- #71
e2m2a
- 359
- 14
K^2 said:Now, what do you have to add to all of this?
I admit I have been wrong about Lagrangian analysis. After seeing your derivations, I have more respect for it and I am beginning to appreciate its beauty and power in analyzing systems. And, if you are absolutely certain your derivation of the angular velocity is constant, then I admit the Lagrangian does yield exact solutions for the equations of motion. I can solve simple Lagrangian problems ( I had to teach myself) concerning pendulums and masses on inclined planes, etc. , but the advanced analysis you have done with Lagrangian multipliers and constraints is beyond my experience right now. I could learn it if I took the time, but I am busy with other things. The reason I doubted an exact parametric equation was possible is based on my experience with this experiment. I have actually been working on it for about 4 years. I have done different versions of it, and always the angular velocity was not constant. I struggled with deciding if this was due to friction or some other fundamental law. If your analysis is correct, then you have answered the question for me and I thank you for that.
I thought I would point out some typos and other errors in your analysis. Under section 4, where you express the Lagrangian with the lambda multipliers, the last term reads lambda33. I think you mean lambda3f3: Under section 6, where you take the first derivative of y2, your first term on the right of the equality is x dot sub y. I think you mean y dot sub 1. In the same section at the end, I think your lambda1 equation is inverted, and I think you should not have an r factor left in the numerator. (I say this because the dimensional analysis of this equation does not make sense.) If you check on the previous page, where you have the two equations without the rigorous proof, I think you’ll see the algebraic error. Since, you have asked me about how I could demonstrate that centrifugal force has anything to do with the increase in speed of the center of mass, and since you have given me the full Lagrangian analysis, I am now going to give you my full explanation of how centrifugal force accelerates the speed of the center of mass. After seeing your Lagrangian analysis I was pleasantly surprised to see something that I think you may find interesting also. I will talk about it at the end.
Bear with me if I seem to go off on tangents with stories, but it is one of the ways I explain or teach a concept. Suppose an individual visits a science museum and after seeing all the standard demonstrations, Foucault’s pendulum, etc., he runs across an exhibit that has a rotating body pivoted at one end inside a glass box that has a near-vacuum inside. The sign on the exhibit states this body has been rotating for 11 years, and has been maintaining a constant angular velocity without the aid of a motor, and rotates solely by its own rotational inertia around high-tech magnetic bearings. The sign further states this exhibit demonstrates Newton’s first law as pertaining to rotation: a body in rotation stays in rotation, unless acted upon by a torque. The individual is intrigued by this. Not having a physics background, the individual wonders how the body could do this without something continuously pushing it. He did read somewhere years ago that the Greek philosopher Aristotle believed that the natural state of bodies was to stay at rest, not to move. In fact Aristotle taught the reason why an arrow continues to fly through the air is because “something” keeps pushing it to maintain its flight. At this time a museum guide notices the individual is perplexed and asks if he has any questions? The individual responds that he does, and the guide starts answering them. The individual asks what force keeps the body rotating? The guide answers its on rotational inertia, there is no outside force necessary to keep it rotating, only an internal force to change its direction. What causes it to change direction, he asks? Centripetal force, the guide answers. It acts perpendicular to the tangential motion of the body, and therefore, only changes the direction of the body, not its speed. Its tangential speed, he continues, is 10 meters per second. Will it always be 10 meters per second, the individual asks? Yes, the guide answers, unless you have relative motion with respect to the body. What do you mean, the individual asks? Wait until the body is at the 6 o’clock position, the guide answers, then quickly step back with continual motion and observe what the speed is at the 3 o’clock position. What has happened to the speed? The individual steps back while still moving and observes the body appears to have gained speed. I guess its going faster when I move back, the individual observes. Right, the guide responds.
Why does it speed up? Has the rotation of the body changed in any way? The individual thinks about it for a moment. No, you said the body by its own inertia continues to rotate on its own at a constant speed, and I don’t see how any force could act on it just by my stepping back. Right, the guide answers. The reason it speeds up is because, relative to you, the rotator at its 3 o’clock position has gained an additional speed, pointing to the right because you are moving to the left. In fact, if you recorded the motion of the body with a high-speed video camera while you moved to the left, and then played it back at slow motion, you would see the body inscribe an arc of a circle from 9 o’clock to 6 o’clock, but the shape of the path would “stretch” out from 6 o’clock to 3 o’clock for the same interval of time. Then measuring the two different paths, you would find the post-6 o’clock path is longer than the pre-6 o’clock path, demonstrating the body has gained in total speed. The individual thinks about it. First he observes the motion of the body without moving back and notices the speed at 3 o’clock does not change. He then moves back beginning at 6 o’clock, and notes the total speed of the body speeds up. He does it over and over again. No movement back, speed is constant. Move back, speed increases. He concludes correctly, its only due to relative motion of the body to the right at the 3 o’clock position that the body has sped up. Now, I think you get the point I’m making. The reason why the center of mass of the rotator-slider system increases is because of the gain in speed of the slider to the right. Period. By you own derivation, since the rotational velocity is constant, there is no other factor necessary to explain the increase in speed of the center of mass. The center of mass has simply acquired an additional velocity to the right when it is at its 3 o’clock position due to the speed of the slider to the right. Furthermore, there is no need to find an external force, such as a rail constraint force, acting on the center of mass of the system to account for the increase in speed of the center of mass because it is only the increase in speed of the system to the right that accounts for the overall gain in speed of the center of mass as demonstrated in my “story”. Surely, this constraint force from the rail, could not act to the right on the center of mass of the system, since this normal force can only act perpendicular to the motion of the slider. By the way, concerning constraint forces, I looked it up, and everything that I read states that the constraint forces used in Lagrangian analysis cannot do virtual work on a body because there is no real or virtual displacement associated with them. More on this later. So here is the crux of the matter concerning the gain in speed of the center of mass. It is due to and only due to the increase in speed of the slider to the right, which speed is added to the speed of the center of mass at the 3 o’clock position. But now, we do have to find a force that is responsible for the acceleration of the slider to the right in the positive x-direction. What causes the slider’s speed to increase to the right? The x-component of the centrifugal force. This is where the inertial force steps in. It never acts in the positive y-direction from 6 o’clock to 3 o’clock. It doesn’t have to. It only has to act in such a way to increase the speed of the slider to the right, and there is an x-component of the inertial centrifugal force that acts in the positive x-direction as the rotator rotates from 6 o’clock to 3 o’clock. Now I am going to show a mathematical derivation how this happens and how this exactly matches your Lagrangian analysis.
To do this, I am going to use my hanging-weight, swinging-weight explanation. Let's begin with a simple high-school experiment. Imagine two bodies. Both are connected by a wire. One body hangs over the edge of a table. We call this body m2. The other body initially rests on a rail on top of the table. We call this body m1. At some point in time we release the hanging body, and the whole two-body system begins to accelerate as the hanging mass accelerates downward. Simultaneously, the speed of the center of mass of the system increases. Obviously, all of this is due to the weight acting on m2, the overhanging mass. Now every physicist on this planet has no problem in understanding the mechanism of this motion and accepting weight as the “real” force that causes it. But now watch what happens next. Imagine this two-body system is a Dali surreal painting. We “morph” the overhanging mass so that it transforms into the rotator, and instead of hanging down, its “lifted” up, so that it can rotate in a plane parallel to the table. As it rotates, an inertial force manifests in the center of mass of the rotator. This inertial force arises by the same mechanism that the weight manifests in the overhanging system – the forces arises from a departure of the bodies from their natural path in spacetime. That is, weight manifests when a body departs from its locally-curved geodesic in curved spacetime; Inertial force manifests when a body departs from its locally-straight line geodesic in un-curved spacetime. Einstein made a big deal about this equivalence between inertial force and gravitational weight, which led to his strong and weak principle of equivalence and to his theory of general relativity. For Einstein, there really was no difference between gravitational weight and inertial force. Yet, it fell on deaf ears then, and it still falls on deaf ears today. But now I will apply this equivalence literally, and show how it predicts the motion of the rotator-slider system. Instead of thinking of centrifugal inertial force acting on the rotating rotator, just replace this with a “weight force”, acting in an outward radial direction on the rotator. Every dynamic and kinematics motion of the system can be predicted by using this substitution.
First, we take your equation from your Lagrangian analysis. Now, at this point I am going to make a change in coordinates to simplify for me the use of the trigonometric functions. I am going to rotate the coordinate system so that the rotator-slider system moves in the positive y-direction. This change will not affect you Lagrangian analysis since it is not bound to any coordinate system or any orientation of a coordinate system. (One of the usefulness of the Lagrangian). Recall after taking the first derivative of your derived equation, we have this for the total constant speed of the center of mass of the system in the positive y-direction, for wt > 0 radians:
y`(t) = (rwm2)/(m1 + m2) equation (1).
Where, m1 is the mass of the slider and m2 is the mass of the rotator.
Now, there are two components that make up this total velocity of the center of mass of the system in the positive y-direction: a component due to the instantaneous speed of the slider in the positive y-direction and the y-component of the tangential velocity of the center of mass of the system as it rotates from 3 o’clock to 12 o’clock. Now, the y-component of the tangential velocity of the center of mass as the rotator rotates from 3 o’clock to 12 o’clock is:
vcom y = [rwm2cos(wt)]/(m1 + m2) equation (2).
Now, I am ready to use the centrifugal or weight force to find the speed of the slider in the positive y-direction. Now this weight force has a postive x-component. This force “pulls” the slider against the rails in the positive x-direction. There is a reaction, normal constraint force to this from the rail that acts on the slider in the negative x-direction. Can this force effect the system in anyway? Possibly, if you assume centrifugal force is a fictitious force and not real. Absolutely not, if you realize the centrifugal “weight” force is as real as a weight force. If we do a free-body diagram of the slider, we find that the x-component of the centrifugal weight force in the postive x-direction cancels out the normal reaction rail force acting on the slider in the negative x-direction. Since it is axiomatic that the net effect on any part of a multi-body system affects the center of mass of the system, this cancellation of forces has no effect on the center of mass. This is no different than the case where we have a body sliding down an inclined plane. There is a normal constraining force that keeps the body at an angle as it slides down. However, this constraining force is always normal to the motion of the body, so it can never do virtual work on the body. And, since we can find a component of the weight acting on the sliding body that cancels this normal force, there is no net force acting in the direction of the normal force that can do virtual work in this direction. Now, we return to the hanging weight system and find the force acting on m1, the mass of the object on the table. Without derivation, we find it is equal to:
f weight m1 = m1m2g/(m1 + m2) equation (3).
Now I “lift” this expression and apply it to the rotator system as follows:
fcenm1y = [m1m2 wsq r sin(wt)]/(m1 + m2) equation (4).
This is the y-component of the centrifugal or weight force acting on the slider in the positive y-direction. Note, I have only substituted wsq r sin(wt) for g. Now I seek the change in velocity of the slider due to this force. I can find it if I first find the change in momentum of the slider by the action of this force. This is easy if I integrate 4 with respect to time. Taking the constants out of the integral, I end up with this:
integral sin(wt) dt. equation 5
Now it turns out the indefinite integral of equation 5 is:
-1/w cos(wt) + C.
Now, taking the definite integral of equation 5 from 0/w seconds to wt/w seconds, I end up with:
-cos(wt) + cos(0 radians). Factoring in the constants taken out of the integral, I get:
py = rwm1m2[ 1 – cos wt]/(m1 + m2) equation (6).
Or:
py = rwm1m2/(m1 + m2) - rwm1m2 cos wt/(m1 + m2) equation (7)
Or, the momentum of the slider in the positive y-direction.
Finally, dividing out by m1, the mass of the slider, we get the velocity of the slider in the positive y-direction, or:
y`(t) = rwm2/(m1+m2) - rwm2 cos wt/(m1+m2) equation (8).
But alas, the second term in the above equation is just the negative of equation 2, the tangential velocity of the center of mass in the positive y-direction. As stated before, the total velocity of the center of mass of the system in the positive y-direction is the sum of the velocity of the slider and the tangential velocity of the center of mass in the positive y-direction. When we add equation 2 to equation 8, it reduces to:
y`(t) = rwm2/(m1+m2) equation (9).
Which is in exact agreement with your derived equation, equation 1, the constant speed of the center of mass of the system in the positive y-direction! All this from assuming a component of the centrifugal force acts on the slider in the positive y-direction!
Now, finally, I am ready to show you the key equation that matches your Lagrangian analysis. Look at my equation 4, the y-component of the centrifugal force acting in the positive y-direction:
fcenm1y = [m1m2 r wsq sin(wt)]/(m1 + m2) equation (4).
I am going to rotate the coordinate system back to its original orientation, changing the above to:
fcenm1x = [m1m2 r wsq cos(wt)]/(m1 + m2) equation (4).
Next, I am going to divide out the mass of the slider on both sides, yielding:
Acceleration x-direction = [m2 r wsq cos(wt)]/(m1 + m2) equation (4).
Now look at your first equation after the “Coordinates for case II” in your analysis, the equation that shows the acceleration of the slider in the positive x-direction. What do we find? They exactly match!
Do you think this is just a coincidence? Or do you think the convergence of your brilliant Lagrangian analysis with my a priori approach to the reality of centrifugal force, which yields the same exact equation for the acceleration of the slider, tells us we have found a mathematical expression that expresses the effect of a subtle but real force of nature?
Last edited: