Cancel
Showing results for 
Search instead for 
Did you mean: 

mom_kin_iks_usage troubles

Phenom
Phenom

Is it possible to get the correct output for helical moves without loading any kinematics from the machine model?

 

Create post with a generic "5-Axis with Dual Rotary Tables"

mom_kin_machine_type "5_axis_dual_table"
4th Axis ZX B
5th Axis XY C

Using OOTB setting (ie using new iks)

The post outputs helical moves hole milling at any angles.
With no machine model loaded the output is safe and correct.
With a machine model loaded and the C table axis is not in line with MCS Z axis the posted output is FUBAR

 

Edit PB_CMD_revise_new_iks
set mom_kin_iks_usage 0
MOM_reload_kinematics

Post only outputs helical moves at B0 and B90.
All other helical moves are output as small linear increments
With a machine model loaded in any orientation output is safe.

 

Any help appreciated

 

Thanks

 

Alex

NX9.0.3.4 MP15 & PB9.0.3

9 REPLIES

Re: mom_kin_iks_usage troubles

Phenom
Phenom

I am using generic 5 axis posts like a core for all of my posts. the best.

 

Ive noticed that NX11.0.2 doesnt output helical moves if operation is in some angle.

I am not using ISV and get output in lines.

I reported this on GTAC. PR 8930075

 

If you dont want overide kinematics from simulation model puts "return" into the top of  PB_CMD_reload_iks_parameters.

 

---------------------------------------------
#♫ PB, 5ax, itnc, nx, vericut ♫ #

Re: mom_kin_iks_usage troubles

Legend
Legend
Should be fixed in NX11.0.2 MP1

Re: mom_kin_iks_usage troubles

Phenom
Phenom

Adding return or not editing PB_CMD_reload_iks_parameters is effectively the same.

 

This leaves the new iks usage enabled.

 

I want the output from post with new iks enabled when there is NO machine loaded.

 

Many of our older parts have one workpiece body/component & blank with multiple MCS's in different orientations for OP1 OP2 etc.

 

Adding a machine model to these and then posting is a recipe for disaster.

 

As far as I can tell there is no way of telling if a machine model is loaded in the post so it can be aborted.

 

 

Re: mom_kin_iks_usage troubles

Phenom
Phenom
I am guessing by overriding "MOM_validate_machine_model" to return zero you can not load a machine if there but still use new kinematic solver.

proc MOM_validate_machine_model {} {return 0}

This would probably need to be done in a sourced file in the first read of the post tcl (user source.)
NX10.03
Windows 7 Pro

Re: mom_kin_iks_usage troubles

Phenom
Phenom
Now that I went into PB - the following should work: rename MOM_validate_machine_model "" In PB - uncomment return out of PB_CMD_reload_iks_parameters I have successfully output helical off-plane without a machine model. I know that head/table will make NX decide not to in some orientations - but table/table should be fine. Many times table/table can be used for head/table when you always G68/ROT into orientation.
NX10.03
Windows 7 Pro

Re: mom_kin_iks_usage troubles

Phenom
Phenom

I always use table-table for any 5ax machine. do you think there is some problem in this logic?

Because Ive never met som troubles with this. Can I face some in the future?

---------------------------------------------
#♫ PB, 5ax, itnc, nx, vericut ♫ #

Re: mom_kin_iks_usage troubles

Esteemed Contributor
Esteemed Contributor

Depends on how you output data, and the types of operations.

If all you do is 3+2 machining, and use G68 (or CYCLE800 or whatever), you should be OK

If you do full 5 axis machining, you may not (depending on what options you use).

I would tets full 4 / 5 axis motion & see what happens

Ken Akerboom Sr CAx Systems Engr, Moog, Inc.
Production: NX10.0.3.5 MP5 + patch/TC11.2
I'd rather be e-steemed than e-diseaseled


Re: mom_kin_iks_usage troubles

Phenom
Phenom

It seems to be OK, I am using it for years (HT,TT,HH).

Maybe there is something... but really dont know.

just thinking...

---------------------------------------------
#♫ PB, 5ax, itnc, nx, vericut ♫ #

Re: mom_kin_iks_usage troubles

Phenom
Phenom

@Study you Legend!

 

Initial tests work, I had never thought to overide system  procs that are not created by postbuilder or in ug_post_base.

 

Thanks

 

I will report back after more testing.

Learn online





Solution Information