Hollow Back Bow no.4

[Jan de Bogenman]

Are you maintaining the same thickness of the spars throughout the U profile? In a feather it is both reducing in width and in thickness over the length of the limb to the tip.

Thanks for pushing me to the next level Philipp! So far I mainatained aproxx the same width over the full length of the ridges. Except for the outer limb where everything get's narrow. This was mainly out of laziness. (Somehow I am always in a hurry to see if things will work..)
I was planning for the next to do what you suggest. To taper the ridges down from the handle to the tips. Thanks for reminding me!

And another thing concerning the feather principle. Think I remember a patent for a bow with limbs filled with foam!

[Springbuck]

This is amazing work, of course....  I have so many thoughts and no focus or time to try things out anymore, but the closest I have come to this is a pyramid bow, backed with two slats of bamboo from a small diameter stalk, side by side.

The first thing I wonder is, while you are definitely saving weight throughout the bow, does this save weight where it matters most?  Which I still assume is in the outer limbs, toward the tips?  Proportionally, I mean.

Jan de Bogenman:  regarding your bamboo test piece...., I tried a couple times to make bamboo bows bend like that, but my efforts were half-hearted and did not do my ideas justice.    However, I think I now have a way of making a bow of split bamboo with the concave side as the back.  I may or may not get around to trying it.

The first possibility is a cable backed bow.  Weight management would be an issue depending on the material and amount of cabling.  I would split the bamboo, maybe splice at the handle, and taper the thickness for good tiller.  By tightly wrapping some spiraling cables (opposite directions) the full length to squeeze it in from the sides.  Then a main "backing" cable could ride atop those wraps, secured here and there.  Basically, a hollow limb, but with a cable.  Biggest problem here is that most bamboo doesn't taper quickly enough, so the whole bow could be super light, but you are saving less weight toward the tips, and again because you have to secure the cable.

[Philipp A]

Cool, I can't wait to see the results of your next progression of this design! In regards to foam filled spars, this has been done in wind turbine rotor blades and other composite structures before as well. It adds stiffness without adding weight. At the end with most of what we do we try to imitate what nature has already designed over millennia before us.

I enjoy the dialogue and like you I love exploring new designs and variations on existing ones. It makes this an infinitely enjoyable hobby and I have yet tons to learn from some of the seasoned experts that regularly post on this forum.

[Jan de Bogenman]

but the closest I have come to this is a pyramid bow, backed with two slats of bamboo from a small diameter stalk, side by side.

Thanks Springbuck, I was thinking of such a bow when working hard with chisels, gouges and scrapers on this bow! Would be great to see it, maybe you've got a picture of it?
And that idea partly ties in with your other question. Mass reduction. If one is to glue up a bow, this design can save lots of material.

Concerning the other interpretation of your question about mass: I think saving mass in the outer limbs remains the most important and most simple way to gain speed. I started this experiment to see if a hollow back could add a little more to that. Or to see if other effects, like flattening out or curling up, will influence performance. For now I think it does some. But 'some' could be very important for, for instance, a flight bow.

About bamboo bows. Sounds interesting, but maybe you can do without the cables if the bamboo is thick enough?

[Philipp A]

Hi Jan, 

If you want to take the time to replicate the turkey feather, when examining the back of the feather closely you will see that the sides of the spars are undercut (hollowed out when viewed from the centre of the groove) and the belly is rounded towards the thick end of the feather and then gradually flattens towards the tip.

[Philipp A]

Hi Jan,

Interesting observation and this would corroborate the Poisson effect. I have just tried what you did with the bamboo slats with a turkey wing feather (the one I took the picture of). You could very much notice that the undercut side spars on the back were starting to close the gap when putting the feather under load and the gaps were opening back up when unloading it. Maybe undercutting them might be required in order to resemble the full elasticity of a turkey feather?! I will try to take a close up video with my cell phone to show the effect tomorrow (I observed it through a magnifying glass, so not sure whether I can replicate it with my cell phone as a close up).

[Springbuck]

About bamboo bows. Sounds interesting, but maybe you can do without the cables if the bamboo is thick enough?
[/quote]

I'm really bad about pics, because I have ADHD and lose them, break my phone before I store them elsewhere, etc..also, the only place I use a computer is at work and what pics I do have are at home....so...... let me see.

Doing without cables.....I know one can flatten the inside of bamboo following the fibers and use the inside surface of the bamboo as the back, as in the Bhutanese bows, but those are made from a very specific specie, very heavy with thick, and they are more like a flatbow in cross section.

It seems that to make either a concave OR hollow bow, you would want to start with bamboo about 3-4 cm diameter, maybe 5 cm max, so you can split it in half  and have that "C" shaped cross section.   The engineering problem I see is that most 3 cm dia bamboo (at the handle) is going to be something like 3.4 cm on the fat end and 2.7 cm on the thin end.  Not enough taper, so unless you want all the bend in the middle of the bow (and fat-looking tips) you have to taper the bamboo.  Right?

So how do you do that?  You have to taper both front/back and side to side at the same time (because it is round).  Because of this, the "C will be a half circle at the handle, but only a quarter circle at the tips, or something like that.  A shallower "C", anyway.   Maybe the limb will be 3 cm across and 1.5 cm "thick" at the handle, but if the bamboo starts out 3 cm dia. at the tip, the tip could be only .4 cm thick and 1.7 cm wide, roughly.   To me, that is too wide and not thick enough.  I want it more like 1 cm wide and 3/4 cm thick at the tips.

Either way, you HAVE to cut across the exposed fibers at the sides of the bamboo to tiller it.   A large diameter bamboo then becomes NOT a hollow or concave bow, and a small diameter bamboo loses stiffness but maintains weight out toward the ends.

[willie]

Philipp

Emulating nature seems like a good idea. I looked thru all my turkey feathers( two wings worth from a florida bird). I see a groove on the underside that is a bit more pronounced than feathers from other species, but I do not see any thing filling the groove. Are you dissecting the feather or looking at internal structure? Perhaps you could sketch the crossection? The "undercuts" you mention are not apparent or understood.
thanks

[Philipp A]

Hi Willie, the groove is not filled but the spar is. Just take a knife and cut through the spar lengthwise and you will find the gummy substance that fills it.

In regards to the undercut I mean, that the groove is hollowed out so the top of the spar has an overhang like a snow drift. Take a magnifying glass and you should see it well. I would not imagine that there is a difference between Florida and Canadian turkeys but who knows.....

[leonwood]

Wow Jan that looks very promising! The weight reduction really shows. I wonder what would happen if you scale it up to 50@28 with a proper heat treat and without rushing the tiller;-)

[simson]

Jan, thanks for doing this testing.
First, I haven't read all the comments on this post.

Here some thoughts:
Your design is a mass saving design, which is ok with tension strong wood (like BL or others). But you can do a lot easier going with a trapped profile or a high crown sapling. The only difference I see is your design is more problematic for torque.
You have made a stiff handled bow. Not good with the excessive grain run out at the fade area. It is probably better to go with a bendy handle bow with no narrowed grip.
Your sketch with the tension and compresssion forces doesn't show the real forces. The farther from the neutral plane the higher is the stress. In other words: the forces are not equal in the compression area (black) or the tension (white) area. Will say the more you can save the (unnecessary) mass near the neutral plane – the greater is the total effect of mass saving. Hope you can follow my bad English. Think on an extreme profile of your design, which will put the neutral plane in the real middle of an U.

At least: To bad your not working on real HLD bows, sorry to say that – but I believe your skills would be better invested in turning the bows back into the belly. But thank you anyway!

[Springbuck]

"Hope you can follow my bad English."

Brother Simson, there is NOTHING wrong with your English.

[willie]

Just as a column can buckle when loaded too heavy, a beam can twist under load if not cross braced adequately.
I think many of us have experienced this when making a tip too narrow. Maybe tip lightening can be accomplished better with a well designed U section? Are the physics behind these unusual crossections related to lateral torsional buckling?

Code: [Select]

https://www.infograph.eu/en/steel-construction

[joachimM]

Just as a column can buckle when loaded too heavy, a beam can twist under load if not cross braced adequately.
I think many of us have experienced this when making a tip too narrow. Maybe tip lightening can be accomplished better with a well designed U section? Are the physics behind these unusual crossections related to lateral torsional buckling?

I fully agree. The hollow tip design allows you to make a broader tip with a lighter mass. The broader tip will be less prone to torsion or sideway bending.
You can even make the tip entirely in the shape of an I-beam, but rotated 90° relative to how it is normally positioned.

[Jan de Bogenman]

Jan, thanks for doing this testing.
First, I haven't read all the comments on this post.

Here some thoughts:
Your design is a mass saving design, which is ok with tension strong wood (like BL or others). But you can do a lot easier going with a trapped profile or a high crown sapling. The only difference I see is your design is more problematic for torque.
You have made a stiff handled bow. Not good with the excessive grain run out at the fade area. It is probably better to go with a bendy handle bow with no narrowed grip.
Your sketch with the tension and compresssion forces doesn't show the real forces. The farther from the neutral plane the higher is the stress. In other words: the forces are not equal in the compression area (black) or the tension (white) area. Will say the more you can save the (unnecessary) mass near the neutral plane – the greater is the total effect of mass saving. Hope you can follow my bad English. Think on an extreme profile of your design, which will put the neutral plane in the real middle of an U.

At least: To bad your not working on real HLD bows, sorry to say that – but I believe your skills would be better invested in turning the bows back into the belly. But thank you anyway!

Simon, thank you for your thoughts. I must admit that I doubted for some time whether I should respond. That you do not agree, okay. But the tone of your posts in particular surprised me. But maybe, because I am a little sensitive on this subject,  I'm mistaken. So, I will try to explain properly why I can not agree with some of your thoughts. It is also a good opportunity to sharpen my thoughts and exchange more information.
I am afraid it became quite a story....

-Tension strong wood: To me every type of wood seems suitable, because even now you can adapt the design to the wood type. For example, an 80lb ash bow in this configuration does not seem like a good idea, but design (width) or wood type (stronger) can be adjusted.
-Trapped design or high crown sapling. Very unlikely to achieve the same amount of mass saving in a (massive) high crown sapling. Opinion is based on several high crown bows I made, better examples or proof very welcome! I do think this mass saving is possible in an extreme form of a trapped back, with hollow edges, a so called T-profile. Do not know if that's less work .... nice experiment though...let's try it? The curling effect may then expire and maybe it is less stable?
-Torque. See the two pictures above in my answer to Badger. The edges straighten out in a hollow back design. In four bows, so far, torque was not an issue. I have experienced it as a stable design.
-Stiff handle bow and grain: Design for outer form and handle was fixed for this experiment. I tried to make a bow that can be compared to 'normal' pyramid straight flatbows that I have made from similar board staves. Comparison, for the experiment, was therefore the most important, not the optimal shape or wood selection (btw, it is a straight grained bias ringed board stave). I will keep your suggestion in mind when I am going for the optimum.
-The sketch. Yes, that's just what it is, a sketch. A simplified representation to support the story. The pressure zone is a suggestion to represent the train of thought. To avoid confusion I had better omitted the measurements, old habit, sorry.
- Removing mass from Neutral plane and make Extreme U-profile. I'm not sure if I exactly understand the shape you discribe, but it sounds very interesting !! Could you please upload a sketch?

- Finally, why not a 'real' HLD? Simply because I wanted to see what a hollow back does. And sorry Simon, and with all respect, but to me, this sounds like you think anything outside of your version of hollow limbs is nonsense. That would be a shame.
I know HLD is your baby. And I love that beautiful baby. But there are other babies and they also deserve attention and respect. (And maybe one of them will grow up to be a real big bow).

Maybe it's not good to make a comparison with HLD here, because this topic is about hollow back. And both have their own characteristics and charms. But if we are talking about it.....
Actually, a few weeks ago I did a 'turn around test'. With a scale model of a bamboo stick bow, about 60 cm long. And frankly the results were not such that I continued with the hollow belly in my quest for saving mass.

See pictures below for the model.
With hollow back this bow draws 7 kg when the string is pulled 21 cm.
When turned around (hollow belly) it dropped to 6 kg. And tiller changed in the high stress sections.
My conclusion was that a hollow back can take more draw weight per bow-mass.
Now, of course, it is all about whether that also results in higher arrow speed. Draw weight, or f-d curves, in it self mean nothing. I shot a normal 400 grain arrow through both sides. Not a very good test, I know. But the hollow back performed noticably better.

Further tests with bamboo, also with adapted cross sections according to HLD, make me come to the following provisional comparison:
- Hollow back: more load per mass, tough draw, stable, rounding of the high edges is dangerous, curling up until sudden flattening and collapse.
- Hollow belly: sweet draw, potential unstable, rounding of the edges creates sweeter draw and more control in tillering, flattening out untill soft collapse.

So if you look for less mass per lb drawweight, hollow back seems to be preferable.
But I suspect two major benefits of hollow belly, please let the experts judge:
- The flattening out ensures that the pressure has a way to escape. Lateral movement instead of being pressed together. Result: little chance of pressure failures and probably little set. That would mean durable, high performance.
- The high thin edges in the belly are very suitable to heat-treating. Like the point of an all wooden arrow. Heat-treating a hollow belly might contribute a lot to better performance.

And there is another thing: These tests were done with straight bows and full-length cavities. In a bow with strong recurve and short working parts, it might certainly becomes another game...
Okay, I'm trotting on, right? ... all that comparising and suggestions might be something for another topic.