We have a reoccurring issue with through the arc tracking using WeldGuide III and can’t seem to find the cause. When we try to track we make corrections in the right directions but every once in a while we get a random twitch. The twitches are unpredictable and in all directions, sometimes only in 1 direction but most of the time in at least 2 about half the time they’re linier and the other half they rotate. We’ve got weld data monitoring software set up through the Lincoln PowerWave 455M so we can watch the amperage and voltage with observer, which stay stable. Most of the movement seems to happen at the sides of the weave but a few are in the middle. We can weld without WeldGuide and have no unpredictable movement. It does the same thing if the part is machine finished and totally free from scale, smoke or debris; or if it's still smoked up from the previous pass.
We've shortened the ground cables (so no loops)
Checked the Resistance (m ohms) 9.4
Inductance (µH) 17.5
Tried tracking in only Z (Y gain at 0)
Tried tracking in only Y (Z gain at 0)
Had both Y and Z gains from 0 - 25 with no change
Tried using track mode 5
Tried multiple configurations some close to singularity and others no where near it.
Tried multiple positions (flat, vertical, and overhead)
Tried moving the current sensor from close to the terminal on the welder to up next to the part.
Electrically isolated the welding torch from the robot using some Black Delrin.
Checked the resistance between the + and - terminals (50 ohms) confirmed with lincoln automation devision this # is correct.
If I watch the WG screen it's doing it's job keeping my current at what I set it to in the weld data.
Tried with preheat and without.
out of ideas.
Any thoughts??