To all experienced builders:

Thanks in advance for your thoughts on this one.

I have been playing around with one blank for about 6 months, changing guide positions and spine positions in order to find a semi-systematic method for documenting how some of these parameters affect casting and feel. I'm noticing that a major factor in casting distance seems to be rod recovery / resonance damping. Interesting, this parameter seems too be most sensitive to spine position. In the blank I use (which is a very high quality blank that most seem to like on this board), I get better distance when ignoring the traditional spine and orienting the rod for maximum resonance damping (which I can only grossly test by feel). My initial conclusion is that even small resonances in the rod can rob it of distance - which makes sense in that the line is trying to shoot out of guides that are moving up and down during the cast, causing more friction against the guides and rod itself.

Any thoughts from some others? And if this is something you use as a building parameter, how do you test for it?

Mark.

You're not alone Emory Harry has some great observations on this subject in articles he has done in RodMaker, seminars, and post here. Use the Search engine check authors only, all dates for the info.

Mark,
The resonant frequency will vary as you rotate the blank. The soft axis (the spine) will have a little lower resonance frequency than it will have on the stiff axis, just as when all other things are equal a more powerful rod will have a higher resonant frequency than a lower powered one.
However, do not confuse the resonant frequency with damping or damping factor. If you mount the butt of the blank down firmly to something like your work bench and then deflect the tip of the blank and then release it, how rapidly it oscillates or vibrates is the resonant frequency but how long it takes the oscillations to damp out or how long it takes for the oscillations to be reduced to some very small level is the damping factor. A given raw blank may have a resonant frequency of several cycles per second but may take several seconds for the oscillations to damp out.
The resonant frequency will have a significant effect on casting distance but damping will not usually have a significant effect with modern graphite blanks. After guides and wraps are added and with line and terminal tackle added and also with your hand added the length of time it takes for the oscillations to damp out is dramatically reduced. In use most graphite rods will tend to damp to a very low level in only a couple cycles so damping does not normally have much effect on casting distance. With glass and bamboo rods the damping will be quite a bit longer and can have more effect on casting distance.




Edited 1 time(s). Last edit at 09/10/2005 08:59PM by Emory Harry.

Mark

Here's what I understand you are talking about--when you stop the rod on the forward stroke to form the loop and send the cast off, the rod tip bobs up and down (oscillates) which puts waves in the line diminishing the cast distance you get, and presses the line against guides and blank, increasing friction on the shooting line. The cause of it is the following. To make a cast of a given distance you have to get the line up to a certain speed regardless of the rod type. Until you stop the rod, line and rod tip are necessarily moving at the same speed. When you stop the rod, the line uncouples from the rod and its kinetic energy makes the cast happen. Unfortunately, that moving rod tip also has kinetic energy and this goes into the oscillation of the rod tip. The amount of this energy varies directly with the effective mass of the rod tip. It tends to be rather greater for bamboo and fiberglass rods than graphite, as Emory points out, although there is also considerable variation in graphite rods. Generally, faster rods will be thinner towards the tip and thus have less mass up there, and put less energy into the oscillation. Thus the amplitude of the oscillation is less and the effect is less noticeable. It's interesting that you have found some dependence on the orientation of the guides on the rod, but I think the effect is going to be there in some measure no matter what you do. I think the largest effect is in the initial choice of the blank and assuming you don't use depleted uranium for the guides and tip top, how you build won't have a large effect. Generally, the higher the frequency of the rod, the less the effective mass of the tip, and the less noticeable the oscillations.

Mike

Mark,

It's a little bit of a technicality, but when you deflect the rod and let oscillate freely, you are experiencing the "natural frequency" of the rod as opposed to the resonant frequency, which is related to a system which is continuously driven. In the simplest of terms, the frequency of the rod is proportional to the (sqrt) stiffness/mass, so if you can find a stiffer axis, or otherwise reduce the mass, the natural frequency will increase....so the rod will vibrate faster. For a material with a uniform cross-section and known modulus, you can calculate the frequency. But in the more complex case of the rod/blank it is much easier to just measure frequency and it usually takes some electronic gear like an oscilloscope to do so with a degree of precision.


As Emory mentioned, the damping function is an altogether different beast than the natural frequency. Here you're talking about how the system dissipates energy and the damping factor is independent of geometry. So changing the orientation won't change the damping, which is somewhat intrinsic to the material. Related to what Mike said...... If however, the change in stiffness or mass affects the deflection (amplitude), then you could well affect the way in which the rod settles over time.

Mark

Thanks to Emory, Mike, and Mark for the comments.

I agree with everything said above. All of what has been described is essentially the physics of a tapered rod:

1. There is an inherent frequency of free oscillation when the rod tip is displaced.

2. There is a "damping factor" governing the dissipation of energy in a rod that has been displaced and is returning to it's original position. This is an aggregate factor that is influenced by material, rod taper, and length.

3. Related to the "damping factor" is the amplitude of the initial couple of oscillations in a rod/blank.

So here are my observations when it comes to a flyrod:

My experiment arose from the need to replace the tip section on my favorite rod. I ordered an extra tip and began playing with guide placement and spine.

1. Guide placement does matter. It does not affect the feel of the rod as much as the inherent "shootability" of line. I measured the amount of line I could shoot in 100 casts with guides placed according to static displacement testing and 20% off in either direction. The difference in shooting starting with about 50ft in the air was 5-7%.

1a. The test that was most sensitive to guide placement was the ability to shoot line on the backcast when double hauling. There definitely was a "sweet spot" for guide placement that allowed for maximum line movement, and this was correlated with how easliy line moved out during the backcast when double hauling.

2. Spine position affected the "feel" of the rod somewhat in false casting. I could generally find a difference in spine orientation by measuring the interval between forward and back casts using a video camera with a known speed in frames per second, controlling for 50ft of line in the air and a consistent loop size. Even though the variation in cycle time was fairly small, as a caster, I felt that the rod was different for the tip section oriented in the stiffest axis versus the least stiff axis.

3. When casting for distance, I began to notice something interesting. There was a huge difference in the response of the rod to EXTRA force when shooting line. By this, I mean that on the final forward stroke using a double haul, certain spine positions yielded fairly large amplitude oscillations when applying extra force. Or to say it another way, when just shooting with the final cast similar to the false cast in applied force, there wasn't much of a difference in distance. However, applying extra power to the forward cast caused a much greater difference in casting distance, and seemed to be correlated with spine. I adjusted the guides with static displacement for each of the spine positions in this tip section. This spine position WAS NOT the stiffest or the least stiff orientation.

I don't think I'm imagining this effect. Subsequent to these experiments, I have tried this on a couple of different rods, including a commercial 2 piece $600 rod. (yes, I cut off the guides and started screwing around). Same effect....and it turns out that the commercial rod was not oriented in the direction to diminish oscillations during a powerful forward cast.

By the way, the difference in most to least oscillations on the powered forward cast was almost 20% at 50 ft in the air. In addition, if I measured now much line I could keep in the air, I could keep a longer length in the air with a spine orientation that minimized oscillation.

Comments? Sanity check?

Mark,
The peroid that it takes for the rod to damp out the oscillations should not be affected a great deal by the allignment of the spine on most rods because all that is changing is the stiffness and the difference in stiffness between the stiffest and the softest plane is not large relative to the overall stiffness of the rod. At least that is the case on well made blanks.
I suspect what you are seeing may be a function of how much your casting stroke is affecting the damping period. The difference in the damping period of a raw blank and a rod in use can be very large. As I mentioned before the addition of guides, wraps, line and your hand reduces the amplitude of the oscillations. A difference in your casting motion can have a significant effect on how long it takes to damp out the oscillations.

Mark Gibson/Mike McGuire,
I stand corrected however the difference between a driven system and a non-driven system, resonant frequency and natural frequency, is not a distinction that we dumb EE's usually make. I guess though, just for fun, I could argue that any system that is oscillating IS or HAS BEEN driven. It will not oscillate unless energy is being put into the system or has been put into the system. Then the question arrises how about the system that is driven at other than the resonant frequency or natural frequency or the system that is aperiodically driven. This sounds more like a fishing rod to me which I think would make the arguement that either term could be used for a fishing rod, either natural frequency or resonant frequency.
While we are on this subject, you physics guys use the term first harmonic. A term that I do not think we EE's use or at least the dumb EE's do not use. What is the difference between the resonant frequency and the first harmonic. To me they are exactly the the same thing.
I think that I should have better sense than to try to start an arguement with both you two heavy weights at the same time.

In archery, they use dampeners and stabelizers to reduce the vibration of the bow after a shot. Since rod recovery and resonance damping, have such a large effect on casting distance, has anyone ever put thought into making a buttcap, with the same dampening abilities. Maybe a fluid filled cap or somthing to make a rod dampen quicker. just an idea but it seems like a good one. especially if you could keep it light weight. Maybe some beads suspended in oil.

Emory,

You're asking some good questions. The main difference between the idea of the resonant and natural frequency relates to how the system is controlled....or not. If you take the rod and pulse it once and allow it to vibrate, you are essentially exciting it's natural frequencies. There are a set of frequencies or harmonics that will show up which are a function of the geometry and material the object is made of. There can be quite a few natural frequencies with different wave shapes, or modes as they are called. Once pulsed, or even in the case where a system which once driven and the driver is removed, the system will go into free decay at it's natural frequency and the amplitude of the vibration will diminish over time.

For resonance to happen, there has to be a continuous external driving force that is very close to one of the natural frequencies.

There are also external forces which will cause the damping of the vibrations, such as air friction or slip damping. As Emory pointed out, you can significantly alter the damping characteristics of the rod simply by how tightly you hold it in your hand. To prove this to yourself, deflect a rod tip that is clamped tightly at the handle compared to one that is loose and you'll see that the rod held loosely can easily become highly over damped to the point where it might not even oscillate at all.

Matthew, You have a very interesting suggestion but I would guess that a damper in the butt section would have a small effect on the working section of the blank. Your general idea has merit though and I'm not sure how many people are aware that Orvis actually has a couple of patents whereby they coat or fill the inside of the rodblank with a visco-elastic material in order to control and damp out unwanted vibrations.


Mark, Your casting results are very interesting. If you can quantify your results you could set up a cause and effect (C&E matrix) which might help you understand the root cause of the effect you are seeing. I agree with Emory that many blanks may not have a noticeable spine, but of those blanks that do have a pronounced spine, I've seen as much as a 10-12% difference in stiffness between the relaxed spine and the stiffest axis.


Mark
-

WOW! I may hold the record for the most words in a single Post or Reply, but this Thread has to hold the record for the most technical, and compact, considering the amount of engineering / physics involved. Thanks to all for such an excellent discussion. And special thanks to Mark Gibson for so carefully outlining in language and equations the relevant system parameters and dynamics. This is what my incomplete understanding has been waiting for.

It seems like this damping effect is most pronunced in fly rods, and probably less so in other types of fishing rods. ... What about the effect of the choice and placement of line guides?

How the magnitude of one (the improvement of casting distance by the proper orientation of the spine with respect to damping) compares to the magnitude of the other (the improvement in casting distance by the proper choice and placement of line guides of low-mass, high frame and larger ring - a path of least resistance) is unknown to me.

It seems like guide choice & placement would also contribute significantly to optimizing overall casting performance and distance. -IMO, Cliff Hall+++

In summary, then, is it fair to say that:
(1) the simplest way to improve rod damping is to hold the rod more loosely for a moment toward the end of the cast; and
(2) the simplest way to maintain a higher resonance frequency is to keep guide mass low (assuming a given modulus material and tip-taper); and
(3) since resonance frequency is only inversely proportional to the SQUARE ROOT of the TOTAL mass of that linear segment of the built rod (mass of rod blank + guide mass + thread wrap + epoxy finish), and not just the guide mass alone, then guide mass would have to vary by a fairly huge amount (like 50% between different styles or sizes) before the effect on resonance frequency would be more than say 10%.
(4) IF I can figure out how to determine the axis of highest (?) resonance (LOL) or damping, I may be able to increase casting distance by 10+%. I may have to rely on orienting the stiffest axis in the plane of casting to achieve this higher resonance [per Frequency = K * SQRT(MODULUS / MASS)], where K = some constant and has the units of Hz (CPS = 1/sec). [Ignorance may be bliss here.]
So, is that about right, Gibson? Thanks in advance for any succinct answers. -Cliff Hall .



Edited 5 time(s). Last edit at 09/12/2005 04:43PM by Cliff Hall.

Mark,
Thank you. but your explaination brings up a number of additional questions.
First, let me ask about your formula for resonant frequency, (sqrt) stiffness/mass. I assume that you used this for purposes of simplification. The formula (sqrt)spring constant/mass is the formula for a beam or a tube, as I am sure that you are very well aware, but a blank is essentially a long cone with varying wall thickness so the formula must include terms for length, the distribution of both the mass and the stiffness and the cross section.
Second, you mentioned other modes. We EEs would call these different wave shapes but I do not understand how a blank resonating can be anything other than a sine wave. There can be harmonics or multiples of the resonant frequency if the driving force is at other than the resonant frequency but they should very quickly drop to zero amplitude for a undriven blank and the blank continue to oscillate at the resonant frequency or natural frequency for the damping period but the oscillation will be a sine wave.
Third, the second question brings up a question about Q. In fact, I have several questions about the Q of a blank. In a resonant circuit if the Q is high the circuit will usually only resonate over a very narrow range of frequencies. Is the same true for a blank? Does the Q of the blank when loaded with guides, line etc. drop significantly? If it does then your points about other modes makes more sense to me. How would you go about measuring the mechanical Q of a blank and what are the main variables affecting it?
Sorry for all of the additional question from a dumb EE.



Edited 1 time(s). Last edit at 09/12/2005 11:50AM by Emory Harry.

Thanks for all the discussion.

Mark G: I am indeed setting up a C/E matrix with my results. However, given the number of variables, I think it may be difficult to discern anything from the data. One thing I am becoming convinced of is that the "static" model that we often apply to flycasting is only part of the picture. I have noticed with my own experimentation, observations of great casters, and my own improvement over the years as a caster, that the dynamics of applied power has a huge influence on the performance of a rod. Another way to say this is that the forces required to keep 30-50 ft of line in the air is NOT the same as the force used to actively shoot line. It is this ability to apply power and accelerate the rod tip that allows a caster to shoot more line. In the next week or so I will try to post the pictures of a 3 piece rod with the tip section oriented at softest, stiffest, most "damped" (by feel) and least "damped" by feel during a forward cast starting with 50 ft of line in the air. I think you will be impressed by the amplitude of the resonance for some of the orientations - and lack thereof in the others.

As far as modifying damping by holding the rod differently, I'm going to try it out this afternoon if I can get out of the office before dark.

Emory: I think there is a difference between a system that is being driven by a constant input versus a system that responds to a dynamic input. Has anyone measured the low frequency modes for a flyrod? ..those modes in the few per second frequency in plane with the applied force would be the most important I would think.

Theorizing aside, Cliff hit the nail on the head...maybe this issue of resonance damping matters in building an optimized fishing tool. As an experimentalist, I think that my observations are changing the way I build rods. Cliff summarized it well by trying to rank order effects. I am now starting by putting on a cork handle and tip top. I then use the following crude test:

Put about 40 ft of line out of the tip top and keep it in the air. Apply more and more power, observing resonance in the rod. It becomes fairly obvious if a "bad resonance" is present.

I start with least stiff axis in all segments.

Second is stiffest axis in all segments.

Third is least stiff all segments changing the tip from stiffest to
Then, just for fun, I try the tip section in the two other orientations not covered by stiffest / least stiff.

I then chose the most damped orientation and proceed to guide placement using a static displacement method.

I then do a lot of casting with the guides wrapped but not epoxied to fiddle with placement to maximize shooting line out the backcast when double hauling.

Sounds like a lot of work, but I have found that the difference between fishing a optimized and non-optimized rod is dramatic for the type of dry-fly fishing that I do.

Mark Li

Mark Li,
Stimulating lower or higher resonances or modes and how much energy is required is going to mainly be a function of the Q of the rod. And frankly I do not know much about rod Q. That is why I asked Mark Gibson the questions that I asked him. If the Q is very high I think that it will be almost impossible to cause the rod to resonate at a lower or higher frequency than its natural or resonant frequency.
I am not sure why you care about lower modes. There are some fly rod user that try with their casting strokes to introduce a second harmonic to the rod so as to increase their casting distance but I am not a fly fisherman and do not really know very much about the technique they use.

Emory,

Yes, I significantly oversimplified the equation in order to reduce the frequency relationship to the stiffness and mass. There are a number of other terms in there, but I just wanted to call out those effects. The stiffness for example will contain modulus and geometry terms (area moment) but it's easier to think about when you combine those. I mainly threw that out to show the general effects, but if you were to try and do a calculation, you’d need the full equation and then you'd have to integrate it down the blank to account for the change in cross-section.

Cliff, yes I agree that the square root will compress the effect of mass on the frequency, but I would guess it could still be significant or measurable esp. with added weight in the tip section or on lighter power rods?


Emory, On the waveforms...when you pulse the rod/blank, you hit it with a broad frequency spectrum of complex waveforms and the natural frequency and multiples will take on sinusoidal shapes. Sometimes interference patterns can distort the shapes, as will any irregularities or discontinuities in the medium that the wave is traveling through.

As far as the Q factor, there are a couple of ways you can look at it. It's typically thought of as the quality or sharpness of a system in resonance. So a higher Q would give you a sharper peak in the frequency response of a driven oscillator. The way you would measure this would be to sweep the frequency and measure the system response (amplitude) to get the so-called resonance curve. The plot of amplitude as a function of frequency gives you a curve with a maximum at the resonant frequency. Keep in mind that the resonant frequency is usually somewhat higher than the natural frequency, and the degree of shift is dependent on the damping. In this case of a driven system, Q is the ratio of the resonant frequency divided by bandwidth of the peak. You can also think of the Q in more general terms for any oscillating systems as related to the ratio of the energy stored to the energy lost per period. The ratio of the energy storage and loss can be easily measured for materials (composites) using an instrument called a Dynamic Mechanical Analyzer….got any composites you want measured? A third approach to measuring the Q of a complete rod or blank would be to drive the rod on one end and measure the degree (delta) to which the response of the rod is out of phase with the driver. The tangent of the phase angle will be inversely related to to Q as 1/tan delta.

mark

To Mark, Mark, Cliff, Matthew, Emory et. al.

I have worked out a way of measuring the intrinsic mass of the rod and the spring constant at least for small oscillations. Emory is familiar with it. The idea is to measure the frequency with different masses attached to the tip and curve fit to get the values of intrinsic mass and spring constant. I have put some effort into making it simple to do this by writing Windows dialog program to do it. Email me, and I will be happy to send you a copy along with how-to-use instructions. I would of course be interested in any comments and feedback. If this all holds together, this could be a significant way that the custom rod builder can further add value to the product.

Mike

Just reading through this series of posts, interesting topic. I saw one of the posts that made me think of a product I had seen for cycling that is designed to eliminate vibrations travelling through the handlebars. Check out this link:

[www.bontrager.com]

Not sure if this totally relates to the topic being discussed, but I thought I would offer it up anyway.

Mark Gibson,
Thank you very much for the education. With your explaination of Q it does not appear that there is any difference in your explaination of mechanical Q, the Q of a blank, and what I am familiar with the Q of an electrical resonant circuit with one exception. I should look it up but if my memory is correct the Q of an electrical circuit is not a function of the bandwidth at the peak. It is measured at the half power point. A high Q circuit, and I assume some mechanical devices, may have virtually zero bandwidth at the peak.

I have never heard of a Dynamic Mechanical Analyzer but I had thought of the solution you mentioned of sweeping the blank through its resonance. Actually I did not think of it. I saw it in a paper written by
Dr. Graig Spolek. Where he uses a synthesizer driving a scotch Yoke mounted to the butt of the rod. Are you familiar with his papers?
It would really be interesting to be able to measure a rods Q and how it lowers when guides and wraps are added and also how it lowers with the addition of line through guides. I think that we would learn something about the material that the blank is made of and even things like potential casting distance, ease of casting, and maybe even rod sensitivity. Can your Dynamic Mechanicial Analyzer also measure the slope of the skirts of the Q? In an electrical resonant circuit the slope can be completely different depending upon the components used and I would assume that the same thing would be true for a mechanical device, blank ,and the construction material.

Thanks again I have read your post over several times and will probably think about it and read it over several more times.

Mike McGuire,
Some time ago I was thinking about the use of your software to aide in finding the optimum casting weight for a give rod or blank. This is very often a question and not just for fly rods. What do you think?

Also I would encourage anyone interested to contact Mike and experiment with his software. It is very clever and very straight forward to use. The measurements can all be made with just a stop watch and counting rod oscillations by eye.

Emory

the optimum casting weight is what you would determine with a CC measurement which would give you an ERN which can be translated into a lure weight. As Tom has pointed out many times, this is the optimum weight for a fly rod to cast. If you are going to cast the canonical 30 feet of line, then this is the line weight number your will be using. Reduce or increase the length of line by about 5 feet and you increase or decrease the line weight number by one to get the same weight to cast. Of course in real fishing situations, we will cast short or long to the extent of our ability to present to a fish. The interesting figure of merit would be the ratio of intrinsic weight of the rod to the ERN weight. That would tell you how much energy you are putting into moving the line or lure versus how much you are putting into what has to be damped and dissipated to keep from screwing up the cast. At this point, my software doesn't calculate this, but I will add it soon.

Mike