The Scythian-type bows from Yanghai (Shanshan)

jack farrell · Post by **jack farrell** » Sun Nov 22, 2009 11:12 pm

Length across is 44". The horn backbone will stand as in the original. Here are the three pieces glued together.

Dusan · Post by **Dusan** » Sun Nov 22, 2009 11:40 pm

Congratulation Jack ! You are in hero way

.I am very couriouse , how its ended.All best to your work.

Bede · Post by **Bede** » Mon Nov 23, 2009 1:22 am

Jack,
I think ancient bowyers probably had the same difficulty with horn length as you and probably solved it the same way. Was that a picture of the horn alone or the horn and its side strips?

redhawk · Post by **redhawk** » Mon Nov 23, 2009 9:05 am

Jack, same question as Bede. Is it the horn alone?Sidewalls?
Laminated wooden core?
The Olon kurin gol bow was made of at least 14 pieces. By far enough to run you crazy.
The core of the Subeixi is a lamination too, made of a center slat and sidewalls. For not talk of the multi- laminated grip.
I still don?t believe that the Subeixi is a shootable bow, so I?m very excited about your trial and if you?ll match the measurements(thickness!!) of the original.

Good luck, Michael

redhawk · Post by **redhawk** » Mon Nov 23, 2009 10:15 am

Jack, if the pic you?ve posted is the horn alone, glue it as soon as it possible to the core or fix it at your jig, otherwise it will return into its original shape.
Michael

jack farrell · Post by **jack farrell** » Mon Nov 23, 2009 1:17 pm

This is only the very center horn piece of the bow. The pieces are not cut from the outer curve or the inner curve but are from the sidewall plates of horn, as Adam has suggested. I am at the beginning only.

Michael I think you are understanding one of the dilemmas of the chore. When wood is bent around a form and later released, there is spring back, a return to a less bent shape. So I bent this horn core by hand till it would stay in this shape independent of the form. (I hope)

In my experience to add heat later is to induce spring back Also to add 15 layers of glue is to induce spring back.

If bent wood is added to this horn they both together should want to spring back.

I do not have construction details of the Subexi bow but am using these from the Yanghai bows. Can anyone provide details of the Subexi bow(s)? Post here. -j-

Andrew Hall · Post by **Andrew Hall** » Mon Nov 23, 2009 8:01 pm

I have some Chinese archaeological reports of the Subeshi cemeteries. The only relevant information on bow construction is a tiny cross sectional drawing showing what is probably similar construction to that of the Yanghai finds, and some text describing the horn between sandwiched between two pieces of wood. I assume they are of the same basic construction.

Best of luck with this challenging reconstruction.

Andrew

redhawk · Post by **redhawk** » Tue Nov 24, 2009 10:49 am

Jack, if you glue the core and the horn together, they will never or only a small amount spring back, that?s one of the reasons for to make laminations.
As Stephen and me are always mentioning, hornbows are a glueing process.

Michael

jack farrell · Post by **jack farrell** » Tue Nov 24, 2009 12:00 pm

Michael; Have you or anyone else successfully bent wood into this small diameter curve? Please describe.

Why would two layers done this way not spring back?

redhawk · Post by **redhawk** » Tue Nov 24, 2009 12:27 pm

Jack, it is not possible to bend wood with such a diameter into such sharp Scythian curves.
That?s why they laminated the bow of so many parts.
Look at my olon kurin gol plans.
They bent small and thin slats into the desired shape and then they glued them together, that?s the only sense- making explanation for making such a multi- laminated construction.

Laminations can?t spring back, cause they are glued.
Take 2 paperstrips, glue them straight and glue them in a curve. If you glue them in a curve you are reducing the glue- line, they can?t spring back.
The more strips you glue this way, the more stable the lam will be. That?s all.

In bowery sidewalls are used to stabilize siyahs, recurves, grips, same technique: adding glue- lines.

Michael

jack farrell · Post by **jack farrell** » Tue Nov 24, 2009 12:37 pm

Michael; I will take your comments under advisement. I know Germans have a proud history of bending wood. Can anyone else comment about bending wood into a tight curve?

Daniel T. · Post by **Daniel T.** » Tue Nov 24, 2009 2:49 pm

Aside from the well-known important things (wet wood, steamed or boiled, no grain violations), there are a few engineering considerations that I think are important.
Michael is right about the laminations. When they are glued together and want to spring back, they must stretch glue. The glue resist this shear stress more or less and helps to maintain the shape. Another advantage is the reduced thickness. The strain on the surface is approximately proportional to the thickness and inversely proportional to the bending radius. So if you want to bend over a smaller radius you must use thinner wood to keep the strain tolerable. If you exceed a certain strain, the wood will break.
This is the simplified elastic theory. With wood, it is not only a matter of geometry. Wood is viscoelastic, which means that it's mechanical properties depend on a number of things (time, temperature, humidity...). The time-dependence causes e.g. the reduction of draw weight as full draw is maintained or the gradual recovery of string follow after unbracing. For bending wood, this time-dependence can be exploited when the bending is performed slowly. When the bending is applied in a slow manner, the wood fibres have more time to rearrange themselves and accomodate the deformation; the stress will be lower than in fast bending. The breaking strain is also larger than in fast bending. A slow bending can therefore increases the chances of success. This benefit is counteracted by the decreasing temperature, so one cannot bend it over 5 minutes, but in my experience half a minute doesn't hurt.

I have not tried it, but perhaps a soft plastic - 4mm polypropylene or something like that - on the outer side lath during bending can help to prevent fibres from lifting.

ginni · Post by **ginni** » Tue Nov 24, 2009 10:10 pm

Excellent work Jack!
I am very curious to know how this bow work and if your theory about griping position is right.

Adam Karpowicz · Post by **Adam Karpowicz** » Wed Nov 25, 2009 12:13 am

Very nice start Jack, please report on the progress.

In my effort to reproduce this bow, I try to find answers to two questions: how such bow was made and why. As to how, I believe the (vertical) horn strip and the side wood were bent together and I assure you (Michael) it is possible and I am sure it is the only logical way to proceed. I will post the results once I finish the bow, but this will take some time.

Adam

Bede · Post by **Bede** » Wed Nov 25, 2009 12:57 am

I haven't made a bow this way, but I found fresh cut wood (green) could be bent a lot further on a smaller radius. However, Jack can tell you that the minimum bending radius of the piece and its thickness are very closely related.

There are several people attempting to reproduce these bows. Everyone should be encouraged to experiment. It might be useful to remember that the originals were made without G clamps and probably without rigid forms. I don't suggest that bowyers should start out that way, but if you come up with a high tech solution, please try and work out how a nomad could have achieved the same result. A working bow should be the first objective. After that, there are many opportunities to discover the ancient ways that bowyers arrived at the same result.