﻿

asp.net mvc 4 and the web api pdf free download : Add multiple pdf files into one online control Library platform web page asp.net winforms web browser mechanics21-part1292

6|Orbits
207
4.
5.
6.
7.
8.
Fig.6.7
Youcanseefromthesepicturesthatthereisnostableorbit.Acircularorbitisjustzerodistance
awayfromanorbitthateithercrashesintothecenteror iesaway.Theinversecubeorbitiswhatyou
shouldexpectinauniversewithfourdimensionsofspace.Foracontrast,lookattheeectivepotential
energygraphfor
F
/1
=r
andyouwillseethatnothing canescapeanythingelseandallstarswould
coalesce. Thisisthepotentialyougetintwodimensions,sothreeisthemagicnumber.Wegotlucky.
Graph
U
eective
fortheinversecubeforce.Thatwillexplainalot.
6.10HyperbolicOrbits
IntheanalysisoftheKeplerorbits,equations(6.15)and(6.16),theeccentricity
waslessthanone.
Whatifitisn’t?ThepolarequationfortheellipsefromEq.(6.17)becomestheequationforahyperbola,
thoughitneedsasignchangewhen
>
1.
r
=
a
(
2
1)
1+
cos
Nowwhen
>
1thedenominatorcan vanish. . Thatisfor
=cos
1
( 1
=
),. Theseangleshave
>=
2and
<
=
2atthesymmetricpositions. Toverifythatthisisahyperbola,putitinthe
morefamiliarrectangularform.
r
+
r
cos
=
a
(
2
1)  !
p
x
2+
y
2+
x
=
a
(
2
1)
x
2
+
y
2
=
a
(
2
1)
x
2
=
a
2
(
2
1)
2
2
a
(
2
1)
x
+
2
x
2
x
2
(1
2
) 2
a
(1
2
)
x
+(1
2
)
a
2
2
(1
2
)
a
2
2
+
y
2
=
a
2
(
2
1)
2
(1
2
)(
x
a
)
2
+
y
2
=
a
2
(
2
1)
2
+
a
2
(1
2
)
2
a
2
(
2
1)
(
x
a
)
2
a
2
y
2
a
2(
1)
=1
Add multiple pdf files into one online - Merge, append PDF files in C#.net, ASP.NET, MVC, Ajax, WinForms, WPF
Provide C# Demo Codes for Merging and Appending PDF Document
apple merge pdf; pdf merge files
Add multiple pdf files into one online - VB.NET PDF File Merge Library: Merge, append PDF files in vb.net, ASP.NET, MVC, Ajax, WinForms, WPF
VB.NET Guide and Sample Codes to Merge PDF Documents in .NET Project
break a pdf into multiple files; break pdf file into multiple files
6|Orbits
208
Fig.6.8
The hyperbola has two branches,but only one applies tothe orbit. . The e other r branch h is the
dashedcurve,anditwillnotbeneededuntilproblem6.54. Inordertodrawthesolidcurveseasily,you
canparametrizethemby
x
=
a
a
cosh
;
y
=
a
p
2
1sinh
(6
:
57)
Forthe other(dashed)branch,changethe signinfrontofthe cosh
in this expression for
x
. The
orbits drawn herehave eccentricities1.15, , 1.3, 1.7|whichiswhich? ? The e orbits of comets can be
ellipseswithverylargeeccentricity,sothattheywillreturnevery76or10000yearsasthecasemaybe.
Cometary orbitscanalsobehyperbolassuchasthese,andsuchacometwillthenappearonlyonce,
thenleaveforever.Thesehyperbolicorbitscorrespondtotheenergy
E
3
atEq.(6.24).
6.11TimeDependence_
AllthecalculationsconcerningtheKeplerproblemresultedintheshapeoftheorbit,buttheydidn’t
givethepositionasafunctionoftime. ThereasonI’veputthatquestionoisthatthereisnoneat,
explicitsolutiontotheequationsintermsoftime. Thereishoweverasolution,justnotanexplicitone.
Thisprocedurewill not produce equationsfor
r
and
asfunctionsof
t
. Insteadyouwill l get
threeequationsthatgiveyou
r
,
,and
t
asfunctionsofafourthparameter.
Thetypicalwaytosolvefor
r
(
t
)fromEq.(6.10)istogettheenergyintegralandthentouse
separationofvariables. Theprocedurehereisclosetothat,butwithatwist. . Theenergyintegralis
Eq.(6.22),andaftersolvingfor _
r
itisseparable,justasinEq.(2.23). Ifyouproceedalongthisroute
theintegraltodotond
t
(
r
)wouldbe
E
=
1
2
m
_
r
2
+
m‘
2
2
r
2
GMm
r
!
dt
=
r
m
2
dr
r
E
m‘
2
2
r
2
+
GMm
r
(6
:
58)
Todotheintegralyoucanmultiplythenumeratorand denominatorofthe
dr
integralby
r
,making
thebig square rootaquadraticpolynomial. . Youthencompletethe e square on thatpolynomial and
integrate. That t will give you
t
(
r
), not
r
(
t
), and you can’tinvertthe resulting expression to get
r
explicitly. Thecleverthingtodoistoredenetheindependentvariable
t
. Indthesimplestwayto
presentitfollowsSundman*andchangesvariablesintheenergyequation.
* Itissometimessaidthatthegeneralsolutionforthetimedependenceofthreemutuallyorbiting
masses is impossible|the three e body y problem. . Poincare e did prove e that t it is s impossible e using the
particularmethodsthatheapplied,butSundmanshowedthatthereareotherways,demonstratinghow
tond aseriessolutioninpowersof
t
1
=
3
,albeitaveryslowlyconvergingone. The
n
-bodysolution
wasn’tfounduntilthe1990’s.
Online Merge PDF files. Best free online merge PDF tool.
Merge PDF, Multiple PDF files. Then press the button below and download your PDF. Also you can add more PDFs to combine them and merge them into one single
combine pdf files; pdf merger online
VB.NET TWAIN: Scanning Multiple Pages into PDF & TIFF File Using
one convenient multi-page document file, like PDF and TIFF This VB.NET TWAIN pages scanning control add-on is an efficient solution to scan multiple pages into
add pdf files together online; best pdf combiner
6|Orbits
209
Doingthis,itiseasiertoworkwiththeenergyequationfromthersthalfofEq.(6.58)andnot
themanipulatedandseparatedformthatisthesecondhalf.Changevariablesfrom
t
to
s
,where
ds
dt
=
r
then
dr
dt
=
dr
ds
ds
dt
=
r
dr
ds
:
Let
r
0
=
dr
ds
E
=
m
2
2
r
2
r
0
2
GMm
r
+
m‘
2
2
r
2
!
1
2
m
r
0
2
GMm
2
r
+
m‘
2
2
2
=
E
2
r
2
(6
:
59)
This is s aharmonic c oscillatorif
E <
0. The e parameter
is still arbitrary and Ican choose it for
convenience. Twoplausiblechoicesare
=1and
=
p
GM
,butIwillletthechoicehangfornow
becauseitisjustaseasytocarrythe
alonguntilyouareforcedtodecide. Isthisreallyaharmonic
oscillator?Yes,justcompletethesquare(remember,
E<
0).
1
2
m
r
0
2
E
2
r
+
GMm
2
E
2
m‘
2
2
2
G
2
M
2
m
2
4
E
2
(6
:
60)
Thisisinthesameformastheenergyconservationequationthatyou’veseenforamassona
spring:
1
2
mv
2
+
1
2
k
(
x
x
0
)
2
=
E
0
=)
x
(
t
)=
x
0
+
A
cos(
!t
+
)
; !
2
=
k=m; kA
2
=
2=
E
0
With this you can read the frequency (
p
k=m
), the center r (
x
0
) and d the e amplitude of oscillation
(
A
=
p
2
E
0
=k
).Thetranslationtableforthepresentcaseisthen
x
(
t
)!
r
(
s
)
; m
!
m; k
! 2
E=
2
; x
0
GMm=
2
E; E
0
m‘
2
2
2
G
2
M
2
m
2
4
E
2
LookbackagainatthepictureofanellipseatEq.(6.17). Thevalueof
r
variesfromamaximumof
a
+
f
downtoaminimumof
a
f
.Translatethisintothelanguageofaharmonicoscillatorandthat
says
x
0
!
a
and
A
!
f
. Theequationfor
r
(
s
)isthen
r
r
(
s
)=
a
f
cos
!
(
s
s
0
)
(6
:
61)
Translatefurther:
!
2
=
k
m
2
E
m
2
;
x
0
!
a
GMm
2
E
;
A
2
=
2
E
0
k
!
f
2
=
2
m‘
2
2
2
2
G
2
M
2
m
2
4
E
2

2
E=
2
=
m‘
2
2
E
+
(
GMm
)
2
4
E
2
(6
:
62)
t
?Take
s
=0when
t
=0,andforconveniencechoose
s
0
=0.
t
ds
dt
=
r
!
dt
=
rds
=(
a
f
cos
!s
)
ds
!
t
(
s
)=
1
as
f
!
sin
!s
(6
:
63)
r
2
_
=
‘;
so
d
dt
=
d
ds
ds
dt
=
d
ds
r
=
r
2
!
0
=
r
(
s
)=
Z
s
0
ds
1
a
f
cos
!s
=
2
!b
tan
1
b
a
(1
)
tan(
!s=
2)
(6
:
64)
=
2
!b
tan
1
r
1+
tan(
!s=
2)
!
=
2
!b
tan
1
cot(
)tan(
!s=
2)
=
2
!b
sin
1
0
@
cos(
)sin(
!s=
2)
q
sin
2
(
)cos
2
(
!s=
2)+cos
2
(
)sin
2
(
!s=
2)
1
A
VB.NET PDF File Split Library: Split, seperate PDF into multiple
that they can split target multi-page PDF document file to one-page PDF Add necessary references Split PDF Document into Multiple PDF Files Demo Code in VB.NET.
c# combine pdf; c# pdf merge
C# Create PDF from images Library to convert Jpeg, png images to
multiple image formats into one or multiple PDF file in C# List<Bitmap> images = new List<Bitmap>(); images.Add(new Bitmap shows how to build a PDF document with
batch pdf merger online; add two pdf files together
6|Orbits
210
Here
b
istheusualsemi-minoraxisoftheellipse,
b
2
=
a
2
f
2
and
f
=
a
andcos2
=
andRyzhik2.553.3)
r
t
s
!
r
=
a
+
f
r
=
a
f
=
c
=0
=2
,
t
=0,
t
=1yr
Fig.6.9
Thesethreegraphsshow
r
,
,and
t
asfunctionsof
s
foranorbitwitheccentricity0.66asin
orbit: theperihelion. . Forminimum
r
=
a
f
,
s
iszero|theleftendofthegraph. Whatis
d=dt
atthatpoint?
d
dt
=
d
ds
ds
dt
=
d
ds
dt
ds
Youcanseethatattheleft(orright)endsofthegraphtheslopeofthe
t
-curveisatitssmallest
andtheslopeofthe
-curveisatitslargest.Thisgives
_
itslargestvaluethere,atperihelion,justas
Kepler’slawsays.Theplanetdoesn’tspendmuchtimeneartheperihelion. Attheotherextreme,the
aphelioninthemiddle,youcanagainlookattheslopesandseethat
_
issmallestthere.
Whyaretheresomanydierentversionsoftheresultfor
(
s
whensolvingacomplicatedproblemyoudon’timmediatelyknowwhichformoftheresultwillbethe
mostuseful. Itisprudenttowriteitinseveraldierentwaysuntilyougetabetterideaoftherouteto
take. Forexample,there’saminortechnicaldicultywhenplottingthegraphof
ifyouusetherst
form.Thearctangentismultiplevalued,soyouhavetousesomecareinevaluatingit. Ifyouuseany
oftherstthreeversionsyoumayneedtowritethetangentas(sineovercosine)andthentotreatthe
youarestarttolookatunbound(
E>
0)orbits.
Doesthiscomplicatedlookingsetofequationsreducetothecorrectresultswhentheorbitisa
circle?
=0,
b
=
a
,
f
=0,so
r
(
s
)=
GMm
2
E
=
a;
t
(
s
)=
1
as;
(
s
)=
2
!a
tan
1
tan(
!s=
2)=
a
s
Thestatementthat
A
=0(novariationin
r
)determines
.
A
2
=
m‘
2
2
E
+
(
GMm
)
2
4
E
2
=0  !
2
G
2
M
2
m
2
E
!
=
GM
p
2
E=m
d
dt
=
(
‘= a
)
ds
(
a=
)
ds
=
a
2
andthatis(
r
2
_
=r
2
)asitshouldbe.
VB.NET Create PDF from images Library to convert Jpeg, png images
Turn multiple image formats into one or multiple PDF file. images As New List(Of REImage) images.Add(New REImage example shows how to build a PDF document with
reader combine pdf pages; attach pdf to mail merge
C# PDF: C#.NET PDF Document Merging & Splitting Control SDK
C#.NET Sample Codes to Merge Multiple PDF Files. C#.NET PDF Merger & Splitter SDK FAQs. Q 1: Using this C#.NET PDF merger & splitter control add-on, can I
reader merge pdf; acrobat reader merge pdf files
6|Orbits
211
6.12PerihelionofMercury_
Thissectionismoredemandingthanmostofthistext,butitshowsthepowerofthetoolsthatyou’ve
studied. Thesolarsystemisawhole e lotmorecomplicated than anythingthatdone sofar,andthis
sectiondevelopsoneaspectofthiscomplexity.
1. Thegravitationalforcesarenotjustfromthesun.
2. Theplanetsmoveonlyapproximately inclosedorbitsalongellipses.
3. Itispossibletogureouttheeectsthattheplanetshaveoneachother,atleasttoaverygood
approximation.(Wecan’tdoeverythinghere.)
Thedevelopmentofperturbedorbitsafewsectionsbackhas anim-
portant application thatatrstmay seem impossibletodo: : Precession n of
theperihelion of Mercury. . Each h planetispulledbythe sun, butitisalso
pulledbyalltheotherplanets,andthatwillaecttheshapesoftheirorbits.
Mercury’sorbitinparticularwas foundtohavemeasurabledeviationsfrom
asimple ellipse. . Its s orbit precessed by y an n amount that was large enough
to be measured as s far back k as 1800, , and this s measurement is associated
withasignicantmilestoneinthehistoryofphysicsbecausethecalculations
of the precession and the measurements of the precession did not agree.*
precise methodstocalculatethe predictedmotionshadbeen developed by then,andformore than
halfacenturythedisagreementremainedavexingproblem. OnlywhenEinsteinpublishedhistheory
ofgravity(general relativity)in1916was theproblemresolved. . Thedrawingis s greatly exaggerated,
bothintheeccentricityofMercury’sorbitandinitsrateofprecession. Thedotsarenearwherethe
perihelionisontheseorbits. TheangleofprecessioninoneEarthcenturyis
532
00
(Newtonian)+43
00
(generalrelativity)=575
00
(measured)
Howdidthey computethisprecession? ? That’swhatIwanttodohere|atleastinpart. . The
person whofound the discrepancy y between n theory and observation was Urbain Le Verrier,and you
predictednotonlytheexistence,butthepositionofNeptune.)
Theperturbationmethodsdescribedsofarinthischapterinvolvedmakingsmallchangesinthe
potential energy andinthe correspondingradialforce,
F
r
(
r
) =
dU
(
r
)
=dr
. In n all casesthesmall
change involved addingsomething that was spherically symmetric. . All l the changes to
U
(
r
anothersmall term
U
1
(
r
),andwhetheritwas acloud ofdustinthesolarsystemorabulgeatthe
Sun’s equatorit was symmetric. . The e pull on Mercury by theplanetsVenus orJupiteris not atall
symmetric;theirattractionsarenottowardtheSunandtheyareverymuchtime-dependent. Sonow
what?
Thereis aclevertrickdue toGauss. . Henoted d thatplanets orbit the Sun in periodsthatare
measuredinyears|1
=
4
yearinthecaseofMercury,165yearsforNeptune. Themeasuredprecession
:
16
inoneEarthcentury,andthisisanextremelyslowmotioncompared
tothatofanyplanet,sowhynotsmearoutandaveragethepositionsoftheplanets?They’rewhizzing
aroundso(comparatively)fastthattheirdetailedpositionasafunctionoftimeshouldn’tmatter,only
theirorbit. Theaverageinthiscasemeansthatyoureplacetheorbitofagivenplanetby y aringof
uniformlinearmassdensity,thentheeectofsucharingonthemotionofMercurywillbesymmetric.
HereIwillmaketheprettygoodsimplifyingapproximationthatalltheplanetaryorbits(exceptMercury)
are circles andallofthem orbitin exactly the sameplaneas Mercury. . Renementsbeyond this are
moredicultandI’llsavementionofthemuntiltheend. Itisquiteamazinghoweverjusthowclose
* Close,butnocigar.
C# Create PDF from CSV to convert csv files to PDF in C#.net, ASP.
multiple sheets CSV file to one PDF or splitting to multiple PDF documents. Add necessary references: Description: Convert to PDF/TIFF and save it into stream.
pdf combine files online; acrobat merge pdf files
VB.NET Word: Merge Multiple Word Files & Split Word Document
Split Word File(s) Created by Multiple Microsoft Word NET Word combining and splitting add-on allows controls, PDF document, image to pdf files and components
merge pdf files; pdf split and merge
6|Orbits
212
Thisisnow,atleastinprinciple,thesamesortofperturbationproblemdevelopedearlierinthis
chapter. The e only initialdicultyisinguringouttheeld ofaringofmass. . The e answerforthe
potentialenergyis
U
1
(
r
)=
GMm
2
Z
2
0
d
1
p
R
2+
r
2
Rr
cos
;
V
1
=
U
1
m
(6
:
65)
M
isthe massoftheouterplanet;
R
is its orbital radius;
m
is Mercury’smass,
U
is the potential
energy,and
V
isthegravitationalpotential. This
r
isintheplaneofthering,andthedenominator
in theintegral comesfrom usingthe lawofcosines. . Don’t t beintimidated. . This s sortofcalculation
becomesfairlyroutineonceyou’vetakenoneE&Mcoursebeyondtherst.(
dM
=
Md=
2
)
Asacheckonthisexpressionforthepotential,whatisitat
r
=0?*
V
1
(0)=
GM
2
Z
2
0
d
1
p
R
2
GM
2
2
R
GM
R
and since the entire mass
M
is at the constant distance
R
, this is s correct. . Ratherthan n spending
time discussinghowtoevaluatethisintegral,Iwill state the resultforitslowest orderpowerseries
representation. Attheendofthissectionyouwillndamorecompleteresult.Outtotermsin
r
2
this
potentialis(seeproblem6.70forhowtodothis.)
V
1
(
r
)
GM
R
1+
r
2
4
R
2
;
U
1
(
r
)=
mV
1
(
r
)
; F
r
(
r
)=
dU
1
dr
+
GMm
2
R
3
r
(6
:
66)
Thenextpicturesshowsaviewfromabovetheplaneoftheplanet’sorbitontherightandasideview
oftheorbitontheleft.Italsoshowsagraphof
V
(
r
),andyoucanseebothitscurved(
r
2
)shapenear
r
=0anditsbehaviornear
r
=
R
. IchosetolabelitastheorbitofVenus,butitcouldaseasilybe
anyoftheotherplanetsfartherout.
Fig.6.10
TopView
SideView
V
(
r
)
R
OrbitofVenus
graphtoputMercury. Italsomeans s thatthebehaviorof
V
near
r
=
R
will notberelevanttothe
problem.
NowfortheeectofthisextraforceontheplanetMercury,andthistakesusbacktosection
6.5. ThecoreequationisEq.(6.14),where
u
=1
=r
and
M
isthesolarmass. Also,
f
dU=dr
f
(
r
)=
GM
m
r
2
+
GMm
2
R
3
r
GM
mu
2
+
GMm
2
R
3
u
d
2
u
d
2
+
u
1
m‘
2
u
2
f
1
u
=
GM
2
GM
2
2
R
3
u
3
* andwhatisitfor
r
R
?
6|Orbits
213
u
(
)=
u
0
+
x
(
),
with
x
u
0
=1
=r
0
,andkeepingtermsonlytorstorderin
x
.
d
2
(
u
0
+
x
)
d
2
+
u
0
+
x
=
GM
2
GM
2
R
3(
u
0
+
x
)3
d
2
x
d
2
+
u
0
+
x
=
GM
2
GM
2
2
R
3
u
3
0
1 3
x
u
0
!
u
0
=
GM
2
GM
2
2
R
3
u
3
0
and
d
2
x
d
2
+
x
=+
GM
2
2
R
3
u
3
0
3
x
u
0
Thesolutionfor
x
isthen
x
(
)=
x
0
cos
!;
where
!
2
=1
3
GM
2
2
R
3
u
4
0
Rememberthatthisisarstorderexpansion,soinputting
u
0
intothis
!
toevaluateit,youshould
keeptermstoaconsistentorder. Theequationfor
u
0
hasthe
GM
termasasmallcorrectiontothe
rstterm,sowhenyouusethistoevaluate
!
,youitwouldbeinconsistentkeepit.Use
2
=
GM
r
0
tosimplifytheequations.
1
r
0
=
u
0
GM
2
!
!
2
=1
3
2
GM
u
0
1
2
R
3
u
3
0
1
3
2
M
M
r
3
0
R
3
r
=
1
u
=
1
u
0
+
x
=
1
u
0
(1
x
u
0
)=
r
0
r
2
0
x
=
r
0
r
2
0
x
0
cos
!
(6
:
67)
with
!
=1
3
4
M
M
r
3
0
R
3
(6
:
68)
Thisiswhatwenowhavetoanalyze,puttinginthenumbersandseeingwhatitpredicts.
First:
!<
1,soinorderthat
!
=2
,theangle
mustbegreaterthan 2
. Mercurydoesn’tget
Second: Howmuchmorethanoncearound?
!
.
(2
+
)=2
,ortothisorder
3
4
M
M
r
3
0
R
3
(2
+
)=2
!
=+
3
4
M
M
r
3
0
R
3
.
2
(6
:
69)
Third: What t is
r
0
? Mercury’s s pathisreally anellipse you recall. . Yousee e fromEq. (6.67)thatits
r
0
r
2
0
x
0
anditsaphelion(at
=
)isat
r
0
+
r
2
0
x
0
.Thatmeansthat2
r
0
is
thewholemajoraxisandthat
r
0
isthenthesemi-majoraxis,
a
.
amountgivenbyEq.(6.69).
total
=
k
=Neptune
X
k
=Venus
3
4
M
k
M
a
3
R
3
k
.
2
(6
:
70)
6|Orbits
214
R=R
E
M=M
E
Mercury
0
:
38710
0
:
05527
Venus
0
:
72334
0
:
8150
1
:
76710
6
Earth
1
:
0
1
:
0=5
:
973610
24
kg
8
:
309510
7
Moon
0
:
0123
Mars
1
:
52371
0
:
10745
2
:
493310
8
Jupiter
5
:
2029
317
:
83
1
:
8524610
6
Saturn
9
:
537
95
:
159
9
:
00510
8
Uranus
19
:
189
14
:
500
1
:
68510
9
Neptune
30
:
061
17
:
204
5
:
20410
10
Pluto
39
:
482
0
:
0022
Sun
333000
:
=1
:
98910
30
kg=
M
Shouldsomeofthelargermoonsbeincluded? Ganymede,amoonofJupiter,is0
:
025themass
ofEarth,butourownmoonishalfthemassofGanymedeandit’svetimesclosertotheSun,soit
willcountformuchmore,andit’sstillasmallcorrection.
This table is expressed as ratios s to o the orbital distance and the mass s of f Earth, and because
Eq.(6.70)involvesonlyratios,thatisallyouneed.
3
2
a
3
M
k
=Neptune
X
k
=Venus
M
k
R
3
k
=
3
2
0
:
3871
3
333000
0
:
815
0
:
723343
+
1
:
0123
13
+
0
:
1074
1
:
52373
+
317
:
83
5
:
20293
+
=4
:
568610
6
=0
:
9423
00
perorbit365
:
24100
=
87
:
97
=391
00
perEarthcentury=0
:
11
perEarthcentury
seethattheorderofin uenceisJupiter,Venus,Earth,withtheotherplanetswellbehind.Ifyougoto
higherordersinthecalculationof
U
1
(
r
),keepingtermsin
r
4
=R
4
:
11
ascomparedto0
:
16
isnotyetgoodenough.
Itis possibletocalculatethisprecessionwithoutmakingtheapproximationinEq.(6.66),that
thepotentialisproportionalto
1+
r
2
=
4
R
2
. Youcanusethemethodsdescribedinthenextsectionto
manipulatethepotentialofEq.(6.65)exactly,orifyouprefer,thereisamorefamiliartypeofmethod
available: UseaseriesrepresentationfromEqs.(6.72)and(6.73). Thatis
V
1
(
r
)=
U
1
(
r
)
m
GM
R
"
1+
1
2
2
r
R
2
+
1
.
3
2
.
4
2
r
R
4
+
1
.
3
.
5
2
.
4
.
6
2
r
R
6
+
#
YouuseeachsuccessiveterminthisseriestocomputeitseectonMercurybyfollowingexactlythe
sameprocedureasforthe
r
2
term. Youwillneedalotoftermstogetareallygoodresult,whichis
why,inthenextsection,Iwanttoshowyoualessfamiliarmethodthatavoidsthisseries.
6.13EllipticIntegrals_
you’veneverheardofbefore,youshouldn’tbeintimidated.Ifyouseea\sineintegral"whenstudying
othertopicsthendon’tbesurprisedbyBesselfunctions,therearebooks,tables,andcomputerprograms
todealwiththem. Gammafunctionstooareubiquitous,andtheyareeasy. Legendrepolynomialsare
6|Orbits
215
Here aretwocommonlyusedfunctions,
K
and
E
: the\completeellipticintegrals s oftherst
andsecondkinds". Alongwiththeirdenitions,herearesomeusefulformulasstraightoutofstandard
references.
K
(
k
)=
Z
=
2
0
d
1
p
k
2sin
2
=
Z
1
0
dx
1
p
(1
x
2)(1
k
2
x
2)
=
1
2
Z
0
d
1
p
1+
k
2
k
cos
E
(
k
)=
Z
=
2
0
d
p
k
2sin
2
=
Z
1
0
dx
r
k
2
x
2
x
2
(6
:
71)
=
2
1
k
0
1
K
E
dK
dk
=
E
(
k
)
k
(1
k
2)
K
(
k
)
k
dE
dk
=
1
k
E
(
k
K
(
k
)
K
(
k
)=
2
"
1+
1
2
2
k
2
+
1
.
3
2
.
4
2
k
4
+
1
.
3
.
5
2
.
4
.
6
2
k
6
+
#
E
(
k
)=
2
"
1
2
2
k
2
1
.
3
2
.
4
2
k
4
3
1
.
3
.
5
2
.
4
.
6
2
k
6
5

#
(6
:
72)
Fromthetwoequationsfor
dK=dk
and
dE=dk
youseethathigherderivativesof
K
and
E
canalso
be expressed interms of the original
K
and
E
functions simply by dierentiating these derivatives
repeatedly|sortoflikethesineandcosine. Equation(6.65)callsfortherstofthesefunctions
K
,
buttheperturbationequationsthatfollowitwillrequiretherstandsecondderivativesof
K
.
U
1
(
r
)=
GMm
2
Z
2
0
d
1
p
R
2
+
r
2
2
Rr
cos
GMm
2
R
Z
2
0
d
1
p
1+(
r
2
=R
2) 2(
r=R
)cos
2
GMm
R
K
r=R
(6
:
73)
Areasonnottoshyfromusingthese integralsisthattherearesomeveryfastandecientways to
evaluate them,andthose methodsare much fasterthen astraightnumerical integration or r aseries
expansionwouldbe. Also,whatweneednowistouseEq.(6.14),whichrequiresknowingonly
f
and
f
0
,andthosecomefrom
dU=dr
and
d
2
U=dr
2
,andthoseinturnrequireevaluating
K
and
E
onlyat
a=R
k
forseveralplanets. Thisisreallynottoomucheortforaveryimportantresult. . Andwhyare
thesefunctionscalled\elliptic"? Seeproblem6.71tondout.
Thegraphaboveshowsthebehaviorof
K
and
E
inthedomain0
k<
1andtheyareobviously
evenfunctions.Theleftendofthegraphat
k
=0iseasytoevaluatefromtheequations(6.71)using
the
d
integrals. Attheotherend,at
k
=1,the
dx
formmakethecalculationof
E
easy,butifyou
lookfor
K
atthesamepointyouwillseethattheintegraldiverges. It’samilddivergence,beingonly
logarithmic,andthebehaviorof
K
near
k
=1is
K
(
k
)
1
2
ln
8
=
(1
k
)
. Thesecondderivatives
of
K
and
E
comebydierentiatingtheexpressionsatEq.(6.72)thatgivetheirrstderivatives. It’s
tediousbutstraightforward,andtheresultsare
d
2
E
dk
2
=
E
+(1
k
2
)
K
k
2(1
k
2)
;
d
2
K
dk
2
=
( 1+3
k
2
)
E
+(1 2
k
2
)(1
k
2
)
K
k
2(1
k
2)2