J. of Thermal Science Vol.12, No.

Condenser Optimization in Steam Power Plant

$iikrii Bekdemir Recep Oztiirk Zehra Yumurtac

Ylldlz Technical University, Mechanical Engineering Department, Istanbul, Turkey
E-mail: [email protected]

In this paper the effects of the condenser design parameters (such as turbine inlet condition, turbine power and
condenser pressure) on heat transfer area, cooling water flow-rate, condenser cost and specific energy generation
cost are studied for surface type condenser. The results are given in the text and also shown as diagrams.

Keywords: condensers, steam power plants, energy cost.

Introduction reach this aim. First one is to increase the turbine inlet
temperature and pressure, the other one is lowering the
The main purpose in steam power plant is to turbine exit pressure, in other words to use a condenser.
generate maximum power at high efficiency. The most If the exit pressure is dropped from 0.1 × 105 Pa to 0.05
effective way for reaching this aim is to use a condenser. × 105 Pa the efficiency and power rise 7%
Condensers increase the enthalpy drops and turbine work approximately when the turbine inlet conditions remain
by lowering the turbine outlet pressure. The lower the constant.
pressure, the higher the efficiency and power are.
Condensers can be classified in two groups: surface and , i
1
direct contact. At present, while the first type is mostly
used in power plants, the other is only used in special
eases. In this paper we investigated the surface type 4

condenser.
The principal cycle of the steam engines is Rankine
Cycle. A schematic of a simple steam power plant with
condenser and its h-s diagram are shown in Fig.1.
Rankine Cycle is an ideal cycle where all the process
take place as reversible. Actual cycle differs from ideal
cycle because of some irreversibilities. Thermal
efficiency of the actual cycle according to Fig. 1 is:
1 /
Oth = [(hi - h2) - (h4- h3)] / ( h i - h4) (1)

and the turbine power: 4 '4

Pe = ms (hi -h2) r/m (kW) (2)

where h is the enthalpy in (kJ/kg), ms the steam mass

flow rate in (kg/s) and r/m the mechanical efficiency of
the turbine.
As seen in equations 1 and 2, to maximize the
thermal efficiency and power, the enthalpy drop in the Fig.1 A schematic of a simple steam power
turbine must be increased. There are mainly two ways to plant and it is h-s diagram
Received 2002
$iikrii Bekdemir et al. Condenser Optimization in Steam Power Plant 177

Condenser Calculations ATi is difference between saturation steam temperature

and cooling water inlet temperature (AT~=TE-Ti), ATe is
From Fig.1 and Fig.2, the heat from the exit steam is: difference between saturation steam temperature and
cooling water exit temperature ATe = (T2-Ta).
as = ms (h2-h3) (kJ/kg) (3)
From Eq.6 condenser heat transfer area is:
and the heat that cooling water gets:
A1 = Q / ( k • AT,,) (m 2) (8)
aw = mw Cpw (TcT~) (kJ/kg) (4)
On the other hand some air enters condenser through
From the equalization of Qs and Qw we get: different ways. It decreases the k value, increases the
condenser pressure and causes the drops in efficiency
mw = [ms (h2 -h3)] / [Cpw(Ta-Ti)] (5)
and power. For this reason the air is removed by using a
where mw denotes the cooling water flow rate in (kg/s), vacuum pump or ejector. This air also cooled by cooling
Cpw the specific heat of the water in (kJ/kgK) and (T,-Ti) water so that it needs additional condenser area. This
the temperature difference of the cooling water. (Ta-Ti) area is added into AI. The cooling area of air can be
can be taken as (6 + 8)°C. computed from the equation given below.
A2=(maCpa/kl){ln[(T2-Ti)/(Tu-Ti)]} (m 2) (9)
ms, 7"2
where ma refers to air mass entering the condenser per
| hour, k the heat transfer coefficient from air to water and
Tu the air inlet temperature.
According to Hoeffertll, ma can be taken as:

mw,T1 m | ml row, Ta ma = 0.0314 (ms/102) °'9 (kg/s) (10)

As a result the total area is A = A1 + A2.

For a safety cooling, this value of A is increased
I| I from 10 percent to 25 percent, so that the final area At is
,Ip between 1.10 A and 1.25 A.
ms ~T2 If we take the length of the cooling water tubes as L,
inner diameter of the tube as di, outer diameter as da and
tube number as n, At can be written as:
Fig.2 A schematic of a surface type condenser At= ~ " da • L •n (m E) (11)

The heat transferring in condenser is: Cooling water velocity in the tubes:

O =A~ • k . ATm (kW) (6) V=mw/[(p. d 2 • n)(~4)] (m/s) (12)

where k is the heat transfer coefficient in (W/mZK), A1 where p is the density of water in (kg/m3).
the heat transfer area of the condenser in (m 1) and ATm Cooling water heat transfer coefficient is:
the log mean temperature difference. From Fig.3: aw=(a, vn/d~')(l+b. 7",,) (W/m2K) (13)
ATm = [(ATi-ATe)] / [ln(ATi/ATe)] (7)
where T,, is the mean temperature of the cooling water
and a, b, m, n are the constant factors. From Hutte, we
can take a=1755, b=0.015, n=0.87 and m=0.13. If we
denote the dirtiness factor as tp then[21:
T
OCw= tp.OCw (W/m2K) (14)
~0value can be taken between 0.60 and 0.70.
The overall heat transfer coefficient k can be drawn
T~
from equation below[3]:

llk=(1/a*)+(SIJ.)+(1/fls) (15)
where ~ is the tube thickness in m, ~ the thermal
A.O. conductivity of the tube material in (w/meK) and as the
heat transfer coefficient of the steam (W/meK) as value
Fig.3 Heat transfer diagram of a condenser is taken between 11000~ 16000 (W/mZK).
178 Journal of Thermal Science, Vol. 12, No.2, 2003

In condenser calculation, the heat transfer coefficient 14000 [ . . . . .

k is assumed firstly, calculation are done according to k
12000
this assumption and finally k is controlled, k value varies
~"Ii"~\ ee = 1O0 M W
from 2000 (W/m2K) to 3500 (W/mZK) in turbine
condensers. 1°°°° I ~ " P 5MW
The power of the condenser cooling water pump is: 8000 \ '., ~ . \

Pp = (0.1 • mw • AP) / rlp (kW) (16)

a: •- "',>,".,~7"-.
where r b is the pump efficiency and AP the total 4ooo :.-:.::>~--.-5%:>,-.
pressure loss.

Condenser Cost 0 i i i i ,
o 04 o 0.5 o.oG o o7" o.oo o o9

Condenser cost equation can be written as below: Condenserpressure ( X 105 Pa)

Ccon= A t . g, • 13 ($) (17) Fig.4 The effect of the condenser pressure

on the heat transfer area
where C~o. is the condenser cost, A~ the total condenser
heat transfer area (m2), gc the specific condenser cost
($/kg) and fl the relation between condenser heat transfer 4000 t
area and condenser mass. The share of condenser cost in
energy generation cost,
gcon = Ceon • / E (S/kWh) (18) 83.~ 3000 I
r
Energy Generation Cost 2500 [ .~d=7/8" "

~ 2000
Electrical energy generation cost consists of three :z I/
items[4]:
1500 / , , , ,
0., 1.0 112 ,.4 1; 11, 20 2'.2 ~.4
C, = Ck + Cy+C,, ($) (19)
Cooling water velocity (m/s)
where C~ is the total cost, Ct the capital cost, Cf the fuel
cost and C,, the operation and maintenance cost per
Fig.5 The effect of the cooling water velocity on
annum. If we divide the total cost by annual electrical
heat transfer coefficient for different tube
energy generation we obtain specific energy cost:
ge = Ct / E = ( Ck+Cf'FCm)/E ($/k'Wh) (20)
180
Annual electrical energy generation can be written as:
E = 8760 • P~ • nt (kWh) (21) ,~ 16o
P.=IO0 M W
where P, is the plant power and nk is the capacity factor. P~=5 M W
140

Conclusion 120

In this paper, the effect of condenser pressure and

100
plant p o w e r on condenser cost and specific energy
k
generation cost are studied. For this aim, we used a
8o
c o m p u t e r p r o g r a m m e running on Matlab R12 base.
Results are given as diagram in Figs.4~6. Fig.4 shows r~ 60 i i i
0 03 0 04 0 05 0 06 0 07 0 08 0 09 l]I
the effect of the condenser pressure on heat transfer area.
Condenserpressure (X 105 Pa)
As seen from that, heat transfer area decreases with
condenser pressure. In Fig.5 the effect of the cooling
Fig.6 The effect of the condenser pressure
water velocity on heat transfer coefficient for different
on the unit energy cost
tube diameter is shown. As can be seen from Fig.5 heat
transfer coefficient increases with the cooling water
velocity. The effect of the condenser pressure on unit (continued on page 192)
192 Journal of Thermal Science, Vol. 12, No.2, 2003

0.9

08
1 1[
07

06

05

04
0 2 4 0 2 4
03 riD riD
02 1 1
0.1

0
0 5 10 15 20 25 o05~
y/D
Fig.5 Concentration of the reaction product in terms 0
of vertical distance from the outlet of the nozzle
0
0 2
' '
4 0 2' 4'
ffD riD
The mean concentration function has two local maximum Fig.6 Numerical results of the mean relative concentration
in approximately 3D and 7D distance from the outlet. of NH4C1aerosol versus distance from the
symmetry axis in different cross sections
Numerical Approach The distance from the outlet of HCI:
(a) lD (b) 2D (c) 5D (d) 10D
The numerical modelling of the reaction was
performed by means of Fluent code. The finite rate
chemistry approach was used together with LES model Acknowledgements
for turbulence. The calculation were performed for This work was supported by Komitet Badan Nau-
boundary conditions from the experiment. The space size kowych, under Contract No. 1443/T 1012000/18.
of the numerical grid node in the vicinity of the nozzle
outlet was 0.05 mm. The profiles of mean relevant
concentration of the reaction product is exhibited in
References
Fig.6. The numerical calculations can give only averaged
results without fluctuations, because the physical model [1] Countess, R J, Heicklen, J. Kinetics of Reaction of
is based on the averaged balance equations. Ammonia with Hydrogen Chloride and the Growth of
In the light of the experimental observation the rate Particulate Ammonium Chloride. The Journal of Physical
constant k proposed by Countess and Heicklen t2] seems Chemistry, 1973, 77(4): 444--447
to be too low, because the numerical concentration in [2] Seinfeld, J H. Atmospheric Chemistry and Physics of Air
terms of vertical space coordinate is monotonically Pollution., Willey, NY, 1986
growing up instead of reaching maximum in a distance [3] Krzywicki, P. Jasnosc Code (in Polish). IFFM Report,
of 5 ~ 8 D from the outlet. Gdansk, 2002

(con~nuedfrompage 178)
energy cost is given for different power in Fig.6. University, 1976. III
ccording to that, specific energy cost decreases with [2] Schr6der, K. GrosseDampfkraftwerke. Springer Verlag,
condenser pressure, but changing is fairly slight. Fig.4 1970
and Fig.6 show that especially below 0.05 × 105 Pa [3] Coulson, Richardson. Chemical Engineering Fluid Flow.
condenser pressure, while the heat transfer area rises Heat Transfer and Mass Transfer, 1996, 5
sharply, specific energy generation cost increases [4] Aybers, N, Sahin, B. Energy Cost. Turkey: Yddlz
slightly. On the other hand, since the analytical method Technical University, 1995
is used with this study, if one changes the parameter [5] Geankoplis, Christie, J. Transport Processes and Unit
another solution may be obtained. Operation. Prentice Hall, 1993
[6] Guyer, E C. Handbook of Applied Thermal Design. Mc-
References Graw Hill Book, 1989
[7] Perry, P H, Green, D. Chemical Engineering Handbook.
[1] Eyice, S. Steam Turbines. Turkey: Yddiz Technical Mc-Graw Hill Book, 1997