AD9578 questions

I would greatly appreciate an answer to my two outstanding questions.

I have another question. I notice that the development board uses ADP7104 regulators for the AD9578. This is a low noise regulator, however the noise is much higher than the ADM7150A, which I used in a previous project to supply two off HMC832.

Is there any advantage in using the ADM7150A to supply the AD9578 ( at considerably higher cost)?

The answer depends upon the supply noise rejection of the AD9578, which is not specified in the data sheet.

regard

Cosmo Little

  • 0
    •  Analog Employees 
    on Jun 16, 2016 12:52 AM

    Hello Cosmo,

    I wish I had a definite answer for you, but we haven't tested it. My educated guess is that you will not see a big improvement with the ADM7150A, and I think the reason is that the AD9578's intrinsic jitter isn't low enough. We have other devices that have lower jitter than the AD9578 if that's what you need.

    -Paul

  • Dear Paul,

    Thank you for the reply.

    I am now designing my prototype schematic, and need an answer to my question concerning the reference input, especially if it is necessary to bias the XO pins externally, as is done on the evaluation board. I have an HCMOS (3.3V) external reference at 50MHz, however this exceeds the maximum voltage for direct connection.

          I am also still unclear as to the effect of the xtal sustaining amplifier, which it appears cannot be disconnected.

       My other question concerning rational mode is now answered as I eventually found a note (buried in the small print!) that the C/D ratio must be less than 1. However I am still unclear as to the effect of reducing the C/D ratio to a minimum fraction.

    regards

    Cosmo Little

  • 0
    •  Analog Employees 
    on Jun 17, 2016 6:31 PM

    Hello Cosmo,

    I'm going to ask one of the AD9578 designers to review this post to be sure I'm answering correctly.

    > I am now designing my prototype schematic, and need an answer to my question concerning the

    > reference input, especially if it is necessary to bias the XO pins externally, as is done on the

    > evaluation board. I have an HCMOS (3.3V) external reference at 50MHz, however this exceeds

    > the maximum voltage for direct connection.

    >    I am also still unclear as to the effect of the xtal sustaining amplifier, which it appears cannot be disconnected.

    It shouldn't impact the performance.

       > My other question concerning rational mode is now answered as I eventually found a note

    (buried in the small print!) that the C/D ratio must be less than 1. However I am still unclear as

    to the effect of reducing the C/D ratio to a minimum fraction.

    The question is best answered by the designer, but I can say that rational mode is provided merely as an easy way to program the dividers, and not really a different mode per se.

    -Paul Kern

  • Cosmo,

    Hi, this is Steve again.   I commented on how to drive XO3,XO4 in my last comment, so I won't repeat it here...

    To address Paul's point on the rational mode; the rational mode is provided so that EXACT ratios can be achieved.  That is if the reference should be multiplied by, say 237/255, it can be.   That is to say 237/255 != 0.92941.    Thus bits in a FEC system would not be lost...

    Regards,

    Steve

  • 0
    •  Analog Employees 
    on Aug 2, 2018 3:29 PM
    This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin