[time-nuts] An example of phase noise

John Ackermann N8UR jra at febo.com
Sun Aug 20 14:59:40 EDT 2006


Poul-Henning Kamp said the following on 08/20/2006 02:37 PM:
> In message <44E89DD9.7090303 at febo.com>, John Ackermann N8UR writes:
> 
> 
>>Thought this might be interesting.  I took a spectrum analyzer shot of
>>my ICS525 "Clock-Block" synthesizer at 250MHz, and overlaid it with a
>>shot of an RF signal generator (Marconi 2202A) at the same frequency,
>>and using the same reference source.
>>
>>The 2202A is designed for bench use and isn't noted for particularly
>>good phase noise, but you can see the dramatic difference between the
>>two units (the ICS525 is in red, the 2202A in cyan).
>>
>>That's why the Clock-Block isn't recommended for RF use!
> 
> 
> Have you played with the various PLL configurations that give the
> same output frequency to see what difference they make ?

No, I haven't done that.  In this case, all three results from the
calculator (lowest error, jitter, and current) used the same jumper
settings, and I haven't messed with the formulas to come up with other
possibilities.

However, some other experiments indicate that the phase noise is better
when only the "V" divisors are used; using the "R" and particularly "S"
divisors results in quite a bit more noise.  If you look at the phase
noise plots at http://www.febo.com/time-freq/hardware/ICS525/index.html
you can see the difference between a simple divisor configuration and a
more complex one.

John



More information about the time-nuts mailing list