[time-nuts] WWVB D-PSK document

Magnus Danielson magnus at rubidium.dyndns.org
Mon Apr 1 07:00:54 EDT 2013


Paul,

On 04/01/2013 04:17 AM, paul swed wrote:
> Bob and Magnus no offense taken at all good comments and thats why I
> shared. Needs other eyes then just mine. I captured the text in a seperate
> document reread tomorrow.
> Easy stuff first. Magnus the concerns on the channel are very real and it
> really does shift phase significantly.

I mearly rephrased Bob's (very valid) point.

> I do see this sometimes at sun rise
> and set. When I went to the 22K instead of the 474 K that seemed to allow
> things to track the dynamics better. I actually do not believe 22K is the
> answer. I believe its at the other end actually of the possible integration
> solution. 1M was definitly out.

One way to combat it would be to re-play the shifts from previous day, 
as this is a systematic shift. That way the loop bandwidth could be held 
lower as dynamics would be knocked down a bit to handle the diffrential. 
Surely these things would shift and be different from day to day, but 
there should be a high degree of similarity in the way that the sun goes 
up and alters the ionsphere from one day to another. Today memory and 
CPU isn't all that hard to combine, and the circular buffer of 86400 s 
doesn't take a masterpiece of software to create.

Another thing to consider is that you might want to reach for a PII^2 
loop to combat the phase-shift-range. Pre-filtering the loop is a good 
tool, but I don't think is gives much benefit here, a second integrator 
will be able to track in the higher drift rates better than a plain PI loop.

> bob you have numbers of comments I need to chew on. The TL072 cutoff well
> below the broadcast band. The other thing noted (Darn good eyes) is the
> arrangement of the 2 Xformers. Yes they are what they are its what was in
> stock at mouser. No one make IF xformers anymore. So the z is high and the
> other side is higher. Great choices. Is what it is.
> What are you going to do no one does DC anymore.

The DC transformers isn't what they used to be... (had to, a day like this)

Have you been able to fully recover the phase-modulated message?

The one thing I haven't seen is if someone has done a modification hack 
to their WWVB in order to transform them into Costas-loops. It should 
not be too hard to do. I don't expect to hear WWVB over here without 
much trouble (including separating out MSF), so I am not playing this 
game, even if I like the challenge. I need to improve my DXing 
capabilities quite a bit first.

Cheers,
Magnus


More information about the time-nuts mailing list