United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S8-83-017 June 1983
Project Summary
Guidelines for the Reduction of
Emissions and Efficiency
Improvement for Refinery Process
Heaters
R. J. Tidona
The manual contains guidelines for
the operation, adjustment, and modifi-
cation of refinery process heaters to
achieve reduced emissions and in-
creased efficiency. Combustion funda-
mentals are summarized and test data
obtained on this contract from previous
subscale and full-scale process heaters
are reviewed. All of the results cited
pertain to natural draft, vertically fired
heaters; however, many of the combus-
tion modification techniques discussed
in the manual are also expected to be
applicable to forced-draft and horizon-
tally fired heaters. Recommended pro-
cedures for adjusting combustion on
process heaters are given. The most
promising combustion modification to
reduce refinery heater emissions and
improve efficiency - staged combus-
tion air lances - is discussed in detail.
Information appropriate to the design,
operation, and maintenance of a staged
combustion air lance system for a pro-
cess heater is presented. Cost effec-
tiveness estimates for such a system
are also given. It is expected that,
through the proper use of a staged
combustion air system and the heater
adjustment procedures described in
the manual, the fuel costs for a typical
refinery heater will be reduced by about
4-5 percent and the NOX emissions
may be reduced to 50-33 percent of
baseline levels.
This Project Summary was developed
by EPA's Industrial Environmental Re-
search Laboratory, Research Triangle
Park. NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Summary and Objectives
These guidelines should assist manu-
facturers and operators of natural draft
process heaters in the petroleum refining
industry in achieving maximum efficiency
with minimum pollutant emission. The
means for achieving such optimized heater
performance have been investigated over
a period of several years through an ex-
haustive field test program encompassing
both sub- and pilot-scale heater tests.
These methods include operating variable
optimization (e.g., amount of excess air
and air distribution and mixing patterns)
applied either alone or in combination with
hardware modifications such as.-staged
combustion air lances, altered fuel injec-
tion geometry, Iow-N0x burners, flue gas
recirculation, and steam injection.
The guidelines focus on the use of
operating variable changes and the imple-
mentation of the most promising of the
hardware modifications investigated for
process heaters - staged combustion air
lances. Combustion adjustments com-
bined with the staged combustion modifi-
cation have been successfully applied to a
natural draft process heater for 30 days.
Their application resulted in significant
increases in the thermal efficiency of the
heater and NOX emission reductions of 60
percent or more.
Methods for measuring emissions, opti-
mizing excess air level, and establishing
the proper burner register settings are
discussed. Also covered are suggested
-------
procedures for monitoring heater perform-
ance once the unit has been properly
adjusted These topics should be of par-
ticular interest to heater operators.
The design, operation, and maintenance
of a staged combustion air system which
uses stainless steel pipes (or "lances") to
introduce combustion air downstream of
the burner injection plane, as well as the
costs and cost effectiveness of this system,
are also discussed. This information should
be of interest to both manufacturers and
operators.
The guidelines include those for com-
bustion modifications which apply to new
or existing natural draft heaters. The test
results described are given for a vertical
cylindrical firebox; however, the combus-
tion modifications are expected to be appli-
cable to vertical rectangular and horizontal-
ly fired heaters as well.
Although the tests were conducted on a
crude oil heater, the modifications and
adjustments discussed are expected to be
applicable to other heater types (e.g., crack-
ing heaters or reformer heaters) since the
process tube wall temperature distribution
does not appear to be significantly altered
by these changes in heater operation. In
addition, the use of burner adjustments
and staged combustion air is not limited, in
principle, to natural draft heaters. Mechan-
ically drafted units, both with and without
combustion air preheat are expected to
lend themselves well to such modifications
based on previous experience with indus-
trial and utility boilers. In a particular
example, staged air lances applied to a
mechanically drafted industrial boiler re-
sulted in a NOX emission reduction of 42
percent (while increasing the unit's thermal
efficiency by 0.5-1 percent) (Ref. 1).
Process Heater Test Program
Tests were conducted on a natural-
draft, vertical cylindrical crude heater con-
taining six John Zink burners, capable of
combined fuel firing. Initial tests were
conducted to determine heater perform-
ance over a range of operating variables
such as excess air, load, and air register
settings for various fuel types. In these
tests, hardware was not modified to affect
combustion. For some tests firing refinery
gas, steam was injected through the oil-
atomizing system, and its effect on NOX
emissions assessed.
Once the performance of the heater was
documented over its normal range of op-
erating parameters, staged combustion air
was implemented A prototype system,
constructed largely of polyvinyl chloride
pipe, fittings, and valves with 24 stainless
steel lances (4 per burner), was built The
system was designed to provide max-
imum flexibility and flow control for min-
imum cost The heater was then re-
evaluated over the same ranges of oper-
ating parameters. During testing of the
staged air system, several additional pa-
rameters were varied, including burner
stoichiometric ratios, staged air insertion
height and staged air lance orientation. An
optimum Iow-N0x -operating condition
was defined as the configuration at which
the lowest NOX concentrations were ob-
tained while still permitting stable heater
operation without a significant increase in
CO emissions. This condition was defined
for gaseous fuel and for a 50/50 mixture
of No. 6 oil and gas.
A 30-day test was then conducted with
the staged air system in continuous oper-
ation firing refinery gas at the optimum
Iow-N0x condition. System performance
and durability were evaluated as well as
the ability to maintain steady heater oper-
ation at the Iow-N0x condition. At the
completion of the long-term test a per-
manent system was designed. This sys-
tem was intended to be suitable for a
typical furnace of the same type as that
tested in this program.
Table 1 summarizes significant results
of the test program. A detailed final report
covering the research activities at this
heater is being published by EPA (Ref. 2).
The cost effectiveness, based on the
permanent system design, of the staged
combustion modification was evaluated
for the gaseous fuel and for the com-
bination fuel, and also at two levels of
stack oxygen. The cost effectiveness is
shown in Table 2 for three heater sizes.
The values for lowered excess air (LEA)
and for staged combustion air (SCA) plus
LEA modification on the largest heater arc
negative, indicating a cost savings in addi-
tion to a NOX reductioa This is a desirable
situation; however, one must realize that i
modification which saves money yet pro-
duces only a small NOX reduction will have
a large negative cost effectiveness ratio
This may not be the most desirable modi-
fication if the level of NOX reduction is
lower than is necessary to achieve com-
pliance with regulations or to offset future
emissions from plant expansions.
Combustion Adjustments to
Process Heaters
To establish the optimum operating con-
ditions on a natural draft process heatei
(ag., excess air level, air register settings,
and stack damper position), one should be
able to accurately measure stack oxygen,
NOX and CO concentrations, stack tem-
perature, tube skin temperatures, and fur-
nace draft (preferably at a minimum ol
three elevations in the heater-at the fire-
box, the convection section inlet or bridge-
wall, and the stack). These measurements
allow the operator to maximize heatei
efficiency by reducing the excess ait
(stack oxygen) to the minimum amount
which can be maintained without causing
combustibles (CO) to appear in the stack
gases. Reducing the excess air will gen-
erally affect the stack gas temperature, as
well. A change in stack gas temperature
also affect heater thermal efficiency and,
therefore, is an important parameter to be
measured so that the efficiency improve-
ment may be quantified.
Lowering the excess air to this min-
imum value also lowers NOX emissions
since it reduces the amount of atmos-
pheric nitrogen contacting the flame. Mea-
Table 1. Summary of Results of Combustion Modification Tests on a Process Heater
Baseline NO,
Heat
Input
MW
10.4
10.4
9.0
13.7
13.2
13.1
13.1
Fuel Type
Fief, gas9
Ref. gas
Ref. gas
Ref. gas
No. 6 Oil +
Ref. Gas
No. 6 Oil +
Ref. Gas
No. 6 Oil +
Ref. Gas
ng/J
66
66
54
61
114
115
115
ppm
dry at
3%02
125
125
105
120
212
214
214
NOX Reduc.
from Baseline
%
60
71
15
2.5
34
53
28
Change in
Fuel Cons.
%
-0.2
-4.8
-2.8
+2.2
-0.6
-4.8
-3.0
Combustion
Modification
SCAkc
SCA + LEA *•"
LEA
Steam inj.
SCA
SCA + LEA
LEA
8 Ref. gas = Refinery gas.
b SCA = Staged Combustion Air.
c Based on results of a 30-day test.
d LEA = Lowered Excess Air.
-------
suring the NOX concentration (N0,p NO
+ N02 concentrations) is thus important
to quantify the amount of emission reduc-
tion obtained by adjusting the heater.
Also important in a natural draft heater
when making air adjustments is the draft
profile. The heater should be maintained at
a negative pressure throughout to avoid
both flashback at the burners and the
escape of toxic and flammable combus-
tion gases to the surroundings through
cracks, ports, etc.
The operating condition which results in
the lowest stack oxygen level and stack
temperature (without affecting the heater
feed rate and while maintaining negative
pressures throughout the heater) must
also result in radiant section tube skin
temperatures which are well within the
limitations of the tube materials and/or
the process constraints. Significant local
increases in tube skin temperatures could
cause coking of the process fluid or tube
failure. It is thus essential that tube skin
temperatures, especially in the vicinity of
the flame, be monitored throughout any
adjustment procedure. Changes in flame
appearance can indicate changes in the
tube temperatures caused by a combus-
tion adjustment
Once the condition is reached at which
stack 02, NOX, and stack temperature
have been optimized with respect to CO
emissions, heater draft profile, and tube
skin temperatures, the proper operating
condition has been reached for that par-
ticular heater feed rate. It then remains to
determine the optimum operating con-
ditions at several other feed rates covering
the normal operating range of the heater.
The NOX emissions resulting from a
heater at various stack oxygen levels at
varying process rates are shown in Figure
1 for gas firing. These data were collected
from the test site discussed earlier. Al-
though baseline NOX emissions are higher
for oil fuels, the general shapes of the
curves are similar for both oil and gas
fuels. These characteristic curves should
be developed for each individual heater in
order to compare results each time the
unit is adjusted. Ideally, for constant fuel
and feed composition, each time the
heater is returned to the same set of
operating conditions (i.e., fuel and feed
flow rates, feed inlet and outlet tempera-
tures and pressures, fuel temperature and
pressure, stack oxygen, register settings
and furnace draft, and combustion air
temperature and humidity), the same NOX
emissions will result Naturally, it is never
possible to duplicate a set of operating
conditions exactly; however, a discrepancy
of more than about 5 percent in NOX
concentration when attempting to repro-
duce a past test condition usually indicates
that some operating parameter has not
been set correctly.
The guidelines contain recommenda-
tions for stack sampling instruments and
techniques useful in implementing a pro-
gram such as that just described. In addi-
tion, helpful tips on conducting spot-
checks on combustion efficiency and peri-
odic comprehensive performance evalua-
tions are also offered.
Staged Combustion Air
Modification
The principle of staged combustion is
not new. It has been used as a NOX
reduction technique on utility and indus-
trial boilers and steam generators for sev-
eral years. The recent interest in applying
NOX reduction technology (which permits
safe reliable operation at minimum addi-
tional cost) to other industrial combustion
devices motivated the design of a staged
air system applicable to many refinery
process heaters. Process heaters, as well
as several other industrial devices, had
been determined to contribute significant-
ly to nationwide NOX emissions in addition
to consuming large amounts of energy in
the form of fossil fuels.
A staged combustion air lance system
flow schematic is presented in Figure 2.
Basically, the system provides combus-
tion air downstream of the fuel injection
plane of each burner by means of a forced
draft blower. The air is injected into the
firebox through pipes or "lances" which
penetrate either the floor of a vertically
fired unit or the walls of a horizontally fired
unit The introduction of air downstream
of the burners permits fuel-rich operation
Table 2. Cost Effectiveness ($/Mg) of Combustion Modifications Applied to Three Natural
Draft Process Heater Sizes
Heater Size
Modification'
16.1 MW
29.3 MW
147 MW
SCAat4%02
SCA + LEA
LEA Firing Gas
LEA Firing Oil and Gas
2636
1089
(5843)
(1459)
2362
700
(6853)
(1712)
1934
(40?
(4813)
(2189)
aSCA = Staged Combustion Air, LEA = Lowered Excess Air.
b ( ) = savings.
110
100
O 90
80
g 70
60
I
1
I
Fuel: Natural gas +
Fuel gas
O = 18.2 kg/s of crude
__ (11.000 bbl/day)
LI =23.1 kg/s
. (14.000 bbl/day)
A = 13.2 kg/s
(8,000 bbl/day)
I I I
123456789
Stack Oxygen, %,dry
Figure 1. NO emissions at three different loads as a function of stack oxygen for a natural draft
process heater firing a natural gas/refinery gas mixture.
-------
316L stainless steel
lances. 4 per burner
(typical for all burners}
Heater
45-deg. weld ells
(typical for all lances)
Calibrated Orifice Manifold
y ^
L*
. _ Flexible Tubing
\f ii ^~ Reducer
-------
R. J. Tidona is with KVB, Inc., Irvine, CA 92714.
Robert E. Hall is the EPA Project Officer (see below).
The complete report, entitled "Guidelines for the Reduction of Emissions and
Efficiency Improvement for Refinery Process Heaters," (Order No. PB 83-206
995; Cost: $11.50, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
United States Center for Environmental Research Feesfpaid
Environmental Protection Information Pnuirnnm«nt=i
Agency Cincinnati OH 45268 Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
0000329
------- |