I have recently set up a new cell in PM for our KR210 -2K with KRC2. We are running Autodesk PowerMill Premium 2020 Version: 2020.0.2.2020023 with the Robot Plugin version 11.0.25.
I have mastered the robot and calibrated a long and short probe. Using the short probe I also calibrated the rotary table. All of this has nothing to do with PM, and I have no reason believe any of it is incorrect at this point.
After making those calibrations, I created the spindle and rotary table calibrations in PM.
Initially it turns out, I misread the label on the robot and set up the PM cell for a KR210 L150, but then discovered that the extension arm had been removed at some point. I discovered this when I set up a calibration test program in PM to verify the dimensions of the milled part. PM simulations were showing me collision-free toolpaths and transitions, yet when I ran the code the robot kept wanting to collide with the stock multiple times throughout multiple toolpaths in sequence. Eventually, I started measuring the distances between the robot's axes, which is how I discovered that the extension arm had been removed.
Thinking this would resolve the problem, I made all the necessary changes in the CAD and migrated these into the dmt components for the mtd files. The same stock, tool, toolpaths, and robot configurations were all used to regenerate the KRL with some orientation adjustments to account for the shorter arm. However, when I went to execute the code, I then got errors with A2 being out of range at the very first point in the program, the home position. It worked with the wrong CAD model, but then doesn't work with the correct CAD model?
I reviewed my whole MTD file setups, which were copied directly from the MTD files that were originally setup by a PM tech for our first robot. I only modified the new MTD file to accept the new CAD. These new robot MTDs all load into PM and simulate and output KRL correctly, as far as I can see and even run the robot. I have also set up configuration files, spindle calibrations, and rotary table calibrations correctly as far as I can see.
I have attached photos showing the short probe sitting directly on the center of the rotary table along with robot position values along with a screenshot of the rotary table calibration in PM so you can verify that the PM cell reflects the real cell quite accurately.
To try to understand the problem better I changed the home position for the robot so that it wasn't so close to A2's axis limit. When I ran the program, it does jog to this position as expected, but then when it attempts to move to the second position, it stops with the same error: A2 software limit, out of range error.
It turns out that the axis positions and cartesian coordinates in the robot do not align with those in PM. These photos show the initial transition program from Home to the start point of the toolpath on the robot alongside the axis and cartesian coordinates. The third screenshot shows what PM things the axis and cartesian coordinates ought to be:
You can see on Line 6 of the program HF.src in the first 2 photos, what the PM axis coordinates being sent are for that start position, then the subsequent PTP moves, then what the robot's position monitor says the axis and cartesian coordinates are. The robot is being fed cartesian coordinates from PM, but is reading those coordinates as significantly different than what PM is trying to create.
This tells me that my CAD and cell are substantially in error. Or the robot is, which really seems unlikely. But I just can't see how. When I draw arcs on the axis face locations in Powershape and look at the position of the centers, they all align according to the specification as well as what I've entered in my MTD file. When I load the robot cell and open the robot configuration window, all the values in the dimensions panel align with what I've found in the specification sheet for this robot. I have gone over my MTD file repeatedly looking for errors in the axis definitions of the machine parts and it all looks like what I'm supposed to do. I've looked over the Machine.dat and Config.dat files on the robot, all the necessary values for Machine_defs, external axis, bases, and tools look to be set correctly. I set up everything correctly in the Axis Configurator on the robot for the external. If any of this was in error, I'd never be able to output code to the robot that the robot could execute, which it can.
The problem appears to be that PM is generating coordinates from axis positions differently than how the robot does it. This has to be part of CAD/MTD setup, right? What else could it be?
I'm really at an impasse here and just don't how to begin to resolve this any further.
I've attached the PM project archive for my calibration test.
Please let me know if I can answer any more questions or clarifications.
Thank you for any help you can offer.
