22-Mec-B6 Advanced Fluid Mechanics · December 2014
Question 2 of 5: Blowdown of a pressurised air canister through a convergent–divergent valve
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Exams, December 2014 —
07-Mec-B6 Advanced Fluid Mechanics. Three hours, OPEN BOOK, any non-communicating
calculator permitted. Five questions are printed; candidates answer any four and all
questions are equally weighted, so each question carries 25 of the 100 marks. A four-page
aid sheet (compressible flow, boundary-layer integral relations, the Navier–Stokes
equations in Cartesian and cylindrical-polar form, and potential-flow building blocks) is
bound into the paper; every relation used below is taken from it. All five questions
are solved here.
Reference texts. Solutions follow the conventions of the texts the
EGBC syllabus recommends for this subject:
F. M. White, Fluid Mechanics, 8th ed. — potential flow and images (§8.3–8.5),
turbulent flat-plate layers (§7.4), turbomachinery similarity (§11.3–11.5).
F. M. White, Viscous Fluid Flow, 3rd ed. — exact solutions of the
Navier–Stokes equations in annular geometry (§3.2–3.3).
J. D. Anderson, Modern Compressible Flow, 3rd ed. — quasi-one-dimensional
nozzle flow and normal shocks (Ch. 3 and Ch. 5).
P. K. Kundu, I. M. Cohen & D. R. Dowling, Fluid Mechanics, 6th ed. —
complex potential, Blasius theorem and image forces (Ch. 6).
R. W. Fox et al., Introduction to Fluid Mechanics, 10th ed. —
entrance-region and duct-flow treatment (Ch. 8).
SI units throughout; air and water properties are those printed in the question, and
Canadian practice (SI, absolute pressures stated explicitly) is followed.
Question 2: Blowdown of a pressurised air canister through a convergent–divergent valve (25 marks)
Find. Mass flow, exit temperature and exit speed at 10.0 MPa and at 5.0 MPa;
the canister pressure that places a normal shock exactly at the exit plane with the mass flow
and post-shock state there; and the canister pressure below which the throat is no longer choked.
Figure 2.1 — Vessel and convergent–divergent valve. The throat sets the mass flow once choked; the area ratio alone fixes the exit Mach number while the nozzle runs full and supersonic.
Approach. The area ratio alone fixes the exit Mach number whenever the
nozzle runs full and supersonic, so the whole question reduces to one isentropic area-Mach
solution, the aid-sheet mass-flow formula evaluated at the choked throat, and the normal-shock
relations applied at the exit plane. The canister is treated as quasi-steady: the flow through
the valve responds instantly to the instantaneous stagnation state inside.
Fix the design exit Mach number from the area ratio. The aid-sheet mass-flow
relation, written once at the throat (M = 1) and once at the exit and divided, gives the
standard area–Mach function
$$\frac{A}{A^{*}}=\frac{1}{M}\left[\frac{2}{\gamma+1}\left(1+\frac{\gamma-1}{2}M^{2}\right)
\right]^{\frac{\gamma+1}{2(\gamma-1)}}$$
With $ A_E/A_T=3.5/1.0=3.5 $ and γ = 1.4 this has two roots. Solving numerically,
$$\boxed{\;M_{E,\text{sup}}=2.800,\qquad M_{E,\text{sub}}=0.1682\;}$$
The supersonic root governs parts (a) to (c); the subsonic root is the one that matters in
part (d). Note how cleanly the paper is designed: the area ratio was chosen to land on
M = 2.80 to three figures.
Record the isentropic exit state that goes with M = 2.800. From the adiabatic
and isentropic relations,
$$\frac{T_o}{T_E}=1+\frac{\gamma-1}{2}M_E^{2}=1+0.2(2.800)^{2}=2.568,
\qquad \frac{P_o}{P_E}=(2.568)^{3.5}=27.14$$
so, with $ T_o = 300 $ K held constant by the canister,
$$\begin{aligned}
T_E&=\frac{300}{2.568}=116.8\ \text{K}\\
a_E&=\sqrt{\gamma R T_E}=\sqrt{1.4(287)(116.8)}=216.7\ \text{m/s}\\
V_E&=M_E\,a_E=2.800(216.7)=606.6\ \text{m/s}
\end{aligned}$$
Because $ T_o $ never changes and $ M_E $ is set by geometry, the exit temperature and
speed are the same for every choked, fully supersonic condition — only the mass
flow falls as the canister empties.
Part (a) — check the nozzle really runs full at 10.0 MPa, then size the flow.
The design exit pressure is $ P_E = P_o/27.14 = 10.0\times10^{6}/27.14 = 368\ \text{kPa} $,
which exceeds the 100 kPa back pressure. The nozzle is therefore under-expanded: the
adjustment to ambient happens outside in a plume of expansion waves, and the internal solution
is the fully supersonic isentropic one. Evaluating the aid-sheet mass flow at the choked throat,
$$\begin{aligned}
\dot m&=\sqrt{\frac{\gamma}{R}}\,\frac{P_o}{\sqrt{T_o}}\,M\left(1+\frac{\gamma-1}{2}M^{2}
\right)^{-\frac{\gamma+1}{2(\gamma-1)}}A\bigg|_{M=1,\ A=A_T}\\
&=\sqrt{\frac{1.4}{287}}\,\frac{10.0\times10^{6}}{\sqrt{300}}\,(1.2)^{-3}\,(1.0\times10^{-6})
\end{aligned}$$
$$\boxed{\;\dot m=0.02334\ \text{kg/s}=23.3\ \text{g/s},\qquad T_E=116.8\ \text{K},
\qquad V_E=606.6\ \text{m/s}\;}$$
Part (b) — halve the canister pressure and use the linearity of choked flow.
At $ P_o = 5.0 $ MPa the design exit pressure is still
$ 5.0\times10^{6}/27.14=184\ \text{kPa} > 100\ \text{kPa} $, so the nozzle is still
under-expanded and still runs full. Choked mass flow is directly proportional to $ P_o $ at
fixed $ T_o $, so it simply halves while the exit state is untouched:
$$\boxed{\;\dot m=0.01167\ \text{kg/s}=11.7\ \text{g/s},\qquad T_E=116.8\ \text{K},
\qquad V_E=606.6\ \text{m/s}\;}$$
The physical picture is worth stating: for the whole first decade of the blowdown the jet looks
identical — same Mach number, same temperature, same speed — and only its density,
and hence its thrust and mass flow, decay.
Part (c) — place a normal shock exactly on the exit plane. A shock
standing at the exit is fed by the undisturbed supersonic flow, $ M_1 = M_E = 2.800 $, and must
discharge directly into the ambient, so the post-shock pressure equals the back
pressure. The aid-sheet shock relation gives the static pressure jump
$$\frac{P_2}{P_1}=\frac{2\gamma M_1^{2}-(\gamma-1)}{\gamma+1}
=\frac{2(1.4)(2.800)^{2}-0.4}{2.4}=\frac{21.55}{2.4}=8.980$$
Requiring $ P_2 = P_b = 100 $ kPa fixes the pre-shock static pressure at
$ P_1 = 100/8.980 = 11.14 $ kPa, and running that back up the isentropic nozzle to the canister,
$$\boxed{\;P_o=27.14\,P_1=27.14(11.14)=302\ \text{kPa}\;}$$
Scale the mass flow to that canister pressure. The throat is still choked at
302 kPa (the shock is downstream and cannot signal upstream), so mass flow again scales linearly
with $ P_o $:
$$\dot m=0.02334\times\frac{302.2\times10^{3}}{10.0\times10^{6}}
=7.05\times10^{-4}\ \text{kg/s}=0.705\ \text{g/s}$$
a factor of 33 below the opening flow — the canister is nearly spent by the time the shock
reaches the lip.
Get the state immediately after the exit shock. The shock is adiabatic, so
$ T_o = 300 $ K carries across it unchanged; only the Mach number and the entropy change. From
the aid sheet,
$$M_2^{2}=\frac{M_1^{2}+\dfrac{2}{\gamma-1}}{\dfrac{2\gamma}{\gamma-1}M_1^{2}-1}
=\frac{(2.800)^{2}+5}{7(2.800)^{2}-1}=\frac{12.84}{53.88}=0.2383
\quad\Longrightarrow\quad M_2=0.4882$$
$$\begin{aligned}
T_2&=\frac{T_o}{1+0.2M_2^{2}}=\frac{300}{1.0477}=286.4\ \text{K}\\
V_2&=M_2\sqrt{\gamma R T_2}=0.4882\sqrt{1.4(287)(286.4)}=165.6\ \text{m/s}
\end{aligned}$$
$$\boxed{\;M_2=0.488,\qquad T_2=286.4\ \text{K},\qquad V_2=165.6\ \text{m/s}\;}$$
The shock converts nearly all of the 607 m/s kinetic energy back into thermal energy: the jet
leaves at little more than half the speed of the earlier plume, and 170 K hotter.
Part (d) — find the pressure at which the throat unchokes. The throat
stops being sonic once the nozzle can pass the required flow entirely subsonically, which happens
when the exit runs on the subsonic branch of the same area ratio,
$ M_{E,\text{sub}} = 0.1682 $, with the exit pressure equal to the ambient. That requires
$$\frac{P_o}{P_b}=\left(1+0.2M_{E,\text{sub}}^{2}\right)^{3.5}=(1.005659)^{3.5}=1.0199$$
$$\boxed{\;P_o=1.0199\,(100\ \text{kPa})=102.0\ \text{kPa}\;}$$
Above 102.0 kPa the throat remains sonic; below it the entire valve is subsonic, the mass flow
starts to depend on the back pressure as well as on $ P_o $, and the blowdown decays
asymptotically instead of linearly. The choked window therefore covers the canister's life from
10 MPa down to about 1.02 bar — essentially the whole discharge.
Check: quasi-steady, isothermal canister. The question
holds To = 300 K constant, which means the canister is heated (or emptied slowly enough
that the walls supply the heat); an insulated blowdown would instead cool along
To ∝ Po(γ−1)/γ and would reach 102 kPa at a
much lower temperature. The steady nozzle relations are used at each instant, which is
justified because the residence time in a 3.5 mm² valve is microseconds while the canister
empties over minutes: at the opening flow the time constant Vρ/ṁ is of order
V Po/(R To ṁ) ≈ 2.5 × 103 s.
Part
Quantity
Result
design
Exit Mach (supersonic / subsonic root of A/A* = 3.5)
2.800 / 0.1682
Po/PE, To/TE at M = 2.800
27.14, 2.568
(a) Po = 10.0 MPa
Mass flow
0.02334 kg/s (23.3 g/s)
Exit temperature
116.8 K
Exit speed
606.6 m/s (under-expanded, PE = 368 kPa)
(b) Po = 5.0 MPa
Mass flow
0.01167 kg/s (11.7 g/s)
Exit temperature
116.8 K (unchanged)
Exit speed
606.6 m/s (unchanged; PE = 184 kPa)
(c) shock at exit plane
Canister pressure
302 kPa
Mass flow
7.05 × 10−4 kg/s (0.705 g/s)
Temperature after the shock
286.4 K
Speed after the shock
165.6 m/s (M2 = 0.488)
(d) unchoking
Canister pressure below which the throat is subsonic