[time-nuts] quick and very dirty phase comparator

[Bruce Griffiths]

Joseph M Gwinn wrote:
> time-nuts-bounces at febo.com wrote on 06/04/2008 08:05:16 PM:
>
>   
>> WB6BNQ wrote:
>>     
>>> Ulrich,
>>>
>>> One of things I noticed when playing with that DDS design 
>>>       
>> tool is you can achieve
>>     
>>> the same output frequency using a different clock frequency, 
>>>       
>> thus removing the
>>     
>>> spurs.  THe trick is to look at clock frequencies that you 
>>>       
>> can lock to your house
>>     
>>> reference for stability.  Also, it might be possible to use 
>>>       
>> cascading DDS chips,
>>     
>>> one providing the clock for the other.
>>>
>>> What are the thoughts on such an approach ?
>>>
>>> Bill....WB6BNQ
>>>
>>>
>>> Ulrich Bangert wrote:
>>>
>>>
>>>       
>> Cascaded DDS systems together with PLLs have been widely investigated:
>> See US patents: 4965533, 5028887, 5146186, 5216389, 5467294, 5517156, 
>> 5598440. 7143125
>>
>>
>> It's much better to use a DDS that employs noise shaping techniques to 
>> shift the spur energy to higher offset frequencies that are easily 
>> cleaned up by a PLL.
>>     
>
> A question occurs to me.  Perhaps we can turn this upsidedown.  While the 
> DDS won't necessarily generate the frequency we want without spurs, can't 
> we set the DDS to generate a nearby frequency that is free or sufficiently 
> free of nearby spurs?  We may not be able to get nice power-of-ten phase 
> amplification ratios, but we will know the ratios we do get quite 
> precisely.
>
> Joe Gwinn
>
>   
Joe

Thats essentially what JPL do with their latest DMTD system which uses a 
commercial synthesizer as an offset generator.
They use a 100MHz mixer input frequency and use a 123 Hz beat frequency.

Bruce

Bruce