Hello All,
Thanks Arvid, it's nice to be thought of as an expert, I'll take that as a compliment but as mentioned I'm sure there are a few other experts here as well, obviously Bob is one but there'd be others that know more than they think they do, often the case with welders.

Sorry it's taken me so long to find this but I've been distracted and had some password issues as well.

I've never heard the term "Block and Chip" before but the joint shown is a pretty common one, the idea is that to minimise distortion all the welds are at 90 degrees to each other rather than at 45 degrees as in a mitre, with the welds at 90 degree the shrinkage effectively shortens the "side length of a rectangular frame where shrinkage at 45 degrees pulls the angle closer together so the resulting frame isn't square.

A further complication being that as the centroid (axis of strength) of angle is not coincident (contained within) the steel but rather floating somewhere in the middle of the angle's "V" the shrinkage isn't directly supported at all by the compressive strength of the metal, this results in a twist being induced, all of this shrinkage can be reduced using certain techniques, but I think first of all to control shrinkage it is a good idea to understand how it works, this will let you decide strategies to reduce warpage on all sorts of connections not just angle or squares etc. I will write this into a thread for those that are interested and try to answer the question at hand here.

When it comes to square tube 99% of designs use mitred joints, and these are very often done without warpage by skilled operators, With Rectangular tube 2 x 6 x 3/8" you need to consider which axis the mitre would be welded across as the 45 degree weld could be shrinking over a length of 2.8" compared to 8.4" so obviously making a huge difference to the overall shrinkage. I will assume the worst case that we are welding across the 6" side so the 45degree weld is 8.4" long.

The following method is a bit OTT but is the best possible method to ensure an even frame. If you did this in a commercial shop on any thing but the most critical work, like a turbine plinth or something like that people would think you were dragging the job out. But it is the right way.

1, cut all the pieces, including mitres, leave a 1/16" gap at each joint.
2, grind a 45degree weld prep on each edge to be welded.
3, lay out the frame over an accurate chalk drawing on a rigid flat steel bench and clamp securely. remember the 1/16" gaps, make sure they are even.
4, tack each of the inside top corners of the frame well, for a 3/8" thick wall your tacks should be 3/8" long. tack in order of opposite corners, so furthest away, lefthand then closest righthand side, then furthest away right hand followed by closest lefthand side. The idea is to keep the heat even across the whole frame. I would mark each corner with a number as I will use the same order for each step, this not only reduces warpage but reduces the chance that you will miss a weld.

5, tack the out side top corners in the same order as step 4.
6, flip the frame check it against the chalk drawing, and repeat 4 and 5.
7, stand the frame on it's edge and fully weld each of the inside fillet welds, in the same order as step 4 again.
8, frame still on edge weld each of the out side corner fillets, step 4 order.
9, lay the frame down on the bench, check it against drawing adjust if necessary, after the next step you won't be able to.
10, weld each diagonal joint from the inside corner out to the midpoint, in the same order as step 4, do not take a rest between welds, you need these to all still be quite hot when the last one is finished.
11, flip the frame and repeat step 10.
12, flip the frame again and repeat step 10 except welding in from the outside corner.
13, finally flip the frame one last and time finish off the last four welds and still in order and as always without resting in between the four welds.

As I said above this is the neurotic method, but it assures that all shrinkage will be counteracted by identical shrinkage on the opposing corner.

the more commonly used method is to simply use experience or a test piece to estimate the effect of shrinkage if you simply weld around each corner continuously and set your pieces up taking the shrinkage into account, so if your test piece measures an 87degree angle set you real pieces up at 93 degrees and use the same method to weld as you did on the test piece.

I hope this has answered your question and frankly, couldn't imagine how I could add any more detail.

I am going to explain the way shrinkage works on another thread if that helps, I will give some tips as to where one can get caught up as well.

Best Regards
Rick