Flat belly, Round belly

[Matt A]

I've been wondering this lately, is there certain woods that prefer a flat belly or round belly or is it only a performance/preference issue?
And is it lets say, all white woods need a flat or rounded belly or other woods like osage need a flat or rounded belly? If you get what i'm saying. 

[rossfactor]

Hi Matt,

In general woods that have greater compression strength (e.g. ability resist compression of the belly cells) can tolerate a rounded belly better than woods with less compression strength.  A more rounded belly allows you to design deeper narrower limbs.  However a wide flat bow, with proper limb design can be as fast as a narrow limbed bow. 

There may be some benefit to narrow rounded limb design in reducing hysteresis (internal friction) as well as vibration, but I would be a bit out of my league in answering that question specifically.

Gabe

[JackCrafty]

A wood that is weak in tension but comparatively strong in compression, like yew or juniper, should have a rounded belly.  A wood that is strong in tension but comparatively weak in compression, like hickory, should have a flat belly.

"Comparatively" is the key word here.  The best way to tell is with an experiment.  Cut two sticks of the same wood in the same way, with one side flat and the other side rounded.  Bend one with the rounded part as the back and the other with the rounded part as the belly.  The stick that takes the least amount of set will tell you if the belly should be rounded or flat.

[rossfactor]

Good compression woods include: yew, osage, cherry, maple, ipe, ash, apple, and jatoba.

Poor compression woods often include: hickory, oak, purpleheart, and walnut.

As you can see high density does not necessarily mean good compression strength.

These lists are based on my experience and research, others may have other opinions.

Gabe

[PEARL DRUMS]

For what its worth every bow I build regardless of wood or style (except ELB) is flat bellied with pea sized radii on the edges.  I like for my limbs to have shoulders. A touch of robustness never hurt a wooden bow of my own building. 

[crooketarrow]

  I like to and do make all my limbs to a shallow oval. Even wood s like hickory which I build bows from a lot seams to get by just find. I built bows from hickory with flat belly and as long as you keep the limbs to a low oval I can't see any diffenence in cast,speed,hand shock or even set and string follow.
  Woods like Osage like rounded bellys again I can find no difference in a shallow oval even though it works good with a deeper oval. I've found any any woods work good with this a shalow oval. I can build my most of my bows to keeping a little reflex no matter the wood with a shallow oval belly. I reflex all my staves while green to have 2 to 3 incks of reflex before I start. Doing it like this
makes a smooth well rounded no sharp edges that just look better when your finished.

[Pappy]

I'm with Pearl.
   Pappy

[Jim Davis]

We have been left without information by the Forest Products Research group. They give statistics for compression strength, but none for tension strength. The relationship between compression and tension is different for every wood species.

Good information above, generally, but most woods are 3 to 4 times as strong in tension as in compression. The known exceptions include, as noted above, yew  and cherry--eastern red cedar too. (I've not seen statistics for the others noted above and have not found ash or maple to be stronger in compression.)

With careful tillering, we can get away with rectangular sections for all the other woods, but the belly is doing all the work. If the backs of bows made of these woods that are stronger in tension is narrowed by 30 percent or so, the back begins to share the work load. That makes it easier to avoid chrysals and frets. It also makes for slightly lighter limbs and better speed.

Hickory is about 3-1/2 times as strong in tension as compression. Elm seems about the same. Both do very well with a flat belly and a narrowed back--even a highly crowned back, as when made from a 2- to 3-inch sapling.

A very good suggestion by jackcrafty to test the shapes of belly and back for a given wood. I have done that with red cedar and the results very clearly showed that red cedar is much weaker in tension than in compression.

As I said earlier, careful tillering can result in a good bow with oval or flat belly, but with most woods, leaving the back full width makes the belly do all the work.

Jim Davis

[mikekeswick]

I agree with the above post.
Trapping of bows backs is underated in my opion. - it can make a real difference to set when used correctly.

[Matt A]

thanks for all the info! how about a real dense wood like HHB? would that more than likely require a flat belly

[scp]

The Wood Database http://www.wood-database.com/wood-identification/ has some interesting information. I don't know what they call it, but if we divide the elastic strength with the crushing strength, we come up with a number that might be useful in making better bows. If we compare the numbers for several bow woods, we might be able to guess its meaning a little better.

pacific yew  166
mulberry  167
sassafras 170
osage orange 179
dogwood 199
black locust 202
pignut hickory 233
white oak 241
red oak 266

What do you think?

[Jim Davis]

Interesting site. Not sure what is meant by elastic strength, so also not sure if the ratio you suggest is useful to us.

The mechanical properties of wood are usually given in such terms as "modulus of elasticity," "modulus of rupture," "work to maximum  load" and such.

Of most direct value to bowyers would be "compression strength parallel to grain" and "tensile strength parallel to grain." The first can be found at http://www.fpl.fs.fed.us/research/centers/woodanatomy/techsheets_display.php?geo_category_id=2&genus_commonname_criteria=c&sorting_rule=1a but the Forest Products Laboratory doesn't offer tensile strength data.

If we had compressive and tensile strength numbers, we could come up with a very useful ratio for every wood type. We could make the back-to-belly width ratio fit the tensile/compressive ratio and be off an running, or some other metaphor. 

I found this tidbit for these woods

Wish I could find that kind of data for all the other common woods. It is exactly what I would want.

Jim Davis

[Badger]

    I very slightly round my bellies so the tools work easier and then durring the final tillering my scraper usually flattens them out pretty well.
    I don't find engineering specs very usefull for selecting bow woods. Strong in compression has nothing to do with good in compression. Many of the strongest compression woods are terrible about chrysaling. How strong something is refers strictly to how much it resists bending. The ratio of tension to compression probably has more effect on where the nuetral plane will fall. The amount of elasticity overall will determine how thick we can make it and how tightly we can bend it.
    A bowyer has little interest in when a wood will fail, a bowyer concentrates his efforts only within the elastic limits of wood. Most woods fall somewhere around 1% plus or minus about .2%. This sounds small but a wood with 1.2% elasticity has almost 50% more elasticity that a wood with .75% elasticity.
      The strength of something probably would interest us most as a ratio of mass weight in the wood or specific gravity. If a wood were not very strrong in tension or compression but was very dense and heavy it would not be good, however if a wood were light but only medium strong in might be a great wood.

[scp]

Interesting site. Not sure what is meant by elastic strength, so also not sure if the ratio you suggest is useful to us.

Terms are explained in the site.

I find it fascinating that the number 166(%) for pacific yew is rather close to the ELB width/thickness ration of 8/5 and the similar number for Sassafras might explain why certain oldtimers liked that wood.

It's just a heuristic hypothesis but the numbers might be useful in getting the starting ratio of width and thickness for each wood. I wonder, though rather extensive, how reliable the data in the site is.

Anyhow, I find it intersting that osage orange and mulberry have rather low numbers like yew and sassafras.

BTW Badger, I'm using elastic strengths, not rupture strengths. I'm still trying to incorporate the specific gravity into the hypothesis.

[Danzn Bar]

This site is incredible !!!!  From one end of the spectrum to the other.  Sicentific comment to the most common sence simple comments.  I love this site ...so cool!