[time-nuts] D-PSK-R wwvb thread

paul swed paulswedb at gmail.com
Thu Apr 24 17:04:57 EDT 2014


Back from travel. I hijacked another thread and that is bad behavior. Sorry.
So here is the thread all by itself.

Background
To remove the new wwvb bpsk so that old phase tracking receivers can work.
As a by-product of that effort you do obtain the new format data from wwvb.
Already released are traditional analog and pseudo digital approaches along
with a wwvb RF receiver and regenerators for classic time receivers like
the Spectracom 8170 and Truetime DC60.

Pretty much all of the approaches so far look for a phase change and invert
that change if it departs from the reference.

But I have always felt there should be more of a pure digital solution. To
add to that the power that is available in the arm evaluation boards is
truly amazing along with costs that are very cheap. They include physical
boards (No soldering required) and amazing development tools for free.
Those combinations are enough to motivate at least me to learn.

To date I have experimented a bit with the STM and TI boards in both FORTH
and C with help from some fellow Time-Nuts.
It has become apparent that the TI launchpad will not be able to hit the
speeds needed to do the work. However its simplicity in many respects is
helpful in learning about the technology along with great training tools.

I am discovering that the STM Discovery board also has pretty good training
tools. They are a bit harder to find. However it absolutely has the speed
to do some real work and plenty of memory. So the real focus has to be on
that board. My challenge to date is with the basics. These processors are
very complex in how the utilize all of the IO. Every pin can do numbers of
functions like A>D, D>A, interrupt, timing, clock, and general IO. The
clock mechanisms are also very complex with PLL multipliers dividers and
such. (Somehow all of the clock stuff has to be very useful)
So for the moment I will work through these basics.

I can see a solution though. Matthias who reads time-nuts every so often
introduced me to some digital techniques on captured wwvb signals that I
sent to him that show great promise. Like 30-50 db of tracking gain. You
need that on wwvb in the summer.

He uses FORTH to program much and I may also. But because of the
documentation available for C for both the STM and the Launchpad I will
start there to come to terms with all of the base register programming of
these chips.

Thats at least the start of this thread.

Regards
Paul
WB8TSL



More information about the time-nuts mailing list