[ntp:questions] Re: Good and cheap GPS receivers (was: Reference clocks - which?)
Sun Oct 31 00:57:09 UTC 2004
"Wolfgang S. Rupprecht"
<wolfgang+gnus20041030T154422 at dailyplanet.dontspam.wsrcc.com> wrote in
message news:x7hdobdd5b.fsf at bonnet.wsrcc.com...
> Bjorn Gabrielsson <bg at lysator.liu.se> writes:
>> "Wolfgang S. Rupprecht"
>> <wolfgang+gnus20041030T140650 at dailyplanet.dontspam.wsrcc.com> writes:
>> > One thing to keep in mind is that Garmin GPS's which will extrapolate
>> > for 30 seconds whenever it loses lock. That tends to give a false
>> > impression that things are working better than they really are.
>> > Graphing the time (or positional) jitter will show you that things
>> > aren't working all that well.
>> Does your Garmin extrapolate on the sky-plot? (Navigate is my wording
>> for "at least 3-4 sats" in previous message)
> The sky plot position is calculated from the almanac as far as I can
> tell. The signal strength bars at the bottom (and skyplot
> highlighting) are real-time and only lightly filtered. Covering the
> antenna with my hand has them go to zero within a second or two. The
> important thing to note is that the signal strength bars on the 5-6
> best satellites are only full-strength when I go outside. Inside the
> best bars are mostly "in the mud" and the rest are non-existent. I'm
> told that the bars aren't really signal strength in the classical
> radio receiver sense, but are really the signal-to-noise ratio.
> Having a low bar means one is seeing a very low signal to noise ratio
> and the recovered data and calculated x,y,z,time will be very noisy.
> I do believe that is true, given some 24hr position plots I've made
> with the gps while it was inside. None of my garmin gps's will output
> time well enough to even speak of. (They have seconds worth of time
> jitter even when outside and locked to many good satellites.)
> I certainly am a big fan of GPS's and think everybody should get one.
> I just have heard way too many horror stories about folks buying a gps
> for their computer room and then being horribly disappointed when it
> didn't work when they put the antenna near a window. They seemed to
> expect it to be more like an am/fm radio where they could make it work
> by just moving it near a window.
In a lot of cases placing the GPS near a window works fine if you don't need
accurate positions. The person I got the modified jupiter driver from said
he had his on his window sill in his apartment for a few years and it worked
reasonably well for time. There is not a lot of difference in the number of
satellites in a 90 degree cone directly above or out the side window. In
fact a window position should give you a 180 degree panorama from horizon to
There is a caveat. I recently replaced the windows on my house. They have
low E rating and it blocks the GPS signal completely. I get a better signal
in the middle of the room then near a window. I suspect most new houses in
colder climates could have the same problem.
More information about the questions