Bamboo Tips - Tips Area
Form Setup - Glue Compensation

< Home < Tips Area < Form Setup < Glue Compensation


To compensate or not to compensate for glue,  that is the question.  I recently did a very poor glue job on a rod by forgetting to use the proper weight on my binder. Rookie move I know but hey, I'm a rookie. Anyway, the rod wasn't bound very tight and I didn't notice until it was too late. Needless to say I have all sorts of problems, the greatest of which is too much glue. How much is too much, well if the typical glue line is .0005 to .001,  I probably have .002 glue lines. Per Ray Gould's assertion, that should mean I my butt is .01 oversized and I'm finding that, that is true in some places.

So here's my current boggle. Before glue up my splines were dead on and if I had glued the rod in my normal manner, things would be fine. So my question is this. What's more important, the size of each individual spline and fiber content (distance of each spline flat to it's apex) or the distance between the outer most fibers along each plane? If I have a particular action and line weight in mind. Will I still get what I'm looking for or do I need to scrape the rod down and possibly remove too many power fibers?

I'm not looking to duplicate an existing rod here, so trying to get the proper flat to flat dimensions to duplicate an existing rod is not the issue. The issue is really, I wanted X amount of cane to get a particular action. I've got x amount of cane but now the flats are .01 wider than planned. I don't think it's a simple issue of simply being a few thousandths off as discussed previously on list and tips site. At least I don't think it is.  (Jim Lowe)

    Sometimes the action you seek will not turn out just right even if your measurements are dead on  and your glue joints are minimal. There are many other variables at work that can affect action as well as line weight. Put the ferrules on and temporarily rig up the grip and guides and test cast before finishing the rod. If you like it, you will be way ahead of starting over without testing the blank.

    You can also scrape or sand a couple of thousandths off and still have plenty of power fibers to spare.

    The issue is really, I wanted X amount of cane to get a particular action. I've got x amount of cane but now the flats are .01 wider than planned. I don't think it's a simple issue of simply being a few thousandths off as discussed previously on list and tips site. At least I don't think it is.  (Steve Weiss)

    You not going to see much of a difference being off .001, Finish out the rod as is.   (Tim Preusch)

      Thanks, but I'm going to be off .01 and not .001, which as I understand it is an entire rod size. On further reflection, I think I'm simply going to throw on a ferrule, cast the rod as is, then shave it down to the correct size, cast it again and then simply forget the whole thing and make a new butt using the correct binder tension.  :)   (Jim Lowe)

    Take my advice, I am not using it anyway.   Just finish the rod and do not worry about the amount of glue. Power fibers farthest from the center have a bigger impact on the rod then does the amount of glue. Glue can and does shrink over time, if you have glue gaps then the cosmetics will suffer but it will still cast fine. If you are able to hit the dimensions you want on cane then you are fine. Different glues will have some thicker, some thinner glue lines.

    Some makers are determined to hit the same numbers as other makers. Well, if you do not know what glue they used and how thick the glue line is you really can not determine the exact dimensions, you can only get close. The fact is rods react to their environment and are constantly changing dimensions. So the numbers often change depending where the rod is and  where it has been also how old it is and what kind  of glue was used.

    Finish it up as normal. And in time you will not even know.  (Adam Vigil)


My planing forms are ready to go (fixed the problem of the station that was to deep - I took them back apart and found a small burr of metal  that was  preventing the  forms from  fully closing at the 8.5 station), so I'm ready to put a final taper on my first rod.

I've read several books and lots of  the  tips  on and just have a couple of questions.

1.  I'm making the PHY Midge #2 from the Hexrod archive.  The action length is 65" and the rod length is 75".  The taper chart stops at the 65" mark (0.237).  What do I set the forms at for the 70" and 75" stations?  I'm assuming that I keep it at 0.237.

2.  My sections have been rough planed, bound, and heat treated and have been hanging in a warm dry closet.  Should I wait until I have time to take all of the strips to final taper before I begin or can I cut the binding thread and work on them slowly over the next few nights. 

3.  Should I set my forms at the exact 1/2 measurement from the taper?  I think I read in Wayne's book that you should add .001 to account for sanding after the glue dries.  If that's the case do you add .001 to the 1/2 measurement?  For example if the 10 station call for a rod  flat to flat measurement of .088, should I set my forms for .044, .045, or .0455.  I already removed most of the enamel with a razor blade so the strips would set neatly in the rough planing forms.  (Aaron Gaffney)

    Congratulations on finishing your forms!

    1. You're right, set the other stations for the same dimensions, those stations are under the grip and reel seat and don't need a taper.

    2. Work on them as you have time, they will adjust to the humidity level there eventually anyway, no matter what you do. If it makes you feel better, keep them in the closet when you aren't actually working with them.

    3. Set the stations for the half dimension minus whatever you think you should allow for glue. In my own experience, Urac will make the sections .006-.010 fatter than they should be, Epon, which  I now  use, adds  about .004,  so I  set my  forms to half -.002. I also have 1 1/2" from the end of the form to the first station and finish planing with the strip even with the end, so I have 1 1/2" leeway on the tip end and I make the strips long enough to have 3" leeway at the butt end. This way, when I'm ready to cut the sections to final length, I can shift the cut points to match the intended taper.  (John Channer)

    The only thing I could add to John Channer's advice is to take some of the cutoff pieces you have, measure them carefully and glue them together with your adhesive of choice, measure the difference and adjust the forms to compensate accordingly.

    Ray Gould states in his book that a .001" glue thickness adds .005" to the taper. That's about a line weight. Which brings up a point I've often trough of but never ask the list. It's not uncommon to see a taper that says "deduct .004" for varnish" but never says anything about glue thickness. Two questions: Do most of you compensate for glue thickness under these circumstances? Should we compensate for glue on all tapers?  (Don Schneider)

      I deduct .002 from my form settings when I am gluing with Epon, that is - .004 total.  None when I use Titebond ll.

      It works pretty well for me.  (Tony Spezio)

        Have you had any long term problems with Titebond II?  I have been pretty happy with it, but I have only been building rods for a couple of years.    (Tom Mohr)

    Most folks will keep a constant measurement under the handle (if the last measurement is .237 @65",  70" and 75" will also be .237).

    Since the cane under the handle and reel seat aren't flexing with rod loading, there is no reason to change the diameter.  (Chris Obuchowski)


It's not  uncommon to see a taper that says "deduct .004" for varnish" but never says  anything about glue thickness.

Two questions:

Do most of you compensate for glue thickness under these circumstances?  Should we compensate for glue on all tapers?  (Don Schneider)

    I might be wrong but I think when they say deduct for varnish.  They measure a vintage rod with varnish on it to get the taper.  So I think that you also deduct for glue??? Anybody? Am I right?  (Bill Tagye)

      After all the hand wringing over fractions of a thousandth is over, we guess the varnish thickness and calculate the amount undersize using our own personally derived formulae, which considers the type of original glue, the type of glue we use, along with a correction factor for the relative "stiffness" of the glue we use! This is such a precise science!

      I tossed out the "measure it with a micrometer" comment last week. This is exactly why I feel that quote is so apropos.  (Larry Blan)

      I agree, you also deduct the glue thickness for whatever glue you are using.

      This didn't occur to me on the first rods I built and was always wondering why they needed 1 more line weight. One day the light went on, Glue Thickness was adding to "d".

      No one ever said, "Don't forget to deduct for glue thickness." It would be nice to have a chart that says how much each adhesive adds to "d". Did the math and didn't believe the answer until I made a scale drawing of a cross section.

      How many times have you heard someone say "I built that rod but it performed better with 1 more line weight". Could not considering glue thickness have been the cause ? Think so.  (Don Schneider)

      I hear what you're saying Bill.  My only problem is that when a rod is finished, on or off the numbers, three different people casting it may like it with three different line weights.  I'm not sure where that leaves us.  Now where's my crayon?  (Ed Berg)

      Okay I'm new to this, but the whole glue line thing got me thinking about the math involved.  It turned out to be a great problem for my high school advanced algebra math class.

      In THEORY 2 times the glue line should be added to the overall dimension of the rod.

      In other words if the dimension of a rod should be .200 (flat to flat) a glue line between the strips of .001 would add .002 to the overall flat to flat dimension increasing it to .202.  A glue line of .002 would add .004 and so on.

      In calculating this I assumed that 1/2 of each glue line belonged to each strip.  What I mean is that if the glue line between strips 1 and 2 is .001, then 1/2 of that or .0005 belongs to strip 1 and .0005 belongs to strip 2.  (Aaron Gaffney)

        Okay, sounds reasonable, but when gluing a porous surface such as bamboo or wood, doesn't the glue penetrate into the pores so the glued surfaces can be right next to each other - essentially no glue lines? I can see where if you were gluing metal to metal there would need to be a glue line since  the glue cannot penetrate into the metal.  (Darryl Hayashida)

    I'm not the sharpest tool in the drawer, but unless you're all are building PMQ's and not hex rods, you have six glue lines to take into consideration. All adding to the extra dimension of the section.

    As someone mentioned you would have to bind dry, measure, unbind, glue and measure again to get to exact specs and then adjust you strip size.  (Pete Van Schaack)

      One would think there were 6 lines, but if you look at a drawing or cut off of a section, there is only one line that affects the flat to flat measurement and it runs from point to point thru the middle of the rod, the caliper jaws are parallel to it when you measure. The other lines don't come into play.  (John Channer)

        I have to disagree. All the glue lines make the section bigger. Thus, the flat to flat measurement would be greater. It would be very miniscule, but never the less the flat to flat would be affected.  (Lowell Davis)

        Get out your Trivial Pursuit game with it's slices of "cheese" make your hex and measure flat to flat. Then slip a piece of paper between each piece and measure again, the entire circumference has expanded.  (Pete Van Schaack)

        Twice the glue line thickness be added to the flat-to-flat measurement?  There is only one glue line between flats, the other two glue lines are diagonally across the one that's between the flats and shouldn't affect the flat-to-flat  measurement.   Of  course,  how do you measure the glue line?   (Henry Mitchell)

          I think the diagonal glue lines  also add to the flat to flat  measurement.  Draw the six triangles with space between  them for the glue.  Better still, check Ray Gould's book, he does the geometry(?)..  (Roland Cote)

          Bind the strips dry, measure flat-to-flat,  then unbind, glue, let the glue set up and remeasure I guess.  (Neil Savage)

            Won't there be some uncertainty introduced when you sand off the dried glue? How will you know that you stopped just at the glue and didn't sand off just a little bamboo?  (Darryl Hayashida)

              In my limited experience, it takes quite a bit of sanding to reduce the measurement by .001".  (Neil Savage)

                Well, I hesitate to add my 2 cents but here goes.  I would think that when one talks about the diameter of a rod and its subsequent casting characteristics one would be talking about those characteristics due to the power fibers of the bamboo itself.  For example if I have strips that are .035 each with a diameter of .070 and I plot that one a stress  curve, the curve is based on the structural characteristics of the bamboo not the structural characteristics of the glue. 

                If I account for a glue line of .002  and reduce the bamboo by that much haven't I changed the rod characteristics significantly.  If I now plot the curve using .002 less diameter I would think that would the curve would look quite different.  All in all, it seems the overall diameter of the rod should take into account just the bamboo strip not the glue and assume that the glue adds nothing to the casting characteristics (other than holding everything together).  Reducing the diameter (to meet the specs) by reducing the bamboo would seem to weaken the rod.  It seems the goal should be to reduce the glue line and keep the cane to maintain a certain diameter rather than sanding or scraping off the bamboo.  (Tom Mohr)

                If you look at the problem from a beam theory point of view, you are changing the characteristics.  Take a simple I beam for instance.  You have two flanges, held a certain distance apart by the web.  For a given thickness of the flange(s), the wider the flanges are held apart by the web, the stiffer the beam will be, up to the point you've exceeded the stress the material can take before failure.  The web, in beam theory is there only to hold the flanges apart, even though it does contribute to the properties of the beam.  A beam, when flexed, will show compression on concave side, and tension on the convex side.  The web, as it approaches the center shows no tension or compression, much the same as our theoretical cane rod, with the flanges being the flats, and the web being the pith.  The further apart we move our flats, the stiffer the rod is going to be.  We've actually increased the cross section, due to glue.  The glue itself will contribute some, but not a lot to the stiffness of the rod, mainly because most of the glue is not in the planes of tension or compression when the rod is flexed.  What is making the difference is the beam has gotten larger, or the flanges moved further from the center.

                Getting back to the thickness of the glue line, I think somebody earlier today said that Epon gave a nominal glue line of about .004" and the polyurethanes about .006".  On a tip section of a 3 wt rod, lets say a flat to flat dimension of .060", .006" is 10% of the cross section.  I think Ray Gould uses roughly .006" increases in the stations to either go up or down a line wt.

                I'm not 100% sure if the software we use to evaluate tapers takes the glue into account as part of the composite matrix.  The stresses for a given cross section of bamboo are somewhat arbitrary anyway, since Mother Nature declines to make our Tonkin cane with any kind of consistency in the depth of the power fibers.  Larry Tusoni, Wayne Cattanach, and John Bokstrom, if you're out there, do your software programs take the characteristics of the laminating adhesive into consideration when calculating the stress curves?  And if you are taking glue into account, how are you taking it into account?  (Mark Wendt)

                  Uh huh



                  But if the glue contributes nothing, why does a rod glued with glue A feel stiffer than a rod glued with glue B?

                  I have a friend who will remain nameless who only halfway jokes that the bamboo is a matrix for holding the glue!  (Larry Blan)

                    Ah, but I didn't say the glue contributes nothing, I said it contributed some.  Other than a few empirical tests where we've made up identical rods using different glues, we've never done any "scientific" testing to determine how much the glue actually contributes to the statics and  strengths  of the rod.   As to the A-B difference, it may be the glue, or it may be the cane.  Do we honestly know or have proved to the gnat's arse either way?  (Mark Wendt)

                  Let me try again..

                  Assuming the rod is a hex, aren't there 12 surfaces with glue on them between the two being measured, and therefore the distance we measure is the partial summation all of the glue joints.  How do you describe how much to take off each stick to account for someone else’s glue.

                  Why would I do this?  (Jerry Foster)

                    Exactly, there are no parallel glue lines between flats.  Each of the six glue lines contributes a little bit to the added flat to flat dimension of the rod.  The math works out in such a way that in theory the flat to flat dimension grows by twice the thickness of the glue between any two strips.

                    How big that glue line is certainly depends on the type of glue, how tight the strips are bound, and probably lots of other variables.

                    Since I've never even glued a rod up before (the first one will be this weekend) I really don't have anything other than theory to contribute.

                    Like someone else said if I consistently find out that I'm .002 over the dimension I sent my forms for after the glue dries I'll learn to set my forms accordingly.  I was just trying to come up with a ballpark figure to begin with for my first set up.   (Aaron Gaffney)

                  Don't y'all just sand the rod down to the right dimensions?

                  Oops..this might of been someone’s trade secret, forget I said anything!  (Jeff Fultz)

                    The guess in glue line thickness is only for the purpose of making a new rod that will come out for size like the rod that was measured.  (Ray Gould)

                  Yes, that was my point..long way 'round.

                  1. I don't think most of us understand what we are measuring in terms of glue joint thickness ergo...correcting for it is another wild card.

                  2. Precisely .. the outside dimensions are all that's important if rod cloning a specific rod..

                  The only accurate way is if you know the original makers intent.  (Jerry Foster)

                    Yes, in a hex rod, you do have 12 surfaces with glue on them, but once bound, you in effect have only six glue "lines."  Assuming you have some kind of constant pressure squeezing the strips together, you should end up with a uniform glue thickness for all six lines.

                    That's the catch isn't it?  You'd not only need to know what kind of glue they were using, but you'd also sorta need to know if the binding was uniform for the length of the section.  Short of destructive testing, there really isn't much of a way to know this.

                    So....  You have to SWAG it.  (Mark Wendt)

              I must say, Darryl, that I am absolutely certain that I don't stop at the glue/enamel interface, as I don't even plan to do so.

              I take off the glue with the bodied scraper, and at the same time I clean up any remaining string marks and tidy any bumps or  wobbles in the surface.

              Admittedly, I have the L-N scraper set pretty fine for this job, but nevertheless I do want to see clear and clean cane all the way before I stop.  I used to sand the glue off, but now I am convinced that I get a better surface with a VERY sharp scraper at about 95 degrees and almost no pressure other than the weight of the tool, and the edges are sharp as well, and don't tend to get a bit "blurred" as it does with sandpaper.

              I certainly can't claim that they mike out exactly correct, but there is only ever a thousandth or so in it, and the rods perform very predictably, which is, I guess, what I want.  (Peter McKean)

              I would think that this would be easily solved. If you  notice your blanks are .002 bigger on a regular basis using your choice of glue. I'm sure after making a couple rods you'll say to yourself, "SELF" I better  adjust my forms to compensate!   (Bill Tagye)

                What I would really tell myself is "Self, aren't we getting a little nit picky here trying to account for what role a glue line plays in the casting characteristics of a rod? There are many other things that will make a bigger difference."

                While it is good to try and keep the factors that change the casting of a rod under control, have you ever tried casting two identical rods, one 0.002 inches larger than the other? I could guarantee no one would be able to tell the difference.  (Darryl Hayashida)

                  I also agree with Darryl although I always get a kick out of the banter on this forum and it does give you something to think about..... but not too much.  Geez, if my rods turn hexagonal I'm pretty happy.  In fact when I was read Engle's book on Splitting Cane I got a kick out of all the builders who started with some classic tapers by Payne or Young and then "tweaked" them. I have always wondered if they were "tweaked" by accident or "tweaked" on purpose.  Heck, when I miss the mark on tapers I just tell everyone that I just "tweaked" the taper a bit to fit my casting style.  (Tom Mohr)

                    One way to measure a glue line is to take into account the enamel that is removed after the glue dries.  Of course you will have to consider the enamel on both sides of your hex.  (Doug Alexander)

    I just always took the approach that if the rod would cast to the fish, not break, no visible glue lines, then the rod is OK. We are making fishing poles here, sometimes things get too complicated with stress curves, stress lines, and  crows feet around the eyes!  (Jerry Andrews)

      It's the Engineers mentality, they seek perfection. Part of the fun for them is the math and theories. Unfortunately, mother nature cannot be engineered. She always throws a curve ball that they can't hit.   (Bill Tagye)

        I don't think they seek perfection at all. They simply understand the math and theories, and want to try to apply them to gain an understanding of what is happening with the taper. At the end of the day, all of it is condensed in some fashion, and applied for general use, to everyone's benefit.  (Larry Blan)

        On the other hand, if we are trying to make a rod that matches a classic taper, maybe it's possible that we do need to take the thickness and type of glue into consideration if we want that rod to cast the same as the original taper.  We do take into account the varnish thickness, and sometimes intermediate wraps.  Each of those things by and of themselves probably don't matter or contribute a whole lot, but taken together, it's quite possible they do affect things.  (Mark Wendt)


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