Effect of Loop Shape on the propagate farther.

Fly casters sometimes say these loops ‘‘stay

aerialized’’ longer and also use the term ‘‘climbing loops’’ to de-
Drag-Induced Lift of Fly Line scribe this effect. The purpose of this note is to explain this ob-
served phenomenon by analyzing the drag on a loop as a function
of its shape. Below, we consider four qualitatively distinct loop
Caroline Gatti-Bono shapes and compare their vertical drag components.
e-mail: [email protected]
Applied Numerical Algorithms, Computational Research
Division, Lawrence Berkeley National Laboratory, 2 Analysis of Loop Drag
Berkeley, CA 94720 Four loop shapes are illustrated in Fig. 2. These loops have
identical length l⫽ ␲ R and diameter 2R, and are assumed to
N. C. Perkins propagate to the right without changing shape. Two loops are
e-mail: [email protected] asymmetrical 共Figs. 2共a兲 and 2共b兲兲 and two are symmetrical 共Figs.
Mechanical Engineering, University of Michigan, Ann 2共c兲 and 2共d兲兲. The first asymmetrical loop 共Fig. 2(a)) has a posi-
Arbor, MI 48109-2125, Fellow ASME tive angle of attack and is termed a ‘‘climbing loop,’’ while the
second 共Fig. 2共b兲兲 has a negative angle of attack and is termed a
‘‘falling loop.’’ The two symmetrical loops consist of a ‘‘pointed
loop’’ and a ‘‘circular loop’’ as illustrated in Figs. 2共c兲 and 2共d兲,
This note explains why casting a loop with a positive angle of respectively. It should be noted that the shapes of these loops do
attack is advantageous in distance fly casting. Several loop not satisfy the steady-state conditions. However, they are close to
shapes, one with a positive angle of attack, one with a negative the shapes observed in real casts and, therefore, they do provide
angle of attack, and two symmetrical loops with zero angle of good insight into the mechanics of a cast. Below we demonstrate
attack are studied. For each loop, we compute the vertical drag how the loop shape significantly affects the vertical component of
component, i.e., the ‘‘lift.’’ It is found that a loop with a positive drag on fly line. We begin by computing the velocity field for an
angle of attack generates lift about four times larger than a sym- arbitrary loop shape.
metrical loop. Thus, loops with positive angles of attack stay ‘‘ae-
rialized longer’’ which is consistent with observations made by 2.1 Velocity Field. Figure 3 illustrates an arbitrary loop
(competition) distance fly casters. 关DOI: 10.1115/1.1778414兴 shape and a control volume that travels with the loop with veloc-
ity vc v ⫽ v c v i. The upper portion of the loop 共assumed horizontal兲
travels with velocity vo ⫽ v o i and the velocity of the bottom por-
1 Introduction tion of the loop 共assumed horizontal兲 is zero as it is attached to the
end of the stationary fly rod 共and the effects of gravity are ne-
Fly casting involves considerable mechanics of both the fly rod glected兲. The velocity of an arbitrary material point P relative to
and fly line as described in several studies, 关1–9兴. For instance, the control volume is denoted as vr and its magnitude is uniform
the angler imparts both rigid and flexible body motions of the fly along the loop since the loop does not deform as it propagates.
rod in accelerating the fly line during the forward and back casting The absolute velocity of point P is
strokes. These strokes end with an abrupt deceleration of the fly
rod, often referred to as the ‘‘stop,’’ after which a ‘‘loop’’ of fly vp ⫽vc v ⫹vr (1)
line is formed as shown in the photograph below. This loop propa-
The velocities of the material points coincident with the top point
gates as a nonlinear wave under the action of fly line tension, air
共AA兲 and bottom 共point BB兲 of the loop are
drag and gravity. The initial conditions that form the initial shape
and velocity of this loop are generated during a short time interval vAA ⫽ v o i (2)
following the stop, 关10兴. Eventually the loop propagates to the end
of the fly line and the attached ‘‘leader’’ and ‘‘fly’’ turn over as the vBB ⫽0, (3)
line straightens at the end of the cast.
The dynamics of this loop is the subject of a number of studies respectively. Using Eqs. 共2兲 and 共3兲 in Eq. 共1兲 leads to the con-
beginning with those that assume idealized semi-circular or square clusion that
loop shapes, 关1–3兴. Further studies, 关5–9兴, relax these assump- vo
tions and compute the loop shape from the kinematics of the at- v cv⫽ v r⫽ (4)
tached fly rod and the equations of motion of the flexible fly line. 2
Fly casting experts 共see, for example, 关11–13兴兲 are fully aware Therefore,
that the shape of the loop has considerable influence on its dy-
namics. For instance, it is well understood that loops with smaller vo vo
diameters propagate farther as they provide less projected area in vp ⫽ i⫹ et (5)
2 2
the flow, hence less air drag. As a result, casting small loops are a
distinct advantage when casting longer distances or into a head where et is the unit tangent vector to the loop at P. Thus, given the
wind. shape of the loop, Eq. 共5兲 can be used to evaluate the velocity of
The purpose of this note is to explain a second advantage that an arbitrary material point. Resolving this velocity into compo-
results from casting loops that are asymmetrical and with a posi- nents tangential and normal to the loop allows one to compute the
tive angle of attack such as shown in Fig. 1. Fly casting loops of drag components due to skin friction and form drag as follows.
this form have a ‘‘pointed’’ top portion followed by a larger and 2.2 Vertical Drag Component. The drag on an element of
more rounded ‘‘belly’’ beneath. The belly forms a positive angle fly line derives from skin friction 共tangent to the element兲 and
of attack in the flow and the air drag along this portion of the loop form drag 共normal to the element兲. The drag coefficients for skin
generates a component that is vertically upwards. The upward friction and form drag are denoted by C dt and C dn , respectively.
drag component acts opposite gravity and allows these loops to Let ␳ a denote the density of air and let d denote the diameter of
the fly line 共considered uniform for this example兲.
Contributed by the Applied Mechanics Division of THE AMERICAN SOCIETY OF
MECHANICAL ENGINEERS for publication in the ASME JOURNAL OF APPLIED ME-
The loops shown in Fig. 2 are composed of straight segments
CHANICS. Manuscript received by the ASME Applied Mechanics Division, August and circular segments that, in total, subtend a semi-circle. For a
28, 2003; final revision, December 16, 2003. Associate Editor: O. M. O’Reilly. straight segment of length l s , the vertical component of drag is

Journal of Applied Mechanics Copyright © 2004 by ASME SEPTEMBER 2004, Vol. 71 Õ 745
 Fig. 3 Kinematics of arbitrary loop

and the contribution due to form drag on the straight segment

共second term兲 is negative due to the negative angle of attack 共Fig.
Fig. 1 The fly line ‘‘loop’’ is formed after the ‘‘stop’’ in a cast- 2共b兲兲.
ing stroke and propagates as a nonlinear wave. This loop is
asymmetrical and possesses a positive angle of attack. Such
Pointed Loop: For this loop, the straight segments BC⫽DE
loops are a hallmark of expert fly casters. ⫽ ␲ R/4 and the semi-circle of radius R/2, composed of the arcs
AB, CD, and EF, are subject to drag. The total vertical drag
becomes

1 1
D Y ⫽⫺ ␳ a ␲ dl s C dt v t 兩 v t 兩 et •j⫺ ␳ a dl s C dn v n 兩 v n 兩 en •j (6)
DY⫽
1
16 冉
␳ a dR v 2o
14␲ 2 ␲ 共 ␲ 2 ⫺4 兲
3
⫹
␲2 冊C dt (10)
2 2 and the contributions due to form drag cancel due to symmetry.
where v t and v n are the velocity components tangential and nor- Circular Loop: The result for this loop follows directly from
mal to the fly line, respectively 共see, for example, 关14兴兲. Note the Eq. 共7兲
contributions of both skin friction and form drag to this result. For
␲
the 共sum of兲 circular segments of radius R, the vertical component D Y ⫽ ␳ a dR v 2o C dt (11)
of drag is 3

DY⫽ 冕冋␲

0
1 1
⫺ ␳ a ␲ dRC dt v t 兩 v t 兩 et •j⫺ ␳ a dRC dn v n 兩 v n 兩 en •j d ␪
2 2 册 and it is independent of form drag as mentioned above.

3 Example and Conclusions

␲ Prior studies of fly line dynamics have used slightly different
⫽ ␳ a dR v 2o C dt (7)
3 values for drag coefficients for skin friction and form drag,
关1–3,5–9兴. Here, we shall assume values C dt ⫽0.015 and C dn
and this drag contribution depends only on the skin friction. We ⫽1 that are typical of those used in prior studies. We also recog-
now employ Eqs. 共6兲 and 共7兲 to compute the vertical drag com- nize that these values depend, in general, on fly line speed 共Rey-
ponent for the four loops of Fig. 2. nold’s number兲, 关2兴. Using these drag coefficients and the results
Climbing Loop: For this loop, the straight segment CD⫽( ␲ 2 above leads to the following table that compares the lift on the
⫹4/4␲ )R and the semi-circle of radius R/2, composed of the arcs four loops shown in Fig. 2.
BC and DE, are subject to drag. The total drag is expressed as The results of Table 1 show that the lift generated by a climbing
1
D Y ⫽ ␳ a dR v 2o
2 冉
共 ␲ 4 ⫹8 ␲ 2 ⫹64兲 ␲
3 共 ␲ 2 ⫹4 兲 2
C dt ⫹
␲ 共 ␲ 2 ⫺4 兲
C
共 ␲ 2 ⫹4 兲 2 dn
冊 (8)
loop is approximately four times greater than that of a semi-
circular loop with the same characteristic dimensions. The source
of this additional lift is the contribution of form drag on the
and the contribution due to form drag on the straight segment ‘‘belly’’ of the fly line that has a positive angle of attack. The
共second term兲 is positive due to the positive angle of attack 共Fig. negative angle of attack for the falling loop shape results in a net
2共a兲兲. negative ‘‘lift,’’ again due to the form drag on the belly. The
Falling Loop: For this loop, the straight segment BC⫽( ␲ 2 symmetrical loops 共circular and pointed兲 generate approximately
⫹4/4␲ )R and the semi-circle of radius R/2, composed of the arcs the same lift. These results may be readily generalized to other
AB and CD, are subject to drag. The total drag is expressed as loops shapes.

冉 冊
This note explains a fact observed by fly casting experts,
1 共 ␲ 4 ⫹2 ␲ 2 ⫹4 兲 ␲ 共 ␲ 2 ⫺4 兲 namely, that a climbing loop is advantageous in distance casting.
D Y ⫽ ␳ a dR v 2o 4 ␲ C dt ⫺ C
2 3 共 ␲ ⫹4 兲
2 2
共 ␲ 2 ⫹4 兲 2 dn How to generate a climbing loop through control of the casting
(9) stroke is left as a 共considerable兲 exercise to the reader.

Fig. 2 Four qualitatively different loop shapes. „a… and „b… l Ä AE , „c… l Ä AF , and „d…
l Ä AB

746 Õ Vol. 71, SEPTEMBER 2004 Transactions of the ASME

Table 1 Comparison of the vertical component of drag for the 关4兴 Hoffmann, J. A., and Hooper, M. R., 1998, ‘‘Fly Rod Response,’’ J. Sound
four loops shown in Fig. 2. Vib., 209共3兲, pp. 537–541.
关5兴 Hendry, M. A., and Hubbard, M., 2000, ‘‘Dynamic Finite Element Simulation
Climbing loop Falling loop Pointed loop Circular loop of Fly Casting and Its Potential Use in Fly Rod Design,’’ Proceedings: The
Engineering of Sport, Blackwell Science, London, pp. 407– 414.
16D y 关6兴 Gatti-Bono, C., and Perkins, N. C., 2002, ‘‘Physical and Numerical Modelling
0.924 ⫺0.450 0.276 0.251 of the Dynamic Behavior of Fly Line,’’ J. Sound Vib., 255共3兲, pp. 555–577.
␳ a dR v 2o
关7兴 Watanabe, T., and Tanaka, K., 2002, ‘‘Modelling the Dynamics of a Fly Line,’’
Proceedings The Engineering of Sport, Blackwell Science, London, pp. 353–
359.
关8兴 Gatti-Bono, C., and Perkins, N. C., 2003, ‘‘Comparison of Numerical and
Analytical Solutions for Fly Casting Dynamics,’’ J. Sports Eng., 6共3兲, pp.
Acknowledgment 165–176.
The authors wish to acknowledge the input of Mr. Bruce Rich- 关9兴 Gatti-Bono, C., and Perkins, N. C., 2003, ‘‘Numerical Model for the Dynamics
ards from Scientific Anglers 共3M兲 who first posed this interesting of a Coupled Fly Line/Fly Rod System and Experimental Validation,’’ J. Sound
Vib., in press.
question to us. 关10兴 Perkins, N. C., and Richards, B., 2003, ‘‘Dissecting Your Casting Stroke,’’ Fly
Fisherman, Primedia Magazines, Dec.
References 关11兴 Krieger, M. K., 1987, The Essence of Flycasting, Club Pacific, San Francisco,
CA.
关1兴 Spolek, G. A., 1986, ‘‘The Mechanics of Flycasting: The Fly Line,’’ Am. J.
Phys., 54共9兲, pp. 832– 835. 关12兴 Wulff, J., 1987, Fly Casting Techniques, The Lyons Press, Guilford, CT.
关2兴 Lingard, S., 1988, ‘‘Note on the Aerodynamics of a Fly Line,’’ Am. J. Phys., 关13兴 Kreh, B., 1991, Modern Fly Casting Method, Odysseus Editions, Birmingham,
56共8兲, pp. 756 –757. AL.
关3兴 Robson, J. M., 1990, ‘‘The Physics of Flycasting,’’ Am. J. Phys., 58共3兲, pp. 关14兴 Sarpkaya T., and Isaacson M., 1981, Mechanics of Wave Forces on Offshore
234 –240. Structures, Van Nostrand Reinhold, New York.

Journal of Applied Mechanics SEPTEMBER 2004, Vol. 71 Õ 747