G68.2 P1 I,J,K format.

[jon5aunders88]

Hi. 

I'm trying to configure a post to output G68.2 P1 with the correct I,J,K format for my machine. 

Machine is a Doosan DNM6700 with a 3+2 Nikken rotary B and C axis. 

what works on the machine is to simply have the I at zero and for the J and K to match the Current B and C axis positions, as below; 

G00 B90. C360.
G68.2 P1 Q231 X0. Y0. Z0. I0. J90. K360.
G53.1

I have tried every combination of the Euler parameters from the post training guide. For example;

var eulerConvention = EULER_ZYZ_R;

But unfortunately none have output the code correctly. 

Has any one come up with a solution to this? I have tried other threads on this topic but so far haven't found a solution. 

Is there a way to force I to zero and ,J,K to current rotary axis positon?   

[serge.quiblier]

Hi @jon5aunders88 

have you tried the convention XYZ_S ?

Because ZXZ_R is the classical "true" euler convention using a regular G68.2 (without P1) IJK on a Fanuc controller.

Regards
______________________________________________________________

If my post answers your question, please click the "Accept Solution" button. This helps everyone find answers more quickly!

[jon5aunders88]

Hi. 

Thanks for your reply, I have tried every convention in the training manual for; 

var eulerConvention = EULER_???

Unfortunately nothing seems to have output the code correctly. Any suggestions welcome. 

thanks. 

[serge.quiblier]

Hi @jon5aunders88 

reset the euler convention to it's initial setting EULER_ZXZ_R

Then search for the setWorkplane function, for the code outputting the G68.2

Then alter it this way :

      var machineABC = abc.isNonZero() ? getWorkPlaneMachineABC(currentSection.workPlane, false, false) : abc;  // added
      writeBlock(gRotationModal.format(68.2), "P1 X" + xyzFormat.format(0), "Y" + xyzFormat.format(0), "Z" + xyzFormat.format(0), "I" + abcFormat.format(machineABC.x), "J" + abcFormat.format(machineABC.y), "K" + abcFormat.format(machineABC.z)); // set frame  // modified code
      skipBlock = _skipBlock;

Check the code, and test carefully on the machine.

Regards

______________________________________________________________

If my post answers your question, please click the "Accept Solution" button. This helps everyone find answers more quickly!

[jon5aunders88]

Hi Serge 

I have tried the code you mentioned but its now failed to post. 

Can you check the format below is correct please?

Have you been able to test these changes with the standard Doosan 3/5axis post? And if so was the G code matching the I,J,K output I requested in my original message?   

Many thanks. 

[serge.quiblier]

Hello @jon5aunders88 

I have tested the code in a modified Fanuc mill post processor.
As I don’t have your Fusion file, I used something else, and check the values in the G0 and G68.2 lines.
If you share your post, and your Fusion file, I can test with the same elements.

Regarding the Euler convention, I have overlooked the Q231.
It is important as it's a clear indication of which Euler convention is used.
Take a look for example at this article:

https://www.linkedin.com/pulse/fanuc-g682-5-axis-tilted-work-planes-tim-markoski

Q231 should be EULER_YZX_S normally.

Are you using a .machine file also? Or the kinematics is hardcoded in the post?

If and when you have an error when posting, sharing the error message can help us guess what is going on.
I am not in front of your computer, looking over your shoulder.

Regards

[jon5aunders88]

Hi Serge. 

I have managed to get a post to output what I wanted. I ended up editing the Fanuc.cps as the Doosan with Fanuc.cps would not post with the edits mentioned above. 

Thanks for the help, everything seems to be working with the edited Fanuc.cps.  

[serge.quiblier]

Hi @jon5aunders88 

do you means the "doosan vmc fanuc.cps", or really  the "doosan with fanuc"?

The first one is the library post, the second one can be any user edited or renamed post.

Can you give me the forkid for this post?

I will check the configuration, but normally these post should be derived one from another, so, they should behave similarly normally. (but they may have different ways of settings things)

The line I am interested in is that one :

FORKID {XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX}

Regards

[jon5aunders88]

Hi Serge. 

Apologies it’s taken a while to get back to you. It was the "doosan vmc fanuc.cps" from the post library. 

I have tested my edits and now have a working post. 

Thanks 

Jon.