]\gt ; 455 XIII .
On the Motion of Two Spheres in W. M. HICKS , M.A. , Fellow of St. John 's College , Cambridge .
municated by Professor J. CLERK MAXWELL , F.R. Received May 16\mdash ; Read June 19 , 1879 .
THE general theory of the motion of a single rigid body through an infinite incompressible fluid is well known , chiefly through the work of THOMSON and TAIT and KIRCHHOFF , and we are able to calculate numerically the results in the case of the sphere , the ellipsoid , and a large number of cylindrical surfaces .
The theory of the motion of two or more bodies in a fluid has naturally not made the same progress , and we are unable to determine the form of the expressions involved for the general motion of any particular solids .
So far as I am aware , the first attempt was made by STOKES , in a paper read before the Cambridge Philosophical Society in 1843 , entitled " " On some cases of Fluid Motion.\ldquo ; In this paper , amongst other problems , he considers the case of two spheres .
He determines the instantaneous velocity potential for two concentric spheres and for two concentric cylinders with fluid between them , and finds that the effect of the fluid is to increase the inertia of the inner sphere by a mass of the mass of the fluid displaced , and that of the inner cylinder by a mass of the mass displaced , , being the radii of the spheres or cylinders .
He also approximates to the cases where one sphere is moving in the presence of another in infinite fluid ; and also in the presence of a plane , the method used being ffist to calculate the velocity potential for any motion of the points of the plane , and then suppose them actually animated with velocities equal and opposite to the normal velocities of the fluid motion at those points if the plane had been removed .
He applies the same method also to the consideration of the motion of two spheres .
In a note in the Report of the British Association at Oxford , 1847 , he states the theorem given by me in S 4 relating to the image of a doublet whose axis passes through the centre , and mentions that this will easily serve to determine the motion .
In 1863 Herr BJERKNES communicated a paper to the Society of Sciences at Christiana , on the motion of a sphere which changes its volume , and in * Not .
Phil. , p. 264 , new edition , p. 330 .
BORCHARDT , Bd .
71 .
Camb .
Phil. Trans. , vol. viii .
but he has given an account of his researches in the 'Comptes Rendutogether with some historical notices on the development of the theory .
He does not seem , however , to have been acquainted with the important paper of STOKES above referred to .
In 1867 THOMSON and TAIT'S 'Natural Philosophy ' appeared , containing general theorems on the motion of a sphere in a fluid bounded by an infinite plane , viz. : that a sphere moving perpendicularly to the plane moves as if repelled by it , whilst if it moves parallel to it it is attracted .
In a paper on vortex motion in the same year ( Edin .
Trans. , vol. xxv THOMSON proved that a body or system of bodies passing on one side of a fixed obstacle move as if attracted or repelled by it , according as the translation is in the direction of the resultant impulse or opposite to it .
In the ' Philosophical Magazine ' for June , 1871 , Professor GUTHRIE publishes some letters from Sir W. THOMSON on the apparent attraction or repulsion between two spheres , one of which is ating in the line of centres .
Results only are given , and he states that if the density of the free globe is less than that of the fluid , there is a " " critical\ldquo ; distance beyond which it is attracted , and within which it is repelled .
problem of two small spheres is also considered by KIRCHHOFF in his ' Vorles . .
Math. Phys pp. 229 , 248 .
In his later papers BJERKNES takes up the question of " " pulsations\ldquo ; as well as vibrations .
Of solutions for other cases than spheres , KIRCHHOFF has considered the case of two thin rigid rings , the axes of the rings being any closed *'Comptes Rendus , ' tom .
lxxxiv .
, p. 1222 , \amp ; c. Not only Herr BJERKNES , but several writers on the Continent seem to be unacquainted with this paper of STOKES , and also with GREEN'S papers .
KIRCHHOFF , in his Vorlesungen uber Physik ' ( second edition , p. 227 ) , says that DIRICHLET first treated the motion of a sphere in a fluid in the Monatsberichte der Berl .
Akad .
' in 1852 , and CLEBSCH that of the ellipsoid in 1856 , in ' Crelle , ' Bd .
52 .
BJERKNES also repeats the same statement , and CLEBSCH in his paper regards DIRICHLET as the first to solve for the sphere .
In his paper DIRICHLET says : " " Wie es scheint , ist bis jetzt fur keinen noch so einfachen Fall der Widerstand , den ein in einer ruhenden Flussigkeit fortbewegter fester Korper von dieser erleidet , aus den seit Euler bekannten allgemeinen gleichungen der Hydrodynamik abgeleitet worden The fact is that GREEN in a paper read before the Royal Society of Edinburgh in 1833 , entitled ' ' Researches on the Vibrations of Pendulums in Fluid Media " " ( Trans. Roy .
Soc. Edin .
; also published in the Reprint of his papers , p. 313 ) , and written without the knowledge of POISSON'S paper of 1831 , " " Sir les mouvements simultanes d'un pendule et de l'air envirolmant , \ldquo ; treated of the motions of an ellipsoid moving parallel to one of its axes .
He obtaius the velocity potential as an elliptic integral for a motion parallel to an axis , which also of course contains implicitly that for the sphere .
He shows that it is necessary to suppose the density of the body augmented by a quantity proportional to the density of the fluid .
For the case of the spheroids moving in their equatorial planes or parallel to their axes he completely determines this quantity , whilst for the sphere he finds that it is one-half the mass of the fluid displaced .
The first place in which I have been able to find the well known form of the velocity potential for a sphere is in STOKES ' paper of 1843 before mentioned .
He obtains it as a particular case of a more general problem , and refers to it as the " " known ' value for the sphere .
The equations of the lines of flow were , I believe , first given by DIItICHLET .
, Bd .
71 , MR. W. M. HICKS ON MOTION OF TWO SPHERES IN A FLUID .
curves the sections by planes perpendicular to the axis being small circles of constant radii , and he arrives at the result that their action on one another may be represented by supposing electric currents to flow round them ; and I have recently solved the problem of the motion of two cylinders in any manner with their axes always parallel .
The velocity potentials for the motion of the two ders are found in general as definite integrals , which , when the cylinders move as a rigid body , are expressed in a simple finite form as elliptic functions of bipolaJ : coordinates .
The functions involved in the coefficients of the velocities in the expression for the energy have a close analogy with those for spheres arrived at in the following investigation .
1 .
Our first aim will be to find the velocity potential for the motion of the fluid in which a sphere is fixed and in which a source of fluid exists .
the image of the source in general is meant that collocation of sources or sinks within the sphere which produces outside of it a motion which in conjunction with the original source has no normal motion across the sphere : in other words , that " " mass " " of positive or negative ces which produces across the surface of the sphere a normal flow equal and opposite to that of the outside source .
When this " " image ' ' is found , the way is eoretically clear to finding the velocity potential when two spheres are fixed in the fluid , and thence , by distributing over the of the spheres sources proportional to the normal motion of the surface at that point , to determine the velocity potential when the two spheres are moving in any manner .
In the case of an electrical point the image is , as is well known , a negative point at the inverse point of the other .
In the case of fluid motion the image is , as will be shown , a positive source at the inverse point , together with a negative line sink stretching from this point to the centre of the sphere .
Fig. 1 .
2 .
Take the centre of the sphere for origin and let the axis of pass through the source S. Let the radius of the sphere be , and the distance of from the centre be .
Then the velocity potential will clearly be symmetrical about S. velocity potential for the unit source at can be expanded in the series Expand the potential due to the sources , \amp ; c. , inside the spheres in a series of spherical harmonics Hence the flow out of the sphere , for points just outside , is and this must be equal to the other , whence and Consider the potential for a source at a point on inside the sphere at a distance from the centre .
Then Comparing this with the expression for , we see that if we make and source i. e. , V is the potential of source at the distance from whose magnitude is equal MR. W. M. HICKS ON THE MOTION OF TWO SPHERES IN A FLUID .
to of the source at , together with a line sink extending from to the distance ' the line density of the sink being source at S. Performing the integration for V , we find finally that the whole velocity potential for a unit source at is ' where It is easy to verify this value for by direct differentiation .
If we the same method to find the velocity potential for the motion of fluid inside a sphere under the luence of a source inside , the integral becomes infinite unless the source is zero .
The case is of course physically impossible since if fluid is generated within the sphele it must pass through the boundary .
But if we also place an equal sink at any point within , the motion is then possible , and the expression becomes finite .
being the source let be its inverse point with reference to the sphere , and any point on the line produced to infinity .
Then the " " image\ldquo ; of is a source at , and a line distribution of sinks of line density from to infinity .
Let be an equal sink , then its image and that of will produce with finite derivatives .
In fact , the potential at will be where , are the angles makes respectively with OS , 3 .
The expression found for the motion when there is a single source outside the sphere enables us to deduce the velocity potential for a single sphere moving through an infinite fluid .
Taking the direction of motion as the axis of , from which we will suppose measured , we may arrange a surface distribution of sources proportional to and integrate over the surface of the sphere , or we may employ the simpler method used in a paper in the ' Quarterly Journal of Mathematics ' for March , p. 128 .
The first gives us the velocity potential when the sphere moves by an integration which would be laborious .
The other gives directly the potential , when the sphere is fixed and the fluid moves past it , by means of an easy differentiation .
Putting a source at and an equal sink at , let these move off to infinity , increasing indefinitely as they do so , yet so that the motion at a finite distance from the origin is finite .
In the limit we clearly get the case of fluid flowing past the sphere .
460 MR. W. M. HICKS ON THE ] MOTION OF TWO SPHERES A FLUID .
We have to the litnit when of When is large and sIllall this is easily shown to be Hence the linnit is If the velocity of the fluid at an finite distance parallel to is towards the origin , then Also impressing on the whole system a velocity , the sphere moves with velocity in an infinite fluid , and the potential function is The well-known form of in this case .
4 .
If now two spheres are present in the fluid , and we consider the series of images resulting from the first image in , we see that they very rapidly become extremely e.g. , the first image is a and line sink ; the image of this in consists of ( 1 ) a source and line sink , ( 2 ) the image of the first sink or a line sink ( segment of a circle ) , and an area source bounded by this last line sink and two straight lines from the .
It is , therefore , hopeless in this way to find first the velocity potential for a source in the presence of the two spheres , and thence the potential for any motion of the spheres .
But now suppose A fixed and moving in any direction .
If A were not present the velocity potential of would be that due to a doublet at its centre , whose axis lies in the direction of the motion of B. The effect of the introduction of A will be to produce a series of images of this doublet , lying inside A and B. This method dispenses with the necessity of integrating over the spheres when we have found the velocity potential for the doublet .
In the special MR. W. M. HICKS ON IOTION OF TWO LUID .
case where is moving in the line joining the centres , the image becomes simplified and reduces to a single doublet .
For let us find the image of a doublet whose axis passes through the centre of a sphere .
The doublet is formed by allowing an equal source and sink to indefinitely approach one another , their magnitudes increasing indefinitely , yet so that is finite .
Now let lie on the line through the centre of the sphere , and let be their inverse points ; moreover , let the limit of .
Then the image of consists of a source at , a sink at , and a line source ( supposing outside , and therefore outside Q ) along with line density , also the quantity , together with the sink at , is equal and opposite to the source at , and we may suppose it added to the at , when they become equal .
Now as approach to coincidence so do , and the image of the doublet at becomes the doublet at , whose magnitude is the limit of i. e. , one of opposite sign and magnitude that at P. The same result can easily be shown to follow from the analytical formula in S 2 .
The case where the doublet has its axis perpendicular to the line joining the centres has more analogy with the case of a source .
The image here consists of a doublet of the same sign at the inverse point , with a trail of doublets of opposite sign extending to the centre .
Fig. 2 .
Let , be equffi source their inverse points with respect to the circle .
Then at we have a source and of magnitude , and in the limit we a doublet Also , if be corresponding points on , we have a line dollsity at nd at we get a line doublet along , whose line magnitude at any point is the limit of to the distance from the centre .
Fig. ( 3 .
5 .
Supposing that the positions of all the images of the doublets and their magnitudes are known when the sphere A is moving along the line , we proceed to find an expression for the kinetic energy .
Let be the distance of the image in A from , and the distance of the image in B. Ako let the magnitudes of the doublets there be respectively .
Let be the velocity potential of the motion , and ' the parts of due to and .
Then denoting the kinetic energy by where ] is the value of at any point on the sphere .
Now and the part of due to will be Now When and the above becomes MR. W. M. HICKS ON THE MOTION OF TWO SPHERES IN A FLUID .
463 Similarly when and it becomes and Also the original doublet , and if be the mass of fluid displaced by the sphere But from what has been shown before Hence ( 1 ) By S 4 we have , if is the distance between the centres , AIso Hence Again whence .
W. M. HTCKS ON Put , and choose we have Fig. 4 .
Now let be the inverse points of the spheres , and the middle point of .
Put .
Then say ( 2 ) . . . . .
( 3 ) Further , being any point on the sphere , denote the constant ratio and let be the similar constant for the sphere B. Then ( 4 ) The equation to now becomes The roots of which are Choosing the positive sign , the equation of differences becomes Now whence writing for we get Here and Whence writing and and But MR. W. M. HICKS ON THE MOTTON OF TW SPHERES TN A FLUID .
5 and Also say and Whence . . .
( 5 ) We shall denote , in when .
follows , by the functional symbol .
6 .
If the sphere is also moving along the line AB , the kinetic energy of the fluid will be of the form It remains , then , to find the value of L. It is easily seen that depends on the part of belonging to the images of 's motion taken over the sphere , together with that belonging to the images of 's motion taken over the sphere B. Let now dashed letters apply to the images , \amp ; c. , of the system , then using the results in S 5 , the part of due to the integration over A is But as before , remembering that now the original doublet is in and and Now as before .
W. M. HICKS ON TH MOTION OF TWO SPHERES IN A FLUID .
and determining A by the condition that we shall find and and Similarly same quantity .
Therefore , denoting by the mass of fluid contained in sphere of radius unity where ( 6 ) Tables for and are given at the end of the paper for equal spheres , and for the case of 7 .
When the sphere A is moving perpendicularly to , the original doublet is one perpendicular to the line , as also its images .
Suppose A is moving along the axis of , A being the axis of .
Then the normal velocity at a point on the sphere is the polar coordinates of ; and the kinetic energy is given by Let be the magnitude of a doublet at a point distant from the centre of ; the part of depending on this is and the part of on this is MR. .
M. HICKS ON TH1 MOTION Ol TWO IN A FLUJi ) .
The integral of which is Writing and for for doublets outside the sphere , we obtain and whence Now any at the distance produces an image in A consisting of a doublet at a distance \mdash ; together with a line sink stretching from this to the centre , whose ' line magnitude is distance from the centre .
Hence the whole amount of the image is Now eyery except forms part of an image of some , and of that only .
Hence and ( 7 ) The 2 extending to the whole mass of inside A. 8 .
If A has also a motion along , together with one perpendicular to it , has no term depending on ; for it is clear that if the sign of is changed , then the kinetic ergy must be the same as before .
If moves also perpendicularly to will have additional terms in , and .
The coefficient of will be analogous to that for , whilst that for MR. W. M. HICKS ON THE MOTION OF TWO IN A FLUID .
in the case of , consists of two parts , depending on the integration over the two spheres .
As in the case of this coefficient is dashed letters referring to the motion of referring to images within A and respectively , and denoting distances from the centres of A and B. This may be reduced as in the former case to the being the images in A of 's motion perpendicular to A , and the images in of 's motion perpendicular to A B. where are the original doublets at the centres of whence ( 8 ) in which last the ratios do not contain In the case of general motion of two spheres , each will have three components of velocity , ; and , in general , the expression for the kinetic ellergy will contain 21 terms .
In the case in question we can easily see that the coefficients of 12 of these vanish .
For .
the term in \mdash ; snppose all zero ; the energy , from the symmetry of the motion , must clearly be unaltered if we reverse the direction of .
And this can only happen if the coefficient of .
In this way we find the terms all vanish except those in Also from symmetry the coefficients of , are equal respectively to the coefficients of In what has gone before we have expressed coefficient of ) , in terms of quantities determined by the radii and distance of the spheres , and have shown how the coefficients of the other terms depend on the images of the motion , whereby we can without much difficulty approximate to their values when the distance of the spheres is large compared with their radii\mdash ; or the distance between their surfaces is large compared with the radius of one of the spheres .
We pass on to are equal and opposite .
But this is due to the fact that in a cylinder the image of a doublet ( or a source ) is a single doublet , whatever be the direction of the axis of the original doublet .
9 .
If be a line through the centre the infinite trail of images of S2 cuts out , and we are left with an image source and sink , and a line sink between them , supposing to be outside .
Let now and approach together and become a doublet whose strength is .
Then we shall get a single doublet as its image whose strength as in the former case .
This we might have deduced at once from the case of the external doublet in S 4 , considered as the image of its image .
If we proceed to find the kinetic energy , as in the previous case , we must clearly be led to the same form for the result , viz. : when A is moving with a velocity from where .
are the strengths of the doublets inside A .
But in this case the relations between the are given by the equations being radii sphere ) whence R. .
M. HICKS ON MOTION OF SPHERES IN A FLUID .
47-1 which differs from the equation for external spheres in having for for all values of .
We may therefore use the same solution and writing here which is the same form as before , only is the inverse of its former value .
And , as A table for when is given at the end of the paper .
10 .
It will be well here , before passing on to the consideration of the motion , to make a digression on the properties of the functions and .
In the first place it is easily seen that the series for the and functions are both convergent , eyen up to the case when the spheres touch , or ; for the ratio of the ?
term to the is and this is always less than , which is less than unity , except in the case when the spheres touch .
In this particular case the term tends to the limit and the series is still convergent .
The value of is the limit when of .
Hence when the spheres are in contact ( 9 ) if The of this may be foumd from LEGENDRE'S table of the -functions .
If the spheres be equal and ( 10 ) Now whence When the spheres are equal .
If in this case denote either or 11 .
We may easily express the general term in terms of .
For writing it the form suppose , we get once the relations ( 2 ) , ( 3 ) , ( 4 ) which , since and MR. W. M. HICKS ON THE MOTION TWO IN A FLUID .
Now suppose , also We shall furthel write Then ( 11 ) and Denote by Then Let . . . . .
Then value of when is value of when is value of Now .
Hence when MR. W. M. ON THE OTION OF TWO IN A LtJlD .
Therefore Also we see that is a rational integral function of When one sphere is inside the other the series for is still convergent .
12 .
When the spheres are concentric and whence ( 12 ) which agrees with the result found by TOKES in his paper of 1843 before referred to .
When the inner touches the outer and lo ( 13 ) where If is an integer say a finite expression , and in this case In the particular cases MR. W. M. HTCKS ON MOTION OF TWO IN A FLUID .
If is of the form Also a finite expression , and in this case In the particular cases The expressions for directly in terms of are the same functions of as the corresponding expressions for external spheres are of 13 .
The series for may , as in the case of , be shown to be convergent .
When the spheres are in contact ( 14 ) Also the general term in is given by ( 15 ) It is easily seen that both for external spheres diminish as x\mdash ; i.e. , increases .
Hence for external spheres ' are both negative .
When one sphere is inside the other , decreases , as .
diminishes .
Hence in this case is positive .
The values of the first three terms of are for ( 16 ) and for both of which are of the order Hence the values of , at contact are .
But though this is the case , the value of at contact is finite .
The term is which is of the order and therefore the whole sum is finite .
Also when the ' term is positive , when the sign depends on the value of .
But by considering the values of , \amp ; c. , in terms of , expanding them in ascending inverse powers of , it can be shown that is positive always .
Further , at contact is a tive quantity , whilst at an infinite distance it is zero .
Hence , on the whole , it must increase with , and if this takes continuously , would always be positive .
Though I have convinced myself that such is the case , I have not been able to prove it in general .
When the spheres are at a great distance the values of and depend only on their first terms , only on the term of , which is of the order .
Hence here also the differential coefficient is positive .
I have calculated and laid down curves representing the magnitudes of the and -functions in the case of equal spheres , and when the radius of one sphere is twice that of the other , and in both cases the value for comes out positive for all distances .
In what follows we shall suppose that this quantity is always positive , but it must be understood throughout as ] proved for the case of equal spheres and the case which the radius of one sphere is double that of the other .
15 .
Although the rapidly increasing complexity of the successive images when the spheres move perpendicularly to their line of centres would lead us to regard the \amp ; MR. W. M. HICKS ON THE MOTION OF SPHERES IN A problem of finding the energy in this case as almost hopeless , yet we can carry approximation to any number of images with less labour than might at first sight appear .
For suppose we wish to take into account in , due to 's motion , that is on whole .
We need only firsb calculate the distribution of doublets for position of the ol.iginal one , in the image in , and find the amount of the first images .
We can then treat the second portion of the images as the images resulting from the different of the nage , and employ our first result to find the amount of the second portion by a single integration .
Suppose we proceed as if we were going on indefinitely : we suppose an original doubIet in A at a distance and calculate the density of the parts of the first image in , say at a distance and thence its amount .
We employ this result to find the density at any point of the second image , regarding it as made up of images of the different parts of the first , and this we do by using the expression found before , substituting for the original doublet at , an amount at a distance / ; and ting with respect to over the first image .
Thus we find the distribution for the second image and its amount , and therefore the amount for the first two images together .
Starting now from this , and proceeding in the same way , we find the tribution and amount of the first.four images , then of the first eight , and so on .
Thus to find the distribution of the image we only require operations , and to find its amount only operations .
Even with this method of proceeding the work would be exceedingly laborious .
But for all practical purposes the first two images in , i. e. , the motion due to four , will be sufficient\mdash ; except when the spheres are in contact .
We proceed then to find the values of the coefficient of and of to this degree of approximation .
Suppose we have at uside A Q ) doublet at a distance whose itxis ; perpendicular to A B. .
First nage in B.\mdash ; Then we have at its inverse point in , a doublet and a line doublet thence to , whose line density .
First in A.\mdash ; The image of this in A consists of two parts , that ) on the single doublet in , and that depending on the line doublet .
Image of .\mdash ; A doublet at whose magnitude is , and a negative line doublet from to A whose line density of tegative line doublet.\mdash ; At a distance from we have tive doublet dr .
This has ( 1 ) a negative doublet at a distance from A. equal to dr .
That is from to A we have a negative line doublet whose density at a distance is and .
density ( 2 ) a line doublet image of each portion .
The doublet produces a positive line doublet a distance .
to , whose line density Hence the density at distance , due to this part , from the whole line doublet in Hence finally the density at a distance from A of the resultant line doublet and the whole amount that the whole amount of the image is or substituting \amp ; c. , in terms of MR. W. M. HICKS ON MOTION OF TWO RES IN FLUT For the first image in Hence ( 17 ) The density at any point of the first image is together with a doublet at a distance The amount of that part of the second image in A which depends on the tter is and the amount of the part due to the portion of the former at a distance is whence the whole due to the former Now say .
Then the above is Wherefore the whole amount of the second image in is writing .
( 18 ) Substituting for and in ( 7 ) we get the part of depending on correct to the second image in A. Interchanging and , the part depending on is found .
In the case of equal spheres 16 .
find the value of the coeflicient of the term in we need to find the amounts of the nages in due to the motion of , and vice The first image in of in A at a distance is at and a negative line doublet thence to , whose line density is The whole amount is therefore For the first and To find the amount of the second image in we start from the first image in already found .
This is , as has been shown , a doublet at and a line doublet thence to , whose line density is MR. W. M. HICKS ON MOTION OF TWO SPHE RES IN A FLUID .
The amount in from the former is and from the second is . . .
So also Whence from ( 8 ) L\fnof ; m ( 19 ) Similarly can be found the coefficient of when one sphere moves inside another .
Motion in the line of .
When the two spheres are moving in the line of centres the kinetic energy is given by where It is to be remarked that are functions only of the distance between the spheres , and that therefore .
Since no forces are supposed to act on the system , both the energy and momentum are constant .
Hence constant ( 20 ) The last equation also follows at once from LAGRANGE'S equation since and may be written We shall transform these equations by referring the motion to the velocity of an arbitrarily chosen point between the spheres , and the distance between them .
Let divide the distance ( r ) in the constant ratio .
Then if is the distance of from the origin , its velocity and ence ( 21 ) which we shall write whence or ( 22 ) in which we are to take the positive or negative sign according as the spheres are separating or approaching another .
The spheres will move as if they repel or MR. W. M. HICKS ON THE MOTION OF TWO SPHERES IN A LUID .
attract one another relatirely according as is positive .
This condition does not depend on their relative motion at any time , but only on their distance and the ratio of the constant energy to the constant momentum .
The above condition may also be expressed , writing ) , as the sign of The last term is positive , for all decrease as increases .
Now must ahvays be since / always .
If we put in the above , the criterion educes to the sign of i.e. , since is always positive to the sign of Now we are led to conclude from the argument in that .
are always positive .
Hence when has its gl.eatest possible value the criterion is positive , much more then is it so .
any other value of .
Hence we are led to collclude that whatever be the relation between the energy the spheres always move so that tends to ease , whilst in the case of equal spheres , or that in which the radius of one is twice that of the other , we know that such is the case .
We cannot prove from this that the spheres move with reference to a fixed as if they repel one another , for it might happen that both the spheres might be accelerated , the extra energy of the motion of the spheres themselves being taken from the fluid 1notion ; or that both are even retarded .
We can easily show , however , that both cannot be accelerated if is positive and both move in the same direction , for the disin this case increases , and therefore so do hence because is constant ?
cannot both increase .
Also if is ve of the same sign the same result holds .
In the case where the are projected so that the momentum is zero and the relation between the velocities of projection that this may be the case is givell by When the spheres equal and the motion is the same as that of a single sphere in a fluid bounded by a plane , and moving perpendicularly to the plane .
For this particular case if denote the velocity relative to the fixed plane , and where are the values of , and at the point of projection .
If the sphere is projected from contact with the plane and at an infinite distance Hence the ratio of the limiting velocity to the initial velocity is where is the density of the sphere .
For densities , 1 , 10 , the values of this ratio are respectively .
The greatest value is when the density of the sphere is zero , and the least is when ( no fluid ) or , the ratio then being , as it ought to be , unity .
In the case where the spheres are unequal and projected with no momentum from contact their initial velocities must be opposite and in the ratio of the quantities and and denoting the quantities .
W. .
HICKS ON MOTION OF TWO SPHLRES A FLUID .
, and , find from tables of the Eulerian integrals the ratio is which when the densities of the spheres fluid are equal becomes We find velocities of the spheres elatively to the fluid by ting n between and whence and Suppose now the same spheres projected with the same initial circumstances except that now the spheres have changed places , and let be the COl.responding velocities at the same distances .
Then and do not depend on the question which of the two is foremost .
Now if we see at once from the expressions given for in terms of the distances that , and hence that the foremost will be most accelerated when it is the smallest .
If now denote the velocities any moment which we may regard as the velocity of projection Writing for the ratio the equation to find , in ordel that may have a given valne , is This enables us to find within what limits must lie , must have real roots , and therefore or Hence may be any positive quantity less than .
The greatest possible value of this is when the spheres are infinitely distant , and then To each value of will correspond two states of motion , the initial velocities in each case being opposite .
For example , if is positive , i.e. , both velocities in the same direction , the two states will be when is the foremost , and when ( b ) is the foremost ; if be negative , the two states will be , one in which the balls begin to move towards each other , the other in which they begin to move from each other .
Thus for every given value of there are four possible states of motion .
If ever then , and the spheres must be at such a distance that Now , supposing given , this can only happen if lies between the greatest and least values of .
The least value is when the spheres are in contact , the gyeatest when they at am infinite distance , the value then being If , then must lie between the greatest least values of .
Now as as If we suppose then , and calling the least values of the above limits Hence if MR. W. M. HICKS ON THE MOTION OF TWO SPHERES IN A FLUID .
the spheres can neither ever come to rest ; if or the small sphere can never come to rest .
The effect of the vibratory ?
notion.v .
18 .
Suppose each of the two spheres attracted to a fixed centre of force where the force varies as the distance .
Let be the distances of the spheres at any time from their respective centres of force measured in the same direction .
Then Also since we neglect squares of small quantities in finding the small vibrations , the equations of motion become . .
and we suppose the spheres so distant , and their motions so small , that we may neglect the small changes in during the motion .
The spheres must not be too close , for at contact , \amp ; c. , are infinite , as was shown in S 14 .
Solving the above equations in the usual manner we find where From this we see that , to the first order of small quantities , mean position of the spheres is not altered , or to that degree of approximation there is no mean attraction or repulsion .
If we regard the spheres as two pendulums swinging in the fluid , in the same horizontal line , of lengths , then the motion is giyen by the aboye equations if we write where are the densities of the spheres compared to the fluid .
If in the above we make ) we get the case of a forced vibration of period In this case If the sphere ( b ) is set free when is for the moment at rest , and the time be reckoned from this moment and the motion of ( b ) consists of two terms whose amplitude is times that of .
Let now the strength of the centre of force on ( b ) diminish indefinitely .
Then and ( b ) would oscillate in the same period as , without bein attracted or repelled towards it except by forces depending on the square of the amplitude of .
To find , then , whether the action of on ( b ) is attractive or repulsive we must take account of quantities of the second order of small quantities .
The full equation of motion of ( b ) is For a first approximation we have BL Write MR. W. M. HICKS ON THE MOTION OF TWO SPHERES IN A FLUID .
48.9 where is of the order at least .
Substituting for and neglecting cubes and higher powers of \mdash ; - and Whence the equation takes the form where in which last form we may neglect in the as it appears in the differentiation .
Hence the mean action of ( a ) on ( b ) is an acceleration towards ( a ) if is the " " velocity of mean square\ldquo ; of .
If the distance of the spheres is so large that we may neglect twelfth and higher inverse powers of the distance , we need only consider the or the first terms in A and B. In this case it will be found that the accelel.ation to is ( 23 ) To find when there is repulsion or which clearly only happen if or the density of the sphere less than the fluid .
In general , then , when the body is denser than the fluid it is attracted .
If its density is less than the fluid there will be a critical point ( as mentioned by Sir W. THOMSON ) , beyond which there will be repulsion , and within which it is attractive .
This critical distance is given by ( 24 ) in using which it must be xlbeyed that if comes out nearly equal to , the formula fails to give a correct value , as it was obtained on the supposition that the distances were large .
It is , howeyer , extremely accurate if we remember that it is true up to inverse powers of the twelfth at least .
If the density of the sphere be 9 the critical distance would be times its radius .
It may be noticed that while the principal term in the acceleration depends on , if the density be the same as the fluid it depends on In the case of a sphere vibrating within another sphere , along the line of centres , the effect of the fluid will be represented by supposing the inertia of the sphere increased by a mass mass of fluid displaced by it where has the value given in S 9 : provided it is not close to the boundary of the containing sphere , as in that case becomes infinite , and the small motions of the sphere will produce great changes in the value of Q. When its mean position is the centre , and may be considered constant when we neglect in our equations of motion cubes of small quantities .
The ] of in this case is , as has been already mentioned , mass of fluid displaced The foregoing serves to solve the problem of a ball pendulum within a spherical when it is so suspended that its centre lies in the horizontal line through the centre of the envelope .
When it oscillates in any other position the value of the coefFicient of inertia may be approximated to as in , 16 .
MR. W. M. HICKS ON THE OF TWO SPHERES IN A FLUID .
19 .
If instead of supposing the sphere ( b ) free to move we suppose it held fast , and require to find the force necessary to do so , we get a different result from the foregoing .
Suppose the sphere ( a ) moving in any manner , the sphere ( b ) being for the moment at rest , and suppose a constant force acting on The equation of motion for is Suppose now that is of such a magnitude that being zero it makes also zero .
Then is the force required to keep ( b ) at rest at the moment when the motion of ( a ) is given by .
Hence Let being small .
Then neglecting cubes of small quantities and This is the force at the time necessary to keep ( b ) at rest .
Hence the mean force is a force towards , which is equal and opposite to the force of on ( b ) .
Since negative , t , he action is an attractive one Taking for only the first term of , which is equivalent to neglecting higher inverse powers of and the force ' weight of fluid by ( / ( 25 ) For example , for equal spheres at a distance ( distance between faces , the mean square of velocity of being the same as for oxygen at a tempebeing measured in feet .
It is that while the force decreases indefinitely with the size , the " " effective\ldquo ; force increases indefinitely .
If vibrate through a distance -inch , 256 times a second , and -inch , the force is 01197 weight of water displacedweight of milligrammes .
In the same mamler call be found the action of on ( b ) when describes any small curve whose plane contains ( b ) .
of for equal spheres .
Q. 037142 023335 015631 008001 002895 001373 VALUES of when for external spheres .
VALUES of when for an internal sphere .
\amp ;

