United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-82-049 Sept. 1982
Project Summary
Evaluation of the Maintenance
Effect on Fugitive Emissions
From Refineries in the South
Coast Air Quality
Management District
R. L Honerkamp, M. L. Schwendeman, and L P. Provost
Seven areas of interest were inves-
tigated in a screening study coupled
with an inspection of maintenance
records and a historical review of the
effects of previous inspections and
maintenance performed under the
South Coast Air Quality Management
District (SCAQMD) Rule 466.1 in two
refineries. The category of pumps has
the highest rate of leaks at 17.2%,
with all other categories at 5.5% or
less. Maintenance could reduce emis-
sions 96% to 97%. Leak frequency
varied from unit to unit and from time
to time. Leak recurrence rates after 3
months ranged from zero to 19%; the
average was 8% of the sources which
had been repaired. Readings taken at
distances other than zero could be
correlated. As a seal aged, it tended to
leak a bit more. The mean time
between seal replacements was 1.0 to
1.5 years; approximately 90% of the
seals are replaced within 3 years. Both
the quantity involved and the disposi-
tion of process samples indicate
minimal contribution to fugitive emis-
sions. Less than 20% of the observed
variation was attributable to the
instrument or screening process. A
mean screening value of X can be
expected to be between X/2.5 and
2.5X for 95% of the time.
This Project Summary was devel-
oped by EPA's Industrial Environ-
mental Research Laboratory, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Seven areas of interest were investi-
gated in a screening study coupled with
an inspection of maintenance records
and a historical review of the effects of
previous inspections and maintenance
performed under the South Coast Air
Quality Management District (SCAQMD)
Rule 466.1 in two. refineries.
The first area, screening of sources,
provided data that indicate that, among
sources in the vapor pressure range of
Rule 466.1, the category of pumps has
the highest rate of leaks at 17.2%, with
all other categories at 5.5% or less.
The second area investigated, the
immediate effects of maintenance,
produced data to show that a weight
percent reduction in emissions of 97%
for open-ended lines and 96% reduction
in valves could be attributed to the
maintenance performed.
The third area, historical leakdetection
and repair data, provided data to
indicate that leak frequency varied from
unit to unit and from time to time, but
that leak recurrence rates after 3
months ranged from zero to 19% with
the average being 8% of the sources
which had been repaired. The different
-------
population of sources tested at each
inspection makes it difficult to draw
conclusions on the effectiveness of Rule
466.1.
The fourth area of concern, how
screening at a distance from the
emission source would change the
readings and how readings taken at
distances other than zero could be
correlated, produced the equation for
pump seals:
screening value = e772H}364""
where d = distance from the pump
seal in centimeters.
A nomograph with confidence intervals,
is included.
The fifth area, pump seal replacement,
provided data which showed a slight
positive relationship between the
screening value and time since last seal
replacement; i.e., as a seal ages, it leaks
a bit more. Both operating and spare
pumps were included in the data. The
mean time between seal replacements
was 1.0 to 1.5 years; approximately 90%
of the seals are replaced within 3 years.
The sixth area, process control
sampling procedures used by the
refinery, yielded information on the
frequency, size, and disposition of liquid
samples taken to monitor the refining
process. The three to six 1 -gal. samples
taken each day are routed either to sour
water treatment or to the API separator
after analysis. Both the quantity involved
and the disposition indicate minimal
contribution to fugitive emissions from
process samples.
The seventh area, quality control,
revealed that there was no bias between
screening teams used in the project.
There was very good agreement between
test teams on whether or not a source
was screening greater than or equal to
10,000 ppmv. Repeated screenings of
the valves indicated that less than 205%
of the observed variation in the screening
values for a given valve was attributable
to variations from the instrument or
screening process. Data indicate that a
mean screening value of X can be
expected to be between X/2.5 and 2.5X
for 95% of the time.
All sources in eight process units with
greater than 10% VOC were scheduled
for screening with a Century Systems
OVA. Screening results are presented
for sources exempt from and subject to
the vapor pressure criterion of SCAQMD
Rule 466.1. Approximately 15% of the
sources could not be screened; location
was the main reason for not screening.
Leaking sources (sources screening
greater than or equal to 10,000 ppmv)
were found at the following rates:
valves, 5.4%; pumps, 17.2%; open-
ended lines, 4.8%; and other sources,
5.5%. The number of leaking sources in
each process unit is tabulated in the
report. For valves in gas service, the leak
frequency inthe eight units ranged from
zero to 5.7%. Control valves had a leak
frequency (16.3%) greater than three
times that of block valves (4.9%). On-
line pumps had a leak frequency (22.6%)
of greater than five times that of off-line
pumps (4.1%)
At the Chevron refinery, all valves and
open-ended lines subject to the rule
with screening values greater than or
equal to 10,000 ppmv were repaired or
taken out of service. Post-maintenance
screening values were recorded for 342
repaired valves and open-ended lines.
Leak rates were estimated from the
screening values and a weight percent
reduction in emissions estimated. The
weight percent reduction was 97% for
open-ended lines and 96% for valves.
Refinery records from previous
screening inspections were examined
to evaluate the impact of inspections
done to comply with Rule 466.1.
Different populations of sources were
tested during each inspection, making it
difficult to quantify the impact. The
percent of sources screening greater
than or equal to 10,000 ppmv at each
inspection represents combined leak
occurrence and recurrence rates. Be-
cause of the different populations and
varying time periods between inspec-
tions, it is difficult to quantify these leak
occurrence and recurrence rates. After
the initial inspection, the leak frequency
ranged from zero (vacuum distillation
unit) to a high of 11.7% (third inspection
of FCC unit). Leak recurrence rates after
3 months ranged from zero to 19%,
averaging about 8% of the sources
repaired.
Pump and compressor seals with a
screening value at the source of greater
than 1,000 ppmv were also screened at
distances ranging from 1 to 20 cm from
the leak source. A functional relationship
between the screening value and the
distance from the source was developed
from the pump data. Significant differ-
ences between sources were observed
for the rate of screening value decrease
with distance. An overall model for the
32 pump seals studied was:
Screening value = e77!H)364 (cm "°m
source)
A nomograph for this model is included
in the report with confidence intervals.
For example, using the nomograph, if
the screening value obtained 2.5 cm
from the source is 1,000 ppmv, the
predicted screening value at the source
is 2,500 ppmv.
Refinery records were examined to
evaluate pump seal replacement prac-
tices. For the pump seals in the vapor
pressure range of Rule 466.1, a slight
positive relationship was seen between
the screening value and the time since
last seal replacement. The total available
historical data indicated that 544 seals
had been replaced in pumps. This
includes both operating and spare
pumps. The mean time between seal
replacement was computed for each
pump seal. The mean time averaged
about 1.0 to 1.5 years. Cumulative
distributions show that about 90% of
the seals are replaced within 3 years.
Information from the Powerine refin-
ery indicates that 1 -gal. process samples
are obtained three to six times a day.
The sample material is either routed to
sourwater treatment facilities or to an
API separator. The sampled material
could range from 8 to 47 Ib/day
depending on the density of the material
sampled and the number of samples. No
information was available on sample
purging.
Quality control procedures were part
of all field testing during the program.
Method 21 performance checks were
performed on each screening instru-
ment. The calibration of each instrument
was checked daily. Slight negative drifts
in calibration were observed for one
instrument. No consistent drifts in the
dilution factor were observed. Three
valves at each refinery were screened
repeatedly. Less than 20% of the total
variation in the screening values for a
given valve was attributable to variations
from the instrument/screening process.
Reported screening values for a source
with a mean screening valueof Xcan be
expected to be between X/2.5 and 2.5X
for 95% of the time. Screening quality
assurance audits indicated no biases
between the different screening teams
used in this project. Of the 38 sources
determined to be screening greater than
10,000 ppmv by either the original or
audit screens, 36 also had screening
values greater than 10,000 ppmv at the
other screening.
-------
R. L Honerkamp, M. L. Schwendeman, and L. P. Provost are with Radian Corp.,
Austin, TX 78759.
Samuel L. Rakes is the EPA Project Officer (see below).
The complete report, entitled "Evaluation of the Maintenance Effect on Fugitive
Emissions from Refineries in the South Coast Air Quality Management
District," (Order No. PB 82-239 260; Cost: $19.50, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
* UJ. GOVERNMENT miNTmaOFFICE: 18B-559-017/0808
-------
, ,- Postage and
United States Center for Environmental Research pees pai(j
Environmental Protection Information Environmental
Agency Cincinnati OH 45268 Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
"1"1 *GENCY
------- |