Vous êtes sur la page 1sur 11

Energy Conversion and Management 48 (2007) 2150–2160


Effect of air ingress on the energy performance of coal fired

thermal power plants
M. Siddhartha Bhatt *

Central Power Research Institute, Energy Conservation and Development Division, P.O. Box 8066, Sir CV Raman Road, Bangalore 560 080, India

Received 21 June 2006; accepted 24 December 2006

Available online 2 March 2007


Ingress of air in boilers leads to drops in energy efficiency. This paper presents the effects of air ingress in the combustion zone, post-
combustion zone and air pre-heater (APH) on the energy efficiency and loading capacity of a coal fired thermal power plant operating on
fuel with high ash (35–45%). The optimal O2 in the flue gas for a pulverized coal fired system is 3.5% (corresponding to 20% excess air).
The operating values are in the range of 4.2–6.0% in membrane type boilers and up to 10% in refractory type boilers (after sustained
periods of operation). The leakage rate of boilers (up to the entrance of the APH) is designed at 0.2% while the average operating values
are 7.25% for membrane type enclosures and 33.61% for refractory enclosures. The leakage rate of the APH is designed at 5.0% while the
operating values range from 13.66% to 20.13% for rotary and tubular APHs. When the O2 in the combustion zone varies from 3.5% to
8.0%, efficiency drops of 2.0% points are experienced in the boiler and turbine separately, and the gross overall efficiency drop is 3.0%
points. The units do not experience any capacity drop up to an O2 in the flue gas of 6.0% before the APH. At an O2 in the flue gas (before
APH) of 7.2%, a mild limitation on the unit capacity of around 2–3% is experienced. When O2 in the flue gas (before APH) reaches a level
of 9.0%, 20% capacity drop of the unit is experienced due to which the plant load cannot be raised higher than 80%. Beyond the level of
9.0% (rare occurrence), the unit is quite difficult to operate and has to be taken off for overhaul.
 2007 Elsevier Ltd. All rights reserved.

Keywords: Air ingress; O2 in flue gas; Excess air; Rotary air pre-heater; Membrane type water walls; Capacity factor; Boiler efficiency; Gross overall

1. Introduction • Primary air during pulverization and to heat and convey

the coal from the bunker to the combustor.
Infiltration of air in the boiler and internal leakage of air • Secondary air injection in the boiler to provide the tem-
in the air pre-heater are operating problems in many coal perature and turbulence required for effective combus-
fired thermal units, especially those using fuels with high tion of fuel.
ash (35–45%), and affect their energy efficiency and loading
capacity. The high ash in coal leads to acceleration of wear The total air supply (primary and secondary air) is the
and erosion of internal components, which affect the air theoretical air (stoichiometric) plus an optimal quantity
tightness. of excess air (20% for pulverized coal combustion, 5% for
Coal boilers perform the dual role of combustion of fuel liquid fuels and 3% for gaseous fuels) [1]. The excess air
and transfer of heat to the working fluid, viz. water-steam. is based on the hydrodynamic efficiency of intimate mixing
The combustor is designed for air supply as: of coal and air streams and is a measure of the inhomoge-
neity of mixing. If the mixing of air with fuel does not take
place, the unmixed air acts as a dilutant without any useful
Tel.: +91 80 23604682; fax: +91 80 23601213. purpose and only enhances the largest energy loss in the
E-mail address: msbhatt@powersearch.cpri.res.in boiler, viz. the dry flue gas loss.

0196-8904/$ - see front matter  2007 Elsevier Ltd. All rights reserved.
M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160 2151


A0, . . . , A2 constants in curve fits t time (h)

AFR air to fuel ratio (dimensionless: 4–6 for coal) v velocity (m s1)
APH air pre-heater VR velocity ratio (dimensionless: 0.8–5.0)
B0, B1 constants in curve fits X independent variable
C calorific value (MJ/kg) Y dependent variable
CF capacity factor (dimensionless: 0–1; 0–100% of
MCR) Greeks
CO/CO2 Carbon monoxide/dioxide (volumetric per- a excess air factor (dimensionless: 1.0–3.0)
centage in flue gas) d leakage rate (dimensionless: 0–1.0)
d tube diameter or transverse spacing (mm) D difference
D0, D1 constants in curve fits g efficiency (dimensionless or %)
E erosion rate (mm/h)
f fraction Subscripts
FD forced draft a ambient
ID induced draft av average
MCR maximum continuous rating (MW) A auxiliary
N2 nitrogen (volumetric percentage in flue gas) B boiler
O2 oxygen (volumetric percentage in flue gas) BA bottom ash
p pressure (Pa, kPa or MPa) e electric
P power input/output (MW) f flame
PA primary air G generator
Q rate of heat addition (MWt) in input
R2 measure of approximation of curve fit to data m mechanical
points. Predictive power of curve fit to related max maximum
data points O gross overall
SEC specific energy consumption (kW h t1) ref reference
SECR specific energy consumption ratio (dimension- S steam
less: 1.0–2.0) T turbine
T temperature (C)

On the other hand, if the excess air is below the optimal Injection of higher than the required quantity of second-
level by over 2% points (i.e., from 20% excess air to 18% ary air is utilized when the fuel is wet or when the unburned
excess air for coal combustion), it can lead to excessive pro- carbon level in the ash is high. In such cases, when excess
duction of CO (due to O2 deficient carbon conversion), secondary air is injected, the flue gas temperature is
increased risks of explosion (release of CO and H2), slag increased up to 10% (due to intensification of combus-
build-up and fouling in the post-combustion zones (due tion) beyond which it decreases (due to dilution of the com-
to temperatures in these zones exceeding the ash fusion bustion environment).
temperatures). The minimum quantity of air required for combustion
Earlier trends in minimizing excess air to their thread (stoichiometric air with a design excess air factor) is the tech-
bare levels led to unusually high levels of NOx. The nical air requirement for the process. Air is injected at a pres-
NOx in flue gas increased by 35% for a decrease of excess sure higher than atmospheric (1–2 kPa; operating tolerance:
air from 20% to 10% for coal [2,3]. Present trends are to ±0.2 kPa) and at high temperature (300–350 C; operating
keep the excess air to the barest level satisfying a guaran- tolerance: ±20 C). The entrance of air in the ambient mode
teed minimum NOx and minimum SO2 to SO3 conversion without design is generally termed as illegal ingress, tramp
[4]. air, infiltration, etc. This air, which enters the combustor
Injection of a higher quantity of primary air than or boiler, is ambient air and does not in any way aid the pro-
required results in mill wear (due to recirculation of coal– cess of combustion or heat transfer but instead acts as a hin-
air mixture), higher erosion of pulverized piping (due to drance in achieving the objective function of the boiler
higher flow velocities), combustion difficulties (due to lean processes. Its entry is mostly in the post-combustion zone,
primary air to fuel ratio), higher pressure drop, increased and even if it enters the combustion zone, it will not be in a
auxiliary power, higher primary fuel jet velocity at the bur- position or location where it can interactively be involved
ner tip (resulting in shifting of flame front away from the in the combustion process. While excess injection of quality
burner) and difficulty in flame scanning and sensing. air from the primary and secondary air fans can be
2152 M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160

controlled, infiltration is quite difficult to trace and control. • Reduced heat transfer efficiency (rate of heat generated
The universally accepted boiler draft is 50 Pa (maintained in hot combustion products/rate of energy input in fuel
negative to avoid pressurization and explosion). Deviation at the mill inlet) (from 92% to 91%) in water walls due
of the boiler draft from this value results in ingress or puffing. to lower flue gas temperatures in combustion and
The excess air in the flue gas could originate from sev- immediate post-combustion zones. The trend is
eral sources such as: reversed in the further post-combustion zones (convec-
tive zones) where the heat transfer is increased due to
• Excess air supplied deliberately to avoid smoke forma- high flue gas velocities. The overall effect is high flue
tion during combustion and to achieve combustion. gas losses and consequent decrease in boiler efficiency
• Excess air supplied through the fans intentionally to from 86% to 83%.
maintain the system hydrodynamic conditions. • Reduction in flame temperature from 1800 C down to
• Air ingress into the system in the combustion area, flue 1300 C with consequent reduction in radiant heat
gas passages and through air pre-heater in-leakage. transfer in the water walls, reduction in flue gas temper-
ature below ash fusion temperature causing slagging in
O2 in the flue gas is a primary indicator of ingress. When some sections of the boiler.
there is no combustion, O2 in the flue gas is atmospheric • Combustion difficulties like instabilities necessitating
(i.e., 21% by volume), which can be taken as the case of fuel oil support.
infinite dilution. O2 in the flue gas is zero under stoichiom- • Fouling of some portions of the boiler, affecting the heat
etric combustion. As the quantity of excess air in the flue transfer efficiency.
gas rises, the O2 content rises from 0% to 21% depending • Increase in volume of air through the boiler, resulting in
on the situation. increase in flue gas velocities from 10–12 m/s [5] to 15–
The excess air factor for the bulk of the flow can be 18 m/s in the main pass combustion zone and from 8–
approximately determined as 9 m/s [5] to 12–13 m/s in the return pass.
• Reduced residence time in the combustion zone from 2
a ¼ ðN2 Þ=½ðN2 Þ  ð79=21ÞðO2  0:5COÞ; ð1Þ
to 1.5 s.
where N2, O2 and CO are volumetric percentages of flue • Increase in heat transfer in the superheater and re-heater
gas components. Considering N2 = 79%, the formula can zones because of lower heat absorption in water walls,
be simplified to increased velocities and higher flue gas temperature at exit
portion of the water walls or entrance of superheaters.
a ¼ 21:0f =½21:0  O2 ; ð2Þ • Requirement of high quantity of superheater and re-
where f is a factor (0.9 < f < 1.0) to account for CO in the heater attemperating sprays to maintain the water-steam
flue gas. When CO is negligible, the formula can be further side heat balance in the sub-cooled, evaporating and
simplified (without significant loss of accuracy) to superheating zones of the boiler.
• Increase of size of ash particles (normal: 0.01–50 lm) to
a ¼ 21:0=½21:0  O2 : ð3Þ above 50 lm size and with higher erosion potential.
The leakage rate (d) with reference to a reference excess • Increase of heat transfer effectiveness (Q/Qmax) (because
air factor (aref) is given by of increased heat transfer coefficient) but decrease of the
heat transfer efficiency (Q/Qin) because flue gas loss is
d ¼ ½ða=aref Þ  1: ð4Þ
also dependent on its mass flow rate. Lower flue gas
In coal boilers of present day design, infiltration can occur temperature does not imply lower energy loss because
in three distinct areas: of increased quantity of flow.

• Combustion zone. Infiltration in post-combustion heat transfer zones of

• Post-combustion zone. the boiler causes dilution of flue gas, which results in effects
• Air pre-heater (APH). such as:

Infiltration in the combustion zone is the most damag- • Increased velocities in the main pass post-combustion
ing of the three and will give rise to problems of: zone from 10–12 to 18–20 m/s, leading to higher ero-
sion rate, lower residence time, reduced gas tempera-
• Reduction in air to fuel mixing efficiency results in high tures and higher flue gas temperatures at the exit of
level of unburned fuel in bottom ash and fly ash. the boiler.
• Expansion/propagation of oxidation zone upwards. • Overloading of induced draft (ID) fans, resulting in
• Reduced combustion efficiency (rate of heat usefully additional auxiliary power (leading to plant capacity
absorbed by water-steam/rate of energy generated in reduction).
hot combustion products) (from 94% to 93%) because • Hunting of motor currents of forced draft (FD) and pri-
of increase in losses due to unconverted/partially con- mary air (PA) fans in designs where these air flows are in
verted fuel. series.
M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160 2153

• Logic failure of FD fans because the flow through the Boiler

FD fan is based on the O2 in the flue gas before the exit
APH. If the O2 in the flue gas is increased, the logic is
designed to decrease the flow through the fans through
the hydraulic motor drive.
Air pre-
Infiltration in the air pre-heater results in:

• Lowering of the volume flow rate of high quality com-

bustion air, thereby leading to additional auxiliary Combusti Air pre-
on zone heater
power and capacity limitations.
• Dilution and cooling of flue gas with cold incoming air. Boiler
• Overloading FD, PA and ID fans.
• Damage of APH seals, resulting in entry of air from the
flue gas into the primary and secondary air streams. on zone
• Erosion in ID fans, leading to reduction in its efficiency.
• Reduced hot air temperature from APH outlet results in
lowering of secondary air injection temperature and
coal–air mixture temperature at the mill.

The objective of this paper is to quantify the air ingress

Fig. 1. Schematic of P type and tower type boilers.
and its effects on the energy efficiency of thermal power
units. Levy et al. [6] have studied optimal excess air for var-
ious coal finenesses from 60% to 95% mass below 75 lm cloth (1 mm) and backed by hydraulically set castable
size. They have shown that up to an excess air of 30%, there refractory (12 mm) and finally a skin casing of mild steel
is no deviation in boiler efficiency or gross overall efficiency sheet (3.15 mm). The leak tightness of the refractory
of the unit. Combustion controls using flame emission enclosure depends on the resistance to thermal cycles
spectroscopy, digital image processing and intelligent soft- undergone by the boiler, method of application and
ware have helped trim down excess air and emissions to the effectiveness of local repairs. In fixed refractory walls
very narrow optimized bands [7]. (which do not move downwards due to thermal expan-
For clarity of the systems studied, a brief review of boi- sion), such as pent house, refractory bricks are used in
ler technology is presented in the following section. place of castable refractory. Hydraulic lock arrangement
is made between fixed and movable refractory (attached
2. Boiler technology and ingress to arch portion of the water walls) with water seal
The boiler design features are as follows:
The present design of boilers is of the membrane type
• Construction: A number of boilers are described by design wherein the water wall tubes (of seamless carbon
Reznikov and Lipov [8]. Of these, only two are popular steel: 75 ± 1 mm B · 6 ± 1 mm t) are pitched in line at
amidst boiler designers, the P type and the tower type. 2d and the space between consecutive tubes is fusion
The construction of boilers for high ash coals is predom- welded with carbon steel/alloy steel panes, thereby enclos-
inantly of the P type (which is composed of two passes) ing the entire boiler in a steel enclosure while also acting as
with water walled combustor, super heater and re-heater fins. The refractory is eliminated, and the steel panels are
platens and pendants in the main pass; and low temper- enclosed by a light resin bonded mattress. The boiler wall,
ature superheaters, economizer and APH in the return bottom sides, 2nd pass and penthouse sides are covered
pass. The other popular construction is the tower type with fusion welded steel tube plate panels followed by
construction in which the entire boiler is in the form lightly bonded mineral wall mattress (25 mm) and with a
of a single conduit in the upward direction. Fig. 1 gives ribbed aluminum skin casing (1.2 mm). The pent house,
a schematic of the P and tower type boilers. roof, bottom and extended side walls are reinforced with
• Enclosure: The earlier designs of boilers were of refrac- refractory bricks. The buck stays holding the casing of
tory type enclosures wherein the water wall tubes are the piping are designed to withstand a pressure of
in line with gaps in between them and backed by a ±0.4 kPa.
refractory wall. Loose tubes (75 ± 1 mm B · 6 ± 1 mm Membrane type walls (designed for an operating life of
t) are pitched at 2 d (for front and real walls) and 15 years) are inherently superior in their response to air
1.2 d (sideways) with gaps in between each tube. In the tightness, thermal shocks and pressure fluctuations.
moving refractory walls, the tubes are followed by a Refractory walls have an operating life of around 6–8 years
gap of 40 mm, insulated reinforced fabric sheet asbestos after which they lead to loosening of the refractory/bricks,
2154 M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160

Mild steel sheet • Type of enclosure (refractory or membrane type and its
Insulation expended life).
Asbestos cloth • Type of APH (tubular or rotary).
Water wall tubes • Operating pressure in the combustion zone (50 to
+50 Pa).
Refractory type water wall

3.2. Excess air measurement through sensing of O2 or CO2

Insulation and skin casing
The methods of measurement of excess air are:
Water wall tubes

Membrane type water wall 1. Chemical analyzers: volume change by Orsat samplers.
2. Electrochemical cells: electrical signals (4–20 mV)
Fig. 2. View of refractory and membrane type water walls.
3. O2 ion conductivity cells-electrical signals (4–20 mA)
through yttrium stabilized zirconium oxide cells with
porous platinum electrodes on the air and flue gas sides
which are a source of ingress, puffing and heat losses. There [9]. Zirconium is an electrolyte that conducts only oxy-
is a tendency to extend the life of the refractory walls gen ions at temperatures above 600 C and gives a volt-
through partial repair or patchwork. This is a cause of seri- age difference (governed by Nernst equation) between
ous air ingress. Fig. 2 gives a schematic of membrane and the electrodes as a function of the ratio of the partial
refractory type water walls. pressures of the flue gas side and its temperature [9].
4. Catalytic combustion thermo chemical analyzers: elec-
• Air pre-heaters: The earlier boilers had shell and tube trical signals (4–20 mV). Magnetic gas analyzers: para-
APHs with air in horizontal flow in the tube side and magnetic generated effects.
flue gas in vertical flow on the shell side. The present 5. Infra red analyzers: electrical signals (4–20 mV).
boilers are fitted with tri-sectored rotary regenerative 6. Gas chromatography of sampled gas.
APHs with flue gas in downward flow and air (primary
and secondary) in upward flow through a regenerative The flue gas duct sizes are generally 3–4 m · 3–5 m, and
heat exchanger matrix, which is constantly rotating since the flow is in the highly turbulent region and aided by
alternatively between the flue gas and air sides at a speed active flow elements (FD and ID draft fans), it is well
of 11–12 rpm. Though the rotary APHs are superior to mixed. In this study, zirconium probes (62 mm B · 2 m
tubular APHs in terms of heat exchange effectiveness length) are used for measurement of O2 in flue gas. The
and surface area per unit volume [350 m2/m3 and vol- probes can access up to 50% of the duct length. The instru-
ume per unit load (m3/kW)], tip sealing is the source ments have an accuracy of 0.1% O2 over a flue gas temper-
of severe air leaks from the air side (pressure: +6.5 to ature range of 20–700 C, environmental temperature
7.5 kPa) to flue gas side (0.3 to 0.8 kPa). When air range of 0–55 C and environmental relative humidity of
leaks from the air side to the flue gas side, there is also 0–80%. The output is in the form of 4–20 mA and can be
some counter leakage of ash particles into the air side. acquired through either a stand alone data acquisition sys-
These can cause erosive damage to the air ducts and tem (off line) or through the plant distributed control sys-
dampers. tem (on line).

3.3. Sample size

3. Experimental work
A total of 44 coal fired units with the following compo-
3.1. Factors on which ingress is dependent sition are studied:

The present work pertains to coal fired thermal power 1. Plant size: 30 MW: 4, 62.5 MW:7, 110 MW: 9, 210 MW:
plants. A study of ingress in over 40 plants of different rat- 20 and 500 MW: 4.
ings, designs and makes has led to identification of the fol- 2. Boiler type: P type: 42, tower type: 2.
lowing factors on which ingress has strong dependence: 3. Enclosure type: Membrane type: 28, refractory type: 16.
4. APH type: Rotary: 26, tubular: 18.
• Plant size (30–500 MW).
• Plant load [50–100% of the maximum continuous rating
(MCR)]. 3.4. Curve fits
• Flow length of the boiler (from bottom ash hopper to
boiler exit) (40–100 m). Based on field studies conducted in the units and studies
• Period elapsed since the last overhaul (0–12 months). at the laboratory, the results are obtained in the form of the
• Type of boiler (P type or tower type boiler). following curve fits:
M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160 2155

1. Curves of Type 1 (Power series) Table 1

Basic parameters of a 210 MW coal fired thermal power unit
Y ¼ A0 þ A1 X þ A2 X 2 þ A3 X 3 : ð5Þ
No. Rated variable/parameters Unit Value
2. Curves of Type 2 (Exponential) 1. Power output (electrical) MW 210
2. Boiler efficiency % 86.0
Y ¼ B0 expðB1 X Þ: ð6Þ
3. Power input in coal (energy in coal) MW 636.6
3. Curves of Type 3 (Non-linear) 4. Power input in steam (energy in steam) MW 547.5
5. Coal flow rate kg/s 41.667
Y ¼ D0 X D1 : ð7Þ Total combustion air flow rate kg/s 269.700
6. Higher heating value of coal MJ/kg 15.257
7. Design coal (permissible variation: ±10%)
Fixed carbon % 30.0
3.5. Efficiency and capacity Volatile matter % 21.0
Ash % 41.0
Moisture % 08.0
Efficiency is the ratio of the rate of energy output to the Size (% mass below 75 lm) % 70.0
rate of energy input to the system under consideration. The 8. Stoichiometric air to fuel ratio kg/kg 4.480
various component efficiencies are defined as: 9. Flame temperature C 1552
10. Specific volume of flue gas m3/kg 7.980
• Boiler efficiency (gB) is the ratio of the rate of energy 11. Feed water flow through water walls kg/s 204.167
12. Main steam flow rate to turbine kg/s 194.444
output of steam produced at the boiler outlet to the rate 13. Flow through re-heater kg/s 166.667
of the energy input in raw coal at the mill inlet. The boi- 14. O2 in combustion zone % 3.50
ler efficiency is the product of the combustion efficiency 15. O2 in APH entrance % 3.53
and heat transfer efficiency. 16. O2 in APH exit % 4.20
• Turbine efficiency (gT) is the ratio of the rate of energy 17. Leakage rate in boiler up to APH % 0.17
18. Leakage rate in APH % 4.17
output at the shaft to the net rate of energy in steam sup- 19. Break-up of heat load to water-steam
plied from the boiler to the turbine. Economizer (sensible heat) % 10.33
• Auxiliary steam efficiency (gS) is the ratio of the rate of Water walls (latent heat) % 12.19
energy output in steam supplied to the turbine to the Super heaters (superheat) % 63.83
rate of energy output produced at the boiler outlet. Re-heater (superheat) % 13.64
20. Boiler dimensions
• Generator efficiency (gG) is the ratio of the rate of elec- Length of the flow passage in the boiler m 63.224
trical energy output at the generator terminal to the rate Width m 10.592
of energy output at the turbogenerator shaft. Depth m 13.868
21. Exist flue gas temperature C 135
The gross overall efficiency (gO) is given by 22. Temperature of air at APH outlet C 336
23. Main steam and re-heat steam temperature C 540
gO ¼ gB g T gS gG : ð8Þ 24. Surface areas of heaters
Water walls m2 2400
The capacity factor (CF) is the ratio of the load on the unit Super heaters m2 8200
to its rated MCR. Re-heaters m2 2600
Economizer m2 5200
Air pre-heater m2 19,000
4. Results and discussion

Table 1 gives the parameters for a 210 MW unit (consid- Most boilers have provisions for burner tilting horizontally
ered as a representative unit size for analysis). from +30 (upwards) to 30 (downwards). The design tilt
is +25 at 50% MCR and 0 at MCR with a buffer provision
4.1. Plant size to go up to 30 to overcome ingress and related problems
that cause lowering of heat absorption in the water walls. It
The curve fits of results of the effect of plant size (for can be seen that at part loads, the O2 in the flue gas is quite
conditions before and after the APH and before and after high, probably because the air fans are not sufficiently mod-
an overhaul) are given in Table 2. It is seen that ingress is ulated and allow a high air flow. However, in practice, part
higher in smaller size units and reduces with size of the load operation below 90% MCR is not very common. The
units. It is also likely that this is caused by the units being problem of excess air is at loads of 90–100% MCR where O2
older that are smaller in size and, hence, older technology, trimming and fine control provides scope for improvement
while larger units are of more recent design. in energy efficiency.

4.2. Plant loading 4.3. Overhaul period

The curve fits of the results for 110 and 210 MW units for The annual overhaul period is usually 12 months. The
the response of O2 with plant loading are given in Table 3. variation of O2 in 210 MW (membrane type enclosure with
2156 M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160

Table 2
Curve fits of results of O2 in flue gas with plant size
No. Particulars X: Plant (unit) size (30–500 MW) A0 A1 R2
1. Y: O2 in flue gas before APH after an overhaul (3.5–5.5%) 5.4229 0.0047 0.79
2. Y: O2 in flue gas after APH after an overhaul (4.0–7.0%) 7.0942 0.0060 0.80
3. Y: O2 in flue gas before APH before an overhaul (4.5–7.6%) 7.5672 0.0079 0.77
4. Y: O2 in flue gas after APH before an overhaul (5.0–9.5%) 9.5400 0.0092 0.76

rotary APH) over the overhaul period is given in Table 4. 4.5. Boiler enclosure type
The total running hours of the boiler since inception do
not have a bearing on ingress, but the total hours before The range of excess air (before APH), O2 in the flue gas
a capital overhaul strongly affects it. Overhauls give a and leakage rate (for varying conditions of plant size, load-
decrease by 1.5–1.75% points. ing, overhaul period, etc.) for refractory and membrane
type enclosures are given in Table 6. The average leakage
4.4. Boiler length rate is around 7.25% for membrane type enclosures,
whereas for refractory enclosures, it is 33.61%. Membrane
The flow length of the boiler is the length of the boiler type enclosures are inherently superior to refractory type,
flue gas flow passage starting from the bottom ash hopper and hence, all refractory type boilers must be structurally
below the combustion zone and terminating in the ash hop- converted into membrane type boilers.
per of the return pass (for P type boilers) and the exit of
the APH for tower type boilers. It is 94.5 m for a tower
4.6. Air pre-heater type
type boiler and 64.5 m for a P type boiler of 210 MW.
The curve fits for variation in temperature, draft and O2
The range of air ingress (rise/change O2 in level) and
in the flue gas along the boiler length are given in Table 5.
leakage rate for tubular and rotary regenerative APHs
are given in Table 7. The average O2 pick up is in the range
Table 3
of 1.80–1.96% points for the two types of APHs. The aver-
Curve fits of results of O2 in flue gas with plant load
age leakage rate is in the range of 13.66–20.13% for the two
No. Particulars X: Plant load (fraction of A0 A1 R2
types. The leakage rates for the rotary regenerative heaters
MCR: 0.40–1.0)
are comparable to those of the tubular APH, but the for-
1. 210 MW membrane type enclosure – P
mer are preferred because of considerations of size and
type boiler, rotary APH
Y: O2 in flue gas before APH (4–7.5%) 11.90 7.7700 0.88 heat transfer effectiveness. Reduction of leakage rates in
Y: O2 in flue gas after APH (4–7.5%) 13.05 6.6500 0.86 regenerative heaters handling high ash laden flue gases is
Y: O2 in flue gas before platen super 11.95 8.7500 0.88 a technical challenge.
heater (4–7.5%)
2. 210 MW membrane type enclosure –
tower type boiler, rotary APH
4.7. Excess air factor and flue gas velocities
Y: O2 in flue gas before APH (design 6.8133 3.6133 0.98
value) (5–7%) For a given section, the typical velocity distribution
Y: O2 in flue gas before APH (5–8.5%) 8.2600 3.9100 0.89 across the section is given by the velocity ratio (VR) with
reference to the average velocity as
3. 110 MW-refractory type enclosure – P
type boiler, tubular APH VRi ¼ vlocationi =vav : ð9Þ
Y: O2 in flue gas before APH (5–8.5%) 8.3993 3.2117 0.98
Y: O2 in flue gas after APH (6–9%) 10.491 3.6132 0.99 The flow of flue gases in the boiler is in the highly turbu-
Y: O2 in flue gas before platen super 8.2123 3.3182 0.98 lent regime in both the combustion and post-combustion
heater (5–8.5%)
zones (Reynolds number of bulk of fluid = 6.4 ·
106 ± 10%). In the post-combustion zones, the presence
of heat exchangers inside the boiler (superheaters and re-
Table 4
heater pendants/platens, economizers, etc.) results in high
Curve fits of results of O2 in flue gas with overhaul period
velocity flows through the spaces in between the heat
No. Particulars X: Overhaul period (fraction A0 A1 R2
exchangers and the boiler enclosure.
of interval between overhauls: 0–1.0; 0:
overhauled boiler, 1: due for overhaul) The VR for the bulk of the fluid at a given ingress level
with reference to that at the design air intake level can be
1. 210 MW membrane type enclosure – P
type boiler computed as
Y: O2 in flue gas before APH (design) 3.530 0
Y: O2 in flue gas after APH (design) 4.200 0
VRa ¼ ½AFRa þ ð1  fBA Þ=½AFRaref þ ð1  fBA Þ: ð10Þ
Y: O2 in flue gas before APH (4–6%) 5.910 1.7800 0.80 Over the range of interest (O2 from 3.0% to 8.0%), the ex-
Y: O2 in flue gas after APH (6–8%) 7.610 1.4700 0.82
cess air factor varies with O2 as follows (R2 = 0.99):
M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160 2157

Table 5
Curve fits of variations along the boiler length for 210 MW boiler (membrane type water wall enclosure, P type)
No. Particulars X: Boiler length (fraction of length: 0–1.0; 0: bottom ash hopper zone of A0 A1 A2 A3 R2
boiler, 1: exit of APH)
1. Y: O2 in flue gas (design) (3.5–4.2%) 48.757 213.630 284.880 125.74 0.93
Y: O2 in flue gas (operating) (4.0–6.0%) 22.958 107.35 143.75 63.509 0.92
2. Y: Flue gas temperature (design) at MCR (135–1200 C) 327.330 3611.80 3785.0 0 0.97
Y: Flue gas temperature (design) at 60% MCR (126 MW) (135–1200 C) 1628.700 180.67 12140.3 0 0.96
3. Y: Enclosure pressure (design) at MCR (50 to 2790 Pa) Y = 0.031e11.407X 0.91
Y: Enclosure pressure (design) at 60% MCR (126 MW) (50 to 2880 Pa) Y = 0.026e11.612X 0.90

Table 6 The variation of boiler effects with VR is as follows:

Excess air, oxygen and leakage rate of refractory and membrane type
i. Convective heat transfer (forced convection in turbu-
No. Particulars Refractory type Membrane type lence flow) over bank of tubes:
enclosure enclosure
1. Oxygen in flue gas (D%) Q=Qref / VR0:6 : ð14Þ
Max 9.40 6.70 ii. Erosion of tube material
Min 5.50 3.70
Average 7.86 4.63 E=Eref / VR2:7 to 3:5
: ð15Þ
2. Excess air factor
iii. Duct pressure drop
Max 1.81 1.47 Dp / VR2:0 : ð16Þ
Min 1.35 1.21
Average 1.60 1.29 iv. Flame temperature
3. Leakage rate T f ¼ C=½AFRa þ ð1  fBA Þ þ T a ð17Þ
Max 50.86 22.38 ðT f  T a Þ / VR1
a : ð18Þ
Min 12.90 1.16
Average 33.61 7.25 v. Ratio of SEC of ID fans:
SECR / VR: ð19Þ

a ¼ 0:8141ðO2 Þ
: ð11Þ The performance of the active flow elements (fans) is
given in Table 8.
The variation of VR (bulk of fluid) with O2 is given by
(R2 = 0.99)
VR ¼ 0:823e0631½O2  : ð12Þ 4.8. Variation in heat absorption rates in the boiler

The local VR varies between 0.80 (over the central core of The heat absorption rates in the heat exchangers of the
the enclosure where the super heaters, re-heaters, etc. are boiler are affected by the following:
placed) and 4.0 (at the free space between the heaters and
the boiler wall), which implies nozzling of the flow at the i. Reduced flame temperature decreases the radiant
edges. heat transfer in the water walls but increases the con-
The transverse spacing of tubes (mm) as a function of vective heat transfer in the superheaters, re-heaters
the flue gas temperature [R2 = 0.95] is given by and economizer.
d ¼ 34:68e0:0022T fg : ð13Þ

Table 8
Table 7
Performance of active flow elements (fans) in a 210 MW coal fired plant
Oxygen pick-up and leakage rate of tubular and rotary regenerative APH
No. Particulars ID FD PA
No. Particulars Tubular APH Rotary APH
fan fan fan
1. Oxygen pick-up in APH (%)
1. Baseline (reference) SEC (kW h/t of air for 1.60 1.50 3.80
Max 5.0 4.90
FD & PA fans; kW h/flue gas handled for
Min 0.2 0.30
ID fans)
Average 1.96 1.80
2. Composite efficiency of fan-motor (%) 56.70 56.90 55.70
2. Leakage rate in APH () 3. SECR (SEC/SECref) (dimensionless) 1.0– 1.0– 1.0–
Max 71.43 42.61 1.538 1.350 1.885
Min 1.43 1.85 4. Power ratio (P/Pref) (dimensionless) 1.0– 1.0– 0.30–
Average 20.13 13.66 1.231 1.359 1.0
2158 M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160

ii. Decrease in flue gas temperatures tends to decrease Table 10 gives the variation of the unit efficiencies with
the heat transfer in superheaters, re-heaters, econo- ingress of O2 in the flue gas. It can be seen that when
mizer and APHs but an increase in velocity has an ingress is present in the combustion zone, the drop in effi-
opposite effect and increases the heat transfer. The ciency is more serious, while it is less so with ingress in the
simulated results of the changes in heat transfer in APH. The boiler and turbine efficiency drop over the entire
the water walls, superheaters and re-heater are given range (O2 in the combustion zone: 3.5–8.0%) is 2.0%
in Table 9. points. The gross overall efficiency drop is 3.0% points.
iii. Table 9 also gives the superheater and re-heater spray The units do not experience any capacity drop up to an
quantities. O2 in the flue gas of 6.0% before the APH. At an O2 in the
iv. The decrease in flue gas temperature in the APH out- flue gas (before APH) of 7.2%, ID fans reach the limit of
let due to excess air is given by their operating margins (20% in pressure and 20% in flow),
2 and their motor currents reach up to 10% above their
T fg ¼ 204:69  15:13ðO2 Þ þ 0:3066ðO2 Þ ½R2 ¼ 0:97:
MCR value. This places a mild limitation on the unit
ð20Þ capacity of around 2–3%. The other auxiliaries are largely
The design exit flue gas temperature is 136 C with a unaffected. When O2 in the flue gas (before APH) reaches a
tolerance of +6 C. level of 9.0%, a 20% capacity drop of the unit is experi-
enced because of:
The lowering of flue gas temperature tends to mislead
operators that the boiler is operating efficiently and effec- • Overloading of the fans; their motors drawing full load
tively transferring the heat to the working fluid. Hence, power at partial plant loads.
the flue gas temperature must be viewed vis-à-vis the excess • Upset of heat loading; under loading of water walls and
air level or O2 level. overloading of superheaters beyond the controllable
capacity of the boiler (through burner tilting, spray
4.9. Overall unit performance attemperation, etc.).
• Excessive velocities leading to erosion necessitate limit-
The boiler, turbine and gross overall efficiencies have ing the loading of the unit to tolerable limits to protect
been experimentally determined for a number of coal the metal from erosive wear losses.
plants of various ratings [10–12]. In this section, the exper-
imental efficiencies are given for three cases of ingress as Consequently, due to these limitations the plant load
follows: cannot be raised higher than 80%.
Beyond the O2 level of 9.0%, the unit is quite difficult to
i. Variation with reference to O2 in the combustion operate and has to be under loaded to keep it in operation.
zone. The boiler efficiency is the product of the com- Such a situation is quite rare and was only once observed
bustion efficiency and the heat transfer efficiency. physically when the unit was due for its capital overhaul.
Both of these are affected by air infiltration. Difficul- Subsequently, the unit was withdrawn for overhauling.
ties in maintaining combustion and hydrodynamic The variation of plant capacity factor with O2 in the flue
stability is some times experienced. gas after the APH (since the ID fan is located after the
ii. Variation with reference to O2 in the APH at a con- APH) is given by
stant 3.5% in the combustion zone.
iii. Variation with reference to O2 in the APH zone Table 10
Curve fits of results of efficiencies of boiler, turbine and overall plant with
considering ingress in and constant leakage in the
variation in O2 in flue gas
No. Particulars A0 A1 R2
1. X: O2 in the combustion zone (3.0–8.0%)
Y: Boiler efficiency (82–85%) 85.629 0.3511 0.94
Y: Turbine efficiency (41–45%) 46.304 0.5310 0.98
Table 9 Y: Gross overall efficiency (33–37%) 38.778 0.5830 0.94
Changes in heat transfer rates (Q/Qref) in the heat exchanges of 210 MW 2. X: O2 in the APH at a constant O2 of
boilers at 100% MCR with variation in O2 in the flue gas 3.5% in the combustion zone (4.0–5.5%)
No. Particulars X: O2 in flue gas before A0 A1 R2 Y: Boiler efficiency (82–85%) 85.298 0.2261 0.98
platen superheater (%) Y: Turbine efficiency (41%) 41.000 0.0 –
Y: Gross overall efficiency (33–37%) 37.125 0.0984 0.98
1. Y: Q/Qref of super heaters (1–1.5) 0.7542 +0.0782 0.99
2. Y: Q/Qref of re-heaters (1–1.5) 0.7603 +0.0761 0.99 3. X: O2 in the APH zone considering
3. Y: Q/Qref of water walls (0–0.40) 1.3656 0.1298 0.99 ingress in boiler and a constant leakage
4. Y: Main steam spray (0–50 t/h) [1 t/ 35.538 +12.617 0.99 in the APH (3.0–10.0%)
h = 0.2778 kg/s] Y: Boiler efficiency (82–85%) 85.715 0.3083 0.98
5. Y: Re-heat spray (0–50 t/h) [1 t/ 30.461 +10.814 0.99 Y: Turbine efficiency (41–45%) 46.519 0.4695 0.92
h = 0.2778 kg/s] Y: Gross overall efficiency (33–37%) 38.994 0.5150 0.91
M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160 2159

CF ¼ 0:7989 þ 0:0861ðO2 Þ  0:0096ðO2 Þ The design changes suggested are:
3 2
þ 0:0002ðO2 Þ ½R ¼ 0:99: ð21Þ
• Conversion of refractory type water walls to membrane
type water walls – either partially in areas prone to high
4.10. Remedial measures –operational adjustment, overhaul ingress or for the entire boiler.
and design modification • Installation of ceramic lined coal burners that enable
very fine air flow adjustment.
The general experience is that ingress cannot be con-
tained through adjustments or modifications in the combus-
4.10.2. Air pre-heater
tion controls (air–fuel controls, draft controls or oxygen
The overhauling of the rotary APH includes the
trim controls). The NOx control strategies, such as fuel flow
biasing in burners, fuel flow biasing across a burner section,
over fire air, low excess air firing, flue gas re-circulation and
• Water cleaning of APH when the heater metal tempera-
use of low NOx burners, which reduce NOx by 20–60%,
ture is 70–90 C with water at 80–90 C to remove sul-
result in excess air flow above the bare combustion require-
phur deposits.
ment. This benign requirement of excess air cannot be clas-
• Cleaning of APH hopper of debris and ash.
sified under ingress. The source of unwanted ingress has to
• Replacement of radial and axial seals.
be identified and plugged. This strategy of suppressing leaks
• Cleaning of cold end baskets and replacement of dam-
in individual equipment is given below.
aged baskets.
• Cleaning and refurbishing of ash hoppers below the
4.10.1. Boiler (excluding APH)
Operational adjustments include:
The design changes include [13]:
• Setting of optimal position of burner air dampers.
• Overhaul and calibration of movements of dampers and
• Introduction of static seal on both sides of the hot end
guillotine gates.
sector plates.
• Equal loading of burners and air flow.
• Introduction of double radial and axial seals with cover-
• Minimizing air flow through non-working burners.
ing in between the sealing surfaces.
• Minimizing unbalance of air/flue gas flow in branched
• Over sizing or rotor post seals and backing with sup-
porting rings.
• Conduct of clean air tests (operation of air fans only
• Packing of the hot end sector plate in board with ther-
without the combustion. process before the boiler is
mal insulation.
lighted) and measurement of pressures and pressure dif-
• Increase of rotor diameter.
ferences in each related section/equipment.
• Reversing the rotation of the APH to gas-secondary air-
primary air to minimize erosion in the primary air
Generally, boiler suppliers recommend that they
must be routinely overhauled in intervals of 6000–8000 h
• Introduction of automatic leakage control system.
(or annually) and capital overhauled in intervals of
35,000–40,000 h (or 5 yearly) intervals to minimize ingress
By these measures, the measured leakage has been
reduced to 5.0% + 1% of the total air handled [13]. Also,
The changes through overhauling include checks for
the leakages are not highly sensitive to the period between
ingress, leak tightness, replacement of sealing of:
• Pent house.
• Sealing trough of bottom ash hoppers. 4.10.3. Ducts
• Water wall bottom header and bottom cover plate. The design modifications include introduction of:
• Side water walls, goose neck portions.
• Peep hole doors, explosion doors and access doors. • Self lubricated type of bearings for dampers.
• Secondary air dampers connections, corner dampers in • Anti-friction bearings for flow gates.
wind box duct and burner air ducts and inter air com- • Low passing dampers.
partmental dampers.
The changes through overhauls include check for ingress,
The changes through overhauls also include: leak tightness, replacement of sealing of the following:

• Repair of eroded baffles and deflectors. • Seals, expansion joints, dampers, gates and liners of
• Renewal of hydraulic lock arrangements between fixed ducts of APH, ID fan, PA fan, FD fan, mills etc. on
and movable portions of water walls. the suction and discharge sides.
2160 M. Siddhartha Bhatt / Energy Conversion and Management 48 (2007) 2150–2160

• Door seals and liners of access doors and observation cannot be raised higher than 80%. Beyond the level of
doors of ducts. 9.0% (rare occurrence), the unit is quite difficult to
• Worn out ducts are to be replaced by carbon steel ducts operate and has to be taken off for overhaul.
(5 mm thick for air and 6 mm for flue gas) backed by viii. Air ingress also leads to increased flow velocities
mineral wool mattress insulation (25 mm) and skin clad- resulting in erosion of internal components, which
ding of aluminum (1.2 mm). affects the energy efficiency through sub-standard
performance. Hence, all refractory type boilers must
5. Conclusions be structurally converted into membrane type boilers.
Major re-design of rotary APH seals is required to
i. Ingress is an operational problem in coal fired boilers contain the long term in leak from the air side to
that can significantly affect efficiency and, to a small the flue gas side to within 5%, which has been a
extent, the unit capacity. chronic problem.
ii. The optimal O2 in the flue gas for a pulverized coal
fired system is 3.5% (corresponding to 20% excess References
air), which is recommended to be kept as a bench
mark/target value for energy efficiency. The operat- [1] Reed RJ. Combustion handbook. North Am. Mfg. Co.; 1995.
[2] Lifshifts V. Ultra low NOx burner. Burlingame (CA): Cogen
ing values are in the range of 4.2–6.0% in membrane
Company Inc; 2005.
type boilers and up to 10% in refractory type boilers [3] Lifshifts V. Experience with low pressure drop, high efficiency low
(after sustained periods of operation). emission burners in power boilers. Burlingame (CA): Cogen Com-
iii. The leakage rate of boilers (up to the entrance of the pany Inc.; 2005.
APH) is designed at 0.2%, while the average operat- [4] Athens L, Claxton M, Richard TW. Effect of fuel composition on
emissions from ultra low NOx burners. Tulsa, Oklahoma: John Zink
ing values are 7.25% for membrane type enclosures
and 33.61% for refractory enclosures. [5] Kumar R, Arokiraj V. Design of coal fired boilers for improved
iv. The leakage rate of the APH is designed at 5.0%, performance. Proceedings of the national conference on best practices
while operating values range from 13.66% to in power generation, 19–20 July 1999. Bangalore: Karnataka Power
20.13% for the two types of APHs. Corporation; 1999. p. 19–31.
[6] Levy EK, Munukutla S, Jibilian C, Crim HG, Cogoli JG, Andrew
v. Air ingress in the combustion zone is much more seri-
FK, et al. Analysis of effects of coal fineness, excess air and exit flue
ous than ingress in the air pre-heater and affects the gas temperature on heat rate of a coal fired power plant, Paper No.
combustion and heat transfer process inside the boi- 84-JPGE-PWR-1. ASME: New York; 1984. p. 1–11.
ler. Ingress is usually higher in smaller boilers (below [7] Schmidt D. Combustion goes digital. Power Eng Int 1999;7(9):77–9.
150 MW) because of their refractory wall enclosures. [8] Reznikov MI, Lipov YM. Steam boilers of thermal power sta-
tions. Moscow: Mir; 1985.
Larger boilers (above 150 MW) are usually designed
[9] ABB, Zirconium oxygen sensors. Howard Road (UK): ABB Kent
with membrane type wall enclosures that are robust Ltd.; 2006.
in their resistance to air ingress. [10] Bhatt MS, Jothibasu S. Performance enhancement in coal fired
vi. For O2 in the combustion zone varying from 3.5% to thermal power plants. Part I: Boilers. Int J Energy Res
8.0%, efficiency drops of 2.0% points are experienced 1999;23:465–87.
[11] Bhatt MS, Mandi RP. Performance enhancement in coal fired
in the boiler and turbine separately, and the overall
thermal power plants. Part III: Auxiliary power. Int J Energy Res
efficiency drop is 3.0% points. 1999;23:779–804.
vii. The units do not experience any capacity drop up to [12] Bhatt MS, Mandi RP, Jothibasu S, Rajkumar N. Performance
an O2 in the flue gas of 6.0% before the APH. At an enhancement in coal fired thermal power plants. Part IV: Overall
O2 in the flue gas (before APH) of 7.2, a mild limita- system. Int J Energy Res 1999;23:1239–66.
[13] Venkataraman GR, Mariraj Anand K. Energy conservation improve-
tion on the unit capacity of around 2–3% is experi-
ments in air pre-heaters of boilers. Proceedings of the national
enced. However, when O2 in the flue gas (before symposium on energy conservation measures in generating sector, 17–
APH) reaches a level of 9.0%, a 20% capacity drop 18 November 2005, vol. V. Bangalore: Central Power Research
of the unit is experienced due to which the plant load Institute; 2005. p. 55–61.