • +1 in reply to Kelly Wang
    TI__Mastermind 34270 points

    And that's while "switching" at 68MHz?

  • 0 in reply to Derek Payne
    Intellectual 550 points

    It is measured when LMK1C1102 is enabled.