Thanks for all the interest. I used Mastercam at a friend of mine. The impeller CAD model was fetched somewhere from the Internet. I wrote my own post processor in Delphi - it reads the NCI file that Mastercam creates. This file has the tip and tail 3D coordinates of the tool in the toolpath in world or part coordinates. From this it was easy to determine the A and C angles for the 4th and 5th axis in part coordinates. The XYZAXkins that I wrote expects part coordinates. For the impeller the part coordinates' origin is in the center of the blank on the top surface. The post-processor also displays the toolpath and tool vector in 3D, but with the tool fixed in space and the part moving (opposite to what AXIS shows).
The setup is just a test - two rotary tables fitted with stepper motors. The bigger rotary table (red) is fitted on the machine's table with its axis parallel to the machine's X-axis (provides the A-axis). The smaller rotary table is fitted on an angle plate which is fitted to the bigger rotary table so the its axis is vertical parallel to the spindle axis or z-axis when in the A = zero position. It provides the C-axis. The machine is a Hermle 780. Its spindle is fixed and the table moves up-down, left-right, and forward-backward. In the movies it may look as if the spindle is moving up-down - this is not the case, the camera was mounted on the table. Backlash was adjusted as best I could - the rotary tables are not the best quality. I was not concerned about the accuracy, just the functional testing of the kinematics component. I was only cutting a soft material at relatively slow speed. The movies were speeded up 200% to shorten their playing times. The component actually used 6 axes. The 6th axis is used to do indexing between the blades of the impeller. The Gcode is provided for preparing the blank (1st movie) and then to remove the material between the blades at one indexing station only (2nd movie). The 6th axis adds a rotational increment for each blade so the Gcode is used over and over again with a relative indexing C-axis move between blades.. I have designed a complete trunnion table to replace my test setup. This has a 140mm C-axis rotary table built in to a yoke including a concealed stepper motor. The yoke is supported at both ends in bearings and is turned at one end by a larger stepper motor. The trunnion table will be built as soon as I have completed the conversion of my latest machine, A Hermle PF1000. If I knew how and where I could post some pictures of the machine used here and the new trunnion table. Hope this background answers most of the questions. Rudy ------------------------------------------------------------------------------ Precog is a next-generation analytics platform capable of advanced analytics on semi-structured data. The platform includes APIs for building apps and a phenomenal toolset for data science. Developers can use our toolset for easy data analysis & visualization. Get a free account! http://www2.precog.com/precogplatform/slashdotnewsletter _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
