Custom Uneven Motor Mixing

Replies

- - - - Joe Gordon > Joe Gordon October 31, 2015 at 1:18pm
        
        I have added the Rotor Layout for the quad to the workbook I have been modifying.
        
        I made a change to the torque factors as the pitch and roll were inverted in sign and for the calculation I needed them to reflect normalized motor coordinates.
        
        The matrix calculation has been added and this sheet will generate the motor factors automatically once Matrix add-in is opened.
        
        Changing the input parameters in line A flows all the way to generated APM motor matrix code..
        
        I put one of your Octo designs into the motor location coordinates on the Oct tab. It has a solution that gives no coupling effects. Any motor layout can be inputted and it will generate the solution but some may have coupling depending on the design. So one sheet for each design type Quad, Hex, Octo would be sufficient to generate motor factors and generate code. Coaxial designs can be input as well - the octo sheet will calculate matrix for a Octo. quad
        
        I tend to just input motor coordinates since other design work I do on a CAD program
        
        The dynamic labels have been added to the graph
        
        matrix.zip
        
        https://storage.ning.com/topology/rest/1.0/file/get/3702947945?profile=original
        
        MR60
        
        Forrest Frantz > Joe Gordon October 31, 2015 at 3:35am
        
        I should probably show you these. Can't wait to see your improvements.
        
        Custom Octa CAD.xlsm
        
        Custom Quad CAD.xlsm
        
        https://storage.ning.com/topology/rest/1.0/file/get/3702947871?profile=original
        
        Joe Gordon > Forrest Frantz October 22, 2015 at 8:53am
        
        Here are a couple of pics of the copter. The Excel file has the coordinates in the U tab.
        
        I may not have been clear in what I meant by equal length arms in opposite rotation. Each set of arms front, middle and rear arms are equidistant from the COG and therefore the counter toque forces for each set would balance out for equal changes in motor RPM. I think this would mean no lift coupling with yaw or yaw on lift.
        
        In this design roll would also have no impact on other axis or lift due to right to left symmetry. But pitch has some minor coupling on lift due to the center arms being lightly behind COG. For this configuration this is what your sheet shows as well. The sum rotational effect of the front and rear motors would be rotation around COG and the impact of the middle motors would be a rotational center slightly behind COG. This was a design trade off due to layout and component interferences.
        
        Since actuation on roll pitch and lift use the torque factor in the matrix if there are not pairs of equal and opposite there will most likely be more severe couplings between axis. I am sure there are solutions in design with unequal arm length pairs where this could be accounted for but KISS.
        
        Looking at the results motor rotation options you have seems to agree with this. All other combinations ended withsevere couplings.
        
        HFCCustomHexaCAD.xlsm
        
        IMG_20150825_142702_854.jpg
        
        https://storage.ning.com/topology/rest/1.0/file/get/3702887373?profile=original
        
        MR60
        
        Forrest Frantz > Joe Gordon October 26, 2015 at 10:10am
        
        great. this gives me the motor locations.
        
        so if i read correctly in an e-mail, you have changed the center motors to the pitch axis to eliminate coupling, so their x/y coordinates are:
        
        -418.17, 0
        
        418.17, 0
        
        i'll try to look at your motor factors tonight.
MR60

Forrest Frantz October 10, 2015 at 8:05am

First, define a U hexa. Is it like the U in the attached Excel file?

Custom Hexa CAD.xlsm

https://storage.ning.com/topology/rest/1.0/file/get/3702107923?profile=original
- Joe Gordon > Forrest Frantz October 13, 2015 at 7:24am
  
  Forrest,
  
  Great job on the spreadsheet. The auto generation of the code for the motor matrix is very helpful and time saving.
  
  The incoporation of the drawing tied tot he design input and the analysis of coupling factors make this great for initial design work. Thanks for posting the updated sheet.
- Joe Gordon > Forrest Frantz October 10, 2015 at 10:49am
  
  My layout is in the PDF attached. It is close to the configuration in your file.
  
  Measurements are from center of gravity of the copter.
  
  The middle motors are just behind COG.
  
  I have done calculations on the factors but when I look at the code I am seeing degrees rather that the matrix used as I have seen noted in the past.
  - Joe Gordon > Joe Gordon October 10, 2015 at 10:56am
    
    File attached
    
    Custom Hex Layout2.pdf
    
    https://storage.ning.com/topology/rest/1.0/file/get/3702722438?profile=original
Joe Gordon October 9, 2015 at 3:15pm

Forrest,

I was wondering what the status is on custom motor mixing for copters. I see you were lobbying for this but I don't think this has made it into the mainstream code. I need to set up custom mix but I think this means I will need to modify and compile code. Is this correct?
MR60

Forrest Frantz March 25, 2013 at 6:50pm

did you ever figure out how to change the motor mixing for pitch, roll, and yaw?

i'm in the final setup of a custom octa-H and need to figure this out also.

of 3

Discussions

Custom Uneven Motor Mixing

Replies

Activity