That is all well and good BUT the reflexed limb would take less set
Assuming that they would take the same set is wrong. The recurved limb has a shorter woking section opposed to a proper pyramid limb of the same intial length. The pyramid can also have significantly less weight at the tip/last 6 inches or so.
These were just short flipped tips, not big hooks. The pyramid flatbow I started with has stiff tips. The flipped tips were shorter than the stiff section, so the working limb length was unchanged (unless there is something there I am missing). I agree the tips can be narrowed significantly on a flat pyramid design, but optimizing that wasn't what I was interested in. I wanted to know the effects of reflexing the tips forward and if one method was better than the other overall to achieve that.
The goal of the experiment was to take a flat bow design and modify it exactly as you could in real life with an existing bow in your hand, by either flipping the tips or reflexing the whole limb without any other changes. I changed nothing on the limb thickness or back profile, so the flipped tips were the same weight/width/thickness as before flipping them. As expected with this approach, draw weight increased along with limb stresses. In reality you would likely want to retiller some, to either get back to the original draw weight or fine tune the bend because of the changes made from the original bow.
Are you including string length in your figuring? If you have two bows like you say, at brace height the reflexed bows string will probably be full length ie tip to tip but the recurve string won't have lifted off yet. This will skew your DW at various DL's.
Brace height was kept constant. I would have to look at the string lengths to see how much they differed. The flipped tips didn't touch the string at brace so there was no lift off point to worry about. I just wanted to see what effect the two options had on the outcome, if the string rode the recurves a bit that would have been one of the effects to watch.
In the end the recurved design stored the same energy as the reflexed design at a slightly lower draw weight, had lower overall stresses and a smoother draw with less stack. The differences aren't huge (because I didn't move the tips a huge amount forward), but they are definitely there.
Mark