EPA-450/2-78-047
OAQPS No. 1.2-116
Control of Volatile Organic
Emissions from Petroleum Liquid
Storage in External Floating
Roof Tanks
Emission Standards and Engineering Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
December 1978
-------
OAQPS GUIDELINE SERIES
The guideline series of reports is being issued by the Office of Air Quality Planning and Standards (OAQPS) to
provide information to state and local air pollution control agencies; for example, to provide guidance on the
acquisition and processing of air quality data and on the planning and analysis requisite for the maintenance of
air quality. Reports published in this series will be available - as supplies permit -from the Library Services Office
(MD35). U.S. Environmental Protection Agency. Research Triangle Park, North Carolina 2771 1; or, for a nominal
fee, from the National Technical Information Service, 5285 Port Royal Road, Springfield, Virginia 22161.
Publication No. EPA-450/2-78-047
OAQPS No. 1.2-116
-------
TABLE OF CONTENTS
Page
Chapter 1.0 Introduction 1-1
1.1 Need to Regulate 1-1
Chapter 2.0 Sources and Types of Emissions ..... 2-1
2.1 External Floating Roof Tanks 2-1
2.2 Primary Seals ........... 2-3
2.2.1 Mechanical Shoe Seal 2-4
2.2.2 Resilient Foam Seal ....... 2-4
2.2.3 Liquid-Filled Seal ....... 2-6
2.3 References 2-7
Chapter 3.0 Control Technology 3-1
3.1 Rim-Mounted Secondary Seal . 3-1
3.2 Wind Induced Emissions 3-3
3.2.1 Shoe Seals on Welded Tanks . . . . . 3-3
3.2.2 Liquid-Mounted Resilient Foam and
Liquid-Filled Seals on Welded Tanks . . 3-5
3.2.3 Vapor-Mounted Resilient Foam Seal on
Welded Tank .......... 3-5
3.2.4 Riveted Tanks 3-7
3.3 References 3-10
Chapter 4.0 Cost Analysis 4-1
4.1 Introduction . 4-1
4.1.1 Purpose 4-1
4.1.2 Scope . . . 4-.1
4.1.3 Use of Model Storage Tanks 4-1
iii
-------
Chapter
4.1.4 Bases for Capital and Annualized
Page
4-5 Page
4.2 Control of Emissions from External Floating
Roof Tanks 4-6
4.2.1 Model Cost Parameters 4-6
4.2.2 Control Costs 4-6
4.3 Cost Effectiveness 4-8
4.4 Economics of Scale 4-13
4.5 References '. 4-15
Chapter 5.0 Recommended Regulations, Compliance Test
Method and Record Keeping ......... 5-1
5.1 Recommended Regulations 5-1
5.2 Compliance Test Method 5-3
5.3 Monitoring and Record Keeping 5-3
Appendix A Selection of Experimental Tests for Wind
Induced Emission Calculations A-l
A. Methodology of Selection . . . . . . . A-l
B. Selection of Welded Tank with Shoe Seal and
Rim-Mounted Secondary Seal A-2
C. Welded .Tank with Shoe Seal and Shoe-
Mounted Secondary Seal A-3
D. Riveted Tank with a Shoe Seal and Rim-
Mounted Secondary A-3
E. Resilient Foam Seal Mounted in Rim Vapor
Space of Welded Tank ........ A-6
References A-8
Appendix B Calculation of Wind Induced Emissions from
Experimental Tests B-rl
A. Equation for Wind Induced Emissions . . . B-l
B. Ep for Test Tank ......... B-l
iv
-------
Page
Appendix C. Emission Calculations B-3
C.I Primary Seal With and Without Rim-
Mounted Secondary B-3
C.2 Shoe Seal with Shoe-Mounted Secondary . B-3
References B-6
-------
LIST OF TABLES
Table 4-1
Table 4-2
Table 4-3
Table 4-4
Table 4-5
t
Appendix A
Table A-l
A-2
A-3
Appendix B
Table B-l
B-2
B-3
Technical Parameters Used in Developing
Gasoline Storage Tank Control Costs . . .
Cost Parameters Used in Computing Control
Costs
Control Cost Estimates for Model Existing
Cost Effectiveness of Controlling Floating
Roof Storage Tanks ....
Factor for Estimating the Cost Effectiveness
of Controlling Crude Oil Storage Tanks . . .
Seal Gap Area in Inspected Welded Tanks with
Shoe Seal Compared to CBI Experimental Tests .
Seal Gap Area in Inspected Riveted Tanks with
Shoe Seals Compared to CBI Experimental Tests .
Seal Gap Area in Inspected Welded Tank with
Non-Metallic Seals Compared to CBI
Experimental Tests
Ep versus Wind Speed j
Emissions from 100 ft 0 Gasoline Tank ....
Emissions from 100 ft 0 Welded Gasoline Tank
with Shoe Mounted Secondary Seal
Page
4-2
4-7
4-9
4-10
4-12
A-4
A-5
A-7
B-2
B-4
B-5
vi
>
-------
LIST OF FIGURES
Page
Figure 2-1 External Floating Roof Tank (pontoon type) . . 2-2
Figure 2-2 Primary Seals ....... 2-5
Figure 3-1 Rim-Mounted Secondary Seals 3-2
Figure 3-2 Emissions from 100 ft Diameter Welded Gasoline
Tank with Primary Shoe Seal at 16.1 kph (10 mph)
Average Wind Speed 3-4
Figure 3-3 Emissions from 100 ft Diameter Welded Gasoline
Tank with Primary Foam Seal at .16.1 kph (10 mph)
Average Wind Speed 3-6
Figure 3-4 Emissions from 100 ft Diameter Riveted Gasoline
Tank with Primary Shoe Seal at 16.1 kph (10 mph)
Average Wind Speed . . . . . . . . . . . 3-8
-------
ABBREVIATIONS AND CONVERSION FACTORS
EPA policy is to express all measurements in Agency documents in metric
units. Dual units are sometimes given in the text for clarity. Listed below
are abbreviations and conversion factors for English equivalents of metric
units. Frequently used measurements are also presented in dual units below
for the reader's convenience., '.'...,
METRIC UNIT
kilogram (kg)
metric ton (m ton) or
megagram (Mg)
kilometer (km)
kilometers per hour'(kph)
meter (m)
centimeter (cm)
liter (1)
Pascal (Pa)
kiloPascals (kPa)
ALTERNATE UNIT
pound (Ib)
ton
pound (Ib)
ton
mile (mi)
miles per hour (mph)
foot (ft)
inch (in)
gallon (gal)
barrel (bbl)
atmospheres (atm)
pounds per square
inch (psi)
atmospheres (atm)
pounds per square
inch (psi)
FREQUENTLY USED MEASUREMENTS
CONVERSION.
kg x 2.205 = Ibs
kg x 1.1 x TO'3 = tons
Mg x 2205 = Ibs
Mg x 1.102 = tons
km x 0.621 = mi
kph x 0.621 = mph
m x 3.281 = ft
cm x 0.394 = in
1 x 0.264 = gal
1 x 6.3 x 10-3 = bbl
Pa x 9.9 x 10~6 = atm
Pa x 6.7 x 10"7 = psi
kPa x 9.9 x 10~3 = atm
kPa x 0.145 = psi
1,600,000 1
150,000 1
422,000 gal
40,000 gal
10,000 bbl
950 bbl
10.5 kPa
13.8 kPa
27.6 kPa
41.4 kPa
69.0 kPa
Oi
1.52 psi
2.0 psi
4 psi
6 psi
10 psi
9.7 kph ^ 6 mph
16.1 kph <\, 10 mph
22.5 kph
-------
Definition of Terms
i-
A. Condensate means hydrocarbon liquid separated from natural gas which
condenses due to changes in the temperature and/or pressure and remains
licjuid at standard conditions.
B. Cost Effectiveness - Cost (or credit) per megagram of controlled
[
emissions. Given in general by: (recovered petroleum liquid value -
net annual control system cost) v (megagrams of controlled emissions) =
cost (or credit) /Mg controlled emissions,
C. Crude oil means a naturally occurring mixture consisting of
hydrocarbons and/or sulfur, nitrogen and/or oxygen derivatives of
hydrocarbons and which is a liquid in the reservoir and at standard
conditions.
!
D. Custody transfer means the transfer of produced crude oil and/or
condensate, after processing and/or treating in the producing
operations, from storage tanks or automatic transfer facilities to
pipelines or any other forms of transportation.
E. External floating roof means a storage vessel cover in an open top
tank consisting of a double deck or pontoon single deck which.rests
upon and is supported by the petroleum liquid being contained and is
equipped with a closure seal or seals to close the space between the
roof edge and tank shell.
F. Internal floating roof means a cover or roof in a fixed roof tank which
rests upon or is floated upon the petroleum liquid being contained, and
is equipped with a closure seal or seals to close the space between the
roof edge and tank shell.
IX
-------
6. Liquid-inounted means a primary seal mounted so the bottom of the seal
covers the liquid surface between the tank shell and the floating roof.
H. Vapor-mounted means a primary seal mounted so there is an annular vapor
space underneath the seal. The annular vapor space is bounded by the
bottom of the primary seal, the tank shell, the liquid surface, and
the floating roof.
I. Petroleum liquids means crude oil, condensate, and any finished or
intermediate products manufactured or extracted in a petroleum refinery.
0. True vapor pressure means the equilibrium partial pressure exerted
_ t
by a petroleum liquid as determined in accordance with methods
described in American Petroleum Institute (API) Bulletin 2517,
Evaporation Loss from Floating Roof Tanks, 1962. The API procedure
may not be applicable to some high viscosity or high pour crudes.
Available estimates of true vapor pressure may be used in special
cases such as'these.
K. Volatile Organic Compounds (VOC) means compounds which under favorable
conditions may participate in photochemical reactions to form oxidants.
>
x
-------
1.0 INTRODUCTION
This document is related to the control of volatile organic compounds
(VOC) from the storage of petroleum liquids in external floating roof tanks.
Methodology described in this document represents the presumptive norm
or reasonably available control technology (RACT) that can be applied to
existing external floating roof storage tanks. RACT is defined as the lowest
emission limit that a particular source is capable of meeting by the application
of control technology that is reasonably available considering technological
and economic feasibility. It may require technology that has been applied
to similar, but not necessarily identical, source categories. It is not
intended that extensive research and development be conducted before a
given control technology can be applied to the source. This does not,
however, preclude requiring a short-term evaluation program to permit the
application of a given technology to a particular source. The latter effort is
an appropriate technology-forcing aspect of RACT.
1.1 NEED TO REGULATE
Control techniques guidelines concerning RACT are being prepared for those
industries that emit significant quantities of air pollutants in areas of the
country where National Ambient Air Quality Standards (NAAQS) are not being
1-1
-------
attained. Storage tanks for petroleum liquids are a significant source
of VOC. A control techniques guideline (CTG) for storage of petroleum liquids
in fixed roof tanks (EPA-450/2-77-036) was published in December 1977. RACT
for fixed roof tanks was defined as the retrofit with internal floating roofs
or equivalent.
The following recommended control measures apply to external floating
roof tanks (EFRT) larger than 150,000 liters (950 bbls) storing petroleum
liquids. They do not 'apply to fixed roof or tanks with or without internal :
floating roofs, nor do they apply to small production tanks. In general,
RACT for external floating roof tanks (EFRT) is defined as follows:
(1) A welded EFRT equipped with primary metallic shoe or liquid-
mounted seals is required to retrofit with a rinwiiounted secondary seal
if the TVP of the stored "liquid exceeds 27.6 kPa (-4 psi).
(2) A welded or riveted EFRT equipped with primary vapor-mounted
seals is required to retrofit with a rim-mounted secondary if the TVP of
the stored liquid exceeds 10.5 kPa (1.5 psi).
(3) A riveted EFRT equipped with primary metallic shoe or liquid-
mounted seals is also required to retrofit with a rim-mounted secondary if
the TVP of the stored liquid exceeds 10.5 kPa (1.5 psi).
Specific recommendations-for regulations, including exemptions, are presented
in Chapter 5.0.
Estimated emissions from the affected EFRT's.during 1978 were 65,000
megagrams/year (71,630 tons/yr). The proposed recommendations would reduce
these emissions to 30,000 megagrams/year (33,060 tons/yr).
The emission estimates used in this document were calculated from data
obtained by Chicago Bridge and Iron Company (CBI) on a 6.1 m (20 ft)
1-2
-------
diameter test tank. Data obtained by Pittsburgh-Des Moines Steel Company (PDM)
on a 10.7 m (35 ft) diameter test tank were used to verify RACT for liquid-
mounted seals which are liquid or foam filled. An American Petroleum Institute
(API) emission test program, scheduled for completion in 1979, is expected
to provide verification of the validity of the scale-up techniques used
herein.
Cost effectiveness of retrofitting rim-mounted secondary seals to EFRT's
is dependent on tank size, product type, product value, average wind speed
and other factors. For example, the installed capital cost for retrofitting
a rim-mounted secondary seal to a 30.5 m (100 ft) diameter welded tank
equipped with a primary shoe seal is about $17,000. The net annual cost
after credit for recovered product is $3,140 when storing gasoline at a
TVP of 41.4 kPa (6 psi) and an average wind speed of 16.1 kph (10 mph). A
welded tank having a vapor-mounted primary seal and a riveted tank having
a primary metallic shoe seal can be retrofitted with a rim-mounted secondary
seal for the same capital cost. However, in these two cases under the same
storage conditions the emission reductions are larger and the net annual
cost is $1930 for the welded tank with the vapor-mounted seal and $1750 for
the riveted tank with the shoe seal. The cost effectiveness for the above
three cases is $373, $117, and $99 per megagram of emissions controlled,
respectively. At lower wind speeds and vapor pressures, the cost effective-
nesk would be higher. At higher wind speeds and vapor pressures, the cost
i :
effectiveness would be lower.
1-3
-------
-------
2.0 SOURCES AND TYPES OF EMISSIONS
There are an estimated 13,800 internal and external floating roof tanks
storing petroleum liquids at refineries, terminals, tank farms and along
pipelines. Of these, 10,700 are storing liquids whose vapor pressures equal
or exceed 10.5 kPa. Data are not available to establish how many of these are
external floating roof tanks.
2.1 EXTERNAL FLOATING ROOF TANKS
An external floating roof tank consists of a steel cylindrical ,
shell equipped with a deck or roof which floats on the surface of the stored
i
liquid, rising and falling with the liquid level (Figure 2-1). The liquid
surface is completely covered by the floating roof except in the small annular
space between the roof and the shell. A seal attached to the roof contacts
the tank wall and covers the remaining area. The seal slides against the tank
wall as the roof is raised or lowered. The primary route of VOC emissions is
by this seal.
When a commercial fit between the seal and the tank wall is maintained,
most losses by the seal are attributable to the wind. ' »4>5'6 Wind induced
losses occur when air flow across the tank creates pressure differences around
the floating roof, causing air to flow into the annular vapor space* on the lee-
ward side and air plus VOC to flow out on the windward side. Improper or loose fit
*
Unless the primary seal is liquid-mounted, the vapor space bounded by the
sliding seal, wall, roof, and liquid surface defines an annular vapor space.
2-1
-------
03
a.
4-1
o
o
+-»
o
O3
C
X
LLI
CM
0}
L.
g)
u.
2-2
-------
of the seal creates gaps or openings between the seal and the tank wall.
These gaps expose the liquid surface directly to the wind and sun, which
combine to increase emissions. The wind flows across the tank, scouring
the vapor space and sweeping away VOC. In addition, leakage through holes
in the envelope (the fabric cover that is used to bridge the space between
the seal and the floating roof) or around the envelope attachment bolts can
be a significant source of loss from shoe seals.
Other causes of emissions are: (1) release of dissolved air saturated
with VOC because of barometric pressure changes; (2) solar heating of liquid
in the rim space which increases liquid vapor pressure and VOC migration;
(3) evaporation of the liquid which clings to the tank wall when the tank
is being emptied (wetting losses)7; (4) breathing of the vapor space due
to changes in the ambient temperature or barometric pressure; or (5) changes
in the bulk liquid temperature. Wind-induced losses are larger than all of
these.
2.2 PRIMARY SEALS
There are basically three types of primary seals; mechanical shoe seals,
resilient foam seals, and liquid-filled seals. Although there are other
designs, these three comprise the vast majority of primary seals in use today.
A weather guard is often installed over primary seals to protect the
seals from deterioration caused by dust, rain or sunlight. Typically, a
weather guard is an arrangement of overlapping thin metal sheets pivoted
from the floating roof to ride against the tank wall. This helps protect
the product from contamination, but its effect on gaps and hence wind-induced
emissions is variable. Some weather guard designs could do little to curb
emissions where other tighter designs may be reasonably effective over certain
types of primary or secondary seals. Because of the uncertainties associated
2-3
-------
with emissions control, the weather guard is not usually considered as
flt
effective an emission control device as a secondary seal. ^
2.2.1 Mechanical Shoe Seal
The mechanical shoe seal is characterized by a 75 to 130 cm (30" to 51")
high metal sheet (the "shoe") held against the vertical tank wall (Figure 2-2a).
The shoe is connected by braces to the floating roof and is held tight against
the wall by springs or weighted levers. A flexible coated fabric (the
"envelope") is suspended from the shoe seal to the floating roof to close
the annular space between the roof and the primary seal.
Emissions from the mechanical shoe seal occur from the exposed liquid
surface in the gap spaces between the shoe and the tank wall, and
through openings in the envelope or shoe. Close fitting primary shoe seals
effectively reduce emissions from the liquid surface in the gap space, as
o
do shoe-mounted secondary seals (Figure 2-2a). Shoe-mounted secondary f|
^y
seals are discussed in Chapter 3.0. Emissions are also affected by the
envelope and shoe conditions. Holes, tears, or other openings in the
envelope or shoe allow direct communication between the annular vapor space
and the atmosphere. Through these openings, the wind can scour the vapor
space, exiting with VOC laden vapors.
2.2.2 Resilient Foam Seal
As illustrated in Figure 2-2c,d, resilient foam primary seals fill the
annular space between the floating roof and tank wall with a continuous
compressible foam log encased in a protective tube. The resiliency of the
foam log allows the seal to adapt itself to some imperfections in tank
dimensions and even to fill or partially fill some protrusions. The foam
log may be vapor-mounted (Figure 2-2c) or liquid-mounted (Figure 2-2d)
O
2-4
-------
SECONDARY SEAL
(WIPER TYPE)
METALLIC WEATHER GUARD
^FLOATING ROOF
VAPOR SPACE!
"-FLOATING ROOF
-LIQUID FILLED TUBE
a. Metallic shoe seal with shoe-mounted
secondary.
b. Liquid-filled seal with weather guard.
METALLIC WEATHER GUARD
METALLIC WEATHER GUARD
c. Resilient foam seal with weather guard
(vapor-mounted)..
d. Resilient foam seal with weather guard
(liquid-mounted).
Figure 2-2 Primary Seals
2-5
-------
When a foam seal is vapor-mounted emissions can be much higher than when
liquid-mounted. A gap between a vapor-mounted foam seal and the wall allows
direct communication between the atmosphere and the vapor space bounded by
the seal, the roof, the tank wall, and the product liquid.
When a foam seal is liquid-mounted, the vapor space is eliminated and
losses are comparable in magnitude to those for the shoe seal.
2.2.3 Liquid-Filled Seal
A liquid-filled seal may be a tough fabric band or envelope filled with
a liquid, or it may be a 20-25 cm (8-10") diameter flexible polymeric tube
filled with a liquid and sheathed with a tough fabric scuff band (Figure 2-2b).
The liquid is commonly a petroleum distillate or other liquid that would not
contaminate the stored product if the tube ruptured. Liquid-filled seals are
mounted on the product liquid surface with no vapor space. They are usually
protected by a weatherguard.
Losses from tanks equipped with liquid-mounted liquid-filled primary
g
seals are comparable in magnitude to shoe seals and liquid-mounted foam seals.
2-6
-------
2.3 REFERENCES
1. Evaluation of Hydrocarbon Emissions from Petroleum Liquid Storage,
EPA-450/3-78-012, March, 1978.
2. SOHIO/CBI Floating Roof Emission Program, Interim Report, October 7, 1976.
3. SOHIO/CBI Floating Roof Tank Emission Program, Final Report,
November, 1976.
4. Western Oil and Gas Association, Metallic Sealing Ring Emission
Test Program, Interim Report, Chicago Bridge & Iron Company, January, 1977.
5. Western Oil and Gas Association, Metallic Sealing Ring Emission
Test Program, Final Report, Chicago Bridge & Iron Company, March, 1977.
6. Western Oil and Gas Association, Metallic Sealing Ring Emission
Test Program, Supplemental Report, Chicago Bridge & Iron Company, June, 1977.
7. SOHIO/CBI Floating Roof Tank Emission Test Program. Supplemental
Report, Chicago Bridge & Iron Company, February 15, 1977.
8. Floating Roof Seal Development - Emission Test Measurement on
Proposed CBI Wiper-Type Secondary Seal for SR-1 Seals, Chicago Bridge & Iron
Company, February 23, 1977.
R
9. Measurement of Emissions from a Tube Seal Equipped Floating Roof
Tank, Pittsburgh-Des Moines Steel Company, October 9, 1978.
2-7
-------
-------
3.0 CONTROL TECHNOLOGY
Recommended control technology for existing external floating roof tanks
with primary foam, liquid-filled, and metallic shoe seals is retrofitting
with a rim-mounted secondary seal. A rim-mounted secondary seal is defined
as a continuous device extending from the floating roof to the tank wall,
and installed over the primary seal.
3.1 RIM-MOUNTED SECONDARY SEAL
A rim-mounted secondary seal is continuous and extends from the floating
roof to the tank wall, covering the entire primary seal. Installed over a
mechanical shoe seal, this secondary seal can effectively control VOC that
escape from the small vapor space between the shoe and the wall, and through
any openings or tears in the seal envelope which would permit direct
communication of the seal system vapor space with the atmosphere (see
Figure 3.1.a).
Rim-mounted secondary seals are effective in controlling emissions from
the liquid and vapor-mounted primary seals shown in Figure 3.I.1)2'3'4*5
The secondary seals can often be rendered inoperative by cooling and
i
hardening of waxy, heavy pour crude oils. These crudes cause a deposit on
the tank wall which is scraped onto the roof when the tank is worked,
damaging the secondary seal.
Another type of secondary seal that is commonly installed on external
floating roof tanks is a shoe-mounted secondary seal. A shoe-mounted
3-1
-------
TANK
WALL
RIM-MOUNTED i
. SECONDARY SEALJ
TANK
WALL!
RIM-MOUNTED SECONDARY SEAL
a. Shoe seal with rim-mounted secondary seal.
t
b. Liquid-filled seal with rim-mounted
secondary seal.
RIM-MOUNTED
.SECONDARY SEAL
^FLOATING ROOF*
RIM-MOUNTED
SECONDARY SEAL
c. Resilient foam seal (yajpoF-mounted) J
with rim-moiinted secondary
d. Resilient foam seal (liquid-mounted).
with-fim-mounted secondary seal
Figure 3-1 Rim-Mounted Secondary Seals
3-2
O
-------
seal extends from the top of the shoe to the tank wall (see Figure 2-2a).
Shoe-mounted seals do not provide protection against VOC leakage through
the envelope. Holes, gaps, tears, or other defects in the envelope can
allow direct communication between the saturated vapor under the envelope
and the atmosphere and the wind can enter this space through envelope
defects, flow around the circumference and exit with saturated or near
saturated vapors.
3.2 WIND INDUCED EMISSIONS*
Three 30.5 m (100 ft) diameter tanks were chosen as base cases for
emission calculations; a welded tank with a primary shoe seal, a welded tank
with a vapor-mounted resilient foam seal, and a riveted tank with a primary
shoe seal. The emission reduction that would occur from installing a
secondary seal over each of these base cases is discussed below.
3.2.1 Shoe Seals on Welded Tanks
When storing a 27.6 kPa (4 psi) vapor pressure product, a rim-mounted
secondary seal installed over a primary shoe seal reduces emissions from 11.2
megagrams per year (Figure 3-2a) to 2.8 megagrams per year (Figure 3-2d).
A shoe-mounted seal installed on a primary shoe seal reduces emissions from
11.2 megagrams per year to 5.3 megagrams per year (Figure 3.2c) for the
same product. Emission reductions for various seal configurations are best
illustrated over a range of product vapor pressures by Figure 3-2.
The amount of emissions curbed for each progressively stricter control
option increases as the TVP of the stored liquid increases. For example, by
subtracting (d) from (a) in Figure 3-2, the emission reduction for installing
* The emission rates throughout this chapter are calculated for a 30.5 m
(100 ft) diameter tank storing 41.4 kPa (6 psi) vapor pressure gasoline with
an average wind speed of 16.1 kph (10 mph). The average vapor molecular
weight was assumed to be 65, typical for gasoline. Emission rates may be
scaled according to Appendix B.
3-3
-------
voc
Emissions
(megagrams/yr)
30
28
26
24
22
20 -
18 .
16 -
14 -
12 .
10
8
6
4
2
a
b
c
d
- Primary shoe seal, no secondary seal
- Primary shoe seal with gapped shoe-mounted
secondary seal
- Primary shoe seal with tight shoe-mounted
secondary seal
- Primary shoe seal with tight rim-mounted
secondary seal
11 psi
10
20
30
40
50
To"
kPa
True Vapor Pressure of Stored Product
Figure 3-2. EMISSIONS FROM 30.5 m (100 ft) DIAMETER WELDED GASOLINE
H°E SEAL A" 16'] kph
3-4
-------
a rim-mounted secondary seal over a primary shoe seal is only about 5.1
megagrams per year for scoring a 27.6 kPa (4 psi) product, but the re-
duction increases to 18 megagrams per year if the stored liquid has a TVP
of 69 kPa lib psi). "' "~ "
Emissions from a tank equipped with a shoe seal and shoe-mounted
secondary seal storing 41.4 kPa (6 psi) vapor pressure product are 5.3
megagrams per year. Retrofitting this tank with a rim-mounted secondary
seal would reduce emissions by only 2.5 megagrams per year. Thus, tanks
now equipped with a primary shoe seal and a shoe-mounted seal are controlled
reasonably well and need not be retrofitted with a rim-mounted secondary
seal. Nevertheless, the susceptibility of this system to envelope leaks
and gaps make good inspection and maintenance practices imperative. A shoe
seal without any secondary seal should not be retrofitted with a shoe-
mounted secondary.
3.2.2 Liquid-Mounted Resilient Foam and Liquid-Filled Seals on Welded Tanks
Liquid-mounted resilient foam and liquid-filled primary seals have
approximately the same emission rates as primary shoe seals and exhibit the
same emission reduction trends with control (see Figure 3-2,c). However
in some cases the stored liquid may be harmful to the seal, making liquid-
mount! nq impractical.
3.2.3 Vapor-Mounted Resilient Foam Seal on Welded Tank
As discussed in Section 2.2.2, this primary seal has the potential for
high emissions when vapor-mounted. These emissions can be effectively con-
trolled by retrofitting with a rim-mounted secondary seal provided the gap
between the secondary seal and tank wall is carefully controlled. This is
illustrated in Figure 3.3
3-5
-------
500 ^
400 -
300
45
40
VOC
Emissions 35
(megagrams/yr)
30
25
20
15
10
5
0
a -
b -
c -
d -
e -
Foam seal mounted in the vapor space, gapped,
no secondary seal
Gapped foam seal with gapped rim-mounted
secondary seal
Foam seal mounted in the vapor space, tight,,
no secondary seal
Gapped foam seal mounted in the vapor
space with tight rim-mounted
secondary seal
Tight foam seal mounted in the
vapor space with a tight rim-
mounted secondary seal
1
2
3
4
5
6
7
8
9
10
11
psi
10 20 30 40 50 60 70 kPa
True Vapor Pressure of Stored Product
Figure 3.3. EMISSIONS FROM 100 FT DIAMETER WELDED GASOLINE TANK WITH
PRIMARY FOAM SEAL, 16.1 kph (10 mph) AVERAGE WIND SPEED
3-6
-------
Emissions from a tank storing a 41.4 kPa (6 psi) vapor pressure product
equipped with a vapor-mounted resilient foam seal are 18.6 megagrams per
year, if the primary seal has a tight commercial fit (Figure 3-3c) and 212
megagrams per year if the primary seal is slightly gapped (Figure 3-3a).
With a tight rim-mounted secondary seal, emissions from a gapped primary
seal are reduced from 212 megagrams per year to 4.3 megagrams per year
(Figure 3-3d). When installed over a tight primary seal, emissions are
reduced from 18.6 megagrams per year to 2.2 megagrams per year (Figure 3-3e).
With a gapped primary seal and a gapped secondary seal, emissions are 95.3
megagrams per year (Figure 3-3b).
3.2.4 Riveted Tanks
Riveted tanks present special problems regardless of primary seal design.
The primary seal must ride over the protruding rivet heads when the tank is
being worked, creating gaps. If the primary seal stops or is riding on a
row of rivet heads, the gaps can be nearly continuous and the wind-induced
emissions extremely high. The portion of the seal riding on the rivets
(and the riveted members) depends on design, and varies with location in
the tank. Emissions based on experimental tests conducted to evaluate a
shoe seal in contact with a "worst case" simulated rivet row are shown in
Figure 3-4a.6
Installation of a rim-mounted secondary seal over this primary shoe seal
reduces emissions from 39.9 to 22.3 megagrams per year based on one test and
to 7.1 megagrams per year based on another (the only difference being the
rivet row design with which the secondary seal was in contact). At more
favorable roof locations in a riveted tank, emissions will be lower. Emissions
from a welded tank with a rim-mounted secondary were 2.8 megagrams per year.
3-7
-------
no
100
90
80
VOC Emissions
(megagrams/yr)
70 -
60 -
50
40
30
20
10
Shoe seal, no secondary seal
Shoe seal with rim-mounted secondary
seal, Position I
Shoe seal with rim-mounted secondary
seal, Position II
Welded tank with shoe seal and rim-
mounted secondary
11 psi
10 20 30 40 50 60 70
True Vapor Pressure of Stored Product
kPa
Figure 3-4. EMISSIONS FROM 30.5 m (100 ft) DIAMETER RIVETED GASOLINE
TANK WITH PRIMARY SHOE SEAL AT 16.1 kph (10 mph) AVERAGE
WIND SPEED
3-8
-------
Rivet heads are particularly harsh on primary seals, and seal condition
may deteriorate more rapidly. Frequent inspections and good maintenance
practices must be followed to control emissions from riveted tanks.
3-9
-------
3.3 REFERENCES
1. SOHIO/CBI Floating Roof Emission Program. Interim Report,
October 7, 1976.
2. SOHIO/CBI Floating Roof Tank Emission Program, Final Report,
November, 1976.
3. Western Oil and Gas Association. Metallic Sealing Ring Emission
Test Program, Interim Report, Chicago Bridge & Iron Company, January, 1977.
4. Western Oil and Gas Association, Metallic Sealing Ring Emission
Test Program, Final Report, Chicago Bridge & Iron Company, March, 1977.
R
5. Measurement of Emissions from a Tube Seal Equipped Floating Roof
Tank, Pittsburgh-Des Moines Steel Company, October 9, 1978.
6. Western Oil and Gas Association. Metallic Sealing, Ring Emission
Test Program, Supplemental Report, June, 1977.
3-10
-------
4.0 COST ANALYSIS
4.1 INTRODUCTION
4.1.1 Purpose
The purpose of this chapter is to present estimated costs for control
of volatile organic compound (VOC) emissions from existing external floating
roof petroleum liquid storage tanks.
4.1.2 Scope
Estimates of capital and annualized costs are presented for controlling
emissions from existing external floating roof storage tanks. The estimates
pertain to welded and riveted steel tanks used for storing gasoline. Current
standards for floating roof tanks require the use of single closure (primary)
seals, so the cost of control is limited to the additional cost of installing
(retrofitting) a secondary seal on existing tanks. Control costs are
developed for a model existing external floating roof tank with a diameter
of 30.5 m (100 ft), a height of 12.2 m (40 ft) and a storage capacity of
8,910,000 liters, A range of cost effectiveness ratios are presented for
storing gasoline that allow for varying operating conditions, locations,
and control costs of tanks.
4.1.3 Use of Model Storage Tanks
Gasoline storage tanks vary in size with typical diameters ranging
from less than 9.1 m (30 ft) to more than 91.5 m (300 ft). Since it would
be impractical to determine costs for all tank sizes, a middle size model
was selected for this cost analysis. Table 4-1 presents the cases evaluated
4-1
-------
UM
ts\
O
0
1
J
r«*x
1 -Illf
fe
8
^.
^>»
-^r-
H-
LU
CD
s
0
CO
LU
e£
1 1
d
CO
s
2:
14
D-
o
LU
ni
Q
"J"
fl^Vt
| || )
O
LU
to
=3
CO
on
LU
Lu to
^fj. i ^ ^-^ Q}
|*y? 1. . \ I \
CC 14- v-
LU CU O CM «
| N CO i CO
co
yy
| £;
*?r CO
cu si >>
r CU CU -P
IQ en -)-»»-»-
CO co CU -E O
f- s_ E cn co
O co *r CL
4J .1- CU CO
co Q a: o
^^ ^
CQ
X
1
a
E
CU
Q.
riZ" ^-^ ^ cjir 5"
(A >rw >r« (/)
CL to to CL *
Q. CL CM
CM O 1
LO O O -CO
» r- CM tO i CU
f* cO CO CO CO CO
_Ct ' Q. Q. CL. Q- I
O "^ 1" CD CU
CD cu »
«« S- co r-* i en '
in 3 r CM * vo
in to
to
cu m
CO S- -~ - -S= -S= -r-
^= CL CL 4-
-o a. E E |i-
c: E LU
co CD ^"
VD r i i
" r r IT J
Q Q. CL CL C
CO NX NX .N^ O
CJ>
-o t~- ^~ m
cr CJ
r O> NO CM O
3 r CM >
to
J3
4-»
^
O
0
s_
cr.
.JE
0
=
o
to
^~
CU
to
cO
a
£=
O
o
cu
to
-a
cu
3
O
E
CO
O
M-
-o
CU
-P
c
3
0
E
a
r-
3
cr
r~
n
'
CO
CU
to
CU
in SD
^
in i
r~
(O r
CU CO
to cu
to
fen
p-1 r
S- !-
1
C7)T3
£ *i^
r- 3
3 cr
O -r-
r
OT3
t CU
cu s=
-P 0
^-> o
1 !
3 3
NJ
CO
0
a
CU >
-a r
r CO
cu a>
3 to'
i
i i
cu
CJ
CO
1
cO
cu
(/)
E
o
q-
-a
cu
"c
3
O
0
CL
CO
>
cO
1
'cO
CU
to
CO
E
r-
CL
a>
d
i
3
O
i
0
M-
CU
^
P
i^
3
-P OO
CO
a
CU 1
-o
a> co
i
i i
i i
cu
to
CO
CJ
*
x~^»
"5^i
LO
^
i
r
CO
CU
CU
o
.G
c/1
CJ
-r~
*
"«
CU
E
CO
1
CO
CM
VU
>j
s-
E
si
CL
en
c
5
0
^^
"o
cu
-p
JC
3
-^
d
-o
cu
cu
-r-
C£.
1
i i
i i
i i
CU
to
CO
CJ
O
4-2
-------
o
CJ
o
I
o
UJ
CD
CO
UJ
o
oo
CT
CD
a.
o
UJ
>
UJ
Q
UJ
oo
a;
UJ
LU
Of
o
O
CJ
CO
o
M-
CO
o
CO
r-
UJ
1 1
~
H
,_,
UJ
oo
=C
O
.c
c
LO
CM
CM
C
'"",
VO
r
j:
f>^
ft
<*
c
LO
ft
CM
CM
.c
c
^
,__
vo
.c
c
-^
I--.
°»
_c
a
^
LO
cv
CM
.G
Q
.is:
,_
»
vo
.c
c
*".
en
g
o
0
Q,
^>
*X2
g^
r
3:
ir>
to
!*"
0
CM
,
""
01
CO
vo
oo
00
«
r^.
CM
LO
CM'
o
oo
o
vo
un
un
'
oo
LO
LO
,_T
<-
o
a. co to
(OS- Q-
> 3 -i£
CO
CO CO
CO CO
35- OO
S- Q- i
H-
00
vo
oo
oo
CM
,-j!
CM
LO
en
o
«^J-
l
CO
CO
*
LO
I""*
o
oo
,_!
^
to
vo
LO
*~
CM
r~
CM
CO
vo
VD
OO
oo
fO
Q-
*^
vo
ft
1"^
CM
CM
oo
uo
in
LO
en
en
00
LO
^3"
CVJ
CO
CO
' ft
LO
CM
O
VO
CO
r
r-»
0
i-
co
en
CT»
r
C\J
CM
r_
r
IO
IO
rO
a.
_N^
" *si~
r
'sj*
o
1^^
r
1
O
LO
CO
LO
OO
ft
CM
LO
0
it,
LO
LO
LO
en
oo
o
o
CM
CM
^.
LO
CM
**
O
en
oo
CM
CO
,_!
to
Q-
\s
O
cr>
(j2
3;
S-
Q.
O
S-
4J
O
co
co
co
CO
i
i
i
<
i
UJ
yj
o
.c
LO
CM
CM
_C
.^
' ' ^
vo
j:
a
^£
^
en
IE
_v,/
LO
e
CM
CM
^:
Q
i*
»
VO
I
a
i-^
en
Q
LO
CM
CM
Q
r
^=
a
_i£
P^^
«
en
r*
O
i
CO
^>
-a
35
t i^O
VO VO
CO OO
t
LO ^i-
CM tO
VO 00
"
CTl CM
r-- CM
00 CO
VD CM
i LO
i CM
i r
VO OO
|
£ ?e
0 «*
* CD
t** m
cr» CD
r~~ r-^
cy> oo
ro CO
-
s-
O CO (O to
a. s- Q. a.
(O 3 y» \s
co co vo
CO CO
3 S- OO 1^
S- Q. i CM
tn
r-.
o
oo
en
CM
CM
CM
oo
LO
00
~"
in
r-;
^i-
vo
«,
CM
LO
CM
t-^
O
CD
LO
r
CO
CM
CO
OO
t
fd
Q_
«^-
r
"*
=3-,
<=}-
*
LO
VO
^J-
*
I>~I
^
O
CO
CO
CM
OO
co
co
en
LO
^j-
vo
vo
CM
^
VO
O
l~~
en
LO
^.
CM
tO
Q-
O
Oi)
VO
to
CD
10
s-
a.
T3
OJ
o
CO
Q.
v>
CO
s-
co
o
CO
W)
V)
to
fO
o
o
CO
CO
a
CO
c:
3
o
vt
CO
trt
to
o
LLJ
o
4-3
-------
o
o
to
o
CO
a:
C£5.
CD
Q.
O
LU
Q
LU
CO
CO
C£
LU
LU
o
o
LU
o
a>
O
O
CO
CT3
-a
0)
o
s-
o
o
en
O
! ,
~
Hj
Li
^
o
Q-
LO
CM
CM
CX
"^
LO
Q.
'""J
en
.E
o.
LO
cv
CM
Q.
^
LC
-E
Q.
j^
Cn
.E
Q.
If
C\
CM
.c
»
XT
O.
LO O>
* en
r-- co
LO r^
CM LO
0 r-
CM i
*^t~ CTJ
LD CM
CO i
CM LO
LO CO
i LO
*
i CM
5-
O CD OS X
> VI
CO CO LO
CD CO
13 s>- co r-^
S- Q- r- CM
h-
p^
LO
^J-
CM
LO
r^
LO
o
I
^~
CO
r--
r
CM
«*
LO
t
LO
cn
CO
en
«^!
i
*
CO
LO
CO
a.
\s
^j-
r
"*
CM
LO
CM
LO
CO
LO
p-^
CO
LO
to
CO
CM
r-.
r*-^
LO
sd"
LO
co
CM
O
CM
O
cn
,_
co
CO
en
r
CO
co
CO
O-
"*S
O
en
LO
CO
T3
CU
a.
a.
cu
o
S-
-------
and the technical parameters used in the analysis. The parameters were
selected as being representative of average annual wind speed on the
United States Gulf Coast, East Coast and West Coast, respectively, and
expected ranges of product true vapor pressure at stored temperatures.
Emissions and emission reductions are based on extrapolations from a
6.1 m (20 ft) diameter test tank to the full size model tank (see Appendix
A and B). It will be noted from Table B-2, Appendix B, that Cases II
and III do not represent predictions of maximum achievable emission
reductions. Accordingly, cost effectiveness for these cases (see Section 4.3)
i
are conservatively high.
4.1.4 Bases for Capital and Annualized Cost Estimates
Capital cost estimates: represent the total investment required to
purchase and retrofit the control systems on existing storage tanks
including the cost of cleaning and degassing tanks. Costs for research
and development, lost time during installation and start-up, and other
highly variable costs are not included in the estimates. These costs vary
so widely from case to case and from situation to situation that it is
virtually impossible to realistically quantify these costs. All capital
costs reflect second quarter 1978 dollars.
Annualized control cost estimates include operating labor, maintenance,
credits for petroleum savings, and annualized capital charges. Cost
estimates were obtained from an EPA contractor, equipment vendors, tank
service companies, local air pollution control reports, and an API contractor.
Credits for gasoline savings due to emission control have been calculated
from the emission reductions projected from the experimental tests.
4-5
-------
The annualized capital charges are sub-divided into capital recovery
costs (depreciation and interest costs) and costs for property taxes,
insurance, and administration. Depreciation and interest costs have been
computed using a capital recovery factor based on a 10 year secondary seal
life and an interest rate of 10 percent per annum. Costs for property
taxes, insurance and administration are computed at 4 percent of the
capital costs. All annualized costs are for one year periods commencing
with the second quarter of 1978.
4.2 CONTROL OF EMISSIONS FROM EXTERNAL FLOATING ROOF STORAGE TANKS
4.2.1 Model Cost Parameters
Cost parameters used in computing secondary seal control costs are
presented in Table 4-2. These parameters are based on actual cost data
2
from an oil industry journal, a National Energy Information Center
monthly publication, an EPA contractor, seal vendors, ' tank service
78 910 : 11
companies, local air pollution control reports, ' an API contractor,
and EPA estimates.
4.2.2 Control Costs
Table 4-3 shows the estimated costs of controlling VOC emissions from
the model floating roof storage tank. The estimates pertain to existing
welded and riveted floating roof petroleum liquids tanks that are equipped
with primary closure seals. The installed capital costs are average industry
costs of retrofitting a secondary seal on the model storage tank. The
annual operating and maintenance costs are estimated based on normal
maintenance and inspection programs. The annualized capital charges consist
of the capital recovery costs using capital recovery factor with 10 percent
annual interest rate and 10 year secondary seal life plus 4 percent of
4-6
-------
Table 4-2. COST PARAMETERS USED IN COMPUTING CONTROL COSTS
* Gasoline Value5
$100.60/m3 ($16.00/bb1)
II. Secondary Seal Value:
A. Installed (Retrofit) Capital Costs;*3
Tank with primary seal: $176 per linear m.
B. Annual Maintenance Cost;0
5% of installed capital cost plus annual inspection charge of $200,
C. Replacement Life: 10 years
Average gasol i ne yalue based on price data from Reference 2_ aricT a're~sR6wrr
~in'fabTe~4-"5. '" ~~" """
Average installed cost of retrofitting secondary seal per References 4,5,
6,7,8,9 and 10.
Annual maintenance cost per EPA estimate and annual inspection charqe per
Reference 12.
Expected replacement life per References 4 and 8.
4-7
-------
installed capital cost for property taxes, insurance and administration.
The total annual control system costs are the sum of the annual operating
and maintenance costs and annualized capital charges. Annual petroleum
credits from controlling (reducing) emissions are not included in these costs.
From Table 4-3, it can be seen that the average installed capital cost
of a secondary seal on a 30.5 m (100 ft) diameter tank is $16,900 and the
total annual control system cost average is $4,400.
4.3 COST EFFECTIVENESS '
Table 4-4 presents the cost effectiveness ratios of controlling gasoline
3
emissions from the model existing floating roof tank. The $100.64/m price '
for gasoline was established by averaging the per barrel prices of regular,
premium and no-lead gasoline from three different areas. The per barrel
price was then converted to a $/m . The cost effectiveness ratios for crude
oil may be approximated by multiplying the cost effectiveness ratios in
Table 4-4 by 1.38. For the development of this factor see Table 4-5.
This factor reflects the different average values and emission rates of the
two liquids. The amount of emissions controlled (reduced) varies with wind
velocity, absolute vapor pressure and control efficiency. Higher wind
velocity, greater vapor pressure and higher control efficiency will result
in a greater quantity of controlled emissions and larger petroleum credits;
opposite (low) values will result in a smaller quantity of emissions con-
trolled and lesser petroleum savings. Since a range of the above controlling
factors is needed to cover the typical range of tank operating conditions
and locations, cost effectiveness ratios have been determined using various
vapor pressure and wind velocity values for the factors and control system
costs.
4-8
-------
LU
03
oo
LU
C3
C£5
OO
ii
X
a
o
f
£
CO
oo
LU
o
o
o
C£.
O
O
CO
I
=*
CU
_a
(O
' i
0
-° i
CD:
o
in
LO
>,
! +J
i ^ _'»r
4-> 0
<*- ! to
O -4-> 10
OH- 0
o to
i_ : CU
CU 4->
4J 4J !-
CU .£ r
^^ d
-'58
ES?
LOCM CTl
OCM CO*
coi
CU
M
r-
00
>>
4-^
r
^-~
r
O
(0
u.
'
E
ro
r-
O
O
o;
en
E
*p
(O
o
u_
External
*
cu
Q.
1^?
>>
4-i
r
T*
i
O
rO
U.
,_
rO
CU
00
S-
rO
^
O
CU
oo
~C3
cu
-u
£
3
O
cu
0
cu
Q
r~*
O
SM
O
0
1
1
i r
LU
00
O'
H ~4
'! -1
,-LU
':OO
.O,
t*H
LU
OO
>
tt
£
t.
a..
C|_.
O
0)
1
CTi f* CO ^i*
>-' * ' ' * * *
*lo >i ro «xf
j
-
O^ r CO «s3-
UO f~" CO ^i"
r '
C3^ r^r OO «sf
tO i OO
rfl CJ CD 10
C O O
re 09- CJ
O £
0 CU E
o -»-> to cu E
-bO- £ CU -4-> 3
T- co to cu
ro S- >,i
4-> S ro CO O
tO .£ S-
o -a cj i <->
CJ E O CUT3
rO r^ ^ fS
ro »-> . .
«a en »-> c cno
+J £Q -r- O E O
i !- a. o < CD
rO ro O CJ r 3 ^ ^
CJ J- O ro i »
CU O -O 3 O V>
"O Q.-t^)- CU C E 4->
CU O N £ -r- -r-
i «r cf "O
r 4J i 4J CU
ro ro to «O r O i-
+J 3 O 3 ro 2: O
to £ O £ 4->
£ E CO
t-i eC
CU £
r--0
rO
3 £
£
CU
us
S- «
o «/>
4J CU
0 2
re tO
CU
O
CU
O.
CU
a
3
CJ
o
E
to
cu
o
-a
t/>
cu
s_
fO
cj
ro
a.
ro
O
-a
cu
N
to
3
to
3
Q-
LU
&.
CU
Q.
4J
(/)
O
o
cu
o
(O
CU
+>
C
ro
E
-a
E
(O
en
E
r
4->
cu
0 Q.
CJ O
cu s-
$- a.
t S-
ro O
_t_\ tt _,
*r~
Q-4->
ro
CJ O
CJ
en
i- "fO
to -p
3 !-
*^* rO
to u-
t/l "O
o a>
0 i
>> rO
CU c/1
> £
O <-
O
CU tf
i- O
Q. 3
(O i
o a.
o
i >
to
O
o
cu
o
£
ra
c
cu
-t-j
E
i
ro
E
T3
C
ro
en
E
M
(O
S- .
O. (/)
o en
E
(O >
3 tO
C l/l
-------
I;
s:
I
g
J
u
o
UJ
J
1.
1
o
» i
"^f
_J
o
JJ
1
2
1
_J
s!
LI
O
ss
l_
U
i
i
o
4
S
-J
T
JJ
CO
UJ
o
CO
_
s
Q-
1
o
i
i
11
LU
00
^£
U
o
LO
CM
CM
i:
c
tO
F
X
C
cn
>,
t
u
c
1
a
c
3:
cS
^£
O
at
to
a
a.
*
a.
to
CM
Q_
co
CO
IO
a.
o
cn
to
Q.
Jb£
2-
*
a.
to
CM
a
CL-
OD
CO
1
10
a.
^
o
at
to
CO
Q.
*
rd
Q_
to
P-.
CM
rd
ex.
CO
CO
a>
n
tn
cn
J-
Q_
S-
o
0.
rd
QJ
s-
1
o cn
cn r**. ^f 01 "'«
CO CM
CO * ^
co in
cn CM t* P-*
i LO
«* CM ^t- «*
r * CM i
^- P-. CO
r CO «* O CO
co
cn t" ^i" co co
* '
CD CO f""»
CM tO "3" fx. CO
* cn
«sj- - * «sj- CO CO
_
co cn
cn to r
r-I CM , m
r- *~* *a- ^ cn
i IO *fr
«3* CM *3- i CO
p-«fc
CO r"» ^f CO CO
,
CM I"- CO
r~ P**« ^j~ tO Ol
tn >--* *?j* co r**
to in tn
co co "fir Q to
* . (
CM ^ " ^f ^J" P**
^~ CO P*1*
co co -^ o ;^
CO r ^ CO CO
in co P»*
r IO ^ P*% CO
. . . . . . o
^" *^* *!" CO O>
CO CO CM
in co «d- o cn
CM «3- -a- m
^^
to P>. <*
r ^- ^- CM in
tn
E °°
TD tn w -
» 4-J 4-
i c:ji co (- o tn
r *r- *-** "O C
S--^^ rd >> 'o S- ET-r-
4J S- co -^ S- o tn
s= >, o 4J ^^ s- tn
O "^ E O C * * (U '-
ocn no OS- +> OLE
S QJ *»*> C_> >> W) LU
CJ 4-> "O
O * i- * - rd O T- OJ
in o a. *r- c ^-^ zj *** S- o a>
r-=j -a <: co^s-s
LU
c£ 3CO rdo ** +J o -^-*
o c o a> o
> «X I E: c->
]
1
i
i
1
fe
c
1
ce
o
Q.
>
LL
CO
i
Lu.
>-
^^
Q^
0.
t
\s
"Si
0
LU
C
3
1
"
LU
or
S
.c
o
LO
CM
CM
t~i
to
a
cn
u
o
"a!
o
c
3
k?'
O
to
ca
Q_
**
S-
0.
i-
o
a.
10
0)
3
I,
1
t^- CM CM - *
r- o to to
d- LO
to r>* CM
n to
r^ CM <3-
^J- I CM
I CO LO
CM *! 1
to o
CO O O
^f
CO CM CO
i * «3- CM i
l~~ fO 1-^
i cn *3- ^s- CM
to * «* co to
^_^
o in *
O CM LO «-»
^- CO *d"
LO LO CM
<"> "^ ^J- -
* rC"
=» * co
^* cn r^
tO CM . . r
o
cn LO o
cn «3- cn o
cn , «3- CM CM
o cn i
. . o
3- - * CO CO
3- 5"
CM O to r^-
*d" CO tO
o ro
CM ' ^1* r
CO
LO tj- Cn CM
cn ^-- <* CM ro
CO ^^ CO
l~- CO -3- LO i
LO ' s- i a> cn in
s-* 10 >> o s- E: v>
4J S- 01 -^ S- 0 -r-
c >> o +j i- e
0---. EO C OJLLJ
cj cn 3 o o »-> Q.
r-«_- O ^^ 1
O ^ S- ---* IO > r i. -O OC
W1O O.-I- C O 3--« i. +>
r-3 -O =CEO CO OCI
S~a i
,
!
J
-*,
r
X.
*>
4-10
-------
oo
I
UJ
o
OO|
si
oo
o
CJ
o i
o) :
3 '
C
_,
^£
UJ
00
t
Ul
o!
3:
oo
«^
1 t
Oi'i
o.
I
» 1
3
zr
Q
E
>
t 1
Q£
I
t 4
j_ |
LU
S
If
c\
Cs
"
^
u
.c
o
jv;
o
r
O
O
"oJ
S»
a
c
'S
re
O_
-i£
o
a>
to
a
. Q.
5
re
a.
-M
to
CM
re
Q.
CO
CO
re
0-
-^
o .
O>
to
to
re
a-
3
re
a.
to
r
CM
re
a.
CO
CO
1
in
VI
0)
i.
a.
i.
0
a.
re
1
o
ti
c
r>
J O^ O^ - *
3 CO «d- <* U5
j r^ ^- co v-'
^_^
to to
CM "*
Lf
o
^ . i
} <* if. <* co §
CO LO ~S -~^
*0 «!j- CvJ CO
^- - t')
1C
tc
1^*
in
r**
» --»
tn tn
»JD «h r^. CTI
» CTt
CM ^J- r
^O rf. I^s. O
. . .10
O i «* CM CM
Ifl
Ol
^_^
T 10 1
r^«. ^- (o Ot'
« m
* " *a- co r
»
CM
Q
CO
to
o
r~~^
^
'
Oro
= >>
o -Q
O 0)
c
o *~^
i TD
U) «! oo r-
fO ^* * ^
to <3-
to «* t*.
-> *co
* CM ql
^^ -: CM
o o>
^- co «*
t . . o
~~-r ^- co to
_^
r^- co
i- ^t- en co
. to
* «3- CO CM
^ 4J tj«
^cua oo -i- o in
'i'R o I i1!
0+3 ~!i "*
§ f~> o t_ jj o *£
"S--- "reo "° *^"°
> ^r-'S
3tJ reo «o <->§ '
C--- 4JO JJ t>0
C O (U O
* t z o
I
TO
3
CT
g,
re
I
Ol
=1
I 4->
Ul
If
O 01
i.
O
S- -O
01
S-
01 -U : 01
> C ' O
00 =
O CJ T3
01 Q)
S. « 55.
3
x, c -o
O i O
c re si
a) *-> +J
o £ §
C -a °
<4- C U>
ui re. c
3
CO
o
o
s-
«s- ^
v>
+>
a. v>
r- 3
re
't-
is
S; ui
Q» **
10 J3
4-n
-------
s
t 4
_J
O
o:
gi»
o
o
LU
co
to
LU
LU
> CO
LU-
LL. LU
LU CD
1 'D£
COO
o co
LU _J
-r- t 4
(_ O
CD LU
* < ' '
H- o;
a: o
l -
LU'
s-
Q.
O S
O
(/I
O) cn
C CU
i- E:
"o >
to +j
CD O
cu
LU
i
O
cu
a
3
S-
o
cu
c
f^
0
j co
[ > £1
1 "^^
w cn
cu
0
CM
O
CM
CO
1
CO
CM
CM
r
CM
VO
O
O
VO
d.
VO
O
o
1
(0
o
3 ^"^
o i
O -Q
S- JQ
Q- ">.
tl ^^^^
O
CO
CU E
3 --.
'«o
>
CO
1
CO
co
o
CO
II
^-^
o
CM
co
,_!
"k^-x
o
CO
vj.
o"
o
CO
**
II
tn"
*~*
*
CM
^ '
X
'"^
^_^
II
s-
o
0
as
LU
0,
-a
(O
,.CM
W)
cu
o
(U
s-
-------
For the model existing gasoline floating roof tank, it should be noted
from Table 4-6 that the cost effectiveness ranges from a cost of $3,665
to a credit of $66 per Mg of controlled emissions. The corresponding cost
effectiveness ratios of crude oil emission controlled using the 1.38 factor,
vary from a cost of $5,044 to a cost of $25 per Mg. Thus, due to the higher
value and emission rate of gasoline, the cost effectiveness for crude oil
ranges from $66 to $1389 higher per Mg of controlled emissions than for gasoline.
4.4 ECONOMICS OF SCALE
The preliminary cost of retrofitting a secondary seal to existing
floating roof tanks were also developed for a 10.7 m (35 ft) diameter tank
and a 53.3 m (175 ft) diameter tank. These were developed to check the
linearity of the scaling effect on cost effectiveness developed for our
model tank. As could be expected there were some dis-economies of scale
in the cost effectiveness of the smaller tank. This resulted in the smaller
tank cost effectiveness ($/Mg of emissions controlled) being approximately
105 percent of the cost effectiveness of the 30.5 m (100 ft) diameter model
tank. Also, as expected, the larger tank had some economies of scale. This
resulted in the larger tank cost effectiveness ($/Mg of emissions controlled)
being approximately 85 percent of the cost effectiveness of the 30.5 m (100 ft)
diameter model tank.
4-13
-------
T3
(U
S
4->
O
O
DJ
*^
>-
in
oo
co
UJ
o
UJ
LO-
LL.
UJ
o
o
vo
I
0)
r-*
JD
(X]
11
1 1
1 1
t 1
t 1
LU
OO
o
-G
OL
LO
CM
CM
CL
l^
VO
r
-S3
CL
^^
2
Q.
LO
CM
CM
CL
v^
-^
VO
r
JZ
CL
-^
^
CT»
CL
.*£
LO
CM
CM
.C
CL
~£
S
Q.
^y\
>j
.i-i
p
o
o
r
0)
>
o
cr
r
S
1^
CO
=1-
cr>
oo
r^»
co
vo
CM
CM
VO
un
f^
o
CO
cb
^:
CO
en
co
vo
r
Ln
vo
oo
<0
__}
en
i
CO
3
s-
f-
.-
^-
'«^-
r
O
vo
CM
^
O
LO
0
co
o
cr>
CM
r
^
VO
oo
oo
oo
o>
o
CO
LO
ns
Q_
^
VO
r^
CM
en
CM
en
o^
CO
CM
CM
LO
f*^
r~
"~
CM
i
oo
LO
«q^-
~
OO
oo
§
o>
rd
Q.
**
^"
i
*
,
VO
VO
'
*
oo
oo
^ '
^
oo
vo
LO
^"
CM
vo
CM
r^
LO
o>
^.
oo
to
0)
CD
O
a>
ai
<4-
a>
ca
4-14
-------
4.5 REFERENCES
1. "Methods for Extrapolating Chicago Bridge and Iron 6.1 M (20 Ft.) Test
Tank Results to Full Size Tank_," EPA-DAPQS Draft Report,
April, 1978.
2. "Refined-products prices", Oil and Gas Journal, March 27, 1978
3. Monthly Energy Review, June 1978, Office of Energy Information and
Analysis, National Energy Information Center.
4. R. Bakshi, Pacific Environmental Services, Inc., Santa Monica, Cal.
Petroleum storage tank and seal cost data memo to file by R.A. Quaney,
U.S. Environmental Protection Agency, dated November 2, 1977.
5. 0. Hunter, Western Petro-Chemical Co., Los Angeles, Cal. Petroleum
storage tank seal cost, data memo to file by R.A. Quaney, U.S. Environmental
Protection Agency, dated November 2, 1977.
6. K. J. Kolkmeier, Pittsburgh-Des Moines Steel Co., Pittsburgh, Pa. Petroleum
storage tank seal cost data memo to file by R.A. Quaney, U.S. Environmental
Protection Agency, dated November 2, 1977.
7. H.R. Wiggs, Tank Service, Inc., Tulsa, Oklahoma. Letter to R. H. Schippers,'
U.S. Environmental Protection Agency, dated May 18, 1977.
8. J. Mulkey, Tank Service, Inc., Wilmington, Del. Petroleum storage tank
seal cost data memo to file by R.A. Quaney, U.S. Environmental Protection
Agency, dated November 2, 1977.
9. California Air Resources Board. Public hearing - Proposed Ammendments to
Rule 463 of SCAQMD, June 25, 1976.
10. California Air Resources Board. Public hearing - Proposed Ammendments to
Rule 463 of SCAQMD, March 25, 1976.
11. Dr. W. Sheppard, Battelle Columbus Laboratories, Columbus, Ohio. Petroleum
liquids transportation cost data memo to file by R.A. Quaney, U.S. Environ-
mental Protection Agency, dated November 2, 1977.
12. R. H. Schippers, U.S. Environmental Protection Agency. Memo on Secondary
Seals on New Petroleum Product Storage Tanks, dated May 25, 1977.
4-15
-------
-------
5.0 RECOMMENDED REGULATIONS, COMPLIANCE
TEST METHOD, AND RECORD KEEPING
The affected facilities are external floating roof storage tanks with
capacities greater than 150,000 liters (950 bbls) containing petroleum liquids
with a true vapor pressure greater than 10.5 kPa (1.5 psi).
5.1 RECOMMENDED REGULATIONS
Recommended regulations for the storage of petroleum liquids in external
floating roof tanks are:
1. Except where specifically exempted (See 5.1.4), all external floating
roof tanks with capacities greater than 150,000 liters shall be retrofitted
with a continuous secondary seal extending from the floating roof to the tank
wall (a rim-mounted secondary) if:
(a) the tank is a welded tank, the true vapor pressure of the
contained liquid is 27.6 kPa (4.0 psi) or greater,and the primary seal is one of
the following:
(i) a metallic-type shoe seal, a liquid-mounted foam seal,
or a liquid-mounted liquid-filled type seal, or
(ii) any other closure device which can be demonstrated
equivalent to the above primary seals.
(b) the tank is a riveted tank, the true vapor pressure of the
contained liquid is 10.5 kPa (1.5 psi) or greater, and the closure device is as
described in 5.1.1 (a).
5-1
-------
(c) the tank is a welded or riveted tank, the true vapor pressure
of the contained liquid is > 10.5 kPa (1.5 psi) and the primary seal is
vapor-mounted. When such primary seal closure device can be demonstrated
equivalent to the primary seals described in 5.1.1 (a), the provisions of
5.1.1 (a) apply.
2. The seal closure devices shall meet the following requirements:
(a) there shall be no visible holes, tears, or other openings in
the seal(s) or seal(s) fabric.
(b) the seal(s) must be intact and uniformly in place around
the circumference of the floating roof between the floating roof and the tank
wall.
(c) the gap area of gaps exceeding 0.32 cm (1/8 inch) in width between
the secondary seal installed pursuant to 5.1.1 (c) and the tank wall shall not
2 2
exceed 6.5 cm per 0.3 m of tank diameter (1.0 in per foot of tank diameter).
3. All openings in the external floating roof, except for automatic
bleeder vents, rim space vents,and leg sleeves, are to provide a projection
below the liquid surface. The openings are to be equipped with a cover, seal
or lid. The cover, seal or lid is to be in a closed position at all times
except when the device is in actual use. Automatic bleeder vents are to be
closed at all times except when the roof is floated off or landed on the roof
leg supports and rim vents are to be set to open when the roof is being
floated off the roof leg supports or at the manufacturer's recommended setting.
Any emergency roof drain is to be provided with a slotted membrane fabric
cover or equivalent cover that covers at least 90 percent of the area of the
opening.
4. The following are specifically exempted from the requirements of
5.1.1:
5-2
-------
(a) external floating roof tanks having capacities less than
1,600,000 liters (10,000 bbls) used to store produced crude oil and
condensate prior to custody transfer.
(b) a metallic-type shoe seal in a welded tank which has a secondary seal
from the top of the shoe seal to,the tank wall (a shoe-mounted secondary).
(c) external floating roof tanks storing waxy, heavy pour
crudes.
5. External floating roof tanks with a closure or other devices
installed which will control VOC emissions with an effectiveness equal
to or greater than the seals required in 5.1.1 (a).
5.2 COMPLIANCE TEST METHOD
1. Compliance for external floating roof tanks does not require
measurement of the primary or secondary seal gap area, except as required
to meet 5.1.2 (c), and can be determined by visual inspection.
2. For compliance with 5.1.2 (c), the secondary seal gap area can
be determined by measuring the length and width of the gaps around the
entire circumference of the secondary seal. Only gaps greater than or
equal to 0.32 cm (1/8 inch) shall be used in computing the gap area. The
area of the gaps can be accumulated to determine compliance.
5.3 MONITORING AND RECORD KEEPING
It is recommended that the routine visual inspections be conducted annually
or at shorter intervals, and that the secondary seal gap measurements be
made annually. Evidence of any type of malfunction (as noted above) is to
be recorded.
When a liquid having a true vapor pressure greater than 7.0 kPa (1.0 psi)
is stored in an external floating roof tank not equipped with a secondary seal
5-3
-------
0-' aoproved alternative control technology (see 5.1.5), a record should he
maintained for no more than two years of the average monthly storage
temperature, the type of liquid, and the Reid vapor pressure of the liquid.
The true vapor pressure may be determined by using the average monthly
storage temperature and typical Reid vapor pressure of the contained liquid
or from typical available data on the contained liquid. Supporting analytical
data can be requested if there is a question on the values reported.
5-4
-------
APPENDIX A
SELECTION OF EXPERIMENTAL TESTS FOR
WIND INDUCED EMISSION CALCULATIONS
An experimental 20' 0 test tank at the Plain-field, Illinois, Research
Center of Chicago Bridge and Iron Company (CBI) has been extensively used by
industry to investigate the mechanisms causing hydrocarbon emissions from
I rt o n f- /-
floating roof tanks. »^>-:>»H-»O'D A large number of tests were conducted on
various types of seals to study the effect of parameters such as wind speed, gap
between the seal and tank wall, and the leak rate of the shoe seal vapor
space system on hydrocarbon emissions. Secondary seal efficiency was
evaluated. Methods for extrapolating specific test results from the 20' 0
test tank to full size tanks have been developed. This appendix describes
the methodology used in selecting specific tests for extrapolation which
are considered representative of average field tank conditions.
A. METHODOLOGY OF SELECTION
The selection of CBI tests on primary seals which represent the
"average" primary seal gap in the field was based on EPA's analysis of
tank inspections made in 1976 by regulatory agencies in California. A total
of 398 tanks were included in this analysis; 163 welded tanks with primary
shoe seals, 141 welded tanks with non-metallic seals of either the foam or
liquid type, and 94 riveted tanks with shoe seals. Tanks equipped with a
weather guard over the primary seal were included. Excluded were 47 tanks
A-l
-------
which were reported to have either a "double" or "wiper" secondary seal.
In the inspections, gaps between the primary seal and tank wall were
measured with probes or rods of varying dimensions. The width and length
of each gap was recorded. It was not possible to derive from the inspection
data an average gap width in the field that was comparable to a specific
CBI test, nor a range of gap width patterns which could be compared to one
or several CBI tests.
The final selection was made by comparing the gap areas in the tanks
2
inspected to the gap area in the CBI tests expressed as in /ft of tank
diameter.
B. SELECTION OF WELDED TANK WITH SHOE SEAL AND RIM-MOUNTED SECONDARY SEAL
A number of tests were made with single shoe seals having gaps up to
1 1/2 inches simulated by forcing the primary seal away from the tank wall with
spacer bar arrangements.7'8'9 In all of these tests the leakage rate for the
seal system (the space bounded by the shoes, the envelope, the rim space
and liquid) averaged about 0.032 SCFM per foot of tank diameter at 1 1/2 inches
of H20 pressure drop. A leakage rate of 0.50 SCFM at 1.5 inches of H20 is
considered commercially achievable. Further research was conducted by CBI
to establish the relationship between shoe seal emissions and leak rate. The
leak rate of seals inspected in California is unknown. The final determination
of the base case for calculating wind induced emissions for a primary shoe seal
was made by; (1) using the methodology described in (A), and (2) using a
test that simulated a leak rate of 0.8 SCFM/ft of tank diameter at 1.5 inches
of HpO pressure drop. This leak rate appears reasonable based on field test data
and the California inspections which revealed relatively few tanks with openings
or tears in the envelope.
A-2
-------
A comparison of the gap area in the inspected tanks with specific CBI
tests is shown in Table A-l. Selected tests on envelope leak rate simulations
are also shown in this table.
It will be noted from Table A-l that 89 percent of the tanks had gap
P
areas (in /ft tank diameter) equivalent to tests where emissions remained
relatively unchanged from a shoe seal with a tight commercial fit to one having a
gap(s) up to one inch.
Test W-12 has a commercial fit and simulates a seal system vapor space leak
rate of 0.8 SCFM per ft of tank diameter at 1.5" H20. W-12 was selected as
the base case for wind induced emission calculations.
A rim-mounted secondary over the W-12 primary seal was then judged to have
at least a 75 percent efficiency based on numerous .secondary seal single seal
combinations tested with varying gaps in the primary, secondary, or both,
3uring the same test. -
C. WELDED TANK WITH SHOE SEAL AND SHOE-MOUNTED SECONDARY SEAL
A secondary wiper seal mounted on the shoe was tested with a tight
commercial fit and with gaps. The efficiency of the wiper in each of these
tests was used to estimate the base case (test W-12) emission reductions.
D. RIVETED TANK WITH A SHOE SEAL AND RIM-MOUNTED SECONDARY
Test W-28 was made with a single shoe seal in contact with simulated
horizontal and vertical rivet rows.11 Table A-2 gives a comparison of the gap
area in the inspected tanks and the gap area in this test configuration.
The gap area in a riveted tank will vary with the position of the roof in the
tank and the rivet patterns in W-28 represent a condition where gaps may be
expected to be at maximum. This is judged to be the reason why the inspected
A-3
-------
CO
E
CO
(ft~
E >>
O cO
r- T3
CO tO
f p,^
LU
r"
cO
CO
CO
E
r-*
to
n.
CO
o
u_
S- CO
CD -r-
r> /"-»
CM ^
JC E
P CO
Cr
i i
CO
CD
S *~~^
eCCM
r JE
CO P
4-> E
O -i-
1 ^^ ^
.
.E .E
4J P
T3 E
r Br
*^g *^-^
0
CO O
CD !- CM
o. co Q n:
o en
r~ CO 4-)
CD ^£ *i LO
> (0 -^
E co s: t
LU _J U.
O 4-5
OO CO
O
to
CD
r-
UO r CO 1 O O
. 1
CO f**- CO 1 r O
LO CM 1 r O
-^
.E cn
A «
P T3 CO 4-
«* CO CO CM IO OO
i CT> CO r LD
g
1
CMCMCMO «* COOOO
to to to r*^ P** r*** *^t~ to LO
r r i CM
LO O LO O
r^. co co r>.
o o o o o
i kj CTl *>J
pi«
OLOIOI^ * oooo
CO CO CO CD
i CM
CM
CM CM ^v.
O'^^.r r OOOO
n~ r I
p«
^i^-x
E
CO P
J3 4-> S-
OCMCM T-"* OOOO
^
LO
CM CM CM CM CO
cocococo o oooo
OOOO O COLOCOLO
OOOCD O Or- CM-vf
CO
>
CM
2
of
3 VO CM 1^. IO LO
"COLOtO C\J i i i i
r i i i i riii
33333 3333
*
t
0
-o
CO
CO
Q.
CO
11
r
-o
^/
c
}^>
4-^
4-
CM
E
LO
i
Iv
O
A
P
S-
co
40
CO
c
r3
T3
J^
c
rO
^
if
"**^
OJ
c
r*
co
^_j-
Iv
LO
r
A
-a
s-
co
4_>
CO
C
o3
^
JK^
c
fC
4J
4->
tf-
**^^,
CM
E
LO
CO
CD
[V
CO
«^j~
A
CD
S-
CD
4-d
CO
pi
niJ
p-
-a
\s
E
t %
^«)
'j
"^^
CM
E
p»^
^f-
r
Iv
LO
co
CD
A
I-"-^
<^J*
"~
A
\s
E
] ^
CD
E
O
>^
r*
E
O
CD
T3
P
E
l l
CD
recorded in inspections; these tanks are included in W-26
4->
o
E
CO
T3
~^
4^
E
CD
CO
to
r«.
c*
4_3
(/>
Q.
CO
CD
-------
Eg
3LU
oo 1-
li i
S|
1 z:
§S
li i C£
t i ^~
LU
o
UJ ^
o ^
LU O
r». |
oo
i 1 1 1 1
f~yf
^. *"*^1.
(/) (/)
r- Q
E^
.
S- !-
CU Q
a.
M E
l£*~ ^J
CJ 1
E
i
tO
CU
^ «*"*
ccc\i
c^
a
rO C
^_j .,_
o
"
^.^
4-5 0
"O C
I *^
*^g V«_*i»
0
s^
tO O
CU !- CM
a. cu Q n:
o en
cu ^ <*- Lo
> to ^
E cu s: i
JLj I 1 '
O 4-5
oo to
.
o
"Z.
4-5
(/I
CU
1
«=3- tO OO CM CO
» *
O O O CM CO 00 O
«^J" ^J~ i " t*^i
r
O O CO O CO CO «!f
oo ^j- i en
oo
z:
I
o
_i
,to =t
CM |
S
1
oo .00'
<^J"
LO «^|- ' CO
r*^ CTi r
LO oo |v ,
O i |V IV
cr» co in
oo i |v r»» co «*
CO O r «3" CTl
\s
<= IA IA |A |A
I
T3
CU
O
r 10 CU
O^ ^""^ p?n
r-^ oo to
^l
to
LO o a.
o to to
CD
O
« *
l\
rd
CD
CO M- SO
3 0 0 -E
o cu 4-5 co
3 Q. O 4-5
E 0 -E 0 <4-
r- | CO CO O
4-5
&H. *^^"fc * "*
O tO r^
CJ -
CO
CM
1
^^
A-5
1
=C
cu
r~
r*>
1
3
O
S-
^
E
^
f" [ %
Q. "i
E
O !-
^"
1 O
0
-a
0) en
CU E
a. -r-
o
"*CJ niJ
E O
i S-
3 Q.
<->
'ro
r- T3
E
a. to
in to
1 0
S-
cu cu
S- E
^3 __j
U) E
cu cu
S- S-
Q- O
E
S-
O t/>
a. a.
(O fO
> CD
-------
gap areas, taken at random roof positions, are considerably lower than W-28.
The inspected gap areas in the riveted tanks are substantially greater than in
the welded tanks in Table A-l. Also, the numbers of gaps in the riveted tanks
were far more numerous than in the welded tanks and gaps in riveted tanks
may exhibit the characteristics of the continuous gap in test W-6.
A rim-mounted secondary over W-28 was tested in contact with two rivet
patterns, tests W-29 and W-31. Emissions were calculated for both these
tests and reductions obtained by subtraction from W-28.
E. RESILIENT FOAM SEAL MOUNTED IN RIM VAPOR SPACE OF WELDED TANK
Two tests were selected for the single seal. A single seal test with a
"tight commercial fit" (Test 13, 16, 20, 21) and the same seal with gaps
(Test 23).12'13
A rim-mounted secondary with a "tight commercial fit" was installed
in each of the above tests. Emissions were then calculated for each (Test 32
and 34A) and emission reduction obtained by difference.
Inspection data for "non-metallic" seals are presented in Table A-3.
The seals were not identified by type (liquid filled or foam) or location
(rim vapor space or in the liquid surface).
For comparative purposes emissions were developed for a secondary seal
with gaps (Test 34B) installed over a primary seal with gaps (Test 23).
A-6
-------
O
I
LU
5 <*
3 ^
J n
-= X
LU O
Q- I
00
a;
2: «C
I-H Q_
=C O
LU CJ
OO
Q- _J
CD UJ
00
UJ
oo
CO
i
LU
_J
CO
-o
CU
U
CU
O.
CO
1 1
CO
V?
c:
^
c:
1
CO
CO
i
,_
(O
^_3
C~
CO
^
r
S-
co
Q.
X
LU
s-
cu
CO
p^
(O
1
O
u_
o
CM
(O
1
tt>
O
S-
cu
ta.
s-
co
CO- -
c: >>
o res
r- T3
CO ~-^
CO CO
i- JD
E-
LU
<
^
to
CO
oo
c:
CO
Q.
(0
CD
4-*
LU
i« fcJ
CU -r-
Q.Q
J NX
J= C
O fO
E H-
i i
fO
CO
<=CCM
t~
t O
to c
4-> ^>
o
1
^^
c~ c~
4J (J
a c
S^-
0
0
CO
CO
H-
LO , r LO O
* r^. cr> cr> tn co
tO r- CO O
1 en o o cr> o
t^ i i r- ro i
-a a!
i o <:
. rrj
CM CO
|
A M
3 00
z:
< ^
0 _I
i <:
CQ i
CM 13 O
LO r~~
CM
o o |v
r CO
CO |V LO CM
Q.
fO O O r
CD
IA |A A
3
CO
^
ro
(
T3
o =* cu
CM 4->
O
O)
CO
C
CM
O ^
O CM
CM
CM
A
to
oo oo
r CM
0)
o
(O
CJ-
s_
3
CO
s-
o
a>
o
Q.
co
o
Q.
(O
a a.
co E
o
i
o
E
r- > -r-
4-> 3
CO
ro
co
a.
-t-j
c: cu
cu s-
-a 3
i CO
co
o cu
a.
cu
.f^ O
(O ^>
c: nJ
ZD >
to <*^
A-7
-------
REFERENCES
1. SOHIO/CBI Floating Roof Emission Program, Interim Report,
October 7, 1976.
2. SOHIO/CBI Floating Roof Tank Emission Program. Final Report,
November, 1976.
3. Western Oil and Gas Association., Metallic Sealing Ring Emission
Test Program, Interim Report, Chicago Bridge & Iron Company, January, 1977.
4. Western Oil and Gas Association, Metallic Sealing Ring Emission
Test Program, Supplemental Report, Chicago Bridge & Iron Company, June, 1977.
5. SOHIO/CBI Floating Roof Tank Emission Test Program, Supplemental
Report, Chicago Bridge and Iron Company, February 15, 1977.
6. Floating Roof Seal Development - Emission Test Measurement on
Proposed CBI Wiper-Type Secondary Seal for SR-T Seals, Chicago Bridge & Iron
Company, February 23, 1977.
7. Western Oil and Gas Association, Metallic Sealing Ring Emission
Test Program, Final Report, Chicago Bridge & Iron Company, March, 1977.
8. Reference 4, Op. Cit.
9. Reference 2, Op. Cit.
10. Reference 6, Op. Cit.
11. Reference 4, Op. Cit.
12. Reference 1, Op. Cit.
13. Reference 2, Op. Cit.
A-8
-------
APPENDIX B
CALCULATION OF WIND INDUCED
EMISSIONS FROM EXPERIMENTAL TESTS
A. EQUATION FOR WIND INDUCED EMISSIONS
Emissions were extrapolated from the 20' 0 test tank selected tests to
the 100' 0 model gasoline storage tank using the following equation:1'2'3
EF = 1.2337 x 10"4 x
P
F
[ 1 - + (1 - .068 PF) "
2.0
X E
p
Where: Ep = Emissions from full size tank, megagrams/yr
Ep = Emissions from test tank at 5.0 psi, Ibs/day
Pp = Vapor pressure of stored product in full size tank, psi
DF = Diameter of full size tank, feet =100
MHC = M°lecular weight of full size tank hydrocarbon emissions
(hydrocarbon vapor molecular weight), Ibs/lb mole = 65
B. Ep FOR THE TEST TANK
In each specific test emissions were measured at varying simulated wind
speeds. These results were then plotted to yield a smooth "Ep vs Windspeed"
4
curve for each test. Ep values for the selected tests and various wind
speeds read from these plots are given in Table B-l.
In the 100' 0 model tank analysis wind speeds of 6 mph, 10 mph, and 14 mph
were used. These represent mean average annual wind speeds on the Vilest Coast,
Gulf Coast and East Coast, respectively.
B-l
-------
TABLE B-l - Ep (Ibs/day) vs Wind Speed (mph)
(201 p Test Tank - 5.0 psi)
Selected Tests
Welded Tank - Shoe Seal
Single Seal (W-l,W-2
W-1R)
Single Seal (W-12)
Rim -Mounted Secondary
(25 % W-12)
Shoe-Mounted Secondary
Tight Fit (6-2)
Shoe-Mounted Secondary
Gaps (C-l)
Riveted Tank - Shoe Seal
Single Seal (W-28)
Rim-Mounted
Secondary (W-29)
Rim-Mounted
Secondary (W-31)
Welded Tank - 'Foam Seal
In Rim Vapor Space
Single Seal (13,16,20,
21)
Single Seal (23)
Rim -Mounted
Secondary (34A)
Rim -Mounted
Secondary (32)
Rim -Mounted
Secondary (34B)
Wind 'Speed (mon)
4
1.80
2.10
0.53
0.92
1.60'
10.10
5.90
1.88
3.70
26.0
0.70
0.62
1.70
6
3.10
3.60
0.90
1.40
2.00
16.00
8.80
2.80
7.00
62.0
1.10
0.82
4.20
.8
4.60
5.20
1.30
1.87
3.12
21.0
11.8
3.75
9.8
98.0
1.80
1.05
12.0
10
6.20
7.30
1.83
2.35
3.80
26.0
14.5
4.60
12.1
>98
2.80
1.41
62.0
12
7.60
10.00
2.50
2.80
4.60
32.0
17.3
5.70
15.0
4.20
2.0
>62.0
14 :
9.20
13.00
3.25
3.30
5.30
36.0
20.0
6.40
17.0
6.00
2.70
B-2
-------
C. EMISSION CALCULATIONS
C.I Primary Seal With and Without Rim-Mounted Secondary
Using the equation in (A) emissions were calculated for each of the
selected tests at wind speeds of 6 mph, 10 mph and 14 mph, and stored gasoline
vapor pressures of 2 psi, 4 psi, 6 psi and 10,psi. Emissions reductions
are the difference between the single seal case (base case) and secondary
seal case (control case). The results for a model 100' 0 tank storing gasoline
whose hydrocarbon emissions have a molecular weight of 65.0 Ibs/lb mole are
given in Table B-2.
C.2 Shoe Seal With Shoe-Mounted Secondary
The shoe mounted secondary was tested on a primary shoe seal with a
vapor space leak rate of < 0.1 SCFM per ft of tank diameter,,(Tests C-l, C-2,
and W-1R). Emissions controlled in Test C-l and C-2 were calculated at various
wind speed and vapor pressure parameters. The emissions controlled were then
subtracted from the emissions in the base case, Test W-12, to determine emissions
from a shoe mounted secondary. The results are given in Table B-3.
B-3
-------
E
0
*
-
CM
0
LO
*
CM
C
1
C
o.
V
« «d- o
in LO cn
CM O r-
a- r- co
CO O 00
ON o Cn
cn in «a"
in «a- i
i O i
CM co cn
o o o
us «a- CM
LO i «3-
O t^ CO
co in r~-
CM r
CM i i
CM OO «a*
i CM CO
r
CM O CM
CO T^* *~
VO » Lf
in cn vo
i I-. CO
CO CD CV
CO *f
in i *;
CO CO CO
CO CO If
CO O C\
in cn 10
in co t
|-^ CD r
Experimental Test Test Number
Welded Tank-Shoe Seal
Single Seal . W-12
Rim Mounted Secondary (25% W-12)
AFmiisinns3
«a- CD
CM CO
co cn
CM i
§ s
O CP»
O if)
IO LO
3 LO
CO OJ
S g
i CM
O CM
co cn
in in
CM O
CM VO
co cr
cn vo
CM VO
CM VO
vo ir
i CM
cn VD
CM 0
P; S
CO t--
CD O
CO
cu
CM .a
i re
01-
r-J CU
o co
=t CO
O
U
cu
CO
-a
01
c
o
CO
o
S-:
CO
oo vo
r-^ i
cn CM
CM i
CO O
cn o
o vo
n vo
00 CO
CM CO
cn VD
CO i
a- vo
O CO
vo t-~
co cn
LO VO
r^ co
co "a-
LO CM
«* C.
CM CO
i O
CM r
" S.
CO f
*a° c
CO VD
CO i
CO CO
CO CD
CM i
r^. o
o *
i-^ CD
CO
CU
T re
01
c? cu
cn «* in
CM in t-~
a- co CD
CM i r
o in LO
cvi CM cn
in o in
CO CO LO
cvi co f
LO CVI CM
cn' co vo
LO LO O
co c:
13
cu o
i E U-
cn <
c E
co a:
^J- CO
O CD
CM en
<* co
cn in
CO LO
cn co
n r^
CM
co in
CM CM
LO CM
CD CD
in o
CD §8
r^ CM
CO
o cn
co cn
^i~ cn
CO CM
cn cv
r-^ cn
vo cn
cn pr
t CO
CO CN
cv
CM CO
VO CO
CM CN
i CO
CM r~-
1 LC
CO
3
re
-a
o
o
cu
CO
-o
01
3
o
E; LL
E
5
i
o
u_ cu
cu to
1^ S-
V- O
,
to i.
re
cu -a
i E U.
cn o
i- CU
to co
CO
cn
CO
CM
cn
0
vo
LO
cn
LO
CM
n CD *
LO t CO
n -3-
.a
cn t eo
r^ co -a-
CM «a- cc
:M cv
CO CM VO
co vo r-*
cn CM vo
CVI CV
«a- i co
VD CM ^*-
cn r-^ a
LO LO
o
cn vo co
cvi co' ai
o cn
CM i
r- cn co
CM CO LO
in r-^ o-
cn cn
i CO CO
cn o a
in u
CM r-- LC
vo o ur
CM CM
CO CC
CM =3-
CO
^-^ to
CO O.
CL re
re cn
cn
o
+j
'2 t
re
cu
r CO
re
cu >,
to i-
re
.£ 1 u
cn o
C 0
i O
co to
CO
CM
O
CM
CO
CO
LO
cn
Lo
CM
CD
CM
CO
CO
a-
CM
cn
CO
5
CO
CO
to
ci-
re
CO O
o.
>
re Z.
a a.
c
o t-
u cu
cu >
000
o
a Emission Reduction used in Chapter 4.
to
"o
o.
to
i.
B-4
-------
H-:
e
LUC
o
a.
E:
i
"S
CO
0.
to
O
c
3
O
*
CM
O
to
-
CM
O
to
CM
Stored Product Vapor Pressure-PSIA
I Experimental Test Test Number
i
fc
10
a
' C
o
(J
co
to
o
cu
<->
C 4->
^
CO -C
o en
toP
O O O * CD «d-
r- CO OO in CO CM
s ° £ 3 2 g>
S 5 § § § §
3- to en en en o
en co r m -sa-
in o m r- tn 10
co co m CM in 10
to CM «3- CD «a- to
en «3* in to in CD
CO r- CM 10 CM CO
en en CD CD o CD
CM to to en to co
O l-x CM CO CM r
CM r CM i r
CO ^- CM CM CM CD
m to en CM en co
en ^ co in r m in
en en CD CM o CM
px, r LO co to CM
in CM -TO to co co
£. t to tn to en
tO O tO r tO «3-
CM r IP CO ^- i
** CO tO CO to CM
* tn to i^ LO CM
o ^- in i tn to
r*»- tn CM co CM r
P-- t ij3 in to en
«=3- CM CM in CM CM
CD r- en to en r^.
en co ir> co to r-*
CM i , CO* ,
«^- o «st- in «^i- r
co to r-^ in ix. co
r CD O i O O
CM -3
5= S O 3 -^O
"^ fO O (O
S -0 0- T3
Jf C 0 !-> C
^- JT O t *t- O
Sen o CD L> a; *^o o E:CMO
OO -J UJ T- O ~O !- r T3 M- O
"" (1) P r t at
"U OJCOf r OJ r-r-tO
W V) O fO 4-> 1 O 03
s r i: O r 4- O
*- its en S:CMaJO
"a co t o j= >>co c: EH DT co jc >>
c cos- ujc_> coC
O OJ -^ QJ 03>i CO=>, n3
aJ ^ScS"0"0"^ CM -" J="a
tO CO _J UJ «r~ O "O *r r T3 «i O
30 i i co
+^ W) in Ora4->IOfO
3 d <_J c_i QJ +J CO cy C3L 4-* CO flJ
O tO 4J c; |^ -|_-i
JBg. CU O QJ C O QJ ry y Q QJ C
C1J -C co cos co_j| COE:
CO 5
tu>-
3 ce
1-1 UJ
to o
to :n
-< to
co
co
B-5
-------
REFERENCES
1. Summary of 20 foot Diameter Pilot Floating Roof Hydrocarbon
Emission Test Results, Chicago Bridge & Iron Company, December 28, 1977.
2. Hydrocarbon Emission Loss Measurements on a 20' 0 Floating Roof
Tank with a Type SR-1 Seal for a Product at Various Vapor Pressures, Chicago
Bridge & Iron Company, October 25, 1977.
3. Methods for Extrapolating Chicago Bridge & Iron Company 20' 0
Test Tank Results to Full Size Tanks, EPA, OAQPS Draft Report, April, 1978.
4. Reference 1, Op Cit.
B-6
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing!
1. REPORT NO.
EPA-45Q/2-7S-n47
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Control of Volatile Organic Emissions from
Petroleum Liquid Storage in External Floating Roof Tc
5. REPORT DATE
December 1978
6. PERFORMING ORGANIZATION CODE
nks
7. AUTHOR(S)
Richard K. Burr, ESED
Kerri C. Brothers, ESED
Jack G. Wright, SASD
8. PERFORMING ORGANIZATION REPORT NO
OAQPS No. 1.2-116
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emission Standards and Engineering Division
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report provides the necessary guidance for development of
regulations limiting emissions of volatile organic compounds (VOC) from
storage of petroleum liquids in external floating roof tanks. Reasonably
available control technology (RACT) is defined in this document; cost
analysis for RACT is included for evaluating the cost effectiveness of
controlling external floating roof tank sources.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Air Pollution
Storage Tanks
Regulatory Guidance
Air Pollution Control
Stationary Sources
Organic Vapors
8. DIS1
iUTION STATEMENT
19. SECURITY.CLASS (ThisReport)'
Unclassified
2O. SECURITY CLASS (This page)
Unclassified
21. NO. OF PAGES
64
22. PRICE
EPA Form 22201 (Rev. 477) PREVIOUS -EDITION is OBSOLETE
-------
-<>"*
------- |