[time-nuts] Z3816a and newbie questions

[John Ackermann N8UR]

Rex wrote:

>A couple weeks ago I obtained a Z3816a (it has an MTI 260 OCXO) and set
>it up. I also got a copy of GPSCon software which I have been running. I
>first brought it up just after Christmas. It locked up fine in less than
>an hour and I think it seems to be working ok. Just in the last few
>days, the EFC seems to finally be nearing horizontal.
>
>Specs I have seen on the Z3801a for 10 MHz output are only 1 x 10^-9 per
>day. Looking at a link to NIST from www.realhamradio.com it seems that
>something around 10^-12 might be closer to possible.  I almost always
>have 8 satellites; never seen less than 6. 
>
>About what accuracy can I reasonably expect from this unit? Is there
>anything in the GPSCon statistics that gives me a better real grasp on
>accuracy of my 10 MHz?
>  
>
Hi Rex --

Tom already posted an excellent answer to your question, but I'll add a 
bit more.  The GPSCon software generates a lot of information and some 
really pretty pictures, but as Tom said, you need to compare against an 
external reference to see what the unit is really doing.  I'm currently 
measuring two Z3801As against each other, and there is a plot (updated 
automagically every 15 minutes) of the results at 
http://www.febo.com/time-freq/plots/z3801a.html.

Anyway, as I write this I have about 5.75 days worth of data, and during 
that time the maximum difference between the two units was about 58 
nanoseconds.  The frequency offset between the two is about 1x10e-14.  
Over time, of course, that offset should reduce to zero as both units 
are being steered by the GPS system.  I've been doing time comparisons 
of the 1pps output averaged over 100 seconds, so my data won't show any 
noise at shorter averaging times -- that's a limitation of my 
measurement system that I'm working on correcting.

I am pretty sure that one unit is contributing quite a bit more of that 
58ns of noise than the other .  I plan to soon compare each unit against 
a Cesium standard so I can determine their individual performance.

If you're interested, I also have some other data comparing five Z3801As 
over about one month (using only the internal statistics gathered with a 
Linux program similar to GPScon) at 
http://www.febo.com/time-freq/gps/z3801a/internal/index.html.

I don't really know whether the Z3816 has better performance than the 
Z3801A or not; I suspect that in GPS-locked operation, there's not much 
difference, though with the different OCXO the unlocked performance 
might be quite different.

Welcome to the gang!

John