R/D Tiller Discussion Continued

[Bayou Ben]

How do you adjust your height Badger?

[Stick Bender]

Ben I worked every Saturday extra for a sewer contractor sucking out sewer systems to be able to cheat with that...lol it was blood money !  This is a great thread let me throw this out why tri lam I understand the Perry reflex part & glue lines for the form shape , but Marc that makes a lot of lam or backed bows says he has experienced no performance difference between  using 2 lams vs 3 lams in a previous thread ? But I have some lams setting in a box collecting dust while I finish the 2 current bows , I was going to try a tri lam Pyramid , core is going to be bamboo,the belly osage , and back hickory , I was going to run the osage belly up the ramps , my core & belly lams are 0.150 with the belly 0.001 taper so I just need to figure out the stack total before grinding the hickory , I need to be close ,not much wiggle room for tiller ! I was going to try to isolate most of the reflex in the outer 1/3 just a exsperment really !

[Badger]

How do you adjust your height Badger?

  I use the height adjustment built into the planer. I also have a long plate that sets on the adjustable side of the sander not seen in the pin. The plate is made from channel iron and inlaid with a 1/2" thick strip of marble for flatness.

[Springbuck]

Stick, I'm not sure there is an "advantage" to a tri-lam, it's just another way to do it.

   Theoretically it might hold the shape off the form better (I think it does, less "spring-back" as you release clamps, but less "Perry reflex" benefit, as well?), two stage glue-ups give you more time to mess with glue and alignment, and you might be able to utilize a lighter core, etc.  but I can't see any real differences.   

The only thing I THINK I saw a difference with was when I was using a lot of bamboo flooring.  That stuff was developing a reputation for taking set, sometimes lots, but never seemed to really hinge or fret.    I started doing two stage glue-ups when I went to TB III, and gluing up belly and core in the form in one stage, then back to the same form for the backing SEEMED to make the bows take less set.

On the other hand, that MIGHT have been about the time I started to have a clue what I was doing.

On that note, has anyone actually tried the method Baker theorizes for laminated bows, where he suggests that gluing up two thin lams in "extreme exaggerated reflex", then gluing each consecutive lam into less and less reflex SHOULD make a laminated bow shoot faster?  I never have, because my laminations aren't that thin, and I can get the profiles I want with two or three lams.

[Bayou Ben]

"Blood Money", haha.  I may have to get you to make me a couple custom sleds with you blood money grinder; I should watch what I say...And if I'm reading correctly, you already have osage at .15" and bamboo at .15" and you are trying to figure stack for the hickory?  In my designs with a dense belly, light core, and bamboo back, I need at least 1/2" total stack to get to 50 lbs.  That would put your hickory pretty thick for the back.  Of course your design will vary some...

I see now Badger.  It's one of those jointer/planer in one units.

Springbuck- why would you think there's less Perry reflex?  Unless my understanding of perry reflex is not right, I would think there should be more.
That is an interesting method you mention about less and less reflex.  Seems that it would have a spring type effect with coiled up potential energy.  I may have to try that out but it will still probably only be with 3 lams.  3 is too much of a headache sometimes, and I couldn't imagine the glue weight on 4 or more lams 

[gfugal]

Set tapers are much more reliable than the eye on these bows. I use a slightly less precision method but I do pre tiller my belly on the belt sander and it gives me a nice shape to start with when I start bending the bow. My eye will always tell me to get it bending more right in the curve and that aint always so.

Badger I remember you saying that at times you will make at least 1-2 bows a week. I wonder if this "pre-shaping" allows you to be quicker in your bow building. I find that getting from roughed out to brace is the hardest and slowest part for me in bow building. I feel if I had a better expectation of what thickness the bow will be beforehand then roughing it down closer to final shape would speed up that process tremendously. It sounds like you have a good grasp on what it will end up being and can thus predict quite accurately how much you need to take off in pre-shapping before you even get to tillering. I wish I could pick your brain but it's probably something that just comes with experience. In general how thick do you rough your bow down before you begin tillering and do you rough out a taper before you begin with all your wood bows on the belt sander?

[gfugal]

That is a hard question to answer depends a lot on working limb and design  the tapper rate changes but I just measured 1 of my osage lever bows with a 20 in. working limb  68 in TTT and the average taper rate is  0.012  I'm seeing a average more like 0.010 for my self bows maybe I'm wrong Steve ?

  That sounds right, it just depends on the bow and the tiller you are after. Most of mine are what I would call semi pyramid. slight limb taper most of the way down. Parallel would likely come out around 10 but I never measure.

Parallel limbs will have a more aggressive thickness taper on my self bows.It's hard to judge.When width taper takes over taper thickness wise slows down a little.

So I'm going to try to deduce based off these comments and feel free to critique if you don't think it logically holds up. So a bow needs to taper to bend evenly, or else you would get all the bend right in the middle if it was a flat board, or all the bend will occur right out the fades if it had a handle on it. It seems you can achieve your taper by either tapering width (pure pyramid bow), thickness (parallel limb bow) or a combination of the two. With beam theory, I believe it's stated that removing 1 unit of thickness on the belly is equivalent of removing 8 times that unit from the sides. If Stick Bender is accurate in saying his bows require 0.010"-0.012" taper, then I want to propose a hypothetical. Let's say Stick Bender's bows are parallel limbed as BowEd kind of suggests cause that's a more "aggressive" taper than most people use in the Trad camp. Whether this is true or not doesn't really matter for this thought experiment. If it is true it may be possible to determine the taper you need from either the thickness, width, or both. Supposing a 0.01" thickness taper rate is all the taper you need on a parallel limbed bow, then using beam theory a 0.08" width taper rate is what you'll need for a uniformly thick pyramid. For a 66" bow with 6" handle/fades and 6" stiff tips this would result in 24" working limbs and would require taking off around 0.96" on each side right before the stiff tips. if you still wanted a half inch wide tip you would have to have the limb be 2.4" wide when it starts right out the fade. I'm not sure if this is realistic, it's just what I'm deducing based off of Stick Bender's measurements and Beam Theory (if it can be applied so simply). Subsequently, you could also determine how much width taper you still need if you don't do a 0.01" thickness taper rate or vice versa. For example if you only have a thickness taper rate of 0.004" then you would need a 0.048" width taper rate as well. I calculated that as such (0.01-0.004)x8 = 0.048". Do you think there's any merit in calculating it like this? Sure it's more important to listen to the wood and how it's tiller looks than getting it shaped based on calculations. But maybe it can help when designing the bow and roughing it out to shape.

[Badger]

Set tapers are much more reliable than the eye on these bows. I use a slightly less precision method but I do pre tiller my belly on the belt sander and it gives me a nice shape to start with when I start bending the bow. My eye will always tell me to get it bending more right in the curve and that aint always so.

Badger I remember you saying that at times you will make at least 1-2 bows a week. I wonder if this "pre-shaping" allows you to be quicker in your bow building. I find that getting from roughed out to brace is the hardest and slowest part for me in bow building. I feel if I had a better expectation of what thickness the bow will be beforehand then roughing it down closer to final shape would speed up that process tremendously. It sounds like you have a good grasp on what it will end up being and can thus predict quite accurately how much you need to take off in pre-shapping before you even get to tillering. I wish I could pick your brain but it's probably something that just comes with experience. In general how thick do you rough your bow down before you begin tillering and do you rough out a taper before you begin with all your wood bows on the belt sander?

  I don't use any power tools on self bows, on lam bows I only use the belt sander prior to glue up and then once again to clean it up after I remove the rubber straps.. I never measure thickness. Sometimes I use a thicker backing than others, I don't have a set system. If it is completely rigid coming out of the glue up I might bring it down a bit with the belt sander but that is unusual,

[Badger]

   I do weigh my pieces, if I am projecting a bow to be 19 oz I will want the pieces in the glue up to weigh about 23 0z. I will do some narrowing and thinning as I build.

[Bayou Ben]

gfugal, beam theory is based on the moment of inertia of a rectangular section, where inertia is proportional to the bending strength of the material or composite.  So the moment of inertia of a rectangular cross section is the following equation, I = [(Width)x (Thickness)^3]/3.  I think there's confusion since people often say double the width and double the strength, but double the thickness and you get 8 times the strength, 2^3=8.

Plug in some of your numbers into that equation and you can see how width and thickness affect strength.  It's a cubed root relationship between width and thickness. 

[gfugal]

gfugal, beam theory is based on the moment of inertia of a rectangular section, where inertia is proportional to the bending strength of the material or composite.  So the moment of inertia of a rectangular cross section is the following equation, I = [(Width)x (Thickness)^3]/3.  I think there's confusion since people often say double the width and double the strength, but double the thickness and you get 8 times the strength, 2^3=8.

Plug in some of your numbers into that equation and you can see how width and thickness affect strength.  It's a cubed root relationship between width and thickness.

Ah, I see. That statement is only true in that specific example. It's not that changes in thickness always 8 times as strong as changes in width. I'm going to have to do some calculations and try and digest this. Thanks man.

[Springbuck]

Yeah, the "twice as thick=eight times as stiff" is not good math.  Just a rule of thumb.

[Springbuck]

Greg:   I feel if I had a better expectation of what thickness the bow will be beforehand then roughing it down closer to final shape would speed up that process tremendously.

   Yeah, and with R/D lam bows I definitely had to learn not to start too thick.  I was gluing up 5/8" slats with a 1/8" bamboo back at first, which caused things like the bow gaining reflex as I tillered down to a reasonable thickness, and how it took me hours to work down to where I could even get limbs to bend with my whole body weight.

  For OTHER than laminated bows, this is what floor tillering is for.

I get the idea of the long single paragraph post, but I can't break it down for you without going over it sentence by sentence.  Let me say this, though, that there is merit in doing it basically this way if you care to, have the time, etc.  HOWEVER, there are flies in the ointment.  As mentioned, bow length and working limb length changes the rate of taper needed.   I imagine deflexed portions would, too.   Perry reflexing makes the limb MUCH stiffer than gluing on a backing flat and varying the radius of that curve will vary that stiffness as well.   Maybe more.

[Springbuck]

"why would you think there's less Perry reflex?  Unless my understanding of perry reflex is not right, I would think there should be more."

 As I understand it, the whole idea behind Perry reflex is that bending the laminations and gluing them COMPRESSES the inner surface of the belly lam and STRETCHES the inner surface of the back.  Likewise, it stretches the exposed belly (slightly) and compresses the exposed back slightly.  Agreed?  Hence, as each tries to relax back or bend, each resists the other doing so at the glue line.

  So, as I understand it, there must be some strain in the system created by bending the lams,in order to create Perry reflex in the first place.  As with a bow limb, you create more strain by bending a thicker lam than a thin one.  If, for instance, you made an entire limb from 1/40" thick veneers (all glued at once) each thin lamination would be too "limp" to create much effect.

I once had a LONG discussion about this with Dan Perry and others on Paleoplanet.net, which culminated with me asking him, "Would it be better to bend one 3/8" belly lam and apply a 1/8" backing, or to Perry reflex two 1/4" lams?"  Which would take best advantage of the effect?  He wouldn't tell me, so I know it was an important question.    He just described how to make models to work with from foam rubber, rubber cement, and marking pens. 

 

[willie]

"Would it be better to bend one 3/8" belly lam and apply a 1/8" backing, or to Perry reflex two 1/4" lams?"  Which would take best advantage of the effect?  He wouldn't tell me, so I know it was an important question. 

https://www.tapatalk.com/groups/paleoplanet69529/calling-dan-perry-how-to-tiller-t53163-s20.html#p579487