This is a fast index of information from various browsing of websites.
Beautiful bows, and discussion of some design strictures at Ballistik.
This guy Dave Canterbury at the pathfinder school shows how to shoot, teaches how to make a bow, and has pretty good physics (in addition to great technique) and beautiful craftsmanship as shown in these quivers and arrows.. Highly recommended, and there's a whole series of them.
Some terminology follows, not so much for the sake of the words, but the ideas behind them.
The "shelf" is to rest the arrow on. It's cut out so that the string travel is coaxial with the arrow. If you think about it, a bow without a cutout to pass the arrow simply can't have the string move in the same straight line as the arrow. The rightward translation of the arrow's nock during the thrust phase sets up a flexure that affects trajectory.
"Center Shot" refers to carving away the center of the riser so the arrow can rest directly in front of the string's direction of travel. This is funny because of the mis-use of that term in the movie Deliverance.
Brace height is the distance between the shelf of the strung bow and the string. Needs to be ~6", dunno why. Far enough to let the fletching fit, anyway. Not too far though because a longer stroke through the draw means more energy to put into the arrow. On recurves, brace height will be smaller for the same length bow & preload. We'll see why bigger is better later, in discussing springs and stacking.
(I was going to post a link to the definitive comparative bow design site, but haven't found it! Wow, there is a lot of bullshit on the internet about bow design & how to get energy into the arrow. Of course it's an (almost) simple integral of the draw force x length. Only "almost" because the limbs are also being accelerated, so light limbs helps. Considering a compound makes clear the main elegant idea: there has to be force to deflect the string back, but it needn't be linear and indeed that's bad because your max force capability is at the beginning of the draw. (...and there's an accuracy cost to holding a heavy weight at the extremity of the draw.) How then to proportionately increase the load early in the draw? The obvious way is to make the spring physically BIG, so the fractional spring deflection & hence force change through your (fixed) draw length is small. Then biasing the load up is possible: if your draw weight goes from 40lb to 50lb through the stroke, that's a lot more energy than if it goes from 25lb to 50lb.
We can probably imagine bows as linear springs with regards to bending vs force. How does recurve geometry make this linear spring behave differently? One way to think of the tip of the limb of the recurve is like a sort of big wheel (or cam!) off which the string rolls. This has a tendency, through the early part of the draw, to raise the contact point and thus retard the decrease of the angle between the string and the direction of the arrow. In the limit, you could imagine the string geometry at (say) 4" and 5" of draw being parallel. If you then provide the same force over a longer lever arm, hence greater bending moment, you're doing what's necessary to have deflected the limbs more. Viola, zero "stacking" which is the benefit we're looking for!. Also at the beginning of the draw you are not bending the terminal few inches of the limb: it's purely in compression. That offers the possibility of variable spring constant as different parts of the wood are successively loaded. They're all tapered, likely mostly for reasons of bending moment (builds towards center) and speed (heavy limbs absorb energy) but as a possible consideration, softer tips would again reduce stacking if geometry keeps them out of play until late in the draw. This is all just reasoning so far. As noted, I'm finding zip on the internet. I've got to think more about this though, to describe it better: maybe later. (comments solicited).
Fletching needs to pass the bow without strongly disturbing the tail of the arrow. Feathers are designed to collapse when this happens. I like the idea of removing one feather entirely. Per Marv Clyncke, "you can't shoot rubber off the shelf. Has to be elevated." Clyncke by the way is a fantastic guy, local, who sold Kevin & I bows from his phenomenal stockpile of traditional equipment. Here's a short video to give a sense of the guy.
Cross section (my term) should be rectangular, or even I-beam in shape. Idea is to preclude limb flex in any direction but where you want, towards the arrow's nock. You can imagine a recurve bow is mechanically "unstable" in that it'd like to twist around & unload into it's unstrung shape.
Finally, this.
Beautiful bows, and discussion of some design strictures at Ballistik.
This guy Dave Canterbury at the pathfinder school shows how to shoot, teaches how to make a bow, and has pretty good physics (in addition to great technique) and beautiful craftsmanship as shown in these quivers and arrows.. Highly recommended, and there's a whole series of them.
Some terminology follows, not so much for the sake of the words, but the ideas behind them.
The "shelf" is to rest the arrow on. It's cut out so that the string travel is coaxial with the arrow. If you think about it, a bow without a cutout to pass the arrow simply can't have the string move in the same straight line as the arrow. The rightward translation of the arrow's nock during the thrust phase sets up a flexure that affects trajectory.
"Center Shot" refers to carving away the center of the riser so the arrow can rest directly in front of the string's direction of travel. This is funny because of the mis-use of that term in the movie Deliverance.
Brace height is the distance between the shelf of the strung bow and the string. Needs to be ~6", dunno why. Far enough to let the fletching fit, anyway. Not too far though because a longer stroke through the draw means more energy to put into the arrow. On recurves, brace height will be smaller for the same length bow & preload. We'll see why bigger is better later, in discussing springs and stacking.
Stacking is the term to describe increased draw weight through the length of the draw. Minimal stacking is best. A compound bow uses complex geometry to make the force actually go down towards the end of the draw. (I think you could even call that negative stacking, but the terminology is a bit fuzzy, IMO. I think they call negative stacking "let-off" on compounds.) A recurve bow, by flexing to change the virtual attach point, has less stacking than a traditional longbow and will therefore store more energy (for the same max draw weight). With less stacking, the preload can be higher, so the integral through the draw will be, too.
(I was going to post a link to the definitive comparative bow design site, but haven't found it! Wow, there is a lot of bullshit on the internet about bow design & how to get energy into the arrow. Of course it's an (almost) simple integral of the draw force x length. Only "almost" because the limbs are also being accelerated, so light limbs helps. Considering a compound makes clear the main elegant idea: there has to be force to deflect the string back, but it needn't be linear and indeed that's bad because your max force capability is at the beginning of the draw. (...and there's an accuracy cost to holding a heavy weight at the extremity of the draw.) How then to proportionately increase the load early in the draw? The obvious way is to make the spring physically BIG, so the fractional spring deflection & hence force change through your (fixed) draw length is small. Then biasing the load up is possible: if your draw weight goes from 40lb to 50lb through the stroke, that's a lot more energy than if it goes from 25lb to 50lb.
We can probably imagine bows as linear springs with regards to bending vs force. How does recurve geometry make this linear spring behave differently? One way to think of the tip of the limb of the recurve is like a sort of big wheel (or cam!) off which the string rolls. This has a tendency, through the early part of the draw, to raise the contact point and thus retard the decrease of the angle between the string and the direction of the arrow. In the limit, you could imagine the string geometry at (say) 4" and 5" of draw being parallel. If you then provide the same force over a longer lever arm, hence greater bending moment, you're doing what's necessary to have deflected the limbs more. Viola, zero "stacking" which is the benefit we're looking for!. Also at the beginning of the draw you are not bending the terminal few inches of the limb: it's purely in compression. That offers the possibility of variable spring constant as different parts of the wood are successively loaded. They're all tapered, likely mostly for reasons of bending moment (builds towards center) and speed (heavy limbs absorb energy) but as a possible consideration, softer tips would again reduce stacking if geometry keeps them out of play until late in the draw. This is all just reasoning so far. As noted, I'm finding zip on the internet. I've got to think more about this though, to describe it better: maybe later. (comments solicited).
Fletching needs to pass the bow without strongly disturbing the tail of the arrow. Feathers are designed to collapse when this happens. I like the idea of removing one feather entirely. Per Marv Clyncke, "you can't shoot rubber off the shelf. Has to be elevated." Clyncke by the way is a fantastic guy, local, who sold Kevin & I bows from his phenomenal stockpile of traditional equipment. Here's a short video to give a sense of the guy.
Cross section (my term) should be rectangular, or even I-beam in shape. Idea is to preclude limb flex in any direction but where you want, towards the arrow's nock. You can imagine a recurve bow is mechanically "unstable" in that it'd like to twist around & unload into it's unstrung shape.
Finally, this.
