Flow oF commessnce FLumos 123 shown in physics texts to be (69) oe) where the subscript S calls attention to the isentropic restraint on the process. IDEAL-GAS EQUATIONS. Subject to assumptions 1 to 6, Eqs. (6.2) to (69) apply to any fluid. In fact, they may be used for incompressible flow simply by assuming that the density p is constant, To apply them to compressible flow, itis necessary that the density be related to temperature and pressure, The simplest relation, and one of considerable engineering utility, is the ideal-gas law (Bq. (1.48)), which for the present purpose may be written in the form R wet (6.10) where R = molar gas-law constant, in units of mechanical energy per mole per degree absolute (M = molecular weight ‘The gas may either be pure or a mixture, but if it is not pure, the composition should not change, Equation (6.10) may be written logarithmically and then differentiated to give dp_dp aT 24S (61) pop T Since the specific heat c, is assumed to be independent of temperature, the enthalpy of the gas at temperature T is H =H +¢(T —T) (6.12) where H = enthalpy per unit mass at temperature T Hy = enthalpy at arbitrary temperature Ty ‘The differential form of Bq. (6.12) is d= aT (613) Acoustical velocity and Mach number of ideal gas. For an ideal gas, an isentropic path follows the equations ppm? = const 6.14) Tp-"- 4” = const (6.15) where 7 is the ratio of c, the specific heat at constant pressure, to ¢,, the specific heat at constant volume. For an ideal gas, 6.16) Since, by assumption, ¢, is independent of temperature, so are c, and 9.124 piu wecnamies The quantity (dp/dp), can be calculated by differentiating the logarithmic form of Eq. (6.14), giving BP io ana (@) - PoP ap)s Substituting into Eq. (6.9) yields ? ‘TR on f= pa @ Equation (6,10) is used to establish the second equality in Ea, (6.17), which shows that the acoustical velocity of an ideal gas is a function of temperature only. From, Eqs. 6.1) and (6.17) the square of the Mach number of an ideal gas is pe Ni, = = (6.18) 9.7 gayTRIM ‘THE ASTERISK CONDITION, The state of the fluid moving at its acoustic velocity is important in some processes of compressible-fluid flow. The condition where u=a and Ny,= 1 is called the asterisk condition, and the pressure, temperature, density, and enthalpy are denoted by p*, Tp, and H* at this tate. 2 - STAGNATION TEMPERATURE. The stagnation temperature of a high-speed fluid is defined as the temperature the fluid would attain were it brought to rest adiabaiically without the development of shaft work. The relation between the actual fluid temperature, the actual fluid velocity, and the stagnation temperature is found by using the total-eneray equativu (6.3) aud the enthalpy equation (6.12) The terminal a in Eq. (6.3) and reference state 0 in Eq. (6.12) are identified with the stagnation condition, and stagnation is denoted by subscript s, Also, terminal bin Eq. (6.3) is chosen as the state ofthe flowing gas, and this subscript is dropped. ‘Then, since the process is adiabatic and Q = 0, Eq, (6.3) becomes w 29d Eliminating H ~ Hg from Eq. (6.19) by substitution from Eq, (6.12) gives, for the stagnation temperature , HH, =H-Hy (6.19) tr, 20 . 2gle, Lead ‘The stagnation enthalpy H, is defined by the equation e H,=H (6.21) pe Hors (6.21) Equation (6.3) can be written Q Hog ~ Huy = (Ty — Tey (622)Flow oF commressine ruins 125 where H,, and Hy are the stagnation enthalpies at states a and b, respectively. For an adiabatic process, Q=0, Z,= Ty, and the stagnation temperature is constant. PROCESSES OF COMPRESSIBLE FLOW “The individual processes to be considered in this chapter are shown diagrammatic- ally in Fig. 6.1. It is assumed that a very large supply of gas at specified temperature and pressure and at zero velocity and Mach number is available. The origin of the gas is called the reservoir, and the temperature and pressure of the gas in the Conergent ‘section Divergent section Reservoir Receiver Adiabatic. Thermal insulation oo) Reservoir Receiver Isentropic section | Friction section Isothermal ‘Heating jacker- fe) FIGURE 6.1 {@) entropic expansion in convergent-divergent nozzle. (9) Adiabatic titional low. () Isothermal feetonal Ros:126 Fup mecnamies reservoir ate called reservoir conditions. The reservoir temperature is a stagnation value, which does not necessarily apply at other points in the flow system. From the reservoir the gas is assumed to flow, without friction loss at the entrance, into and through a conduit. The gas leaves the conduit at definite temperature, velocity, and pressure and goes into an exhaust receiver, in which the pressure may be independently controlled at a constant value less than the reservoir pressure, Within the conduit any one of the following processes may occur: 1, An isentropic expansion. In this process the cross-sectional area of the conduit must change, and the process is described as one of variable area. Because the process is adiabatic, the stagnation temperature does nat change in the conduit. Such a process is shown diagrammatically in Fig. 6.la. 2. Adiabatic frictional flow through a conduit of constant cross section. This process is irreversible, and the entropy of the gas increases, but as shown by Eq. (622), since Q = 0, the stagnation temperature is constant throughout the conduit. This process is shown in Fig, 6.16, 3. Isothermal frictional flow through a conduit of constant cross-sectional area, accompanied by a flow of heat through the conduit wall sufficient to keep the temperature constant, This process is nonadiabatic and noniscntropic; the stagnation temperature changes during the process, since T is constant, and by Eq, (6.20), T, changes with uw. The process is shown in Fig. 6.1c. The changes in gas temperature, density, pressure, velocity, and stagnation temperature are predictable from the basic equations. The purpose of this section is to demonstrate how these three processes can be treated analytically on the basis of such equations.t Flow through Variable-Area Conduits A conduit suitable for isentropic flow is called a nozzle, As shown in Fig. 6.14, a complete nozzle consists of a convergent section and a divergent section joined by a throat, which is a short length where the wall of the conduit is parallel wit the axis of the nozzle. For some applications, a nozzle may consist of a divergent section only, and the throat connects directly with the receiver. The configuration of an actual nozzle is controlled by the designer, who fixes the relation between S, the cross-sectional area, and L, the length of the nozzle measured from the entrance. Nozzles are designed to minimize wall friction and to suppress boundary. layer separation. The convergent section is rounded and can be short, since separation does not occur in & converging channel. To suppress separation in the TA sencraized treatment, including general heat transler to and from the gas, injection of gas ino the conduit, variations in specie heat end molecular weight, chemical reactions, the drag of internal bodies, and change of phase, i given in Ref. 3.FLOW OF CompnessiaLe FLUIDS 1Z7 ivergent section, the diverging angle is made small, and this section is therefore relatively long. The nozzle entrance is sufficiently large relative to the throat to permit the velocity at the entrance to be taken as zero and the temperature and pressure at the entrance to be assumed equal to those in the reservoir. ‘The purpose of the convergent section is to increase the velocity and decrease the pressure of the gas. At low Mach numbers the process conforms essentially to the usual Bernoulli relation for incompressible flow [Eq. (4.25). In the convergent section flow is always subsonic, but it may become sonic at the throat. Mach numbers greater than unity cannot be generated in a convergent nozzle. In the divergent section, the flow may be subsonic or supersonic. The purpose of the divergent section differs sharply in the two situations, In subsonic flow the purpose of the section is to reduce the velocity and regain pressure, in accordance with the Bernoulli equation. An important application of these nozzles is the measure- ment of fluid flow, which is discussed in Chap. 8. In supersonic flow, the usual purpose of the divergent section is to obtain Mach numbers greater than unity for use in experimental devices such as wind tunnels, Flow through a given nozzle is controlled by fixing the reservoir and receiver pressures. For a given flow through a specific nozzle, a unique pressure exists at cach point along the axis of the nozzle. The relation is conveniently shown as a plot of p/pg versus L, where po is the reservoir pressure and p the pressure at point L. Figure 6.2 shows how the pressure ratio varies with distance and how changes in receiver pressure at constant reservoir pressure affect the pressure distribution, ‘The pressures at the throat and in the receiver are denoted by p, and p,. respectively. If p, and po are equal, no flow occurs and the pressure distribution is, represented by the line aa’. If the reeciver pressure is slightly below the reservoir pressure, flow occurs and a pressure distribution such as that shown by line abc FIGURE 62 ‘Variation of pressure ratio with distance from novale inlet.125 FLUID MECHANIC is established. Pressure recovery in the convergent section is shown by line be. The maximum velocity occurs at the throat. If the receiver pressure is further reduced, the flow rate and the velocity throughout the nozzle increase. A limit is attained when the velocity at the throat becomes sonic. This case is shown by line ade, where p,= p*, t= a, and Ny. = 1. The ratio p*/py is called the critical pressure ratio and is denoted by r...Flow is subsonic at all other points on line ade. As the receiver pressure is decreased from that of point a’ to that of point , the mass flow rate through the nozzle increases. The flow rate is not affected by reduction of pressure below that corresponding to critical flow. Figure 63 shows how mass flow rate varies with the pressure ratio p,/p9. The flow rate attains its maximum at point , which is reached when the pressure ratio in the throat is reduced to its critical value. Further reduction in pressure p, does not change the flow rate. The reason for this is as follows: when the velocity in the throat is sonic and the cross-sectional area of the conduit is constant, sound waves cannot move upstream into the throat and the gas in the throat has no way of receiving a message from downstream. Further reduction of the receiver pressure cannot be transmitted into the throat. IF the receiver pressure is reduced to the level shown by point f in Fig, 62, the pressure distribution is represented by the continuous line adgh: This line is unique for a given gas and nozzle, Only along the path dghf is supersonic flow possible. If the receiver pressure is reduced below that of point f, for example, to point k, the pressure at the end of the nozzle remains at that of point f and flow through the nozzle remains unchanged. On issuing from the nozzle into the receiver the gas suffers a sudden pressure drop from that of point f to that of point k. The presaure change is accompanied by wave phenumena in the receiver If the receiver pressure is held at a level between points e and f, pressure. distribution curves of the type dag'i and dhhij are found. Sections dg and dh wes. Gonstont tow rote 4___ Maximum How rate Mass flow rote ri 7 TO FIGURE 63 Pressure ratio p, 0, ‘Mass flow rate through nozae.Flow OF comPressrnLe FLUIDS 129 represent isentropic supersonic flow. The sudden pressure jumps gg’ and hi represent shock waves, where the flow changes suddenly from supersonic to subsonic. Shocks are thermodynamically irreversible and are accompanied by an increase in entropy in accordance with the second law of thermodynamics. Curves g'i and Mj represent subsonic flows in which ordinary pressure recovery is taking ace ree Tye area below line adghf is not accessible to any kind of adiabatic lo Isentropic flow is confined to the subsonic area aded’a and the single line dghf. ‘The qualitative discussions of Figs. 62 and 6.3 apply to the flow of any compressible fluid. Quantitative relations are most easly found for ideal-gas flow. EQUATIONS FOR ISENTROPIC FLOW. The phenomena occurring in the flow of ideal gas through nozzles are described by equations derivable from the basic ‘equations given earlier in this chapter. Change in gas properties during flow. The density and temperature paths of the gas through any isentropic flow are given by Eqs. (6.14) and (6.15). The constants are evaluated from the reservoir condition. This gives P_Po PP Tr To prin (6.23) (624) ‘These equations apply both to frictionless subsonic and supersonic flow, but they must not be used across a shock front. Velocity in nozzle. In the absence of friction, the mechanical-energy balance [Eq. (6.6)] becomes simply aa (=) (25) Eliminating p from Eq, (6.25) by substitution from Eq. (6.23) and integrating from a lower limit based on the reservoir, where p = Po, p = Po. and u = 0, gives LG)--E Ls jo 2G) Po dnoP ae [: - ey (626 ‘A Mach-number form of Eq, (6.26) is convenient. It is derived by substituting w from Eq. (6.2) into the first equality of Eq. (6.18) and eliminating p/p, by130 FLUID Mechanics substitution from Eq, (6.23), This gives PG)" Solved explicitly for the pressure ratio, Ea. (6.27) becomes e 1 po + fy yzINR ‘The critical pressure ratio, denoted by y,, is found from Eq. (6.28) by substituting p* for p and 1.0 for Ny (2,)" 1m i= yF1, (627) y (6.28) me Po The mass velocity is found by calculating the product of u and p, using Eqs. (623) and (6.26), Gap Pueeate(2)" fy (2) 630 EFFECT OF CROSS-SECTIONAL AREA. The relation between the change in cross-sectional area, velocity, and Mach number is useful in correlating the various cases of nozzle flow. Substitution of p from Eq, (6.25) into the continuity equation (62) eves du aS fdp\ udu du a8 _ (ap) wa 6: aS -(b)a-° ne r, Subscript 5 is used to call attention to the fact that the flow is isentropic. From Eq. (69) ‘ap a) (6.32) (2), a Maes Eliminating (dp/dp), from Eqs. (6.31) and (6.32) gives and, by substituting Ny, from Eq. (6.1), Hoa, -) (6.33) Equation (6.33) shows that for subsonic flow, where Ny, < 1, the velocity increases with decreasing cross section (converging conduit) and decroases with increasing cross section (diverging conduit). This corresponds to the usual situation of incompressible flow. Lines abc, ade, g', and ij of Fig. 6.2 represent examples.Flow oF compresses FLus 131 For supersonic flow, where Ny. > 1, the velocity increases with increasing cross section, as in the diverging section of the nozzle. This conforms to line dghf of Fig. 6.2. The apparent anomaly of supersonic flow is a result of the variation in density and velocity along an isentropic path, Since the mass flow rate is the same at all points in the nozzle, by continuity the cross-sectional area of the nozzle must vary inversely with the mass velocity up. The velocity steadily increases with Mach number, and the density decreases. However, at Ny, = 1, the value of G goes through a maximum. In the subsonic regime the velocity increases faster than the density decreases, the mass velocity increases, and $ decreases. In the supersonic regime, the increase in velocity is overcome by a sharper decrease in density, the mass velocity decreases, and S increases to accommodate the total mass flow. This behavior of G is demonstrated by studying the first and second derivatives of Eq, (630) in the usual manner for investigating maxima and minima. Example 6.1. Air enters a convergent-divergent nozzle at a temperature of 555.6 K (10G0°R) and a pressure of 20 atm. The throat area is one-half that of the discharge of the divergent section. (a) Assuming the Mach number in the throat is 0.8, what are the values of the following quantities at the throat: pressure, temperature, linear velocity, density, and mass velocity? (8) What are the values of p*, T*, u*, and G* covtesponding to reservoir conditions? (¢) Assuming the nozle is to be used supersonically, what is the maximum Mach number at the discharge of the divergent section? For air y = 14 and M = 29, Solution () The pressure at the throat is calculated from Eq. (6:28): 2” fate on pe= 13.12 atm From Eq. (6.10), since R = 82.056 x 10-* atm-m®/kg mol-K 2x29 82056 x Substituting po/po from Eq. (6.10) into Eq. (6.26) gives, for the velocity in the EO" i My = 1) cy Fra wage x TOS : BROS TOTES page 291.41) aeoeee = 1.1175 (m?-atjig)!* ‘These units are converted to meters ‘atm = 1.01325 x 10° N/m’, By de 12.72 kplen® (0.395 by) Po 5556 x second as follows. From Appendix 2, ion, 1N = 1 kgm/s®. Hence y= 1117S /LO1S35 10? = 355.7 ms (1167 0/5)132 uur wecraies The density at the throat i, from Eq, (623), ‘n a eo) = 12.72 x 0.65644 = 9.41 kgm? (0387 Ib?) ‘The mass velocity atthe throat is G,= up = 385.7 «9.41 = 3348 kgm? (686 Ibs) [he mass velocity can also be calculated directly by use of Eq. (6.30):] The temperature at the throat is, from Eq. (6.24), T= x(2) = 5556 x 0656-44 = 492.5 K 886.5°R) () From Eq, (629) rari. From Eqs. (6.24) and (6.29) mass 7) lari ( 2 ) ON 20528 ph = 20 0528 = 1056 am 463 K (833.4°R) £.(e)" 0 \Po, pt = 12.72 x 0.5284!" = 8.06 kg/m* (0.503 lb/ft?) From Eq, (6.30) PERCE SES ESO E Ss of 2 x 20x 10 o.s2"4 fos = 3476 keg/m?-s (712 Ibyft-s) o 376 SA a 431 mys (4 pape = BE MACS Hs) (6 Since, by continuity, G cc 1/S, the mass velocity at the discharge is 1738 kgym?.s (356 lbjit.s) From Bq, (623) From Eq, (630) RIA TRE RIO x i" THT p, us ine & 27 7 T0325 x10 P-® 2) Oe ‘This equation is solved for py iP to give >» 0.0939