Calibration: MAG values outside of the range

[Mlozano]

In order to improve the results that I exposed in the post:

viewtopic.php?f=35&t=2997

, I have done three statics calibrations in the same place where I did the calipso dance (travelling 10 min with a small isothermic bag). The temperature scale is -12 to 55 ºC.

The MAE Value of these file are better than before (MAE 0 = 0.000124, MAE 1 = 0.000123, MAE 2 = 0.000135) and the IMU_RATE graphs are similar each other.

The HUD is better leveled when the AQ is on the frame than when isn't onboard

no_onboard:

HUD no onboard

on the table

on the table

on_board:

HUD_onboard

on the frame

on the frame

Also I was expecting to improve the MAG and ACC magnitude in any orientation, but not. Now ACC magnitude is 9.6 in any orientation (its ok but not great), but MAG magnitude it depends of the orientation (2.4; 2.6; 2.4;2.2), far away of the range 1.95 to 2.05.

Maybe the ESC4in1 that I have below the AQ board is the responsable of the MAG Value desviation.



The summary of the calculation can be downloaded from:



I would appreciate any help. I'm exhausted with the calibration process...

Thanks...

[Mlozano]

STATIC RATE

[Mlozano]

STATIC ACC SESSIONS

[Mlozano]

STATIC MAG SESSIONS

[Mlozano]

DYNAMIC SESSION

[Mlozano]

TEMPERATURE COMPENSATION

[LordPikaChu]

Have you tried flying your copter with this calibration data? I've had similar issues where AQ couldn't find the correct level and was drifting badly and I blame MAG noise coming from power distribution and other components for that. What I can tell you is that having ESCs directly below AQ is definitely not a good idea, at least until the DIMU (which supposedly lets AQ not use MAG data form flying) comes out.

[Mlozano]

No, I didn't try to fly because I'm a newcomer and I don't know the consequences of the MAG desviation. To take a desicion, I need to confirm that the statics and dynamic files are ok. If the files are not ok I will try to improve them, if not maybe I will try to fly.

What's the effect of mag desviation? Is it a great problem in order to fly?
Can I control the problem with the navigation PIDs? Can I fly it with some security?

In my opinion, the problem could come from an interference because the results with differents files are similar.

Do you know a way to confirm that possibility?

In that sense, last night I was trying to compare differents MAG logs with differents positions and situations: with and without electronics components and with and without frame. Before replacing any component, If I can confirm the interferences, I will try an EMI shield.

Please, see this link:
http://diydrones.com/profiles/blog/show ... 4#comments

Unfortunately, now I have a more serious problem because I can't comunicate with the board. If I can resolve the problem I will try to finish the test. I would appreciate if you could help me to find out a solution. viewtopic.php?f=27&t=3033

Thank you for your answer,

Mlozano

[chschmid]

[Mlozano]

I'm trying to connect with 115200 baud rate, but nothing. Always I connected with 115200 baud. I have checked the continuity with a tester and the solders are ok, but I will solder the pins again. Sorry for my english but I don't know what you say with "Maybe you can try Putty to see if you can get a connection".

Thanks
Mlozano