Periods of N N -Body Systems Determined through Dimensional Analysis
Dan Jonsson · 2026
Read Periods of N N -Body Systems Determined through Dimensional Analysis by Dan Jonsson (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-10559.txt.
Introduction
Jonsson derives Kepler style period laws for many body systems by dimension counting. His broadened dimensional analysis first recovers the two body law up to a constant. The same tool then stretches that law to n-body systems of both classical and quantum kinds. Results match Sun's classical guess and the Semay and Sun quantum guess. The work thus finds periods without ever solving the motion equations.
Catalog record
| CREATOR | Dan Jonsson |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65015 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.10559
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Periods of N N -Body Systems Determined through Dimensional Analysis Dan Jonsson Abstract . A generalization of classical dimensional analysis, presented in a separate article, makes it possible to derive Kepler’s third law without solving the equations of motion. Here we show how to derive generalizations of Kepler’s third law to n n -body systems by the same technique. Our results agree with conjectures by Sun on the period of a Newtonian n n -body system and by Semay and Sun on the quantum-theoretical counterpart of an n n -body system. 1. Introduction The classical theory of n -body systems concerns n point masses, “bodies” or “particles” \mathopen{}\mathclose{{\left\{p_{1},\ldots,p_{n}}}\right\} that move in space under influence of their mutual gravitational attraction in accordance with the laws of Newtonian mechanics. The equations of motion for a collisionless n -body system can be derived directly from the force law \mathbf{F}=m\mathbf{a} and Newton’s law of gravitation \mathbf{F}=G\frac{m_{1}m_{2}}{r^{2}}\mathbf{r} , where \mathbf{r} is a unit vector pointing in the same direction as \mathbf{F} . In vector form, the equations are (1.1) \begin{cases}m_{i}\frac{\mathrm{d}^{2}\mathbf{q}_{i}}{\mathrm{d}t^{2}}=G\sum_{j\in\mathbf{n}\setminus\mathopen{}\mathclose{{\left\{i}}\right\}}m_{i}m_{j}\frac{\mathopen{}\mathclose{{\left(\mathbf{q}_{j}-\mathbf{q}_{i}}}\right)}{\mathopen{}\mathclose{{\left\|\mathbf{q}_{j}-\mathbf{q}_{i}}}\right\|^{3}}&\mathopen{}\mathclose{{\left(i=1,\ldots,n}}\right)\end{cases} where \mathbf{n}=\mathopen{}\mathclose{{\left\{1,\ldots,n}}\right\} , t is time, G the gravitational constant, m_{i} the mass of p_{i} and \mathbf{q}_{i} its radius vector as a function of t . For a spatially bounded n -body system there may exist some time T_{0} such that at a later time T_{0}+T each body has the same position and velocity as at T_{0} . Then it has the same position and velocity at any time T_{0}+nT for any positive integer n since the equations of motion determine the future trajectory of each body given only its position and velocity at an arbitrary time. We call a trajectory followed from T_{0} to T_{0}+T an orbit , an n -body system with trajectories made up of orbits an orbital (periodic) n -body system, and T the period of the orbital system. In the two-body case, we obtain, under suitable initial conditions, the familiar elliptical orbits, with the two bodies in addition obeying Kepler’s second and third laws. Kepler’s third law concerns the period of an orbital two-body system. It can be shown from the equations of motion that (1.2) T^{2}=4\pi^{2}a^{3}G^{-1}\mathopen{}\mathclose{{\left(m_{1}+m_{2}}}\right)^{-1}, where T is the period and a the semi-major axis of the elliptical orbit [ 1 , eq. 2.37] . Let us regard two bodies in an n -body system from the point of view of kinematics as a two-body system where one body p_{i} orbits a central body p_{j} , the radius vector of which is set to \mathbf{0} . The mean extremal distance between p_{i} and p_{j} is defined as (1.3) \mu_{ij}=\frac{\mathopen{}\mathclose{{\left\|\mathbf{r}_{a}-\mathbf{0}}}\right\|+\mathopen{}\mathclose{{\left\|\mathbf{r}_{p}-\mathbf{0}}}\right\|}{2}, where \mathbf{r}_{a} and \mathbf{r}_{p} are the radius vectors of the apoapsis and periapsis, respectively, of p_{i} . In the proper two-body case, with elliptical orbits, we have \mu_{12}=\mathopen{}\mathclose{{\left\|\mathbf{r}_{a}-\mathbf{r}_{p}}}\right\|/2 and a=\mu_{12} . Also, a is a characteristic distance , that is, T depends separately on the mean extremal distance between the bodies (up to a multiplicative constant). It turns out that T depends on any d\not\propto a (for example, the semi-minor axis) only in conjunction with a dimensionless shape parameter s (such as the eccentricity of the orbital ellipse), so that T^{\mathfrak{K}}=\varphi\mathopen{}\mathclose{{\left(m_{1},m_{2},a,G}}\right) becomes T^{\mathfrak{K^{\prime}}}=\varphi^{\prime}\mathopen{}\mathclose{{\left(m_{1},m_{2},d,s,G}}\right) . Let E be the total mechanical energy of an orbital n -body system, normalized by letting its potential energy tend to zero as all distances between bodies approach infinity. Then it can be shown that in the two-body case we have (1.4) E=-\frac{Gm_{1}m_{2}}{2a}. E needs to be negative since otherwise we would not have spatially bounded trajectories, so that T would not be defined. ( 1.4 ) implies that a^{3}=\frac{\mathopen{}\mathclose{{\left(-E}}\right)^{3}G^{3}\mathopen{}\mathclose{{\left(m_{1}m_{2}}}\right)^{3}}{8} , and substituting this in ( 1.2 ) gives (1.5) T^{2}=\frac{\pi^{2}}{2}\mathopen{}\mathclose{{\left(-E}}\right)^{-3}G^{2}\mathopen{}\mathclose{{\left(m_{1}m_{2}}}\right)^{3}\mathopen{}\mathclose{{\left(m_{1}+m_{2}}}\right)^{-1}. Unfortunately, the equations of motion ( 1.1 ) do not admit closed solutions for n>2 . Even the Newtonian three-body problem is much more difficult than the two-body problem. While the latter was solved completely by Newton [ 2 ] and Johann Bernoulli [ 3 ] , the former is a famous problem that has remained recalcitrant despite efforts by Newton himself, Euler [ 4 ] , Lagrange [ 5 ] , Poincaré [ 6 ] and many others; it is still an active area of research. Powerful computers have made it practicable to generate numerical solutions of selected equations of motion, but the theoretical understanding of the problem is still a work-in-progress. Recently, BoHua Sun [ 7 ] has formulated a bold conjecture which relates to Kepler’s third law, namely a generalization of ( 1.5 ). He suggests that the orbital period of a three-body system is given by the formula (1.6) T^{2}=\frac{\pi^{2}}{2}\mathopen{}\mathclose{{\left(-E}}\right)^{-3}G^{2}\mathopen{}\mathclose{{\left(\mathopen{}\mathclose{{\left(m_{1}m_{2}}}\right)^{3}+\mathopen{}\mathclose{{\left(m_{1}m_{3}}}\right)^{3}+\mathopen{}\mathclose{{\left(m_{2}m_{3}}}\right)^{3}}}\right)\mathopen{}\mathclose{{\left(m_{1}+m_{2}+m_{3}}}\right)^{-1}, and, more generally, that the period of an orbital n -body system is given by 1 1 1 There is an inaccuracy in [ 7 ] , where the double sum is written as \sum_{i=1}^{n}\sum_{j=i+1}^{n}\mathopen{}\mathclose{{\left(m_{i}m_{j}}}\right)^{3} . (1.7) T^{2}=\phi_{T}\mathopen{}\mathclose{{\left(m_{1},\ldots,m_{n},E,G}}\right)=\frac{\pi^{2}}{2}\mathopen{}\mathclose{{\left(-E}}\right)^{-3}G^{2}\sum_{i=1}^{n-1}\sum_{j=i+1}^{n}\mathopen{}\mathclose{{\left(m_{i}m_{j}}}\right)^{3}\mathopen{}\mathclose{{\left(\sum_{i=1}^{n}m_{i}}}\right)^{-1}. In addition to the analogy between ( 1.5 ) and ( 1.7 ), this conjecture is supported by known numerical solutions [ 8 , 9 ] of equations of motion describing certain three-body systems. Let \sigma be a permutation of 1,\ldots,n and \overline{\sigma}:p_{i}\mapsto p_{\sigma\mathopen{}\mathclose{{\left(i}}\right)} a corresponding relabeling of the bodies in the n -body system \mathopen{}\mathclose{{\left(p_{1},\ldots,p_{n}}}\right) . Each \overline{\sigma} generates permutations m_{i}\mapsto m_{\sigma\mathopen{}\mathclose{{\left(i}}\right)} and \mathbf{q}_{i}\mapsto\mathbf{q}_{\sigma\mathopen{}\mathclose{{\left(i}}\right)} of the n -tuples of scalars \mathopen{}\mathclose{{\left(m_{1},\ldots,m_{n}}}\right) and vector-valued functions \mathopen{}\mathclose{{\left(\mathbf{q}_{1},\ldots,\mathbf{q}_{n}}}\right) , respectively. It is shown in Section 2.2 that \overline{\sigma} does not affect the system of equations of motion and their initial conditions, and that therefore \phi_{T} , defined by ( 1.7 ), must satisfy (1.8) \displaystyle\phi_{T}\mathopen{}\mathclose{{\left(m_{1},\ldots,m_{n},E,G}}\right)=\phi_{T}\mathopen{}\mathclose{{\left(m_{\sigma\mathopen{}\mathclose{{\left(1}}\right)},\ldots,m_{\sigma\mathopen{}\mathclose{{\left(n}}\right)},E,G}}\right) for any permutation \sigma of 1,\ldots,n , as ( 1.7 ) indeed does. A deeper question concerns the uniqueness of those functions \phi_{T} that satisfy the symmetries in ( 1.8 ). To the extent that \phi_{T} is shown to be unique by dimensional analysis, there exists theoretical support from this point of view for Sun’s conjecture. The main goal of this article is to investigate the uniqueness of \phi_{T} , using augmented dimensional analysis [ 11 ] , a generalization of dimensional analysis that, in contrast to classical dimensional analysis, takes symmetries of the kind considered here into account. Hence, information can be extracted from the parameters of an n -body system without actually solving it, evading the obstacle that such solutions are typically not known for n>2 . Augmented dimensional analysis is surveyed in Section 2 , where we also discuss the symmetry assumptions used. In Sections 3, 4 and 5, we apply dimensional analysis to two-body systems, three-body-systems and n -body systems, respectively, providing support for uniqueness. In Section 6, an assumption used in generalizations to the n -body case is discussed, and in Section 7 we derive the values of coefficients of proportionality. In Section 8 four main generalizations to the n -body case are highlighted and compared to results obtained by Sun [ 7 ] and Semay [ 12 ] . Section 9 contains final comments on the uniqueness of the equations derived. 2. Preliminaries 2.1. Augmented dimensional analysis Dimensional analysis can be described as a method of transforming ”physically meaningful” equations of the form t=\phi\mathopen{}\mathclose{{\left(t_{1},\ldots,t_{n}}}\right), describing relationships among quantities, into equivalent but more informative equations, using data about the dimensions of t,t_{1},\ldots,t_{n} . This can be done by representing the unknown function \phi as