2010/6/5 Andy Pugh <[email protected]>:
>
> You would create a new file viesturskins.c (and possibly a .h) and put
> it in the src/emc/kinematics/ directory, then recompile. The module
> would then be available to load in HAL. You might need to add it to a
> makefile or header file somewhere. Hopefully someone with a bit more
> knowledge of the build process can advise here.
>

With some help from Przemek Klosowski I now have downloaded the source of EMC.
I looked in .../emc2-dev/src/emc/kinematics folder. There are .c and
.h files for different kinematics modules.

1) Is there some source of information, where I can find out, what
does each file do (cubic, genserkins and pumakins has both - .c and .h
files, 5axiskins, gantrykins and scarakins - only .c file) and what is
the difference?

2) Where can I find some information about the syntax, used in the .c
and .h files? I encounter the word "double" in beginning of so many
lines, that I have a suspicion that it is not meant to be mathematical
action. I feel that my lack of programming skills is making me
reconsidering, if I ever want to try this :)

3) Where are first two files, mentioned below? They are mentioned also
in several .c files, but I do not see them in this folder...
#include "rtapi_math.h"
#include "posemath.h"
#include "genhexkins.h"
#include "kinematics.h"
And what is the purpose of these files? What does each of them define/control?

4)
>
> My own preference (which is irrelevant, as it isn't my machine) would
> be for a kinematics module that takes X, Y, Z, A where the A is an
> angle set by the Gcode, but isn't directly related to the conventional
> A axis, but is in fact a kerf-angle or cutting angle, with positive to
> the right of direction of travel, and negative to the left. The
> kinematics module would measure the instantaneous X and Y velocities
> to determine the tangent, and do the maths from there. Output of the
> kinematics module would be pass-through of X, Y, Z for the X1 X2 Y Z
> joints, and calculated positions for the two head joints.
>

Currently it seems to me that this type of solution can manage vast
majority of my needs for 5 axis cutting (assuming that I can change
the angle of tilt with appropriate G-code commands for different
G01/G02/G03 moves or even during them).
Any ideas, how to implement this? It seems to me that this whole thing
has to be separated in following parts:
a) kinematics for A and B rotary axis, where offset from end of nozzle
to center of rotary axis is handled

b) implementing tangent calculation:
> The kinematics module would measure the instantaneous X and Y velocities
> to determine the tangent

c) implement A and B move calculation to keep tilt angle perpendicular
to movement direction

Are there any examples, where something similar has been achieved?

ANY help, tips, advices, opinions or commentaries will be greatly
appreciated :))

Viesturs

------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to