Return speed test (?)

[blackhawk]

If all the lams are the same dimensional size wouldn't the different wood densities scew the results? Wouldn't a lighter piece return faster,or vice versa? To me they'd all have to weigh about the same mass wise,and also have the same bending weight(like draw weight). 

Someone correct me if I'm wrong as I'm not the brightest crayon in the box 

[huisme]

That's why I was thinking I'd need to many lams and didn't want to bend them especially far, just enough that they're taught through the whole process, and then use averages from dozens of tests noting the first, second, third, etc use of each lam.

Can you think of an alternative way to get dozens of dimensionally uniform pieces of a few woods? I know I can't get completely identical pieces out of locust, and this is fairly consistent wood.

Blackhawk, that's what I was thinking I'd need a good planer for, to reduce some of the denser woods to the same weight as standard thickness maple or whatnot. I figured leaving them taperless was the most reliable way to avoid better/lesser tiller interfering with results.

[avcase]

Measuring the return speed of a sample doesn't require very sophisticated equipment. You can firmly clamp one end of a sample, pull back the free end and let it go. The samples with a faster return rate will vibrate at a faster frequency/higher pitch.  The samples that the vibration dies out quicker have a higher degree of hysteresis. You can rank a pretty large number of samples from best to worse if you have a bit of a musical ear. There are apps for phones and pc's that can take it to the next level of sophistication if you want.

Alan

[son of massey]

 If you are concerned about mass and density you could certainly correct for that. Weigh each sample and find the volume of it and you have the density for that piece-you don't even have to use the general published average values. If you then find the [(return rate)/(density)] for each sample you should have comparable results.

SOM

[DavidV]

If you are concerned about mass and density you could certainly correct for that. Weigh each sample and find the volume of it and you have the density for that piece-you don't even have to use the general published average values. If you then find the [(return rate)/(density)] for each sample you should have comparable results.

SOM

+1

I guess the most accurate way would be to grind your own and make them the same weight but then you would need a drum sander :p

[Badger]

  I would try making them all the same length and thickness then adjust the width so they all had the same pull down load.

[huisme]

but then you would need a drum sander

Eeeeexactly

Badger might be on to something, maybe I need to be looking for the width/thickness at which a sample bends to the desired six inches under equal weight. I do have a belt sander and access to a planer, but it's going to be a pain in the butt either way... But I will be doing this with weighed samples.

Avcase, I hadn't thought of that at all! I'm a drummer, so the rest of the band would say I'm no good for this but the fact is I just so happen to hear all the times the other instruments are out of tune

[Eric Garza]

I think this is a neat experiment, but density differences from one piece of wood to another will make it challenging to isolate the quality you want.

For instance, you might make your slats a standard length and thickness and vary width so that their masses are identical. This way you've factored out mass, and can measure return speed without that confounding variable. This would yield useful information.

But by keeping mass constant by varying width, you would be ignoring the fact that skilled bowyers commonly vary width to get different levels of performance from wood. So you might also try using slats of a standard length and width but varying thickness to hold mass constant. This would also yield useful information, but it would yield subtly different information than the above test.

And, of course, you could hold length, width and thickness constant and vary the mass of the slat. This would also provide useful information, but again the information would be subtly different from what you learn by the two above tests.

Because wood varies so much from tree to tree and even within a single tree in terms of its physical properties, you can't test a slat from a single species and generalize the data you gather to that species. You've made a single observation, and you have no idea where that single observation falls on the bell curve that represents the totality of possible values from that species. If you tested several samples from several trees of that species, you'll get a better sense both for the average value and how variable the values are among different wood samples. Of course the better you want to understand average performance and the variability around that average, the more samples you need to test and the more time and resources you'll need to invest in testing.

Finally, I'll second the idea that milled slats might not be the best option for this test. Sure, they're easy to get in quantity. But they don't represent wood that we use for bows, at least those of us who make self bows and try to use an uncompromised ring for the back of our bow. Using slats with uncompromised wood fibers adds more complexity to your experiments in the sense that you need to source vary particular slats, but in the long run I think it makes the results more applicable to bow making.

My two cents...

[Badger]

  Allens suggestion I think would give you the most useful information. I didn;t see his post till just now.

[PatM]

I just don't se this having any practical value. People are still going to just select a piece of wood and judge how that piece will react to being made into a certain design.

[huisme]

you would be ignoring the fact that skilled bowyers commonly vary width to get different levels of performance from wood.

I actually do intend to ignore that. I'm looking for what the wood does, not what we can do with the wood. It seems like as long as the wood is shaped consistently, probably edge instead of broad face reduction, results should be relevant.

I was thinking thirty samples from various trees in various locations for most woods and separate stacks representing average and exemplary ratios in ring porous woods like osage and locust. Furthermore, each piece would be tested five times to note degradation.

I can think of no replacement for milled slats though. This is already going to take me a long time to prepare and execute; add splitting, ring chasing, and sanding to bend uniformly and you're talking possibly two generations worth of work just to get the material in the testing jig. I'm open to suggestions if anyone knows how to get clean-backed slats, but otherwise the ~8" of bend over several feet doesn't seem like it's going to completely ruin the samples.

I just don't se this having any practical value. People are still going to just select a piece of wood and judge how that piece will react to being made into a certain design.

We will not have refrained from learning because it's easier to do so. We will know more about what makes a bow wood, and more about ourselves when we use or disregard information.

[Badger]

  I do think it is difficult to apply things like this, I am constantly trying. I do believe every variety of wood has an optimum thickness for a given radius. Finding it is the challenge.

[SLIMBOB]

I'm with PatM.  I'm not sure what you will come away "knowing" that you don't already know.

[huisme]

I will know how quickly a slat of given wood at a uniform length and thickness reduced in width to bend eight inches under uniform weight will return to two inches of bend, after which I can compare woods. I will know which wood, on average, degrades faster from said testing as well. I will, in effect, know if some woods are faster as bows, and I won't be persuaded to be ignorant.

[PatM]

The point is that you will figure all this out and then go out and cut down  a fresh tree and be right back to making an educated guess when you start building the next bow.

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