I have been analyzing some guide spacing charts and have discovered an interesting fact. Here is an example of a published spacing for a 9' fly rod....

4.5 9.5 15.5 22 29 36.5 44.5 54 65 76

Distance between guides...

4.5 5 6 6.5 7 7.5 8 9.5 11 11


Increase of distance...

.5 1 .5 .5 .5 .5 1.5 1.5 0


Now, knowing that a bent rod describes a uniform curve, these non-uniform spacings seem odd to me. Wouldn't these spacings be better?

4.5 9.5 15 21.25 28.25 36 44.75 54.5 65.25 77

Distance between....

5 5.5 6.25 7 7.75 8.75 9.75 10.75 11.75

Increase...


.5 .5 .75 .75 .75 1 1 1 1


This represents a more uniform increase, without rapid increases or decreases: 2 @ .5, 3 @.75, 4 @ 1


Crudely graphed, the increases look like this...


* *
* *
* * *
* * *
Original * * * * * * * *
* * * * * * * *
----------------*


* * * *
* * * * * * *
New * * * * * * * * *
* * * * * * * * *
-----------------


I think that this is a better plan for initial (static) guide placement than a non-uniform-increase. I've checked a number of published spacings with this method, and most have similar errors. What do you think?

Not all rods flex the same on the way to that uniform curve. The article on static placement in the library shows you how to locate guides in three stages. It's better than any chart, really.

But there is no reason you couldn't use your new chart or system for initial spacing. Maybe others will as well and then get back to you and let you know what they think.

..........

I am not sure what you mean by uniform. I frequently have the first 4 guides equidistant, or with only 1/4 inch increase from the preceeding guide. The reason is because the rod bends progressively. As I work down, I will add an extra inch or two between guides, more for the last 3. Then adjust for a good static distribution.

Oops! The graphs sure didn't come out right. If you really want to see them, correctly, I'll try again using spaces instead of tabs....

* *
* *
* * *
* * *
Original * * * * * * * *
* * * * * * * *
----------------*


* * * *
* * * * * * *
New * * * * * * * * *
* * * * * * * * *
-----------------


The point I'm trying to make is that no rod that I want to fish with (at least these days of self ferrules) increases the bend from the tip, then flattens out, then increases again, then flattens out again, then...

Don, your reply illustrates what I'm thinking. The distance between guides should increase as you move down from the tip,as yours do. The sample chart shows an increase, then a lesser increase, then a huge increase, then no increase.

Remember guys, this is theoretical; it just makes more sense mathematically. Anyway, it satisifes my Virgo soul.

Oops again with the charts. I guess the software removes whitespace. 8^(

Jim. I re-read the factory spacing. That is interesting. Your proposal is pretty close to my practice, except for the last couple of guides, where I may add a couple of extra inches between. Also, I put guides on the ferrules first, and then distribute the guides in between the known guides with the gradual progression you describe. From the tip the increase might be slight: 4.5, 4.75, 5.0, 5.5. It might then be 6.0, 7.0, 8.0, 10.0, 13.0. You will find your proposal is a pretty good start for the static distribution test.

Keep in mind that most of the spacing charts are based on charts that were developed and used by Fenwick over 25 years ago now.

A few companies have devised their own rod guide spacings, but many still use the same generic charts that have been in existence for decades now. This is just one reason that most serious custom rod builders don't use them for anything more than to arrive at a starting place.

.....

Jim,
I think that you are trading off several different things when selecting the number and positioning of guides. The first is what I would call line line control, the second is stress distribution and the third is getting as high a performance out of the rod as possible.
In my judgement there are a number of methods that have been used to position guides, all of which have advantages and disadvantages in one or more of these areas but I feel that the static method in the Library is the best that I have seen because it tends to position the guides on the blank consistant with the blanks natural curve and every blank will be a little different. This means that using this system you will be putting the guides on the blank in positions that will end up with the best stress distribution, if you position guides so that the rod is forced to flex on other than its natural curve you will end up with higher stress in one or more spots on the rod. And I would argue even more importantly it also tends to minumize the number of guides that you use. More guides than are necessary just means more weight and the additional weight has a direct and significant affect on lowering the rods performance, casting distance, ease of casting, even the sensitivity of the rod.
Also if you look closely I think that you will find that almost all blanks have a flat area at the tip and as a result most charts or formulas for guide positioning tend to put the first couple of guides at the tip of the rod too close together. And this is the point on the rod where the added weight of the guides will have the most affect on lowering the performance.

If you noticed a few days ago Tom mentioned that there is going to be an article in the next RodMaker on this very topic.

Hi folks.

I couldn't help but get interested in this topic of spacing. I agree with Jim Benenson's assertion that if we assume natural curvature to be mathematically smooth and monotonically increasing in first derivative (in absolute value), then the spacings should increase not a fixed amount between guides, but a varied amount that continues to increase. But I'd argue further that as a first guess, the increases are better approximated by geometric fraction. In other words, the next guide should have a separation defined by a fixed percentage increase and not some incremental increase of 2 guides @0.5 inches increasing differential, [email protected], and 4@1 inch increases. Instead, the increase would be different for each and increase at a normal rate, e.g. [email protected], [email protected], [email protected], [email protected], [email protected], etc...

Writing some simple software to compute this, I can take an end point, say 76 inches as the fixed position of the last (stripper) guide, and put in a geometric taper of 1.11 (11% increase each step) and assume 10 guides. Computing, I get [1]4.54, [2]9.59, [3]15.19, [4]21.41, [5]28.3, [6]35.96, [7]44.46, [8]53.9, [9]64.37, [10]76.0 as the guide spacing to 2 decimal places. This would be a starting point, after which I'd have to judge whether the tip top area of the blank was nearly solid graphite and thus stiffer so I'd move the first guide further toward the butt by a quarter inch or so. And of course, I'd then check the ferrule area and account for stiffness there. If the taper of the fly blank were slightly faster, I might start with a 12% increase, which would yield: [1]4.33, [2]9.18, [3]14.61, [4]20.7, [5]27.51, [6]35.15, [7]43.69, [8]53.27, [9]63.99, [10]76.0.

If anyone wants to use the software, I've written a small Java applet to compute it. It runs directly in any Java-enabled web browser without need to download or install anything else if you already have the Java plug-in. Just go to: www.gyozadude.com and select the Guide spacing applet.

No guarantees obviously on the exactness of spacing for any blank but it is an attempt at a first guess. It does take some familiarity with the tool to guess what the geometric taper should be. I've been using it for a while and can almost tell what value of geometric taper to compute with by flexing the blank and looking at the relative deflection. The results, at least give me a starting point without the need to find my guide spacing booklet, which my young kids have probably used as a colouring book.

I've gotten to where I don't use a chart anymore. I hand flex the blank in varied amounts and eyeball where I think the guides need to be, starting at the tip and mark each spot with a bit of tape, drawing an imaginary line between each one. Then put a tip and guides on and do my static testing. I find I wind up nearly always using one and sometimes 2 guides less than if I started with a chart. When everything is done, I measure for records. I've had a few rods that did not end up with progressive spacing between guides.

James,
I think that you will find that the curve of most blanks will be roughly logarithmic but only roughly plus this logarithm will change depending upon the load on the rod or how much the rod is deflected. If you take the first derivative of this curve and use that for guide placement I do not think that you will end up with either the optimum number or the optimum positioning of the guides. If you can take the second derivative I think that it will show you that the curve is not monotonic.
If it is something that you are having fun analyzing then go for it and more power to you but if you want the optimum guide positioning I would take a more pragmatic approach and use the static method.

Emory:

The rod curve itself from butt to tip has ever increasing slope under load, and hence is monotonic because the first derivative is always increasing in absolute value isn't it? That is, if I brace the blank at the butt and extend it horizontally out, then dangle a weight at the tip, the slope of rod deflection is always getting steeper as I go toward the tip. That is the definition of monotonicity, no? In fact, all deflection loads on cantilevered members must result in monotonic curves in principle unless the blank is horribly bent in the middle. Or am I completely wrong?

I agree that the 2nd derivative might not be monotonic, but my formula makes a simple approximation and assumes that geometric taper has constant exponent. This certainly is contrived, but without doing any analysis, it seems like a good approximation since, as you said, most blanks are roughly logarithmic. Plus, I can't really know if a certain manufacturer maintains constant wall thickness and just increases blank diameter, if a manufacturer varies both wall thickness and blank diameter (which I'm sure at least some makers do). I just trust that most blanks sort of follow the standard slow-moderate-fast-xfast taper descriptions you see in catalogs.

If we're trying to be pragmatic, that's something I completely agree with. So if the choice is a static lookup table in a booklet or a simple formula I can exercise through my web browser, I'll choose the browser option any day and tweak from there. Like you, I'm aiming to find optimal guide placement with my little software tool. But since the deflection curve changes with load, the big question is whether an optimum spacing for guides even exists through the whole range of load. I seriously doubt it. So guide spacing is a compromise anyway.

I'm not doing this just as a mental exercise. It's a true starting point that actually comes quite close to the static tables when I've put in some standard exponents into the formula, except my applet prints computed measurements to two decimal places, and doesn't contrive a measure to the closest 1/2 inch like the static tables provide usually. Folks can continue to use static tables if they want. But what guarantees that static tables are any better than the formula I've implemented? The spacings are suspect since the incremental increases are inconsistent step functions and therefore are less likely to be a general starting point for all blanks, yet you're recommending that people use them anyway. I don't think it's a big issue, because the static tables are probably good enough for general guide spacing applications, and those who aren't comfortable with using a computer or internet browser don't have to go online with static tables. But for those who misplace their tables (like me), and those who can never find the right table that shows the right number of guides for the right rod length, a software application that gives them sufficient good guide placement to achieve a starting point is perhaps more pragmatic.

On the topic of ideal numbers of guides, I think this is subjective - or at least influenced by many parameters not related directly to rod blank action. Certainly, my simple formula does NOT provide any information on optimum NUMBER of guides. In fact, the applet requires the user enter that value. And the reason why it should be obvious to all of us why this number can change is that it's strongly dependent on factors such as line slap, low frame versus high frame, ring size, guide material, guide mass, single/double foot, spinning versus casting versus fly rod, guide foot length, stiffness of the guide feet, etc. Even line type used in fishing can impact the number of guides and type, since a little line rub during a fight with small fish on mono isn't a big problem, but get a big fish on spectra braid and line rub can cause catastrophic failure for a conventional rod.

Guide spacing and number can be influenced by many variables and looking up a sufficient number of combinations to reach some acceptable compromise is more easily done, IMO, in software where one can model the numbers and ask 'what-if" questions much more quickly, than through a static table lookup. In fact, some static tables might not even have spacings for 11 ft rods or any rod with more than 9 guides. Which is why I think software is the way to go.

James,
Because the stiffness of the blank will not change uniformly, the sign of the derivitive will change, I would say that it is not monotonic. That is why I suggested a look at the second derivitive.
You make a very good point about number of guides. I feel that the positioning of the guides is probably less important than the number and weight of the guides because this will have a larger effect on the resonant frequency and there tip velocity (casting distance, ease of casting) and alos on the efficiency (sensitivity and feel.
If you can wait until the next issue of RodMaker comes out there is an article in it that addresses this issue. If you can not wait send me an e-mail.

Emory:

I think I see now what you are referring to as "stiffness" and second derivate. I was analyzing spacing purely from total physical deflection path and first derivative that any blank would make on a horizontal jig table with a well defined chart in the background. But a more accurate analysis would take into effect the second derivate so it would better deal with extreme load situations better; case in point would be when the rod is doubled over or even back on itself (e.g. an Ugly stik can do this).

I wonder... Do the static tables take such deflection into account?

It gives me food for thought, and I probably need to pull out the old mechanical engineering texts on structural loading on cantilevered composites. The relative "stiffness" of a blank should always be increasing monotonically with diameter, but the deflection under load won't be because as the rod loads under real conditions, the total load of line tension is borne by each part of the rod blank above it such that at very high loads, the majority of the flex and stress is at some midpoint, and the tip section may be bent little if at all.

I look forward to the upcoming article. It may be enough to give me the hints I need to improve the guide spacing model in software.

James,
No, I think that the guides spacing charts have been arrived at empirically. Yes, the blank stiffness will increase with diameter at the 4th power of diameter, twice the diameter will result in 16 times the stiffness, assuming the wall thickness, modulus, etc. stays the same. However, it is good to remember that stiffness and power are not the same thing. The rods power is really the spring constant. It is a single number because a given amount of deflection is chosen at the point where it is measured. Stiffness on the other hand is a curve, the more the rod is deflected the stiffer it becomes. It is a function of sectional diameter, sectional wall thickness, modulus of the material, etc.
I am afraid that your mechanical engineering text are a good place to start but are going to be of limited use because they will be dealing with beams and tubes not a tapered tube or cone with varying wall thickness. It turns out that a rod blank is actually a pretty complex structure.
You're right the total stress will naturally be a function of the load on the rod but where on the rod that the stress and the strain will be is a function of the load and also the angle to the load. As the angle to the ground or water is increased the stress and strain tends to move up the rod toward the tip and be concentrated in a shorter and shorter section of the rod, parcially dependent upon the rods action. A low angle and the stress and strain are back more toward the butt. A high angle and the stress and strain move up toward the tip. This is what we call "high sticking a rod".
If you want to pursue this farther you might want to send me an e-mail. There are probably not many others rod builders that are really interested in this.

Personally, I think it would be a shame if there aren't a lot of rodbuilders following this. As I have always seen it, a software guideline is no better than a published chart for the same reason that charts do not work that well. You are assuming that all blanks are the same. The guidelines are being set for a predetermined number of guides. It doesn't work that way. If your particular blank mathemetically requires 9 guides, but your static testic tells you to you use 8 or 10, what are you going to use? Math cannot be applied to blanks in the grand sense. Blanks are too dynamic and too inconsistent in construction to have a formula applied to them. Any chart, any mathematical formula, or any other black magic that is associated with them will not work across the board. There's a reason they call this craft custom building. You take the materials given to you and you make the best product you can. As far as I'm concerned, no real builder worth his salt will follow "rules" based on a cool new math formula, any more than they would a published list of guide locations. They do some work on the rod they are building and change things accordingly.

Mick,
I completely agree with you. Blanks are surprisingly complex and I don't think that there is any single method or formula that works for all rods under all conditions. I generally prefer the static method but it too has weaknesses. But I also applaud James effort to gain a more in depth understanding. He may well come up with some insight that helps us all.

James,
I was able to get back to this and wanted to offer apologies to you for possibly sounding harsh. I think you may be on to something similar to the CCS though. If enough guys started measuring their guide locations on all their blanks, (even though it will be largely subjective) I think that then a computerized formula might be of certain use. A new builder could look up their particular blank and see an average of what other people have done as far as positioning guides. But for a general starting point, again, I just can't see a formula working. I've done several rods using the same blank mfgr and models, grips, guides, seats, and the guide positions vary. When you start comparing similar blanks between mfgrs, the guide locations can vary much more and I have noticed that on some blanks, in what I would call the upper 2nd quarter, the guides can sometimes wind up closer to each other than on the tip and the lower end of the blank. It's just the way some rods bend.

Mick,
Once again I think that you are right on. The vast majority of blanks actually have a flat spot at the tip. There is a few inches where the blank is actually stiffer at the tip than it is a few inches back from the tip, or as you suggest does not deflect as much near the tip as it deflects a few inches back from the tip. This means that we are often putting the first guide or even the first couple of guides too close to the tip adding unnecessary weight and lowering the rods performance.

Exactly Emory. The sweet spot of blanks (at least in the usual stuff I build) is in that upper 2nd quarter or 1st third. That's where it all happens and where things can drive a builder insane trying to figure out the best spot to locate a guide depending on the deflection of the blank with X amount of pressure. It all boils down to static deflection. I will give people my measurements on blanks, but I will not guatnee that they will be be the optimum locations. Math associated with static testing will work great. Math by itself with no properties of the blank being worked is not going to give a person a good rod.