Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Yaskawa Motoman YRC1000 and MT1 S2HD postprocessor

#1
MT1 S2HD is not continuous rotation. I got some help previous where one of the fixes was to increase the amount of turns on this table. Measured I get +- 450 degrees before end to end stop. There need to be a rewind function in the PP as is common with 5 axis cnc mill.

Postprocessing coordinates, attatched files:
A.JBI: pulses on everything, works well. Useless to quickly edit files to quickly iterate between tests, which I do a lot with our research. Can be used together with the positioner with sync

        Puts out this
///GROUP1 RB1
///GROUP2 ST1

       Instead of this
///GROUP1 RB1,ST1

I feel like the file with one group seem to run smoother. I MAY be mistaken... but.

V2.JBI: From manually programming the robot from the controller.
      Works great with the only exception that I have to manually add a userframe and placement withing to    
      manually change this if I ever bothered to use this approach instead of removing robot sync. Not useful    
      for other then a weld here and one there.

PATH.JBI: From Cartesian only post
          Missing the ///POSTYPE PULSE part which is critical.
          As Albert stated in an other thread, it posts pulse format, just not the header for it.
          Group are separate here to.
Also be aware that mixing pulses in like this is not an relative job (rj) so it also fails there until the header is fixed

Files are cut to size for better overview so no large files here


Attached Files
.jbi   V2.JBI (Size: 1.07 KB / Downloads: 128)
.jbi   A.JBI (Size: 1.65 KB / Downloads: 111)
.jbi   PATH.JBI (Size: 1.76 KB / Downloads: 102)
#2
I believe the way to output the GROUP1 tag depends on how you have configured the turntable with the controller (if this is an option). I understand that turntables should be a sperate motion group and rails should be inside the same motion group. Can you confirm? 

Also, are you saying RoboDK should output this for turntables:
Code:
///GROUP1 RB1,ST1

Instead of this?
Code:
///GROUP1 RB1
///GROUP2 ST1
Also, I believe POSTYPE PULSE should not be output unless pulse data is exported. Did you see any issues loading this program?

Can you share your RoboDK project file?
#3
(09-08-2024, 06:30 PM)Albert Wrote: I believe the way to output the GROUP1 tag depends on how you have configured the turntable with the controller (if this is an option). I understand that turntables should be a sperate motion group and rails should be inside the same motion group. Can you confirm? 

Also, are you saying RoboDK should output this for turntables:
Code:
///GROUP1 RB1,ST1

Instead of this?
Code:
///GROUP1 RB1
///GROUP2 ST1
Also, I believe POSTYPE PULSE should not be output unless pulse data is exported. Did you see any issues loading this program?

Can you share your RoboDK project file?

For the ///GROUP  I am not sure, but it seems like one group moves smoother than two. I cannot find any data online and my dealer does not have a clue what I am talking about.
I think however that with one group the axis is moved together so they reach points at the same time. Whereas with two groups is seems like shortest rotational distance reaches its goal before other axis, not smooth at all. I have not had this confirmed but that is my take on the groups.
I am hesitant to say I want it posted differently as I have cells with several robots that would work as several groups, but I would have the option.

POSTYPE PULSE should post when pulsedata is exported. We Agree.
But if you sync motoman positioners, they always post pulse data, but not POSTYPE PULSE header. See PATH.JBI. When you sync this positioner it ALWAYS post pulse data but not POSTYPE PULSE when posting in cartesian for the other robot.

I have to remove sync to be able to post cartesian. If not it mixes mm and pulse data, and the job is also not considered a relative job. I want it to post like V2.JBI. That is the correct way.

No relevance to the robodk file, but I'll include it


Attached Files
.rdk   Yaskawa1.rdk (Size: 2.8 MB / Downloads: 132)
#4
I'm confused based on the RoboDK file you shared. I don't see the programs you are referring to. Also, did you customize your post processor?

Also, can you try adding the Relative Job flag manually to see if the program loads? I've been able to load programs with a mixture of pulse data and cartesian values without issues on a Yaskawa/Motoman robot controller.
  




Users browsing this thread:
1 Guest(s)