Graphical Mechanics

A particle of mass m m is released from origin on a wedge shaped as y = x 4 y = -x^4 and given a very slight push to the right . After falling through height h h , it loses contact with the wedge. Find h h .

If h = a b where a , b N and gcd ( a , b ) = 1 \displaystyle h = \frac{a}{b} \text{ where } a, b \in \N \text{ and } \gcd(a, b) = 1 , Enter answer as a + b a + b .


Details and Assumptions:

  • Take acceleration due to gravity g = 10 m / s 2 \color{#3D99F6}{g = 10m/s^2}
  • Very slight push means initial velocity u 0 \color{#3D99F6}{u \approx 0}
  • All surfaces are smooth \color{#3D99F6}{\text{smooth}}

Inspiration Aniket Sanghi


All of my problems are original


The answer is 5.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

4 solutions

Aryan Sanghi
Aug 17, 2020

Applying conservation of energy

1 2 m v 2 = m g h \frac12mv^2 = mgh v = 2 g h ( 1 ) \boxed{v = \sqrt{2gh}} \ldots (1)


Radius of curvature R R after falling through height h h of y = x 4 y = -x^4

R = ( 1 + ( d y d x ) 2 ) d 2 y d x 2 3 2 R = \frac{\bigg(1 + (\frac{dy}{dx})^2\bigg)}{\frac{d^2y}{dx^2}}^\frac32

R = ( 1 + 16 x 6 ) 12 x 2 3 2 ( 2 ) \boxed{R = -\frac{(1 + 16x^6)}{12x^2}^\frac32} \ldots (2)


Slope after falling through height h h is

tan θ = d y d x \tan \theta = \frac{dy}{dx} tan θ = 4 x 3 \tan \theta = -4x^3 cos θ = 1 1 + 16 x 6 ( 3 ) \boxed{\cos \theta = -\frac{1}{\sqrt{1 + 16x^6}}} \ldots (3)


FBD after falling through height h h if Normal reaction = N = N is

m g cos θ N = m v 2 R mg\cos \theta - N = \frac{mv^2}{R}

When particle loses contact with wedge, N = 0 N = 0

m g cos θ = m v 2 R mg\cos \theta = \frac{mv^2}{R}

g cos θ = 2 g h R ( by (1) ) g\cos \theta= \frac{2gh}{R}\ldots (\text{by (1)})

h = R cos θ 2 h = \frac{R\cos \theta}{2}

h = ( 1 + 16 x 6 ) 12 x 2 3 2 . 1 1 + 16 x 6 2 ( by ( 2 ) and ( 3 ) ) h = \frac{-\frac{(1 + 16x^6)}{12x^2}^\frac{3}{2}.-\frac{1}{\sqrt{1 + 16x^6}}}{2} \ldots (\text{by } (2) \text{ and } (3)) h = 1 + 16 x 6 24 x 2 h = \frac{1 + 16x^6}{24x^2} From equation y = x 4 , when y = h , x = h 1 4 \text{ From equation }y = -x^4, \text{ when } y = -h, x = h^{\frac14} h = 1 + 16 h 6 4 24 h 2 4 h = \frac{1 + 16h^{\frac64}}{24h^{\frac24}} 24 h 3 2 = 16 h 3 2 + 1 24h^{\frac32} = 16h^{\frac32} + 1 8 h 3 2 = 1 8h^{\frac32} = 1 h 3 2 = 1 8 h^{\frac32} = \frac18 h = 1 4 \color{#3D99F6}{\boxed{h = \frac14}}

Therefore a = 1 , b = 4 , a + b = 5 \text{Therefore } a = 1, b = 4, a + b = 5

@Aryan Sanghi .go for my latest discussion

SRIJAN Singh - 9 months, 4 weeks ago

Log in to reply

Hey thanks :) I'll try it right away.

Krishna Karthik - 9 months, 2 weeks ago

I have to ask a question; is the spring's natural length 2 meters or 2 times the mass?

Krishna Karthik - 9 months, 2 weeks ago
Steven Chase
Aug 17, 2020

Here is the math I used in conjunction with a numerical simulation. Derive an acceleration constraint equation:

y = x 4 y ˙ = 4 x 3 x ˙ y ¨ = 4 x 3 x ¨ 12 x 2 x ˙ 2 y = -x^4 \\ \dot{y} = -4 x^3 \dot{x} \\ \ddot{y} = -4 x^3 \ddot{x} - 12 x^2 \dot{x}^2

Let N N be the magnitude of the normal force, and let u \vec{u} be a unit vector in the direction normal to the curve. v \vec{v} is a non-unitized version of u \vec{u} .

v = ( v x , v y ) = ( 4 x 3 , 1 ) u = v v \vec{v} = (v_x, v_y) = (4 x^3, 1) \\ \vec{u} = \frac{\vec{v}}{|\vec{v}|}

Write the Newton's Second Law equations:

m x ¨ = N u x m y ¨ = N u y m g m \ddot{x} = N u_x \\ m \ddot{y} = N u_y - m g

Substitute these two equations into the acceleration constraint equation, yielding:

N = m g 12 m x 2 x ˙ 2 u y + 4 x 3 u x N = \frac{m g - 12 m x^2 \dot{x}^2}{u_y + 4 x^3 u_x}

Then substitute N N back into the Newton's Second Law equations to solve for the accelerations. Numerical integration takes care of the rest. Run the simulation until the normal force becomes negative for the first time. This is the time at which the particle loses contact with the curve. The particle falls through a height h = 1 4 h = \frac{1}{4} before this happens.

Excellent solution sir. Thanku for sharing it with us.

Aryan Sanghi - 9 months, 4 weeks ago

Log in to reply

Thanks. Fun problem too

Steven Chase - 9 months, 4 weeks ago

Log in to reply

Ohk. Thanku for compliment sir. :)

Aryan Sanghi - 9 months, 4 weeks ago

@Steven Chase Your solution is smallest among all solution still you have 7 upvotes. You are a very powerful guy.

Talulah Riley - 9 months, 4 weeks ago

Log in to reply

He is one of brilliant's top members. Famous guy indeed.

Krishna Karthik - 9 months, 3 weeks ago
Karan Chatrath
Aug 17, 2020

I have solved it exactly the same way as @Steven Chase has although my nomenclature and terminology is a bit different. One may refer to his solution for details. I am attaching the simulation code here for anyone who is interested.

A plot of the trajectory of the particle is as follows. The simulation runs till the particle loses contact with the wedge. The 0.25 -0.25 y coordinate can be clearly spotted in the plots.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
clear all
clc

% Acceleration due to gravity:
g         = 10;

% Initial conditions position:
x(1)      = 0.05;
y(1)      = -x(1)^4;

% Initial conditions velocity:
dx(1)     = 0;
dy(1)     = 0;

% Time initialisation:
t(1)      = 0;
dt        = 1e-5;

% Reaction force initialisation:
N(1)      = g;

% Loop variable initialisation:
k         = 1;


% While loop until the normal reaction becomes zero:
while N(k) >= 0

    % Sine term in Equation of motion:
    S         = (4*x(k)^3)/(sqrt(1 + 16*x(k)^6));

    % Cosine term in Equation of motion:
    C         = (1)/(sqrt(1 + 16*x(k)^6));

    % Linear system of equations solving for N and acceleration components:
    A         = [1 0 -S;0 1 -C;4*x(k)^3 1 0];
    b         = [0;-g;-12*x(k)^2*dx(k)^2];

    % Matrix inversion:
    S         = inv(A)*b;
    ddx       = S(1);
    ddy       = S(2);
    N(k+1)    = S(3);


    % Numerical integration:
    dx(k+1)   = dx(k) + dt*ddx;
    dy(k+1)   = dy(k) + dt*ddy;
    x(k+1)    = x(k)  + dt*dx(k+1);
    y(k+1)    = y(k)  + dt*dy(k+1);
    t(k+1)    = t(k)  + dt;

    k         = k + 1;
end

Answer = y(end)

This "acceleration constraint equation" method is generally just about as good as Lagrangian mechanics, it seems. Probably even better, since it can handle non-conservative forces too

Steven Chase - 9 months, 4 weeks ago

Log in to reply

Moreover, in contrast to Lagrangian mechanics, this method allows one to compute reaction forces. However, I don't think this that this constraint equation approach is as robust as LM. This is because, as you numerically integrate for a long enough period of time, errors accumulate to a point where the constraint equation gets violated. In the case of LM, this does not happen as the constraint is very naturally accounted for, in the formulation. So in many commercial software packages that simulate rigid body dynamics and rely on the 'constraint equation' based modelling approach, the numerical integrator is combined with some sort of error minimisation optimisation scheme for the constraint equations. I read this in a source that I cannot recall.

Karan Chatrath - 9 months, 4 weeks ago

Log in to reply

I have not tested the above claim myself but the illustrative example that the source used involved the simulation of a chaotic system (double pendulum). It was seen that over time, the links break apart. One can imagine that the small variations between successive iterations can result in large errors over time.

Karan Chatrath - 9 months, 4 weeks ago

Log in to reply

@Karan Chatrath Yes, I think that is certainly right. In fact, when I use this Newtonian method, I always keep track of the maximum value of a "residual" quantity. For example, for a sphere, the residual would be x 2 + y 2 + z 2 R 2 | x^2 + y^2 + z^2 - R^2| . I only have confidence in the answer if the residual is very small at the end of the simulation. So Newton allows for computations with non-conservative forces, but Lagrange ends up being better for long-duration simulations.

Steven Chase - 9 months, 4 weeks ago

Log in to reply

@Steven Chase Precisely. But in order to have small residuals, it wouldn't be surprising if you used smaller time steps.

Karan Chatrath - 9 months, 4 weeks ago

Log in to reply

@Karan Chatrath Have you solved the follow-up to this? It doesn't like my answer

Steven Chase - 9 months, 3 weeks ago

Log in to reply

@Steven Chase Just solved it. And I get the expected answer.

Karan Chatrath - 9 months, 3 weeks ago

Nice one. I used Matlab to solve as well; this is a hard problem.

Krishna Karthik - 9 months, 4 weeks ago

Log in to reply

@Krishna Karthik Do you have purchased MATLAB?

Talulah Riley - 9 months, 4 weeks ago

Log in to reply

I'm on free trial at the moment; but when the free trial expires, I intend to purchase a license.

Krishna Karthik - 9 months, 4 weeks ago

Python's free, but I like Matlab a lot.

Krishna Karthik - 9 months, 4 weeks ago

Excellent solution sir. Thanku for sharing it with us. :)

Aryan Sanghi - 9 months, 4 weeks ago
Mark Hennings
Aug 18, 2020

The particle's position vector, velocity and acceleration are r = ( x x 4 ) r ˙ = x ˙ ( 1 4 x 3 ) r ¨ = x ¨ ( 1 4 x 3 ) x ˙ 2 ( 0 12 x 2 ) \mathbf{r} = \binom{x}{-x^4} \hspace{1cm} \dot{\mathbf{r}} = \dot{x} \binom{1}{-4x^3} \hspace{1cm} \ddot{\mathbf{r}} = \ddot{x}\binom{1}{-4x^3} - \dot{x}^2 \binom{0}{12x^2} and so the equation of motion of the particle is m x ¨ ( 1 4 x 3 ) m x ˙ 2 ( 0 12 x 2 ) = m R ( 4 x 3 1 ) + m g ( 0 1 ) m\ddot{x}\binom{1}{-4x^3} - m\dot{x}^2 \binom{0}{12x^2} \; = \; mR\binom{4x^3}{1} + mg\binom{0}{-1} Here m R ( 4 x 3 1 ) mR\binom{4x^3}{1} is the normal reaction. Taking the scalar product of this equation with ( 4 x 3 1 ) \binom{4x^3}{1} gives 12 m x 2 x ˙ 2 = m R ( 1 + 16 x 6 ) m g -12mx^2\dot{x}^2 \; = \; mR(1 + 16x^6) - mg and hence R = g 12 x 2 x ˙ 2 1 + 16 x 2 R \; =\; \frac{g - 12x^2 \dot{x}^2}{1 + 16x^2} Conservation of energy gives 1 2 m ( 1 + 16 x 6 ) x ˙ 2 m g x 4 = 0 \tfrac12m(1 + 16x^6)\dot{x}^2 - mgx^4 \; = \; 0 and hence R = g ( 1 8 x 6 ) ( 1 + 16 x 6 ) 2 R \; =\; \frac{g(1 - 8x^6)}{(1 + 16x^6)^2} Thus the particle leaves the surface when R = 0 R=0 , so when x 6 = 1 8 x^6 = \tfrac18 , or x 2 = 1 2 x^2 = \tfrac12 , and and hence when h = 1 4 h=\tfrac14 . This makes the answer 1 + 4 = 5 1+4=\boxed{5} .

Excellent solution sir. Thanku for sharing it with us. :)

Aryan Sanghi - 9 months, 3 weeks ago

@Mark Hennings try this problem sir Moving Wedge

Aryan Sanghi - 9 months, 1 week ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...