Dynamics of an Inverting Tippe Top

The existing results about inversion of a tippe top (TT) establish stability of asymptotic solutions and prove inversion by using the LaSalle theorem. Dynamical behaviour of inverting solutions has only been explored numerically and with the use of certain perturbation techniques. The aim of this pa...

Повний опис

Збережено в:

Бібліографічні деталі
Дата:	2014
Автори:	Rauch-Wojciechowski, S., Rutstam, N.
Формат:	Стаття
Мова:	English
Опубліковано:	Інститут математики НАН України 2014
Назва видання:	Symmetry, Integrability and Geometry: Methods and Applications
Онлайн доступ:	http://dspace.nbuv.gov.ua/handle/123456789/146837
Теги:	Додати тег Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:	Digital Library of Periodicals of National Academy of Sciences of Ukraine
Цитувати:	Dynamics of an Inverting Tippe Top / S. Rauch-Wojciechowski, N. Rutstam // Symmetry, Integrability and Geometry: Methods and Applications. — 2014. — Т. 10. — Бібліогр.: 18 назв. — англ.

Репозитарії

Digital Library of Periodicals of National Academy of Sciences of Ukraine

id	irk-123456789-146837
record_format	dspace
spelling	irk-123456789-1468372019-02-12T01:23:35Z Dynamics of an Inverting Tippe Top Rauch-Wojciechowski, S. Rutstam, N. The existing results about inversion of a tippe top (TT) establish stability of asymptotic solutions and prove inversion by using the LaSalle theorem. Dynamical behaviour of inverting solutions has only been explored numerically and with the use of certain perturbation techniques. The aim of this paper is to provide analytical arguments showing oscillatory behaviour of TT through the use of the main equation for the TT. 2014 Article Dynamics of an Inverting Tippe Top / S. Rauch-Wojciechowski, N. Rutstam // Symmetry, Integrability and Geometry: Methods and Applications. — 2014. — Т. 10. — Бібліогр.: 18 назв. — англ. 1815-0659 2010 Mathematics Subject Classification: 70F40; 74M10; 70E18; 70E40; 37B25 DOI:10.3842/SIGMA.2014.017 http://dspace.nbuv.gov.ua/handle/123456789/146837 en Symmetry, Integrability and Geometry: Methods and Applications Інститут математики НАН України
institution	Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection	DSpace DC
language	English
description	The existing results about inversion of a tippe top (TT) establish stability of asymptotic solutions and prove inversion by using the LaSalle theorem. Dynamical behaviour of inverting solutions has only been explored numerically and with the use of certain perturbation techniques. The aim of this paper is to provide analytical arguments showing oscillatory behaviour of TT through the use of the main equation for the TT.
format	Article
author	Rauch-Wojciechowski, S. Rutstam, N.
spellingShingle	Rauch-Wojciechowski, S. Rutstam, N. Dynamics of an Inverting Tippe Top Symmetry, Integrability and Geometry: Methods and Applications
author_facet	Rauch-Wojciechowski, S. Rutstam, N.
author_sort	Rauch-Wojciechowski, S.
title	Dynamics of an Inverting Tippe Top
title_short	Dynamics of an Inverting Tippe Top
title_full	Dynamics of an Inverting Tippe Top
title_fullStr	Dynamics of an Inverting Tippe Top
title_full_unstemmed	Dynamics of an Inverting Tippe Top
title_sort	dynamics of an inverting tippe top
publisher	Інститут математики НАН України
publishDate	2014
url	http://dspace.nbuv.gov.ua/handle/123456789/146837
citation_txt	Dynamics of an Inverting Tippe Top / S. Rauch-Wojciechowski, N. Rutstam // Symmetry, Integrability and Geometry: Methods and Applications. — 2014. — Т. 10. — Бібліогр.: 18 назв. — англ.
series	Symmetry, Integrability and Geometry: Methods and Applications
work_keys_str_mv	AT rauchwojciechowskis dynamicsofaninvertingtippetop AT rutstamn dynamicsofaninvertingtippetop
first_indexed	2025-07-11T00:44:35Z
last_indexed	2025-07-11T00:44:35Z
_version_	1837309294482030592
fulltext	Symmetry, Integrability and Geometry: Methods and Applications SIGMA 10 (2014), 017, 18 pages Dynamics of an Inverting Tippe Top Stefan RAUCH-WOJCIECHOWSKI and Nils RUTSTAM Department of Mathematics, Linköping University, Linköping, Sweden E-mail: strau@mai.liu.se, ergoroff@hotmail.com Received September 05, 2013, in final form February 18, 2014; Published online February 27, 2014 http://dx.doi.org/10.3842/SIGMA.2014.017 Abstract. The existing results about inversion of a tippe top (TT) establish stability of asymptotic solutions and prove inversion by using the LaSalle theorem. Dynamical be- haviour of inverting solutions has only been explored numerically and with the use of cer- tain perturbation techniques. The aim of this paper is to provide analytical arguments showing oscillatory behaviour of TT through the use of the main equation for the TT. The main equation describes time evolution of the inclination angle θ(t) within an effective potential V (cos θ,D(t), λ) that is deforming during the inversion. We prove here that V (cos θ,D(t), λ) has only one minimum which (if Jellett’s integral is above a threshold value λ > λthres = √ mgR3I3α(1+α) 2 √ 1+α−γ and 1 − α2 < γ = I1 I3 < 1 holds) moves during the inversion from a neighbourhood of θ = 0 to a neighbourhood of θ = π. This allows us to conclude that θ(t) is an oscillatory function. Estimates for a maximal value of the oscillation period of θ(t) are given. Key words: tippe top; rigid body; nonholonomic mechanics; integrals of motion; gliding friction 2010 Mathematics Subject Classification: 70F40; 74M10; 70E18; 70E40; 37B25 1 Introduction A tippe top (TT) is constructed as a truncated axisymmetric sphere with a small peg as its handle. The top is spun on a flat surface with the peg pointing upward. If the initial rotation is fast enough, the top will start to turn upside down until it ends up spinning on its peg. We call this interesting phenomenon an inversion. It is known that the TT inverts when the physical parameters satisfy the conditions 1−α < γ = I1 I3 < 1+α where 0 < α < 1 is the eccentricity of the center of mass and I1, I3 are the main moments of inertia. The TT and the inversion phenomenon has been studied extensively throughout the years, but the dynamics of inversion has proven to be a difficult problem. This is because even the most simplified model for the rolling and gliding TT is a non-integrable dynamical system with at least 6 degrees of freedom. The focus in many works has been on the asymptotics of the TT [1, 5, 9, 10, 13] or on numerical simulations for a TT [3, 11, 18]. In this paper we study equations of motion for a rolling and gliding TT in the case of inverting solutions and analyse dynamical properties of such solutions through the main equations for the TT [12, 14]. We study the main equation for the TT for a subset of parameters satisfying 1 − α2 < γ < 1 and 1−γ γ+α2−1 = mR2 I3 when it acquires a simpler form, which enables detailed analysis of deformation of the effective potential V (cos θ,D, λ) during the inversion. We show that, during the inversion, a minimum of the effective potential moves from the neighbourhood of θ = 0 to the neighbourhood of θ = π and therefore the inclination angle θ(t) oscillates within a nutational band that moves from the north pole to the south pole of the unit sphere S2. We give also estimates for the period of nutation of the symmetry axis. mailto:strau@mai.liu.se mailto:ergoroff@hotmail.com http://dx.doi.org/10.3842/SIGMA.2014.017 2 S. Rauch-Wojciechowski and N. Rutstam 2̂ θ O C M 3̂ ẑ AvA a αR s Figure 1. Diagram of the TT. Note that a = Rα3̂−Rẑ. 2 The tippe top model We model the TT as an axisymmetric sphere of mass m and radius R which is in instantaneous contact with the supporting plane at the point A. The center of mass CM is shifted from the geometric center O along its symmetry axis by αR, where 0 < α < 1. We choose a fixed inertial reference frame (X̂, Ŷ , Ẑ) with X̂ and Ŷ parallel to the supporting plane and with vertical Ẑ. We place the origin of this system in the supporting plane. Let (x̂, ŷ, ẑ) be a frame defined through rotation around Ẑ by an angle ϕ, where ϕ is the angle between the plane spanned by X̂ and Ẑ and the plane spanned by the points CM , O and A. The third reference frame (1̂, 2̂, 3̂), with origin at CM , is defined by rotating (x̂, ŷ, ẑ) by an angle θ around ŷ. Thus 3̂ will be parallel to the symmetry axis, and θ will be the angle between ẑ and 3̂. This frame is not fully fixed in the body. The axis 2̂ points behind the plane of the picture of Fig. 1. We let s denote the position of CM w.r.t. the origin of the frame (X̂, Ŷ , Ẑ) and the vector from CM to A is a = R(α3̂ − ẑ). The orientation of the body w.r.t. the inertial reference frame (X̂, Ŷ , Ẑ) is described by the Euler angles (θ, ϕ, ψ), where ψ is the rotation angle of the sphere about the symmetry axis. With this notation, the angular velocity of the TT is ω = −ϕ̇ sin θ1̂+ θ̇2̂+ (ψ̇ + ϕ̇ cos θ)3̂, and we denote ω3 := ψ̇ + ϕ̇ cos θ. The principal moments of inertia along the axes (1̂, 2̂, 3̂) are denoted by I1 = I2 and I3, so the inertia tensor I will have components (I1, I1, I3) with respect to the (1̂, 2̂, 3̂)-frame. The axes 1̂ and 2̂ are principal axes due to the axisymmetry of TT. The equations of motion for TT are the Newton equations for the rolling and gliding rigid body ms̈ = F−mgẑ, L̇ = a× F, ˙̂3 = ω × 3̂, (1) where F is the external force acting on the TT at the supporting point A and L = Iω is the angular momentum w.r.t. CM . We assume that the TT is always in contact with the plane at A, so ẑ · (a+ s) = 0 holds at all times. This system is known to admit Jellett’s integral of motion λ = −L · a = RI1ϕ̇ sin2 θ −RI3ω3(α− cos θ) (without loss of generality, we will assume in this paper that λ is positive). The contact condition determines the vertical part of the external force, but the planar parts must be specified to make system (1) complete. We assume that the contact force has the form F = gnẑ−µgnvA, where gn ≥ 0 is the normal force and −µgnvA is a viscous-type friction force, acting against the gliding velocity vA. The quantity µ(L, 3̂, ṡ, s, t) ≥ 0 is a friction coefficient. For this model of the rolling and gliding TT, it is easy to see [4, 15] that the energy is decreasing Ė = F · vA < 0 and that the ŷ component of the friction force is the only force creating the torque necessary for transferring the rotational energy into the potential energy, Dynamics of an Inverting Tippe Top 3 thus lifting the CM of the TT. This mechanism shows that the inversion phenomenon is created by the gliding friction. The asymptotic properties of this model have been analysed in previous works [1, 5, 9, 10, 13]. In the nongliding case, vA = 0, the possible motions for the TT are either spinning in the upright (θ = 0) or in the inverted (θ = π) position, or rolling around with fixed CM with an inclination angle θ ∈ (0, π). The inclined rolling solutions are called tumbling solutions. If 1− α < γ < 1 + α, where γ = I1/I3, every angle in the interval (0, π) determines an admissible tumbling solution. Further, by a LaSalle-type theorem [13], it is known that for initial conditions such that the absolute value of the Jellett integral \|λ\| is above the threshold value λthres, only the inverted spinning position is a stable asymptotic solution. For a TT built such that it satisfies the parameter condition 1 − α < γ < 1 + α and for initial conditions with L · ẑ such that λ > λthres, the inversion can take place. Since we are primarily interested in the dynamics of inversion and we want to consider solu- tions describing an inverting TT, the basic assumptions are that the TT in question satisfies the parameter constraint and that we have initial conditions such that λ is above the threshold. Then we have a situation where an inverting solution becomes the only stable asymptotic solution, so the TT has to invert. Our aim is to describe the dynamics of inverting solutions. In our particular model, the assumptions about the reaction force F and the contact constraint yield the reduced equations of motion for the rolling and gliding TT: d dt (Iω) = a× (gnẑ − µgnvA) , mr̈ = −µgnvA, ˙̂3 = ω × 3̂, (2) where r = s−sẑ ẑ. We write the gliding velocity as vA = νx cos θ1̂+νy2̂+νx sin θ3̂, where νx, νy are the velocities in the 2̂× ẑ and 2̂ direction. Equations (2) can be written in the Euler form and then solved for the highest derivative of each of the variables (θ, ϕ, ω3, νx, νy). We then get the system θ̈ = sin θ I1 ( I1ϕ̇ 2 cos θ − I3ω3ϕ̇−Rαgn ) + Rµgnνx I1 (1− α cos θ), (3) ϕ̈ = I3θ̇ω3 − 2I1θ̇ϕ̇ cos θ − µgnνyR(α− cos θ) I1 sin θ , (4) ω̇3 = − µgnνyR sin θ I3 , (5) ν̇x = R sin θ I1 ( ϕ̇ω3 (I3(1− α cos θ)− I1) + gnRα(1− α cos θ)− I1α ( θ̇2 + ϕ̇2 sin2 θ )) − µgnνx mI1 ( I1 +mR2(1− α cos θ)2 ) + ϕ̇νy, (6) ν̇y = − µgnνy mI1I3 ( I1I3 +mR2I3(α− cos θ)2 +mR2I1 sin 2 θ ) + ω3θ̇R I1 (I3(α− cos θ) + I1 cos θ)− ϕ̇νx, (7) which, if we add the equation d dt(θ) = θ̇, becomes a dynamical system of the form (θ̇, θ̈, ϕ̈, ω̇3, ν̇x, ν̇y) = (h1(θ, . . . , νy), . . . , h6(θ, . . . , νy)). The value of the normal force gn can be determined from the contact constraint (a+ s) · ẑ = 0 and it is gn = mgI1 +mRα(cos θ(I1ϕ̇ 2 sin2 θ + I1θ̇ 2)− I3ϕ̇ω3 sin 2 θ) I1 +mR2α2 sin2 θ −mR2α sin θ(1− α cos θ)µνx . (8) We see that we get a complicated, nonlinear system for 6 unknowns. 4 S. Rauch-Wojciechowski and N. Rutstam 3 The main equation for the tippe top For further study of inverting solutions we need to clarify the logic of applying the main equation for the tippe top (METT) to analysing motion of TT. We need also to recall properties of TT equations when the TT is only rolling on the supporting surface and the gliding velocity vanishes vA = ṡ+ω×a = 0. It is the well known [2, 7] integrable case of the rolling axisymmetric sphere that was first separated by Chaplygin. We need to explain how the structure of separation equations motivates the introduction of the METT and how this equation differs from the classical separation equation. For the purely rolling axisymmetric sphere, the constraint vA = ṡ + ω × a = 0 implies that the equations of motion (1) reduce to a closed system for the vectors 3̂ and ω: d dt (Iω) = ma× ( gẑ − d dt (ω × a) ) , ˙̂3 = ω × 3̂. (9) For this system the external force is dynamically determined: F = mgẑ −m d dt(ω × a). In the Euler angle form the equations give a fourth order dynamical system for (θ, θ̇, ϕ̇, ω3). The system (9) admits three integrals of motion. Since the system is conservative, the energy E = 1 2 mṡ2 + 1 2 ω · L+mgs · ẑ = 1 2 ( I1ϕ̇ 2 sin2 θ + I1θ̇ 2 + I3ω 2 3 ) +mgR(1− α cos θ) + 1 2 mR2 [ (α− cos θ)2(θ̇2 + ϕ̇2 sin2 θ) + sin2 θ(θ̇2 + ω2 3 + 2ω3ϕ̇(α− cos θ)) ] (10) is an integral of motion. We also have Jellett’s integral λ = RI1ϕ̇ sin2 θ −RI3ω3(α− cos θ) as well as the Routh integral D := ω3 √ I3I1 +mR2(α− cos θ)2 +mR2I1 sin 2 θ = I3ω3 √ d(cos θ), where d(z) = γ + σ(α− z)2 + σγ(1− z2), σ = mR2 I3 and γ = I1 I3 . They allow to eliminate ω3 = D I3 √ d(cos θ) and ϕ̇ = λ √ d(cos θ)+RD(α−cos θ) RI1 sin 2 θ √ d(cos θ) from the expression of the energy (10) to get the separation equation E = g(cos θ)θ̇2 + V (cos θ,D, λ), (11) where g(cos θ) = 1 2I3 ( σ((α− cos θ)2 + 1− cos2 θ) + γ ) and V (z = cos θ,D, λ) = mgR(1− αz) + (λ √ d(z) +RD(α− z))2 2I3R2γ2(1− z2) + (R2D2 − σλ2) 2R2I1 . The separable first order differential equation (11) for θ determines the motion of the rolling TT. It is the Chaplygin separation equation for an axisymmetric sphere [2]. We shall show that for certain choice of parameters the effective potential V (z,D, λ) is convex in z ∈ [−1, 1] so, since V (z,D, λ) → ∞ as z → ±1, it has one minimum in the interval [−1, 1]. This means that for fixed E the solutions θ(t) describe nutational motion of the rolling TT between two bounding angles θ1, θ2 determined by the equation E = V (cos θ,D, λ). Dynamics of an Inverting Tippe Top 5 The rolling and gliding TT only has λ as an integral of motion. It is useful however to consider D(θ(t), ω3(t)) = I3ω3(t) √ d(cos θ(t)) being now a time dependent function. Its derivative we calculate using the equations of motion (3)–(7) for the rolling and gliding TT: d dt D(θ, ω3) = γmR sin θ√ d(cos θ) (ϕ̇νx + ν̇y) = γm α √ d(ẑ · 3̂) (ẑ × a) · v̇A. For the total energy of TT, E = 1 2m(vA−ω×a)2+ 1 2ω ·L+mgs · ẑ, we know that Ė = F ·vA < 0. The part of E that does not depend on vA, Ẽ(θ, θ̇, ϕ̇, ω3) = 1 2 m(ω × a)2 + 1 2 ω · L+mgs · ẑ = 1 2 ( I1ϕ̇ 2 sin2 θ + I1θ̇ 2 + I3ω 2 3 ) +mgR(1− α cos θ) (12) + 1 2 mR2 [ (α− cos θ)2(θ̇2 + ϕ̇2 sin2 θ) + sin2 θ(θ̇2 + ω2 3 + 2ω3ϕ̇(α− cos θ)) ] , we will call the modified energy function. The derivative of this function is d dtẼ(θ, θ̇, ϕ̇, ω3) = mv̇A · (ω × a). With the use of the functions D(θ, ω3), Ẽ(θ, θ̇, ϕ̇, ω3) we can write the TT equations of motion (3)–(7) in an equivalent integrated form [12, 15] as d dt λ(θ, θ̇, ϕ̇, ω3) = 0, d dt D(θ, ω3) = γm α √ d(ẑ · 3̂) (ẑ × a) · v̇A, d dt Ẽ(θ, θ̇, ϕ̇, ω3) = m(ω × a) · v̇A, d dt mṙ = −µgnvA. These equations are as difficult as the equations (3)–(7). However, if we treat D(θ(t), ω3(t)) =: D(t), Ẽ(θ(t), θ̇(t), ϕ̇(t), ω3(t)) =: Ẽ(t) as given known functions, then fromD(t) = I3ω3 √ d(cos θ) and λ = RI1ϕ̇ sin2 θ−RI3ω3(α−cos θ) we can calculate ϕ̇, ω3 and substitute into expression (12) for the modified energy to obtain the METT [12, 15] that involves only the function θ(t): Ẽ(t) = g(cos θ)θ̇2 + V (cos θ,D(t), λ). This equation has the same form as equation (11), but now it depends explicitly on time through the functions D(t) and Ẽ(t). Solving this equation is therefore not longer possible. It is a first order time dependent ODE which we can study provided that we have some quantitative infor- mation about the functions D(t) and Ẽ(t). The functions D(t), Ẽ(t) are usually unknown but for inverting solutions we have qualitative information about their behaviour due to conservation of the Jellett function λ. Thus we consider the motion of the TT as being determined by the three functions (λ,D(t), Ẽ(t)) and governed by the METT. Of particular interest regarding the inversion movement is the initial and final position of the TT. The TT goes (asymptotically) from an initial angle close to θ = 0 to the final angle close to θ = π which means, since λ = −L · a is constant, that λ = L0R(1 − α) = L1R(1 + α) (where L0 and L1 are the values of \|L\| at θ = 0 and θ = π, respectively). This implies that D0 = L0 √ d(1) = λ R(1−α) √ γ + σ(1− α)2 and D1 = −L1 √ d(−1) = − λ R(1+α) √ γ + σ(1 + α)2, and also that Ẽ0 = λ2 2R2I3(1−α)2 +mgR(1− α) and Ẽ1 = λ2 2R2I3(1+α)2 +mgR(1 + α) [14]. 6 S. Rauch-Wojciechowski and N. Rutstam The values (D0, Ẽ0) and (D1, Ẽ1) can be interpreted as the boundary values for the unknown functions (D(t), Ẽ(t)). So we assume that for inverting solutions (D(t), Ẽ(t)) t→∞−→ (D1, Ẽ1) and (D(t), Ẽ(t)) t→−∞−→ (D0, Ẽ0). The aim of the following sections is to analyse dynamical properties of the inverting solution as the symmetry axis of TT moves from a neighborhood of θ = 0 to a neighborhood of θ = π. In order to simplify the technical side of analysis we choose special values of the parameters in METT so that the effective potential V (cos θ,D, λ) becomes rational, but we expect that the whole line of reasoning can be repeated in the general case when the potential depends algebraically on z through √ d(z). We show that V (z,D, λ) is strictly convex and therefore has one minimum zmin. We show also that, for inverting solutions when D(t) moves from D0 to D1, the potential deforms so that zmin = zmin(D,λ) moves from a neighborhood of z = 1 to a neighborhood of z = −1. Thus as the potential V (z,D(t), λ) deforms and the modified energy Ẽ(t) changes from Ẽ0 to Ẽ1 the angle θ(t) performs oscillatory motions between two turning angles θ±(t) satisfying the equation V (cos θ±, D(t), λ) = Ẽ(t). On the unit sphere S2 the angle θ(t) performs nutational motion within the nutational band [θ−(t), θ+(t)] that moves from the neighborhood of the north pole to the neighborhood of the south pole. We shall give an estimate for the relation between the inversion time Tinv and the maximal period of nutation TV (Ẽ(t), D(t)), so that if Tinv is an order of magnitude larger than TV , say Tinv > 10TV , the angle θ(t) performs oscillatory motion within the moving nutational band. 3.1 The rational form of the METT The effective potential in the separation equation (11) is an algebraic function in z, which complicates the analysis. We can however make a restriction on the parameters so that the second degree polynomial d(z) can be written as a perfect square. This makes the term √ d(z) a linear function of z and the potential becomes a rational function [2, 16]. We see that if 1 − α2 < γ < 1, and if we let the parameter σ = 1−γ γ+α2−1 > 0, then d(z) = γ + σ(α − z)2 + σγ(1 − z2) = (α−(1−γ)z)2 γ+α2−1 . This is a perfect square, so for γ in this range we can find physical values for σ such that √ d(z) is a real polynomial in z: α−(1−γ)z√ γ+α2−1 . Note that (1−α2, 1) is a subinterval of (1−α, 1+α), the parameter range for γ where complete inversion of TT is possible. When σ = 1−γ γ+α2−1 , we can rewrite the functions in the separation equation E = g(cos θ)θ̇2 + V (cos θ,D, λ) as g(z) = I3 2 ( σ ( 1 + α2 − 2αz ) + γ ) = I3 2 1 γ + α2 − 1 ( α2 + (1− γ)2 − 2α(1− γ)z) ) , and V (z,D, λ) = mgR(1− αz) + ( λ(α− (1− γ)z) +RD √ γ + α2 − 1(α− z) )2 2I3R2γ2(γ + α2 − 1)(1− z2) + R2D2(γ + α2 − 1)− (1− γ)λ2 2R2I1(γ + α2 − 1) . (13) This rational form of the effective potential is simpler to work with. We should note that the restriction on the parameter σ implies that the moments of inertia I1 and I3 are dependent on each other; if σ = 1−γ γ+α2−1 then I1 = I23+mR 2I3(1−α2) I3+mR2 < I3. Dynamics of an Inverting Tippe Top 7 a b αγλ 1+α γλ − αγλ 1−α D1 D0 αγλ Figure 2. Linear relationship between the parameters a and b. This line intersects the lines b = a and b = −a when D = D1 and D = D0, respectively. 4 Convexity of the rational potential V (z,D, λ) The range of parameters making the effective potential a rational function provides a simplest non-trivial situation in which we can study properties of the potential in greater detail. We consider the potential function Ṽ (z,D, λ) = −mgRαz + ( λ(α− (1− γ)z) +RD √ γ + α2 − 1(α− z) )2 2I3R2γ2(γ + α2 − 1)(1− z2) =: 1 2I3R2γ2(γ + α2 − 1) ( −βz + (az + b)2 1− z2 ) , (14) where Ṽ (z,D, λ) = V (z,D, λ) − ( mgR + R2D2(γ+α2−1)−(1−γ)λ2 2R2I1(γ+α2−1) ) , since the constant does not affect the shape of V (z,D, λ) and the position of minimum zmin. The parameters in the function f(z) = −βz + (az+b)2 1−z2 (the expression inside the parentheses on the r.h.s. of (14)) are therefore defined as a = −(1− γ)λ−RD √ γ + α2 − 1, (15) b = αλ+ αRD √ γ + α2 − 1, (16) β = 2mgR3αI3γ 2(γ + α2 − 1). (17) Remember that 1 − α2 < γ < 1 and the range of parameters a, b is determined by the range of D. We observe that a + b = 0 if D = D0 = λ R(1−α) (α−1+γ)√ γ+α2−1 and a − b = 0 if D = D1 = − λ R(1+α) (α+1−γ)√ γ+α2−1 . The parameters a and b satisfy the relation b + αa = λγα. It is illustrated in Fig. 2, where the lines b = a and b = −a correspond to D = D1 and D = D0, respectively. Proposition 1. The effective potential V (z,D, λ) of (13) is convex for z ∈ (−1, 1) and for all real values of D and λ. Proof. We must show that d2 dz2 V (z,D, λ) ≥ 0 for z ∈ (−1, 1). Due to the form of V , it is enough to show that the rational function (az+b)2 1−z2 is convex for all a, b. Suppose first a 6= ±b and ab 6= 0. We look at the second derivative of this function: d2 dz2 (az + b)2 1− z2 = 2 (1− z2)3 (2abz3 + 3(a2 + b2)z2 + 6abz + a2 + b2) = 2q(z) (1− z2)3 , 8 S. Rauch-Wojciechowski and N. Rutstam and have to show that the third degree polynomial q(z) = 2abz3+3(a2+ b2)z2+6abz+ a2+ b2, has no roots in the interval [−1, 1]. To do this we apply the Sturm theorem [17]. We generate a sequence of polynomials (q0(z), q1(z), q2(z), . . . , qm(z)) recursively by starting from a square-free polynomial q(z): q0(z) = q(z), q1(z) = q′(z) and qi = −rem(qi−1, qi−2) for i ≥ 2. Here rem(qi−1, qi−2) denotes the remain- der after polynomial division of qi−1 by qi−2. By Euclid’s algorithm, this will terminate with the constant polynomial qm. Let S(ξ) be the number of sign-changes in the sequence (q0(ξ), q1(ξ), q2(ξ), . . . , qm(ξ)) at the point ξ. The Sturm theorem states that for real numbers c < d the number of distinct roots in (c, d] is S(c)− S(d). For our polynomial q(z) the algorithm described above yields four polynomials (q0, q1, q2, q3). When we look at this sequence of polynomials at the points z = −1 and z = 1 we have q0(−1) = 4(a− b)2, q0(1) = 4(a+ b)2, q1(−1) = −6(a− b)2, q1(1) = 6(a+ b)2, q2(−1) = − (a2 − b2)2 ab , q2(1) = (a2 − b2)2 ab , q3(−1) = −6ab, q3(1) = −6ab. We thus see that the number of sign changes for the Sturm sequence at both points −1 and 1 is the same, either 1 or 2, depending on whether ab is positive or negative. Thus according to Sturm’s theorem, q(z) has no roots in (−1, 1]. Since q(0) = a2 + b2 > 0 (and q(−1) > 0), q(z) is positive in [−1, 1] and we can conclude that (az+b)2 1−z2 is convex when a 6= ±b and ab 6= 0. Routine checking confirms that this is also convex if a = ±b and as well as if ab = 0. Thus the function is convex for all arbitrary values of the parameters a, b ∈ R. � 4.1 Estimates for position of minimum of V (z,D, λ) The result that V (z,D, λ) is convex for z ∈ (−1, 1), combined with V (z,D, λ)→∞ as z → ±1, gives that the potential will have exactly one minimum zmin in [−1, 1]. We use this to analyse how this minimum shifts as D goes from D0 to D1, for λ > λthres. We show below that for every (small) ε > 0 we can find δ+(ε, λ) > 0 and δ−(ε, λ) > 0 such that for 0 < δ < δ+(ε, λ) and 0 < δ < δ−(ε, λ) the potential V (z,D0 ∓ δ/(R(1 − α) √ γ + α2 − 1), λ) has a minimum in [1− ε, 1] and V (z,D1 ± δ/(R(1 + α) √ γ + α2 − 1), λ) has a minimum in [−1,−1 + ε]. This means that as the inclination angle θ goes from a value close to 0 to a value close to π and D(t) goes from a neighborhood ( D0 − δ R(1−α) √ γ+α2−1 , D0 + δ R(1−α) √ γ+α2−1 ) to a neigh- borhood ( D1 − δ R(1+α) √ γ+α2−1 , D1 + δ R(1+α) √ γ+α2−1 ) , the minimum of V (cos θ,D, λ) moves from a neighborhood of θ = 0 to a neighborhood of θ = π. As explained before, it is sufficient to study the position of zmin for the function 2I3R 2γ2(γ+ α2 − 1)Ṽ (z,D(a, b), λ(a, b)) =: f(z) with f(z) = (az + b)2 1− z2 − βz. The derivative is f ′(z) = p(z) (1−z2)2 , where p(z) = 2(az + b)(bz + a)− β(1− z2)2. Our question about the position of zmin can be reduced to the question: for which values of the parameters a, b, β does the polynomial p(z) have one zero in one of the subintervals [1− ε, 1] and [−1,−1 + ε] of the interval [−1, 1]? We see that p(−1) = −2(a− b)2 and p(1) = 2(a+ b)2, which confirms that f ′(z) has a zero and changes sign in [−1, 1] (remember that a 6= ±b if D 6= D1 and D 6= D0). Dynamics of an Inverting Tippe Top 9 Proposition 2. Assume that λ > λthres = √ mgR3I3α(1+α)2√ 1+α−γ . i) For any (small) ε > 0 there is δ−(ε, λ) > 0 being a positive solution of the equation α(1− α)γλδ + αδ2 + δ2 ε2 (1− ε)(1 + α)2 = α2γ2λ2 − β 2 (2− ε)2(1 + α)2, such that for every positive δ < δ−(ε, λ) the potential V (z,D, λ) has a minimum zmin in the interval [−1,−1 + ε] for D = D1 ± δ R(1+α) √ γ+α2−1 . ii) For any (small) ε > 0 there is δ+(ε, λ) = min{δ1, δ2} > 0, where δ1 = γ(1 + α)λ and δ2 is the positive solution of the equation δ2 ε2 (1− ε)(1− α)2 + αγλδ(1 + α)− αδ2 = α2γ2λ2 + β 2 (2− ε)2(1− α)2, such that for every positive δ < δ+(ε, λ) the potential V (z,D, λ) has a minimum zmin in the interval [1− ε, 1] for D = D0 ∓ δ R(1−α) √ γ+α2−1 . Proof. We know that V (z,D, λ) has one minimum in [−1, 1] and p(z), the numerator of the derivative f ′(z), has one zero in [−1, 1]. The question is to formulate conditions for the para- meter D so that this zero is located in the interval [−1,−1 + ε] or [1− ε, 1]. i) We have to show that for any (small) ε > 0 and λ > λthres there is δ−(ε, λ) > 0 such that for any δ < δ−(ε, λ) the potential V ( z,D = D1 ± δ R(1+α) √ γ+α2−1 , λ ) has a minimum in the interval [−1,−1+ ε]. To find this δ−(ε, λ) we have to study the signs of p(−1) = −2(a− b)2 < 0 and of p(−1 + ε) = p(−1) + 2(a− b)2ε+ 2(ab− 2β)ε2 + 4βε3 − βε4. To have minimum zmin in [−1,−1+ε] we need to have p(−1+ε) > 0. If D = D1± δ R(1+α) √ γ+α2−1 then (a− b)2 = δ2, ab = 1 (1+α)2 (αγλ− δ)(αγλ+ αδ) and p(−1 + ε) = −2δ2 + 2δ2ε+ 2abε2 − βε2(2− ε)2 = −2δ2(1− ε) + 2ε2 (1 + α)2 ( α2γ2λ2 − αγ(1− α)δλ− αδ2 ) − βε2(2− ε)2. For positivity of p(−1 + ε) we must have α(1− α)γλδ + αδ2 + δ2 ε2 (1− ε)(1 + α)2 < α2γ2λ2 − β 2 (2− ε)2(1 + α)2. (18) The r.h.s. of (18) is positive since (αγλ)2 − β 2 (2− ε)2(1 + α)2 = (αγλ)2 [ 1− 2β λ2 (1 + α)2 α2γ2 (2− ε)2 4 ] , and for λ = Cλthres, C > 1, the term in the square brackets can be shown to be positive. We see this when we write 2β λ2 (1 + α)2 α2γ2 (2− ε)2 4 = 4 [ (γ + α2 − 1)(1 + α− γ) α2(1 + α)2 ] 1 C2 (2− ε)2 4 . (19) 10 S. Rauch-Wojciechowski and N. Rutstam Clearly 1 C2 (2−ε)2 4 < 1 and the function ρ(α, γ) = (γ+α2−1)(1+α−γ) α2(1+α)2 < 1 4 for 1− α2 < γ < 1, 0 < α < 1. So the r.h.s. of (18) is indeed positive. The l.h.s. of (18) has the form δ[α(1 − α)γλ + αδ + δ ε2 (1− ε)(1 + α)2] with all terms positive. If we define δ−(ε, λ) as the positive solution of α(1− α)γλδ + αδ2 + δ2 ε2 (1− ε)(1 + α)2 = α2γ2λ2 − β 2 (2− ε)2(1 + α)2, (20) then for all positive δ < δ−(ε, λ) we have p(−1 + ε) > 0 and the minimum zmin ∈ [−1,−1 + ε]. ii) We have to show that for any (small) ε > 0 and λ > λthres there is δ+(ε, λ) > 0 such that for any positive δ < δ+(ε, λ) the potential V ( z,D = D0 ∓ δ R(1−α) √ γ+α2−1 ) has a minimum zmin in the interval [1− ε, 1]. To find this δ+(ε, λ) we need to study the signs of p(1) = 2(a+ b)2 > 0 and of p(1− ε) = p(1)− 2(a+ b)2ε+ 2(ab− 2β)ε2 + βε3(4− ε). To have minimum zmin in [1 − ε, 1] we need to show how to find δ so that p(1− ε) < 0. If D = D0 ∓ δ R(1−α) √ γ+α2−1 then (a+ b)2 = δ2, ab = − 1 (1−α)2 (αγλ− δ)(αγλ− αδ) and p(1− ε) = 2δ2(1− ε)− 2ε2 (1− α)2 ((αγλ− δ)(αγλ− αδ) + 2β(1− α)2) + βε3(4− ε). For negativity of p(1− ε) we must have δ2 ε2 (1− ε)(1− α)2 + αγλδ(1 + α)− αδ2 < α2γ2λ2 + β 2 (2− ε)2(1− α)2. (21) The r.h.s. of (21) is obviously positive. The l.h.s. of (21) has the form δ [ δ ε2 (1− ε)(1− α)2 + α((1 + α)γλ− αδ) ] , which is certainly positive when λ > δ γ(1+α) since δ > 0 can always be chosen so small that this is true, by the chosen convention λ > 0. If we define δ+(ε, λ) = min{δ1, δ2}, where δ1 = γ(1 +α)λ and δ2 is the positive solution of δ2 ε2 (1− ε)(1− α)2 + αγλδ(1 + α)− αδ2 = α2γ2λ2 + β 2 (2− ε)2(1− α)2, then for all positive δ < δ+(ε, λ) we have p(1− ε) < 0 and the minimum zmin ∈ [1− ε, 1]. � Example 1. We exemplify the above proposition for some physical values for the parameters. We let m = 0.02 kg, R = 0.02 m and α = 0.3. We shall assume I3 = 2 5mR 2, and we get a rational potential (13) when I1 = I3(I3+mR2(1−α2)) I3+mR2 = 131 350mR 2, which means that γ = 131 140 ∈ (1 − α2, 1) = (0.91, 1). Jellett’s integral λ will for simplicity be assumed to be two times the threshold value λthres, so λ = 6.88 · 10−6 kg ·m3 · s−1. Suppose first that ε = 0.1 and consider the condition for a minimum zmin to stay in the interval (−1,−1 + ε). We have D1 = −6 · 10−4 and by solving (20) we get δ− = 1.48 · 10−7, so if D = D1 + δ, with δ < δ− R(1+α) √ γ+α2−1 ≈ 3.5 · 10−5, the potential has its minimum in the interval. If ε is reduced to 0.01, the bound for δ is tightened one order of magnitude as well. Dynamics of an Inverting Tippe Top 11 5 Oscillation of θ(t) within the deforming rational potential V (cos θ,D(t), λ) As a toy TT inverts, we can see that the symmetry axis performs oscillations, or equivalently we say that it nutates. This is also apparent in simulations of the equations of motion [3, 11, 18] where graphs of the evolution of the inclination angle θ(t) show that it rises in an oscillating manner from an angle close to θ = 0 to an angle close to θ = π. Here we say that a solution θ(t) is oscillatory on a time interval [0, T ] when θ̇(t) changes sign a number of times in this interval. We consider solutions of METT with D(t), Ẽ(t) describing inverting solutions of TT equa- tions, under assumption that D(t), Ẽ(t) are slowly varying regular functions moving from a small neighborhood of (D0, Ẽ0) to a small neighborhood of (D1, Ẽ1). They are regular since D(t) = D(θ(t), ω3(t)) and Ẽ(t) = Ẽ(θ(t), θ̇(t), ϕ̇(t), ω3(t)). We further have assumed that (D(t), Ẽ(t)) t→∞−→ (D1, Ẽ1) and (D(t), Ẽ(t)) t→−∞−→ (D0, Ẽ0). In the limiting case of constant D and Ẽ when the METT describes purely rolling TT, the oscillating behaviour of θ(t) follows from the dynamical system representation of the second order equation for θ(t). For the rolling TT the energy Ẽ = g(cos θ)θ̇2 + V (cos θ,D, λ) (22) is an integral of motion for the θ-equation obtained by differentiating (22): θ̈ = sin θ 2g(cos θ) ( g′z(cos θ)θ̇ 2 + V ′z (cos θ,D, λ) ) , where g(z = cos θ) = I3(α2+(1−γ)2−2α(1−γ)z) 2(γ+α2−1) . This means that Ẽ = g(cos θ)y2 + V (cos θ,D, λ) is an integral of motion for the dynamical system θ̇ = y, ẏ = sin θ 2g(cos θ) ( g′z(cos θ)y 2 + V ′z (cos θ,D, λ) ) . (23) Trajectories of system (23) are lines of constant value of energy Ẽ and they are closed curves. The closed trajectories describe periodic solutions [2] with period T defined by the integral T (Ẽ) = 2 ∫ θ2 θ1 √ g(cos θ)dθ√ Ẽ − V (cos θ,D, λ) , where the turning latitudes θ1 < θ2 are defined by Ẽ = V (cos θ1,2, D, λ). The modified energy Ẽ(t) is bounded. That entails boundedness of ω(t) = −ϕ̇ sin θ1̂+ θ̇2̂+ (ψ̇+ ϕ̇ cos θ)3̂ and thus \|θ̇(t)\| < B for some positive B. Since the potential V (z,D(t), λ) θ→0,π−→ ∞ the curve (θ(t), θ̇(t)) of each inverting solution is confined to the open rectangle (θ, θ̇) ∈ (0, π)× (−B,B). The picture that emerges is that, for slowly varying Ẽ(t), D(t), the inverting trajectories of METT stay (locally) close to the trajectories of system (23) and are traversed with almost the same velocity. The time of passing T = t2 − t1 between two turning angles given by V (cos θ1, D(t1), λ) = Ẽ(t1) and V (cos θ2, D(t2), λ) = Ẽ(t2) is close to the half-period of the non-deforming potential. Thus initially the trajectory moves around (θmin, θ̇ = 0), with θmin close to 0. As D(t)→ D1 the minimum θmin moves toward θ = π (see Proposition 2) and for sufficiently slowly va- rying D(t) the trajectory goes several times around (θmin, θ̇ = 0) and drifts toward the point 12 S. Rauch-Wojciechowski and N. Rutstam 0 1 2 3 4 5 6 7 8 Time (sec) 0 π/2 π θ (r a d ) a) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 θ (rad) −15 −10 −5 0 5 10 15 θ̇ (r a d /s e c) b) 0 2 4 6 8 10 Time (sec) 0 π/2 π θ (r a d ) c) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 θ (rad) −8 −6 −4 −2 0 2 4 6 8 10 θ̇ (r a d /s e c) d) Figure 3. Plots of (t, θ(t)) (left) and (θ(t), θ̇(t)) (right) obtained by integrating equations (3)–(7) and (8) for two sets of parameters and initial values. Plots a and b correspond to parameter values for rational potential in Example 1 and µ = 0.3. Plots c and d correspond to parameter values provided in [3], corresponding to an algebraic potential. The equations are integrated using the Python 2.7 open source library SciPy [8]. (θ = π, θ̇ = 0). The trajectory describes a selfcrossing and contracting spiral with the center moving toward (π, 0) in the (θ, θ̇)-plane. The solution curve for θ(t) is a superposition of a small amplitude oscillating component with a logistic type curve going from θ = 0 to θ = π. In Fig. 3 we see this. The plots a and b present a graph of θ(t) and a curve (θ(t), θ̇(t)) obtained by solving equations of motion (3)–(7) for the rolling and gliding TT, with gn given by equa- tion (8) and with values of parameters taken from Example 1. So the effective potential V given by (13) is rational. We have taken µ = 0.3 and we have integrated the equations numerically for initial values θ(0) = 0.1 rad, θ̇(0) = ϕ̇(0) = 0, νx(0) = νy(0) = 0 and ω3(0) = 155.0 rad/s (corresponding to the value of λ = 2λthres). Plot a shows oscillations of θ(t) as it rises to π. The time of inversion from the moment when θ(t) starts to rise, at approximately 3 seconds, is about 4–5 seconds. Larger values of µ give shorter time of inversion. Plot b shows the trajectory in the (θ, θ̇)-plane. In Fig. 3 c and d the same type of numerical integration is performed, for parameters giving an algebraic (not rational) effective potential. The parameters are the same as used in Cohen [3] and they are close to values taken in Example 1. Here m = 0.015 kg, R = 0.025 m, α = 0.2, I1 = I3 = 2 5mR 2 and µ = 0.3 with initial values θ(0) = 0.1 rad, θ̇(0) = ϕ̇(0) = 0, νx(0) = νy(0) = 0 and ω3(0) = 100 rad/s. We may note that Cohen uses a Coulomb-type gliding friction force, so the equations solved in his article are slightly different than equations (3)–(7). In the next section we shall estimate the maximum value of the period of oscillations for all values of the modified energy Ẽ and D. This will allow us to formulate a condition for the time of inversion needed for oscillatory behaviour of θ(t). Basically Tinv has to be an order of magnitude larger than Tupp – the maximal period of oscillations within potential V (cos θ,D(t), λ). Dynamics of an Inverting Tippe Top 13 5.1 Estimates for the period of oscillation A direct way of estimating the period of nutation is to study the explicit integral defining the period and expanding it w.r.t. a small parameter ε = 2β b2 , with b and β given by (16) and (17), respectively. This is the technique used in [16], but in that paper the analysis is based on the assumption that ω3 is large. Here we use the more general assumption that λ is only above the threshold value, so that λ = Cλthres with C > 1. To simplify estimates we also assume that we consider curves (D(t), Ẽ(t)) such that D1 < D < D0. Then by using that b = αλ+ αRD √ γ + α2 − 1 satisfies αγλ 1+α < b < αγλ 1−α for D1 < D < D0, we have the estimate ε = 2β b2 < 2β(1 + α)2 λ2α2γ2 = 4 C2 · [ (γ + α2 − 1)(1 + α− γ) α2(1 + α)2 ] < 4 C2 · 1 4 < 1, (24) for 1− α2 < γ < 1, 0 < α < 1, similar to the estimate for equation (19). In the following we shall determine the dependence of the period T (ε) on the small parameter ε = 2β b2 and we shall find an estimate for the maximal value of T (ε) that is valid for all D ∈ (D1, D0) and all values ε < 1, which means all values of the Jellett integral that are above the threshold value. As known from the asymptotic analysis these initial values of λ lead to inversion of the TT. The period of oscillations for the potential V (z,D, λ) is given by the integral T = 2 ∫ z2 z1 √ g(z)dz√ (1− z2)(Ẽ − V (z,D, λ)) , (25) where z1 < z2 ∈ (−1, 1) are two turning points defined by Ẽ = V (z1,2, D, λ). This equation always has two solutions for Ẽ > V (zmin, D, λ) since the potential is convex and lim z→±1 V (z,D, λ) =∞. In terms of the the parameters a and b, the potential (13) reads V (z,D, λ) = mgR(1− αz) + (az + b)2 2I3R2γ2(γ + α2 − 1)(1− z2) + α2a2 − (1− γ)b2 2R2α2I3γ2(γ + α2 − 1) , and (by definition) the left turning point z1 is given implicitly by the equation Ẽ = mgR(1− αz1) + (az1 + b)2 2I3R2γ2(γ + α2 − 1)(1− z21) + α2a2 − (1− γ)b2 2R2α2I3γ2(γ + α2 − 1) . In the following we shall parametrise (similarly as in [16]) the remaining roots of (1− z2)(Ẽ − V (z,D, λ)) by z1. Due to this they become solutions of a quadratic equation. The function in the denominator of (25) we write as ( 1− z2 ) (Ẽ − V (z,D, λ)) = 1( 1− z21 )(−mgRα(z1 − z)(1− z21)(1− z2) + (az1 + b)2(1− z2)− (az + b)2(1− z21) 2I3R2γ2(γ + α2 − 1) ) = β(z1 − z) 2I3R2γ2(γ + α2 − 1) ( z2 + a2 + b2 + 2abz1 β(1− z21) z − 1 + (a2 + b2)z1 + 2ab β(1− z21) ) . (26) 14 S. Rauch-Wojciechowski and N. Rutstam Notice that z = ±1 are not roots because V (z,D, λ) has singularities at the points z = ±1. The quadratic polynomial in the parentheses of (26) determines roots z2, z3, for any given turning point z1. They are thus solutions of the quadratic equation z2 + ( a2 + b2 + 2abz1 β(1− z21) ) z − 1 + (a2 + b2)z1 + 2ab β(1− z21) = 0. (27) By z3 we denote the root satisfying z3 < −1 and by z2 (>z1) the right turning point. The polynomial (1− z2)(E − V (z,D, λ)) is then factorized: ( 1− z2 ) (E − V (z,D, λ)) = β(z1 − z)(z2 − z)(z3 − z) 2I3R2γ2(γ + α2 − 1) = mgRα(z1 − z)(z2 − z)(z3 − z). We substitute this into the integral for the period T (25). By the mean value theorem for integrals there exist a z∗ ∈ [z1, z2] such that we have T = 2 √ g(z∗)√ mgRα ∫ z2 z1 dz√ (z1 − z)(z2 − z)(z3 − z) , where g(z∗) = I3(α2+(1−γ)2−2α(1−γ)z∗) 2(γ+α2−1) . The integral can be transformed to a standard complete elliptic integral of the first kind through the change of variables z = z2 + (z1 − z2)s2 [6]: T = 4 √ g(z∗)√ mgRα 1√ z2 − z3 ∫ 1 0 ds√ (1− s2)(1− k2s2) , (28) where k2 = z2−z1 z2−z3 < 1 is a positive parameter. This integral has the standard expansion K ( k2 ) = ∫ 1 0 ds√ (1− s2)(1− k2s2) = π 2 ∞∑ n=0 ( (2n− 1)!! (2n)!! )2 k2n = π 2 ( 1 + k2 4 +O ( k4 )) .(29) By using the roots of equation (27) and by expanding k2 = z2−z1 z2−z3 w.r.t. ε = 2β b2 , we can show that k2 = O(ε) if ε is small. Indeed, after solving equation (27) we get that the quantity k2 can be written: k2 = z2 − z1 z2 − z3 = 1 2 − 1 2 ( z1 + a2 + b2 + 2abz1 2β(1− z21) ) × ( (a2 + b2 + 2abz1) 2 4β2(1− z21)2 + 1− (a2 + b2)z1 + 2ab β(1− z21) )− 1 2 = 1 2 − 1 2 ( z1 2β ( 1− z21 ) a2 + b2 + 2abz1 + 1 ) × ( 1 + 4β2(1− z21)2 (a2 + b2 + 2abz1)2 − 4β(1− z21)((a2 + b2)z1 + 2ab) (a2 + b2 + 2abz1)2 )− 1 2 . (30) With the parameters ε = 2β b2 and w = a b , the expansion of k2 is: k2 = 1 2 − 1 2 ( z1(1− z21) w2 + 1 + 2wz1 ε+ 1 ) × ( 1 + (1− z21)2 (w2 + 1 + 2wz1)2 ε2 − 2(1− z21)((w2 + 1)z1 + 2w) (w2 + 1 + 2wz1)2 ε )− 1 2 Dynamics of an Inverting Tippe Top 15 = − (1 + z1w)(w + z1) ( 1− z21 ) (1 + w2 + 2wz1)2 ε+O ( ε2 ) . (31) It should be noted that the minus sign here is misleading. When (31) is expressed by parame- ters D, λ, one sees that the factor at ε is positive. We consider the parameter k2 for (z1, w) ∈ [−1, 1]× [−1+ δ, 1− δ] with certain small δ. The values w = a b = ∓1 correspond (as Fig. 2 shows) to the upright and inverted spinning solutions, which are asymptotic solutions of TT equations and are never attained during the inversion. The inverting TT starts with w0 = −1 + δ0 and angle θ0 close to 0 and moves to the value w1 = 1 − δ1 where the angle θ1 is close to π. So we need to estimate the function h1(z1, w) = − (1+z1w)(w+z1)(1−z21) (1+w2+2wz1)2 , being a factor in front of ε in (31), on a rectangle [−1, 1] × [−1 + δ, 1 − δ] =: Rδ with δ < min{δ0, δ1}. The function h1(z1, w) has on [−1, 1] × (−1, 1) two critical points ( 1√ 3 , 0), (− 1√ 3 , 0), which gives a local minimum h1( 1√ 3 , 0) = − 2 3 √ 3 and a local maximum h1(− 1√ 3 , 0) = 2 3 √ 3 . In the whole region Rδ the function h(z1, w) satisfies the inequality − 2 3 √ 3 ≤ h1(z1, w) ≤ 2 3 √ 3 and, therefore, it is bounded \|h1(z1, w)\| ≤ 2 3 √ 3 . This clarifies how k2 depends on the small parameter ε. To estimate 1√ z2−z3 in (28), we write 1√ z2 − z3 = 1√ 2 (( a2 + b2 + 2abz1 )2 4β2(1− z21)2 + 1− ((a2 + b2)z1 + 2ab) β(1− z21) )−1/4 = √ β ( 1− z21 ) b √ 1 + w2 + 2wz1 × ( 1 + (1− z21)2 (1 + w2 + 2wz1)2 ε2 − 2(1− z21)((1 + w2)z1 + 2w) (1 + w2 + 2wz1)2 ε )−1/4 = √ ε 2 √( 1− z21 ) √ 1 + w2 + 2wz1 ( 1 + (1− z21)((1 + w2)z1 + 2w) 2(1 + w2 + 2wz1)2 ε+O ( ε2 )) . (32) The function h2(z1, w) = (1−z21)((1+w2)z1+2w) 2(1+w2+2wz1)2 satifies the inequalities −1 < h2(z1, w) < 1 on the rectangle (z1, w) ∈ Rδ, so it is bounded: \|h2(z1, w)\| ≤ 1. Thus the nutational period behaves as T (ε) = 4 √ g(z∗)√ mgRα 1√ z2 − z3 ∫ 1 0 ds√ (1− s2)(1− k2s2) = 4 √ g(z∗)√ mgRα √ ε 2 √ 1− z21√ 1 + w2 + 2wz1 [ 1 + h2(z1, w)ε+O ( ε2 )] × π 2 [ 1 + 1 4 h1(z1, w)ε+O ( ε2 )] = 2π √ g ( z∗ ) √ 2mgRα √ 1− z21 1 + w2 + 2wz1 √ ε [1 +O(ε)] , where z∗ is some value between z1 and z2. In the leading factor at √ ε we have ∣∣∣ 1−z21 1+w2+2wz1 ∣∣∣ ≤ 1 on any rectangle Rδ and the function g(z∗) is decreasing for z∗ ∈ (−1, 1) with supremum g(−1) = I3(α+1−γ)2 2(γ+α2−1) so that 2π √ g ( z∗ ) √ 2mgRα √ 1− z21 1 + w2 + 2wz1 √ ε < 2π √ g(−1) √ ε√ 2mgRα = 2π ( RI3γ(α+ 1− γ) b ) =: Tmax. We summarize these results in a proposition. 16 S. Rauch-Wojciechowski and N. Rutstam Proposition 3. i) For D ∈ (D1, D0) = ( −λ √ d(−1) R(1+α) , λ √ d(1) R(1−α) ) , Ẽ > V (zmin, D, λ) with λ = Cλthres = C √ mgR3I3α(1 + α)2√ 1 + α− γ , C > 1, the period of oscillations behaves as 2π √ g(z∗)√ 2mgRα √ 1− z21 1 + w2 + 2wz1 √ ε [1 +O(ε)] , as ε = 2β b2 → 0. ii) The leading factor is bounded 2π √ g(z∗)√ 2mgRα √ 1− z21 1 + w2 + 2wz1 √ ε ≤ 2πRI3γ(α+ 1− γ) b =: Tmax for w and z1 in any closed rectangle (z1, w) ∈ [−1, 1]× [−1 + δ, 1− δ]. Remark 1. Notice that the upper bound is taken uniformly w.r.t. both variables (z1, w) over Rδ. It does not take into account the dependence of the root z1(a, b) on a, b and that z∗ ∈ [z1, z2] belongs to a subinterval of [−1, 1]. A better estimate could be difficult to find due to the complexity of the expression for V (z,D, λ). It is actually not needed when we do qualitative analysis of oscillations within a deforming potential. Here we wanted to see how the period of oscillations within the potential V (z,D, λ) depends on the value of Jellett’s integral λ, as stated in Proposition 3, in order to relate the time of inversion Tinv to this period. The dependence Tmax ∼ 1 b ∼ 1 λ implies that the frequency of oscillations within the potential behaves as 2π Tmax ∼ λ. For formulating a sufficient condition for having oscillating behaviour of θ(t) we need to know that there is an upper bound for the period of oscillations within V (z,D, λ). To find a universal bound independent of choice of λ > λthres could be difficult because the functions h1 and h2 have singularities at the boundary of the rectangle (z1, w) ∈ [−1, 1]× [−1, 1]. Finding a universal bound would require detailed analysis of the interdependence between z1 and w during inversion. Therefore we restrict our estimate to the region w ∈ [−1+ δ, 1− δ] with a certain suitable δ and ε < 1 C2 , meaning λ > Cλthres (see (24)). We consider period T given by (28) as a function of ε, w = a b and z1 ∈ [−1, 1], but we drop here the assumption ε→ 0. Let us take ε < 0.9, w ∈ [−1 + δ, 1− δ] with δ = 0.0001. These are physically well justified values since ε < 0.9 means λ > 1.054λthres and w = ±0.9999 corresponds to extremely vertical initial angular momentum L that is practically never taken by a toy TT. For (z1, w) ∈ Rδ = [−1, 1]× [−1+δ, 1−δ] with δ = 0.0001 and ε < 0.9, the function appearing in the calculation of 1√ z2−z3 (32): h3(z1, w) = ( 1 + (1− z21)2 (1 + w2 + 2wz1)2 ε2 − 2(1− z21)((1 + w2)z1 + 2w) (1 + w2 + 2wz1)2 ε )−1/4 , can be shown, by checking the maximum and minimum using a computer algebra system since h3 does not have singularities in Rδ, to satisfy 0 < h3(z1, w) < 3.15. Then we get 1√ z2 − z3 = √ ε 1√ 2 √ 1− z21 1 + w2 + 2wz1 h3(z1, w) ≤ 2.23 √ ε. Dynamics of an Inverting Tippe Top 17 The parameter k2 (30) can similarly be shown to satisfy 0 < k2 < 0.342 in this Rδ, which implies that the elliptic integral in (29) satisfies π 2 < K(k2) < 1.74. Thus for the period T (ε) we have the estimate T (ε) ≤ 2.23 · 1.74 4 √ g ( z∗ ) √ mgRα √ ε < 21.95 ( RI3γ(α+ 1− γ) b ) =: Tupp, provided that ε < 0.9. Notice that Tupp = 21.95 2π Tmax is (as expected) larger than Tmax that is providing a bound for the leading factor of T (ε). Proposition 4. Let (z1, w) ∈ Rδ with δ = 0.0001, ε < 0.9 so that b2 > 20β 9 , which cor- responds to λ > 1.054λthres. Then the period of oscillations is bounded by T (ε) < Tupp = 21.95 ( RI3γ(α+1−γ) b ) . The estimate Tupp is not the best possible estimate because it is taken uniformly over Rδ, but it is of the same order of magnitude as Tmax (Tupp ≈ 3.5Tmax) that also has been taken uniformly. For large angular velocities (implying small ε) Tmax provides better estimate of magnitude of period of oscillation. Real oscillations within V (z,D, λ), as observed in numerical simulations, for most of the time have shorter period. If the time of inversion is an order of magnitude larger than Tupp, say Tinv > 10Tupp, then θ(t) changes sign many times and it has oscillatory behaviour. It should be stressed that behaviour of a correctly built TT with 1− α < γ < 1 + α depends strongly on the initial conditions and it also depends on the friction function µ(s,L, 3̂,vA, t). The majority of solutions do not demonstrate inverting behaviour and even inverting solutions do not necessarily have to be oscillatory. The estimate found here provide a natural condition for the oscillatory behaviour of TT. Notice the dependence of Tmax ∼ 1 λ and Tupp ∼ 1 λ . A high initial angular momentum L with L almost parallel to the vertical axis ẑ implies a high value of Jellett’s integral λ = RI1ϕ̇ sin2 θ − RI3ω3(α − cos θ) since −RI3ω3(α − cos θ) > 0. Then the period of nutations becomes small and θ(t) performs many oscillations within a given inversion time Tinv. Example 2. By using again parameters from Example 1, we see that Tmax = 2πRI3γ(α+1−γ) b = 2.1816 · 10−8 · 2πb . Since b has the lower bound λγα 1+α = 1.4851 · 10−6 and the upper bound λγα 1−α = 2.7581 · 10−6, the maximal period of nutation satisfies 0.0497 < Tmax < 0.0923 seconds. 6 Conclusions An analysis of the dynamics of an inverting top has been performed in the special case 1− α < 1− α2 < γ < 1 < 1 + α and I1 = I23+mR 2I3(1−α2) I3+mR2 for which the effective potential V (z,D(t), λ) is a rational function of z = cos θ. We have shown that this potential is strictly convex, has only one minimum (Proposition 1) and that this minimum zmin moves from the interval [1 − ε, 1] to the interval [−1,−1 + ε] (Proposition 2). The high frequency period of oscillations Tmax and the maximal period of oscillations Tupp within the potential V (z,D, λ) has been estimated in Propositions 3, 4. When the time of inversion Tinv is an order of magnitude larger than the maximal period of oscillations Tmax within the potential V (z,D, λ), then the velocity θ̇(t) has to change sign many times and the solution is oscillatory. It is visible as nutational motion of the symmetry axis 3̂ on the unit sphere S2 between two latitudes while the nutational band moves from a neighborhood of the north pole into the neighborhood of the south pole when TT inverts. 18 S. Rauch-Wojciechowski and N. Rutstam Acknowledgements We would like to thank Hans Lundmark for discussions and for valuable help with numerical simulations. References [1] Bou-Rabee N.M., Marsden J.E., Romero L.A., Tippe top inversion as a dissipation-induced instability, SIAM J. Appl. Dyn. Syst. 3 (2004), 352–377. [2] Chaplygin S.A., On a motion of a heavy body of revolution on a horizontal plane, Regul. Chaotic Dyn. 7 (2002), 119–130. [3] Cohen R.J., The tippe top revisited, Amer. J. Phys. 45 (1977), 12–17. [4] Del Campo A.R., Tippe top (topsy-turnee top) continued, Amer. J. Phys. 23 (1955), 544–545. [5] Ebenfeld S., Scheck F., A new analysis of the tippe top: asymptotic states and Liapunov stability, Ann. Physics 243 (1995), 195–217, chao-dyn/9501008. [6] Erdélyi A., Magnus W., Oberhettinger F., Tricomi F.G., Higher transcendental functions, Vol. II, McGraw- Hill, New York – Toronto – London, 1953. [7] Glad S.T., Petersson D., Rauch-Wojciechowski S., Phase space of rolling solutions of the tippe top, SIGMA 3 (2007), 041, 14 pages, nlin.SI/0703016. [8] Jones E., Oliphant T., Peterson P., Open source scientific tools for Python, available at http://www.scipy. org. [9] Karapetyan A.V., Qualitative investigation of the dynamics of a top on a plane with friction, J. Appl. Math. Mech. 55 (1991), 563–565. [10] Karapetyan A.V., Rubanovskii V.N., On the stability of stationary motions of non-conservative mechanical systems, J. Appl. Math. Mech. 50 (1986), 30–35. [11] Or A.C., The dynamics of a tippe top, SIAM J. Appl. Math. 54 (1994), 597–609. [12] Rauch-Wojciechowski S., What does it mean to explain the rising of the tippe top?, Regul. Chaotic Dyn. 13 (2008), 316–331. [13] Rauch-Wojciechowski S., Sköldstam M., Glad T., Mathematical analysis of the tippe top, Regul. Chaotic Dyn. 10 (2005), 333–362. [14] Rutstam N., Study of equations for tippe top and related rigid bodies, Linköping Studies in Science and Technology, Theses No. 1106, Matematiska Institutionen, Linköpings Universitet, 2010, available at http: //swepub.kb.se/bib/swepub:oai:DiVA.org:liu-60835. [15] Rutstam N., Tippe top equations and equations for the related mechanical systems, SIGMA 8 (2012), 019, 22 pages, arXiv:1204.1123. [16] Rutstam N., High frequency behavior of a rolling ball and simplification of the separation equation, Regul. Chaotic Dyn. 18 (2013), 226–236. [17] Sturm C., Mémoire sur la résolution des équations numériques, Bull. de Ferussac 11 (1829), 419–425. [18] Ueda T., Sasaki K., Watanabe S., Motion of the tippe top: gyroscopic balance condition and stability, SIAM J. Appl. Dyn. Syst. 4 (2005), 1159–1194, physics/0507198. http://dx.doi.org/10.1137/030601351 http://dx.doi.org/10.1070/RD2002v007n02ABEH000199 http://dx.doi.org/10.1119/1.10926 http://dx.doi.org/10.1119/1.1934091 http://dx.doi.org/10.1006/aphy.1995.1097 http://dx.doi.org/10.1006/aphy.1995.1097 http://arxiv.org/abs/chao-dyn/9501008 http://dx.doi.org/10.3842/SIGMA.2007.041 http://arxiv.org/abs/nlin.SI/0703016 http://www.scipy.org http://www.scipy.org http://dx.doi.org/10.1016/0021-8928(91)90024-O http://dx.doi.org/10.1016/0021-8928(91)90024-O http://dx.doi.org/10.1016/0021-8928(86)90053-5 http://dx.doi.org/10.1137/S0036139992235123 http://dx.doi.org/10.1134/S1560354708040060 http://dx.doi.org/10.1070/RD2005v010n04ABEH000319 http://dx.doi.org/10.1070/RD2005v010n04ABEH000319 http://swepub.kb.se/bib/swepub:oai:DiVA.org:liu-60835 http://swepub.kb.se/bib/swepub:oai:DiVA.org:liu-60835 http://dx.doi.org/10.3842/SIGMA.2012.019 http://arxiv.org/abs/1204.1123 http://dx.doi.org/10.1134/S1560354713030039 http://dx.doi.org/10.1134/S1560354713030039 http://dx.doi.org/10.1137/040615985 http://arxiv.org/abs/physics/0507198 1 Introduction 2 The tippe top model 3 The main equation for the tippe top 3.1 The rational form of the METT 4 Convexity of the rational potential V(z,D,) 4.1 Estimates for position of minimum of V(z,D,) 5 Oscillation of (t) within the deforming rational potential V(cos,D(t),) 5.1 Estimates for the period of oscillation 6 Conclusions References

Dynamics of an Inverting Tippe Top

Репозитарії

Схожі ресурси