Weird observations with 3.6.0 RC 12

I installed 3.6.0 rc12 today…man that fixed a number of problems but one observation I have is this.
Doing a compass calibration…I first noticed the beep tone from the quad was different…Not a big deal just different. But then I noticed at 51% I got the tone of your all done…but it was 51%.
so I did it again this time the your all done tone at 33%.
So I did it again…worked perfectly.

Here is the next weird item.
I ran an auto review of a log and found this message.

Test: Motor Balance = FAIL - Motor channel averages = [1338, 1319, 1149, 1124]
Average motor output = 1232
Difference between min and max motor averages = 214

Now this machine is just on the bench. It’s not flying so I have no clue why there would be a difference in output to the motors… They are not trying to balance the machine…its not moving.

Wow, same for me too on RC12 NuttX. But I also have a number of another issues with my compass so I even didn’t accent on that.

Do you have an external compass and what chip it is?

Hey there.
Yes external compass, sorry I don’ t know the chip but if I can find out I will let you know.

I have a theory about the compass calibration issues in my instance. I was using a telemetry modem and I am thinking that it was this that was causing the funny result. typically I used a USB cable but this quad is a 1 meter wheel base and so tangles up the cable all the time. So I used the telemetry modem.

for the strange Motor output I am going to try a new ESC calibration…but I am not buying that as the fix. I also notice a bad ahrs from time to time.

Copter-3.4 (and higher) supports “Onboard Calibration using RC Controller stick gestures” meaning that the calibration routine runs on the flight controller with no GCS.

In my case there was onboard calibation just initiated by MissionPlanner (wi-fi link) not by sticks.
May be GCS connection could break the onboard calibration process halfway, but that doesn’t look good.

Hey thanks ambattuhari for that suggestion.
I gave it a try and had no issues calibrating the compass. I was able to set it to external and the calibration ran perfectly. Wa hooo.

Still seeing this strangeness.
Test: Motor Balance = WARN - Motor channel averages = [1283, 1259, 1151, 1136]
Average motor output = 1207
Difference between min and max motor averages = 147
This machine is sitting on the bench not moving so you would think they would be closer then this.

No Bad AHRS now. No compass drifting with throttle up.
So I am much happier. So far…I have more to look at then maybe put it in the sky this weekend

The motor balance was “off” because the aircraft was not airborne where it belongs. The instant you arm the motors the PID controllers become active and the motors will behave erratically. BTDT.

The only valid motor bench test is Initial Setup > Optional Hardware > Motor Test, and all this test does is verifies motor order and motor spin direction.

Now that makes sense.
Thanks for that. Seems I don’t have an issue…this is goodness.
Now if I can solve the last couple issues I am good to go.

thanks again.

PS whats BTDT

Been There, Done That

Where I got my first taste of this behavior was when I transitioned from flying helicopters with Bell-Hiller Fly Bar rotor heads to flying helicopters with Fly-Bar-Less rotor heads.

With the Bell-Hiller heads it is normal to spool up slowly and use the cyclic to keep the rotor head level and then execute a “scale like” take off.

With an FBL head you also have a 3 Axis gyro to contend with. If you spool up slowly and “chase cyclic” things go wrong because the gyro is trying to keep the fuselage level and not the rotor head. So, while you are dinking around chasing cyclic, the gyro errors are building, and if you add enough collective and throttle to get the aircraft light on the skids, it promptly tips over causing the main rotor blades to mimic a high speed tiller…

This same thing happens with a multirotor. Large aircraft will tend to either pitch or roll over. Small aircraft like racers tend to flip.

Been there done that…doh…I should have known.
Thanks for your input…I get it now. Thank you