Custom Uneven Motor Mixing

Replies

MR60

Forrest Frantz October 13, 2015 at 7:30am

did you have the quad, hex, or octa worksheet. the hex worksheet need more optimization. the other two are good.
- Joe Gordon > Forrest Frantz October 14, 2015 at 8:00am
  
  The CustomHexaCAD spreadsheet. I am using the U Hexa tab.
  
  Are there issues or just improvements you are noting for optimization?
  
  The motor matrix values corresponded to my calcs very closely so I think these are good.
  
  Working on compiling now but running into some issues with errors = but first time on this computer so I guess to be expected.
  - MR60
    
    Forrest Frantz > Joe Gordon October 15, 2015 at 1:30pm
    
    the motor factors were my best quick guess versus optimal thought given the prop directions:
    
    CW CCW
    
    CCW CW
    
    CCW CW
    
    but, prop rotation does not alternate (CW to CCW to CW to CCW) as on other ships. i'm not sure if this matters in terms of noise or effienciency.
    
    is there a solution if the props alternate is in the following?
    
    CW CCW
    
    CCW CW
    
    CW CCW
    
    if there is, then all the props move the same direction when passing each other.
    
    what are your thoughts on this?
    - Joe Gordon > Forrest Frantz October 16, 2015 at 8:52pm
      
      I have mine setup with alternating prop rotation. as in your second case. Vertical thrust factors would be the same in both cases. Torque factors for yaw might vary but I am not sure the code accounts for varying arm length. My assumption was that keeping sets of arms of equal length from COG to opposite rotation would produce the best balance.
      
      I am not seeing any ill effect of this setup. Seems to fly well.
      
      I would think this would be best for efficiency since prop tips and airflow would be moving in the same direction. I would expect this to be secondary to the overall effects of prop size and RPM etc particularly if these is a significant distance between prop tips. Also any turbulence due to props moving in opposite direction would seem to be minimized and these could be somewhat quieter although I am not sure if this is a big effect.
      - MR60
        
        Forrest Frantz > Joe Gordon October 17, 2015 at 12:50pm
        
        pitch and roll motor factors are affected by arm length (radius to CG axis).
        
        yaw is only affected by rotation direction and the radius of it's rotating mass around itself. thus yaw motor factors can be altered to remove coupling between roll and yaw and pitch and yaw.
        
        you may have minor coupling between roll and yaw where yaw will also impart a bit of roll and roll will impart a bit of yaw.
        
        i don't understand how a U can can each arm the same distance from CG. can you send me two things:
        
        1) the x/y coordinates for the left arms
        
        2) the roll, pitch, and yaw motor factors for each motor (this will tell me if there is coupling
        
        Joe Gordon > Forrest Frantz October 25, 2015 at 3:22pm
        
        Forrest,
        
        I have worked on the the calculations a bit and done some cross checks.
        
        I have added a Excel add in to allow me to use the Moore-Penrose Pseudo Inverse method to calculate the motor command factors for non symmetrical copters (also works for standard symmetrical configurations).
        
        And redone the coupling check calculations and verified with standard HES X configuration.
        
        There alternating motor rotation has a solution. Form my config it shows the minor pitch lift coupling I expected but for the case where the middle motors are on pitch COG it shows no coupling on any axis
        
        Torquw factors
        
        Rotor Roll Pitch Yaw
        
        1 0.772 -0.899 -1.000
        
        2 -0.772 -0.899 1.000
        
        3 1.000 0.074 1.000
        
        4 -1.000 0.074 -1.000
        
        5 0.488 0.899 -1.000
        
        6 -0.488 0.899
        1.000
        
        Motor Command factors
        
        Motor Roll Pitch Yaw
        
        1 -0.631 -0.924 -0.501
        
        2 0.631 -0.924 0.501
        
        3 -1.000 0.076 0.643
        
        4 1.000 0.076 -0.643
        
        5 -0.369 0.924 -0.524
        
        6 0.369 0.924 0.524
        
        Coupling Check
        
        Impact of Roll on Impact of Pitch on Impact of Yaw on
        
        Lift 0.000 0.152 0.000
        
        Roll 0.000 0.000
        
        Pitch 0.000 0.000
        
        Yaw 0.000 0.000
        
        MR60
        
        Forrest Frantz > Joe Gordon October 26, 2015 at 9:15am
        
        grea
        
        Joe Gordon > Forrest Frantz October 30, 2015 at 8:49pm
        
        I have added an octocopter sheet and it is working well. I need to add code generator for 2 added motors.
        
        Matrix calculations now happen automatically once add in is loaded in Excel.
        
        Made some changes to the graphic copter display to dynamically update motor number and prop rotation on graph.
        
        Doing quad sheet next.
        
        Joe Gordon > Joe Gordon November 3, 2015 at 7:08am
        
        Latest rendition with labels added to all three sheets.
        
        Some corrections to coupling checks that I somehow messed up
        
        I am trying to figure out whether the yaw should be normalized to the highest yaw value or the highest overall value. I think probably the highest yaw but currently is normalized to highest value in output matrix.
        
        I believe this would limit yaw authority.
        
        JSG ModCustomHexaCAD2.xlsm
        
        https://storage.ning.com/topology/rest/1.0/file/get/3702948081?profile=original
        
        Joe Gordon > Joe Gordon October 31, 2015 at 3:03pm
        
        Use this file Found same reversal of sign in Oct and Hex sheets.
        
        Now changed.
        
        Also put the S octo input into sheet and comes out with no coupling solution.

of 3

Discussions

Custom Uneven Motor Mixing

Replies

Activity

Rotor	Roll	Pitch	Yaw
1	0.772	-0.899	-1.000
2	-0.772	-0.899	1.000
3	1.000	0.074	1.000
4	-1.000	0.074	-1.000
5	0.488	0.899	-1.000
6	-0.488	0.899	1.000

Motor	Roll	Pitch	Yaw
1	-0.631	-0.924	-0.501
2	0.631	-0.924	0.501
3	-1.000	0.076	0.643
4	1.000	0.076	-0.643
5	-0.369	0.924	-0.524
6	0.369	0.924	0.524

	Impact of Roll on	Impact of Pitch on	Impact of Yaw on
Lift	0.000	0.152	0.000
Roll		0.000	0.000
Pitch	0.000		0.000
Yaw	0.000	0.000