Return speed test (?)

[huisme]

Wood will just never be a strip of s-glass or carbon.

True.

But I will know if there is a diference between sixty samples of osage, sixty samples of locust, thirty maple, thirty hickory, thirty ipe, etc, and which ones average higher or lower in this test. I'd love to get some greenheart too, that stuff sounds insane.

I know I'm coming across as defensive but I don't have time to consider how useless more data might be, and there's no way to avoid suggestions I do exactly that.

[SLIMBOB]

Don't get me wrong huisme, the results from such a test will be interesting and I will be following to see the results if you pursue this.  The point I am making doesn't change as a result of my admitted curiosity.  Is this information new?  Will it translate into building better bows?  I am recovering from a long love/hate relationship with Black Cherry.  Well, actually I was in love with the idea of loving Black Cherry.  This infatuation was based on some of the engineering values associated with it, and some of the things Tim Baker had written about it.  After multiple runs at building my "dream bow" from BC, and ending up with nothing remarkable at all, I realized that yes, I can build a bow from BC if I build it the way BC bows need to be built.  The low hysteresis values were of little value in the real world, as you had to modify the design to compensate for its brittle nature.  You compensate for it and end up with a bow that performs like all the other bows from all the other woods.  Each having their own compensating factors.  Some were dogs that just didn't perform, some were exceptional, and most were somewhere in the middle, regardless of any engineering value.  A simple bend test on each piece of wood, for me anyway, tells me what I need to know about THAT piece of wood.

[PatM]

I'm just not sure what you intend to do with the info in a bow building application.

[huisme]

Post it in a few places, maybe write a new/more informed bow wood article for the wood database, maybe even contribute to the wood database itself, refer people to the information so they understand why I use some woods and not others, and otherwise enjoy the information.

Slim, that's pretty much what I intend to find out. It's not like anyone could look at this one test any more than any other number and know how every piece of everything tested is going to perform, but everyone who values information like that in the wood database already knows that and can function accordingly.

[Eric Garza]

But I will know if there is a diference between sixty samples of osage, sixty samples of locust, thirty maple, thirty hickory, thirty ipe, etc, and which ones average higher or lower in this test.

If you use slats created as described, then no you won't. You'll know if there is a difference between X samples of these woods that have unknown levels of wood fiber integrity. Which means, in the end, you won't know much of anything that's relevant to self-backed bows. You might choose to draw conclusions anyway, but those conclusions may and probably will be erroneous because you've decided to design your experiment to be easy to carry out rather than useful to carry out.

[huisme]

How much variance do you think there will be? Is a board bow so much worse than a single ring backed bow?

[PatM]

 Well Jay Massey was of the opinion that lumberized Osage lost most of its good qualities.
 What sort of slats were you thinking of?
 Your test will also skip the effects of trapping and crowns etc.

[huisme]

The crown issue was actually a concern at first as I like locust better with just a bit of crown, but since we're going for raw material properties rather than optimal anything (for this first test anyway, later on I could try shaping while keeping equal mass) it's okay that we don't shape anything to perform its best.

Which actually makes me think reducing thickness to match flex under x pounds would be better than width, then compression/tension doesn't come into play quite as much I think...

I don't know why it would when other woods can be used as backing strips. It has a decent MOR doesn't it? What's different? And how can I test it?

[Eric Garza]

What's the value of doing raw materials tests if the goal is to apply the data you generate to making bows? It seems like design is pretty integral to this experiment to me. Regarding the impact of fiber integrity, I think the variability it adds can be huge. Again though, it will depend on your wood choice. If you get perfectly straight grained boards that have been milled perfectly to follow that grain, then the difference between that and a shaped stave will not be so big. Especially with woods known to hold together well, like elm and hickory. But other woods are less tolerant of grain runoff (I'm thinking osage and black locust, among others), so for those woods fiber integrity will make a huge difference in how they perform in your test.

[PatM]

I meant what type of bow will this cause you to make  with the wood?

[Jim Davis]

I think a length of 18 inches or so has been mentioned for test slats??? At that length or longer, when you start narrowing them to have equal  mass you are going to start getting a different air drag, which will affect your readings.

The Forest Products Laboratories tested hundreds of samples of each wood to come up with their averages  for mechanical properties. Any smaller number of tests is going to render results that are accordingly less significant.

I have read this whole thread. I see a danger of introducing too many variables. Density within a species, moisture content, grain orientation, ratio of  spring wood to summer wood, and other factors would need to be held within very close tolerances to produce useful information.

The testing of non-homogenous materials just has too many variables to permit the extraction of some  types of useful data.  Average bending strength and work to maximum load are about  the only useful predictors in my estimation.

Jim Davis

[huisme]

Ash, that's why I've decided to return to just reducing thickness.

My results aren't going to be as thoroughly tested, but they will be results. If anyone intends to do hundreds of tests for each species I'll gladly sit by and watch, but so far I'm the only one who seems to be interested in justifying tests like this.

As for density, do you think reducing thickness over the whole slat would be better to control mass or bend under uniform weight? I intend to keep the slats in a controlled climate to help control moisture content and I'll take a reading as part of each test. I intend to use quarter-sawn samples that could have been used as self board bows; seems like if they'll throw an arrow they can be tested. Early/late wood ratios are certainly on my list of priorities.

Pat and Eric, it could cause me to make more from one species or less from another. If one material moves faster than another I look at that material and consider the other variables to decide if that's what I want after doing the same with other fast materials. Is this particularly fast wood available in large quantities in my area? Does is require ring chasing? Doe sit glue well? Is it suitable for laminates? Belly or back wood?

[Badger]

  I think Allens frequency checking test would be the most useful. If you are checking the speed of return for any accuracy you would need to know how much work it is doing returning. This would be very hard to figure out. All you are looking for in this test is histerias and frequency of vibrations would best reflect this.

[Jim Davis]

Hey Steve, here's another monkey wrench for me to brandish.

How about uniform slats bent to a particular distance and released, this time with a miniature microphone attached near the base and the output fed to an oscilloscope (or it's current equivalent) to read amplitude, frequency and duration of vibrations. It seems this would give a real comparative test of the resilience of each wood and the rate of loss of energy to histerisis.

What say you?

Jim

[Badger]

  Jim, I think that is similar to what Allen was proposing, I would agree with that as well.