[time-nuts] Accuracy/drift of Garmin GPS 16 HVS 1 PPS output under invalid fix conditions...

[Bob kb8tq]

Hi

The spec sheet from Garmin states that the “time output is only correct when there is a GPS lock” (or something 
close to that).  I doubt that comes as a surprise. That strongly suggests that holdover was *not* a feature they 
worried about on the design. Without an intentional “design in”, holdover likely is highly variable. The ~ 1ppm 
stability required might happen by chance this time / in this environment  and not happen next time / in that 
environment. Your unit might do one thing, my unit (if I had one) could easily do something radically different. 

The 1 us accuracy stated in the spec sheet is what you normally see on crystal (as opposed to TCXO) based 
GPS units.  Again another indication that holdover (at the level you are looking for) could be a problem. The 
much looser temperature stability of the XO could mess things up a lot. 

I’m not saying that the device will not work at the level you are looking for, only that I would be a bit surprised if it does. 
I’m simply trying to read the tea leaves after working with these sort of devices for the past few decades. There
is not much to go on, so I am very much guessing. There’s not much here to trust so you are left with verify ….

I think that’s about as close to a “helpful” answer as any of us have (Yes, I realize it simply suggests you are unlikely 
to get there with the unit you have ….. thus not exactly helpful …. sorry about that). It also is suggesting that if
you do test it, it’s likely to vary a bit. You probably should run several tests if this is important. 

Bob

> On Mar 12, 2019, at 6:40 AM, Steve Olney <t0502 at internode.on.net> wrote:
> 
> Hi Guys,
> 
> Just a reminder that the original and remaining question was "Accuracy/drift of Garmin GPS 16 HVS 1 PPS output under invalid fix conditions...".
> 
> I was hoping someone could answer this.
> 
> However, it seems the answer is "test it yourself" - which I will do.
> 
> Thanks.
> 
> Cheers
> 
> Steve Olney
> 
> HawkRAO
> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.