August 30, 2007
about a year ago
Thanks, yeah tried this, and also increasing the baud rate. But I found the solution to my problem myself. Obviously, the GPS is only having wrong timestamps for a few minutes. When it gets a 3D position fix, it changes the timestamps to the correct value. This happens usually 2-5 minutes after the 3D fix. During the first few minutes, the timestamp value seems to be in a range of +/- 1 second or so. This is what confused me, sometimes it is ok, sometimes it was 1 second too early. But now it’s consitently fine, around 130ms late at 38400 baud.
So the only problem was, that I was not patient enought during my initial tests.
These work really great for positioning. I wanted to use it, to get an exact time, but this seems to be impossible. The 1PPS pulse is perfect and very exact, but the binary as well as NMEA timestamps are shifting by up to 1 second. This makes it impossible to assign the exact seconds to a 1PPS pulse. My Logic analyzer shows, that sometimes the timestamp is around 100ms to 500ms late (would be ok I guess), but then it is 500ms to 1000ms earlier, then the actual time. To be precise: After each reboot, the modul seems to have a different time offset to the actual time.
Any help as to how I can assign the exact time to the pulses would be much appreciated. Im currently guessing, that this is not possible with a gps module.
No public wish lists :(
Forgot your password?
No account? Register one!