Accelerating a fly rod - constant or constant increasing

[Merlin]

Bernd,

A two-stage cast, or an overpowered cast with a brutal input from the wrist is not effective because the rod response is delayed by the change in acceleration. The result is an increase in rod deflection, without speed benefit, because the extra speed you would expect can only happen after the line is released.

So a good acceleration is always progressive, whatever it is (constant, linear, squared, cubic and anything in between). Anyway we are unable to perform an acceleration at a given rate. If we ask ourselves to accelerate with a square form, we just don't know what to do.

Merlin

[Bernd]

Hi Merlin,
keep it "progressive" sounds good to me.

That was exactly what I was trying to say here:

Don't decrease the rate of acceleration before it's maximum

Paul said:

It's completely impossible to decrease your rate of acceleration before it is maximum!

If Paul is right, wouldn't that mean the acceleration will be progressive anyway?

Thanks,
Bernd

[Merlin]

Benrd,

That makes sense to me, you do not change the way you accelerate until you decelerate.

The only case for which it is not progressive is when it is constant, if you can make it (see Gordy's post on Bruce analysis of an ideal acceleration).

Merlin

[Bobinmich]

Paul,

Didn't we have the "constant acceleration" discussion a year or two ago? Seems to me the conclusion was that any approximation of a constant acceleration cast might be used in shorter distance casting when working for accuracy. Any attempt at distance would be more of a case of trying to get the maximum energy in the line and paying any attention to constant acceleration would be lost in the moment. We also thought that a constant acceleration cast would be nearly impossible because you can't sense acceleration and we have never gotten to the place where you can quantify what you are actually feeling (an thereby trying to control). I'll dig back and see what I can find.

Bob

[John Waters]

How does a javelin thrower's hand accelerate during the throw?

John

[gordonjudd]

How does a javelin thrower's hand accelerate during the throw?

Here is paper that deals with, "THE DYNAMIC ANALYSIS OF THE APPLIED FORCE ON JAVELIN DURING FINAL THRUST BY AN ELITE JAVELIN THROWER" that answers that question.

The authors did not show a diagram of the javelin centered frame they used for their analysis, but say:

The axial force and torque were the force and torque along the z-vector (longitudinal axis); the lateral force and torque were the resultant force and torque of the other two vectors (x & y).

Here is a plot of how the applied forces in those direction vary with time.

You can see the force applied along the long axis of the javelin had a quadratic (or probably a higher time exponent) variation with time. The resulting acceleration of the javelin along that axis would vary as force/mass so it would be accelerating as [a* t.^exponent]; i.e. non-linear function as well.

The force applied normal to the longitudinal appears to be more linear (i.e. acceleration in those directions would be linearly increasing over the time of throw)

Javelin throwers have the same physics as we do in that they want to apply the maximum force to the javelin over the limited distance they accelerate over with their arm movement. The prescription for doing that is to hold off on the acceleration for as long a possible (to preserve distance) and then hit it with all of the force (or acceleration) they can muster after they plant the front foot. That means they will use an acceleration function that has a large time exponent.

Using constant acceleration would result in a much smaller release velocity for them, just as it does in limiting the line speed you can produce in casting.

Gordy

[John Waters]

Thanks Gordy, appreciated. Affirms a number of coaching tenets for extreme distance casting. Also sheds some light on the penchant for stiffer rods by some power light line casters. The same proclivity exists amongst leading shooting head casters in the heavier line categories and I dare say, for the same reasons. The comment about relaxing the arm muscles equates to the stretching the flyline concept in distance casting.

Thanks for posting,

John

[paul brown]

Many thanks guys, this is all very interesting analytical information.

[Bobinmich]

I can't find the original discussion on constant acceleration casts but I think it was early 2009. I think Magnus was involved. I was quite parochial then, fishing mostly soft rods on small streams. It became evident to me that people cast quite differently and there are many roads to Rome. This becomes especially true when comparing casting styles of a distance caster to someone throwing a 4wt boo on a little midwest (US) stream. Realize that we have never quantified what a caster actually is "feeling" and consequently, what type of acceleration he is striving for, in different types of casts. Until we are able to do that, we will never be able to justify concepts concerning derived quantities like acceleration.

Bob

[Eugene Moore]

Bernd,
For application of the cycloid. Acceleration will build smoothly to max acceleration which will occur slightly before mid cast. This will also be max deflection of the rod. Acceleration continues at a diminishing positive acceleration until the cast is almost complete. This will keep the rod loaded but at lower deflection levels. Acceleration will stop while the cast is still moving forward at a constant velocity allowing the rod to start unloading until the rod is deccellerated quickly to a stop.
At least mechanically this would be the smoothest use of applied energy in time.
The period of reduced acceleration and constant velocity allow rod flex to partially reduce lengthening effective rod length and providing a straighter tip path.
Acceleration starts at zero increases to max then retuns to zero prior to the stop. Velocity starts at zero increases quickly before becoming constant then stops.
What I observed in tailing loops was the application of too much acceleration late in the stroke with the rod already partially flexxed.
This deflected the rod abruptly causing the rod tip to drop below the cast path creating a lower loop in the line with the extending line above it.
Wonderful camera work to all of you.
Thanks. Maybe now I can teach myself to stop.

[Bernd]

Thanks Eugene,
what would the accel. look like when producing an early dip of the tip just after starting the rotation?
Greets
Bernd

[Eugene Moore]

Bernd,
Consider acceleration and rod flex as synonomous.
The higher the accell the more the rod deflection as load increases with acceleration. As the acceleration reduces, but still positive the rod deflection will decrease. Max rod bend at just prior to mid stroke. I believe what you really want is the max acceleration at mid stroke with a reduction as you go past TDC. You will still be applying power, but at a decreasing level. Follwed by a constant velocity not constant acceleration. This will allow the rod to begin to unload increasing velocity as effective rod length increases. By allowing the rod to unload while velocity is rising the tip will be in a straighter path while the line is still accelerating. I believe what you will really want to achieve is a constant velocity. Not a constant accel. Most of the acceleration will be done through rod translation of the arm and body. The constant velocity will be mostly rod rotation from the wrist.
If you wish I could graph displacement, velocity and acceleration versus a translation and rotation increment. I'll need a couple days.
My thoughts are to use arm and body to create acceleration (force) and the wrist rotation to create high tip velocity. The arm and body being stronger for applied force and the wrist quicker to achieve high velocity without high force. All this with lower power requirements for rod and angler. No good to fold the rod with power only to wait for it to unload at delivery.

[gordonjudd]

I believe what you will really want to achieve is a constant velocity. Not a constant accel.

Eugene,
I am not sure how or especially why you would want to apply constant velocity to the line.

The whole idea is to apply work energy (force over distance) to the line so that you increase its K.E. from being nominally zero when it starts forward with a small initial velocity to a higher value at the end of the end of the cast.

Constant velocity by definition would not increase the work energy so you would end up with a launch velocity that was the same as it was at the start or some earlier point in the cast.

Consider acceleration and rod flex as synonymous.

Because the rod is acting like a forced harmonic oscillator it does not react immediately to changes in its input as would happen for the solid linkages in a car engine. That means it will have a variable deflection profile even when the angular acceleration applied at the butt is constant.

The loaded oscillation frequency related to the spring constant of the rod and the mass of the line will tend to smooth out rapid changes applied at the butt. So the only way to predict what will happen is to run different acceleration profiles through the model and see what deflection profile it will actually produce.

With that in mind, do you have a plot of the acceleration vs. time characteristics of your cycloid profile? I could run that profile through Grunde's model and see what sort of a deflection variation it would actually produce on the rod.

Gordy

[Eugene Moore]

Gordon,
Actually because it is under deflection, due to the previous acceleration, the constant velocity input will allow the rod to straighten increasing effective rod length and actually providing a rod tip speed increase for virtually no power output. The rod must straighten whether done by the caster or under it's own deflection rate. By reducing acceleration to zero you get the benefit of some free energy.. In hard linkage it allows the components to unwind prior to load reversal increasing time and reducing shock.

[gordonjudd]

In hard linkage it allows the components to unwind prior to load reversal increasing time and reducing shock.

Eugene,
Do you have an acceleration profile for that case? I should be able to simulate the expected response for any acceleration vs time curve.

I am not sure how something that works for a hard linkage can be applied to our forced harmonic oscillator case, but it might be instructive to see how the cycloid acceleration profile applied at the butt will work out in casting.

Thanks,
Gordy