Bamboo Tips - Tips Area
Rod Design - Converting Geometries


< Home < Tips Area < Rod Design < Converting Geometries

Rule

I have received, seen, numbers varying from 0.89 to 0.93 X (multiplying the next taper values by 0.89 to 0.93) to convert hex to quad tapers.  Could we have a discussion from those who have tried the different conversions?  Is 0.89 to whippy and 0.93, shall we say brisk?  (David Smith)

    The accepted conversion factor by the majority is 0.92 and this is a purely mathematical derivation based on comparing like to like areas at any given point.  I got this information from the late J Irgens from Rice Lake, WI.  (Paul Blakley)

    I use equivalent cross sectional area.  (Darryl Hayashida)

      I didn't explain at all. I match the stress curve using the same formula as a hex, but using a square to calculate the area. I find that the differences are very small going with the straight equivalent cross sectional area.  Calculate the area of a hex at a diameter, and use the same area, but in a square.  (Darryl Hayashida)

    Strictly speaking, it's the integral of the cross sectional area. I have done some research on the stress formulas and have a small PDF file (Acrobat Reader 4.0 or later needed) that explains it all.  It also develops the stress equations for quads and pentas.  (Claude Freaner)

      I am probably making this too simple, but don't you just want to make sure the dimension to be converted (in this case the quad dimension) will produce an identical moment of inertia as the hex dimension?  (Kyle Druey)

    The approximate equivalent cross-sectional area of a square compared to a hexagram is the "hex * 0.93"... not exact, but very very very close.  However, when making a fly rod, you have to consider a couple of things. Is your hex a true hex and is your square a true square?  If you plane or sand the outer flats on the rod to "flat" then yes they are, however, many of us leave the outer flats to the shape of the convex of the original culm.  Now, when you get down to dimensions as small as our rods, they you have to take into consideration the material that will be contained in that part of the convex that is above what would be the "normal" flat of the rod, which will NOT be consistent, but will change with each piece of cane you use, and believe me, when you leave that outside natural, removing only the enamel, there is a marked difference from culm to culm, therefore from rod to rod.  So if you leave the outer flat rounded, you have to consider that the .93 conversion factor just doesn't work in all cases.  The dimensions are so small in the tip section that it will probably be close enough, but in the butt section, you have a greater area, therefore the differences are of a greater magnitude.  Think about it like this.  Take the 4 flats in the butt section of a rod and glue to it a piece of rounded sliver from a culm that's 2" diameter and a couple of thousandths thick... it WILL change the action of a rod, therefore, in my opinion, which was greatly influenced by Claude Freaner, you must consider the diameter and radius of a culm that you make a rod from.

    Anyone interested in this should contact Claude Freaner.  He has done some VERY extensive work on this, which he was kind enough to send me, including the explanation of the math, and it allows for factors that we must consider, especially if we leave the outer shape of the culm as we build the rod.  I think the math, if you care to delve through it, will explain a lot more about the considerations that must be made in converting hex to quad, and when it all comes down to it, there are a lot more considerations than are normally in place with a straight 93% conversion factor.  For those of you that don't presently build quads or are just starting, reading Claude's paper is very very helpful.  (Bob Nunley)

    A quad is stiffer if you use the same flat to flat measurement as a hex.  (Darryl Hayashida)

    The diameter or strip height of the quad strips should be .931 times that of the hex diameter or strip height to yield the same cross-sectional area.  That does not take into account leaving the curve of the culm on the outside of the strip, a practice used by some makers of quads.  (Bill Lamberson)

    If you use geometric conversions like .92 you will get a quad that is noticeably slower than its hex equivalent. To get an equivalent action you need a totally different quad stress curve. John (the quad master) Zimny would be the one to give us more insight.  (Bill Fink)

    If you make say the Sir D as a quad at equal dimensions to the hex, it will throw a 6 wt through a screen door. At .93 all the way down, it will be a pretty fast version of the original. If you make the tip at .93 and then  progress down to .87 or so in the butt, you will get a nice conversion, but it may be weaker in the butt than some like. Personally, I have been tapering down from .93 to about .90 in the butt. I like the stiffness.  It all really depends on the action you are trying to match. This is based on 5 or 6 rods I built on the conversion factor this year. The straight .93 Sir D and Driggs were nice, and really quick. The straight .93 Para 15 was a broomstick, but some guys liked it. They were all guys who could cast a broomstick if needed. The .93-.90 versions of the Driggs and Para 15 were much nicer.

    Something like this is what I mean...

    In a hex rod to quad conversion...

    • Make the tip dimensions .93 of what the hex rod is.
    • Make the butt section .93 at the ferrule, .92 at the 5" and 10" station (from the tip of butt ferrule), .91 at the 15" and 20" station (from the tip of the butt ferrule), and the rest .9 of the hex size. Experiment with tapering down the butt, but I would not go with a factor lower than .87 or so. .90 seems to make a nice stiff butt, a good dry fly action.

    For example:

    Paul Young Driggs 7'2" for../p>

    Stat    Hex     Quad
    0       .070    .065    (times .93)
    5       .091    .085    (times .93)
    10      .109    .101    (times .93)
    15      .127    .118    (times .93)
    20      .142    .132    (times .93)
    25      .152    .141    (times .93)
    30      .160    .149    (times .93)
    35      .185    .172    (times .93)
    40      .200    .186    (times .93)
    45      .211    .196    (times .93)
    50      .219    .201    (times .92)
    55      .229    .211    (times .92)
    60      .245    .223    (times .91)
    65      .259    .236    (times .91)
    70      .264    .238    (times .90)
    75      .265    .239    (times .90)
    80      .265    .239
    86      .265    .239

    I have since given this up and have been building Edward's tapers and "my own" tapers further derived from the original conversions mentioned above. I have used Claude's spreadsheet to convert some tapers, but I have not really had time to build these experimental rods yet.

    I am really more interested now in straight tapers as a basis for quads, tweaking here and there. I really see little use in trying to create a fantastic quad from a hex taper. Why not just build the hex? (Just a devil's advocate sort of thought). I love quads, and do not get me wrong, but I think some of their appeal is real and some is imaginary. A good rod is a good rod, and it should track well. I really believe the Milward idea he proposes that you should be able to build just about any action into any blank if you know the right factors to manipulate. I am trying to develop some totally from scratch quadrate tapers now because I think that will better unlock those factors for me.   (Bob Maulucci)

    I've actually done both  on the same rod with good results. I start with 0.93 and progress down to 0.89. Distributing the difference over the entire length of the rod. This compensates for the tendency of quad to stiffen a bit in the butt. I converted a P. Young rod which needs to flex in the bottom. ( the rod is still too stiff in the butt). The next one I will try 0.83 in the butt. On Payne rods I do a straight conversion with 0.93. Per Brandin told me he just does a straight 7% conversion on all his rods.  (Joe Amaral)

    You can get a lot of opinion on this one. My .02 is that .93 will give you a softer feeling rod, although it will have plenty of power. In short rods ( less than 7 feet), I prefer .95, which will be a bit quick. I'm not so sure about larger tapers. I have heard arguments that the relationship is not linear, and that quads become stiffer is sections over .250 thick. I would stick with .95 until I prove otherwise. The actual conversion is .93, therefore my rods will be a bit heavier than the hex. If that bothers you, you can hollow build, and have both a lighter and more powerful rod. Stresses are low in a quad, so there is still good strength if you hollow build.  (Tom Smithwick)

Rule

One more thought: Anytime you convert a hex into a quad you will have a slower rod, regardless of 2 or 4 strip geometry.  (Bill Fink)

    The resulting action from hex to quad is dependent upon how the conversion was accomplished.  (Kyle Druey)

      Many of us would  really like to know how you convert a hex to quad while maintaining the same action. I've done a good bit of work in this area and found that if you use the straight geometric conversion from hex to quad the result is a nice but slow rod. I've tried using the same stress curve to convert and that doesn't seem to work either. I do make what I consider nice quads by copying the basic John Zimny stress curve which seems to be totally independent of any hex taper.  (Bill Fink)

        Disclaimer: I still haven't gotten my fanny in gear enough to finish all the honey-do's so I can start building...

        However, if you use the same length, same number of sections, same line weight, same heat tempering, etc., matching the stress curves exactly should give you a hex rod and a quad rod that cast exactly the same.  There are software programs that help with this. Wayne's original HEXROD and Frank's web-based one, (both on Rodmaker's site here)..  Any of these can be used to change a taper from 2-piece to 3-piece (or more) - change from a 6 wt to a 4 wt with the same characteristics - change from a hex to a quad or even penta.

        If I had the time, I'd love to take a nice taper, and make it as a hex rod from 3 weight to 7 weight with the same stress curve, and then make the same rods again as a set of quads with the same stress curve as the hex rods...

        <light bulb blinks on> Say - that could almost turn flyfishing into a sport similar to golf...  Can't you just see the Gillie telling the Sport: "well, you've got just a bit of a breeze to cast into, and you need to use a #14 brown drake here, so you probably should use the #5 instead of the #3 on this fish..."  (Claude Freaner)

          I have the time, and I have gotten my fanny in gear, and I've made a number of  Quads based on matching the stress curve EXACTLY . All my quads based either on taper conversions or stress curve conversions cast  slower than their hex counterpart. I invite you to try them. Or ask Tom Smithwick.

          I responded to your message because I hoped you had found a new and unique way of converting hex to quad.  Meanwhile I will continue to use the entirely new quad stress curve based on John Zimny's quad tapers. They make great rods.

          Incidentally I've found that using Garrison's stress curves makes a very nice penta. But not a Quad. What is it about Quads?  (Bill Fink)

      Conversion based upon equivalent radius of bending along the length of the rod, this is not the same idea as conversion based upon equivalent Garrison stress curves.  If the conversion is based upon equivalent cross-sectional area, which is what I think you mean by geometric conversion, the rod will indeed be different because you will have made the resulting quad much stiffer than the original hex.  (Kyle Druey)

Rule

Can anyone help me with the conversion factors for a 60.5 degree hex, a 92 degree quad and 73.5 degree penta???? Have found the conversion factors - thanks to the archives - for standard 60, 90 and 72 degree strips:

Hex: W = 2 x tan(60/2) = 2 x tan(30) = 1.155

Quad: W = 2 x tan(90/2) = 2 x tan(45) = 2

Penta: W = 2 x tan(72/2) = 2 x tan(36) = 1.453

Need math help with the 60.5. 73.5 and 92 degree triangles though.  (Rob Smith)

    It's the same formula.  If you have a trig calculator you could solve them yourself.  (Mine cost $10.95)

    That said, 60.5, W=1.166; 92, W=2.071; 72.5, W=1.493.  (Neil Savage)

    Yes, Neil, or any halfway decent scientific calculator, or even a plain calculator and a book of trigonometric tables.  (Peter McKean)

Rule

Does a Para 14 translate well to a Quad using the standard .93 conversion ratio.

Does anyone on the list have some comments on the resulting rods action and how it compares to its hex counterpart?  (Jon Babulic)

    If you do the conversion from hex to quad you will be keeping the relative stiffness the same.  The difference is that a quad comes out 2% lighter than a hex.  Can you feel that “in hand”  No.  Can you feel that while casting? Doubtful.  (Al Baldauski)

      I've done a good bit of trying to match hex performance in other geometries. Of one thing I'm certain: no straight conversion constant from hex to quad will give equivalent action. The quad will be slower.  To get a quick action quad I use a different stress curve. To get a nice penta action I use Garrison's stress curve. Go figure?

      By the way, my thanks go to Al and The Brit for a stimulating discussion. Even I could follow most of it.

      And regarding aging, it's a rear guard action. You must fight off every  limiting of your possibilities and keep going. I'm 82 and I chainsawed a half cord of wood this AM.  But man I'm tired. Brandy time.  (Bill Fink)

        Maybe one factor is that a quad will have more air resistance as you swing it.  I bet it would only take a slight difference in air resistance over the length of the rod to change the feel in the hand.  (Frank Stetzer, Hexrod, Taper Archive, Rodmakers Archive)

          And the quad will be one line size larger/faster if you do a straight across conversion (build the same taper hex to quad).  (Chris Obuchowski)

          Funny you should mention that.  On my first quad, I noticed while testing the guide spacing that the rod really whistled/hummed on the power stroke.  To compensate, I eased the corners a bit before finishing, which really helped with the noise.  (Paul Gruver)

        Interesting perspectives.  So what does constitute a good conversion candidate (from hex to quad) from a stress curve perspective?  I’ve found light tipped faster  actions  like  the Sir D, Cross sylph & PY midge to come out with a remarkable reserve of long range umph while still being delicate.  I built a strait conversion of an 8013 with a scarf joint ferule & it was almost too smooth and powerful for my taste, sort of boring.  Edwards tapers seem to have crazy complex compound stress curves but are thought to be some of smoothest actions.  Are there really any “rules of thumb” for good quad taper stress curves?  (Jon Babulic)

          My experience, after discussion with others who also build quite a few quads, is a graduated conversion figure of .93 for the tip end, progressing down to .88-.86 somewhere in the butt section.  How low & deep into the butt depends on personal taste/feel.

          Having said that, I much prefer to develop my own tapers for quads, rather than try to convert hexes (if you want the feel of the hex, build a hex).  (Chris Obuchowski)

      Can you feel that while casting? Doubtful.

      My comments on the conversion are based on math only.  As Tom and others have pointed out, it doesn’t seem to work with a straight conversion.  And based on the comments about wind resistance and noise generated during casting, that’s probably the reason.  The greater wind resistance of a quad will cause a greater load on the rod which is going to be most noticed in the butt.  Now there’s a design problem difficult to quantify with a formula!!  (Al Baldauski)

        Now there’s a design problem difficult to quantify with a formula!!

        There is a witches brew of variables when changing geometry. The ratio of power fiber to pith changes, the number and placement of glue lines changes, the impact of the curved face of the cane is different, and as has been pointed out, air resistance changes, and it's a bigger factor than we imagine. I have come to believe that it is impossible to make a quad or penta that feels exactly the same as a hex. Impossible is a big word to me, and not one I often use, but the reason I use it here is that the different cross sections react differently to load. So if you got a quad and a hex to feel the same with 10 feet of line out, they won't be the same with 40 feet.  The problem for someone starting out, is that there are so few published tapers for quad or penta, that it's tough to get the initial traction to get started. As Bill suggests, there is something non linear going on with the conversions, which is why I advise against taking on a compound taper, unless you have reason to think you know what you are doing. In addition, a conversion scheme that works with a mild progressive taper may not work so well with a fast taper, and I think that's why you get different theories from experienced people.  There are more people building quads and pentas now, and I hope we will one day have some sort of a data base of tapers to help everyone sort all this out.  (Tom Smithwick)

        Well, you could modify the quad by adding two extra strips, just to reduce wind resistance and noise.  (Steve Weiss)

          Now that’s a novel idea!  (Al Baldauski)

        I did some research into the coefficient of drag for a cylindrical section (hex rod close enough) versus a square section and was surprised to find a huge difference.  The square section has a Cd = 2.05 (flat facing forward) or Cd = 1.55 (corner facing forward, the cylinder Cd = 1.17.

        If you do the calculations assuming you move the rod at a velocity of 5 feet per second and the rod is an 8 foot straight taper 0.060 tip to 0.300 butt, then you get:

        For a quad the force due to wind resistance is 44 ounces distributed along the length

        For a hex (cylinder) the force is 25 ounces.

        That difference is certainly going to be felt!!!!!!! (Al Baldauski)

          I was under the impression that one of the advantages of using a quad profile was that for the same cross sectional area the D was less and you had more of the tension fibers farther away from the centerline. So you could use a smaller D and cross sectional area (weight) for the same stiffness. That would mean to compare Apples to (maybe) Apples the quad D should be less than the Hex D.

          Or is that just some delusional memory of mine.  I've read way too much about bamboo rods and some of the older stuff is being squeezed out of my brain by the newer factoids.  (Larry Swearingen)

            The moment of inertia for a quad of a given D is greater than the moment of inertia of a hex of the same D.  Therefore, you could decrease the D for a quad to get the same static stiffness.  As far as power fibers being farther away from the centerline, some are some aren’t.  If you have decreased the D to keep MOI the same then some fibers are closer to the centerline. Interestingly, the MOI of a quad is the same whether you orient the casting plane from flat to flat or from corner to corner.  That means its equally stiff either way.

            If you have followed this thread, though, some folks have brought up the issue of the quad being less stiff in the butt.  My calculations on wind resistance support this idea.  Dynamically, when you accelerate the rod to some speed, there is a significant additional load on the rod from the wind resistance.  This causes the rod to bend more that the hex and its felt in the butt section.  (Al Baldauski)

              Is it just me or is there a bit of a contradiction here,  some say quads are soft in the butt while others advocate using a conversion of .89  for  the  butt and .93 for the tip?  Wouldn’t this conversion make the butt softer still?  I’ve never tried the .89 but conversion,  rarely have I felt a cane rod and said that butts way to stiff, hex or quad.  (Jon Babulic)

                I’d agree with you.  Maybe the 0.89 was a typo and should have been 0.98.  That would make more sense.  (Al Baldauski)

                Is it just me or is there a bit of a contradiction here,

                Yes there is. I have heard others suggest the same type of decreasing dimensions in the butt. The rods I have built would not work that way, I don't know if it has something to do with the way I build a quad or something else. My suspicion is that the other guys started with steeper tapers, and that something non linear is at work here, but I have never been able to figure it out, and the other guys do build good rods. I'm usually starting with something like a Garrison taper.  (Tom Smithwick)

              My point was that you were making your wind resistance calculations base on the same D for the quad and a Hex.   If the MOI for a Quad is greater than the MOI for a Hex you should use a smaller D for the Quad on your calculations.  Still the difference would probably not be enough to make the calculations equal.

              I don't have any real experience with a quad other than casting one of Chris Raine's 8 1/2' for 5wts.   That rod made me consider giving up rodmaking.   Didn't think I'd ever be able to make one to compare to the sweetness of the action.  I'm still not sure I'll be able to.  {:>)  I might mention that I never heard any wind noise from dragging that square shape through the air or felt any "softness" in the rod's butt.  (Larry Swearingen)

                OK, if you reduce the Quad by 7-8% to compensate for the larger MOI then you are reducing the frontal area and therefore the drag by the same amount.  So correcting my numbers for the decreased quad dimension:

                Quad Drag force becomes 41 ounces

                Hex Drag force stays         25 ounces.

                Still a significant difference.

                And I don’t have any quad experience either.  I’m looking from strictly a mathematical point of view.

                When it comes to feel, everyone is different and in most cases it’s a difficult concept to quantify.  Even if you are doing a  side-by-side comparison.

                You like Chris Raine’s rods and maybe that “sweet action” is due to more flexing in the butt than you’re used but not so much more that it’s obvious.

                If Chris made one so can you.  (Al Baldauski)

                  I don't think your drag force calculations are correct, because of the assumption that the hex is a cylinder.  If you put a hex and quad in a wind tunnel with the casting planes (flats) into the wind, their basic shapes are very close.  The quad is all flat whereas the hex is a flat and two "wings" off to either side.  I have no idea what the Cd of such a section is, but I honestly can't see it being anywhere near as efficient as a cylinder.  My gut reaction is that the hex will come slightly better in the final calculation, but marginally so.

                  What you've done is given a strong argument in favor of cylindrical synthetic rods, which is amplified by their smaller frontal area.  Dang -- I hate when that happens.  (Rich Margiotta)

                    I'm with Rich on this. . .the few quads built straight across conversion from hexes become one line weight heavier and faster to boot.

                    Al, have you built quads before. . .or are your posts based on engineering formulas (I'm not challenging you, just asking).

                    As I said before, if you like the way a hex casts, build it as a hex.  If you want to take advantage of the inherent strengths of the quad configuration (lighter weight for equal stiffness) then develop quad tapers to suit the purpose (which is what I do).  (Chris Obuchowski)

                    I don't think your drag force calculations are correct, because of the assumption that the hex is a cylinder.  If you put a hex and quad in a wind

                    Rich,

                    The Cd for a quad with the flats forward is 2.05

                    The Cd for a quad with a corner forward is 1.55

                    The Cd for a cylinder                           is 1.17

                    From this progression you can guess that a hex section is going to be much closer to a cylinder than either quad.  Even if you assume the hex falls in the middle of the two at Cd = 1.36 the difference in drag force is still very significant.  41 ounces for a quad and 29 ounces for a hex.  (Al Baldauski)

                      The Cd's of shapes with a flat plane perpendicular to the wind are similar; see the figure here.

                      When casting, both the hex and the quad have a "plane in the wind".  The area of the flat plane is smaller for the hex, but then you've got those wings out to the side.  I couldn't find  the Cd for such a shape, but I believe it is very similar to the square shape, nowhere near the rounded shape of a cylinder.  (Rich Margiotta)

          Good work Al. If I remember correctly drag increases as the square of velocity. Try doubling the velocity. I think the drag will be increased by a factor of 4!  (Jerry Drake)

            Of course you’re right.  But we have to keep our thoughts close to reality.  I don’t have an easy way to measure arm/rod speed but I believe we won’t exceed 5 feet per second in normal casting.  If you invert your question: what if you decrease the speed to ½?  Then, maybe you’re in the range of casting a mid to slow action rod.  Under that situation your drag load would be ¼ that which I calculated.  Still significant.  (Al Baldauski)

    Does a Para 14 translate well to a Quad using the standard .93 conversion ratio.

    No, I think you will find the resulting rod too soft in the butt to perform well. In any parabolic, the relationship between the stiff middle, and the flexible butt is critical. My guess is that it would take you several tries to get it right. I would suggest you stick with progressive tapers. The way a quad casts, they tend to handle a lot of line in the air, and I don't really see the need for a parabolic quad. (That ought to get a few rises).  (Tom Smithwick)

      I cast a quad version of a  Perfectionist at CRR two years ago. I've been lusting after that rod ever since. Some day I will get around to building some quads.  (Mark Shamburg)

      I built a  Cattanach 8062  as a  quad this past year.  It's semi-parabolic.  I used the .93 conversion factor and I can attest to the fact that there is a lot of flex in the butt, which makes it roll cast well, but also makes it easy to overpower on an overhead cast.  I really do like the rod and it's action, but I have never cast a hex version, so I have nothing to compare to.

      My suggestion is to try it, you might like it.  A PMQ is so easy to make, you aren't out much if you find you don't like the action.   (Paul Gruver)

    To those who followed the thread on quad stiffness and conversion:

    I apologize for an error in my calculations on wind resistance. Rich questioned the Cd factor I used for a hex section and that got me looking at everything I calculated and found that the value I used in the Drag Force equation for air density was in the wrong units.  The effect, when changed to the correct value, is to reduce the drag force by a factor of 32.

    Drag force on quad was 41 ounces.  Should be 1.28 ounces

    Drag force on hex was   25 ounces.  Should be 0.78 ounces.

    While these results still fall in line with the original reasoning, the new values of drag force are so small they probably are unnoticeable.   This is based on a rod speed of  5 feet per second.  If you double the speed, the drag force is increase by 4 times.  That might be noticeable.

    I tried to measure hand speed while casting last night and came up with an absolute max value of 25 feet per second.  A speed I certainly would not attain in normal casting conditions.  Maybe half that. (Al Baldauski)

      Thanks for clarifying that.  It corresponds more closely to actual experience, IE, there is more drag, but it's not significant.  It also partially explains why a quad sometimes twists a bit during the power stroke.

      As a side note, this seems to suggest that, on a quad, it would make sense to mount the guides on the apex, not on the flat.  (Paul Gruver)

        Yeah, guides on the corners would probably be indistinguishable from a hex unless the rod speed is VERY high.

        But there's another consideration: a lot of people cast with some rotation as well as translation. So the tip speed is going to be higher and that aggravates the drag problem.  (Al Baldauski)

Rule

What is the formula for converting a hex taper to a Penta taper? I know RodDNA does it but I wanted to cross check it.  (Olaf Borge)

    The formula for equivalent area is to multiply the hex * 1.09.  The formula for equivalent Moment of Inertia is to multiply the hex * 1.05.

    The strip height is then .447 * the Penta's flat to apex measurement.

    I have a spreadsheet (contact me to get a copy) that you are welcome to use that converts hex measurements to Penta or quad using different user defined variables.  It was made so that I could quickly put in a hex value and get out MHM settings, though those are optional as the taper numbers are also generated.

    For it to run properly, you need to have Excel and the following Add-Ins installed:  Install Analysis ToolPak, Analysis Toolpak - VBA, Conditional Sum Wizard and Solver Add-In.  (Chris Carlin)

Rule

Well here is an interesting question that is looking for some answers. I have used Al Baldauski's software program to convert an Edwards 7' from a Quad to a Hex (this is the Edwards #25 7' 4 wt quad rod taper). The conversion went fine based on matching rod deflection contours. Now, the issue is that the original Edwards Quad has lots of discontinuities (hitches or jumps) in the stress curves that I do not like and of course they convert to the Hex geometry. So here is the question - How much liberty should/does one take in "smoothing out the stress curve" to create a modified Hex rod.

One response could be, well don't do anything to the conversion and keep the discontinuities. Well if you know me as a mechanical engineer, discontinuities in stress curves create local high stress area concentrations along a taper - which suggest to me a higher probability of rod failure or fracture. So I have done an eye ball smoothing of the converted Hex stress curve to a smoother one that seems to be OK. I can then compare the various stress curves and decide how I like them. How valid is this approach?

I could of course first smooth the original Edwards Quad taper data and then convert the rod to a Hex putting the smoothing into the Quad before changing it to a Hex. Would this be a better approach and why? Whatever happens here, I will make the rod with one of these approaches to create a Hex version of the Edwards rod and see if I like it. 

What are folks thoughts since I don't think there is a correct answer, just lots of questions and decisions. (Frank Paul)

    Stress curves are very deceiving, they tend to magnify the slightest bump in the actual shape of the rod, look at the TAPER graph first and see if it makes any more sense to you than the stress curve shows. A lot of times what might show as a large blip in the stress curve might only be a very small increase or decrease in diameter that could be due to any number of things, uneven varnish, a heavy hand on the file, what have you.  (John Channer)

    Well, obviously, the ultimate solution is to make two rods - one with the discontinuities and bumps, and one with them smoothed out.  Then take 'em both out and cast them.  Then you can decide which one is better.  Sometimes them old guys designed those "discontinuities" in there for a reason.  (Mark Wendt)

    I agree with your thoughts on smoothing stress discontinuities.

    My opinion is this:

    You can smooth the quad dimensions then convert to hex or vice versa and there will be little difference.  The hex conversion will be carrying a little more weight because the hex cross section is not quite as stiff as the equivalent quad so the conversion fattens it up to compensate.  That extra weight will result in a slightly slower feel.  The weight difference is only 1.98 percent.

    If you look at the deflection of the original quad then make a smoothed version of it and compare the deflection of the smoothed version, it’s less than 1% difference.  So the smoothing shouldn’t make a big change in feel.  I did this exercise so if you want the numbers to look at I’ll send them to you.  (Al Baldauski)

    Conversely....How hard is it to convert a hex to a quad?  (Tom Key)

      Not hard but you have to decide on how you want to convert.  If you think keeping deflection constant then you need to keep the Area Moment of Inertia the same at each station.  If you want to keep the stress curve the same (and I don’t agree with this approach) then you convert keeping the Section Modulus the same at each station.

      To make it simple:

      Deflection the same: multiply each hex station dimension by 0.922 to get a quad dimension

      Stress curve the same:  multiply each hex station dimension by 0.897 to get a quad dimension

      If you use the deflection approach you’ll get a stiffer rod than the stress approach.  (Al Baldauski)

        Of course, there are other options.  Are you building a 4-strip quad, or a 2-strip quad?  

        For 2-strip quads (PMQ's) you can maintain the same physical mass as the hex version by using a .93 conversion factor at each planing station.  This will have a softer feel than the hex version, though, so many of us use a .95 conversion factor for a slightly firmer action.

        The main point here, though, is that you cannot build a hex and a quad version of the same rod that will have the same feel and same casting characteristics.  The geometry of the rod prevents that.  (Paul Gruver)

          I probably was not clear; I am creating a HEX taper based on a QUAD taper, so the rod to be built will be a HEX conversion.  I am matching the rod deflection curves in the conversion, so the rod weights are almost the same - within hundredths of ounces.  I am sure there is some difference in weight distribution because of section geometry. I am not sure about "feel differences" unless one builds both a QUAD and HEX and tries to make a qualitative comparison - I think that will be good, but I find it difficult to do.  (Frank Paul)

            Since all the quads I have done have been conversions from a standard hex taper, it never occurred to me that there were any "native" quad tapers.  Anyway, it's just the reverse of the process.  Take the finished flat to flat measurement, and divide by .93 for the same physical mass, or by .95 for a slightly lighter rod that is closer to the feel of the quad original.

            The same holds true for Al's engineering examples given below.  The factor's are the same, you just divide instead of multiplying.  The hex will always be slightly larger, flat to flat, simply because a hexagon has less area than a square.  (Paul Gruver)

            I am not sure about "feel differences" unless one builds both a QUAD and HEX and tries to make a qualitative comparison

            A lot of people have done this. You will find a feel difference between the geometries, it's just there no matter what you do as far as I can tell. Al's formula will get the rods in the same range as far as casting the line, but I think you will notice a difference, specifically, I would be surprised if the hex did not have a stiffer feel than the quad. There are still a couple worms left in the can, too. There are two less glue lines in the quad. What does that do? The flats on the quad are wider, and have more curvature, especially in the butt. If you follow the curvature when scraping, etc, how much difference does that make?

            I'm with you as far as smoothing the taper. I would approach it the other way, however, and smooth out the taper chart into a fair curve. A funny thing will happen, the stress chart will smooth out, too. Try it both ways, I'll bet the resulting tapers are pretty close.  (Tom Smithwick)

              I would argue that the quad to hex conversion will feel less stiff for two reasons:

              1.  To get the equivalent stiffness in a hex the cross section it will have 2% greater weight.  Accelerating a greater mass with the same stiffness rod will make it feel softer.

              2.  The larger diameter of the hex rod mean there will be proportionally more pith in the butt strips, making it less stiff.

              I grant that these differences are small but together MAY result in a softer/slower feel.  I suspect there is more variation in bamboo and in dimensional tolerances that override the above.   (Al Baldauski)

                I would counter argue that the power fibers in the hex rod are further away from the centerline, and that this factor is the over riding factor. I have done this experiment a couple times, and cast rods made by several other people,  including John Long's Grand Experiment, which is in "The Best of the Planing Form vol. 2". He concluded that quads, built on an = area conversion (.93),  felt softer. However, I remember them as fine casters, as long as the person could deal with the deeper bend. It will be interesting to see what Frank comes up with.   (Tom Smithwick)

                  Did you mean area conversion or diameter conversion??

                  MY factor of 0.922 is base on the diameter not the area.

                  If you convert a 0.200 section from quad to hex using my factor of .922 you get an equivalent hex diameter of 0.217

                  If you convert from quad to hex based on area using the same factor you get 0.215.  If you use an area conversion it will be softer yet.

                  And you have to make sure the conversion was done right.

                  As Paul pointed out earlier:

                  Hex to quad MULTIPLY the diameter (or area if that's what you want) by 0.922. Quad to hex DIVIDE the diameter by 0.922

                  My factor was derived by maintaining the Area Moment of Inertia the same at each station (to maintain deflection the same).  This takes into account the greater distance from the neutral axis.  It doesn't account for the greater amount of pith in the butt sections, though.  (Al Baldauski)

                    Did you mean area conversion or diameter conversion??

                    I was talking about area of the quad and hex remaining equal. I see what you are doing, but still predict Frank's hex rod will feel a bit stiffer. It may be some sort of a perception thing based on recovery rate, or something else, but quads and hexes do feel different. That's based on a lot of rods by different builders.  (Tom Smithwick)

                      Perception is a funny thing and I don't doubt that there are those who have built a hex and then quad equivalent (and vice versa) and believe they are different.  Even if they were built from the same culm there is enough difference among the bamboo strips to give two different feeling rods.  Even if they were both hex.

                      The math says the quad should be stiffer so far as I've carried it and I'll be the first to say I may have missed something.  I'm looking for enlightenment.  (Al Baldauski)

                Well I did both conversion approaches and the HEXes come out at a few taper locations within 1 to 3 thousands of each other - some are the same. The compared deflection curves differ by only 1% - very small. Anyway, the resulting HEX taper stress contour is much different than I normally try to create, so I am going to make the rod and see how it casts and performs - sometimes one is surprised. I also did a modification of the original converted taper as I normally like the stress curves to look, so I will make that one as well and have two different rods to play with - of course, if they turn out poorly, there is always plant stakes in the yard.  (Frank Paul)

Rule

Site Design by: Talsma Web Creations

Tips Home - What's New - Tips - Articles - Tutorials - Contraptions - Contributors - Search Site - Contact Us - Taper Archives
Christmas Missives - Chat Room - Photo Galleries - Line Conversions - The Journey - Extreme Rodmaking - Rodmaker's Pictures - Donate - Store