QGC shows incorrect heading after Onboard Magnetic Cal
Posted: Sat Jun 14, 2014 4:01 pm
Hi,
In stalled a DIMU on my AQ board. The firmware is 7.0-EXPR r399 b1758.
I looked up my magnetic inclination (66d 21m) and declination (0d 32m).
I used the AQ tool to convert it to decimal degrees: -66.35 and 0.53 (I reversed the sign on inclination as described in the documentation). I entered these 2 numbers in the Misc. settings, Mounting & environment page. The IMU rotation I kept at zero, because my AQ is mounted with the front EDGE of the PCB forward. I stored the parameters to the permanent memory.
I then performed the Onboard Magnetic Calibration and the TARE calibration. In the QGC log it shows the cal is successful, but when I point the quad to the north, the "primary flight display" in QGC shows a heading of 310 degrees. That is 50 degrees off...
QGC log also shows the settings were saved correctly, however: if I cycle power to the quadcopter, then QGC shows the heading is 068 degrees after I cycled power???
I wonder what is going on?
Here is what the log tab shows in QGC:
(MAV 228) AQ FW ver: 7.0 - EXPR rev399 b1758, HW ver: 6 rev1
(MAV 228) SUCCESS: Executed CMD: 245
(MAV 228) Parameters saved to onboard EPROM.
-> this is when I updated the magnetic inclination & declination
(MAV 228) Calibration complete
(MAV 228) Disarmed
(MAV 228) DIMU: wrote calibration parameters to EEPROM
-> this is after magnetic cal + TARE + store (all with the sticks)
(MAV 228) Radio init
(MAV 228) GPS init
(MAV 228) Nav init
(MAV 228) Command interface init
(MAV 228) GCS: UNABLE TO DECODE MESSAGE NUMBER 23
(MAV 228) Motors init
(MAV 228) Control init
(MAV 228) Control task started
(MAV 228) GPS: trying new baud rate
(MAV 228) RTC set: 2014-06-14 14:50:24 UTC
-> this is after the power cycle.
So I think I have 2 problems:
1) after the calibration, it does not show a correct heading (310 degrees when facing north)
2) after a power cycle, it shows a different heading (068)
Am I doing something wrong? Or is there something wrong with my setup or one of my parts?
Thanks & kind regards,
Johan.
In stalled a DIMU on my AQ board. The firmware is 7.0-EXPR r399 b1758.
I looked up my magnetic inclination (66d 21m) and declination (0d 32m).
I used the AQ tool to convert it to decimal degrees: -66.35 and 0.53 (I reversed the sign on inclination as described in the documentation). I entered these 2 numbers in the Misc. settings, Mounting & environment page. The IMU rotation I kept at zero, because my AQ is mounted with the front EDGE of the PCB forward. I stored the parameters to the permanent memory.
I then performed the Onboard Magnetic Calibration and the TARE calibration. In the QGC log it shows the cal is successful, but when I point the quad to the north, the "primary flight display" in QGC shows a heading of 310 degrees. That is 50 degrees off...
QGC log also shows the settings were saved correctly, however: if I cycle power to the quadcopter, then QGC shows the heading is 068 degrees after I cycled power???
I wonder what is going on?
Here is what the log tab shows in QGC:
(MAV 228) AQ FW ver: 7.0 - EXPR rev399 b1758, HW ver: 6 rev1
(MAV 228) SUCCESS: Executed CMD: 245
(MAV 228) Parameters saved to onboard EPROM.
-> this is when I updated the magnetic inclination & declination
(MAV 228) Calibration complete
(MAV 228) Disarmed
(MAV 228) DIMU: wrote calibration parameters to EEPROM
-> this is after magnetic cal + TARE + store (all with the sticks)
(MAV 228) Radio init
(MAV 228) GPS init
(MAV 228) Nav init
(MAV 228) Command interface init
(MAV 228) GCS: UNABLE TO DECODE MESSAGE NUMBER 23
(MAV 228) Motors init
(MAV 228) Control init
(MAV 228) Control task started
(MAV 228) GPS: trying new baud rate
(MAV 228) RTC set: 2014-06-14 14:50:24 UTC
-> this is after the power cycle.
So I think I have 2 problems:
1) after the calibration, it does not show a correct heading (310 degrees when facing north)
2) after a power cycle, it shows a different heading (068)
Am I doing something wrong? Or is there something wrong with my setup or one of my parts?
Thanks & kind regards,
Johan.