Hi On Sep 16, 2013, at 9:48 AM, Jim Lux <[email protected]> wrote:
> On 9/16/13 6:03 AM, David J Taylor wrote: > >> >> I am thinking about exact time measurement - getting your PPS edge >> exactly on the nanosecond. People can add in the length of the cable as >> an offset, so they must also need to enter any delay through any >> filters, mustn't they? >> >> Agreed that for position alone it doesn't matter as much. It's the >> antenna's approximate position which will be measured. >> >> Your points about dispersion in the filter, and temperature coefficient >> of delay are good ones. >> > > > It's trying to get nice flat group delay in the filter that causes all the > issues with Light Squared. *small*, *inexpensive* brickwall bandpass filters > tend not to have nice delay properties, or at least ones that are temperature > stable. Spectrum regulators know this, of course, and assign adjacent > services accordingly. > > > If you're only worrying at the few nanosecond level, you probably don't have > to take into account continental drift (periodic resurveys of location to > account for several cm/year?) and solid earth tides (on the order of 30-50 > cm). And, really, for a lot of applications, you're interested in relative > timing, so the solid earth tide shift of 1-2 ns every day isn't a big deal. Well maybe it is…. If 1) Your antenna is ~ 2 ns out of position 2) you have a small number of sats 3) Everything else is doing very well Then as you take sat's in and out of the mix, you will could get 2 ns more "pop" than you would have otherwise. Bob > > > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
