All ancient authors - both Greek and Roman - agree that sinew (animal tendons) were the best material for springs of a torsion engine. Horsehair, or occasionally women's hair, was stated as an acceptable but still inferior substitute.
However, until Stevenson's (1995; 1997) work no academic had managed to create sinew rope for torsion engines. This reflects the gap between craftsmens and academics, as there is really no magic in making cordage from natural materials such as sinew. in fact, the whole process of transforming any kind of fibers into a rope is described in great detail by several authors (e.g. Baker 2000c: 187-258; Edholm & Wilder 1999: 149-151). In addition, there is an abudance of resources and knowledge on the Internet in various primitive archery communities.
Baker (2000c: 231-233) as well as Douglas Spotted Eagle (1978: 51-52) specifically describe making sinew bowstrings, which is one of the closest analogies to a torsion spring's sinew rope one can hope to get. In addition sinew rope was used in Inuit cable-backed bows (e.g. Baker 2000b: 85-87; Callahan 2001: 119-133). In both of these use-cases the sinew cord is subject to at least as large forces as the spring cord in torsion engines. Stevenson's (1995; 1997: 13-17) describes making sinew cordage for the Cheiroballistra, albeit fairly briefly.
Extracting sinew fibers from tendons Edit
This content has moved here.
Making sinew thread Edit
This content has moved here.
Making sinew rope Edit
This content will be here.
Notes on twisting and stretch Edit
Short fibers require more twisting to make a durable thread, which also means that outer fibers are stressed more as they have to travel a longer path than inner fibers (e.g. Baker 2000c: 191-192). However, according to Karpowicz (2008: 59) sinew can stretch 5% without any noticeable loss in it's energy storage capabilities and that 10% stretch will result in a small loss of about 5%. Baker (2000d: 108) has reached similar figures. This means that if one is using shorter sinew fibers, they can be twisted together more tightly to create only a slightly weaker thread than one made from longer fibers.
One important insight can be obtained from the above. Consider the following scenario which applies to sinew thread, bowstrings and the finished torsion springs:
- Outer fibers/cords are stretched 2% by twisting
- Inner fibers/cords are stretched less than this (0-2%), depending on their distance from the center of the torsion spring bundle, bowstring or thread.
- Maximum safe stretch is assumed to be 8%
- In sinew springs energy is stored by twisting the bundle of cords. This means that the stress of the outer cords rises more sharply than that of the inner cords. When outer cords have stretched by 8%, the innermost fibers may have stretched only, say, 4%.
- When sinew thread or a sinew bowstring is pulled, all fibers stretch roughly equally. The outer fibers still reach the maximum (8%) first, but inner fiber are close behind (~6%).
This means the sinew rope should be prestretched as much as possible when being wound around the washer bars. The washers themselves should be rotated as little as possible to allow outer fibers to store as much energy (by reaching their breaking point as late as possible). According to Baker (2000d: 109) energy storage curve of sinew thread is linear, meaning that 1% of stretch will store 10 units and 2% stretch 20 units of energy. Consider what this means in the context of torsion springs:
- Torsion spring A: outer cords stretched by 1% when wound wound around washers. Stretched an additional 4% by rotating the washers. Maximum stretch 8%.
- Torsion spring B: outer cords stretched by 1% when wound wound around washers. Stretched an additional 1% by rotating the washers. Maximum stretch 8%.
Outer cords in torsion spring A can only be stretched by additional 3% when arms are pulled back, whereas those in torsion spring B can take 6% stretch. This means B can store twice the energy of A, provided that arms can be rotated enough to reach 8% stretch. From this we can also deduce that the cordage of an inswinger with more arm rotation should not be prestretched as much as that of an outswinger.