This is where a sketch would be helpful.
The individual posting the question included reference to T-, Y-, and K-connection, which I can only assume, is the proper callout. A sketch of the joint(s) in question could eliminate any misinterpretation of the actual conditions and joints being welded.
The commentary of D1.1 makes it clear that there is a difference between welding a pipe or a box tube (pipe and box tubes are considered to be hollow structural shapes or HSS) and plate. The difference between welding plate and HSS is that welding a HSS involves welding “around” the pipe and the small radius of the box tube. It would appear there is no distinction between welding the HSS to a plate or to another HSS or equal size or different sizes. The situation does become even more complicated if the intersection between the plate (end plate?) and HSS is at a skew. Then the dihedral angle comes into play. Also note that the box tube requires the welder qualification to include a macroetch at the radius to ensure the welder obtained complete fusion. My initial opinion was changed somewhat after reviewing the commentary.
AWS D1.1-2008 clause 4.26(3) references Figure 4.24(B) for qualifying welders to weld butt joints, T-, Y-, or K-joints for production welding involving such joints in HSS where backing is utilized. In this case the initial root bead and second intermediate layer of weld would be considered backing for the purpose of welder qualification.
Again, I may not see the whole picture or understand the question. I don't see where qualifying the welders in the 3G and 4G positions on plate meet the intent or the requirements of D1.1 in this situation.
I believe it is understood that a welding standard cannot address every conceivable joint configuration imaginable. At some point a meeting of the minds between the owner, engineer, fabricator, and inspector must be considered to address the weird or unusual. AWS D1.1 has provisions allowing the Engineer (the owner’s representative) some latitude when it is unclear what the code requires or if in the Engineer’s judgment an alternative to the code requirement is appropriate.
In any event, when there is a question with regards to the intent of a code requirement, short of waiting for many months while an official code interpretation from the code committee is obtained, the Engineer should be the arbitrator and provide direction to the fabricator and the inspector. It is not the inspector’s prerogative to grant a waiver to or to modify the requirements of the code.
Best regards - Al