Interpretation of my calibration results

Static, Dynamic, Log File Analysis, Troubleshooting

Interpretation of my calibration results

Postby HeliHenkie » Wed Apr 10, 2013 8:37 am

Hi all,

I've been experimenting with my aq board for about 6 months now and produced about 10 parameter files for different frames. Yesterday I took some time to load these files on my board and created a couple of 1 minute logs.
Looking at these logs it appears that all the time my calibration has been off.
In none of my logs the accx and accy were exactly zero when put on a flat surface. My accz was always more than 0.1 m/s2 off from 9.806 m/s2.
This poses the following questions:

- What would be the effect on alt/pos hold?
- Will these values get better if I let the calculations run for a couple of days?
- What is the probable cause of this problem? My static or dynamic calibration?
- Did any of you get the accz sensor calibrated within 0.1 m/s2 from 9.806 m/s2?

Best regards,

Jan Willem
HeliHenkie
 
Posts: 93
Joined: Wed Oct 24, 2012 5:33 pm

Re: Interpretation of my calibration results

Postby Max » Wed Apr 10, 2013 10:20 am

Hi Jan,

Another value you can check out using the Telemetry view in QGC (with a recently released firmware) is the "ACC Magnitude". This should be as close as possible to 9.806 when the board is in any orientation. If you hold the board steady on each edge for bit, you should see this value settle down to 1g fairly quickly. In the Values Chart view, you can also see a running average of this value, which can be helpful. If you see differing values for different sides, that is an indication of poor calibration. At least this is how Bill explained it to me once :) "If you find that this number varies a lot from one orientation to another, you need to re-calibrate your sensors."

Similarly, the "MAG Magnitude" is useful for checking the magnetic calibration. Bill says, "It is should be calibrated such that the Earth's magnetic field registers as exactly 2.0."

One effect I noticed from imperfect ACC calibration was that when flying in automatic modes, my quad would slowly loose altitude as it traveled horizontally. PID changes had no positive effect, and Bill suggested checking the calibration via the above method. Sure enough, it was off... I was getting readings in the ~9.8 to 10.02 range, depending on orientation. My other board which didn't have the altitude problem gets readings in the ~9.93 to 9.78 range, which Bill said also wasn't great. :) None of my calibration attempts have been especially careful... though I do let the calcs run until the MAE stops dropping. Now that the weather is getting decent, it's time to dust off the dancing shoes again and try to get some better flights with the quad.

Cheers,
-Max
Max
 
Posts: 2814
Joined: Mon Aug 13, 2012 9:45 pm
Location: Near Ithaca, NY, USA

Re: Interpretation of my calibration results

Postby chschmid » Wed Apr 10, 2013 10:26 am

Hi Max

These comments shoud definitly be included in the wiki. And if there are more such Information, it would be great to get a hold on them.

Cheers

Christof
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Re: Interpretation of my calibration results

Postby Max » Wed Apr 10, 2013 11:13 am

Yeah... I know... any volunteers? :)

-Max
Max
 
Posts: 2814
Joined: Mon Aug 13, 2012 9:45 pm
Location: Near Ithaca, NY, USA

Re: Interpretation of my calibration results

Postby chschmid » Wed Apr 10, 2013 4:05 pm

Hi Max

If you send me more information (just in raw) I can prepare the text for you to copy/paste. Maybe after checking for proper English. ;)

Best
Christof
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Re: Interpretation of my calibration results

Postby Ruffio » Thu Apr 18, 2013 9:50 pm

MAx, for me this is realy important information "How to check correct calibration".
I use 3 AQ, and one OCTO flyes realy grate, 2nd OCTO has a little YAW problems and altitude problem, and 3rd HEXA has problems with altitude in fly forwad and big YAW problems. I couldn't find what is difference, beyond YAW in HEXA and big noise from ESC.
I checked how is "ACC Magnitude" and "MAG Magnitude" look like on all boards.
This "amazing" how well described test confirms my observations.
So what is repair process look like:
- is it enoughto do dynamic calibrations and calculations ?

regards
Ruffio
Ruffio
 
Posts: 73
Joined: Sun Nov 18, 2012 9:43 pm
Location: Poland


Return to Calibration

Who is online

Users browsing this forum: No registered users and 15 guests

cron