On 10/17/2015 06:56 PM, reise tiet wrote: > Hi, > Doing with the source codes of the version 2.7.0 pre. With the puma560 > sim, when doing the linear move only specifing the XYZ values, like: > % > g00 x9 y9 z9 > g01 x14 y12 z11 > g01 x16 y10 z13 > % > I capture that the TASK issuess a constant default value of the ABC to > MOTION when everytime the tpAddLine() is called. And the ABC are changing > when moving. > > But in the end,of the move, the values of ABC are not the default values > issued. > > And the ABC's values in the end of the move are different when doing the > linear move above different times. > > Bug?
Yep, looks like a bug in the "genserkins" kinematics that the puma560 config uses. The vismach model looks like it's moving right, which means the joint positions of the machine are right, and the motion is executed correctly. But the reported axis positions of the A, B, and C axes are wrong, as you point out, which points to a problem with the inverse kinematics. > or, where's the ABC's value calculated during the move in the source code? The code for this is in src/emc/kinematics/genserkins.c. -- Sebastian Kuzminsky ------------------------------------------------------------------------------ _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
