Proceedings of International Conference on Heat Exchanger Fouling and Cleaning - 2011 (Peer-reviewed)

June 05 - 10, 2011, Crete Island, Greece

Editors: M.R. Malayeri, H.
A.P.
Mller-Steinhagen
Watkinson and H.and
Mller-Steinhagen
A.P. Watkinson
Published online
www.heatexchanger-fouling.com

DESIGN AND OPERATING ASPECTS INFLUENCING FOULING INSIDE RADIANT COILS

OF FIRED HEATERS OPERATED IN CRUDE OIL DISTILLATION PLANTS

Z. Jegla, J. Kohoutek and P. Stehlik

Brno University of Technology, Institute of Process and Environmental Engineering,
Technicka 2, 616 69 Brno, Czech Republic, E-mail: [email protected]

ABSTRACT
Crude oil heated in a fired heater flows inside tube coil
consisting of two distinct sections: a convective section and
a radiant section. While the flow of fluid in coil of
convective section is one-phase liquid flow, the fluid flow
inside radiant tube coil is accompanied by boiling and
evaporation process, i.e. two-phase flow occurs there.
Fouling process on the fluid side of radiant tube coils causes
several problems which make outwardly as operating
troubles of different kind according to operating conditions
and type of application (distillation).
The most frequent problem associated with fouling is
rapid deposition of coke in tube coil resulting in fast
increasing of fluid pressure drop and necessity of the plant
shut down and decoking of fired heater. Another problem is
undesirable cracking of fluid especially in fired heaters of
vacuum distillation plants. Cracking process results in
formation of light cracked products, which burden a vacuum
system of distillation plant and influence quality of products
and also in formation of carbonaceous and coke deposits
which adversely influence plant operation. Both above
mentioned and often interrelating problems are usually
caused by too high or non-uniform heat flux of radiant coil
or low fluid flow velocity or combination of both.
The contribution discusses the most important design
and operating parameters influencing fouling process in
radiant tube coils and through industrial examples
documents disposable tools of designer and operator and
also limitations of designer and operator which restrict
possibilities of fouling process mitigation inside radiant tube
coils.
INTRODUCTION
Fired heaters operated in crude oil distillation plants are
large and complex items. Two main types of fired heaters
are used.
Cabin (or box) heater type, as those shown in Fig. 1, is
preferred for large heat duty applications (approximately
20 MW and more) and contains horizontally oriented tubes
(or tube banks) in both radiant and convection parts of
heater.
Vertical cylindrical heater type, as those shown in Fig.
2, is preferred for small and or medium heat duty
applications (below 20 MW) and contains vertically
oriented tubes in radiant chamber and horizontally oriented
tubes (or tube banks) in convection parts of heater.
Following further description of individual parts of fired
heater (of two above mentioned types) and operating
parameters gives us an idea about situation and basic

possibilities of designers (and operators) with relation to

fouling process inside and outside tube coils.

Fig. 1 Cabin type of fired heater (according to Jegla, 2006).

Fig. 2 Vertical cylindrical heater (according to Pelini, 2008).

Jegla et al. / Design and Operating Aspects

Fired heater parts and operating parameters

Arrangement and design of individual parts (tube coils
and tube banks) of fired heaters (together with operating
parameters of heated hydrocarbon fluids and combustion
products) dominantly influences their tendency to fouling
from inside and outside of the tubes. Following description
presents typical situation in fired heaters operated on
atmospheric and vacuum crude oil distillation plants.
Radiation (combustion) chamber
In the radiation chamber takes place the combustion of
fuel on burners, which leads to very high flue gas
temperatures (reaching 900 to 1,500C). Tubes usually
located close to cabin walls are exposed to high heat fluxes
and to radiation-convection mechanisms of heat transfer
with dominant radiation component. These high flue gas
temperatures (due to burn-off and direct effect of fouling
components) do not allow applications of extended surfaces
or any other techniques of heat transfer enhancement and
compact solutions. The only feasible method is to use plane
surfaces (i.e. plane tubes). It is possible to optimize their
location with respect to economically effective utilization of
released heat by tube system in radiation chamber (Jegla,
2008).
Crossover from radiation chamber to convection section
the shield section
Part of tube coil(s) which is located in flue gas stream
between radiation chamber and convection section is called
shield tubes (or shield section). This part of tube system is
the most exposed part of the heater tube system since due to
common action of radiation heat flux from radiation
chamber as well as intensive convection flue gas heat flux
coming from radiation chamber to convection section. Flue
gas temperatures at the crossover between radiation and
convection sections typically range from 900 down to
700C, they are obviously lower than in radiation chamber.
However, intensive heat transfer disables practical
application of enhanced and compact solutions mostly due
to heat transfer surface burn off. This section then also
enables only plane surface solutions (i.e. plane tubes) (Jegla,
2006)
First part of heater convection section
First part of heater convection section is located in the
area of convection section above the shield section in the
area of flue gas temperatures reaching from 500 to 700C.
Lower values of flue gas temperature pertinent to decreasing
influence of radiation component. This section enables
enhanced (compact) solutions depending on fouling and
burn off properties of flue gas. If there is no risk of fouling,
application of fins is feasible and achieving of compact
design can be possible. If there is any chance of fouling by
flue gas, lower degree of heat transfer enhancement can be
used, such as studded surfaces (studded tubes) (Jegla, 2006).
Second part of heater convection section
First part of heater convection section is followed by
second part of heater convection section which is located in

the area of flue gas temperatures reaching from 300 to

500C. Actual value of lower boundary of the mentioned
flue gas temperature region is influenced by temperature
differences between heater inlet temperature of heated fluid
(i.e. into second convection section) and heater outlet flue
gas temperature see diagram in Fig. 3. (Typical inlet
temperature of fluid heated inside heater tubes, (Tf,inlet),
ranges for example in case of heaters for atmospheric and
vacuum distillation of crude oil typically from 250 to 350C.
Thus, outlet flue gas temperature from this section (Tfg,outlet)
reaches from 300 to 350C.)
It can be note that in this convection section (flue gas
temperature range) it is usually possible to full utilization of
potentials of compact and enhanced solutions respectively
with respect of consider compact and enhanced solution
(with respect to fouling properties of flue gas).

Fig. 3 Typical flow scheme of fluids in fired heater.

Depending on possibilities (with respect to fouling
properties of flue gas), it is possible to obtain here even
higher degrees of compactness since burn off is no longer a
problem in this temperature range. Thus, higher density of
fins or studs can be applied than in first part of furnace
convection section. However, the selection of a concrete
degree of compactness or enhanced solution should be
supported by results of operating and technical and
technical-economic balances (optimization) (Jegla, 2006).
Utilization of flue gas waste heat from heater
Flue gas usually comes out of furnace convection
section with temperature reaching 300C. Potential of flue
gas temperature utilization (approximately 300-150C)
greatly depends on temperature of flue gas dew-point.
Effective utilization of this type of low potential waste
heat requires compact or enhanced solution depending again
especially on flue gas fouling properties. This type of heat is
commonly used for preheating of combustion air for furnace
burners, which allows to importantly reducing the amount of
noble fuels (Jegla et al., 2005).

www.heatexchanger-fouling.com

Heat Exchanger Fouling and Cleaning 2011

FOULING PROCESS INSIDE RADIANT COIL

OF REFINERY HEATERS AND ASPECTS OF ITS
MODELLING
Arrangement of flow of oil heated in fired heaters
operated on atmospheric and vacuum distillation plants is
schematically shown on Fig. 3. Oil firstly enters to second
part of heater convection section and then flows inside
convective tube bank systems in counter-current direction to
flue gas.
Oil preheated by this way than though the crossover
tubes flow to the radiant chamber. Here the arrangement of
the flow of crude oil is dictated by the technical possibilities
and requirements not only from thermal-hydraulic point of
view of heated crude oil but also by the distillation column
and connection pipelines (so called transfer lines)
requirements.
Thus in case of cabin type of heater the crude oil can
flow from top to bottom of radiant chamber and/or from
bottom to top. Moreover the most of applications is
processed for large capacities and thus the multi-pass
arrangement of radiant (and convection) tube coils systems
is necessary. This requirement importantly influences
especially arrangement of radiant coil system. As illustration
of this aspect, Fig. 4 and 5 show the typical multi-pass
arrangements of tube system in refinery fired heater for sixpass and four-pass and cases.

Fig. 5 Typical four-pass arrangement of (vacuum) fired

heater (Martin, 1998).
Fouling inside radiant coils take effect in the form of
rapid deposition of coke from the oil in tube coil. The
presence of coke results in fast increasing of fluid pressure
drop and necessity of the plant shut down and decoking of
fired heater. Another result is the appearance of hot spots
and overheating of tube resulting to tube failure. Fig. 6
demonstrates the effect of coke (fouling) layer on tube metal
temperature in comparison with the case without presence of
coke on example of radiant tube from crude vacuum heater.

Fig. 4 Typical six-pass arrangement of (crude) fired heater

(Martin, 1998).
Oil preheated in the convection section of heater is still
in liquid-phase only. However, during flow in radiant
section the oil starts boiling and vaporization, because the
role of heater is to preheat and prepare (i.e. evaporate) oil to
requirements of distillation column. The presence of twophase fluid flow inside the radiant coil plays an important
role in relation to modeling of fouling process of oil inside
radiant coils of heater.

Fig. 6 Effect of presence of coke layer on tube metal

temperature of vacuum heater radiant tube in
comparison with situation without presence of coke
layer (according to Martin, 1998).
Another problem associated with fouling is undesirable
cracking of oil especially in fired heaters of vacuum

www.heatexchanger-fouling.com

Jegla et al. / Design and Operating Aspects

distillation plants. Cracking and coking bring from operating

point of view following typical problems.
Oil cracking in crude vacuum units typically causes:
Production of non-condensable gases which can cause
flooding of existing vacuum systems or an enormous
investment cost of a new vacuum system;
Deterioration of products quality (viscosity index and
stability of products);
Heavy cracking coke particles could be found in the
bottom product with all the disagreeable consequences.
Oil coking means deposits in the tubes which lead to:
Higher metal (surface) temperature of tubes, which can
cause tube walls to fail prematurely;
Higher pressure drop due to reduced (by deposit) tube
cross flow area;
Parts of coke could peel off and accumulate inside the
column.
From above observations it is clear that the fouling
inside the radiant coils is caused by chemical reaction of the
oil with the hot tube surface. Current models for modeling
of fouling caused by chemical reaction in oil refining
processes (Wilson et al., 2002) are based on interaction of
deposition and removal term and with increasing
complexity continuing the development started by first
research works (Nelson 1934; Kern and Seaton, 1959;
Watkinson and Epstein, 1969;etc.).
However, models are focused on fouling of oil in the
liquid phase only which is the most frequent case in heat
exchanger oil preheat trains. Moreover, models are based on
constant heat flux, constant mass flow rate and the bulk fluid
temperature of oil. Thus, in case of radiant coils of fired
heaters and two-phase flow of heated oil the application of
these models is limited and indicative only. An important
observation made by Atkins (1962) was that in the fired
heater tubes fouling deposits appeared as two layers; an
outer porous and a hard crust of deposit next to the tube
wall.
In practice, the heater designers and operators work
with strongly limited tools for design and operation of fired
heaters from oil fouling point of view. The most important
aspects will be now discussed.

Bulk vs. film temperature

Oil flowing inside radiant coil is heated by the tube
metal temperature driven by the heat flux from hot flue gas
outside the coil. Generally the heat input rate is controlled to
achieve desired bulk oil temperature exiting heater.
However, in being heated to the required outlet, the oil film
in contact with the inner heating tube surface is always
exposed to a higher temperature than the bulk oil. In the
most applications it is the oil film temperature, not the bulk
oil temperature, which limits the duty of the heater and
usable life of the oil (Pelini, 2008).
The bulk oil temperature at the outlet of the heater is
easily measured and it is usually the process variable used to
control the heat input to heated oil. In contrast to bulk
temperature, the oil film temperature is not usually measured
directly, and it has to be calculated by the heater designer.
In the given local part of radiant coil the oil film
temperature (Tf) can be calculated from oil bulk temperature
(Tb) and film temperature rise (Tf) as follows:
Tf = Tb + Tf = Tb + qri/hi

(1)

where qri is a local heat flux to radiant coil (local radiant

heat flux) and hi is a film heat transfer coefficient.
Thus, this equation gives us clearly answer why the two
different heaters (i.e. with different qri and/or hi) operated
with the same duty, oil flow rate and heater bulk oil outlet
temperature may have significantly different film
temperature profiles.
There is also, in refinery practice, very frequent case
when heater placed in refinery unit is operated alternately for
heating two or more similar types of oils for the same duty
(processing of several crude oils from different suppliers).
Moreover, operation of one heater with alternately heated two
similar types of oils for the same duty and bulk inlet/outlet
temperatures can have (due to slightly different oil
composition) significantly different bulk temperature profiles.
For demonstration this fact the Fig. 7 and Fig. 8 show the oil
bulk temperature profile of two alternately heated oils in the
radiant coil in case of crude vacuum heater.
FURNACE TUBE COIL - TEMPERATURE DIAGRAM
Job :
Heater KR-1 & TL as Vac.Flasher
Job No.: Revamp TL & heater rad.coil - w ith coil steam - case A

FURNACE
425 INLET

Sc.35-37
ID=120mm

Sc.33-34
ID=154mm

Sc.26-32
ID=203mm

Sc.24-25
Sc.18-23
Sc.16-17
ID=254mm
ID=304mm
ID=387mm

420
415
410
405
400
395
390
385
380
375
370
365

Temperature [C]

DESIGN AND OPERATING ASPECTS OF HEATERS

INFLUENCING FOULING INSIDE RADIANT COILS
Chemical reaction of oil fouling inside radiant coil is
dependent mainly on the oil composition, temperature and
residence time in the radiant coil. Other words, in term of
thermal-hydraulic parameters, it can be state that oil
composition influences the actual flow conditions
(evaporation of oil, oil activation energy, etc.) at given
operating parameters (i.e. pressure, temperature), oil film
temperature is influenced by heat flux and oil film heat
transfer coefficient and oil residence time is function of oil
velocity (which is influenced by tube size and actual flow
conditions). Main important identified parameters
influencing the oil fouling process will be now further
discussed.

360
355
350
345
340
335
330
325
320
315
310
305
300
295
290
285
640630620610600590580570560550540530520510500490480470460450440430420410400390380370360350340330320310300290280270260250240230220210200190180170160150140130120110100 90 80 70 60 50 40 30 20 10 0
Distance from the furnace outlet [m]
KOCH-GLITSCH Brno, a.s.

8.10.2010

Fig. 7 Illustrative software screen: bulk temperature profile

for oil 1 heated in radiant coil of crude vacuum
heater (x-axis: radiant coil length, y-axis: bulk
temperature).

www.heatexchanger-fouling.com

10

Heat Exchanger Fouling and Cleaning 2011

As can be seen from Figures 7 and 8, oil composition

(properties) importantly influences oil bulk temperature
profile.
FURNACE TUBE COIL - TEMPERATURE DIAGRAM
Job :
Heater KR-1 & TL as Vac.Flasher
Job No.: Revamp TL & heater rad.coil - w ith coil steam - case B

425 FURNACE
INLET
420

Sc.35-37
ID=120mm

Sc.33-34
ID=154mm

Sc.26-32
ID=203mm

Sc.24-25
Sc.18-23
Sc.16-17
ID=254mm
ID=304mm
ID=387mm

415
410
405
400
395
390
385
380
375

As example demonstrating typical situation in

determination of two-phase film heat transfer coefficient the
Table 1 (adapted from Hewitt et al, 1994) comparing film
heat transfer coefficients for nucleate boiling of light oil
calculated according to different authors, is presented.
Thus, the calculation of two-phase film heat transfer
coefficient still depends of heater designer experience and
actual correlation preference.

370

Temperature [C]

365
360
355
350
345
340
335
330
325
320
315
310
305
300
295
290
285
640630620610600590580570560550540530520510500490480470460450440430420410400390380370360350340330320310300290280270260250240230220210200190180170160150140130120110100 90 80 70 60 50 40 30 20 10 0
Distance from the furnace outlet [m]
KOCH-GLITSCH Brno, a.s.

8.10.2010

Fig. 8 Illustrative software screen: bulk temperature profile

for oil 2 heated in radiant coil of crude vacuum
heater (x-axis: radiant coil length, y-axis: bulk
temperature).
Moreover, heat requirements for simultaneous heating
and evaporation of some oils are (for given flow and tube
size) too high in some local parts of radiant coil and the heat
flux can not cover local heat requirement. Such situation
results in appearance of temperature peak on bulk
temperature profile in these local parts of radiant coils, as
can be seen from example on Fig. 7.
Film heat transfer coefficient
Mentioned equation for oil film temperature contains the
heat transfer coefficient (hi). This thermal parameter
generally depends on oil properties and oil flow regime and
velocity. In case of heating of oil in one phase (i.e. liquid
only or gas only) the determination of film heat transfer
coefficient is relatively reliable. However, determination of
film heat transfer coefficient for two-phase oil flow is
complicated and unreliable. Therefore, some heater design
standards (for example API Standard 530, 2003)
recommend for determination of two-phase film heat
transfer coefficient an approximate solution. However, these
approximations are very rough (and suitable for tube
material selection purposes only) and dont reflect real
situation of flowing oil. In contrast of this, the theoretically
derived correlations for accurate determination of two-phase
oil film heat transfer coefficient are too complicated and,
moreover, results are significantly differ author from author.

Two-phase oil flow regimes and velocity profile

Two-phase flow of heated oil inside heater radiant coil is
accompanied by boiling and increasing of oil evaporation.
Thus, the simultaneous and continuous changes of two-phase
flow regime are a result of this heating process. The
orientation of tube coil significantly influences the own nature
and character of changes of these two-phase flow regimes.
Moreover, individual two-phase flow regimes (for example
bubble flow, slug/plug flow, churn flow, annular flow, mist
flow etc.) are accompanied by significantly different values of
film heat transfer coefficient. In case of vertically oriented
radiant tubes (i.e. vertical cylindrical heater, Fig. 2) some
undesirable two-phase flow regimes cannot arise (for example
stratified flow). However, their appearance can flourish in
case of two-phase flow inside horizontally oriented radiant
tubes (i.e. cabin type heater, Fig. 1) improperly sized (or
operated significantly out of heater design).
Appearance of such undesirable two phase flow regimes
is the reason for local overheating of some parts of radiant
tubes (due to low film heat transfer coefficient) due to
absence of liquid film on some parts of inside surfaces of
radiant tubes.
However, on the other side, proper design of radiant
coil for achieving the safe two-phase oil flow regimes
(such as annular two-phase flow) cannot be the reliable
solution from fouling point of view, when the (liquid) film
heat transfer coefficient (in relation to local heat flux) is not
sufficiently high.
For example of this, Fig. 9 shows the heavy fouled
(coked) radiant tube of crude vacuum heater caused by the
annular two-phase flow regime characterized by not
sufficiently high film heat transfer coefficient, and thus, by
exceeding the limiting oil film temperature.

Table 1. Comparison of results of correlations for film heat

transfer coefficient for nucleate boiling of light oil (adapted
from Hewitt, et al., 1994).
Author(s) of
Correlation
Foster-Zuber
Rohsenow
Mostinskij
Bier et al.
Cooper

Film heat transfer coefficient

[W/(m2.K)]
5,512
1,182
2,528
5,134
23,212

Fig. 9 Heavily fouled (coked) radiant tube of vacuum crude

oil heater.

www.heatexchanger-fouling.com

11

Jegla et al. / Design and Operating Aspects

Moreover, velocity profile of two-phase (and or single

phase) flow in radiant tubes is also significantly deformed
by local flow resistances (i.e. bending) caused by presence
of 90 or 180 elbows presented in radiant tube coils.
The Fig. 10 shows the formation of velocity profile and
flow in a 90 curved tube. The maximum velocity (vm) is (in
comparison with average velocity, va) displaced to the
outside due to wall friction influences and transverse eddies
arise.

indicator of fouling (coking) potential of most heaters. First

step in the determination of average radiant heat flux is a
determination of the radiant heat duty (Qr).
In generals, the calculation of Qr depends on many
factors including the furnace geometry, combustion
conditions and fluid temperature. Detailed procedures for
radiant section calculation are beyond the scope of this
paper, but there are well-established quasi-theoretical
techniques for determining radiant chamber performance
(for example Wimpress, 1978).
Once, the radiant heat duty (Qr) is calculated, the
average radiant heat flux (qr) is the Qr divided by the total
radiant heat transfer area (Ar). The heater vendor reports the
purchaser the average and local-maximum radiant heat flux.
In simplified form it can be state, that local-maximum
radiant heat flux qm can be related to average radiant heat
flux qr as follows:
qm = FC . FL . FT . qr

Fig. 10 Formation of velocity profile and flow in a 90

curved tube (according to Wagner, 1997).
Deformation of flow and velocity profile can in
conjunction with local heat flux of the elbow lead to the
fouling (coking) inside the elbow in the part with low oil
velocity. Fig. 11 documents this fact on the concrete
example of heavily coked a 90 curved tube from the radiant
coil of crude vacuum heater.

(2)

where FC is a tube-circumference heat flux variation factor,

FL is a tube-longitudinal heat flux variation factor and FT is
a tube-surface temperature heat flux variation factor. In case
of API heaters, the determination of local-maximum radiant
heat flux qm is more complicated task than is presented
above and is described, for example in (Pelini, 2008).
The determination of individual correction factors FC,
FL and FT is a very difficult to quantify accurately, since
there depends on the fuel characteristics, burner design,
flame length, combustion conditions, arrangement and
dimensions of radiant chamber and radiant coils, flue gas
flow inside radiant chamber, etc.
For example, the API Standard 530 (2003) helps the
furnace designers by the simplified recommendations in
form of useful graphs for suitable and quick choice of values
of above mentioned individual correction factors based on
dominant influencing parameters.

Fig. 11 Heavily fouled (coked) 90 curved tube from the

radiant coil of crude vacuum heater.

Fig. 12 Variation of FL correction factor for typical heater

application with classic burners (according to Pelini,
2008).

Average vs. maximum radiant heat flux and radiant heat

flux imbalance in heater
Generally, the refinery fired heaters are designed for
allowable average radiant heat flux (qr). However, radiant
section (radiant chamber) average heat flux is not a good

However, variation of values of individual correction

factors is considerable and heater designer in the design
(and or rating) heater solution stage is not able to accurate
quantify these factors for actual solved heater application. It

www.heatexchanger-fouling.com

12

Heat Exchanger Fouling and Cleaning 2011

can be clearly illustrated on example of variation of tubelongitudinal heat flux factor FL, through Figs. 12 and 13
demonstrate in simplified manner its depending on burner
type and combustion conditions.

radiant chamber heat transfer system (see Figs. 14 and 15)

are not practically applicable in crude refinery heaters due to
thermal-hydrodynamic requirements of oil flowing inside
tubes and current combustion, design, operating and
realisation technical limitations.

Fig. 15 Equalization of heat flux along the flame zone of

cylindrical shape by changing the tube cross-section
(from Wagner, 1997 - adapted for refinery heaters).
Fig. 13 Variation of FL correction factor for heater using
high intensity burners with short flames (according to
Pelini, 2008).
As can be seen from Figs. 12 and 13, the typical locallongitudinal maximum radiant heat flux is obviously in first
one third of radiant chamber height (from chamber bottom).
However, from own experience, the importantly different
cases of heaters are known, where the local-maximum
radiant heat flux was identified at top of radiant chamber as
a result of combination of combustion process fluctuation on
number of burners and changed flue gas flow through
radiant chamber influenced by transverse eddies arise.
Computational prediction of such real situations in
radiant heat flux distribution is very limited, expensive and
requiring highly experienced engineer specialised on
numerical heater modelling. Moreover, such computational
prediction can be performed in some specific cases only,
using Computational Fluid Dynamic (CFD) calculations,
when burners detail design and operation (necessary for
CFD simulations) are known and achievable.

Fig. 14 Equalization of heat flux along the flame zone by

adapting heating surface to flame shape (from Wagner,
1997 - adapted for refinery heaters).
From above discussion it is clear, that theoretically
derived solutions of equalizing of radiant heat flux in whole

Due to above mentioned aspects; the real operating of

refinery fired heaters is accompanied by more or less
significant heat flux imbalances. However, radiant heat flux
imbalances can cause high fouling (coke) formation rates
and high tube metal temperatures.

Fig.

16 Ideal crude vacuum heater performance

characterized by equal pass flow rates and equal outlet
oil bulk temperatures (according to Martin, 1998).

Fig. 17 Typical situation of vacuum crude heater with heat

flux imbalances - equal pass flow rates produce
different outlet oil bulk temperatures (according to
Martin, 1998).

www.heatexchanger-fouling.com

13

Jegla et al. / Design and Operating Aspects

These imbalances reduce unit capacity, lower distillate

yields or require unscheduled and premature heater
shutdowns to remove the coke, all of which affect refinery
profitability, as discussed in detail by Martin (1998).
Operation of multi-pass arrangements of high capacity
heaters influenced by important a heat flux imbalance (see
Figs. 4 and 5) is very difficult. Operators of such heaters can
practically to solve the situation only by heater pass
balancing in relation to measured heated oil heater outlet
(bulk) temperatures. Situation is demonstrated through Figs.
16, 17 and 18 on example of crude vacuum heater operation.
Fig. 16 presents ideal situation of heater operation. Due to
presence of heat flux imbalances to individual passes the
usually real situation in heater operation is presented in Fig.
17. For equalizing of oil bulk temperatures at heater outlet
operators have to make pass balancing. Successful pas
balancing is demonstrated in Fig. 18. However, in some
cases, the pass balancing within allowable control range of
individual pass flow is not successful, as demonstrated in
Fig. 19. Such situations usually lead to heater revamp
requirement.

Fig. 18 Successful vacuum crude heater pass balancing the

flows through each heater pass are balanced to
produce the same oil bulk outlet temperatures
(according to Martin, 1998).

Fig. 19 Unsuccessful vacuum crude heater pass balancing

the flows through each heater pass are unsuccessfully
balanced to produce the same oil bulk outlet
temperatures (according to Martin, 1998).

CONCLUSION
Paper discusses the most important design and
operating parameters influencing fouling process in radiant
tube coils of fired heaters operated in crude oil distillation
plants. Through industrial examples paper documents
disposable tools of heater designers and operators and also
limitations of designers and operators which restrict
possibilities of fouling process mitigation inside radiant tube
coils.
Dedication and acknowledgement
Paper is dedicated to the memory of Dr. Jiri Zachoval.
Financial support of the Ministry of Education, Youth
and Sports of the Czech Republic within the framework of
research plan No. MSM 0021630502 is gratefully
acknowledged.
REFERENCES
API Standard 530, 2003, Calculation of Heater Tube
Thickness in Petroleum Refineries (ISO 13704:2001(E)) (5th
Ed.), API Publishing Services, Washington, DC, USA.
G.T. Atkins, 1962, What to do about high coking rates,
Petro/Chem. Eng., Vol. 34, April 20-25.
G.F. Hewitt, G.L. Shires and T.R. Bott, 1994, Process
Heat Transfer, Begell House, Inc., New York, NY, USA.
Z. Jegla, 2006, The Conceptual Design of Radiant
Chamber and Preliminary Optimization of a Process
Tubular Furnace, Heat Transfer Eng., Vol. 27, pp. 50-57.
Z. Jegla, 2008, Optimum Arrangement of Tube Coil in
Radiation Type of Tubular Furnace, Heat Transfer Eng.,
Vol. 29, pp. 546-555.
Z. Jegla, P. Stehlik and J. Kohoutek, 2005, Combined
Approach Supporting Integrated Furnace Design and
Retrofit, Heat Transfer Eng., Vol. 26, pp. 58-64.
D.Q. Kern and R.E. Seaton, 1959, A theoretical
analysis of thermal surface fouling, Brit. Chem. Eng., Vol.
4, pp. 258-262.
G.R. Martin, 1998, Heat-flux imbalances in fired
heaters cause operating problems, Hydrocarb. Process.,
Vol. 77, pp. 103-109.
W.L. Nelson, 1934, Fouling of Heat Exchangers,
Refining and Natural Gasoline Manufacturer, Vol. 13, pp.
271-276.
R.G. Pelini, 2008, Heat Flux and Film Temperature in
Fired, Thermal-Fluid Heaters, Chem. Eng. New York, Vol.
115, pp. 34-40.
W. Wagner, 1997, Heat Transfer Technique with
Organic Media (2nd Ed.), Dr. Ingo Resch GmbH, Grfelfing,
Germany.
A.P. Watkinson and N. Epstein, 1969, Gas oil fouling
in a sensible heat exchanger, Chem. Eng. Prog. Sym., Ser.
65, pp. 84-90.
D.I. Wilson, G.T. Polley and S.J. Pugh, 2002,
Mitigation of Crude Oil Preheat Train Fouling by Design,
Heat Transfer Eng., Vol. 23, pp. 24-37.
N. Wimpress, 1978, Generalized Method Predicts
Fired-Heater Performance, Chem. Eng. New York, Vol.
85, pp. 95-102.

www.heatexchanger-fouling.com

14