After switching between flight modes, Quad responds incorrectly to RC inputs

I have seen this more than once.
when I switch between flight modes, auto to position hold or position hold to altitude hold, the Quad copter will respond somewhat backwards to my RC inputs, on the aileron and elevator control stick. Stabilize mode will normally fix the problem to get the quad home safely.

latest example: on 3/25/19 I planned a simple auto mission and it executed perfectly, it was a little windy but quad handled it well. At the end of the mission it returned, hovered overhead and I resumed control with position hold flight mode. flew around a few minutes and landed.
using mission planner, changed battery, uploaded another auto mission, and executed, as the quad was flying that mission the wind was a little too much for the forward speed that was set so I used a return to home switch on my RC transmitter to bring the quad home. the return to home executed perfectly as well and the quad hovered overhead, I resumed control again with a position hold setting on the transmitter. Expecting to fly around awhile with the somewhat fresh battery, and my dad watching the monitor from the onboard camera, I realize that the quad is responding to my RC inputs on the aileron and elevator somewhat backwards, if not exactly backwards. I find myself high enough to clear most obstacles so I decide to experiment with the controls. position hold is still working fine, and I verify heading with my dad looking through the camera, yep controls are backwards.
Like I said, I have seen this happen before and have been fighting this problem for sometime.
I would love for someone to look at the telemetry/data log files to see what is happening.
I am not experienced enough to find the problem but have looked at the log files for hours and can not put my finger on the problem.
The setup is new and purchased from GetFPV They treated me good when I could not get the Neo-M8N to calibrate correctly. GetFPV sent me a new one and now there are no compass inconsistant or calibration errors.

I am flying a Holybro Pix32 “PixHawk” Flight Controller, HolyBro Pix32 Neo-M8N GPS, (which I have replaced once and this one seems to be ok), an FrSky Taranis X9D Transmitter, FrSky X4R-SB - 3/16 Channel Receiver w/ SBUS, running arducopter 3.6.6 and using mission planner.

Here is a link to the .bin file.

.bin log or it did not happened :smiley:

Thanks for the reply.
I just added the .bin file link
i see a couple of errors entries that indicate potential thrust loss, but the quad did not have any issues staying in the air or holding position.
i basically figured out what stick inputs would get it home and limped it back.
I sent it on a second auto mission, realized the wind was too strong for the battery flight time that I had and issued a RTL command. When the quad got closer and I switched flight modes, that is when I realized that it was having a control problems. It would position hold ok but the right had stick responded backwards or almost perfectly backwards.
Any help is greatly appreciated.

Hmm, if I look to the roll and pitch inputs versus desired roll and pitch and put yaw above it, it exactly looks like as super simple mode.
You also have super simple mode on ch7 set up.
However i cannot see any activation of super simple, it could be a software glitch (3.7 is in dev). Problem is that super simple mode status is not logged, just the event when it is enabled or disabled.
Try to remove option 13 from rc7_option, just for testing, and add again. Let’s see what happens.

Also perhaps because the wind, motor 1 went to max throttle and that initiated thrust loss detection. However I don’t see any correlation between the two event. (not related, but 0.6-0.7 loiter thrust output is a little high for my taste, I would say your copter is slightly underpowered.)

Again, thanks for looking at this problem for me.

I did assign super simple to a transmitter channel 7 but have not tried it out.
I will remove the option 13 and fly again to see what happens.

it was really windy and but the quad handled it fairly well.

thanks again, I’ll let you know the out come.


Not sure if it’s the same problem, but on 3.7 my quads always start in Simple mode. I reported details of this here a few weeks ago. There’s a message in the log very early, always just after the barometer calibration completes.