Chilingarian G.V. et al. Surface Operations in Petroleum Production, II
Подождите немного. Документ загружается.
255
TABLE
7-VIII
Peripheral heat
flux
distribution for radiant tubes
13233,4
Point
No.
Peripheral heat distribution ratio
Direct Reradiation Total
1
1
.oo
0
1
.oo
2 0.918
0
0.918
3 0.676 0.027 0.703
4 0.336 0.115 0.451
5
6
7
8
9
10
11
12
0.080
0
0
0
0.080
0.336
0.676
0.918
0.232
0.315
0.342
0.315
0.232
0.115
0.027
0
0.312
0.315
0.342
0.315
0.312
0.451
0.703
0.918
’
Point
No.
locations are
shown
in Fig.
7-15.
Center-to-center spacing of tubes is two times the nominal tube diameter.
Tubes exposed to direct radiant heat from one side
only.
Tube centerline is spaced
1.5
nominal tube diameters from the refractory.
temperatures are not experienced on the front face of the tubes. The peripheral heat
distribution problem is particularly severe in waterwall steam generator designs in
whch all of the radiant heat is adsorbed on the front face of the tubes exposed to
the flame envelope.
Most oilfield steam generators are fired from one end only, which introduces a
longitudinal variation in heat flux over the length of the radiant tubes. With present
burner designs incorporating two-stage combustion to reduce
NO,
formation, and
with burners designed to produce relatively long, small-diameter flames, the maxi-
mum longitudinal heat flux is achieved at about
45%
of
the horizontal tube length as
measured from the burner end. For this reason, the ratio of radiant coil diameter to
radiant tube length is important in minimizing the longitudinal variation in heat
flux. The first several feet of the flame envelope has relatively low radiation
intensity until combustion is complete and the temperature
of
the flame envelope
has been reduced by the heat extracted by the radiant tubes, with a resulting lower
thermal driving force to transfer heat to the radiant tubes. It is recommended that
the ratio of radiant tube length to coil diameter should not exceed
4
to
1
for oilfield
steam generators fired from one end only.
In a well-designed oilfield steam generator, the maximum local heat flux in the
radiant section should not exceed the average heat flux by more than
78%.
This
percentage represents the combined variations in both peripheral and longitudinal
heat flux distributions. The susceptibility
of
the steam generator to total vaporiza-
tion of the liquid film and resultant deposition of solids must be evaluated on the
256
basis of the maximum point heat flux calculated for the specific design under
consideration.
The heat release in Btu/ft3-hr acheved in the radiant section is also an
important consideration in the design
of
successful oilfield steam generators. The
heat release is calculated as follows:
A
EXV
H=-
(7-11)
where:
H
=
heat released per unit of combustion volume, Btu/ft3-hr;
A
=
total heat absorption for whch the steam generator is designed, Btu/hr;
E
=
net thermal efficiency of the steam generator,
%;
and
V
=
combustion volume measured inside the radiant coil, ft3.
The heat release represents an experience factor for evaluating the suitability of a
radiant section for a given performance. It is both a measure of the longitudinal
heat flux distribution, which will be achieved, and the susceptibility of the design to
direct flame impingement on the radiant tubes. Experience has shown that heat
release values of
15,000-42,000
Btu/ft3-hr are satisfactory for oilfield steam genera-
tors, provided the limitation of radiant coil diameter to radiant tube length of
4
to
l
is
also observed.
Radiant tubewall temperature
The tubewall temperature experienced in the radiant section
of
an oilfield steam
generator is a function of the rate of heat absorption from the flame, the thickness
of the tubewall, and the rate
of
heat removal by the water or steam-water mixture
flowing inside the tube. The rate of radiant heat absorption is a function
of
flame
temperature, radiation characteristics of the fuel, tubewall temperature, and physical
location
of
the tube with respect to the flame envelope. In order to calculate the
maximum tubewall temperature, it is essential to use maximum local heat flux.
The maximum outside tube skin metal temperature may be determined as
follows:
R
h
AT,
=
-
Rxt
AT--
2-
K,
(7-12)
(7-13)
and
RXf
=
Kr
(7-14)
251
Fig.
7-18.
View
of
a pyramid-type convection section installed horizontally to reduce unit height.
(Courtesy
of
Struthers Thermo-Flood Corporation.)
where:
AT,
=
temperature rise across the tube side fluid film,
OF;
AT,
=
temperature rise across the tubewall,
F;
AT,
=
temperature rise across the scale,
F;
R
h
=
film coefficient of the water or water-steam mixture corrected to the
r
=
tubewall thckness, in.;
K,
=
thermal conductivity of tube, Btu/ft2-hr-
O
F-in.;
f
=
scale thickness on tube wall, in.; and
Kf
=
thermal conductivity of the scale, Btu/ft2-hr-
O
F-in.
The three temperature rises as calculated above are added to the main body
temperature of the flowing stream to give the maximum outside tubeskin tempera-
ture.
It is obvious from the above formulas that the rate at which heat is removed from
the tube is
an
important factor in establishing the maximum tubewall temperature.
=
maximum point heat
flux,
Btu/ft*-hr;
outside tube diameter, Btu/ft2-hr-
F;
258
TABLE
7-IX
Thermal conductivity
of
metals
'
(Courtesy of
TEMA
Standards,
6th
ed.,
1978)
Temperature Thermal conductivity (Btu/ft2-hr- F-in.)
(OF)
Steel Carbon-
1%
Chrome-
2-1/4%
Chrome-
1/2%
MO
1/28
MO
1%
MO
200
300
400
500
600
700
800
900
360 348 324
348 336 324
336 324 312
324 312 300
312 300 288
300 300 288
288 288 276
216 216 252
300
288
216
276
264
264
252
252
~~ ~~
'
Use values
of
8-12
for
thermal conductivity
of
scale
or
deposited solids.
It
is for this reason that a once-through oilfield steam generator, with a high
volumetric percentage of steam in the radiant coil, requires a much more conserva-
tive radiant section design than do conventional boilers with high water recircula-
tion ratios. The effect
of
any significant solids deposition on the tubewall tempera-
ture also becomes apparent from the above formulas. Table
7-IX
shows thermal
conductivities for typical tube materials used in oilfield steam generators.
Convection section design
Most oilfield steam generator designs currently available incorporate convection
sections to improve the overall thermal efficiency
of
the steam generators. Inlet
feedwater temperatures vary from
60
to
210"F,
depending on whether or not
produced water is used and the type of water treatment employed. In any event, the
inlet feedwater temperature is generally at a sufficiently low level to allow efficient
heat recovery in the convection section without resorting to the preheating
of
combustion air.
Most convection sections are of the cross-flow, straight-tube type with several
rows of bare shock tubes followed by multiple rows of extended surface tubes. The
two conventional types are
(1)
a rectangular design of uniform cross-section
designed to produce a uniform flue gas mass velocity, and
(2)
a tapered design of
variable cross-section designed to produce essentially
a
uniform linear velocity
through the convection section. The latter design, classified as a tapered or pyramid
convection section, has been proven in extensive field testing to increase the length
of time between cleanouts
of
the convection section when firing the steam genera-
tors with the heavier crude oils (see Fig.
7-18).
In the case of steam generators fired with heavy crude oils, metallic salts
contained in the crude oil and soot, which is formed in the combustion process, tend
to accumulate on the convection tubes, gradually reducing the heat transfer. Thus, a
259
TABLE
7-X
Estimated flue gas dewpoints
'
(after Monrad,
1932)
Sulfur content
SO,
in flue gases
SO3
content Estimated flue gas
(wt
%)
(PP4
iwt dewpoint
(
F)
0.1
0.25
0.5
0.75
1.00
1.50
2.00
2.50
3
.oo
3.86
9.65
19.29
28.94
38.58
57.87
77.16
96.45
115.74
0.00039
0.00097
0.00193
0.00289
0.00386
0.00579
0.00772
0.00965
0.01157
247
263
274
283
288
296
303
307
310
'
Table is based
on
the use
of
10%
excess air for combustion and
2.5%
of
the
sulfur
being converted to
so,.
convection section whch cools the flue gases to
400
O
F
when it is clean, may only
cool the flue gases to
600-650°F
after
8-10
weeks
of
full-load operation. It has
become conventional practice in California to shutdown a steam generator when the
exit flue gas temperature reaches the
600°F
level, in order to water wash it for a
period of several hours to clean the outside of the tubes. The pyramid convection
section has been shown to increase the time between water washngs by roughly
60%
over that required for a uniform cross-section design; however, it does require a
higher draft loss and increased combustion blower horsepower.
When oilfield steam generators are fired with sulfur-bearing oils, it is essential to
design the convection sections in such a manner that
no
metal surfaces are below
the dew point of the SO,-bearing flue gases.
In
typical oilfield steam generators,
about
3.5%
of the sulfur
in
the fuel is converted into
SO,
during the combustion
process. The presence
of
SO,
in the flue gases sharply elevates the flue gas dew
point over that which would be experienced with
no
sulfur in the fuel. Table
7-X
shows calculated flue gas dew points with various concentrations
of
SO,.
Because
of
the dew point elevation problem, when firing with sulfur-bearing fuels, most oilfield
steam generators are provided with feedwater preheaters designed to preheat
feedwater to at least
200
OF.
This is accomplished by exchanging heat between water
at some intermediate point in the convection section and the feedwater in a
double-pipe heat exchanger.
Steam generator thermal efficiency
The thermal efficiency of an oilfield steam generator is determined by dividing
the total heat added to the feedwater between the steam generator inlet and outlet
nozzles by the heat released by the burner. The heat used in preheating the fuel oil
and in generating atomizing steam for the burner is lost as far as energy transferred
to the feedwater is concerned, and should not be included in the efficiency
260
Fig.
7-19.
Chart
for
approximate determination
of
oilfield steam generator heating requirements.
calculation. Similarly, the heat lost from the steam generator to the atmosphere
should not be included in the net steam generator thermal efficiency determination.
Relationship between the enthalpy and steam quality is presented in Fig. 7-19.
It has become standard practice to specify the thermal efficiency
of
oilfield steam
generators on the basis of the lower heating value
(LHV)
of the fuel. Figures 7-20
and 7-21 show the relationships between the oilfield steam generator net thermal
efficiency and flue gas temperature for typical gas and oil fuels at two different
excess air ratios. An allowance of 1.5% for the steam generator radiation
loss
to the
atmosphere has been included in the natural gas chart. An allowance of 1.75% for
(1)
the steam generator radiation loss to the atmosphere, (2) the heat required to
preheat the fuel oil, and
(3)
the heat necessary to generate the fuel oil atomizing
steam, has been included in the fuel oil chart. Most oilfield steam generators are
designed for operation with 10% excess air and a stack temperature of about 400°F
with a clean convection section. The heat loss from bare pipe to the atmosphere in
still air, with wind velocity correction factor, is shown in
Fig.
7-22.
Feedwater deaeration
In order to prevent corrosion of the tubing, it is recommended that feedwater to
once-through oilfield steam generators be deaerated to reduce the oxygen content
261
92
91.
90
89
5
88
i
w
u
87
2
86
U
85
4
:
84
I
t-
W
t-
83
82
81
8C
79
7E
Fig.
7-20.
Relationship between the thermal efficiency and flue gas temperature of natural gas-fired
steam generators. The efficiency is based on the lower heating value
(LHV)
of the gas, and includes an
allowance of
1.5%
for steam generator setting loss to the atmosphere.
down to less than
0.01
ppm. Both chemical oxygen scavenging and steam deaeration
have been used successfully to remove dissolved oxygen from the feedwater.
Selection of the deaeration method to be used generally is dictated by the size
of
the
steam injection project and by personal preference.
Two-phase pressure drop
Most once-through oilfield steam generators are designed for a substantial
tube-side pressure drop
of
between inlet and outlet nozzles varying between
100
and
300
psi, depending on the size of steam generator. The high pressure drop is a result
of the relatively
high
two-phase flow velocities for which this type of steam
generator is designed in order to ensure good heat transfer and adequate cooling of
the tubewalls. Figure
7-3
allows the approximate determination
of
pressure drop
of
85%-quality steam flowing in tubes. About
90%
of the pressure drop through a
once-through oilfield steam generator is experienced in the two-phase flow region.
Steam quality measurement instrumentation
The development of the once-through, 80%-quality steam generator produced a
need for a convenient and rapid method
of
determining the quality of the steam
262
Fig.
7-21.
Relationship between the net thermal efficiency and flue gas temperature of oil-fired steam
gcnerators. The efficiency is based
on
the lower heating value (LHV)
of
the oil, and includes an
allowance of
1.5%
for steam generator setting
loss
to the atmosphere.
0
100
200
300
400
500
600
700
800
TEMPERATURE DIFFERENCE BETWEEN
PIPE WALL AND AMBIENT
AIR,
OF
Fig.
7-22.
Heat
loss
from bare pipe to the atmosphere
in
still air with wind velocity correction factor.
263
whch is being produced by the steam generator. The need was particularly critical
because exit steam quality determination is the principal guide for the operator in
establishing the burner firing rate in relationship to the feedwater flow rate.
The primary method, which has been in use since the early
1960's
for measuring
steam quality on oilfield steam generators, has been the determination
of
the
increase in salinity of the water.
A
small separator is used to remove a sample of
liquid from the steam-water mixture leaving the steam generator. The sample is
cooled before it is depressurized
so
as to prevent flashing and additional concentra-
tion, and then its conductance is measured by a standard salinity meter. By
comparing the conductance of the sample removed from the steam generator outlet
with the conductance of the feedwater, one can immediately establish the increase in
salinity which has occurred.
A
fivefold increase in salinity would indicate 80%-qual-
ity steam.
In recent years, advanced continuous steam quality measurement systems utiliz-
ing microprocessors have been developed and are in commercial use. Figure
7-23
DUST PROOF CONSOLE
1
I
C
AL
I
E
R
ATORS
r-------
I
STEAM GENERATOR SYSTEM
FT
FLOW TRANSMITTER
-
STEAM QUALITY MEASUREMENT SYSTEM
PT
PRESSURE TRANSMITTER
A+-
AIR SIGNAL P/A PRESSURE /ANALOG CONVERTER
*DIGITAL SIGNAL
A/D
ANALOG-TO/FROM-DIGITAL CONVERTER
.+Ir-
ORIFICE ASSEMBLY
P/A
D
IFF
E
R
E
N
T
I
A L PRESSURE -TO -AN
A
LOG
ANALOG SIGNAL
Fig.
7-23.
A
typical continuous steam quality measurement system
for
50-
to 85%-quality steam.
264
shows the arrangement of such a system. The quality measurement is made by
determining the two-phase flow pressure drop across
a
calibrated orifice plate in
accordance with field test data established by Collins and Gacesa (1971). The
microprocessor is programmed to correct for variations in steam pressure and
temperature. The microprocessor can also be programmed to perform a number of
control functions based on the continuously measured steam quality.
A
number of
larger steam injection projects in the United States and Canada are now using this
arrangement for continuous steam quality measurement.
Low
-NO,
burners
The increasingly more stringent limitations on emissions of oxides of nitrogen
being implemented in the United States have forced oilfield steam generator
manufacturers to develop techniques whereby the units will perform satisfactorily
with significantly reduced
NO,
emissions. The problem of reducing NO, emissions
is particularly difficult when burning heavy crude oil with a high fixed nitrogen
content.
Several methods have been used to reduce NO, emission levels by oilfield steam
generators:
(1)
Ammonia injection into the flue gases leaving the radiant section of a steam
generator.
(2) Use of improved excess air control systems to reduce oxygen levels in flue gas
to
between 1.0 and 1.5%.
(3)
Flue gas recirculation to reduce the flame temperature.
(4)
Catalytic conversion of
NO,
to free nitrogen and water.
(5)
Utilization of two-stage combustion burners designed to operate in a reduc-
ing mode, followed by final oxidation of the fuel to complete the combustion.
Conventional oilfield steam generators without any provisions for
NO,
reduction
and firing crude oil with up to 0.8% fuel-bound nitrogen will produce flue gases
containing 225-380 ppm of
NO,.
With any of the above methods, NO, emissions in
the flue gases can be reduced to 125 ppm or lower.
At the present time, the most widely used method for reducing oilfield steam
generator NO, emissions is the recently developed low-NO, burner as supplied by
several burner manufacturers. The low-NO, burner is designed to perform the
combustion in two stages. The first stage of combustion is performed in a reducing
or neutral atmosphere in order to decrease
NO,
formation. The combustion is then
completed in a second stage, following the introduction
of
additional air to increase
the oxygen level in the flue gases to between 1.0 and 1.5%.
Feedwater treatment
Whereas once-through oilfield steam generators can accept feedwater of much
lower overall quality than would be considered suitable for use in conventional
high-pressure power boilers, there are still certain quality requirements which must
be met in order to insure satisfactory operation. Feedwater treatment procedures