Greetings,
I am running a KR6-2 with a Fronius welding torch/set up and recently my TCP has begun to wander off path.
For example, after teaching a new assembly I am able to get about 4 hours of operation before the TCP has eventually drifted 3mm, causing the torch nozzle to rub against the assembly triggering the crash box/a collision event (at low speeds of 8mm/sec). (We weld inside some rather tight steel pockets). The drift is gradual, as I can plainly see the weld positions shift with each assembly.
The torch, crash box & assembly fixture (rotisserie) have been checked and are in good order, all the weld assemblies are fit to spec.
The coordinates have not changed, but the physical space the TCP reaches is incorrect.
A set of axes mastering checks over the course of several hours of running the arm back and forth through empty space. I was hoping to find a trend in the axes deviation but I didn't. Plotted on a graph, axes 5 & 6 just fluctuate wildly, up to +/- 0.04° (well outside the expected 0.002°).
To make things weird and I absolutely can't imagine why, but when I take the TCP to it's reference point (close to the robot base/root), it's pretty freaking close. As if the TCP is only off when the arm is stretched out.?
Any thoughts?
Thanks in advance,
Chris