saturation with standard mixing table hexa+

[gionag]

Hello guys,
today i discovered a very strange thing...

i have an hexa PLUS, weight about 3.0kg AUW.
t-motor 2814/10 770kv and 11x5 cfk props. driven by esc32 with latest firmware. no calibration.

i configured my motor with the same principle as the wiki. so i ended with that:

motor desc thr pitch roll yaw
m1 FRONT 100 75 0 -100
m2 FRONT RIGHT 100 -100 -100 100
m3 REAR RIGHT 100 -100 -100 -100
...
m8 FRONT LEFT 100 100 100 100
m9 REAR LEFT 100 100 100 -100
m10 REAR 100 75 0 100

NOTE: my yaw is reversed

when i fly, i hear strange noise from the motors and looking at the log on every motor i see clipping...
if i lower the valus in the mixing table for example turning all the 100s in 50s and all the 75s in 25s... the problem seems to be gone but i think i'm only masking the real problem.

what is your experience about ?

[Max]

How does it fly? Is it easy to maintain a steady altitude in manual mode, or is it sensitive/unsteady?

In Alt. hold mode, does the motor throttle graph look like it's oscillating up/down in a steady rhythm? Or does it look more random?

You can check your motor yaw, pitch, and roll log values for imbalance (individually, to make it clearer). Balanced outputs would be centered around the zero line. It's amazing what the AQ can compensate for w/out it being obvious during flight.

And of course check the vibrations -- acc x, y, z plots.

-Max

[gionag]

really not stable in altitude in manual, seems to be sluggish in gaining altitude either on loosing... the hover point is quite impossibile to center.
tomorrow i'll try for the alt hold mode.. not tested yet.

for the imbalance what i have to check precisly ?

the vibration seems fine to me... but tomorrow i'll post some logs...

thank you

[mr_westie]

Is this another build? from your previous post you had a quad.
You will need to redo the PID tuning again for a new frame.
Can you post the log
And image of your setup.

[Max]

Try lowering your Tilt Rate D by several hundred and test. My quad was doing something very similar to what you describe and I had to lower TR D by 2000. Now no motor saturation and steady altitude For alt. hold I had to drop Nav. Alt. Speed I a bit (try lowering in increments of .1-.2 if the copter oscillates up/down in alt. hold).

Here's a graph showing what I mean about motor output imbalance. In this case you can see the AQ is compensating for what turned out to be a motor very slightly out of horizontal (on a round arm) -- notice how most of the yaw commands are below the zero line. The average should theoretically be close to zero.

uneven-yaw.png (11.85 KiB) Viewed 11083 times

-Max

[chschmid]

Here is an interesting discussion about mixing tables and theories. Maybe that helps you.


AQ can compensate many mistakes. Yesterday I found that loading a new motor mix was only stored for one side of my hexa (X) and she still fley quite well. (Left side a total of 100, right side a total of 275)

Best
Christof

[gionag]

[gionag]

now it's raining... hope later to post some logs...

[gionag]

[gionag]

here the logs...
the 100-75 one is with all DEFAULTS from r43. mixing table as stated in wiki with the motor mixer at 100 on all motors and 75 on the front and rear for the pitch.

THIS CONFIGURATION IS NOT FLYABLE AT ALL. WOBBLE OF DEATH EVERYWHERE.

the 50-25 one is with all DEFAULTS from r43. mixing table with 50 on roll axis and 25 on the pitch except the front and rear at 50.

MUCH BETTER HERE. STILL WOBBLE. BUT IN ONE ORDER OF MAGNITUDE LESS.

the only manouver is the hovering. no movements at all.
in the second log... i can fly but very twichy and not stable. but almost flyable. in the second log i eganged the alt/pos hold. look at channel_radio_5 for understand where it engage.

thanks again to all of you...