How Low Can You Get?

A small particle P P of mass m m is sliding on the surface of a rough hemispherical shell of radius a a , which is fixed to the horizontal ground. The coefficient of friction between the particle and the hemisphere is 1 2 \tfrac12 . The particle starts at the top T T of the sphere, travelling horizontally with initial speed a ω a\omega , where ω > 0 \omega > 0 . There is a constant gravitational force of m g mg acting vertically downwards on the particle. One of three things might happen:

  • the particle slides on the hemisphere until it comes to rest a height h > 0 h > 0 above the ground,
  • the particle slides on the hemisphere until it reaches a height h > 0 h > 0 above the ground, at which point it leaves the surface of the hemisphere,
  • the particle slides on the hemisphere all the way until it reaches the ground, keeping in contact with the hemisphere at all times, reaching a height h = 0 h=0 above the ground.

depending on the value of ω \omega . In each case, h h is the lowest height above the ground attained by the particle while still in contact with the hemisphere. What is the infimum of the value of h h as ω \omega varies? Your answer will be k a k a for some constant k 0 k \ge 0 ; give the value of k k .


The answer is 0.48484014.

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.

2 solutions

Steven Chase
Jun 26, 2020

I took an empirical approach, sweeping the initial angular velocity over a range. For each initial angular velocity, a time-domain simulation was run until one of the three termination conditions was reached. With a radius of 1 1 , I ended up with a minimum height of h m i n 0.485 h_{min} \approx 0.485 occurring at initial angular speed ω 0 1.426 \omega_0 \approx 1.426 . The angle θ \theta is taken with respect to the vertical.

Below is a plot of initial angular speed vs the minimum height. If the particle has very little speed initially, it doesn't travel far along the surface, and ends up with a height close to the radius of the sphere. If the particle has very high speed initially, it more or less immediately flies off, resulting in a large final height. Only when the initial speed is at an intermediate value is the final height minimized.

 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
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
import math

R = 1.0    # sphere radius
m = 1.0   # particle mass
g = 10.0  # gravity
u = 0.5   # friction coeff

dw0 = 0.001  # increment in initial angular speed
dt = 10.0**(-4.0)  # time step

####################################

w0 = dw0              # w0 is initial angular speed

hmin = 9999999.0
w0_store = 0.0

while w0 <= 10.0:    # evaluate w0 values over a range
                    # for each w0, run the simulation.........
                    # until one of three termination conditions is satisfied

                    # Initialize time domain simulation
    t = 0.0

    S = 0.0         # Circumferential distance covered
    Sd = R*w0
    Sdd = 0.0

    theta = 0.0     # angle with the vertical
    thetad = Sd/R

    # Initialize termination flags to zero

    flag1 = 0   # set high when particle comes to rest
    flag2 = 0   # set high when particle leaves surface
    flag3 = 0   # set high when particle reaches ground without leaving surface 

    while (flag1 == 0) and (flag2 == 0) and (flag3 == 0):

        S = S + Sd*dt         # numerical integration
        Sd = Sd + Sdd*dt

        theta = theta + thetad*dt
        thetad = Sd/R

        Fc = m*(Sd**2.0)/R                # centripetal force
        # Fc = m*g*math.cos(theta) - N

        N = m*g*math.cos(theta) - Fc      # normal force

        F = m*g*math.sin(theta) - u*N      # accelerating force along surface

        Sdd = F/m                        # acceleration along surface

        # Evaluate termination flags

        if Sd <= 0.0:     # come to rest on surface
            flag1 = 1

        if N <= 0.0:      # leave surface
            flag2 = 1

        if theta >= math.pi/2.0:   # reach ground without leaving surface
            flag3 = 1

        t = t + dt

    h = R*math.cos(theta)      # final height upon termination of time-domain simulation

    if h < hmin:               # store hmin and corresponding w0
        hmin = h
        w0_store = w0

    w0 = w0 + dw0

####################################

print dw0
print dt
print ""
print hmin
print w0_store

#>>> 
#0.001
#0.0001

#0.485019792659
#1.426
#>>> 

@Steven Chase sir nice solution, I just upvoted.
BTW what is the meaning of ‘infimum’ in the problem?

A Former Brilliant Member - 11 months, 2 weeks ago

Log in to reply

Thanks. Basically, it means "minimum"

Steven Chase - 11 months, 2 weeks ago

Log in to reply

@Steven Chase As I predicted while solving.

A Former Brilliant Member - 11 months, 2 weeks ago

Log in to reply

@A Former Brilliant Member Another phrase for infimum is greatest lower bound. The infimum of a set is the largest number less than or equal to all members of that set. Thus the infimum of the set of strictly positive reals is 0 0 , but of course that set has no minimum.

In this case, we cannot actually achieve a lowest height of a cos θ 1 a\cos\theta_1 , since if we use the value α 0 \alpha_0 , the particle stops sooner. However, we can achieve any height greater than a cos θ 1 a\cos\theta_1 .

Mark Hennings - 11 months, 2 weeks ago

Log in to reply

@Mark Hennings @Mark Hennings Thankyou sir

A Former Brilliant Member - 11 months, 2 weeks ago
Mark Hennings
Jun 26, 2020

Let θ \theta be the angle T O P \angle TOP . While the particle is in contact with the hemisphere and still moving, the particle will experience a normal reaction m X mX and a frictional force 1 2 m X \tfrac12mX as indicated in the diagram (note that the motion is, essentially, two-dimensional). We note that θ = 0 \theta = 0 and θ ˙ = ω \dot{\theta} = \omega at time t = 0 t = 0 . For convenience, we introduce the dimensionless parameter α > 1 \alpha > 1 , where a ω 2 = 1 2 ( α 1 ) g a\omega^2 = \tfrac12(\alpha-1)g .

The position vector, velocity and acceleration of the particle are r = a ( sin θ cos θ ) r ˙ = a θ ˙ ( cos θ sin θ ) r ¨ = a θ ¨ ( cos θ sin θ ) a θ ˙ 2 ( sin θ cos θ ) \mathbf{r} \; = \; a{\sin\theta \choose \cos\theta} \hspace{1cm} \mathbf{\dot{r}} \; = \; a\dot{\theta}{\cos\theta \choose -\sin\theta} \hspace{1cm} \mathbf{\ddot{r}} = a\ddot{\theta}{\cos\theta \choose -\sin\theta} - a\dot{\theta}^2{\sin\theta \choose \cos\theta} so we get the equation of motion m a θ ¨ ( cos θ sin θ ) m a θ ˙ 2 ( sin θ cos θ ) = m X ( sin θ cos θ ) 1 2 m X ( cos θ sin θ ) + m g ( 0 1 ) ma\ddot{\theta}{\cos\theta \choose -\sin\theta} - ma\dot{\theta}^2{\sin\theta \choose \cos\theta} \; = \; mX{\sin\theta \choose \cos\theta} - \tfrac12mX{\cos\theta \choose -\sin\theta} + mg{0 \choose -1} Taking radial and transverse components, we obtain a θ ¨ = 1 2 X + g sin θ a θ ˙ 2 = X g cos θ a\ddot{\theta} \; = \; -\tfrac12X + g\sin\theta \hspace{2cm} -a\dot{\theta}^2 \; = \; X - g\cos\theta Eliminating X X from these equations, we obtain a ( 2 θ ¨ a θ ˙ 2 ) = g ( 2 sin θ cos θ ) d d θ [ a θ ˙ 2 e θ ] = g ( 2 sin θ cos θ ) e θ a θ ˙ 2 e θ = A 1 2 g ( 3 sin θ + cos θ ) e θ \begin{aligned} a(2\ddot{\theta} - a\dot{\theta}^2) & = \; g(2\sin\theta - \cos\theta) \\ \tfrac{d}{d\theta}\left[a\dot{\theta}^2e^{-\theta}\right] & = \; g(2\sin\theta - \cos\theta)e^{-\theta} \\ a\dot{\theta}^2e^{-\theta} & = \; A - \tfrac12g(3\sin\theta + \cos\theta)e^{-\theta} \end{aligned} for some constant A A . Applying our initial conditions, it follows that A = 1 2 g + a ω 2 = 1 2 g α A = \tfrac12g + a\omega^2 = \tfrac12g\alpha , so we deduce that a θ ˙ 2 e θ = 1 2 g [ α ( 3 sin θ + cos θ ) e θ ] X e θ = 1 2 g [ 3 ( sin θ + cos θ ) e θ α ] \begin{aligned} a\dot{\theta}^2e^{-\theta} & = \; \tfrac12g\left[\alpha - (3\sin\theta + \cos\theta)e^{-\theta}\right] \\ Xe^{-\theta} & = \; \tfrac12g\left[3(\sin\theta + \cos\theta)e^{-\theta} - \alpha\right] \end{aligned} Thus motion continues provided that θ ˙ > 0 \dot{\theta} > 0 and X > 0 X > 0 , so provided that f ( θ ) < α < g ( θ ) f(\theta) < \alpha < g(\theta) , where f ( θ ) = ( 3 sin θ + cos θ ) e θ g ( θ ) = 3 ( sin θ + cos θ ) e θ f(\theta) \; = \; (3\sin\theta + \cos\theta)e^{-\theta} \hspace{2cm} g(\theta) \; = \; 3(\sin\theta + \cos\theta)e^{-\theta} Even if it were true that θ ˙ > 0 \dot{\theta} > 0 for all 0 θ 1 2 π 0 \le \theta \le \tfrac12\pi , the fact that X = a θ ˙ 2 < 0 X = -a\dot{\theta}^2 < 0 when θ = 1 2 π \theta = \tfrac12\pi means that X X will become zero, and so the particle will leave the surface, before it reaches the ground. Thus the third option given in the question never occurs. To make the final height h h as small as possible, we want to be able to reach as large a value of θ \theta as possible.

We note that f ( θ ) < g ( θ ) f(\theta) < g(\theta) for all 0 < θ < 1 2 π 0 < \theta < \tfrac12\pi . Moreover f ( θ ) = 2 ( 2 cos θ sin θ ) e θ f'(\theta) = 2(2\cos \theta - \sin \theta)e^{-\theta} , and so f ( θ ) f(\theta) reaches a maximum at θ = θ 0 = tan 1 1 2 \theta = \theta_0 = \tan^{-1}\tfrac12 . On the other hand g ( θ ) = 6 sin θ e θ g'(\theta) = -6\sin \theta e^{-\theta} , and so g g is strictly decreasing on ( 0 , 1 2 π ) (0,\tfrac12\pi) . Define α 0 = f ( θ 0 ) 1.40645358 \alpha_0 = f(\theta_0) \approx 1.40645358 . If 1 < α α 0 1 < \alpha \le \alpha_0 , then the particle comes to rest on the hemisphere, and the maximum value of θ \theta is achieved is the solution of f ( θ ) = α f(\theta) = \alpha that is less than or equal to θ 0 \theta_0 . Thus no value of θ \theta greater than θ 0 \theta_0 is possible in this case. If α 0 < α < 3 \alpha_0 < \alpha < 3 then the particle does not stop, but eventually leaves the hemisphere when θ = g 1 ( α ) \theta = g^{-1}(\alpha) . If α 3 \alpha \ge 3 then the particle leaves the hemisphere immediately, and so the largest value of θ \theta that is achieved is 0 0 . Thus the supremum of the values of θ \theta that can be achieved is θ 1 = g 1 ( α 0 ) 1.06461595 \theta_1 = g^{-1}(\alpha_0) \approx 1.06461595 . Thus the infimum of the final height above ground that can be achieved is a cos θ 1 a\cos \theta_1 , so that k = cos θ 1 = 0.48484014 k = \cos\theta_1 = \boxed{0.48484014} .

@Mark Hennings Good problem
I also solved this way only.
Sir can you upload more problems like this ?

A Former Brilliant Member - 11 months, 2 weeks ago

Indeed, a nice problem. Thanks for posting

Karan Chatrath - 11 months, 2 weeks ago

Log in to reply

@Karan Chatrath Sir which method you follow for this problem
MATLAB Or Analytical?

A Former Brilliant Member - 11 months, 2 weeks ago

Log in to reply

I performed the theory, and then used Mathematica to do the numerical calculations...

Mark Hennings - 11 months, 2 weeks ago

Log in to reply

@Mark Hennings @Mark Hennings sir is it free?? Does it work in mobile and computer both??
Can you show your work in Mathematics?
Thanks in advance.

A Former Brilliant Member - 11 months, 2 weeks ago

Log in to reply

@A Former Brilliant Member Mathematica is not free, but Wolfram Alpha is. I only used it to calculate the value of θ 1 \theta_1 , since I know exact expressions for θ 0 = tan 1 1 2 \theta_0 = \tan^{-1}\tfrac12 and α 0 = 5 e θ 0 \alpha_0 = \sqrt{5}e^{-\theta_0} . Having defined the function g ( t ) g(t) and calculated α 0 \alpha_0 , all I had to type was

 FindInstance[{g[t] == alpha0,0 < t < pi/2},t]

to get the desired value of θ 1 \theta_1 . In Wolfram Alpha you would simply type in the explicit formulae for g g and α 0 \alpha_0 inside the FindInstance command.

Mark Hennings - 11 months, 2 weeks ago

Log in to reply

@Mark Hennings @Mark Hennings sir do you have purchased mathematics???
I checked it on website.
Sir i am just asking you have purchases it for what pupose Industry, education, Home and hobby???
I also wants to purchase??
Is it good for me to buy??
Thanks in advance.
Hope I am not disturbing you.

A Former Brilliant Member - 11 months, 2 weeks ago

Log in to reply

@A Former Brilliant Member As a teacher, I have an educational licence. Even so, it is quite expensive, but it is much better than WA.

Mark Hennings - 11 months, 2 weeks ago

Log in to reply

@Mark Hennings @Mark Hennings is it 533 pounds per year??? I am a school student , for me it is 9 US dollar.
Sir in your opinion , will it help?
Can you tell me its features?

A Former Brilliant Member - 11 months, 2 weeks ago

Log in to reply

@A Former Brilliant Member To be honest, it will probably not be as much help to a school student as a good graphical calculator - you won't be able to use it in exams! If you want to spend $7 for your own fun, that is up to you. I do not spend anything like £533 a year on it!

Mark Hennings - 11 months, 2 weeks ago

Log in to reply

@Mark Hennings @Mark Hennings yes sir it will be like a cheating in exam. But can I get solution of hard math questions, I have used wolfram, is it similar with that or have some extra features?
How much do you spend on mathematica in one year?
Thank in advance.

A Former Brilliant Member - 11 months, 2 weeks ago

@Mark Hennings can you help me in solving this equations
Thanks in advance.

A Former Brilliant Member - 11 months, 2 weeks ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...