Mathematical Modelling and Design of an Advanced

Once-Through Heat Recovery Steam Generator

Marie-Noëlle Dumont 1, Georges Heyen
LASSC, University of Liège, Sart Tilman B6A, B-4000 Liège (Belgium)

Abstract
The once-through heat recovery steam generator design is ideally matched to very high

temperature and pressure, well into the supercritical range. Moreover this type of boiler is

structurally simpler than a conventional one, since no drum is required.

In a conventional design, each tube plays a well-defined role: water preheating, vaporisation,

superheating. Empirical equations are available to predict the average heat transfer coefficient

for each region. For once-through applications, this is no more the case and mathematical

models have to be adapted to account for the disappearance of the conventional economiser,

boiler and superheater. General equations have to be used for each tube of the boiler, and the

actual heat transfer condition in each tube has to be identified. The mathematical complexity

as well as the number of equations is increased. A thermodynamic model has been selected

and implemented to suit very high pressure (up to 240 bar), sub- and supercritical steam

properties. Model use is illustrated by two case studies : a 180 bar once-through boiler (OTB)

and a conventional boiler superheater and reheater.

Keywords :Once-through boiler; heat recovery steam generator (HRSG); water flow pattern

1
Corresponding authors. Tel:+32 4 366 35 23 Fax: +32 4 366 35 25 ; E-mail address: [email protected]
1. Introduction

Nowadays combined cycle (CC) power plants become a good choice to produce energy,

because of their high efficiency and the use of low carbon content fuels (e.g. natural gas) that

reduces the greenhouse gases production. CC plants couple a Brayton cycle with a Rankine

cycle. The hot exhaust of the gas turbine (Brayton cycle) delivers energy to produce high-

pressure steam for the Rankine cycle. The equipment where the steam production takes place

is named the heat recovery steam generator (HRSG).

High efficiency in CC (up to 58%) can be achieved for two main reasons:

1 Improvements in the gas turbine technology (i.e. higher inlet temperature);

2 Improvement in the HRSG design

We focus here on the second point. The introduction of several pressure levels with reheat in

the steam cycle in the HRSG allows recovering more energy from the exhaust gas (usually

available between 600°C and 700°C). Exergy losses decrease, due to a better matching of the

gas-cooling curve with the water/steam curve in the heat exchange diagram (Dechamps,

1998). Going to supercritical pressure with the OTB technology is another way to better

match those curves and thus improve the CC efficiency.

New improvements are announced in near future to reach overall cycle efficiency as high as

60%.

In the present work we propose a mathematical model for the simulation and design of the

once-through boiler. The modelling approach used for the simulation of a conventional boiler

has to be revised, since the heat transfer regime in each tube can not be fixed by the

equipment design. General equations have to be used for each tube of the boiler. Moreover

there is a more significant evolution of the water/steam flow pattern type due to the complete

water vaporization inside the tubes (in a conventional boiler, the circulation flow is adjusted
to reach a vapour fraction between 20% and 40% in the tubes and the vapour is separated in

the drum).

Changes of flow pattern induce a modification in the evaluation of the internal heat transfer

coefficient as well as in the pressure drop formulation. The right equation has to be selected

dynamically according to the flow conditions prevailing in the tube.

The uniform distribution of water among parallel tubes of the same geometry subjected to

equal heating is not ensured from the outset but depends on the pressure drop in the tubes.

The disappearance of the drum introduces a different understanding of the boiler’s behaviour.

Effects of the various two-phase flow patterns have to be mathematically controlled.

2.Thermodynamic model

To estimate water and steam properties, we make use of "IAPWS Industrial Formulation for

the Thermodynamic Properties of Water and Steam" (Wagner et al, 1998). It replaces the

previous industrial standard IFC-67. This formulation provides a very accurate representation

of the thermodynamic properties of water and steam over a wide range of temperature and

pressure with a formulation that is designed for fast computation.

The IAPWS Industrial Formulation 1997 consists of a set of equations for different regions

that cover the following range of validity:

0°C < T < 800°C p<1000 bar

800°C < T <2000°C p<100 bar

Figure 1 shows the 5 regions into which the entire range of validity of IAPWS-IF97 is

divided.
3. Mathematical Model

The model described hereafter should be applied to horizontal tube bundles (boilers with

vertical gas path, figure 6 shows a typical tube layout). It has been developed for once through

boilers but could also be used with conventional boiler. The complete set of equations

developed here after has to be applied to each tube row or part of tube row of the complete

tube bundle (for a tube bundle with 50 rows of tubes, the set has to be applied at least 50

times).

3.1. Heat transfer

3.1.1. Water side

Heat transfer equations must be formulated for steady state, forced flow through tubes.

Mathematical models for conventional boilers are usually based on empirical equations

corresponding to each region of the boiler: the economizer, the boiler and the superheater.

Those three parts of boiler are clearly separated thus it is not difficult to choose the

appropriate equation. In a once-through boiler this separation is not so clear. We have first to

estimate the flow pattern in the tubes, and on this basis to select the appropriate heat transfer

equation. “Liquid single phase” and “vapour single phase” conditions are easily identified

from temperature and pressure data. According to Gnielinski (1993) the equation 1 applies for

turbulent and hydrodynamically developed flow.

(ξ / 8) ( Rel −1000) Pr α *d 1
Nu = = ; ξ= (1)
1 + 12,7 (ξ /8) Pr 2 / 3 −1 λ
(1,82log10 Re−1,64)

During vaporization different flow patterns can be observed, for which the rate of heat

transfer also differs. In stratified-wavy flow pattern incomplete wetting has an effect on the

heat transfer coefficient. A reduction appears for this type of flow pattern. Computing
conditions where a change in flow pattern occurs is useful. Steiner (1993) clearly exposed a

method to establish a flow pattern map in horizontal tube for given pressure and flow

conditions. This method has been used in this study. The different flow pattern in the

vaporisation zone of the OTB are given in figure 2. The heat transfer coefficient is estimated

from numerous data. It is a combination of convective heat transfer coefficient and nucleate

boiling heat transfer coefficient.

α ( z ) = 3 α ( z )3 + α ( z )3 (2)
conv B

−2.2
−2 0.37
α (z)conv 0.01 ρliq
= (1− x) +1.2x0.4 (1− x) (3)
α ρvap
lo

−2
0.67
αgo 0.7 ρliq
+ x0.01 1+8(1− x)
α ρvap
lo

n( p* )
q 1.6 p*6.5
α ( z ) B = ψ *18418* 2.692 p*0.43 +
15*104 1 − p*4.4
(4)
0.5 0.133 0.25
0.01 Ra G * 0.1 q
. 1− p
d 10−6 100 qcr , PB

0.66. p*
n = κ (0.8 − 0.13*10
with qcr , PB = 9.19 *106 p* 0.4 1 − p* ( )
p p
p* = =
pc 220.64

The correction coefficients ψ and κ are functions of the heat conduction (λws) of the tube

wall. They have to be applied when (λws) < 0.7 W/K, which is the case for HRSG (Table 1)

For bubble flow, ψ is set to 1.

αLO is the heat transfer coefficient with total mass velocity in the form of the liquid.

αGO is the heat transfer coefficient with total mass velocity in the form of the vapour.

Evolution of the internal coefficients in a typical OTB is presented in figure 3.

32.1.2. Fumes side

The same set of equations can be used used for a conventional heat recovery boiler and a once

trough heat recovery boiler. The main contribution to the heat transfer coefficient is due to

convection, since the fumes temperature is rather low and the tube spacing is short. Radiative

heat transfer plays a secondary role. The effect of the turbulence has been introduced to

reduce the heat transfer coefficient in the first few rows of the tube bundle.

The main difficulty to evaluate the heat transfer coefficient for the fume side comes from the

fins that enhance the heat transfer, but could also produce other sources of resistance in the

heat transfer, such as fouling on the surface of fins or inadequate contact between the core

tube and the fin base.

Several methods have been implemented in the model to evaluate the heat transfer coefficient

on the gas side. Equation 6 is a general equation, which evaluates the Nusselt number in cross

flow over pipes.

n
A
Nu = C Rem Pr l (6)
d d A
b

Values for parameters “C”, “m”, “n” and “l” are given in table 2.

Fins efficiency is estimated from following equations:

ea da
Hr = l f 1+ 1 + 0.35ln (7)
2l f d

2α f
X = Hr (8)
ea λa

tanh( X )
ηf = (9)
X
Finally an apparent heat transfer coefficient is computed from equation 10:

A A
po fo
α =α * +η (10)
app f A f A

Some tube manufacturers supply specific correlations whose coefficients have been tuned to

match extensive data for their specific tube design. For instance, the tube manufacturer

ESCOA recommends (ESCOA, 1979):

α f = j * G * c p * Pr −2 / 3 (11)

d a 0,5 Tb 0,25
j = C1 * C3 * C5 *( ) *( ) (12)
d Ts

For solid fins and staggered arrangement for tubes :

C1 = 0, 25Re−0,35 (13)

( −0,25l f / s f )
C3 = 0,35 + 0,65e (14)

2
C5 = 0,7 + 0,7 − 0,8e( −0,15 Nr ) * e( −lp / tp ) (15)

3.1.3. Overall heat transfer coefficient

Finally the overall heat transfer coefficient is obtained from equation 16.

1 1 e 1
= + + (16)
α α A A
app λ* w α * i
A i A

The overall heat transfer for each tube is computed with:

Q = α * A * ∆T (17)
sl

We call ∆Tsl “semi logarithmic temperature difference” (equation 18). It is the best

compromise between pure logarithmic temperature difference that has no sense here (only one

tube) and pure arithmetic temperature difference that does not allow following the evolution

of water properties along the tube.

∆T =
(Tw2 − Tw1 ) (18)
sl
T −T
mf w1
ln
T −T
mf w2

T +T
f1 f2
T = (19)
mf 2

The heat exchange diagram of a typical OTB is presented in figure 4.

3.2. Pressure drop

3.2.1. Water side

Water flows in several parallel channels, submitted to slightly different heating patterns, thus

the flows distribution will be influenced by pressure drops.

64
2 f = (laminar)
f ⋅ ρ ⋅V l Re
∆P = with (20)
2g d 0.3164
i f = (turbulent)
4 Re

The coefficient f depends on the Reynolds number for flow within the tube. In laminar flow,

the Hagen-Poiseuille law can be applied. In turbulent flow the Blasius equation is used. The

main difficulty is the evaluation of water pressure drop during transition boiling. The pressure

drop consists of three components: friction (∆Pf), acceleration (∆Pm) and static pressure (∆Pg).

In once-through horizontal tubes boiler ∆Pg=0. The Lockard-Martinelli formulation is used to

estimate the friction term.

∆P ∆P
= ⋅ Φ2 (21)
L L ftt
2 phases liquid

20 1
Φ2 = 1 + + (22)
ftt X X2
 ∆P 0,5 0,125
L liquid 1− x
0,875
ρ go ηlo
X= = (23)
∆P x ρlo η go
L vapor

The acceleration term is defined with equation 24 where ε is the volume fraction of vapour

(void fraction defined by equation 25).

x
2 x2 (1 − x )2 2 (24)
∆P = G * +
vap (
m ε *ρ 1− ε )* ρ
liq x
1

−1
1.18 (1 − x ) gσ ( ρl − ρ g )
0.25
x x 1− x
ε=
ρg
(1 + 0.12 (1 − x ) ) ρg
+
ρl
+
Gtot ρl0.5
(25)

It is recommended to discretize each tube in several short sections in order to obtain more

accurate results. Figure 5 shows local pressure drop evolution in a tube, as calculated by this

model. The main pressure drop is observed where vaporisation takes place.

3.2.2. Fumes side

The pressure drop in a tube bundle is given by equation 26. In this case the number of rows

(NR) plays an important role in the pressure drop evaluation. For solid fins and staggered

arrangement for tubes, the ESCOA correlation has been selected (ESCOA, 1979):

( f + a) ⋅ G 2 N r
∆P = (26)
ρb

1+ β 2 1 1
a= ρb − (27)
4 Nr ρ 2 ρ1

An
β= (28)
Ad

0.5
da
f = C2 * C4 * C6 * (29)
d

C2 = 0, 07 + 8 Re −0,45 (30)
 ( −0,7(l f / s f )0,2 )
tp
C4 = 0,11 0, 05
d (31)

C6 = 1,1 + 1,8 − 2,1e( −0,15 Nr ) * e( −2lp / tp )

2

− 0, 7 − 0,8e( −0,15 Nr ) * e( −0,6lp / tp )

2

(32)

4. Examples

The following examples are based on CMI boilers design. CMI Utility Boilers is a company

active in the design, construction, erection and commissioning of heat recovery boilers

associated with high capacity gas turbines used in combined cycle power plant.

4.1.Once through boiler

The once through HRSG design features are:

Vertical gas path.

Economizer and vaporizer are combined (no intermediate header).

Water is flowing down from top to bottom in countercurrent arrangement.

Ready to work with the latest generation of gas turbines.

Fuel can be natural gas, distillate, heavy oil or crude oil.

Results have been obtained for an OTB of pilot plant size presented in figure 6. Main design

parameters of the steam generators is given in tables 3 and 4.

The simulation model has been implemented in VALI software (BELSIM, 2002). Each tube

or tube section can be represented by a separate simulation object. The graphical user

interface allows easy modification of the tube connections and the modeling of multiple pass

bundles (figure 8).The simulation of the OTB described here is performed by connecting at

least 42 modules, one for each tube row. Since VALI implements a numerical procedure to

solve large sets of non-linear equations, all model equations are solved simultaneously.

Convergence is smooth and is achieved in a few iterations (figure 7).

4.2. Conventional boiler and reheater

Other results were obtained for a superheater and a reheater of a conventional boiler. Boiler

description is given in table 5.

This allowed comparison of the detailed modelling presented here, with the more global

design approach that has been applied previously.

Although our design tool has been developed to model once through boilers, there is no

limitation and it can also be applied to model conventional boiler. Figure 3 compares the

predicted evolution of heat transfer coefficient in each tube, with average values predicted

with the design procedures applied in the past.

Economiser and vaporiser are well described with traditional empirical equations however the

extra modelling work required by the new method provides interesting information for

superheater and reheater.

The example shown here allowed to understand some failures in the tubes of a super heater.

Figure 9 clearly shows that the fumes temperatures (shown in italic) does not evolve

homogeneously in the boiler. Gas temperature may vary by more than 30° across the flue

channel. Water temperature profiles also differ for parallel tubes thus pressure drops patterns

also differ. In order to balance the pressure drop in parallel tubes, the water flow rate has to

differ. The heat transfer might become degraded in some tubes, which results in overheating

and increases the risk of failure.

The module is now used during the design to analyse the water temperature profile when

selecting between alternative tube patterns.

5. Conclusions and Future Work

The mathematical model of the once-through boiler has been used to better understand the

behaviour of the boiler. Future mathematical developments have still to be done to improve

the OTB design. The criteria for flow stability have to be reviewed since it is certainly
different in an OTB design or in an assisted circulation boiler design. Automatic generation of

alternative bundle layouts in the graphical user interface is also foreseen.

6. Acknowledgements

CMI Utility boilers (Belgium) has financially supported this work, and provided design data

for the examples.

7. Nomenclature

A total area of outer surface (m²)

Ab bare tube outside surface area

Afo fin outside surface area (m²)

Ai inside surface area (m²)

Apo free area of tube outer surface

Aw mean area of homogeneous tube wall

cp specific heat capacity at constant pressure (J/kg/K)

di tube internal diameter (m)

d tube external diameter (m)

da fin diameter (m)

ea fin thickness (m)

∆P pressure drop

f pressure drop coefficient

g acceleration due to gravity (m²/s)

G mass flux (kg/m2/s)

H enthalpy flow (kW)

lf fin height = (da-d)/2 (m)

lp longitudinal tube pitch (m)

tp transverse tube pitch (m)

Nu Nusselt number Nul = α ⋅ l

λ

Nr number of rows in the direction of flow

P pressure (bar)

c p ⋅η
Pr Prandl number Pr =
λ

Q exchanged heat (kW)

q heat flux (W/m²)

G*d
Re Reynolds number Re d =
η

Ra Arithmetic mean roughness height (m)

sf fin spacing (m)

T temperature (K)

Ts average fin surface temperature (K)

Tb average outside fluid temperature (K)

V fluid velocity (m/s)

x vapour mass fraction

xi component flow rate (kg/s)

α heat transfer coefficient (kW/m²/K)

αf external (fumes side) heat transfer coefficient (kW/m²/K)

αi internal (water side) heat transfer coefficient (kW/m²/K)

α(z) local heat transfer coefficient

ε volume fraction of vapor

λ thermal conductivity (W/m/K)

λa fin thermal conductivity (W/m/K)

ρ density (kg/m3)

η dynamic viscosity (Pa.s) or (kg/m/s)

ηf fin efficiency

8. References

BELSIM (2002) VALI III users guide, Belsim s.a., Saint-Georges-sur-Meuse, Belgium

Dechamps, P.J. (1998), Advanced combined cycle alternatives with the latest gas turbines,

ASME J. Engrg. Gas Turbines Power 120, 350–35

ESCOA (1979), Fin Tube Manual, ESCOA corp.

Gnielinski, V. (1993), VDI heat atlas, GA,GB, VDI-Verlag, Düsseldorf, Germany

Rizhu, Li, Huaiming, Ju (2002), Structural design and two-phase flow stability test for the

steam generator, Nuclear Engineering and Design 218, 179-187

Steiner, D. (1993) VDI heat atlas, VDI-Verlag, HBB, Düsseldorf, Germany

Wagner, W., Kruse (1998), Properties of Water and Steam / IAPWS-IF97, Springer-Verlag,

Berlin, Germany
9. Figures

Figure 1 : IAPWS-IF97 range of validity

Figure 2 : Flow pattern in the boiling zone for horizontal flow (VDI,1993).

Dots show flow evolution in a typical OTB

Mist

Annular

Plug or Slug
Wavy

Stratified

0.01 0.1 1 10 100

X-Martinelli parameter
Figure 3 : Internal heat transfer coefficient evolution in the once through boiler compared to average

coefficients of a conventional steam generator (large dots)

25000
Internal heat transfer

20000
kcal/h/m²/K
coefficient

15000
VAPO
10000
ECO SUP
5000

Once-through HRSG Traditional HRSG

Figure 4 : Fumes and water temperature evolution through out the boiler with the “OTB” mathematical

formulation

450 700

400
600
350

TEMPERATURE (C)
500
LOAD (kW)

300

250 400

200 300
150
200
100
100
50

0 0
Subcooled Boiling Superheated
region region region
LOAD FUME IN WATER OUT
 Figure 5 : local pressure drop evolution in a continuous flow path from inlet water to outlet superheated

steam

160 100

90
140
Pressure drop (mmH2O)

80

Vapour fraction (%)

120
70

100
60

80 50

40
60

30
40
20

20
10

0 0

Subcooled Boiling Superheated

region region region
Pressure drop vapour fraction
Figure 6 : Details of the pressure parts assembly of the once-through boiler

FUMES

Tube Material

ECOVAPO 15Mo3

10CrMo910

SUPERHEATER SA335T91

DETAILS OF PRESSURE PARTS ASSEMBLY

Figure 7: Convergence evolution for the OTB example
Figure 8 : Bundle layout automatically generated with the appropriate connections in the graphical user

interface of Vali software

Figure 9 : Results obtained for a superheater and a reheater of a conventional boiler.

185986 kg/h A B C
81.740 bar
296.5 C 473.4 478.5 471.3
INLET
338.5 380.7 410.4

349.2 398.1 428.3

499.7 473.0 441.7 210065 kg/h
515.3 490.2 459.4 22.080 bar
SHP1 374.7 C

567.9 556.5 529.9 INLET

538.4 506.6 453.5

547.3 517.4 462.9

556.1 528.7 473.2

565.1 540.8 484.8
SMP2

575.2 590.2 601.1 52473.8 kg/h

581.7 595.1 604.7 52201.8 kg/h

570.7 591.3 604.0 52832.7 kg/h

578.8 597.5 608.6 52556.7 kg/h

614.1 618.6 620.7

531.7 556.3 574.7

OUT LET
547.4 570.9 587.5
SHP2

93559.0 kg/h 615.7 605.4 592.0

0.0 92427.0 kg/h 624.0 615.3 603.4 604.6 C
21.655 bar

OUTLET

619.8 C 649.0 649.0 649.0

79.206 bar
10.Tables

Table 1: correction coefficients for nucleate boiling heat transfer coefficient

min value max value value for OTB example

κ 0.72 1 0.8

ψ (annular flow) 0.74 1 0.86

ψ (slug flow) 0.65 1 0.79

ψ (stratified and wavy flow) 0.45 0.86 0.61

Table 2: VDI coefficients for Nusselt correlation (fumes side)

VDI C m n l

In-line arrangement 0.22 0.6 -0.15 0.333

Staggered arrangement 0.38 0.6 -0.15 0.333

Table 3: OTB description

ECOVAPO SUPERHEATER

number of rows 36 6

tubes in parallel 2 2

tubes / row 13 13

tube diameter (d in m) 0.025 0.0269

tube thickness (m) 0.0029 0.0042

length (m) 6 6

fins /m 200 200

fin diameter (da in m) 0.049 0.049

transverse pitch (tp in m) 0.083 0.083

longitudinal pitch (lp in m) 0.073 0.073

Table 4: OTB operating conditions

Flow rate Inlet temperature Outlet temperature

(t/h) (C) (C)

WATER 10.25 44 500

FUMES 72.5 592 197

Table 5: conventional boiler description

SUPERHEATER REHEATER

number of rows 8 8

tubes in parallel 2 4

tubes / row 104 104

tube diameter (d in m) 0.038 0.0445

tube thickness (m) 0.0042 0.0033

length (m) 6.8 6.8

fins /m 290 285

fin diameter (da in m) 0.068 0.068

transverse pitch (tp in m) 0.091 0.091

longitudinal pitch (lp in m) 0.079 0.079

Tin water (°C) 296.5 374.7

Pin water (bar) 81.74 22.08

Tout water (°C) 619.8 604.6

Flow rate water (t/h) 185.986 210.065