Hi All,
I haven't been around sub arc welding since the late 70's. I do have an old SAW troubleshooting chart for stainless weldments which indicates that possible causes for porosity and gas pockets are the following:
Shallow composition burden, improper cleaning, contaminated manual weld backing, insufficient penetration in double welds, improper fitup in welds with manual backing, and moisture on the plate or in the flux. It also indicates that once the arc is started, the welding current is the most influential variable, followed by welding voltage, and then travel speed. Also, I don't know if this is in any way related to your problems or not, but it goes on to state that the sub arc process is not recommended where a weld deposit is needed that is fully austenitic or is controlled to a low ferrite content below 4%.
The pock marking problem is addressed in the troubleshooting guide for carbon and low alloy steels, but not stainless. I don't know if the same corrective action can be used on stainless, but for what it's worth, it lists in order of importance to increase voltage, decrease current, decrease speed. It also goes on to say that heavier plate than normal will cause pocking.
Also, for what it's worth, here are a few causes of porosity in the sub arc welding of carbon and low alloy steel. Again, I don't know whether or not some of this could cause the same problems with stainless:
Contaminants in the joint: remove oil, dirt, moisture, etc., from the exterior surfaces and abutting edges, as Chet mentioned. Insufficient flux: if an inadequate amount of flux is used, the arc flashes through the flux and causes scattered surface porosity. The weld may also contain porosity if the slag spills off the weld before it has solidified. Contaminants in the flux: may be picked up by the flux recovery system and deposited in subsequent welds. Entrapped flux: the possibility of porosity exists wherever there is an opportunity for flux to become trapped between the bottom of the bead and the opposite side of the joint. Porosity from entrapped flux is most commonly encountered in butt welds. If the gap between plate edges is 1/32" or more, flux may spill into the gap ahead of the arc. The remedy is to back the joint with a manual or semiautomatic weld or a backing strip. The bead must penetrate the backing, or clear the backing by at least 5/32". Segregation: The composition of the base metal has a bearing on the porosity occurring in the weld. Even when composition is within acceptable limits, possible local segregation of constituents may promote porosity. Sulfur in particular tends to segregate within steel alloys. When it is necessary to weld steels with sulfur segregation, use procedures that give minimum admixture (low current, negative polarity, and large electrode diameters). Inappropriate polarity: the effect is slight and not normally an important factor, but positive polarity generally produces less porosity than negative polarity. However, negative polarity results in less porosity with sulfur bearing steels or with other alloys where penetration and admixture are to be minimized. Excessive travel speed: as GRoberts mentioned, reducing travel speed reduces porosity, because slow speeds allow gaseous materials to boil out of the molten weld metal. Slag residue: from some types of electrodes may cause porosity where tack welds are covered with sub arc welds. Lastly, the depth of the flux layer must also be considered. If it is too thin, there will be too much arcing through the flux or arc flash. This also may cause porosity.