[time-nuts] DMTD phase shifter

[wa1zms]

bruce-

do you have a favorite ZCD schematic that you can share?

many thanks in advance!


-Brian, WA1ZMS

On Jul 25, 2009, at 9:36 AM, Bruce Griffiths  
<bruce.griffiths at xtra.co.nz> wrote:

> Bert
>
> If you intend to use a USB interface then its probably essential to  
> use
> an optical isolator or equivalent between the zero crossing detector
> (ZCD) output and the counter input to minimise potential low frequency
> ground loop problems. JPL found that low frequency ground loops limit
> the performance if the zero crossing detector outputs aren't isolated
> from each other.
>
> A resolution of better than 1E-14/tau should be relatively easy to
> achieve with a good ZCD design.
> Maintaining such resolution for long tau will be dependent on the
> stability of the mixer and ZCD temperatures.
> Mixer drift can be as large as 10ps/C or more at 10MHz.
> The tempco of the low pass filter components in the first few stages  
> of
> the ZCD will dominate the ZCD phase shift tempco.
>
> Using a linear amplifier and filter in front of a zero crossing  
> detector
> built into the counter is not the best way to implement such a system.
>
> The amplifier filter limiter chain should be designed to suit the  
> offset
> frequency.
> The goal is to increase the zero crossing slope to a point where the
> noise of the counter input circuitry doesnt contribute significant  
> jitter.
> To do this linear amplification is counterproductive, every stage of
> slope amplification needs to incorporate a low pass filter and  
> clamping
> to reduce the output noise whilst increasing the output slew rate. As
> long as the signal level is sufficient to ensure that each stage is
> driven well into limiting, linear amplification serves little purpose
> other than increasing the noise by more than necessary.
>
> Bruce
>
> EWKehren at aol.com wrote:
>> Bruce, Thanks for your comments. It is my opinion that a dual mixer  
>> system
>> can be built including five counters for a material cost of less  
>> than $
>> 200. The  computer interface would be USB. My challenge is the  
>> programming of
>> two u  processors and the software that need to be written for the  
>> computer.
>> How ever  there is so much expertise in this group to make it  
>> happen as a
>> joint effort, if  the interest is there. By integrating a dual  
>> channel counter