• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 21 subscribers
  • Views 297 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.

LMX2595: OSCin impedance

Leon Loeser
Prodigy 90 points
Part Number: LMX2595

In our design, the reference clock comes in at a very low power level as a 50Ohm signal. As the clock-input of the LMX is of the CMOS-variety, the goal is to use a high impedance signal (not 50Ohm terminated, but higher) just requiring the use of an amplifier with a high voltage-amplification but not waste space/resources with a high current amplification aswell, since the power would only be wasted in the termination and the amplifier would be located less than 10mm from the LMX, guaranteeing signal integrity.

I was wondering what the actual impedance of the input was and wether there was a maximum recommended impedance to be used or if there were other suggestions on how to handle low reference-clock-power-levels. The current temporary solution is to use a 30dB amplifier to boost the 50Ohm terminated signal to 0dBm 50Ohm terminated on the LMX, but the amplifier is huge and wastes a lot of power.

  • +1
    TI__Mastermind 33430 points

    Hello Leon,

    If you expect that the high voltage amplification without the high current amplification has sufficient drive to send an appropriately high slew rate signal with minimal error due to reflections, there shouldn't be much trouble providing a non-50Ω input over such a short distance. The pin impedance is 10s of kΩ range if I recall correctly, but we don't specify this as a parameter so the values have some spread across PVT. The input structure looks like an AC-coupled BJT input, so whatever amplifier is used should be able to drive that (relatively light) load.

    The key constraint is getting the minimum input voltage, a high enough slew rate to not add noise to the reference, and not introducing additional noise due to the flicker noise of the input amplifier. You may find that getting sufficiently low flicker noise from the amplifier ends up necessarily pushing you toward higher-current and higher-power amplifiers anyway - I can't say I've ever tried what you're suggesting, so maybe there's some spectacularly clean amplifier that can handle 30dB amplification on a >10MHz signal without requiring high current or 50Ω loading, but my gut feeling is that this is hard to achieve even with lots of power to throw at it.

    Regards,

    Derek Payne