Showing results for 
Search instead for 
Did you mean: 

G68.2 for Hurco

Solution Partner Experimenter Solution Partner Experimenter
Solution Partner Experimenter

The G68.2 for the Hurco post I am writing does not use the standard PRY output.

My understanding of the standard PRY is that it calculates the rotation about the first axis and then calculates the rotation of the second axis based on the orientatin after the first axis has been moved.  similarly, the third axis rotation is calulated fron the postion after the second rotation.


The Hurco does not work like this.  instead each rotation is calculated from the static or original coordinate system

The hurco also uses ABC instead of IJK, so dont be confused by the g68.2 line examples below


In the attached image, the G68.2 line output using standard NX euler type output (in this case XYZ)  to get the coordinate system to rotate to the 45 degree anged face is G68.2 X0 Y0 Z0  A180 B-45 C0  However the machine requires the following result G68.2 X0 Y0 Z0 A0 B135 C180.  There is an image attached of the machines controller simulation when using the A180 B-45


In NX I am using auto_3d to define the tool orientation, which uses the following rotation matrix to make the A180 and B-45 calculation.  How do I get the output from this that the machine requires or is there another / better way to get the required output?


rotation_matrix m0 is 0.70710678118654757
rotation_matrix m1 is 0.0
rotation_matrix m2 is -0.70710678118654746
rotation_matrix m3 is 0.0
rotation_matrix m4 is -1.0
rotation_matrix m5 is 0.0
rotation_matrix m6 is -0.70710678118654746
rotation_matrix m7 is 0.0
rotation_matrix m8 is -0.70710678118654757



Re: G68.2 for Hurco

Gears Phenom Gears Phenom
Gears Phenom

I dont know hurco but:


G68.2 - it should be euler angles - Z'XZ''  (P omited - default P0 option)


G68.2 P.. - there should be many options how it can interpret IJK addresses - projection, direct, spatial, vector, additional, points, ... as well.


other example:

G68.2 P1 Q123 - P1 some kind of angle interpretation, Q - angles order in IJK


(sorry I if discribe something wrong)

#♫ PB, 5ax, itnc, nx, vericut ♫ #

Re: G68.2 for Hurco

Like Juraj mentioned - if I understand you right it is equivalent to P1 Q123 option on Fanuc. I haven't looked for this in the OTB NX12 postbuilder code but it may not be implemented (they seem to prefer Euler.) The problem with Euler is that you can have errors from the translation portion of the transformation (slight angular difference can mean a lot if far away from origin.) For G68.2 fanuc engineers recommend that the origin be moved close to the work. Anyway maybe there is an example in OTB posts. Look at Sim posts - particularly 09 and 15 - that seems to be where the latest post development was done in postbuilder (they concentrate mostly on PC for higher than 16.) Math libraries included in (I think) ugpost_base.tcl have routines for matrices. Between that and what you find in those posts or in custom commands I suspect all of the math code is available. Unfortunately there is more to do than just to get a good matrix. Then you have to get all the XYZ output in terms of it and it also may be wise to turn the rotaries before activating (to be sure how you get there,) This is probably also demonstrated in OTB code. So - I suspect the code is not written OTB but the pieces are there.
Windows 7 Pro

Re: G68.2 for Hurco

Siemens Legend Siemens Legend
Siemens Legend

Attached is some documentation from Hurco about G68.2. I first checked with the representative who gave this to me if it was ok to upload this.


Let me know if this helps.

Re: G68.2 for Hurco

Solution Partner Experimenter Solution Partner Experimenter
Solution Partner Experimenter

I have now found the soluton to the problem.




The Hurco G68.3 works in the same way as a standard fanuc G68.2 but with XYZ  order.



Learn online

Solution Information