Constant Acceleration

[Merlin]

Hi Few

Let’s consider the rod as a rigid body, and define R as the radius of the pulley. The angular acceleration once the door is opened is constant and equal to

MgR / sum of moment of inertia (rod+reel+pulley)

The change is abrupt, there is a “jerk” in terms of angular acceleration. In this example the situation is a bit idealized. If the rod is flexible, then things change, the tip kicks back once the door is opened and kicks forward as the rod butt is stopped with possibly a rebound backwards of the butt as the rod is “shocked” by the stop. In reality we cannot produce a jerk in acceleration, the change is always progressive. When one speaks of constant acceleration in the case of rod casting this relates to the main part of the acceleration phase of the cast, not the very beginning and not the end (deceleration). And this is not the most performing type of acceleration for distance, in that case it is better to produce an increasing acceleration followed by a deceleration.

Now I have a question for you: why do you pay so much interest to the jerk which is theoretical and does not occur in practice?

Merlin

[gordonjudd]

I cannot see an advantage by using a mostly constant acceleration for rod butt rotation in distance competition.

Merlin,
As noted in this thread on the old board constant acceleration of the rod rotation is the last thing you would want to do in terms of producing maximum line speed in a distance cast.

I think that you will find that most accuracy casters do not use constant angular acceleration in their casting stroke either. You can see that the angular velocity curve (green curve) of Maxine McCormick's cast is far from being linear.


Gordy

[gordonjudd]

Here is the angular acceleration curve (blue) measured in the above accuracy cast of Maxine McCormick's.

angular_omega_accel_mccormick.jpg (42.6 KiB) Viewed 248 times

You can see that her acceleration curve is at least quadratic, but is buttery smooth. Her angular deceleration was also quite high and was about 1.8 times larger than the max acceleration observed up to RSP1.

Now I have a question for you: why do you pay so much interest to the jerk which is theoretical and does not occur in practice?

Fortunately the spring in the rod will tend to smooth out discontinuities produced in the acceleration of the butt rotation. This acceleration curve (which has negative values since the rod rotation was clockwise) produced in this cast had a sizeable acceleration discontinuity (jerk) in in the middle of the cast at around -.24 s, but still produced a good loop.

angular_accel_firm_grip.jpg (41.8 KiB) Viewed 248 times

Gordy

[Few]

Thanks, all, for humoring me. I fully understand that constant acceleration is not what folks are arguing for.

My interest in "jerk" stems only from the posts earlier in this thread. There it sounded like the large jerk associated with going from zero to constant acceleration was cited as the reason such an acceleration profile could not be achieved by a caster. I follow the mathematical argument for the existence of jerk, but the "so a caster can't do it" conclusion feels suspect. The velocity clearly can't jump instantaneously because that would entail infinite acceleration and require infinite force. But turning on a force very quickly (granted not instantaneously) doesn't seem unphysical to me. My thought experiment was just an attempt to think of a simple physical system that exhibits that behavior--something a bit more relevant than the first example that entered my head: an electron between the plates of a capacitor that is suddenly charged!

Anyway, thanks again for the patient responses. I don't need to drag everyone else any further through my learning process!

Few

[Walter]

When I look at the acceleration graphs I see a curve not a straight line. To get “constant acceleration” the beginning of the cast is ignored.

Reminds me a bit about discussions regarding the straight(ish) line path of the rod tip.

It’s all approximation. In this case even that is an approximation.

[Paul Arden]

Yes Walter, but I’m not convinced the conclusion that approximation of constant acceleration is what we want either. We might have something like that appearing on a short cast if that’s what you are expecting, but the argument I’ve heard put is that increasing acceleration will cause a tail. However that’s exactly what we have at the start of the stroke to avoid one.

Cheers, Paul

[Bernd Ziesche]

When I look at the acceleration graphs I see a curve not a straight line. To get “constant acceleration” the beginning of the cast is ignored. Now one could say it’s because we are human. Or perhaps it’s because we need to start the stroke slowly?

Hi Paul,
I've no time to read all postings.
So I comment on your first post. You overlook translation first, then rotation. Obviously you need to define what kind of acceleration you are talking about.
Main rotation should be smooth enough to not create a dip in tip path. That's all that matters.
The highest rate of accel. you can control allows for the smallest arc for the desired line speed.
More technical:
Don't ever reduce the rate of accel. within the casting stroke (deceleration) = rod straightening starts. Leaving aside angle rod-line and a few more aspects not to over complicate here.
Smooth accel. = no too sudden increase in force application.
Cheers
Bernd

[Walter]

Yes Walter, but I’m not convinced the conclusion that approximation of constant acceleration is what we want either. We might have something like that appearing on a short cast if that’s what you are expecting, but the argument I’ve heard put is that increasing acceleration will cause a tail. However that’s exactly what we have at the start of the stroke to avoid one.

Cheers, Paul

But the FFI definition of stroke is "movement of the rod sufficient to form a loop" so the stroke doesn't start until you have accelerated the line to the point where it would form a loop if/when you stop the rod. In the frictionless world that would happen as soon as you got the line moving. In the real world there would be some threshold where the line has enough speed/momentum to continue moving forward after the rod stops pulling it. My guess is that the threshold is considered to where the acceleration appears to be constant on the casting analyzer charts. The nonlinear part at the beginning would be considered as not part of the casting stroke because it isn't sufficient to form a loop.

It's not something I find particularly satisfying because it means if I stop the rod at exactly the threshold then I have a zero-length casting stroke.

[Paul Arden]

Yes that was an idea I spun out there, a very long time ago. I can’t believe they ran with that. It doesn’t work because how do you know when a stroke has begun, and why is it always later that the caster thinks he has started his stroke?

I abandoned it and the FFI ran with it. Isn’t life surreal sometimes. Life is full of mysteries Walter

Cheers, Paul

[Lasse Karlsson]

It also means someone like me would have a longer strokelength with my accuracy stroke, than my distance stroke...

Cheers
Lasse