Dr.

Albrecht Kaupp Page 1

The soot and scale
problems
Issue
Soot and scale do not only increase energy
consumption but are as well a major cause of
tube failure.
Learning
Objectives
Understanding the implications of soot at
the fire side
Understanding the implications of scale at
the water side
Estimating the energy losses through soot
and scale
Knowing additional negative side effects
of scale and soot build up
Relating boiler performance parameters to
soot and scale build up
The soot and scale problems Page 2
NOTES
1. Introduction
Boilers are delivered clean with no soot, slag and scale.
onse!uently a soot and scale problem is a classic management and
operational problem that has very little to do with boiler design.
Soot and slag is a mi"ture of solid carbon, ash, and molten ash that
stic#s to the fire side of the tube and prevents heat transfer. Slag
will also cause corrosion.
Scale is a hard coating or layer of chemical materials on internal
surfaces of the boiler e"posed to the water side. Scale mitigates heat
transfer and may lead to corrosion as well.
Scale and soot prevention is one of the most important tas# of a
boiler operator besides reducing stac# gas losses.
$he cause of scale will be e"tensively discussed in lecture %&. Soot
and slag deposits at boiler tube surfaces are mainly a firing problem
and'or mismatch of the fuel and burners.
2. Soot and slag generation
(n any combustion process of a fuel there will be always some
unburned carbon )soot* generated and some ash carried with the
stac# gas stream. Soot, ash, and molten ash )slag* will accumulate
at the tube ban#s of the heat e"changers. Some ash will even melt
down at the tube surface. $he final result is a layer insulating the
tubes against the hot combustion gases.
Soot generation has various causes such as
a) Freezing the chemical reaction
$he flame comes too close to the boiler walls, +free,ing- the
#inetic reaction. $his effect is best demonstrated with a #nife
one passes through a candle flame. $he flame will blac#en the
#nife.
.ajor causes are overfiring of the boiler, too much e"cess air,
and worn out burner no,,les.
b) Insufficient atomization of fuel oil
The soot and scale problems Page
NOTES
/e tal# about the three +$0s- in combustion, turbulence,
temperature and time. 1ow oil line pressure and lac# of
turbulence will result in poor atomi,ation of fuel oil, resulting
in larger droplets leaving the flame envelope partially unburned.
.ajor causes are low oil line pressure, lac# of primary
combustion air, or low fuel oil preheating temperature.
c) High moisture content in the fuel
$oo much water in a fuel leads to a +cold- fire causing
e"cessive smo#e generation because there is not enough fuel
energy available to generate a sufficiently high flame
temperature. 2owever some water helps to speed up #inetic
reaction. (t also shortens the flame length.
.ajor cause of this type of soot generation is a combination of
high moisture content and too much e"cess air.
d) Erratic feeding of solid fuels
Solid fuel firing is not as +smooth- as li!uid or gaseous fuel
firing. E"cess air is changing constantly, causing cold poc#ets
in the furnace, where too much fuel and not enough +fire- and
air are present. $he result is smo#e generation.
.ajor causes are bad combustion air distribution and
malfunctioning or badly designed feeding mechanism.
e) ri!!ing burner
3ccasionally li!uid oil drips directly from the burner down into
the fire tube and forms a pile of soot. $his soot burns up and
generates smo#e.
(n particular in boilers with superheaters, soot and slag accumulates
at the superheater tube ban#s first because the superheater is the
first heat e"changer passed by the products of combustion. 4ll
superheaters have +soot blowers- that are activated periodically to
blow off the soot with steam.
". Fire tube and #ater tube differences
The soot and scale problems Page !
NOTES
$here is a basic difference between fire tube and water tube
arrangements, with respect to the location of the soot and scale.
5igure % shows a water or superheater tube with soot at the outside
of the tube and scale at the inside. (n a water tube boiler the hot
combustion gases pass a ban# of tubes at the outside and release
their energy to the water or steam flowing inside the tube.
Figure 1: Water tube with soot outside and scale inside
Figure 2: Fire tube with soot inside and scale outside
The soot and scale problems Page "
NOTES
5igure 6 shows a fire tube with soot inside the tube. (n a fire tube
boiler the hot combustion gases pass through a bundle of tubes and
release part of their energy to the water at the outside.
5ire tube boilers are usually smaller )% to 6& t'h* and mostly in the &
to 67 bar range. Soot cleaning is simple, re!uiring only to open the
bac# and front door of the boiler to e"pose the hori,ontal fire tubes.
Soot cleaning of water tube boilers is much more complicated, since
they are larger and more complicated built.
$. Stac% gas tem!eratures &ersus
soot and scale de!osits
$he rated heat output of a boiler is based on firing a specific fuel at
a specific e"cess air factor. (n a clean boiler this will result in a well
defined stac# gas temperature for a specific fuel firing rate. Stac#
gas temperatures depend very much on the firing rates. $he stac#
gas temperature is usually lower at +low fire-, while it reaches its
pea# at +high fire-.
.ost manufacturers specify what stac# gas temperature is to be
e"pected at +low fire-, +high fire- or rated heat output. $he stac#
gas temperature always refers to a point in the system where the
combustion gases do not release heat any longer to the water8steam
circuit. $his boiler specific stac# gas temperature is our yardstic#
and the best we can achieve in the field.
2igh stac# gas temperatures are therefore a sure sign that boiler
heat e"changer surfaces have accumulated soot and'or scale.
4 highly recommended method of reducing fuel consumption is to
set a so called best stack gas temperature at +high fire- and to
instruct operators to clean the boiler if stac# gas temperatures
e"ceed this value by %7 to 67
o
.
$his practice is in particular easy to implement with fire tube
boilers, where cleaning of the fire side can be accomplished in half a
day. 9escaling of the water side is usually complicated and done
yearly if necessary.
(mplementation of this practice would re!uire to record stac# gas
temperatures about three times a day at the high fire setting. :ote,
The soot and scale problems Page #
NOTES
that the stac# gas temperature at a lower firing rate will be
considerable lower. 4lso, increasing the e"cess air lowers stac# gas
temperature. 5urthermore most in line temperature sensors get
easily fouled, with soot deposits at the stem. 5ouled temperature
sensors show a lower temperature because the soot layer insulates
the stem against the hot stac# gas.
4s discussed, any 67
o
decrease of stac# gas temperature could
easily save % ; fuel and is therefore worthwhile to consider.
'. ( crash course in a!!lied heat transfer
(n a fire tube boiler, the energy of hot combustion gases are
transferred through the soot layer, the tube wall and the scale layer
to the water.
Both, the soot as well as the scale act as insulators. $he effect is a
hotter than usual stac# gas temperature because the combustion
gases were prevented from efficiently transferring their energy to
the water side.
9epending on the water chemistry, different types of scale e"ist.
2igh density scale ) < 6,77786,=77, < 7.=86.>>*
.edium density scale ) < %,77786,&77, < 7.%&8%.%?*
1ow density scale ) < >778%,677, < [email protected]>*
Roh, )*, is the scale density in #g'm
>
, while Lambda, )*, stands for
the thermal conductivity in /'m
o
.
$he thermal conductivity of soot is between 7.7> and % /'m
o
,
depending on how much slag has been accumulated at the surface.
5or comparison, good insulation of steam distributions line have <
7.7& to 7.> /'m
o
. Boiler tube walls have a thermal conductivity of
&7 to ?7 /'m
o
and are very good heat conductors.
4n insulator causes a high temperature drop across its thic#ness,
while a very good conductor causes a very small temperature drop
across its thic#ness.
Since on both sides of the tube we have a fluid flowing )stac# gas,
water, or steam* there are also two so called film heat transfer
coefficients, h
263
, for the water or steam side and h
gas
for the gas
The soot and scale problems Page $
NOTES
side. $he film heat transfer coefficient, h
gas
, describes how easily the
gas transfers its energy to the metal or soot surface, while h
263
describes how easily the water or steam pic#s up the heat that
penetrates through the metal tube or the scale surface.
$he combined effect is e"pressed in a well #nown e!uation
describing heat transfer through tubes with three layers )scale, metal
and soot*.
( )
Q
L T T
r h
r r
k
r r
k
r r
k r h
a b
<
6
% %
7 7
% 7
%
6 %
6
> 6
> > >

+ + + +

ln ln ln /att
$a < temperature of the fluid )steam, water, stac# gas* inside the tube,
o

$b < temperature of the fluid )gas, water* outside the tube,

o

r7 < inner radius of the free cross section of the pipe, m

r% < outer radius of the first layer, m
r6 < outer radius of the second layer, m
r> < outer radius of the third layer, m
#i < thermal conductivity of layer i, /'m
o

1 < tube length, m

h7 < inner film heat coefficient, /'m
6 o

h> < outer film heat coefficient, /'m

6 o

$o demonstrate some of the temperature and heat loss effects it is

sufficient to assume the followingA
5ilm heat transfer li!uid water side, h
263
< %7,777 /'m
6 o

5ilm heat transfer steam side, h

263
< %,777 /'m
6 o

5ilm heat transfer gas side, h

gas
< %77 /'m
6 o

of scale < % /'m

o

of soot < % /'m

o

of metal tube < &7 /'m

o

$he above numbers are used in the e"ercise section.

). Heat e*changer tube failures
$here is a specific set of circumstances, where you as a consultant,
trying to reduce fuel consumption, may unintentionally contribute to
premature tube failure in heat e"changer tubes.
The soot and scale problems Page %
NOTES
Some boiler operators neither care about soot nor scale and operate
their boiler inefficiently. Recall that scale as well as soot are barriers
to heat transfer. Bour recommendation is to clean the fire side of the
boiler and remove the soot regularly, while descaling of the water
side can wait until the ne"t major shutdown for overhauling of the
boiler.
onse!uently one barrier for heat transfer is removed, but the other
remains. :ote that the barrier on the very hot combustion gas side
was removed, and subse!uently the metal surface temperature will
go up significantly )thermal stress*, depending on the scale
thic#ness on the water or steam side. $his phenomena would not
happen to this e"tend if both barriers, scale and soot, are removed.
Keep in mind that under certain circumstances cleaning the fire side
without cleaning the water or steam side will increase the danger of
thermal stress of boiler tubes. $his happens in particular at the tube
seats of the first pass of fire tube boilers.
(t is in general bad practice to fre!uently clean the fire side and
totally ignore scale built up at the water side. $his will reduce fuel
consumption at the e"pense of more fre!uent retubing of boiler
tubes. $ube corrosion is also accelerated at high temperatures. $he
costs of retubing and repairing water walls damaged by overheating
are much higher than any additional fuel costs due to reduced
efficiencies. 3ne should therefore always in!uire about descaling
practice and feedwater treatment.
+. Fuel reduction !otential
3nly +rules of thumb- figures for scale and soot can be given as
shown in 5igure > and 5igure C. 1osses are e"pressed in percent of
fuel input as a function of the scale or soot thic#ness. (naccuracies
are not caused by rough estimates of film heat transfer coefficients.
$heir value has no significant impact on the result. .ore critical is
the thermal conductivity of scale and soot layers that may vary
widely.
:evertheless even very thin layers )7.& to % mm* of soot and scale
cause significant fuel losses. (n addition scale built up causes
thermal stress to the tubes and increases repair and maintenance
costs.
The soot and scale problems Page &
NOTES
'((ect o( )oot on *uel +onsumption
0
2
4
6
8
10
0
.
5
1
.
5
2
.
5
3
.
5
Thic,ness o( )oot La-er. mm
/

*
u
e
l

'
n
e
r
g
-

L
o
s
s
Figure 3: Fuel energy loss due to soot
'((ect o( )cale on *uel 'nerg- Losses
0
2
4
6
8
10
12
0
0
.
2
5
0
.
5
0
.
7
51
1
.
2
5
1
.
5
1
.
7
52
)cale Thic,ness.mm
/

*
u
e
l

L
o
s
s
High density
Medim !ensity
"o# !ensity
Figure 4: Fuel energy loss due to scale
,. -easurement techni.ues
$he stac# gas temperature of a boiler, if continuously recorded over
longer periods )wee#s, months*, gives very important information
about the efficiency of a boiler and its operating hours. $emperature
recording is accurate, ine"pensive and easy to perform.
(n case a boiler re!uires closer observation one may first
install automatic single channel data loggers that measure the
temperature in % to %7 minute intervals over days and wee#s.
E!uipment costs are USD >&7 only.
The soot and scale problems Page 10
NOTES
4 long term temperature profile of a boiler is li#e a
electrocardiogram. (t can show on a time scale the followingA
$he temperature built up due to scale and soot formation
$he cycling behavior of the oil burner or solid fuel feeding
system
$he low and high fire intervals
$he boiler operating hours
onsultants are encouraged to continuously record boiler
temperature in case they intend to engage in a long term contract to
improve boiler efficiency.
'1'2+I)')
Tas% 1
4ssume there is a scale with < % /'m
o
at the inside of a boiler
water tube )39 < >@.% mm, wall thic#ness <>.C mm*. $he steam
and water mi"ture is at C7 bar and 6&7
o
. $he overall heat transfer
is C77 #/'m
6
inner pipe surface. 5ollow the steps to calculate the
temperature increase across each of the surfaces from the inside to
the outside )/ater film, scale, tube wall*.
)tep Input Data or '3uation 2esult
%* Steam temperature,$
S
, E 6&7 2'/ 01
6* (nner radius, r
7
, m 7.7%F7&
>* /ater film h
/5
, /'m
6
E %7,777
C* 2eat input, !
o
, /'m
6
C77,777
&* $emperature difference across
water film, $
water film
, E
!
h
7
/5
?* Scale thic#ness, t
s
, m 7.77%
=* onductivity scale
s
, /'m E
%
@* $emperature difference across
scale, $
scale
, E
r
7
!
7

ln
r t
r
s 7
7
+

s
F* $ube wall thic#ness, t
$
, m 7.77>C
The soot and scale problems Page 11
NOTES
%7* onductivity tube,
$
, /'mE
&7
%%* $emperature difference across,
tube, $
tube
, E
r
r t t
r t
s T
s
7 7
7
7
$
!


+ +
+

ln

%6* $ube surface temperature, $

surface
sum of rows %, &, @, %%
%>* Soot thic#ness, t
soot
, m 7.77%
%C* onductivity soot,
soot
, /'mE
%
%&* $emperature difference across
soot layer, $
soot
, E
r
r t t t
r t t
s T soot
s T
7 7
7
7
soot
!


+ + +
+ +

ln

%?* Soot surface temperature, E sum of rows %6 and %&

S486%7 arbon steel shall not e"ceed an o"idation temperature of
C&7E. $his e"ercise gives a very educational insight view on the
effect of scale and soot on wall temperatures of tubes. $he case
refers to the water wall in a furnace that receives its energy input
mostly by radiation and not by convection. :ote the rather high heat
transfer of C77 #/'m
6
.
Tas% 2
3ften an energy auditor is in no position to measure water or steam
temperatures. 4ssume a situation where the temperature gages are
missing or you don0t trust the reading, but you are able to measure
the surface temperature of the bare pipe correctly ) 6
o
*.
9id you ma#e a large error by assuming the fluid temperature is 6;
higher than the measured surface temperatureG
Repeat this e"ercise under the assumption that there was a % mm
scale inside the tube, you didn0t #now about. Use the same
procedure as in tas# %. 4sssume steam at 6&7E and a film heat
transfer coefficient of %777 /'m
6
E. $he ambient temperature is
>&E. 9epending on the emissivity of the bare pipe surface )shiny
or blac#* we may have a heat transfer of between C.C #/ 'm
6
and
=.& #/'m
6
from the pipe surface to the ambient.
Tas% "
The soot and scale problems Page 12
NOTES
.easuring surface temperatures re!uires special surface
temperature sensors and some s#ills. (t is always advisable to apply
enough pressure on the sensor and clean the surface, prior to
measuring. $est the effect of paint at the pipe surface. Even a very
thin layer of paint will change the surface temperature of a pipe
carrying hot feedwater at F& by about >8= .
9oes the temperature go up or down as compared to the surface
temperature of the clean pipeG