DAQ-2024-006973

1

DAQC-261-24
Site ID 10335 (B4)

MEMORANDUM

TO: STACK TEST FILE – TESORO REFINING AND MARKETING COMPANY

THROUGH: Harold Burge, Major Source Compliance Section Manager

FROM: Paul Morris, Environmental Scientist

DATE: March 15, 2024

SUBJECT: Location: 474 West 900 North, Salt Lake City, Salt Lake County, Utah
Contact: Crystal Beffa – 606-369-4370
Tester: Advanced Industrial Resources, Inc.
Sources: Fluid Catalytic Cracking Unit (FCCU) Wet Gas Scrubber (WGS), Crude
Unit Furnace (N2C) H-101 and Ultraformer Unit (UFU) F-1
FRS ID #: UT0000004903500004
Permit# AO DAQE-AN0103350075-18 dated January 11, 2018
Subject: Review of Pretest Protocol dated March 12, 2024

On March 13, 2024 the Utah Division of Air Quality (DAQ) received a protocol for testing of the Tesoro
Refining and Marketing Company’s FCCU WGS, N2C H-101, and UFU F-1 located in Salt Lake City,
Utah. Testing will be performed the week of April 29, 2024, to determine compliance with the emission
limits found in AO Condition II.B.3.a, II.B.4.c, and II.B.7.a.

PROTOCOL CONDITIONS:

1. RM 1 used to determine sample velocity traverses: OK
2. RM 2 used to determine stack gas velocity and volumetric flow rate: OK
3. RM 3A used to determine dry molecular weight of the gas stream: OK
4. RM 4 used to determine moisture content: OK
5. RM 5B used to determine Non-sulfuric Acid PM emissions: OK
6. RM 7E used to determine NOx emissions: OK

DEVIATIONS: No deviations were noted.

CONCLUSION: The protocol appears to be acceptable.

RECOMMENDATION: Send protocol review and test date confirmation notice.

ATTACHMENTS: Stack testing protocol.

6 , 3
Salt Lake City Refinery
474 West 900 North
Salt Lake City, UT 84103-1494fii\\*ggY"
A subsidiary of Marathon Petroleum Corporation
March 12,2024
Mr. Paul Morris
Division of Air Quality
Department of Environmental Quality
195 North 1950 West
P.O. Box 144820
Salt Lake city, uT 84114-4820 Hand Submittat
Tesoro Refining and Marketing Company's
Salt Lake City Refinery
FCCU Coke Burn Emissions Test Protocol
Ultraformer F-l Furnace Nitrogen Oxides Performance Test Protocol
Crude H-101 Heater Nitrogen Oxides Performance Test Protocol
Dear Mr. Morris:
Enclosed please find the Fluid Catalytic Cracking Unit Wet Gas Scrubber (WGS) Method 58
Particulate Matter Test, the Ultraformer Unit (UFU) F-l Furnace Nitrogen Oxides Performance
Test, and the Crude Unit (N2C) Heater Nitrogen Oxides Performance Test Protocols scheduled
for the week of Apri129,2024.
The purpose of the Fluid Catalytic Cracking Unit Wet Gas Scrubber (WGS) Method 58
Particulate Matter Test is to demonstrate the compliance status of the source with respect to
NSPS Subpart Ja PM emission standard of I pound per 1000 pounds coke burn off. The PM test
results will be used to determine the new operational parameters for the WGS (differential
pressure and liquid-to-gas ratio) and the ESP (primary power and secondary current).
Please contact me at [phone] if you have any questions.
Sincerely,
Tesoro Refining &
Marketing Company LLC
.sffi,?fiEi,?ililtsi,
Li/'fl 1.i' ,,
14and
DI\,,ISION
(A'klB
.#, {1,,^ 1
Environmental Engineer
Enclosure
f-)
PllJ(s/.,,
^i\\*ry"
March 12,2024
Mr. Robert Leishman
Division of Air Quality
Department of Environmental Quality
195 North 1950 West
P.O. Box 144820
Salt Lake city, uT 84114
Tesoro Refining and Marketing Company's
Salt Lake City Refinery
COBB RATA Protocol
Dear Mr. Leishman:
Salt Lake City Refinery
474 West 900 North
Salt Lake City, UT 84103-1494
Tesoro Refining &
Marketing Company LLC
A subsidiary of Marathon Petroleum Corporation
I,TAH DEPARTMENT OF
ENVIRONMENTAL OUALIW
[,{AR 1 3 2A24
"k:t'^!e{*Environmental Engineer
Enclosure
il q, / Deii N,rer)
DIVISION OF AIR QUALTTY
HAND DELUERED
Enclosed please find the Protocol for the Relative Accuracy Test Audits (RATA) for the
redundant Carbon Monoxide Boiler Bypass (COBB) CO, 02, NOx and SO2 analyzer
scheduled for the week of Api129,2024.
Please contact me at [phone] if you have any questions.
Sincerely,
Srun Spnctrtc Tnsr PnnN
PTnTORMANCn TrSrSz FCC/WGS, UFU F-L, & N2C H-
I0t
Rnl,a.rrvr AccuRACv Trsr Auorr (RATA)z CO Bomn
Byp.,qss
Pnnpanpo Fon:
Mn,narHoN PnrnoLEUM ConpoRATIoN
TnSono RETTNING AND ManrrTING COvrp,I.NY
474 Wnsr 900 Nonnr, S^Lr,r L.qxE Crry, UT 84103
Pnppanso Bv:
AovaNCED [NousrRrAL RrsouRCEs, INC.
3407 Novrs Ponrrn
Acwonru, GnoRGr,q. 30101
PnrpaRpo: Mmcn 412024
Ts,sr Dares: Wunx or Apnn 2912024
MARATHOiI
Tlnr.n or CoNTENTS
1.0 INTRODUCTION 1
I.1 BACKGROUI\DINFORMATION
2.0 SOI]RCE DESCRIPTION
3.0 TEST PROGRAM SUMMARY
3.1
3.2
s.0 REFERENCE METHODS
RELATIVE ACCURACY TESTING OF CEMS
PERFORMANCE TESTING
4
5
4.0 SAMPLING LOCATIONS 8
6.0 OUALITY ASSURANCE / OUALITY CONTROL MBASURES 11
6.I GAS ANALYZER CALIBRATION
6.2 NOz-NO COIWERSION EFFICIENCY
6.3 INSTRUMENTINTERFERENCERESPONSE
6.4 INSTRUMENT RESPONSE TIME
6.5 SAMPLE HAI\DLING AIID CHAIN OT'CUSTODY PROCEDTIRES
6.6 SAMPLING TRAIN LEAK CHECK
6.7 PORBE NOZZLE DIAMETER CHECKS
6.8 PITOT FACE PLAIIB ALIGNMENT CHECK
6.9 METBRING SYSTEM CALIBRATION
6.10 TEMPERATT]RE GAUGE CALIBRATION
6.11 BAROMETRIC PRESSURE DATA ACQUISITION
6.12 DATA REDUCTION CIIECKS
6.13 EXTERNAL QUALITY ASSURANCE PROGRAM
6.13.I AUDIT SAMPLE PROCEDURES
6.13.2 SITE.SPECIFIC TEST PLAN EVALUATION
6.13.3 ON-SITE TESTEVALUATION
11
t2
12
13
13
t4
t4
15
15
16
t6
t6
t6
16
l7
t7
7.0 REPORTING AND DATA REDUCTION REOUIREMENTS 18
Tnrr,r or CoNTENTs
8.0 PLAI\T ENTRY AI\D SAFETY
8.1 SAFETYRESPONSIBILITIES
8.2 SAFETY PROGRAM
8.3 SAFETYREQUIREMENTS
9.0 PERSONNEL RESPONSIBILITIES AI\D TEST SCHEDULE 20
9.1 TEST SITE ORGANIZATION
9.2 TESTPREPARATIONS
9.3 TEST PERSONNEL RESPONSIBILITIES
9.4 TEST SCHEDULE
19
19
19
20
20
20
2t
Appendices:
Appendix A:
Appendix B:
Final Test Report Outline
Example Calculations and Nomenclature
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page I of27
1.0
1.1
Introduction
BACKGROUND INFORMATION
Tesoro Refining and Marketing Company Marathon Petroleum Corporation operates the
Marathon Petroleum Salt Lake City Refinery (Marathon) in Salt Lake City, Utah located
at 474 West 900 North, Salt Lake City, Utah 84103. The facility includes a variety of
petroleum refining processes and associated equipment and is permitted under the Utah
Department of Environmental Quality Division of Air Quality (UDAQ) issued Approval
Order (AO) Number DAQE-AN103350075-18 issued January 11, 2018.
In accordance with the applicable sections of the referenced Approval Order periodic
performance testing and relative accuracy test audits (RATAs) are required on various
process emissions and gaseous continuous emissions monitoring systems (CEMS).
Specifically in this test program, performance tests will be conducted on the exhaust
stacks of the Fluid catalytic Cracking Unit (FCCU) Wet Gas Scrubber (WGS) system
stack, Crude Unit Furnace Q.{2C) H-101, and the Ultraformer Unit (UFU) Fumace F-1.
Additionally, a RATA will be conducted on the secondary CO Boiler Bypass CEMS for
annual certifi cation purposes.
Marathon has contracted Advanced Industrial Resources, Inc. (AIR) to perform the
referenced performance tests and CEMS RATAs. The offrces of AIR are located at 3407
Novis Pointe, Acworth, Georgia 30101, but all correspondence should be directed to P.O.
Box 846, Marietta, Georgia 30061. If there are any questions or comments concerning
this Site-Specific Test Protocol, please contact either Ms. Crystal Beffa of Marathon or
Mr. Ross Winne of Advanced Industrial Resources.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: Much9,2024
Page 2 of 27
I.2 KEY PERSONNEL
The key personnel who will be coordinating and executing this Test Protocol and their
telephone numbers are:
Crystal Beffa, Environmental Engineer, Marathon
Morgan Sites, Environmental Engineer, Marathon
Ross Winne, Technical Director, AIR
Derek Stephens, VP - QA Director, AIR
John Hendry, Chief Operations Officer, AIR
Scott Wilson, Program Director, AIR
(606) 36e-4370
(38s) 306-461s
[phone]
(404) s20-s028
[phone]
(800) 224-s007
Advanced Industrial Resources, [nc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 3 of 27
2.0 Source Description
The FCCU Regenerator operates in conjunction with the CO Boiler which is a CO
oxidizer as well as a Heat Recovery Unit. The system's emissions are controlled by
COAIO* oxygen injection, arnmonia injection, electrostatic precipitator (ESP), and wet
gas scrubber/LoTOx system (WGS). The emissions from the WGS are vented to
atmosphere via single stack. The FCCU is equipped with redundant CEMS that
continuously monitor the emissions of oxygen (Oz), nitrogen oxides (NO*), sulfur dioxide
(SOz), and carbon monoxide (CO) as well volumetric flow rate. In accordance with AO
Condition II.B.4.c, II.B.4.f and 40 CFR 60 Subpart Ja, the particulate matter (non-sulfuric
acid PM) emissions of the FCCU WGS are limited to 1.0-pound PM (non-sulfuric) per
1000 pounds ofcoke burned based upon a 3-hour average.
The CO Boiler Bypass stack is also equipped with a CEMS designed to continuously
monitor the emissions of Oz, NO*, CO, and SOz.
The Ultraformer Unit (UFU) Furnace F-l and Crude Unit Fumace H-101 are each fired
with plant gas and employ ultra-low NO* burners to control NO* emissions. In
accordance with AO Conditions ILB.3.a (H-101) and II.B.7.a (F-l), the respective stacks
are tested annually to determine the correct emission factor for use in calculating the NOx
emissions for contribution towards the facility's NO, Cap emissions rate of 1.988 tons
per day and to demonstrate compliance with their respective NOx emission limits.
Specifically, NO* emissions from CFU H-I01 are limited to 0.054 lbilvlMBtu based on a
3-hour average as listed in AO Condition II.B.3.a while NO* emissions from UFU F-l
are limited to 0.065 lbAvlMBtu based on a 3-hour average as detailed in the Tesoro
Consent Decree lodged July 18,2016.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: Much9,2024
Page 4 of 27
3.0 Test Program Summary
3.I RELATIVE ACCURACY TESTING OF CEMS
Relative accuracy test audits (RATA) will be conducted on the back-up (redundant)
CEMS operated on the CO Boiler Bypass stack. Primary analyzer testing was completed
in January 2024. Below is a summary of the CO Boiler Bypass' CEMS.
CO Boiler Bvpass
Parameter: Oz / SOz A.{Ox - Redundant
Make: Horiba
Model: GI-761CU
SN: Y4WE0JGY
Spans: 0-25% 02
0-200 & 0-1,000 ppmvd SOz
0-200 & 0-1,000 ppmvd NO*
0-1,000 ppmvd CO
The table below provides of list of the source CEMS where RATAs will be conducted
including the methodologies to be employed and relative accuracy units of comparison:
Table 3-l: RATA Test Sumeel ma
Source CEMS
Parameter
Method /
Performance
Specification
Sampling
run
durations
(min.)
Repetitions Em.
Standard
Relative
Accuracy
requirement
CO
Boiler
Bypass
Oz Method 3,{ /
PS3 >21 9-12 NA +1.0 diff. or
20%RA
SOz
Method 6C /
PS2
>21 9-t2
18 ppm
@0% Oz
-7 day
rolling
avs.
20% RA or
10% using
em. Std. in
RA calc.
NO*Method 7E /
PS2
>21 9-t2
20 ppm
@0%oz
-7 day
rolling
avg.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 5 of 27
CO Method l0 21 9-t2
500 ppm
@0%02
(7-dav)
<:100
(RM) or <:
5% (AS)
AIR will conduct reference method tests so they will yield results representative of the
pollutant concentrations from the referenced source emission stacks which can be
correlated with the respective continuous emission monitoring system measurements.
Each RATA will consist of a minimum of nine (9) separate test runs, each a minimum of
twenty-one (21) minutes in duration. RATAs are conducted in accordance with
respective 40 CFR Part 60 Appendix A promulgated methodologies and Appendix B
Performance Specifications. Methodologies are described in further detail in Section 5.0.
Relative accuracies for the CEMS will be assessed in the required emission standard units
of reporting as listed in Table 3-1 above.
To properly correlate individual CEMS data with the reference method data, the
beginning and end of each reference method test run (including the exact time of day)
will be annotated on the individual chart recorder(s) or other permanent recording
device(s).
AIR will confirm that the monitor or monitoring system and reference method test results
are on consistent moisture, pressure, temperature, and diluent concentration basis. AIR
will also consider the response times of the continuous emission monitoring systems to
ensure comparison of simultaneous measurements.
The facility will ensure that the unit is operating at greater than 50%o of its normal
operating load and the CEMS data acquisition system is generating continuous l-minute
averages with time-stamps throughout the RATA.
A RATA cannot routinely be performed on the CO Boiler Bypass CO CEMS because the
analyzer is operated above span during normal operation and is maintained through a
routine CGA. If the bypass stack were to be utilized, Operations would shift the FCCU
to full bum mode resulting in CO within range of the CEMS.
3.2 PERFORMANCE TESTING
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 6 of27
Emissions performance tests will be conducted on the FCCU WGS, UFU F-1, and N2C
H-101 stacks to demonstrate compliance with applicable emission limits and to establish
annual emission factors, as applicable. Testing will be conducted in accordance with the
methodologies presented in 40 CFR 60 Appendix A and this test protocol. Each
performance test will be conducted via 3 repetitions lasting for minimum durations of 60
minutes per test run.
The facility will ensure the units are operating at the desired rates and stabilized at least
15 minutes prior to commencing the performance tests. Additionally, all applicable
process and control device operating data shall be monitored and recorded throughout
testing for inclusion into the final test report. The facility will also provide the plant gas
F-factor (dscf/MMBtu) for use in calculating the emissions in the units of lb
NOx/MMBtu for N2C and UFU.
Methodologies to be employed are delineated in Section 5.0. Table 3-2 below
summarizes the test program to be conducted on the respective stacks.
Table 3-2: Performance Test Program Summary
Source Analytes
(units of reporting)
Sampling
Methods
Sampling
run
duration
(minutes)
Repetitions
(per test)
Emission
standards
FCCU Wet
Gaq Scnrhher
Volumetric flow
rate (dscfm)
EPA Methods
t-4
>60 a
J
1.0 lb PM i
1,000 lbs
Coke
burned
Oxygen / Carbon
Dioxide (%, dry)
EPA Method
3A
Non-sulfuric Acid
Particulate Matter
(mg/dscm, lb/hr, lb
PNl/1,000 lbs coke
burned)
EPA Method
5B
UFU F-I
Oxygen (%, dry)EPA Method
3A
>60 J
0.065
lb/MMBtu
based on a
3-hour
average
Nitrogen Oxides
(ppmvd)
EPA Method
7E
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
PageT of27
Heat input-based
emissions
(lb NOx AvIMBtu)
EPA Method
l9
N2C H-l0l
Oxygen (%, dry)EPA Method
3A
>60 3
0.054
lbA{MBtu
based on a
3-hour
average
Nitrogen Oxides
(ppmvd)
EPA Method
7E
Heat input-based
emissions
(lbNOxIvIMBtu)
EPA Method
t9
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: Merch9,2024
Page 8 of 27
4.0 Sampling Locations
The stack outlet sampling locations are all located at least two (2.0) equivalent diameters
downstream from the nearest upstream flow disturbance and at least one-half (0.5)
equivalent diameters upstream from the nearest downstream flow disturbance or stack
exhausts. Each of the exhaust stacks are circular and are equipped with a minimum of
two sampling ports oriented 90 degrees to one another on a parallel plane-oriented
perpendicular to the exhaust gas flow direction. A maximum of twenty-four sampling
points (twelve traverse points in each of the two sampling ports) will be used for EPA
Methods 2, 3, 4, and 58 sampling on the FCCU/WGS stack, in accordance with EPA
Method 1 requirements.
A stratification test will be conducted in accordance with Section 8.1.2 of Method 7E
when conducting performance tests using instrumental methods (EPA 3A, 6C,7E)) to
determine the quantity and location of the traverse point locations. When conducting the
annual RATAa, the quantity and location of traverse points will be determined in
accordance with the results of the pre-test stratification check described in Performance
Specification 2, Section 8.1.3.2 where a minimum of three (3) traverse points will be used
during each RATA.
*It is noted that when Method 3,A' is being conducted to determine molecular weight of
the stack gases only, in accordance with Section 8.1, a single integrated traverse point
may be used and no stratification assessment conducted.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 9 of 27
5.0 Reference Methods
RATA and Performance Tests will be conducted according to the methodology detailed
below, as applicable:
. EPA Method I will be used for the qualification of the location of sampling ports
and for the determination of the number and positions of stack traverse points, as
applicable to sample traverses for Method 2.
o EPA Method 2 will be employed for the determination of the stack gas velocity
and volumetric flow rate during stack sampling using the Type ooS" Pitot tube.
EPA Methods 3 and 3A will be used for the calculation of molecular weight of
the stack gases and for the determination of the oxygen and carbon dioxide
emission concentrations using properly calibrated instrumental analyzers.
o EPA Method 4 will be used for the determination of moisture content.
o EPA Method 58 will be used on the FCCU/WGS for the determination of non-
sulfuric acid particulate matter. The probe, filter box, and filter exit gas stream
will be maintained at 320+25oF throughout testing. The sample recovery reagent
will be reagent grade acetone. Additionally, prior to gravimetric analysis, the
filter will be cooked for 6 hours at 320oF and then desiccated for 2-3 hours prior
to obtaining a constant weight.
o EPA Method 6C will be used for the determination of sulfur dioxide (SOz)
emissions using a properly calibrated instrumental analyzer. The instrument
detection limit is 0.1 ppm while the method detection limit is 2%o of the span
value.
o EPA Method 7E will be used for the determination of nitrogen oxides (NO,
emissions and where applicable, reported as NOz. A converter efficiency test will
be completed at least once per day utilizing a NOz cylinder of concentration 40 to
60 ppm. The instrument detection limit is 0.1 ppm while the method detection
limit is 2Yo of the span value.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: Much9,2024
Page l0 of27
. EPA Method 19 will be used to calculate the emissions in the units of lb
NOxA4MBtu using the facility provided F-factor (dscf/MMBtu) and Equation 19-
1 of Method 19.
. EPA Method 205 may be used to verift gas dilution system certification.
o Performance Specification 2 will be used to assess the relative accuracy of the SOz
and NO* CEMS.
o Performance Specification 3 will be used to assess the relative accuracy of the Oz
CEMS.
o Performance Specification 4 will be used to assess the relative accuracy of the CO
CEMS.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page ll of27
6.0 Quality Assurance / Quality Control Measures
The quality assurance/quality control (QA/QC) measures associated with the sampling
and analysis procedures given in the noted EPA reference methodology and in Subparts
A of 40 CFR Part 60 and in the EPA QA/QC Handbook, Volume III (EPA 600/R-
941038c) will be employed, as applicable. Such measures include, but are not limited to,
the procedures detailed below.
6.I GAS ANALYZER CALIBRATION
6.1.1 Calibration Gas Concentration Verification Calibration gases that are analyzed
following the Environmental Protection Agency Traceability Protocol No. I will be used.
Certification from the gas manufacturer that Protocol No. 1 was followed will be obtained.
6.1.2 Measurement System Preparation AIR will assemble each measurement system
by following the manufacturer's written instructions for preparing and preconditioning each
gas analyzer and, as applicable, the other system components. l.IR will make all necessary
adjustments to calibrate the analyzers and the data recorders.
6.1.3 Analyzer Calibration Error AIR wilrl conduct the analyzer calibration elror check
by introducing calibration gases to the measurement system upstream of each gas analyzer.
After the measurement system has been prepared for use and immediately prior to starting
the RATA, AIR will introduce the zero and high-range to calibrate the analyzer to establish
calibration curve. Then the mid-range gas will be introduced into the analyzer to check
linearity. During this check, AIR will make no adjustments to the system except those
necessary to achieve the correct calibration gas flow rate at the analyzer.
6.1.4 Sampling System Bias CheckAIR will perform the sampling system bias check by
introducing calibration gases at the calibration valve installed at the outlet of the sampling
probe. Immediately prior to starting each RATA run, a zero gas and the mid-range gas
(which most closely approximates the effluent concentrations) will be used for this check.
AIR wtll introduce the zero calibration gas and record the gas concentration displayed by
the analyzer. AIR will then introduce mid-range calibration gas and record the gas
concentration displayed by the analyzer. During the sampling system bias check, AIR w:,ll
operate the system at the normal sampling rate and make no adjustments to the
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 12 of27
measurement system other than those necessary to achieve proper calibration gas flow rates
atthe analyzer.
6.1.5 Zero and Calibration Drift Checks At the end of each RATA test run and
whenever adjustments are necessary for the measurement system, AIR will repeat the
sampling system bias check procedure described in Section 6.1.4. If either the zero or
upscale calibration value exceeds the specification in Section 6.1.6, then the sampling
data from the previous run will be considered invalid and AIR will repeat both the
arnlyzer calibration error check procedure and the sampling system bias check procedure
before continuing with additional data collection.
6.1.6 Analyzer Error, Bias and Drift Check Specifications Analyzer calibration error
will be less than +l-2 percerrt of the span for the zero, mid-range, and high-range calibration
gases. Sampling system bias will be less than +/-5 percent of the span for the zero and
mid-range calibration gases. Zero drift will be less than */-3 percent of the span over the
period between zero drift checks. Calibration drift will be less than +/-3 percent of the span
over the period between calibration drift checks.
6.2 NOz-NO CONVERSION EFFICIENCY
Prior to the test, AIR will evaluate the NOz to NO conversion efficiency of the reference
method analyzer. AIR will introduce a concentration of 40 to 60 ppmv NOz certified gas
directly into the calibration valve assembly of the analyzer. The analyzer's response
must show an NOz-NO conversion efficiency of at least 90%. The conversion efficiency
will be calculated using Equation 7E-7 presented in Section 12.7 of Method 7E. If the
conversion efficiency of the analyzer is found to be less 90%o, corrections will be made to
the converter before repeating the check or the analyzer will be replaced.
6.3 INSTRUMENTINTERFERENCERESPONSE
AIR wrll obtain instrument vendor data that demonstrates the interference performance
specification is not exceeded for the test gases of EPA Method TF,Table 7E-3.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: Much9,2024
Page 13 of27
6.4 INSTRUMENT RESPONSE TIME
To determine response time, AIR will conduct a response time test. The response time
test will be repeated if changes are made in the measurement system. l1R will first
introduce zero gas into the system at the calibration valve until all readings are stable;
then, switch to monitor the stack effluent until a stable reading is obtained. AIR will
record the upscale response time. Next, AIR will introduce high-level calibration gas into
the system until the system has stabilized at the high-level calibration concentration, then
switch to monitor the stack effluent and wait until a stable value is reached. AIR will
record the downscale response time. AIR will repeat the procedure three times. A stable
value is equivalent to a change ofless than one (1) percent ofspan value for 30 seconds
or less than five (5) percent of the measured average concentration for two (2) minutes.
The greater of the average upscale or downscale response times will be taken as the
"response time" for each analyzer.
6.5 SAMPLE HANDLING AND CHAIN OF CUSTODY PROCEDURES
All samples are transported in a closed sample box, the security of which is the
responsibility of the Test Director. These samples are received, checked, and numbered
by the Test Director and custody records are written. The QA Director then checks again
the integrity of the samples and their identifications.
For samples that require off-site analysis, the samples collected during the emissions
testing projects, will be stored in appropriate containers with sufficient insulation to
prevent breakage during shipping. All samples in a shipping container will be listed on
the chain-of-custody form enclosed in the shipping container. Once the samples are
securely packaged, the container will be sealed with tape and several custody seals will
be placed over the top edge so that the container cannot be opened without breaking the
custody seal.
If samples are not driven directly to the laboratory by AIR personnel, the samples will be
shipped via courier service (e.g., Federal Express, UPS, etc.) and will include a separate,
signed chain-of-custody form enclosed in the shipping container. A shipping document
for the courier service will be completed for each shipment. Shipments will occur
approximately on a daily basis to minimize the storage of samples at the site and to
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 14 of27
maximize the holding time period available to the laboratory for sample extraction and
analysis.
6.6 SAMPLING TRAIN LEAK CHECK
Determinations of the leakage rate of the sampling train are made before and after each
sampling run using the procedure detailed in Section 8.1.4 of EPA Method 5. Before the
sampling run, after the sampling train has been assembled and probe and filter box
temperatures have had time enough to settle at their appropriate operating values, the
probe nozzle will be plugged and the system will be evacuated to a pressure of 15 inches
of Hg below ambient pressure. The volumetric leakage rate will be measured by the dry
gas meter over the course of one (1) minute. If the leakage rate is less than 0.020 cfm,
then the test proceeds. After the sampling tun, before the train is disassembled the probe
nozzle will be plugged and the system will be depressurized to a vacuum equal to or
greater than the maximum value reached during the sampling run. The volumetric
leakage rate will be measured by the dry gas meter over the course of one (l) minute. If
the leakage rate is less than 0.020 cfm, then the test run will be considered valid. If a
component change must be made during the test run, a leak check will be conducted
immediately before and after the component change is made using the procedure outlined
above.
The Type "S" Pitot tube assembly will also be checked for leaks before and after
sampling runs using the procedure in Section 8.1 of EPA Method 2. The impact opening
of the Pitot tube will be blown through until a pressure of at least 3 inches of water
registers on the manometer. The impact opening will quickly be plugged and held for at
least 15 seconds, during which time the manometer reading must hold. The same
operation will be performed on the static pressure side of the Pitot tube, except suction
will be used to obtain the pressure differential.
6,7 PORBE NOZZLE DIAMETER CHECKS
Probe nozzles will be calibrated before each field use by measuring the internal diameter
of the nozzle entrance orifice along three different diameters. Each diameter is measured
to the nearest 0.001 inch, and all measurements will be averaged. If the total variation of
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: Much9,2024
Page 15 of27
the three measurements exceeds 0.004 inch, the nozzle shall be reshaped, sharpened, and
recalibrated before further use.
6.8 PITOT FACE PLANE ALIGNMENT CHECK
Before field use, each Type S Pitot tube is examined in order to verifu that the face planes
of the tube are properly aligned, per Method 2 of 40 CFR Part 60, Appendix A. The
external tubing diameter and base-to-face plane distances are measured in order to verifu
the use of 0.84 as the baseline (isolated) pitot coefficient. At this time the entire probe
assembly (i.e., the sampling probe, nozzle, thermocouple, and Pitot tube) is inspected in
order to veriff that its components meet the interference-free alignment specifications
given in EPA Method 2. If the specifications are met, then the baseline pitot coefficient
is used for the entire probe assembly.
After each field use, the face plane alignment of each Pitot tube is checked. If any
damage has occurred to the tube orifices, the tube will be reshaped, realigned, and
recalibrated before further use.
6.9 METERING SYSTEM CALIBRATION
Every three months each dry gas meter (DGM) console is calibrated at five orifice
settings according to Method 5 of 40 CFR Part 60, Appendix A. From the calibration
data, calculations of the values of Ym and AH6 are made, and an average of each set of
values is obtained. The limit of total variation of Y, values is +0.02, and the limit for
AH6 values is +0.20.
After each field use, the calibration of the DGM console is checked by performing three
calibration runs at a single intermediate orifice setting that is representative of the range
used during field testing. The limit of acceptable relative variation from Y", is 5.0%.
Alternate post-test DGM calibration verifications may also be conducted in accordance to
EPA's EMC Approved Alternative Method 5 Post-test Calibration procedure identified as
ALT-009 where the post-test DGM calibration check value (Yqu) must be within + 5o/o of
the DGM's established Y-.
Advanced Industrial Resources. Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 16 of27
6.10 TEMPERATURE GAUGE CALIBRATION
After each field use, the temperature measuring instruments on each sampling train are
calibrated against standardized mercury-in-glass reference thermometers. The limit of
acceptable variation between the absolute reference temperature and the absolute
indicated temperature is 1.5%.
6.1I BAROMETRIC PRESSURE DATA ACQUISITION
The barometric pressure at the test site is measured by contacting the National Weather
Service prior to and immediately following the testing. Reported barometric pressure
will be adjusted for actual elevation at the test site.
6.12 DATA REDUCTION CHECKS
The Test Director will conduct spot checks on-site to assure that data are being recorded
accurately. After the test, the Technical Director will check the data input to assure that
the raw data have been transferred to the computer accurately. The Technical Director
will run an independent check (using a validated computer program) of the calculations
with predetermined data before the field test. Volumetric flow rates, velocities, moisture
levels, and emission concentrations and rates will be compared to test results from
previous sampling events, as applicable and made available, to identifr potential
inconsistencies in test data.
6.13 EXTERNAL QUALITY ASSURANCE PROGRAM
6.T3.1 AUDIT SAMPLE PROCEDURES
No audit samples are currently required for this test program.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 17 of27
6.13.2 SITE.SPECIFIC TEST PLAN EVALUATION
ln submiuing this Site-Specific Test Plan in advance of testing, Marathon and regulatory
personnel have been afforded the opportunity to review and comment upon the test and
quality assurance procedures to be used in conducting this Site-Specif,rc Test Plan.
6.13.3 ON-SITE TEST EVALUATION
In submitting the schedule in Section 9.4 of this Site-Specific Test Plan, Marathon and
regulatory personnel have been afforded the opportunity for on-site evaluation of all test
procedures . AIR will provide all Marathon and regulatory personnel complete access to
all activities during the execution of this Site-Specific Test Plan to ensure transparency of
the process.
Advanced lndustrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page l8 of27
7.0 REPORTING AND DATA REDUCTION REQUIREMENTS
Upon completion of field-testing activities and upon receipt of all CEMS data, AIR will
reduce the collected data and prepare a Final Test Report. To demonstrate the relative
accuracy of the CEMS, AIR will calculate relative accuracy in accordance with 40 CFR
Part 60 Appendix B Performance Specification 2, 3, and 4, as applicable, to provide
sample calculations and summarize calculation results.
An outline of the Final Test Report and a brief explanation of points are presented in
Appendix A. The Final Test Report will be submitted to Ms. Crystal Beffa of Marathon,
who will forward the required copies to the appropriate regulatory personnel within sixty
(60) calendar days of completing the test field data collection activities, unless otherwise
required or requested.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page 19 of27
8.0
8.1
Plant Entry and Safety
SAFETY RESPONSIBILITIES
The Test Director is responsible for ensuring compliance with plant entry, health, and
safety requirements. The Facility Contact has the authority to impose or waive facility
restrictions.
SAFETY PROGRAM
AIR has a comprehensive health and safety program that satisfies Federal OSHA
requirements. The basic elements include: (1) written policies and procedures, (2)
routine training of employees and supervisors, (3) medical monitoring, (4) use of
personal protection equipment, (5)hazard communication, (6) pre-mobilization meetings
with Marathon personnel, and (7) routine surveillance of the on-going test work.
8.3 SAFETY REQUIREMENTS
The sampling locations on the exhaust stacks will be made safely accessible by elevators,
stairwells, or permanent ladders. There is little risk of exposure to harmful amounts of
toxic materials at the sampling locations. However, test personnel will be instructed to
minimize their exposure to exhaust gases at all times.
All test personnel will adhere to the following standard safety and precautionary
measures as follows:
Confine selves to test area only.
Wear hard hats at all times on-site, except inside the equipment resource vehicle.
Wear steel-toed shoes or boots, protective shatter-resistant glasses or goggles,
FRC, H2S monitor, and hearing protection as appropriate at the test site.
Know the locations of readily available first aid equipment and fire extinguishers.
Be familiar with Emergency Response Procedures at the facility, including
evaluation signals and procedures. Test personnel will maintain constant contact
capabilities with mill personnel by utilizing facility provided two-way radios.
8.2
a
a
a
a
o
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page20 of27
9.0 Personnel Responsibilities and Test Schedule
9.1 TEST SITE ORGANIZATION
The key tasks and task leaders are:
Facility Contact: Crystal Beffa, Marathon
Facility Contact: Morgan Sites, Marathon
Program Director: Scott Wilson, AIR
VP/QA Director: Derek Stephens,llR
Technical Director: Ross Winne, AIR
Test Director: Andy Winkler,llR
Report Preparation Director: Steven Haigh, AIR
Operations/Safety Director: John Hendry, AIR
Laboratory Director: Russell Barton, AIR
9.2 TEST PREPARATIONS
9.2.1 Modifications to Facility. There are no facility modifications required for
this site.
9.2.2 Services Provided by Facility. Marathon will furnish sufficient electrical
utilities to adequately energize all of the necessary sampling and associated
equipment. AIR will provide all other services.
9.2.3 Access to Sampling Sites. There are no special requirements, problems, or
safety issues in gaining access to the remaining testing locations.
9.3 TESTPERSONNELRESPONSIBILITIES
AIR personnel will arrive at the plant about 1.5 hours before the start of the first run on
each of the days scheduled for sampling. Pretest activities on these days will include the
following:
o Meet with the Facility Contact to review the test objectives.
Advanced Industrial Resources, Inc.
Site-Specific Test Plan - Performance Tests & CEM RATA
Marathon Tesoro SLC Refinery
Revised: March9,2024
Page2l of27
o Prepare and set-up including leak checks the sample trains and analyzers at
sampling location.
o Calibrate instrumental analyzers and verifr that the data acquisition systems
functioning properly.
o Veriff communication links between team members/leaders/plant personnel.
The Facility Contact will:
o Verifu that the sources have stabilized at the designated load for at least 15
minutes prior to the start of the tests.
Verifu that the sources are operating at the specified load throughout testing, and
document the reasons if normal load is not achieved or if source operations are
otherwise disrupted in such a way as to invalidate test results for any time periods.
As applicable, ensure the CEMS is operating and recording data in accordance
with the facility's monitoring plan.
TEST SCHEDULE
The test program is scheduled to be conducted the week of April 29,2024, using the
following week as contingency. The test crew will set up at the facility, establish facility
personnel communications and perform initial leak, calibration and response time tests at
least several hours prior to commencing testing on each source. Testing is scheduled to
be completed on weekdays unless otherwise necessitated by facility operations.
In the weeks prior to the testing, a schedule will tentatively be defined and upon request
can be conveyed to regulatory personnel. However, it should be noted that while efforts
are made to adhere to a specific daily test schedule, due to unforeseen facility process
scheduling or equipment issues, the schedule may need to be modified accordingly at any
given point.
the
9.4
Advanced Industrial Resources, Inc.
Appendix A
Final Test
Report Outline
Ex^{vrpr,r OUTLINE oF THE FrNa.r Trsr Rrronr
1.0 Introduction
o Summary of Test Program
o Key Personnel
2.0 Plant, Process, and Sampling Location Descriptions
o Process and Operation Descriptions
o Sampling Locations
3.0 Summary and Discussion of Test Results
. Objectives
o Field Test Changes and Problems
o Presentation of Test Results
o Process Monitoring
4.0 Sampling and Analytical Procedures
5.0 Quality Assurance Activities
o Internal Quality Assurance (equipment, methodology, etc.)
o External Quality Assurance (SSTP and on-site evaluation)
6.0 Data Quality Objectives
Appendices
Appendix A Test Results
Appendix B Field Data Reduction
Appendix C Example Calculations and Nomenclature
Appendix D Raw Field Data
Appendix E Laboratory Analytical Reports
Appendix F Facility Process Monitoring Data
AppnNurx B
Exaupm CarcuLATroNS
AND NoUTNCLATURE
Exavrpl,n C,LrcuLATIoNS
An:Dn2nl4
Ar:Dr2 x I 4
B*r: Vw(std) i (V.frtal * V*(sto))
canaryte : ((rl*ulyt. / Vqrta)) (35.31466 ft3lm3;
'canaryte : (IrLnalyte / V-Gta)) (0.015432 grlmg)
canalyte :'canalyte MWanalyte 1 24.04 Umol
CC : to.szs (So 1 nrrzl
d: l/n lldr)
DE : (E,n,., - Eou,t.r) / E,rnt., x l00o/o
Eanaryte: (m*4yt. / V.rsto)) Qro (60 min/hr) (2.2046x10-6 lb./mg)
Eanaryte: canaryte Q,a (60 mir/hr) (2.2046x10-6 lb./mg)
I : 100 T' (K: Vr. * Y. V. P, / T,,) I (60 0 v, P, An)
where K: : 0.002668 (in. Hg ft3) / (mL "R)
Kr : [(2.0084x107 AHo) A" (1 - B*,)]',(Mo / M,) (T. / T,) (P, / P,")
Ma : 0.44 (% COz) + 0.32 (% Oz) + 0.28 (% Nz + % CO)
Mr: Mo (1 - B*') * M* B*,
P : Qsd / F-Factor x 60 x (20.9-Oz) I 20.9
P.: Pu. * LH I 13.6
Pr:Pu*+prl13.6
Qu: (60 s/min) v, A.
Q,a : (60 s/min) (1 - B*,) v, A' (T',0 / T') (P' / P'to)
RA: [Abs(d) + Abs(CC)]/RM
So : [(Id'' - (Idi)' ln)l (n-l)]ttz
Trn:tn +460o
Tr: t, + 460"
Vm(std) : 17.636 V* Yrn P* / T.
Vw(std) : Ml. 0.04716)
vs: Kp C, [Ap]"' [T, / (P, Mr)]"'
NovrnxcLATURE
Symbol Units Description
Abs(x)dimensionless Absolute value of parameter x
An ftz Area of the nozzle
A.1f Area of the stack
Btns d mensionless Volume proportion of water in the stack gas stream
cp d mensionless Type S pitot tube coefficient
Canalyte mgldscm Concentration of analyte in dry stack g?S,
standardized
I Canalyte gr./dscf Concentration of analyte in dry stack goS,
standardized
rCanalyte ppm Concentration of analyte in dry stack g&S,
standardized
CC dimensionless One-tailed 2.5Yo enor confidence coefficient
d ppm Arithmetic mean of differences
di ppm Difference between individual CEM and reference
method concentration value
Dn inches Internal diameter of the nozzle at the entrance orifice
D.inches Intemal diameter of the stack at sampling location
DE percent Destruction efficiencv
AH inches HzO Average pressure differential across the meter orifice
AH6 inches HzO Orifice pressure differential that corresponds to 0.75
cfm of air at 68 oF and29.92 inches of Hg
Ap inches HzO Velocity head of stack gas
Eanalyte 1b./hour Emission rate of analyte, time basis
I percent Isokinetic sampling ratio expressed as percentage
Kr dimensionless K-factor, ratio of DH to DP, ideal
Kp ft[(lb/lb-mol)(in.
Hs)lt''
Type S pitot tube constant,
s[("R)(in. HzO)]r/2 : 85.49
Lp cfm Measured post-test leakage rate of the sampling train
Mo lb./lb.-mole Molecular weight of gas at the DGM
M.lb./lb.-mole Molecular weight of gas at the stack
NovTTNCLATURE
Symbol Units Description
Mtn lb./lb.-mole Molecular weight of water,
: 18.0
Illanalyte mg Mass of analyte in the sample
n dimensionless Number of data points
P MMBtu Fuel firing rate
Pnr.inches Hg Barometric pressure at measurement site
Pinput tonslhour Process dry mass input rate
Pg nches HzO Gauge (static) pressure of stack gas
P,nches Hg Absolute pressure of meter gases
P.inches Hg Absolute pressure of stack gases
Pstd inches Hg Standard absolute pressure
:29.92
Q,cfm Volumetric flow rate of actual stack gas
Qro dscfm Volumetric flow rate of dry stack gas, standardized
R (in. Hg)(ft')Ideal gas constant,
(lb-moleX'R): 21.85
RA percent Relative accuracy
RE percent Removal efficiency
RM ppm Average reference method concentration
fw 1b/ml-Density of water,
:0.002201
fa glmL Density of acetone,
:0.7899
Sa dimensionless Standard deviation
T,OR Absolute temperature of dry gas meter
Ts "R Absolute temperature of stack gas
Tstd OR Standard absolute temperature,
:528
to.szs dimensionless 2.5 percent error t-value
t-"F Temperature of DGM
ts OF Temperature of stack gas
e minutes Total sampling time
NovlnUCLATURE
Symbol Units Description
Mtc oE Total mass of liquid collected
V,dcf Volume of gas sample as measured by the DGM
Vm(std)dscf Volume of gas sample as measured by the DGM,
standardized
Vw(std)scf Volume of water vapor in the gas sample,
standardized
Vs ft./sec Velocity of stack gas
Y*d mens onless DGM cal bration coefficient
Y.d mens onless DGM cal bration check value
Y*d mensionless Reference (wet) gas meter calibration coefficient
o/o COz percent Percent COz by volume, dry basis
o/o Oz percent Percent Oz by volume, dry basis
7o Nz percent Percent Nz by volume, dry basis