304 PLESSIS "The Method of Repeating Variables Step 1 List the parameters in the problem snd count thei total number n Step 2: Lis the primary dimensions of each of then parameters, Step 3: Set the reduction jas the number ‘of primary dimensions. Caeulite k ‘he expeced numberof I's, ken-j Step 4: Choose repeating parameters. Step 5: Construct the Ks, and manipulate as necessary. Step 6: Wate the ial functional relationship and check your algebra, FIGURE 7-22 A concise summary of the six steps that comprise the method of repeating variables, ‘wy =a versal speed = gravitational feceleaton inthe = (datum plane) FIGURE 7-23 Setup for dimensional analysis of a ball falling in a vacuum. Elevation z is a function of time , initial verti- cal speed wy, initial elevation za, and gravitational constant g og Detailed description of the six steps that comprise the method of repeating variables’ Step 1 List the parameters (dimensional variables, nondimensional variables, and dimensional constants) and count them. Let nbe the total number of parameters in the problem, including the dependent variable. Make sure that any listed independent parameter is indeed independent of the others, ie, it cannot be expressed in terms of them. (E.g., don't include radius rand area A= ar’, since rand A are not independent.) Step 2 List the primary dimensions for each of the parameters. Step 3 Guess the reduction j, As a first guess, set j equal to the number of primary dimensions represented in the problem. The expected nur ber of I's (k) is equal to minus j, according to the Buch theorem, The Buckingham Pi theorem: a1) If at this step or during any subsequent step, the analysis does not work out, verfy that you have included enough parameters in step 1. Otherwise, go back and reduce j by one and try again. Step 4 Choose j repeating parameters that will be used to construct each I Since the repeating parameters have the potential to appear in each TI, be sure to choose them wisely (Table 7-3). Step Generate the II's one at a time by grouping the j repeating parameters with one of the remaining parameters, forcing the product to be dimensionless. In this way, construct all k's. By convention the first II, designated as TI, is the dependent II {the one on the left side of the list). Manipulate the I's as necessary to achieve estab- lished dimensionless groups (Table 7-5). Step Check that all the I's are indeed dimensionless. Write the final functional relationship in the form of Eq. 7-11 clevation z of the ball must be a function of time f, initial vertical speed wo, initial clevation zp, and gravitational constant g (Fig. 7-23). The beauty of dimensional analysis is that the only other thing we need to know is the pri- mary dimensions of each of these quantities. As we go through each step of the method of repeating variables, we explain some of the subtleties of the technique in more detail using the falling ball as an example. Step 1 ‘There are five parameters (dimensional variables, nondimensional variables, and dimensional constants) in this problem; n = 5. They are listed in fun tional form, with the dependent variable listed as a function of the indepen- dent variables and constants: List of relevant parameters: z Ut WZ) n= 5Step 2 ‘The primary dimensions of each parameter are listed here. We recommend writing each dimension with exponents since this helps with later algebra. z ' Mo % 8 wy (e) {Ue} {Ly (Lt?) Step 3 AAs a first guess, jis set equal to 2, the number of primary dimensions repre- sented in the problem (Land 0), Reduction: j=2 I this value of jis correct, the number of TI’s predicted by the Buckingham Pi theorem is Number of expected W's: k Step 4 We need to choose two repeating parameters since j ~ 2. Since ths is often the hardest (or atleast the most mysterious) part ofthe method of repeating variables, several guidelines about choosing repeating parameters are listed in Table 7-3 Following the guidelines of Table 7-3 on the next page, the wisest choice of two repeating parameters is wand zy Repeating parameters: Wp and < Step 5 Now we combine these repeating parameters into products with each of the remaining parameters, one at atime, to create the TT's. The first ITs always the dependent TT and is formed with the dependent variable z Dependent Il: I, = wget 745) where a; and b; are constant exponents that need to be determined. We apply the primary dimensions of step 2 into Eq. 7-15 and force the II to be dimensionless by setting the exponent of cach primary dimension to zero: ) = (Lue Dimensions of fly: (Uh) = (L} = { Since primary dimensions are by definition independent of each other, we {equate the exponents of cach primary dimension independently to solve for exponents a, and bj (Fig. 7-24) Time: Wate = =a Length: (1) = (LILMLN)0= 1 bay + by ben Equation 7-15 thus becomes, 7-16) FIGURE 7-24 The mathematical rules for adding and subtracting exponents during multiplication and division, respectively.DIMENSIONAL ANALYSIS AND MODELING. 306 oe) Guidelines for choosing repeating parameters in step 4 of the method of repeating variables” Guideline Comments and Application to Present Problem 1 Never pick the dependent variable. Otherwise, it may appear in all the T's, which is undesirable. The chosen repeating parameters must not by themselves be able to form a dimensionless group. Otherwise, it would be impossible to generate the rest of the II's. ‘The chosen repeating parameters must represent all the primary dimensions in the problem, Never pick parameters that are already dimensionless. These are T's already, all by themselves. Never pick two parameters with the same dimensions or with dimensions that differ by only an exponent. ‘Whenever possible, choose dimensional constants over dimensional variables so that ‘only one TI contains the dimensional variable. Pick common parameters since they may appear in each of the IIs Pick simple parameters over ‘complex parameters whenever possible. Inthe present problem we cannot choose z, but we must choose from among. the remaining four parameters. Therefore, we must choose two of the following parameters: f, Wo, 2, and g. In the present problem, any two of the independent parameters would be valid according to this guideline. For illustrative purposes, however, suppose we have to pick three instead of two repeating parameters. We could not, for example, choose f, mg, and zp, because these can form a II all by themselves (tw) ‘Suppose for example that there were three primary dimensions (m, L, and t) and two repeating parameters were to be chosen. You could nat choose, say, a length and a time, since primary dimension mass would not be represented in the dimensions of the repeating parameters. An appropriate choice would be a density and a time, which together represent all three primary dimensions in the problem. ‘Suppose an angle @ were one of the independent parameters. We could nat choose ‘as a repeating parameter since angles have no dimensions (radian and degree are dimensionless units) In such a case, one ofthe IT's is already known, namely 0. In the present problem, two of the parameters, z and zy, have the same dimensions (lengt). We cannot choose both of these parameters. (Note that dependent variable z has already been eliminated by guideline 1.) Suppose one parameter has dimensions of length ang another parameter has dimensions of volume. In dimensional analysis, volume contains only one primary dimension (length) and is not dimensionally distinct from fength—we cannot choose both of these parameters. If we choose time t as a repeating parameter in the present problem, it would appear in all three II's. While this would not be wrong, it would not be wise since we know that ultimately we want some nondimensional height as 2 function of some nondimensional time and other nondimensional parameters). From the original four independent parameters, this restricts us to wg, 2, and g. In fluid flow problems we generally pick a length, a velocity, and a mass or density (Fig. 7-25). It is unwise to pick less common parameters like viscosity 0r Surface tension a, since we would in general not want yor «to appear in each of the II's In the present problem, wo and 2, are wiser choices than g. It's better to pick parameters with only one or two basic dimensions (e.g., a length, atime, a mass, or a velocity) instead of parameters that are composed of several basic dimensions (e.g, an energy of a pressure). These auidelines, heat nai, nlp you to pick epetng parameters tat ual led to estab shed nonimensionl groups wih minal eo, In similar fashion we create the first independent TI (II,) by combining the repeating parameters with independent variable 1. First independent UL Tl, = wgeh Dimensions of zz (U;} = (LX"} = {rwge's) cL yeLhyEquating exponents, Time: engi: (L°) = (LAL) gis thus wt on, Finally we create the second independent IT (II,) by combining the repeat- ing parameters with g and forcing the II to be dimensionless (Fig. 7-26). Second independent TI: Ty = gwiixh Dimensions of sz (UIs) = {1%} = (gwh2h) = (LALA LY Equating exponents, Time: p= (rs) 0 Length: (12) = (LILLD) Ttatby by ly is thus a 18) All three II's have been found, but at this point it is prudent to examine them to see if any manipulation is required. We see immediately that IT, and TH, ate the same as the nondimensionalized variables z* and r* defined by Eq, 7-6—no manipulation is necessary for these. However, we recognize that the third IT must be raised to the power of —} to be of the same form as an established dimensionless parameter, namely the Froude number of Eq. 7-8: Modified Uy o-19) Such manipulation is often necessary to put the II's into proper estab- lished form, The I of Eq. 7-18 is not wrong, and there is certainly no ‘mathematical advantage of Eq. 7-19 over Eg. 7-18. Instead, we like to say that Eq. 7-19 is more “socially acceptable” than Eq. 7-18, since it is a named, established nondimensional parameter that is commonly used in the literature. In Table 7-4 are listed some guidelines for manipulation of nondimensional II groups into established nondimensional parameters. ‘Table 7-5 lists some established nondimensional parameters, most of which are named after @ notable scientist or engineer (see Fig. 7-27 and the Historical Spotlight on p. 311). This list is by no means exhaustive, When- ever possible, you should manipulate your IT's as necessary in order to con- vert them into established nondimensional parameters, 307 erie FIGURE 7-25 Itis wise to choose common parameters as repeating parameters since they may appear in each of your dimensionless [I groups FIGURE 7-26 ‘The TI groups that result from the method of repeating variables are guaranteed to be dimensionless ‘because we force the overall exponent of all seven primary dimensions to be zero,308 DIMENSIONAL ANALYSIS AND MODELING. ue Guidelines for manipulation of the I's resulting from the method of repeating variables" Guideline Comments and Application to Present Problem 1. We may impose a constant (dimensionless) exponent on a Il or perform a functional ‘operation on a IL 2, We may multiply a Il by a pure (dimensionless) constant. 3. We may form a product (or quot ‘of any I with any other II in the problem to replace one of the IT's nt) 4, We may use any of guidelines 1 to 3 in combination. 5, We may substitute a dimensional parameter in the Il with other parameter(s) of the same dimensions. We can raise a Il to any exponent 1 (changing it to II") without changing the dimensionless stature of the II. For example, in the present problem, we imposed an exponent of ~1/2 on Ils. Similarly we can perform the functional operation sin(II), exp(ll), etc., without influencing the dimensions of the IL ‘Sometimes dimensionless factors of a, 1/2, 2, 4, etc. are included in a II for Convenience. This is perfectly okay since such factors do not influence the dimensions of the TI. ‘We could replace Il, by IlsIf,, Ily/llp, ete. Sometimes such manipulation is necessary to convert our Il into an established Il. In many cases, the established IT would have been produced if we would have chosen different repeating parameters. In general, we can replace any II with some new II such as All3® sin(I1,°), where A, B, and Care pure constants For example, the II may contain the square of a length or the cube of a length, for which we may substitute a known area or volume, respectively, in order to make the TT agree with established conventions. se guns are ust stp 8 of he me 5 rapeatng variables anaes alo you canvert your aneimensona I gros into stance, FIGURE 7-27 Established nondimensional parameters are usually named after ‘notable scientist or engineer. Step 6 We should double-check that the II's are indeed dimensionless (Fig. 7-28), ‘You can verify this on your own for the present example. We arc finally ready to write the functional relationship between the nondimensional parameters. Combining Eqs. 7-16, 7-17, and 7-19 into the form of Eq. 7-11, z) r, in terms of the nondimensional variables z* and 1* defined previously by Eq, 7-6 and the definition of the Froude number, Final result of dimensional analysis: 2 = fit*, FO) (7-20) Relationship between W's: WI, = flys) > It is useful to compare the result of dimensional analysis, Eq. 7-20, to the exact analytical result, Eq. 7-10. The method of repeating variables properly predicts the functional relationship between dimensionless groups. However, ‘The method of repeating variables cannot predict the exact mathematical {orm of the equation, ‘This is a fundamental limitation of dimensional analysis and the method of repeating variables. For some simple problems, however, the form of the equation can be predicted to within an unknown constant, as is illustrated in Example 7-7.ug ‘Some common established nondimensional parameters or II's encountered in fluid mechanics and heat transfer’ Name Archimedes number Aspect ratio Biot number Bond number Cavitation number Darcy friction factor Drag coefficient Eckert number Euler number Fanning friction factor Fourier number Froude number Grashof number Jakob number Knudsen number Lewis number Lift coefficient Definition Ratio of Significance pal Gravitational force Ane OP ‘Viscous force Lok Length Length Baw D Wide “Diameter wi te Surface thermal resistance & Intemal thermal resistance ste, — pe Gravitational force 7 Surface tension free P=, Press = Vaperpecsnue eo dar Tnertial pressure 2P sometines = _ 8 Wall friction force - pve Tnertial force fo Dag force Oo ~ Sev Dynamic fore ARE YOUR PI’S Kinstic energy DIMENSIONLESS? Entilpy Pressure difference ‘Dynamic pressure Br, ‘Wall friction force Tnerial force Physical time ‘Thermal diffusion time FIGURE 7-28 A quick check of your algebra is always wise, — V2) _tnenial force Val eL) Gravitational force PrN Buoyancy force “Viscous foree Sensible energy Tate energy Mean fre pathlength “Charteris Fensh “Therma fusion Lift force Dynamic free (Continued)310 n Definition Ratio of Significance Flow speed h num 1a (sometimes a Mach number Ma (sometimes M) peal are Convection heat transfer it number ‘Conduction heat transfer Nusselt number Gonduction heat wanster ‘Bulk heat transfer Peclet number Conduction heat transfer Power number Prandtl number ‘Thermal diffusion Static pressure difference Pressure coefficient Dynamic pressare _ SBIATIL Buoyancy force ae Viscous force _ VL _ VL Inertial foree Rayleigh number Ra. Reynolds number Re= = Viscous force Les Buoyancy force Richardson number Ri = =22P — pV Tetial foree Viscous diffusion Species diffusion Overall mass diffusion ‘Species diffusion Enthalpy Internal energy Heat transfer ‘Thermal capacity ‘Schmidt number ‘Sherwood number Specific heat ratio Stanton number 2 PPV Particle relaxation time ‘Characteristic flow time (Characteristie flow time Period of oscillation VL Inerial force %, ‘Surface tension force Stokes number Stk (sometimes St) asf ‘Strounal number St (sometimes $ or 1) Weber number we Aig» charactors are, Dis a character camel, is» character raquency ie, Lis a characte iste engin t's. sorctenave tne, Tis a eraractestc absolute) temperate, Ves ehractersi on, iis sche wah sachet som wns castrate angi vey a) Get ‘evaporation k thermal vty, P= pressure, Tay ~ saturation te 0, Vola flow at, (2 thermal difusivty B= coecant of thernal pation, A mean fee path ength ~ Scot 1 = kireate sce p= ld ders. = la dens, ~ parte ers, ~ Sol NSai roan Ep- Poe cn J Guest Author: Glenn Brown, Oklahoma State University \ Commonly used, established dimensionless numbers have been given names for convenience, and to honor persons ‘who have contributed in the development of science and engineering. In many cases, the namesake was not the first 10 define the number, but usually he/she used it or a similar parameter in his/her work. The following is a list of some, but not all, such persons. Also keep in mind that some numbers may have more than one name. Archimedes (287-212 wc) Greck mathematician who defined buoyant forces. Biot, Jean-Baptiste (1774-1862) French mathematician who did pioneering work in hea, electricity, and elasticity. He also helped measure the are of the meridian as part of the metric system development. Darcy, Henry P.G. (1803-1858) French engineer who per- formed extensive experiments on pipe flow and the frst ‘quantifiable filtration tests. Eekert, Emst R.G. (1904-2004) German-American engineer and student of Schmidt who did early work in boundary layer heat transfer. Euler, Leonhard (1707-1783) Swiss mathematician and associate of Daniel Bernoulli who formulated equations ‘of fluid motion and introduced the concept of centrifugal machinery. Fanning, John T. (1837-1911) American engineer and textbook author who published in 1877 a modified form ‘of Weisbach’s equation witha table of resistance values, ‘computed from Darcy's data. Fourier, Jean B. J. (1768-1830) French mathematician who did pioneering work in heat transfer and several other topics. Froude, William (1810-1879) English engineer who ‘developed naval modeling methods and the transfer ‘of wave and boundary resistance from model to prototype. Grashof, Franz (1826-1893) German engineer and eduea- tor known as a prolific author, editor, corrector, and dispatcher of publications. Jakob, Max (1879-1955) German-American physicist, ‘engineer, and textbook author who did pioneering work in heat transfer. Kaudsen, Martin (1871-1949) Danish phys ‘develop the kinetic theory of gases. Lewis, Warren K. (1882-1975) American engineer who researched distillation, extraction, and fluidized bed reactions. Mach, Ernst (1838-1916) Austrian physicist who was first to realize that bodies traveling faster than the 1 who helped speed of sound would drastically alter the properties of the fluid. His ideas had great influence on twentieth- century thought, both in physics and in philosophy, and influenced Einstein's development of the theory of relativity. ‘Nusselt, Wilhelm (1882-1957) German engineer who was the first to apply similarity theory to heat transfer, Peclet, Jean C. E, (1793-1857) French educator, physicist, tnd industrial researcher. Pranddl, Ludwig (1875-1953) German engineer and develop- er of boundary layer theory who is considered the founder ‘of modem fluid mechanics. Lord Raleigh, John W. Strut (1842-1919) English scientist ‘who investigated dynamic similarity, cavitation, and bubble collapse. Reynolds, Osborne (1842-1912) English engineer who investigated flow in pipes and developed viscous flow equations based on mean velocities. Richardson, Lewis F. (1881-1953) English mathematician, physicist, and psychologist who was a pioneer in the application of fluid mechanics to the modeling of atmospheric turbulence. ‘Schmit, Ernst (1892-1975) German scientist and pioneer in the field of heat and mass transfer. He was th first to measure the velocity and temperature Field in a free convection boundary layer. Sherwood, Thomas K. (1903-1976) American engineer and. educator. He researched mass transfer and its interac tion with flow, chemical reactions, and industrial process ‘operations. Stanton, Thomas E. (1865-1931) English engineer and student of Reynolds who contributed to a number of areas of fluid flow. Stokes, George G. (1819-1903) Irish scientist who devel- ‘oped equations of viscous motion and diffusion, Strouhal, Vincenz (1850-1922) Czech physicist who showed thatthe period of oscillations shed by a wire are related to the velocity ofthe air passing over it Weber, Moritz (1871-1951) German professor who applied similarity analysis to capillary flows.