Hello Guest it is July 05, 2020, 11:52:16 PM

Author Topic: Help with Formulas for robot control (ASEA IRB-60) SCARA  (Read 3404 times)

0 Members and 1 Guest are viewing this topic.

Help with Formulas for robot control (ASEA IRB-60) SCARA
« on: October 05, 2015, 11:25:39 PM »
Hi all,

New member here, and first post!


I have a large spot-welding robot that used to work on a production line at a closed down car factory. See attached. I have connected new servo amplifiers and resolvers to the motors, and have installed and tested my Windows XP PC with a Galil motion control card. I have tested all the (5) axis and all the hardware is going well. This has been a lot of work so far, as I received the robot with no instructions, no manuals and no controller or power supply! So the next step is to get it to mill something...... 


The easiest way I can think of using the robot is to use the rotating base as the Z axis, and the two arm motors for the X and Y axis. (do you follow me so far? The "X Y Table" would be vertical) Think of a SCARA robot. That's basically what I need to do.... The X and Y coordinates have to be translated into A and B coordinates of the two motor arms.

This would be to cut 2D cuts into say MDF with a router..... So I wouldn't be utilising many of the robot's full capabilities, but you gotta start somewhere....



I wonder if anyone has any experience using formulas to do this?


Another less convoluted way of asking the same question without my specific complicated details, is whether anyone has ever used a SCARA robot like this for instance http://www.directindustry.com/prod/adept-technology/product-6076-153983.html, and had any luck using Mach3 formulas to control the two motors to follow a standard X Y router table.

Cheers Haxby (From Melbourne Australia)



Offline Tweakie.CNC

*
  • *
  •  8,345 8,345
  • Super Kitty
    • View Profile
    • Tweakie.CNC
Re: Help with Formulas for robot control (ASEA IRB-60) SCARA
« Reply #1 on: October 06, 2015, 01:43:17 AM »
There has been a lot of discussion over the years regarding Kinematics and that Mach3 does not support this type of control.
To a certain extent this thread applies http://www.machsupport.com/forum/index.php/topic,29706.0.html

I don't know if it ever came to anything, perhaps you could contact the author via his website ?

I would be most interested in following your progress.

Tweakie.
KEEP SAFE !
Re: Help with Formulas for robot control (ASEA IRB-60) SCARA
« Reply #2 on: October 06, 2015, 02:03:54 AM »
Thanks tweakie. I am not after full on kinematics. Just a basic 2D version, which I think can be done using the "Formulas" input of Mach3.  Maybe even a script to change the G-code before feeding it into mach3.

I'm not sure if I am explaining myself properly. The lower arm motor and the upper arm motor would be routing the X and Y coordinates of the G-code. The Base of the robot would move the spindle in and out of the flat (2D) workpiece..... The work piece would be in place vertically........ so hard to explain this without a video! I might try to animate what I want to do to explain myself better.
Re: Help with Formulas for robot control (ASEA IRB-60) SCARA
« Reply #3 on: October 06, 2015, 09:24:49 AM »
linuxcnc has the kinematics to do this sort of thing. (joint to world and back) It is not for the faint of heart..

https://www.youtube.com/watch?v=yqmYnZS-RUs
Re: Help with Formulas for robot control (ASEA IRB-60) SCARA
« Reply #4 on: December 18, 2015, 02:23:22 PM »
I bought a robot ASEA provided you do something to control your robot
greeting

Offline BR549

*
  •  6,932 6,932
    • View Profile
Re: Help with Formulas for robot control (ASEA IRB-60) SCARA
« Reply #5 on: December 20, 2015, 06:00:39 PM »
What you are describing IS full blown kinematics (;-) Mach3 cannot process that as world point motion . BUT IF you have a CAM that you can make understand your machine kinematics then "IT" can create Gcode for you to use.  The CAM can translate the joint geometry needed to move as an XYZ machine in a vertical plane.  Mach3 can run 6 axis  rotary/linear motions . It just does NOT understand kinematics motion.

Formulas will ONLY get you basic coordinated motions not kinematic motion  and it is only all or NONE type motion. You cannot pick and chose as needed.

LCNC CAN do what you need.

(;-) TP