Tig welding torch movement

I was taught to move at a constant speed and add the wire by dipping at a consistent frequency. Pausing during the wire addition makes sense. Backing up onto the bead to deposit filler doesn't make much sense to me.

In automated cold wire feed GTAW it is common to run a constant travel speed and a constant wire feed speed. In more exotic automated welding, the wire feed may be "sync-pulsed" in conjunction with the pulsed current, thereby feeding at a higher rate during the high current pulse, and cutting back to a lower feed during the background current part of the cycle. A little more exotic than that, the wire and current pulses can be sycronized with the torch side-to-side oscillation, so normally the high current and high wire occur during a pause at the weld's outer edges, and the low current and feed occur as the torch traverses across the weld center. In addition to this, the automatic voltage control (AVC) which controls electrode to work distance based on arc voltage, can be incorporated into the scheme to change arc length in sync with the torch weave, current, and wire. There are some applications where an automated welder may be programmed to pause the forward travel in sync with the high pulse of current and wire.

I had some lessons from a guy who was a highly qualfied instructor, able to conduct tests for welder coding.

He showed me to back up a little each time you add filler............this I would imagine if the text book 100% correct way to do it.

I guess I should have added that I generally use gas lenses and extend the tungsten out close to cup diameter. I understand that if you tungsten is flush to a few diameters beyond the cup it would be difficult to add filler without moving the torch back. Is this why that is recommended?

I have been teaching TIG for 10 years and I do NOT advocate moving the torch back. A constant forward travel, no weave. The filler is tapped into the leading edge of the puddle as you go in a steady rhythm.

Tap, tap, tap.

Float the torch at a 70 degree angle.

Thanks - that is exactly how I do it. Just wanted to confirm that I wasn't the only one.

I'm missing something here.

If your torch is moving in a constant forward motion, and all filler added is to the front of the puddle, why isn't the resulting weld smooth instead of having the "stack of dimes" effect?

Don W.

An autogeneous weld (without added filler) will be very smooth if the torch is moved at a constant rate with no extraneous side to side or up and down motion. As wire is dipped into the puddle, or if the torch motion is changed, the effect is seen in the solidification of the weld puddle, you get ripples. Attached are photos of an autogeneous weld where all variables were held constant and the current was pulsed to create overlapping spot welds.