5-1

Engineering Thermodynamics - Prof. Monica Costea

Chap.5. HIGH SPEED GAS DYNAMICS (HSGD)

5.1. Introduction
HSGD is studying thermal processes in which the fluid Ekin it is
modified or flow processes of gases and vapors through
channels of variable section.
- it is widespread in technique:
• flow of gases through the nozzles of Gas Turbine (GTE)
and Jet Engines
• flow of air or gases through different channels
• steam flow through the nozzles of a Steam Turbine (ST)

The processes that accompany the flow of gas and vapors are
characterized by:
➢ Q exchange
➢ W exchange
➢ the phenomenon of friction due to the viscosity of the
fluid, or roughness of the flow channel.
Classification of flow processes by:
a) parameters variation over time
- steady-state (stationary) flow: fluid parameters and speed
in a section of space are constant over time:
𝜕𝑉̅
= 0, 𝑉̅ = speed, τ = time
𝜕𝜏
- non steady-state flow: speed and fluid parameters in a
section are modified over time:
𝜕𝑉̅
≠0
𝜕𝜏
b) flow regime (how fluid layers move)
- laminar flow: current layers are parallel to each other;
- turbulent flow: the particle speed also has a ⊥ component
on the flow direction.
The laminar or turbulent character is specified by the Reynolds
number (invariant):
 5-2
Engineering Thermodynamics - Prof. Monica Costea
̅ 𝑑𝑒
𝑉
𝑅𝑒 = , with  – kinematic viscosity of fluid
𝜈
de – equivalent diameter
4𝐴
𝑑𝑒 = 𝑃 , A –the cross-sectional area of the flow
P – wet perimeter
4(𝜋𝑑𝑖2 /4)
Ex: for circle: 𝑑𝑒 = = 𝑑𝑖
𝜋𝑑𝑖
4𝑎2
for square: 𝑑𝑒 = =𝑎
4𝑎
The simplifying hypotheses of the study:
i. in a flow section, the speed and gas parameters are
considered constant;
ii. friction forces will be considered concentrated at the
contact surface of the fluid with the wall, and W consumed
to overcome friction is transformed into Q stored in the
mass of the fluid.

5.2. Fundamental equations of gas flow

1. Energy conservation equation (FL)
2. Thermal equation (SL)
3. Continuity equation
5.2.1. Energy conservation equation
First Law of Thermodynamics expression for Open System (with
flow) for 1 kg of gas is:
̅22 −𝑉
𝑉 ̅12
h2 – h1+ 2 + g (z2 – z1) = q1-2 – wsh, 1-2 + qf – wf
As for friction qf = wf and neglecting the potential energy term
epot, it result:
̅22 −𝑉
𝑉 ̅12
h2 – h1 + = q1-2 – wsh, 1-2
2
̅2
𝑉
Notation: h + = h* = stagnation enthalpy = gas enthalpy when
2
its ekin is fully transformed in compression work.
 5-3
Engineering Thermodynamics - Prof. Monica Costea

It results:
h*2 – h*1 = q1-2 – wsh, 1-2 energy equation
For a  small process:
dh* = q – wsh, with dh*= dh + 𝑉̅ d𝑉̅
5.2.2. Thermal equation
𝛿𝑞+𝛿𝑞𝑓
It is known that: 𝑑𝑠 = 𝑇
, with qf – due to friction
 T ds = q + qf thermal equation
5.2.3. Continuity equation
In the case of stationary flow, the mass flow of the gas is
constant along the flow channel:
𝑚̇ [kg/s] = const.
̅
𝐴𝑉
𝑚̇ = 𝑉̇ 𝜌 = 𝐴𝑉̅ 𝜌 = , with v – specific volume
𝑣
̅1
𝐴1 𝑉 ̅2
𝐴2 𝑉
 = =. . ..
𝑣1 𝑣2
̅
𝐴𝑉
If 𝑚̇ = 𝑣 = ct  ln𝑚̇ = ln A + ln 𝑉̅ – ln v
By differentiation
𝑑𝐴 𝑑𝑣 𝑑𝑉̅
𝐴 = − ̅
continuity equation in differential form
𝑣 𝑉

5.3. Flow with small pressure variations

- in this case the variation of T and  (density) of the gas can be
neglected
- it can be treated as incompressible fluid flow, where:
1
𝜌 = ≅ 𝑐𝑡
𝑣
- is governed by Bernoulli's equation:
̅22 −𝑉
𝑉 ̅12
p1 – p2 =  +  g (z2 – z1)  pst + pdyn + pgeom = ct
2
 5-4
Engineering Thermodynamics - Prof. Monica Costea

OBS: if friction is taken into account, Bernoulli's ec. becomes:

̅22 −𝑉
𝑉 ̅12
p1 – p2 =  +  g (z2 – z1) + p [N/m2]
2

with p = pf + pl, where  pf – pressure loss due to friction

pl - local pressure loss
𝑙 ̅2
𝑉
pf = 𝑑 𝜌 [N/m2], with  – loss coefficient
𝑖 2
l – length of section with diameter di
w – average speed of the fluid
64
 = 𝑅𝑒 for laminar regime
0,3164
= 𝑅𝑒 0.25 for turbulent regime (smooth pipes)
-2 𝐾 0,314
= 10 ( 𝑑 ) for turbulent regime (rough pipes)
̅2
𝑉
pl = 𝜌 [N/m2]
2
 - considers the local source of pressure losses,
𝐴1 2
i.e., flow section narrowing (A2 < A1) →=(1 − 𝐴 )
2

5.4. Flow with large pressure variations

During the flow process in this case, T and  of the gas generally
change  the equations giving the variation of the density with
the pressure must be accounted for.
Study hypotheses:
i. adiabatic flow processes are considered: q = 0
ii. friction is neglected
iii. the fluid is a perfect gas, in steady state flow
5.4.1. Speed of the sound
= the propagation speed of the pressure perturbation in an elastic
environment.
 5-5
Engineering Thermodynamics - Prof. Monica Costea

Sound propagates as longitudinal waves in the air. Thus, its

expression is given by:
𝐸
𝑎 = √𝜌 = √𝑣𝐸, with E – elasticity module [N/m2]
v – specific volume [m3/kg]
where according to Hooke's law:
E = /, with  [N/m2] - unitary effort
 [-] – relative elongation
𝜎 𝛥𝑝 𝛥𝑝 𝛥𝑝
But 𝐸 = 𝛥𝑙/𝑙 = − 𝛥𝑙𝐴 = − 𝛥𝑉 = − 𝛥𝑣
𝑙𝐴 𝑉 𝑣
Moving to the limit for a  small variation:
𝑑𝑝 𝑑𝑝
𝐸=− 𝑑𝑣 = −𝑣 𝑑𝑣
𝑣
𝑑𝑝
 𝑎 = √𝑣𝐸 = √−𝑣 2
𝑑𝑣
It is considered that the medium in which the sound propagates
under successive adiabatic compressions and expansions. Thus:
pvk = ct ln p + k ln v = ct
𝑑𝑝 𝑑𝑣 𝑑𝑝 𝑝
 +𝑘 =0  = −𝑘 𝑣
𝑝 𝑣 𝑑𝑣
This results in the speed of sound in perfect gas:
𝑎 = √𝑘𝑝𝑣 and p v = R T 𝒂 = √𝒌𝑹𝑻
where R = RM/M [J/kg K]
! ao = √1.4 ⋅ 287 ⋅ 290 ≅340 [m/s], speed of sound in the air at t0
5.4.2. Flow through nozzles
Nozzle = tube with variable section and small size in which the
gases or vapors expand
- at the entrance: fluid with p and 𝑉̅ 
- at the exit: fluid with p and 𝑉̅ 
 5-6
Engineering Thermodynamics - Prof. Monica Costea

- are component parts characteristic of rotary machines

- given their shortness and high flow rate, the expansion process
in the nozzle is considered adiabatic (q = 0)
- no work is collected (wsh = 0)
During the flow, the decrease in the enthalpy of the fluid
creates an increase in fluid velocity as h → ekin.
5.4.2.1. Differential equations of the nozzles
In the hypotheses of fluid flow through aid, the equations used
are:
(1) dh + 𝑉̅ d𝑉̅ = 0 energy equation, where q, wsh =0
𝑑𝑝 𝑑𝑣
(2) +𝑘 =0 adiabatic equation
𝑝 𝑣
𝑑𝐴 𝑑𝑣 𝑑𝑉̅
(3) = − ̅
continuity equation
𝐴 𝑣 𝑉
(4)-v dp = 𝑉̅ d𝑉̅ impulse equation

Objective: to find a relationship between p, A, 𝑉̅ removing dv/v

𝑑𝑣 1 𝑑𝑝
(2) → 𝑣 = − 𝑘 that is replaced in (3)
𝑝
𝑑𝐴 1 𝑑𝑝 𝑑𝑉̅
 = −𝑘 − ̅
(5)
𝐴 𝑝 𝑉
From (4) the differential of the pressure is expressed as:
̅𝑑𝑉
𝑉 ̅
dp = − (6)
𝑣
𝑑𝐴 ̅ 𝑑𝑉
1𝑉 ̅ 𝑑𝑉̅ 𝑑𝐴 𝑑𝑉̅ 𝑉̅2
Next, (5) + (6)  =𝑘 − ̅
or = (
̅ 𝑘𝑝𝑣
− 1)
𝐴 𝑝𝑣 𝑉 𝐴 𝑉
But k p v = k R T = a2
𝑑𝐴 𝑑𝑉̅ 𝑉̅2
= (
̅ 𝑎2
− 1) first differential equation of nozzles (I)
𝐴 𝑉

The second differential equation of nozzles results similarly:

 5-7
Engineering Thermodynamics - Prof. Monica Costea
𝑣𝑑𝑝
From (4) we have 𝑑𝑉̅ = − 𝑉̅ that is replaced in (5) and
𝑑𝐴 1 𝑑𝑝 𝑣𝑑𝑝
 = −𝑘 + ̅2
𝐴 𝑝 𝑉
𝑑𝐴 1 𝑑𝑝 𝑝𝑣𝑘
Furthermore,  =𝑘 ( 𝑉̅ 2 − 1) ;
𝐴 𝑝
but k p v = k R T = a2
It result:
𝑑𝐴 1 𝑑𝑝 𝑎2
= ( ̅2 − 1) second differential equation of nozzles (II)
𝐴 𝑘 𝑝 𝑉

̅
𝑽
Notation: 𝒂 = Mach number  "sonic wall"
If M < 1 (𝑉̅ < a) subsonic flow regime
M = 1 (𝑉̅ = a) sonic flow regime
M > 1 (𝑉̅ > a) supersonic flow regime
5.4.2.2. Types of nozzles
A. Converging nozzle dA < 0
̅ > 0: the fluid is accelerated - the most frequent application
A1. d𝑽
𝑑𝐴 𝑑𝑉̅ 𝑉̅2
In differential equation (I) = (
̅ 𝑎2
− 1)
𝐴 𝑉
we have: dA < 0 and d𝑉̅ > 0  𝑽 ̅<a
The gas is accelerated to a speed less or = speed of sound
𝑑𝐴 1 𝑑𝑝 𝑎2
In differential equation (II) = ( ̅2 − 1)
𝐴 𝑘 𝑝 𝑉

we have dA < 0 and 𝑉̅ < a  dp < 0

𝑘−1
𝑇2 𝑝̶2
The flow being adiabatic  < 1  dT < 0
𝑘
= (𝑝 )
𝑇1 1
 5-8
Engineering Thermodynamics - Prof. Monica Costea

̅ < 0: the gas is braked, thus the pressure increases.

A2. d𝑽
𝑑𝐴 𝑑𝑉̅ 𝑉̅2
In differential equation (I) = (
̅ 𝑎2
− 1)
𝐴 𝑉
we have: dA < 0 and d𝑉̅ < 0  𝑽 ̅ > a (supersonic)
To brake the gas in a convergent nozzle, its speed in the
̅ 1 ≥ a1
entrance section should be 𝑽
𝑑𝐴 1 𝑑𝑝 𝑎2
In differential equation (II) = ( ̅2 − 1)
𝐴 𝑘 𝑝 𝑉

we have: dA < 0 and 𝑉̅ > a

 dp > 0  dT > 0
It is an application used in ejectors, i.e., in the refrigerating
machine for air conditioning (pressure increase is low).

Converging nozzle: dA < 0

A1: d𝑉̅ > 0 A2: d𝑉̅ < 0

𝑉̅ 2 𝑉̅ 1
𝑉̅ 1 𝑉̅ 2
p1 p2
T1 p2 p1 T2
T2 T1
 5-9
Engineering Thermodynamics - Prof. Monica Costea

B. Diverging nozzle dA > 0

̅ > 0: corresponds the acceleration of the fluid that expands
B1. d 𝑽
𝑑𝐴 𝑑𝑉̅ 𝑉̅2
In differential equation (I): = (
̅ 𝑎2
− 1)
𝐴 𝑉
we have dA > 0 and d𝑉̅ > 0  𝑽 ̅ > a (supersonic)
To accelerate the gas in a diverging nozzle, it results that its
speed should be 𝑽̅ 1 ≥ a1 at the entrance section
The gas is accelerated until its complete expansion (p2 = pamb),
reaching supersonic speeds at the exit section.
𝑑𝐴 1 𝑑𝑝 𝑎2
In differential equation (II): = ( ̅2 − 1)
𝐴 𝑘 𝑝 𝑉

dA > 0 and 𝑉̅ > a leads to  dp < 0, respectively dT < 0

̅ < 0: the gas is braked, the pressure increases
B2. d𝑽
𝑑𝐴 𝑑𝑉̅ 𝑉̅2
In differential equation (I) = (
̅ 𝑎2
− 1):
𝐴 𝑉
we have: dA > 0 and d𝑉̅ < 0  𝑽
̅<a
To brake the gas, its speed in the entrance section is 𝑉̅ 1 < a1
𝑑𝐴 1 𝑑𝑝 𝑎2
In differential equation (II): = ( ̅2 − 1)
𝐴 𝑘 𝑝 𝑉

dA > 0 and 𝑉̅ < a  dp > 0  dT > 0

Diverging nozzle: dA > 0
B1: d𝑉̅ > 0 B2: d𝑉̅ < 0

𝑉̅ 2 𝑉̅ 1
𝑉̅ 1 𝑉̅ 2
p1 p2
T1 p2 p1 T2
T2 T1