[time-nuts] RasberryPi, timing and GPS receivers
attila at kinali.ch
Wed Oct 17 15:27:28 EDT 2012
On Wed, 17 Oct 2012 11:42:06 -0700
"Tom Van Baak" <tvb at LeapSecond.com> wrote:
> > A new LEA6-T single piece cost IIRC 190EUR from the webshop.
> > A LEA5-T was IIRC half price.
> > And a little caution: the LEA5 modules use considerably more power during
> > aquisition than the LEA4 did and the LEA6. IIRC we measured up to 150mA
> > (compared to <100mA with LEA4). I don't know the Rasberry, but i guess
> > that should be not a proble, but you should check for it nevertheless.
> I'd say the u-blox 5T or 6T is rather over-kill for George's
> application, both in price and performance (and learning curve and
> integration). Remember that the world of NTP is milli- and microseconds.
> State-of-the-art GPS timing receivers and GPSDO are in the nano- and
> picosecond accuracy/jitter range
Yes, that's why i explicitly mentioned the LEA4-S, which already is
way more accurate than ntp can get over the network. A LEA5-S sells
for 16USD on ebay. For 30USD you already get a soldered board with an
NEO6-M and patch antenna. All you need to do to use those is to solder
a wire for the PPS pin and you're set.
With such little money you get a top modern GPS module with high
sensitivity (even the LEA4 could get reasonable indoor reception
with a simple passive patch antenna, and the LEA6/NEO6 are much
better) and more than good enough PPS output (<60ns w/o sawtooh correction)
At these prices, i'm surprised that not more people are using
the u-blox modules for their stuff.
There is no secret ingredient
-- Po, Kung Fu Panda
More information about the time-nuts