C_R

3762 Reputation

21 Badges

6 years, 235 days

MaplePrimes Activity


These are Posts that have been published by C_R

Just an observation.

I was wondering if less obvious errors than in the below can be avoided with future versions of the AI assistant. Maybe a warning that a formula uses special Maple symbols is possible.

Formulas without dimensions are more susceptible to undetected errors.

Deflection of a circular cantilever

(a first attemp with the AI formula assistant)

_local(I)

I

(1)

AI prompt: Deflection of a circular cantilever with a  force applied at the end

Correct formular inserted ->
delta = F*L^3/(3*E*I)

delta = (1/3)*F*L^3/(E*I)

(2)

AI prompt:  Moment of inertia of a circular cross-section

Correct formular inserted ->

I = (1/4)*Pi*R^4

I = (1/4)*Pi*R^4

(3)

subs(I = (1/4)*Pi*R^4, delta = (1/3)*F*L^3/(E*I))

delta = (4/3)*F*L^3/(E*Pi*R^4)

(4)

params := R = 25*Unit('mm'), F = 200*Unit('N'), L = 1.*Unit('m'), E = 210000*Unit('N'/'mm'^2)

R = 25*Units:-Unit(mm), F = 200*Units:-Unit(N), L = 1.*Units:-Unit(m), E = 210000*Units:-Unit(N/mm^2)

(5)

subs(R = 25*Units:-Unit(mm), F = 200*Units:-Unit(N), L = 1.*Units:-Unit(m), E = 210000*Units:-Unit(N/mm^2), delta = (4/3)*F*L^3/(E*Pi*R^4))

delta = 0.1034759757e-8*Units:-Unit(N)*Units:-Unit(m)^3/(Units:-Unit(N/mm^2)*Units:-Unit(mm)^4)

(6)

simplify(%)

delta = 0.1034759757e-2*Units:-Unit(m)

(7)

NULL

The dimension of m^9 for a deflection clearly indicates an error.

A better prompt to avoid this error (caused by automatic simplification) could not be found

Download AI_formula_assistant.mw

P.S.:

This is a real example that happend to me where I did not notice the minus sign in Maples output in equation (1). The error  can easily be fixed by adding "local I" as the first statement of the document and the deflection becomes 1 mm.

MaplePrimes offers spell check and correction

for the function "Contact Author"

 

Suggestion:
For the sake of message clarity (and to save time) it would be desireable to have spell checking and correction in other MaplePrimes message bodies of as well.

 

Keywords: Intermediate axis theorem, Tennis racket theorem, Dzhanibekov effect, Coriolis force, Euler equations

In 1988 I witnesses the instability of the rotation about the intermediate axis of a foam brick.

Since then I have been fascinated by this effect. It was one of the many experiments which enriched a lecture series on kinetics and on that day Euler equations were on the agenda. Colored surfaces of the brick made it possible to observe the effect without micro gravity and slow-motion equipment.

This post is about reproducing an “intuitive” visualization of an explanation of the effect by Terry Tao from 2011 using 4 rigidly connected point masses. 8 years later the explanation was animated in a YouTube video (The Bizarre Behavior of Rotating Bodies) and considered to be the “best intuitive” explanation.

Motivated by the video, I wondered whether a similar animation with acting forces is possible with MapleSoft products and whether there might be a better intuitive explanation without the use of centrifugal forces. Initially I saw this more as a good test of MapleSim’s visualization capabilities. Finally, it took over 3 years and numerous attempts (mostly during vacation, kind of a substitute for drawing circles in the sand...) to come to a conclusion on the effect.

Intermediate_axis_theoreme_with_3_point_masses.msim


About the model:

Unlike the YouTube video, I decided to simulate 3 identical point masses because a 3-mass model fits better to a T-handle (overlayed in the animation above), video footage from space experiments and discussions in this forum (221298, 225760, 228066).

The movement of the model generates acceleration forces on each mass. The clip displays the corresponding opposing forces that act in the model (i.e. act on the massless T-structure). The blue mass, which is not perfectly centered on the axis of rotation at the start of the simulation perturbs the orbits of the red and the green masses. That was my initial intuitive attempt to explain the effect.

The 3 masses form an isosceles triangle. Here it is helpful to think of a rotating arrowhead where the shape determines stability of the rotation. The aspect ratio (the ratio of the height to the base length) of the triangle determines the stability of rotation about the mirror symmetry axis of the triangle (i.e. the symmetry axis of the T-structure). An obtuse triangle (“blunt”, aspect ratio < sqrt(3)/2) is unstable when rotating about an axis that is slightly inclined with respect to this axis of symmetry. The inclination can be in the plane of the triangle or out of plane. An acute (“pointy”) triangle only wobbles.

About the MapleSim model:

A supplementary rigid body component without mass and rotational inertia is used at the center of mass of the three masses to impose initial conditions. Rotating the triangle at the start of the simulation about the center of mass of the 3 masses prevents the triangle from drifting laterally away from its initial position. This effect of lateral drift is visible in video footage from space with the T-handle.

The rotational inertia of the other rigid body components is set to zero. Without rotational inertia it could be assumed that only Newtonian mechanics are used in the simulation (i.e. no Euler equations are integrated). This is however wrong. MapleSim generates automatically from a system with 3x6=18 coordinates a system with 3 Newtonian equations for translation and 3 Euler equations for rotation.

Forces and moments are measured with sensor components. Visualization is done with force and moment visualization components. These components are “abused” to display the following other physical quantities:

The angular momentum of the masses

The vectors of the angular velocity and the angular acceleration

Moments of the forces with respect to the center of mass

Moments of the forces with respect to the center of the base of the triangle

For a clean model, sensor components and mathematical components to calculate physical quantities are grouped in three subsystems (one per mass, indicated with a colored dot in the image below).

The model contains parameter sets for in plane and out of plane inclination of the axis of the T with respect to the initial axis of rotation (the x-axis).