"If I have a nice straight 40# bow. 6" brace height and 26" draw length so it's about 2# an inch. Now I deflex the bow 4", string it back to 6" brace, will I have a 32# bow at 26" draw? Would it be 40#@30" draw?"
So, the final draw weight numbers may or may not be correct depending on the bow design and length, but for illustration, we can say they are close. However, draw weight has as much to do with leverage, as dictated by geometry, as it does with simple stiffness of the limb. But, let's look at what ELSE really happens.
First, by deflexing the bow, you are straining (by bending) the limbs less to get them to brace height. This means that the string is less TIGHT at brace height. This means that your draw WEIGHT for the first INCH of draw on the deflexed bow will be less than on the straight bow. The straight bow might pull 12 lbs at one inch, while the deflexed bow pulls 8 lbs at one inch. So, that also means that at two inches the straight bow may pull 14 lbs, but the deflexed bow might pull 10. This trend, of storing less energy at every increment at draw continues through the whole draw cycle. This is why we plot force/draw curves.
Now, remember that the MOVEMENT OF THE STRING added to the draw weight stores the energy. If you pulled on a giant iron bow with a hundred pounds of force, but the string didn't move and the limbs don't bend, you haven't stored any energy, and the arrow can't go anywhere.
Anyway, the deflexed bow will probably end up at a lower full draw weight, BUT, it is possible that because of limb geometry, both bows may end up at 40 lbs at 26", or darn close. HOWEVER, the deflexed bow will store less total energy, AND since it started lower, toward the end of the draw, it will stack, rising abruptly to the 40 Lb weight. This is why we work so hard to avoid set. However, the deflexed bow will be under much less strain at full draw, and therefore be less likely to take further set, fret, or break.
A R/D bow has VERY HIGH string tension for their draw weight. They store even more energy at every inch of draw than the straight bow. If the first inch is 8lbs for the deflexed bow and 12lbs for the straight bow, it might be 16 lbs for the R/D bow, even though they all end up at 40 lbs. In addition, toward the end of the draw, the R/D maintains a favorable, higher leverage string angle later in the draw cycle. So, not only are you storing more from having a tight string in the early draw, you also have better leverage and are storing even MORE by moving a stiffer limb more easily LATER in the draw. Recurve bows also do this, as do longer bows (compared to shorter) bows, and lever bows like the mollegabet styles, to some degree. But, energy storage is not a free lunch, and is not the only important factor in bow design.
A R/D has a couple more advantages. They are usually made as laminates, for one thing, and thus take advantage of the Perry reflexing principles. And because the inner limb (close to the handle) is already deflexed, it can be less strained (like the deflexed bow), massive (to protect it from failure), bend very little (while still translating to a lot of movement out at the tip), and still store a lot of energy total. So, good.
But, it's not a free lunch of course. High physical limb weight can cut into efficiency, and the curve of the limb makes it more prone to vibration than a straight- limbed bow. Incidentally, both of these are also inherently magnified in a recurve: the recurved outer limb often requires more mass to maintain stability, so there is often either a weight issue, or a vibration issue, or both. Also, they are tricky to tiller, sometimes, easy to hinge, and just generally take more trouble, time, and technique to make.
"If so, if I could magically increase the weight back to 40#@26, I could do so by adding thickness or width. If I used thickness would I be limiting the draw length? By width shouldn't change it other than draw weight. I'll stop there to see if anyone bites. I won't feel bad if no one does."
As I said, you may not need to "magically" increase the draw weight: it may simply be about the same, due to the geometry and leverage. More than likely, it would be a bit less, but since the deflex is present, the limb will be less strained, too. You could probably safely do either, given any given wood. You could offset the strain you lowered by deflexing by increasing the strain by thickness. The increased thickness would be more efficient, because it would take less weight to increase the stiffness. Hiowever, R/D bows and recurves basically MUST (in my experience) have flat limbs, either lenticular or rectangular in cross section. If you tried to R/D a bow with a round or square cross section, there is nothing to stop the bow from bending sideways as much as back to front, and will want to twist, turn, shed strings, etc.
Essentially the same thing is done on any R/D bow or deflexed recurve bow. If I want to make a bow with a big, high angle recurve, I generally deflex it and widen the limb close to the handle. The bigger the recurve (angle, diameter, and % of limb it occupies) the more deflex, at least in close to the handle.
A R/D bow is a LITTLE different, but principally the same. Appropriate amounts and ratios of R to D will depend on the length of the bow and the wood, the backing, and the limb width, just like the design of a straight self bow depends on these parameters. Basically, I go for a finished bow with tips ahead of the handle, but not WAY ahead. My forms deflex about 1-1/2" -2" from the front of the handle, and put the tips at LEAST that far ahead of the handle before tillering. So, back 2", and foreward 4". However, as you release a bow from the form, the amount of reflex and deflex will both settle into less than the form, so maybe 1.25" and tips only 1" ahead. So plan for that.
For maximum benefit from Perry reflexing, I generally glue up the core and backing, deflex the core with heat, or splice the core at an angle like Marc showed on that front page article a while ago. If I start with a 1.5" deflex and a 1.5-2" reflex (tips ahead of handle) I will end with tips about evenwith handle, or even, like 3/8" behind. Even with the tips a bit behind, you get the benefits of smooth draw and high string weight, but less and less so as they move in line with the deflexed portion.
