Mechanical Properties of Wood for Building Bows and Arrows

[pnwarcher]

Hi PA Community! I'm new to posting here, but have been stalking these forums for quite awhile. I've gleaned a lot of knowledge from y'all, and I think it's time I gave something back.
Like many of us, I am a total wood nerd and am always eyeing trees and shrubs for potential use making bows and arrows. I'm also a mechanical engineer, so naturally I got to thinking it would be interesting to build a database of wood properties and do some analysis to help identify potential new woods. When I discovered The Wood Database http://www.wood-database.com/, I knew I'd struck gold. I used this database to calculate some properties relevant to bow and arrow performance, and tabulated everything in this handy Google spreadsheet (also attached as a PDF):

https://docs.google.com/spreadsheets/d/191XdO8Gb0uo2lW5oYLw__SOSh5Mv9Sfk91qr4D6BWAM/edit?usp=sharing

The header cells have notes describing each mechanical property. Of course, the raw numbers must be taken with a grain of salt since wood is a natural material and there can be wide variations within a species. Maybe someday I'll try to quantify these intraspecies variations for some of our favorite woods.

If you don't see your favorite niche bow wood, it's likely because it is not widely available in the wood market, and therefore hasn't been tested by Eric Meier (the author of the wood database). He does accept wood samples for testing, so I encourage you to send him a specimen! http://www.wood-database.com/donating-wood-samples/
Below are notable woods missing from the database, which can likely hold their own with the best. I have access to the first six of these, so testing them is on my to-do list. The other five, I would love to get my hands on someday.
- Ocean Spray
- Vine Maple
- Hazel
- Lilac
- Hawthorn
- Scotch Broom
- Serviceberry
- Red Stopper
- Laburnum
- Mountain Mahogany
- Syringa

Let me know if you have any questions about the spreadsheet, and feel free to give feedback or make suggestions for improvements or additional features.

Happy bowmaking!
Stephen

[pnwarcher]

If there's one table worth studying in that massive spreadsheet, it's probably this one (see attached pic). This is the list of woods sorted by breaking strain, which is the amount of deformation or bending a wood can withstand before breaking. Woods with a high breaking strain are woods that are strong (high modulus of rupture) and also flexible (low stiffness or elastic modulus). Breaking strain is one of the most important mechanical properties for bow wood. Woods with higher breaking strain allow you to make a bow with shorter NTN length, longer draw length, more extreme recurve/reflex, and narrower, thicker limb cross sections. They also give you more margin for error when using less extreme designs.

Interestingly, while yew and osage orange are near the top of the list, they are not the very best in this metric alone. Makes me really want to try Madagascar rosewood... too bad it's endangered and banned from international trade.

[joachimM]

pnwarcher,

I did a similar exercise on some 100 species of wood from the wood database data.

https://drive.google.com/open?id=0B3YYA3Sr_3gqb0NpazZqdUFJSTQ
Also with information on backing materials.

Since then, I found out that you need to take these things with a decent grain of salt. Some of this error / variation is due to moisture content; the data in the wood database refer to 12% MC, but we often have lower MC, and heat treatment lowers MC on the belly, not the back (that's why it has such a dramatic effect).
Interesting for bow wood selection and testing, though. Certainly not the definitive answer to all your bow wood questions.
See also earlier threads, like http://www.primitivearcher.com/smf/index.php/topic,50077.msg765904.html#msg765904

[Sir Failalot]

This is gold.
Yes I will take it with a bit of salt. Then salty gold 

 Thank you very much!

[willie]

Welcome to PA, Stephen. Its nice that you are making a contribution with your first post.
MOE does seem to be a good indicator for usability in bows, but sometimes there are surprises when using woods further down the list. What sets some staves apart from the expected/estimated is something that all of us wish we knew more about.

[Del the cat]

Interesting that Leyland Cypress(Leylandii) is up near the top. This fast growing tree is prevalent in the UK where it is planted for screening purposes and is often cut because it grows too big too fast. Dunno it it is easy to find clean knot free lengths tho' as it grows bushy (hence it's good for screening)
Del

[Dances with squirrels]

Yew and osage are up there pretty high. I'm gonna try those 

Welcome pnw!

[gfugal]

It's funny I started building my own spreadsheet too a while back. JoachimM  then shared his with me too. We had a good discussion on a topic I started here http://www.primitivearcher.com/smf/index.php/topic,59771.0.html

[gfugal]

If you don't see your favorite niche bow wood, it's likely because it is not widely available in the wood market, and therefore hasn't been tested by Eric Meier (the author of the wood database). He does accept wood samples for testing, so I encourage you to send him a specimen! http://www.wood-database.com/donating-wood-samples/
Below are notable woods missing from the database, which can likely hold their own with the best. I have access to the first six of these, so testing them is on my to-do list. The other five, I would love to get my hands on someday.

I think it's awesome we have an engineer in this form. I'm not an engineer, but I'm a science major and love applying the physics I learned to Bowery and archery in general. My brother is a mechanical engineer and I'm always going to him with questions. Do you happen to know how Eric Meier does his wood testing? Is there a way we could do that ourselves on species not listed?

[Jim Davis]

The factor most needed but not included is YIELD strength not breaking strength (though the two correlate to a degree). If you stay clear of the breaking strength but exceed the yield strength, the wood takes "set."

And I always am slightly annoyed that  Eric Meier is given credit for all that testing when in fact, the information has been lifted from the Forest Products Laboratories' work done nearly a century ago. Unless I am mistaken, which I doubt, Meier has no equipment to test MOR, MOE, or WML. I will admit his arrangement of the information is easier to find and read.

And again, it should be recognized that NOBODY has those moduli numbers for DRY Osage. For some reason, that data was not recorded in the original testing and all we have are the data for green Osage.

[willie]

Is there a way we could do that ourselves on species not listed?

yes, a simple bend test. Actually, Tim Baker developed a test for evaluating bow woods, I think there is a description in the bibles, but a simple three or four point bend test similar to spine-ing arrows is an alternative entirely doable by a bowyer who wishes to assess his materiel.

most needed but not included is YIELD strength

Its a shame that the FPL has not published the graphs that they made to get the data points in their tables. The graph is nothing more than what we would draw when making a F/D curve on a bow.  Wood typically yields in a gradual manner, and a "yield strength" is somebody elses idea of the usefulness the materiel being worked beyond the elastic limit. Being able to see the differences in shape, slope and length of the curves for different species would be much more useful for us bowyers. The image below shows a test where the piece was unloaded and reloaded. What we call set is labeled "plastic deformation"

What would be really sweet is to find a way to quantify the rebound rate of wood near the proportional limit with a test procedure that more closely mimics the draw and release speed of an archer.

[joachimM]

Interesting that Leyland Cypress(Leylandii) is up near the top. This fast growing tree is prevalent in the UK where it is planted for screening purposes and is often cut because it grows too big too fast. Dunno it it is easy to find clean knot free lengths tho' as it grows bushy (hence it's good for screening)
Del

I cut and shredded 15 big leyland cypresses two weeks ago, but kept a few logs just in case. I’ll try this shortly!

[Del the cat]

Interesting that Leyland Cypress(Leylandii) is up near the top. This fast growing tree is prevalent in the UK where it is planted for screening purposes and is often cut because it grows too big too fast. Dunno it it is easy to find clean knot free lengths tho' as it grows bushy (hence it's good for screening)
Del

I cut and shredded 15 big leyland cypresses two weeks ago, but kept a few logs just in case. I’ll try this shortly!

I'll wait with interest
Del

[gfugal]

Is there a way we could do that ourselves on species not listed?

yes, a simple bend test. Actually, Tim Baker developed a test for evaluating bow woods, I think there is a description in the bibles, but a simple three or four point bend test similar to spine-ing arrows is an alternative entirely doable by a bowyer who wishes to assess his materiel.

This is helpful, yes, but this doesn't produce numbers like those displayed in the database. So really there is no way of comparing woods on the database unless you test them all yourself.

[willie]

but this doesn't produce numbers like those displayed in the database.

I am not sure why you think it can't. Could you be more specific as to what useful values in the database can't be produced with a simple bend test?, or why you would need to test them all yourself to get what you need to know about the one you are working with?