BudWalker
First Officer
Flight distance : 5966247 ft
United States
Offline
|
cameracollector Posted at 3-23 14:29
Okay I just uploaded the two most recent FLYxxx.DAT files, since the timestamps did not correspond to anything else I had. Hopefully they have the flights where the GPS kicked out on them.
I installed the DATCON program so I could look at them, but even though I have the latest java, I still cant open it.
FLY014.DAT shows the problem you were referring to. Looks to me like the compass wasn't calibrated correctly. Have you attached an external device such as a tracker?
In the following plot magYaw is derived from just the magnetometers. Yaw is the value computed by the FC and is derived from the accelerometers, gyros and, to a lesser extent, the magnetometers. Ideally Yaw - magYaw should be a constant 13.89 ° since the geomagnetic declination is 13.89°. But, in this flight Yaw-magYaw varies with Yaw over a 2π period, a clear indication that the compass isn't calibrated correctly.
As Mark The Droner explained the several different data are required for location solution. In this case the FC can see that Yaw and magYaw differ, and therefore, can't be confident in the location solution. That confidence level is expressed as gpsLevel or gpsHealth and has a value in [0, 5] with 5 being the highest. When gpsLevel drops below 4 the P4P FC will switch to ATTI mode, as it did in this flight. Note that the numSats wasn't the problem.
As you might imagine gpsLevel or gpsHealth is a very confusing label. That's why DatCon uses the label NavHealth.
|
-
|