Lawrence,
If my memory serves me correctly, we had to take some wire out (WFS) during the down hill side and some back-ground amperage. This was during the 80's and we were using a Gold Track III or IV with a F head. You had a pendant box that you would use to make adustments. The main problem we had, is we had to constantly be looking at the silica floating on the puddle to make sure we would not wash over the top of it. (ER70S-3) So we put a piece of tungsten in a wooden file handle and when the silica got to be about 3/16" of an inch we would reach up and touch the tungsten to it and pick it out of the puddle. It seemed to be worse on the down hill side. The technology has improved so much since then, I'm sure there are many more bells and whistles available now.
Jim
Lawrence,
Like Jim said, we usually decreased wire speed and current on the downhill side when doing 360 welds. Lots of variables though, when we were welding XXwall pipes with high currents (carrying lots of metal) the downhill side, due to gravity, was too risky so I decrease wire speed and current to achieve a safer puddle. On the downhill side it is more critical to manipulate the wire entry to find the sweet spot in the puddle. It's less forgiving on the DH side compared to the UH side. Welding uphill you have the classic tear drop to feed the wire but DH doesn't provide the same "target". It was also normal to have to strip or run an extra layer or two on the DH side to flush it out evenly with the UH side before capping.
I can't remember the magic pipe diameter size (36" I think. It takes coordination to run two weld heads simultaneously on one joint) where we began using two weld heads both vertical uphill instead of one weld head doing 360's. The slower option was one weld head welding vert. UH with two wire spools and wire feed motors to where you had to rewind the head from 12 to 6 o'clock, switch the direction of travel, switch the wire feed motor and put about 3-5 degrees of lead angle in the tungsten before firing up. Of course you can use that same set-up to weld 360, down slope, switch everything to weld the opposite direction then fire back up if WPS and job specs allow 360 UH/DH welding. That is the most productive method if pipe size is too small for two machines. Arc time is boss, not rewind time.
Every job had its own requirements that dictated the weld progression, weld head set up, etc. Usually it was production but not always. PQR's, WPS's and WQTR's were always in order as most or our work was nuclear and high pressure steam pipe on fossil fuel plants. DH Machine GTAW was very safe but with inexperienced operators it can be dangerous...think avalanche :)