2.4: FBW_A and Stabilize Don't Agree on where Level Is

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Setting the plane to FBW_A on the bench

What is the expected output? What do you see instead?
I would expect to see no control surface change between FBW_A and Stabilize, as 
they both yearn to start @ level.  Yet FBW_A wants to pitch up about 20 degrees 
and roll right about 15.  Stabilize is perfect, as usual.

What version of the product are you using? On what operating system?
I'm using APM2.4.1 hardware and APM 2.4 code.

Please provide any additional information below.
I've reloaded 2.4, loaded up an older param file, and re-leveled using the 
mission planner.  All with no change.

Original issue reported on code.google.com by hagfe...@gmail.com on 13 Aug 2012 at 3:31

[GoogleCodeExporter]

I don't mind if this isn't normal... how can I fix it?  Do I reload something, 
delete something, etc?  There must be a way to fully revert the thing and 
reload it, isn't there?

Original comment by hagfe...@gmail.com on 16 Aug 2012 at 11:53

[GoogleCodeExporter]

I've figured this out.

It seems that using manual level does not update the offsets for both fbw and 
stabilize; just stab is kept current.  If you allow the plane to level 
everything at boot, both modes will agree.

I discovered this after the problem followed me from 2.4 to 2.6.  

The most effective solution is blowing away the config and starting over.  I 
simplified this by taking a screenshot of the PID's before clearing the config, 
since  that's the most time-consuming part to redo.

Then I erased the eeprom by entering terminal mode and typing 'setup erase'.  
Then I restored to factory defaults by typing 'setup reset'.  FBW and stab 
agreed after that.  I just re-entered my PID values, recalibrated my radio, set 
the modes, changed settings here and there, and done.  10 minutes, tops.

Original comment by hagfe...@gmail.com on 2 Sep 2012 at 2:58