Question about full temperature compensation

Static, Dynamic, Log File Analysis, Troubleshooting

Question about full temperature compensation

Postby HeliHenkie » Wed Feb 11, 2015 8:02 am

Hi all,

I've been flying with my small quadcopter for a couple of weeks now using onboard mag calibration and tare.
My setup consists of an AQ6.1 board, DIMU1.1, KISS esc's and Quatos.
Stabilization is perfect, but I think alt-hold could be improved on. When I put my quadcopter in alt/pos hold and let it hang for a couple of minutes it slowly drops about two meters. This behaviour is logged in the GPS_HEIGHT parameter perfectly. The outside temperature in the Netherlands is around 0 degrees C so that calls for 'full temperature compensation'.

I've created a couple of static log files with a temperature range of -10 to +50 degrees and a couple of dynamic log files.
Calculations run without any problems with relative low MAE's compared to the old AIMU calibration procedure, but .....

Temperature compensation does not work that well. For all files the ACC MAG changes with 0,5 (for example from 10.2 to 9.7) over a range from 5 to 35 degrees.

What could be the cause of this?
Should I create more static logs?
Does it matter how fast the board is warmed up?
Should I create more dynamic logs?
In what step is the compensation for temperature drift done? Does it help to let that step run longer?

Hope someone can help.

Best regards,

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

Return to Calibration

Who is online

Users browsing this forum: No registered users and 12 guests

cron