Ex. 5. A mass of fluid is in motion so that the lines of motion lie on the surface of co-axial cylinders. Show that the equation of continuity is op .16 @ P= (pu + (pv) =0, a0 39 PO +5, OY) where u,v are the velocity perpendicular and parallel to z. [Agra 2003; Rohilkhand 2002, Kanpur 2000, 08; Meerut 1999, 2002, 2012] Sol. Consider a fluid particle P, whose cylindrical coordinates are (r, @,z). With P as one comer construct an element (curvilinear parallelepiped PORS, P'Q'R'S') with edges PQ =8r PS =r80 and PP’ = Let P be the density of the fluid at P. Since the lines of motion lie on the surface of co-axial cylinder, there is no motion along PQ. Hence the rate of the excess of the flow-in over flow-out along PQ vanishes. Again, we have Rate of excess of flow-in over flow-out along ps = ~r80 Su briz) $7 ri Rate of excess of flow-in over flow-out along Pp! =~ §z 2 (pvr s08r) oz Again, the rate of increase in mass of the element = Leprivarse ) i Hence the equation of continuity is given by Zeorsvarae) = -20 E cou 3r2)—82 Zoowrsoar) or rir bs Pr 80825 (pu) +rbr 8082 (pv) =0 or ete ony Zeon=0 % a ree Ex. 6. If the lines of motion are curves on the surfaces of cones having their vertices at the origin and the axis of z for common surface, prove that the equation of continuity is , ° 20H cosec® 0 yet Fp Pw) =0. where w and w are the velocity components in the directions in which r and $ increase. [Agra 2001; Garhwal 2000; Meerut 2001, 02,03, 04, GN.D.U. Amritsar 1998; Rohilkhand 2004]5.20 FLUID DYNAMICS Jemark, Since the velocity is unique at a point, so usually no two streamtinss intersect 2och, ‘thes. Put some flow fields may have singularities, where the velocity vector is not vnique. Sources and sinks are examples of singularities of a flow field because infinitely many stream lines meet at such points as indicated in the figures (i) and (i), 5.12. Source and sinks in two-dimensions. [Garhwal 2002; Kanpur 1999; Meerut 2010] In two-dimensions a source of strength m is such that the flow across any small curve surrounding is 2xm. Sink is regarded as a source of strength ~m. Consider a circle of radius r with source at its centre. Then radial velocity q, is given by a) or 2) ‘Then the flow across the circle is 2xrq,. Hence we have 2arg = 2am or fagsim @) Lav) _ or (24) =m. bya Inigating and omiting constant of ictgrtoa, ve got yeund @ ‘Using (2) and (3), we obtain as before $=-nbgr 6) Equation (4) shows that the streamlines are 0 = constant, ie., straight lines radiating from the source. Again (5) shows that the curves of equi-velocity potential are r = constant, (e., concentric circles with centre at the source 5.13. Complex potential due to a source. [Kurukshetra 2004; Meerut 2001, 2012; Kanpur 2009] Let there be a source of strength m at origin. Then w= $+ f= —mlog r—im0 =—m (logr + iloge®) = —mlog (re) =—mlogz If, however, the source is at 2’, then the complex potential is given by w = —m log (z~ 2") ‘The relation between w and z for sources of strengths m,, ma, my, situated at the points 252, 2,2. is given by w= —m, log (¢~2;) ~ m, log (@ ~ 2,) ~ m, log (@ ~ z,) ~ leading to 4 = =m, log r, — my log r, ~ my log ry and =m 0, —m;02~m, 05 ~.. where ty 212-291 and 0, = arg (2), n=1,2,3, 5.14. Doublet (or dipole) in two dimensions [Agra 2005; Garhwal 2000, 04; Rohillhand 2000; Kanpur 1999, 2002, 07] ‘A combination of a source of strength m and a sink of strength ~ m at a small distance és apart, where in the limit m is taken infinitely great and as infinitely small but so that the product ins remains finite and equal to 1, is ealled a doublet of strength yt, and the line s taken in the sense from — 1 to + m is taken as the axis of the doublet.Doublet (or dipole) in two dimensions [Agea 2008; Garhwal 2000, 04; Rohilkhand 2000; Kanpur 1999, 2002, 07] A combination of a source of strength m and a sink of strength ~ m at a small distance & apart, where in the limit m is taken infinitely great and &s infinitely small but so that the product ‘ns remains finite and equal to, is called a doublet of strength ju, and the line és taken in the sense from - m to + m is taken as the avis ofthe doublet. MOTION IN TWO-DIMENSIONS AND SOURCES SINKS 521 Complex potential due to a doublet in two-dimensions [GND.U. Amritsar 2004, 06; Kanpur 2000, 05, 07; Meerut 2002, 09, 10; Purvanchal 2004, 05] Let A, B denote the positions of the sink and source and P be any point. Let AP BP =r + jy and 2 PAB =9. Let 6 be the velocity potential due to this doublet a mage log) «mg 28 ee mloe{1-®) g=-m™, 10 fist onkr of apposition (2) Let AM be perpendicular daw frm B on AP. Then, AM =AP -MP= r~(r+ir)=-Br iris sotht. «r= from te 020 end a ses = sregth ofthe doublet os on oos0 A or Oy. _100 Integrating it with veapct to 0, we get ° Now, « Using @) and @), reduces to 4(-1se2) [228 pn] cx f(0)=0 so that f(A)» constant Hence omitng the ave constant, ()redoes to y= Hane © Using (2) and (5), the complex potential due to a doublet is given by weagriy = (cos0-ising) =He-® = He oH Note 1, Equi-potential curves are given by = constant, ie., by (1c0s6)/r = constant or (C0s6)/r=C eos =CP or x= Cty), ‘hich represent citeles touching the y-axis atthe originbeeaicthente:_aleniateermmmnbrsmeine_peaatneaenntl. spans. et ne. Namen nen ‘Agra 2005; Garhwal 1999; GN.D.U, Amritsar 2000, Kanpur 2002, 2010] {et us determine the image of a source of strength m at a point A with respect tothe cic vith O as entre, Let OA = fand lt B be inverse point of A with respect o the cite I abe the radius ofthe ciccle, then OA PB = o 0 that OB = aif Let 2) bean arttary point inthe plane of tbe cic Let there bea source af strength m at BI w be the amples P potential due to sources at and B, then we get Wann log @-f)mlog @- ah Ye y cinch arocrainergeeitd tf ora (4 Us 200d er (on0 stn) renders) ae MOTION IN TWO-DMENSIONS AND SOURCES AND SINKS sat Writing w= + acting re pars, we pt 4==0n12)[log 72000 - 9?-+ xin) tog rcosd-a°/ 9? 4¢rsn0)] [ou -tsatnee?8-22 ot 2% m[_207- Foose) 2ir—(a?/ frooss} Or 2 [Fe F2frooad ea 7? —2r(a" /Pyeosd, Hence normal velocity at any point Q on the ete “ee | Now, if we place a source of strength ~ m at O, the normal velocity due to it at Q will be “(mva) and hence the normal velocity of te system vill weduce to 20. Hence the image system for a source ouside a circle consists of an equal source ot the ‘inverse point and an equal sink athe centre ofthe circle {Kanpur 2002, 08] 522. Image of a doublet with regard to a circle. [Kanpur 2003, 06; Kurukshetra 2000; Rajasthan 1998] Let us determine the image of @ doublet 44" with its axis making an angle o with QA, ‘outside the citele, there being a snk ~ mat and source m at A’ Join Oat and Ot" Let Band {be the inverse points of and with rogard tothe stele with © as cente Tis ah oso ed o/ ate nn cir nor erp eran dann oe ED ant ape a Sebo died cabooses ft ano peo ch ler pee ee ts deh te pe Sate Then lim (ma) ° From (1) Osios" » OB'108, o showing tht tangles OAUA' and OB'B are similar. From these similar tiangles, we have BB on, 08" ox w strength of doublet B78 = hm (88) = fin = =a? wsing 2) and taking OA = 04" = f Thus the image of two-dimensional doublet at A with regard oa circle is another doublet atthe snverse poo B, the ace ofthe doublets making siplementary angles withthe rads Oba542 FLUID DYNAMICS: 5.23A. The Milne-Thomson circle theorem or simply the circle theorem. Statement : Let f(z) be the complex potential for a flow having no rigid boundaries and such that there are no singularities of flow within the circle | z | = a. Then, on introducing the solid circular cylinder | z | = a into the flow, the new complex potential is given by w=f(2)+ fla" /z) for |z|2a. [Rohilkhand 2002, 03, 05; Kanpur 2000, 09; Garhwal 2003, 05; Meerut 1998] Proof. Let C be the cross-section of the circular cylinder | z | = a. Then on C, =a” or Z=a"/z, Hence for points on the circle, we have w= f(2)+f@ !2)=f@)+/® or otiv=S@+/@ ~a) Since the quantity on R.H.S. of (1) is purely real, equating imaginary parts (1) gives w= 0 onC. Hence C is a streamline in the new flow. By hypothesis all the singularites of f (2) (at which sources, sinks, doublets or vortices may be present) lie outside the circle | z | = a and so the singularities of f(a*/z) lie inside the circle |z| =a, Hence the singularities of f(a?/2) also lie inside the circle | z | = a. Thus we find that the additional term 7(a?/z) introduces no new singularities into the flow outside the circle | z | = a. Hence | z | = @ is a possible boundary for the new flow and w= f(z)+f(a?/z) is the appropriate complex potential for the new flow. Remark 1. In the above proof of circle theorem we have used the following important results : Let u(d) and v(f) be real functions of a real variable ¢. Let f() = u(f) + iv () so that {@ is a complex function of the real variable ¢. Then conjugate of f(/) is denoted and defined as FO =uO-i9. On replacing real variable ¢ by the complex variable z (= x + iy), lz) and F(z) are defined as follows : S@=u@+iv@, F@ =u(2)-i9@) Again, S@ =u@+iv@), F@="@-we) On comparing the forms of f(z) and f(Z), we find that, since z = x + iy, F value of F(z) is obtained from lz) by replacing i throughout by — i. It then follows that F(z) is iy, the merely the complex conjugate of f(z) and accordingly, we write 7(Z) = (2) Remark 2, When a circular cylinder is present in the field of sources, sinks, doublets or vortices, the above theorem provides an easy method for determining the image system. Furthermore the theorem can also be used to determine modified flows when a long circular cylinder is introduced into a given two-dimensional flow. Consider the following application of “Circle theneam?™523B. To determine image system for a source outside a circle (or a circular cylinder) of radius a with help of the circle theorem. Refer figure of Art. 5.21. Let OA = f- Suppose there is a source of strength m at A where z =f, outside the circle of radius a whose centre is at O. When the source is alone in the fluid the ‘complex potential at a point P (2) is given by J) =—m log @-) Then Fie) =-mlog (ef) Fla | z) =-mlog(a?/z-f) MOTION IN TWO-DIMENSIONS AND SOURCES AND SINKS: 5.43 When the circle of section | z | = a is introduced, then the complex potential in the region [z|2a is given by w= f(2) + Fe /z) =—mlog (2-f)- mlog iz-f) = mop emg =<) =~ mlog (¢— f)- m log (a - zf) + mlogz m log (2 — f) — m log (J) - aif) + m log z m log (z — f) — m log (2 — @/ f) + m log z-m log(-f) x we log =) - m log = iY) + me log 2 + constant, a the constant (real or complex, — m log (- )) being immaterial from the view point of analysing, the flow. (1) shows that w is the complex potential of (@asource mat A,z=f (iia source mat B,z=a"/f (iif) a sink — mat the origin Since OA + OB = a°, A and B are the inverse points with respect to the circle | z | = a and so B is inside the circle. ‘Thus the image system for a source outside a circle consists of an equal source at the inverse point and an equal sink at the centre of the circle.2.11. The equation of continulty in spherical poler coordinates. [Mecrut 2008; Garhwal 1995, 96; Rajasthan 1997; Rohilkhand 2000] Let there be a fd particle at P whose spherical polar coordinates are (7, 0,6), where 720,05$528,050x. Let p(7. 0.4.0) be the density ofthe uid at Pat any time & With P 88 one corner construct @ small curvilinear parallelopiped (PORS, P{’R'S’) with its edges PP’ =r, are PQ=80, are PS=rsin 0}. Let grado and dy be the velocity components in the direction ofthe elements PP", ae AQ and are PS respectively. Then, we have 218 FLUID DYNAMICS Ma he id at pes in og ee PORS prin, par ne = 0:99), sy o se hd tt ost igh be opts POS 10810 <0.09)08e Berar » xing Th eve) The at nn ms pe ni kine win be de dome pape (PONS. OHS) dow gh FOR ak POMS ym 0 nt) Ms ta gh a ORS = Mase at ne gh ee POR =p1a9-[s.samir 29.96 =-8¢-2. 704), to the fast onder of approximation 2 or’ = 2 (pr sin0 4,508), by C) eo Similarly the net gain in mass per unit time within the clement due to flow through the faces PSP’ and ORR'Q" 2 1882 (9. br-rsind86 (0:8 -r086-45) ® and the net gain in mass per unit time within the element duc to flow through the faces. POOP” Le and SRR'S’ =n 086 gO Brg) ©KINEMATICS OF FLUIDS IN MOTION 219 .: Total rate of mass flow into the elementary parallelepiped ~=#rs0e| sno 2or4,yor Z onindar 20m] © Again, the mass of the fluid within the chosen element at time ¢ = - p8r-r80-r sin 039 Total rate of mass increase within the element =L0r sin 0 8r 8086) = Psindr 506522 oO 7 7 Suppose that the chosen region of the fluid contains neither sources nor sinks. Then by the law of conservation of the fluid mass, the rate of increase of the fluid within the element must be ‘equal to the rate of mass flowing into the element. Hence from (6) and (7), we have ~ér a0se[sno 24) +r Zosinon) +r 2.099 |= r sino5r 50592 a, 10,5 @ 1 or Be ~ (psinog,)+ 2 oq) =0. Bt pe ar OY + Fano a0 PsN Ome) +g HOH © Which is the desired equation of continuity in spherical polar coordinates and it holds at all points of the fluid free from sources and sinks. 2.11A. Generalised orthogonal curvilinear coordinates Let the rectangular cartesian coordinates (x, y, 2) of any point P in space be expressed in terms of three independent, single-valued and continuously differentiable scalar point functions uy, ty, us as follows =x (M2, Us) Y= (uy. ta, us) qa 2=2 (HM, M5) a Suppose that the Jacobian of x, y, 2 with respect t0 14, U>, ty does not vanish, that is, 8% ¥,2) OC a ts) can be inverted, i¢., uy, ta, us can be expressed sp in terms of x, y, z giving um = M(t, ¥.2), Hy = Uglts ¥s2)s ty = W466, ¥, 2). .Q) Thus to each point P(x, y, 2) we can assign a unique set of new coordinates (uy, 13) called the curvilinear coordinates of P. In this sense the equations (1) or (2) may be interpreted as defining a transformation of coordinates. The surfaces w(x, y, 2) = Cy, wget, ¥, 2) = Cy, ty (X, ys 2) = Cy, where Cy, Cy, C3 are constants, are called coordinate surfaces and each pair of these surfaces intersect in curves called coordinate curves or lines. ‘The surfaces u, = C, and u, = C, intersect in a curve along which the coordinate *u,” alone varies and hence it is called w,-curve or line. Similarly, we have u,- line and ‘us- line, ‘The coordinate axes are determined by the tangents PQ,, PQ, and PQ, to the coordinate curves uy = Cy, uy = Cy uy = Cy, Note carefully that the directions of these coordinate axes depend on the chosen point P of space and consequently the unit vectors associated with them are not necessarily constant #0. Then the transformation (1) Figure. Curvilinear coordinate system2 eS ee 5 ee oe Lc aahcoce | [Agra 2008, Himachal 1998, Meerut 2000, 01, Garhwal 2000, Rajasthan 1998], [Let there be a Muid particle at P whose eylindvial coordinates ae (0,2), whete 720,05052n, we =, (1) selons te OF +? — 2) i614 r=) ends? ya at'ce r= a; which are sreenilines, Hence the rough sketch of the streamlines is as shown in the following figure. In this figure there is a source of unit strength at origin O and there are two sinks cach of unit strength at A (a, 0) and a, 0), y,Ex. 3. Two sources, each of strength m are placed at the points (— a, 0), (a, 0) and a sink of strength 2m at the origin. Show that the streamlines are the curves (x? + y*)? =@ (xy? + x29) where 9, is a variable parameter. (U.P. P.CS. 1999; LA.S. 1999, 2003] Show also that the fluid speed at any point is (2ma?)/(r, r ry) where ry ry ry are the distances of the points from the sources and the sink. [LA.S. 1999, 2003; Meerut 2000; Garhwal 2005; Rohilkhand 2002] Sol. First Part. ‘The complex potential w at any point Pz) is given by w =~ m log (2 - a) —m log (2 + a) + 2m log z a Then the complex potential at a point z is given by —m log (@~ ¢) ‘Then ~m log @ - 6) and so. F@ 12) =m log (@/ 2) - c} Let a circular cylinder of section | z | = a be introduced. Then the new complex potential by Milne-Thomson’s circle theorem is given by w=f@+F@/2 for lzlz@ ie. w =~ m log (2c) ~ m log (a? /2)—e} or w =~ m log (z — c) ~m log {z ~ (a /c)} + m log z + constant, .) the constant (real or complex) being immaterial for the discussion of the flow. The point z= a/c is the inverse point of the point z = c with regard to the circle | z Hence (1) shows that the image of a line source in a right circular cylinder is an equal line source through the inverse point in the circular section in the plane of flow together with an equal line sink through the centre of the section, Ex. 12. Determine image of a tine doublet parallel to the avis of a right circular cylinder. Sol. Let there be a uniform line doublet of strength jt per unit length through the point = =c> a Furthermore let the axis of the line doublet be inclined at an angle a to x-axis, Then the ‘complex potential at point 2 is given by S@ =e) Kz - 0) Then F@=e'z-o Jia? (a? jz) = =e (OO Cy Let a circular cylinder of section | z | = a be introduced. Then the new complex potential by et Milne-Thomson’s circle theorem is given by w ye (@/2z)-¢514 FLUID DYNAMICS Ex. 17. In irotaional motion in two dimensions, prove that (qa)? +(@qld)* = qq (Agra 2012; Kanpur 2002; Meerut 2002,05) Sol Since the motion is iotatina, the velocity potential exists such that Vip =0°o/ae? +7967 a Again, = (04/2x)? +(0b/2" @ Differentiating (2) partially w.rt rand y respectively, we get 24. 280° 2 Fae" Oe Be * Oy Oxy io a) 24 (a) 28 Oe (3) deo ° . 4. (24 )-(2t) 6 si o5t-(3) eo) Adding (5) and (6) and simplifying, we get AB) -4 = 2O*H/2x7}? + 20*4/dxdy)" o ‘Next, squaring and adding (3) and (4), we get ‘(a-8)-a\e MOTION IN TWO-DIMENSIONS AND SOURCES SINKS 545, ‘Thus, (P1047)? + (1 Oxdy)? = (0g /ex)? + (Bq /2y)? 8) From (1) and (8), we find aq (Ogle)? + (Cal By = 21(6q/ ey" + Ca/3)"] or a¥?q= (Bq/dx)? +(2q/ avy.