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.

LMK00304: LVCMOS Termination

Part Number: LMK00304


Had to make a new question, as the system marked my last one as resolved when it is not.

If the output impedance of the LMK00304 REFout is 50 ohms, what is the recommended termination scheme?  A basic Signal Integrity simulation is showing is showing a massive line reflection without any external termination.

  • Andrew,

    There's two options for termination:

    • Terminate to 50Ω and match impedance with the source. This cuts the amplitude of the output significantly, you might only get 0V to VCC/2 instead of 0V to VCC. You could add a DC-blocking cap between the source and load to maintain full amplitude, but this would center your LVCMOS signal at the receiver around GND potential (or whatever AC-GND is selected for the 50Ω load - conceptually you could find a stiff 1.25V/1.65V source and get full 2.5V/3.3V LVCMOS swing, but this is usually far more trouble than it's worth).
    • Add a small series damping resistor e.g. 22Ω (and possibly a small load capacitance e.g. 4.7pF). Amplitude will remain 0V to VCC. If the termination is placed close to the LVCMOS receiver, reflections shouldn't be too bad, but there will be significant reflections at the driver. This will reduce the slew rate at the receiver.

    Keep the limitations of the LVCMOS output in mind - it's only designed for up to 250MHz, and should probably be run over only short distances. LVCMOS isn't really intended as a high-speed signaling standard the same way as LVPECL or LVDS. It's not uncommon to see very high reflections on LVCMOS drivers. Hence, they are usually only used for lower frequency signaling where the reflections have less of an impact, or they are damped with series resistance to minimize the effect of these reflections at the receiver side.

    Re: Resolved, you should still be able to continue the thread on "resolved" questions if you have more follow-up questions. If you are unable to comment on resolved questions within 30 days of the last reply, that's actually a bug and we need to fix it. You're welcome to try commenting again on this thread if we manage to resolve your question.

    Regards,

    Derek Payne

  • Thanks for the info, this should be enough for me to move forward with.  I don't know why I could not reply to the other thread once it was marked resolved, seems to be working here.

  • Alright, thanks for confirming. Unsure what happened with the last thread... our audit log claims even now we should both be able to reply. I'll pass this to site IT for additional investigation.