My Math Forum Acceleration components of gravity vector

 Geometry Geometry Math Forum

 May 19th, 2016, 03:28 AM #1 Newbie   Joined: Jun 2014 From: UK Posts: 20 Thanks: 0 Acceleration components of gravity vector Hello, for "fun", I thought I'd have a go at modelling the orbit of planets around the sun in Excel. I'm trying to complete this worksheet I found online: Building Planetary Orbits I've stumbled at the first hurdle, though; the solution requires me to devise equations for the $\displaystyle x$ and $\displaystyle y$ components of acceleration of a planet orbiting the Sun, from the equation for vector gravitational force: $\displaystyle \vec{F_g}=-G\frac{M.m}{r^2}\hat{r}$ ...where $\displaystyle G$ is a gravitational constant, $\displaystyle M$ is the mass of the Sun, $\displaystyle m$ is the mass of the planet, $\displaystyle \hat{r}$ is the unit vector between the Sun and the planet and $\displaystyle r$ is the radius of the planets orbit around the Sun. I've come up with a solution (because I had another model that did something similar, so I just copied that) but I don't understand why it works: $\displaystyle a_x=\vec{F_g}\frac{x}{r}$ $\displaystyle a_y=\vec{F_g}\frac{y}{r}$ ...where $\displaystyle x$ and $\displaystyle y$ are the planets position in 2D space - I have a known starting position. So...why do these two equations give me acceleration? I appeciate that this might be a very elementary maths question and, before anybody suggests it, I have an introductory book on orbital mechanics on order from Amazon Any help would be appreciated, though, Thanks
 May 19th, 2016, 06:18 AM #2 Math Team   Joined: Jan 2015 From: Alabama Posts: 3,264 Thanks: 902 What you have, $\displaystyle a_x= \vec{F_g}\frac{x}{r}$ and $\displaystyle a_y= \vec{F_g}\frac{y}{r}$, are clearly incorrect for two different reasons. I presume you know that "force equals mass times acceleration" so that "acceleration equals force divided by mass". Since force is proportional to mass, acceleration due to gravity does not depend upon mass. But M is still part of $\displaystyle \vec{F_g}$ so that your components of acceleration are proportional to mass. Second, since both $\displaystyle a_x$ and $\displaystyle a_y$ are numbers times $\displaystyle \vec{F_g}$ you have both components pointing parallel to the force vector not parallel to the x and y axes as they should be. Thanks from eigenvexed
 May 19th, 2016, 06:33 AM #3 Newbie   Joined: Jun 2014 From: UK Posts: 20 Thanks: 0 Well, that's a very definite $\displaystyle WRONG!$ then - thanks for clearing that up Can you explain why they work? I use them to model the path of a planet around the Sun and the orbit is as expected... More helpful, though, would be a suggestion as to what the formulae for $\displaystyle a_x$ and $\displaystyle a_y$ should be... Thanks for your thoughts on this
 May 19th, 2016, 08:07 AM #4 Newbie   Joined: Jun 2014 From: UK Posts: 20 Thanks: 0 OK... I've had a sit and a think about this (plus I read all of this - page 19 is very helpful!). I've revised my calculations as follows: Because $\displaystyle r=\sqrt{r_x^2+r_y^2}$ and $\displaystyle \hat{r}=\frac{\vec{r}}{r}$, we can derive $\displaystyle \hat{r_x}=\frac{r_x}{r}$ and $\displaystyle \hat{r_x}=\frac{r_x}{r}$ So $\displaystyle \vec{F_g}=\frac{G.M.m}{r^2}$ can be expanded into its x/y components as follows: $\displaystyle F_g(x)=\frac{-r_x}{r}.F_g$ and $\displaystyle F_g(y)=\frac{-r_y}{r}.F_g$ where $\displaystyle r_x$ and $\displaystyle r_y$ are the $\displaystyle x,y$ co-ordinates of the planet. Thus, as we know $\displaystyle \vec{a}=\frac{\vec{F_g}}{M}$; $\displaystyle a_x=\frac{F_g(x)}{M}$ and $\displaystyle a_y=\frac{F_g(y)}{M}$ I've substituted these equations into my model and they also return the orbits expected - hooray! The reason my model worked was because the mass $\displaystyle M$ I'm using is set to 1 for simplicity (i.e. the Sun=1 Solar Mass), so $\displaystyle \frac{F_g}{1}=F_g$ in this case... I'm off to read the rest of this Newtonian Mechanics paper, anyway... Thanks Last edited by eigenvexed; May 19th, 2016 at 08:10 AM. Reason: A sort of sudden epiphany...
May 19th, 2016, 09:40 AM   #5
Math Team

Joined: Jul 2011
From: Texas

Posts: 3,018
Thanks: 1603

sorry, but $a \ne \dfrac{F_g}{M}$

$a = \dfrac{F_g}{m}$, where $m$ represents the planetary mass.

$F_g = -\dfrac{GMm}{r^2} \hat{r}$

$ma_r = -\dfrac{GMm}{r^2}$

$a_r = -\dfrac{GM}{r^2}$

$a_x = a_r \cos{\theta} = -\dfrac{GM \cos{\theta}}{r^2}$

$a_y = a_r \sin{\theta} = -\dfrac{GM \sin{\theta}}{r^2}$

where $\theta$ is the angle between positive x-axis and the orbital radius
Attached Images
 orbit_calcs.jpg (16.9 KB, 2 views)

 May 19th, 2016, 10:51 PM #6 Newbie   Joined: Jun 2014 From: UK Posts: 20 Thanks: 0 Ah, Ok - to be honest, I found the use of $\displaystyle M$ and $\displaystyle m$ a bit ambiguous in what I was reading - thanks for clearing that up. The original equation for the gravitational force vector is: $\displaystyle \vec{F_g}=G\frac{M.m}{r^2}$ I've read that, where $\displaystyle M$ is significantly greater than $\displaystyle m$, a gravitational field around the larger mass can be defined as $\displaystyle g=-\frac{GM}{r^2}\hat{r}$ which is where the equations I'm using then derive from - perhaps this is why they work, despite my transposing $\displaystyle M$ and $\displaystyle m$...? The equations... $\displaystyle a_x=a_rcos\theta$ and $\displaystyle a_y=a_rsin\theta$ ...were my first thought on how to approach this, too - the problem I found was that there was no easy way of calculating $\displaystyle \theta$...how do you do it? Thanks for your help
 May 20th, 2016, 01:26 AM #7 Math Team     Joined: Jul 2011 From: Texas Posts: 3,018 Thanks: 1603 you don't need to calculate $\theta$ ... $a_x=a_r\cos{\theta}=a_r \cdot \dfrac{x}{r} = -\dfrac{GM}{r^2} \cdot \dfrac{x}{r} = -\dfrac{GMx}{r^3}$ similarly, $a_y=-\dfrac{GMy}{r^3}$ Thanks from eigenvexed

 Tags acceleration, components, gravity, vector

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post jemes Physics 3 December 5th, 2014 01:54 AM mac_alleb Linear Algebra 4 February 1st, 2014 06:57 PM micle Physics 1 June 17th, 2013 10:17 PM Sefrez Physics 1 March 18th, 2010 05:41 PM rmas Advanced Statistics 1 January 1st, 2010 12:57 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top