[time-nuts] GPS W/10KHz

[Perry Sandeen]

List,
 
Wrote: You're looking for the older
Rockwell/Conexant/Navman Jupiter-T ones. Some default in Motorola binary
compatibility mode, with only 8 channels visible. Due to scarcity they are
getting way too pricey...
 
You might be better off with the newer uBlox NEO/LEA-6T,
with configurable output(s).
 
Hi,
 
Actually, I was just looking them up in response
to the list request.
 
Personally, I have several Oncore boards I bought
years ago. When I use them up and then might give the Ublocs a try.
 
TVB hit the nail on the head for my situation. I
know as much about micro-processors as a cat knows about swimming. 10 KHz loop
filters are easier to learn and tame than the 1PPS. I want to make it to use
it. Nothing is more depressing than building something complex and it doesn’t
work and you can’t figure out how to make it work.
 
Many comments made to the list about solving TN
problems assumes that one can write code and load it into a device. Many of us
can’t and then to learn it the results aren’t worth it. And microprocessor
solutions don’t always work out exactly. An example:
 
Bertrand Zauhar, VE2ZAZ designed a DDS replacement
oscillator replacement for the HP 3586 SLV.
It was to replace the non-phase-locked IF
oscillator frequencies (13,775Hz to 17,475Hz). It is an awesome effort.
 
But herein lies the problem. Depending which model
one owned, and which production run it came from, you chose one of five
different code files.
 
After you load the code the 3585 still has errors
of .1 to .2 Hz and change.
 
So what does a poor slob like me do when I have
five 3585’s and they have several different frequencies for the IF oscillator?
 
My solution was to spend $250 at a hamfest for a
good working HP 3336. I then disable the HP 3586 IF oscillator; plug in the
3326, set my frequency and Wallah. I’m ready to go W/O an offset error. If I go
to another 3586 with a different IF oscillator, I just change the dial and I’m
back in business again.
 
Now, before someone flames me, please read this
twice: Burt’s solution is great for someone who 1.) Knows how to AND likes
programming microprocessors and 2.) The DDS error is OK for their purpose.
 
For me, my more expensive solution was 1.) One I
could do it with my limitations and 2.) It had the accuracy I desired.
 
As TVB stated: we have to work with the
limitations we have.
 
When I make my GPSDO, I will go for the simplest
technology I can find to meet my needs. And I want to make it as simple and
reliable as an AK-47
 
Fortunately I have a good supply of HP 10811 and
Lucent Xtal oscillators and a couple of Lucent Rubidium’s for test subjects. I
have some old HP 1U cases for building complete standalone GPSDO units.
 
All I need now is for us to sell our house to be
able to move to our condo in CA and fire up ye olde soldering iron.
 
Regards,
 
Perrier