[time-nuts] Garmin GPS 10 LVC - no position hold.

[Mark Amos]

Hal,

Yes, your assertion about the Garmin position hold question was correct.  I contacted Garmin a 
few months ago on this question and I've included the response I got from Brad Wiseman below 
including some other interesting information.

I ended up using Lotus M12Ms and have been happy with them.

Mark

-----------------------------------------------------------------------------------------------
-------------------

Dear Mark,

The GPS 18 LVC is not optimized for timing accuracy in that manner. Going from 3D to 2D will 
not help unless you constrain the altitude in 2D operation to be the correct altitude.

One thing that would likely help is the use of position averaging, which is controlled by 
field 3 of the PGRMC1 sentence. See section 4.1.4 of the GPS

18 tech specs that you can download from here:

http://www.garmin.com/products/manual.jsp?product=010-00321-05 
<http://www.garmin.com/products/manual.jsp?product=010-00321-05> 

On the other hand, doing this is not likely to help much. Here's why:

The GPS 18 LVC's measurement pulse output signal is based on a reference oscillator whose 
period is approximately 61.5 ns and we are simply gating the correct rising edge of that 
oscillator through to form the rising edge of the measurement pulse output signal. That means 
that the best we could ever do, jitter-wise, is about 40 ns. From a practical point of view, 
we do about 150 ns rms with about 50 to 100 ns rms jitter with a clear view of the sky and no 
nearby interferers. Compare that to the 10 ns or so improvement that you could get by reducing 
the position error to 0 and you'll see that the 10 ns improvement is nearly inconsequential. 
The error processes in timing and in positioning are nearly independent of each other so they 
would add with each other in an rms fashion. So, 150 ns is the result of that addition. That 
means that if you completely removed the 10 ns error that corresponds to positioning error, 
you still would have 149.67 ns rms oferror. I claim that you'll never notice the difference 
between 150 ns rms error and 149.66 ns rms error.

Regards,

Brad Wiseman"

-----Original Message-----

To: 'Wiseman, Brad'

Subject: RE: Tech support request for GPS 18 LVC OEM (hardware only)

Brad,

I have seen some specs for other GPS's that have a "position hold" function

- where the known position is held constant and processing is optimized for precise time.

Is there such a function with the GPS 18? I assume that going from 3D to 2D would have a 
similar effect. One other GPS calls "position hold" "0D"...

Thanks again for your help!

Mark