[time-nuts] Accurate Thunderbolt position

Mark Sims holrum at hotmail.com
Mon Jul 13 22:04:36 UTC 2009


While playing with getting the Thunderbolt to supply a more accurate/reliable position,  I came up with a way to filter the lat/lon/alt data in such a way that you can get a pretty darned accurate location (a quick test shows sub 1 foot accuracy should be doable).

The only problem is there seems to be no way to get the tbolt to accept and/or store that position.  The 0x32 command that sets the "accurate" position only uses single precision numbers which are good to around 5 decimal points.  The command that saves the survey position has no way of letting the user enter that position.   Does anybody know of a way to get the Tbolt to accept an externally generated high-accuracy position?

The only way that I can think up that might work is to calculate the high accuracy position,  put the unit into 3D fix mode,  wait until it randomly comes up with a fix that is very close to the desired value (which may never happen),  then put the unit into position hold mode...  craptastic...
_________________________________________________________________
Bing™ finds low fares by predicting when to book. Try it now.
http://www.bing.com/travel/deals/airline-ticket-deals.do?form=MTRHPG&publ=WLHMTAG&crea=TXT_MTRHPG_Travel_Travel_TravelDeals_1x1


More information about the time-nuts mailing list