[time-nuts] Re: Phase microsteppers

[Carl Walker]

Hi Brooke:

The micro-steppers in question used for LORAN timing were indeed the
2055A units, if memory serves. I don't believe they supported 10 MHz,
but I didn't really pay much attention since the LORAN timers only used the
5 MHz output of the Cs standards for all requirements.

A 1 nS step size wouldn't be too difficult to accomplish in a home-brew 
unit.
Finer resolution would be nice as you suggest, and is worth considering. 
Phase
step insertion could easily be made programmable with a minimal processor,
and that's what I was thinking about implementing. I'd rather write code and
control the unit via serial port than cut a big square hole in the front 
panel for
thumb-wheel switches ;-)

I'm going to have to consider going further with the design effort for 
one of these,
as I believe it would be a useful addition to the tool kit here. If I 
could find one
commercially (at surplus prices) I'd rather go that route, but judging 
by what I've
seen out there and what others have said I don't think there are too 
many of them
circulating on the market. Googling doesn't turn up much, either.

-Carl, WA1RAJ


> Hi Carl:
>
> In an old Austron brochure is says their model 2055A Phase Microstepper
> works with either 1 or 5 MHz signals (no mention of 10 MHz).
> The adjustment range is from 0.00000 ns/second to +/- 9.99999 ns/sec.  
> (i.e. 6 decades).  "Exact time base frequency adjustments having a
> resolution of 1E-14 may be made."  Note that 0.00001 ns/sec is 1E-14.   
> It also has provision for changing the phase at a constant rate
> determined by the thumb wheels on the front panel.
>
> When I was trying to make a 1 PPS saw tooth corrector for the Motorola
> GPS receivers I found an IC that was a 1 ns phase shifter and had just
> enough range to correct the GPS receiver.  I did get it to work but
> never could figure out how to get around the bug in the saw tooth data.  
> This type of chip would be one way to get down to 1 or using another
> part number a fraction of a ns time step.  But some other method must be
> used for finer steps.
>
> 73,
>
> Brooke Clarke, N6GCE
>