-------
MEASURED VALUES
(1) Total Particulate Weight Collected, m , mg.
CONTAINER
NUMBER
1
2
BLANK*
CORRECTION
TOTAL (ran)
WEIGHT OF PARTICULATE COLLECTED,
mg
FINAL WEIGHT
TARE WEIGHT
WEIGHT GAIN
+ mg
+ mg
mg
mg
Blank Correction = Residue (Blank) mg x (Vp/VR)
where V2 = Volume of acetone wash in container number 2,
im, and
VD = Volume of acetone blank, usually 100 m£.
D
(2) Volume of gas sample through the dry gas meter at standard conditions
to three significant digits.
Vm(std) = °'3855 Vm
AH
3.6
m
(3) Moisture content of stack gas to two significant digits.
(0.00134 m3/mA)V1
ws
Vm(std) +
dimensionless.
Figure 4. Sample data form for particulate anissions determinations
(continued).
44
-------
FINAL
INITIAL
LIQUID COLLECTED
TOTAL VOLUME COLLECTED (Vlr)
V*
VOLUME OF LIQUID
WATER COLLECTED
IMPINGER
VOLUME,
ml
SILICA GEL
WEIGHT,
mg
*
m&
Mass of water in mg = volume of water in ma.
(4) Particulate concentration at standard conditions on a dry basis,
.3
cs = (o.oi54
(5) Stack gas molecular weight on a wet basis
Ms =
- Bws) + 18 Bws
g/g-mole
where Md = 0.44(%C02) + 0.32(%62) + 0.28(100 - %C02 - %6"2) .
(6) Average stack gas velocity
1/2
avg
•p (/4P)avg ' S
m/sec
(7) Percent of isokinetic sampling
4.323 Vm(std) Ts
percent.
Figure 4. Sample data form for particulate emissions determinations
(continued).
45
-------
(8) Volumetric flow rate, dry basis, standard conditions
Qs = 1388(1 - Bws) (Vs)ayg A(PS/TS ) - m3/hr-
(9) Parti cul ate mass emission rate calculated on a sample concentration
basis
PMR = Cs x Qs = _ g/hr.
SUMMARY OF RUNS
(10) Average participate mass emission rate for three sample runs
PMR, + PMR9 + PMR,
- ] -- _J - 3 = - 9/hr.
(ll) PMR" with 90 percent confidence limits (for 3 runs) where FWR° is the
grand mean that would be obtained from a large number of tests
PMR - 2.92 sfPMR}//! < PMR < PMR + 2.92 s{PMR}//3
(PMR, - PMR)^ + (PMR9 - PMRY + (PMR. - PMR)^ 1/2
where s{PMR} ' ^ J
Figure 4. Sample data form for particulate emissions determinations
(continued).
46
-------
2.4.2.4 Molecular Weight of Stack Gas. Determine the dry molecular weight
of the gas stream by Method 3 (ref. 2). If the particulate concentration
is to be adjusted to 12% C09, it is recommended that the sample be of the
integrated type for two reasons: 1) the possibility of a more representa-
tive sample and, 2) the convenience of taking the sample at the stack and
being able to transport the sampling bag to a more suitable area for Orsat
analysis.
2.4.2.5 Stack Temperature and Velocity Heads. Set up and level the dual
inclined manometer and determine the minimum and maximum velocity head
(AP) and the stack temperature (T ). This is done most efficiently with
s
a type-S pitot tube, with a temperature-sensing device attached as shown
in Figure 5-1 of Appendix A. The AP's are determined with an inclined
manometer by drawing the pitot tube across the stack diameter in two
directions (circular stack with 90° traverses). This must be done in
order to pick the correct nozzle size and to set the nomograph. Incor-
rect selection of nozzle size and/or setting of the nomograph may result
in not being able to maintain the isokinetic sampling rate, thereby void-
ing the sample. Determine the static pressure as directed in the Quality
Assurance Document of this series for Method 2 (ref. 1).
2.4.3 Sampling
The on-site sampling includes making a final selection of proper
nozzle size, setting the nomograph (if used), loading the filter into the
filter holder, preparing and assembling the sampling train, making an ini-
tial leak-check, inserting the probe into the stack, sealing the port,
sampling isokinetically while traversing, recording the data, and making
a final leak-check of the sampling system. Sampling is the foundation of
source testing. Critical problems in testing result from poor or incor-
rect sampling more frequently than from any other part of the measurement
process. The analytical process (laboratory) can never correct for errors
made in the field resulting from poor judgment or instrumental failure.
If the initial site survey, apparatus check and calibration, and prelim-
inary measurement and setup on-site have been implemented properly, the
47
-------
testing should go smoothly with a minimal amount of effort and crises.
2.4.3.1 Preliminary Setting of the Nomograph. The setup of the nomograph
using the parameters obtained in subsection 2.4.2 is given in detail in
Appendix B of this document. A procedure is included in the appendix for
checking the nomograph for correct design (accuracy).
Note
_3: If the coefficient, C , of the type-S pitot tube being used is
ide the range of 0.85 +; 0.02, compute the ratio (C /0.85)2 and mul-
tiply this constant times the correction factor, C, obtained from the
nomograph. Use this new "C" factor in setting the nomograph for iso-
kinetic sampling (see Appendix B for further discussion).
2.4.3.2 Selection of Nozzle Size. After the nozzle size and appropriate
probe length have been selected, insert the nozzle in the probe sheath
union and tighten the union. Do not use wrenches; finger-tight is suf-
ficient in most cases. Using uncontrolled pressure in tightening the
union will result in a broken or cracked inner-liner. Keep the ball
joint and nozzle tip protected from dust and dirt with a serum cap or
equivalent.
2.4.3.3 Assembling of Sampling Train. Assemble the glass impinger train
as follows:
1. Measure 200 m£ of distilled water in the graduated cylinder
5-m£ divisions) and place approximately 100 m£ of the water in each of
the first two impingers
2. The third impinger is left empty.
3. Place approximately 200 g of preweighed indicating silica gel
into the fourth impinger.
The first, third, and fourth impingers are modified Greenburg-Smith
while the second impinger is of the standard Greenburg-Smith design.
Place the impingers into the sample box and assemble the sampling train
using the appropriate U-joint.s. A very light coat of silicone grease
(acetone insoluble) should be; applied carefully to the joints, to avoid
leaks in the system. Depending upon the design of the impinger, apply
the lubricant in a manner that precludes its contact with the sample.
The loading of the impingers into the sample box can be done in the
48
-------
laboratory by sealing the inlet to the first impinger, the outlet of the
third impinger, and inlet and outlet of the silica gel impinger. This
is practical only when the sampling site is near and the logistics are
suitable.
In view of the fact that the reference method does not consider the
analysis of the condensible fraction, the impingers are not needed as
such. A condenser can be employed consisting of a coiled tube of stain-
less steel or copper and a reservoir that has a capacity of at least
1500 mH, A good design would incorporate a drain valve in the bottom of
the reservoir to allow the tester to drain and measure the total volume
of condensate for moisture determinations. In all cases the impingers
or condenser are maintained in an ice bath during sampling to remove the
condensibles and keep the exit gas at or below 20°C (68°F). When a con-
denser is utilized, a drying tube downstream is used to protect the dry
gas meter and vacuum pump.
2.4.3.4 Load Filter. Load the filter as follows:
1. Preweighed filter (desiccated for 24-hours and weighed to the
nearest 0.1 mg) is removed with teflon covered tweezers from its sealed
container and placed in the filter holder. The filter should have an
identifying number and the filter holder should be numbered with a semi-
permanent marker to preserve the integrity of the sample. Make certain
that the filter is centered correctly in the holder with the sample side
toward the probe. The filter holder should be tightened until the two
halves are secure. Over-tightening the two halves can break the filter
holder or tear the filter.
2. Place the filter holder into the sample box and connect the exit
of the filter holder to the inlet of the first impinger. Plug the inlet
of the filter holder with a glass ball to check for leaks. Connect the
meter box (vacuum port) to the sample box with the appropriate umbilical
cord.
2.4.3.5 Perform Leak-Check. Leak-check the sampling train by plugging
the inlet to the filter holder, turning on the vacuum pump, and opening
49
-------
the valve system until the vacuum in the system reaches 380 mm Hg (15 in. Hg)
3 3
A leakage rate not in excess of 0.0006 m /min (0.02 ft /min) at 380 mm Hg
vacuum is acceptable. Release the pressure in the system but do not turn
off the pump until the following sequence has been completed:
1. Slowly release the pressure in the system by carefully opening
(twisting) the glass ball in the inlet of the filter holder.
2. Shut the coarse valve (main vacuum valve).
3. When the vacuum gage reads zero vacuum, remove the glass ball
and shut down the pump.
2.4.3.6 Installation of Probe. Mount the probe in the sampling box.
Check the configuration of the stack gas temperature measuring system
as shown in Figure 5-1 in Appendix A. Connect the probe to the inlet of
the filter holder and leak-check in the following manner:
1. Seal the inlet of the probe nozzle with a serum cap.
2. Turn on vacuum pump.
3. Open the valve system and adjust the vacuum to 380 mm Hg (15 in.
Hg). A lower vacuum may be used provided that the vacuum used here is
not exceeded during the test.
4. Check the leakage rate on the dry gas meter. A leakage rate less
than 0.0006 m3/min (0.02 ft:>/min) at 380 mm Hg (15 in. Hg) vacuum is
acceptable.
Note 4: If an asbestos string is used in the fabrication of the probe
nozzle to the probe liner connection, leak-check at 25 mm Hg (1 in. Hg)
vacuum only. If a leak-free connection in the nozzle is employed, the
total train, filter and probe, can be initially checked at 380 mm Hg
(15 in. Hg) vacuum.
5. After completion of the leak check, release the pressure as fol-
lows :
a) Slowly release the vacuum by carefully opening (squeezing)
the serum cap until the system pressure is back to ambient (monitor with
built-in vacuum gage).
b) Turn valve system off (coarse valve).
c) Turn off the vacuum pump.
50
-------
Operations in subsections 2.4.3.5 and 2.4.3.6 can be combined, there-
by requiring only one initial leak-check. Record leakage rate on the form
in Figure 4.
2.4.3.7 Taking of Sample. Turn on the sample box and fill the impinger
train container with crushed ice. The meter box operator should now
recheck the setting of the nomograph while the sample box operator checks
the filter box temperature gage to confirm that it is coming up to operat-
ing temperature; likewise, he can touch the probe to see if it is heating.
It is recommended that a thermocouple be mounted next to the glass liner
so that the probe temperature can be monitored.
As soon as the filter box temperature and probe temperature are up
to the desired level, the test itself can be performed:
Remove the plug or cap from the sampling port and remove the dust
(particulates) on the port walls by utilizing a wire brush or its equiva-
lent. Remove the serum cap from the nozzle tip. Record the initial
volume of the test meter on the data log sheet of Figure 4.
2. If the sample gas is hot, start at the traverse point farthest
from the port and draw the probe out as the test continues. Asbestos
gloves should be used in handling hot sampling probes and pitot tubes.
3. Attach a proper electrical ground to the probe and sampling
system.
4. Insert the probe to the farthest traverse point with the nozzle
pointing directly into the gas stream. Seal the port and immediately
start the pump. Adjust the coarse and fine control valves until isokinetic
conditions are obtained. Note the time and record it on the data log
sheet of Figure 4.
5. Maintain isokinetic conditions during the entire sampling period.
Sample for an equal amount of time at each traverse point. The time
period at each traverse point should be long enough to set the sample
rate and record the required data. The time period at each traverse
point must be long enough to obtain a total sampling period representa-
tive of the process being monitored. The time at each traverse point must
3 3
be sufficient to obtain a total sample volume of at least 1.7 m (60 ft )
51
-------
at standard conditions. While sampling, reset the nomograph if:
a) The temperature in the stack changes more than + 14°C (25°F) .
b) T (average temperature of meter) varies more than + 6°C
Adjust the sampling rate for every point and maintain the isokinetic
rate by continuous observation. Record the meter volume when sampling
has been completed at each individual traverse point. Take readings at
each sampling point, at least every 5 minutes (or during sampling period
at each traverse point) : £.11 the readings and adjustments should not be
attempted for time intervals of less than 2 minutes. When significant
changes in stack conditions are observed, compensating adjustments in
flow rate should be made to maintain isokinetic conditions. Record on
the data log sheet of Figure 4 the traverse point number, stack tempera-
ture (T ), velocity pressure head (AP, mm H00 or in. H90) , (orifice pres-
S jL L-
sure differential (AH, mm Kg or in. H«0) , gas temperature at dry gas meter
(Tm. ) and Tm or T , °C or °F) , sample box temperature, condenser
min mout avg
temperature, and the probe temperature if the probe has a thermocouple
and appropriate readout.
6. When sampling at one traverse point has been completed, move
the sampler to the next point as quickly as possible. Close the control
valve only when transferring the sampler from one sample port to the
other. Exclude the time required to transfer the sampler from one port
to another from the total sampling time.
Note 5 : Since moving from port to port is time consuming, it is recommended
that longer probes be employed to allow only one move during a test if
a circular stack is involved. Probes up to 3m (10 ft) long can be managed
without too much difficulty, provided that adequate space is available
on the sampling platform.
Upon transfer of the sampler to another port, the following proce-
dures should be followed:
a) Monitor the vacuum through the system. An increase of vacuum
is an indication of particulate buildup on the filter. Loss in vacuum is
an indication of a broken impinger, connector, filter, or a loose con-
nection.
52
-------
b) Keep the impingers iced down (i.e., monitor the condenser
temperature) to hold the temperature of the exit gas below 20°C (68°F).
Add salt to the ice bath if necessary.
c) Check the line voltage with a voltmeter if a digital temper-
ature system is utilized.
Note 6: Digital temperature systems may read erroneously with a drop in
line voltage and/or interference from electromagnetic fields.
d) Make sure that the dual inclined manometer is level and that
the pitot tube and pitot tube lines are unobstructed. A signal of trouble
would be AP's that are not representative of the velocity heads obtained
in the velocity traverse made during the preliminary site visit.
e) All data should be recorded on a data log sheet as depicted
in Figure 4.
7. At the completion of the test, close the coarse control valve on
the meter, remove the probe from the stack, and turn off the pump. Remove
the probe carefully from the stack, making certain that the nozzle does
not scrape dust from the inside of the port. Keep the nozzle elevated
to prevent sample loss. After the probe cools place a serum cap or equi-
valent over the nozzle tip and leak-check the system at 50 mm Hg (2 in. Hg)
vacuum above the operating vacuum during the test. The vacuum during this
post sampling leak check should be no greater than 380 mm Hg (15 in. Hg).
(Do not boil the water in the impingers.) Follow the same leak-check pro-
cedures as outlined in subsection 2.4.3.5. Seal the end of the nozzle.
Disconnect the pitot tube lines and umbilical. Protect the pitot tube
and the umbilical connections with tape or an appropriate equivalent.
3 3
Record on the data log sheet of Figure 4 the leakage rate in m /min (ft /min)
and the vacuum at which the leak check was performed. Check all connectors
such as umbilical connection, pitot tube lines, glass connections, etc.
for evidence of malfunction. Record all abnormalities on the data log
sheet. The logging of abnormalities will not necessarily void the sam-
ple, but it may help to improve the quality of sampling performance.
53
-------
2.4.4 Sample Recovery
Move the sampling train and probe to the sample recovery area. Care
should be taken to prevent loss or contamination of the sample. If the
probe must be removed before movement to the recovery area, the probe
should be sealed at both ends (serum caps) and the inlet of the filter
plugged with a glass ball. The sample recovery area should be a well-
lighted, relatively clean room with enough table-top work space, about
2 2
4.6 m (50 ft ), for two crew members to change the filter and wash out
the sampling train.
2.4.4.1 Container No. 1. Wipe the exterior of the filter holder sur-
face to remove any excess dust or extraneous material. Remove the fil-
ter from the holder, place the filter in its original container and seal
it. It is recommended that a piece of paper (inert, smooth surface) be
placed under the filter holder as the filter is being removed to prevent
loss of particulates. Removal of the filter is more efficient utilizing
a set of tweezers. Teflon-tipped tweezers and a teflon scalpel should
be used to handle filters. If a filter is torn, all pieces must be saved,
conditioned, and weighed. Record date, time of test, location of test,
and the number of run on this container. This data should also be recorded
on the data log sheet of Figure 4.
2.4.4.2 Container No. 2. Wash all internal surfaces of the sampling train
from the nozzle tip up to the backside of the filter holder with atomic
absorption grade acetone. Determine the volume to the nearest ml and
transfer to the container. A brush with a handle as long or longer than
the probe should be used to loosen the particulate matter. It is recom-
mended that the probe be washed by attaching a calibrated (125 m& or larger)
cyclone flask to the end of the probe and washing the probe contents into
the container. This wash may require emptying the flask into container No.
2 several times (the flask should be filled to the mark or to the nearest
major division with acetone each time so that the quantity of wash trans-
ferred is accurately known). Also, a graduated cylinder should be used to
accurately measure volumes of acetone (10 to 25 m£) to be used for rinsing
54
-------
the residue from the flask.
Accurately measure in a graduated cylinder an appropriate volume (e.g.,
200 ml) of acetone to serve as a blank.
The measured volume of acetone wash is quite critical if the acetone
is not residue free. The total volume of wash obtained by summing the
volumes transferred to container No. 2 is recorded to the nearest m£.
Record total volume of acetone wash, date, time of test, location and
run number on the container and on the data sheet of figure 4.
2.4.4.3 Container No. 3. Transfer the silica gel from the fourth itnpinger
into its original preweighed container. Label container with date, time
of test, location, and any other pertinent data. This information should
also be recorded on the data log sheet. All sample containers should be
glass with caps lined with teflon.
2.4.5 Total Volume
Measure the total volume of condensate by transferring the contents
of the first three impingers into a graduated cylinder with divisions of
2 mH or less. Record this volume of condensate on the data log sheet of
figure 4 to the nearest m£.
2.4.6 Sample Logistics (Data) and Packing of Equipment
The above procedures are followed until the required number of tests
are completed. The following is recommended at the completion of testing.
1. Check all sample containers which must be properly labelled.
(Time and date of test, location of testing, number of test, and any other
pertinent documentation.) This aspect should be performed at the end of
each individual test or prior to such test if the impingers are to be
utilized in further tests before returning to the laboratory.
2. All data recorded during field testing should be recorded in
duplicate (carbon paper). One set of data should be mailed to the base
laboratory and the other to be hand-carried. This is a recommendation that
can prevent a very costly mistake.
3. All sample containers should be properly packed in a sample box
55
-------
for shipment to the base laboratory. All boxes should be properly labelled
to prevent loss of the samples.
4. The sampling equipment should be inspected as it is disassembled
for packing. Any signs of damage that could have had an influence on the
precision/accuracy of the measurement should be documented on the data
sheet and that item of equipment checked in the laboratory (if possible) to
determine the magnitude of error that may have resulted from the damaged
equipment.
2.4.7 Data Validation
Following the directions given in subsection 2.5.3, calculate and/or
determine the following:
1. Moisture content of the stack gas, B
ws
2. Stack gas molecular weight on a wet basis, M .
o
3. The average stack gas velocity, (V )
4. The percent of isokinetic sampling for the sample run, I.
Compare these measured values to theoretical values derived from
combustion nomographs (ref. 12) or to values obtained by other measurement
methods, e.g., measuring B by the wet-bulb/dry-bulb method.
ws
Any large unexplainable differences in measured and theoretical values
should be noted and special care taken to reduce the variability of that
specific parameter for the next run. If the percent of isokinetic sampling
is outside the range of 0.90 to 1.10, the run should be repeated unless it
is known (must have been approved by the administrator) that the particle
size distribution is below, about 5 urn.
Values of B and M as measured by the first run should be used in
ws s •*
setting isokinetic conditions for subsequent runs unless there is reason
to doubt their validity as compared to the values derived from preliminary
measurements or estimates.
2.5 POSTSAMPLING OPERATIONS (Base Laboratory)
2.5.1 Apparatus Check
A postsampling check of the equipment can serve to validate the data
56
-------
from the just completed field test. The least it can do is aid the field
team in making an honest estimate of the accuracy of the field measurements.
Any malfunctions uncovered during the postsampling check should be
reported immediately and in detail to the supervisor. It should also be
documented in the laboratory log and/or calibration log as applicable.
The decision of whether to correct the data, repeat the test, or just
report the error with the data is one that the supervisor must make after
considering 1) the magnitude of the error involved, 2) the precision/
accuracy of the measurement process, and 3) the ultimate use of the field
data.
2.5.1.1 Type-S Pitot Tube. The type-S pitot tube is checked according to
the Quality Assurance Document of this series for Method 2 (ref. 1).
2.5.1.2 Dry Gas Meter and Orifice Meter (Sampling Train). A postcheck (a
postcheck for one test can in some instances serve as the presampling check
for the next test) should be made of the sampling train to check for proper
operation of the pump, dry gas meter, vacuum gage, and dry gas meter
thermometers. Leak-check the vacuum system. Determine y and AH@ at three
points in the operating range. This is a check on the system for future
testing and gives confidence in the data from the previous field test.
This is a recommended procedure to improve the data quality and to prevent
field sampling under assumed conditions.
2.5.2 Analysis (Laboratory)
The requirements for a precise and accurate analysis are minimal in
the reference method. The analytical balance should be checked with a set
of calibrated weights before the weighing of the first filter or at any
time a problem is indicated. Record the actual and measured weights in the
laboratory logbook along with the date and initials. If the actual and
measured values agree to within + 0.3'mg proceed; otherwise, report it to
the supervisor before proceeding. Glass wash bottles should be utilized
in transfer and the laboratory area should be free of any grinding or
dust-producing activities. Blanks are required in the reference method
and are always required in any analytical method striving for data validation.
57
-------
2.5.2.1 Container No. 1. Transfer the filter and any loose particulate
matter from the sample container to a tared glass weighing dish, desiccate
and dry to a constant weigat. Weigh and report results to the nearest
0.1 mg. After opening and closing, the humidity of the air in a desic-
cator may be very different from that of the atmosphere. Since the rate
of removal of moisture by the desiccant may be quite slow, the initial
period of desiccation should be at least 24 hours. After the initial
period of desiccation, eacn succeeding period of desiccation should be
at least 6 hours. Granular or fused anhydrous calcium sulfate is most
frequently used as the drying agent. It has a good capacity, but is not
a powerful desiccant, although sufficiently satisfactory in this respect
to be usable in this work.
It is recommended that the sample be considered to be at constant
weight when two successive readings separated by 6 hours vary by <^ 0.5 mg.
An analytical balance which weighs to 0.1 mg should be utilized.
2.5.2.2. Container No. 2. Transfer the (AA) acetone washings to a tared
beaker and evaporate to dryness at ambient (laboratory) temperature and
pressure. The transfer process should include washings (10 m£ volumes
premeasured in a graduated cylinder with 1 m£ divisions) with (AA) acetone
until a complete transferral (no signs of residual matter) is obtained.
The total volume of these washings must be added to the wash volume obtained
in the field (subsection 2.4.4.2). An accurately measured blank (100 m&
measured in a graduated cylinder) of the acetone taken in the field is
run concurrently with the samples. This blank will account for any
residue in the acetone or any other laboratory condition that would
tend to affect the final weight of the samples. After evaporation, the
samples and blanks (one blank for each field site) should be desiccated
until a constant weight is obtained. The same procedure as outlined in
section 2.3.2.1 should be adhered to. Report results to the nearest 0.1 mg.
2.5.2.3 Container No. 3. Weigh the spent silica gel to the nearest 0.5 g.
A top loading (trip) balance is sufficient for this weighing. It is recom-
mended that the silica gel be weighed in Container No. 3 before the field
test and returned to Container No. 3 following the test. This negates
58
-------
the error involved in not getting complete transferral of the silica gel
from the original container.
All data (weights, volume of condensate collected in the first three
impingers) are recorded on permanent data sheets such as depicted in
Figure 4.
2.5.3 Calculations
Calculation error due to procedure or mathematical mistakes can be a
large component of total system error. Therefore, it is recommended that
each set of calculations be repeated, starting with the raw field data,
preferably by a team member other than the one that performed the original
calculations. If a difference greater than the typical round-off error is
observed, the calculations should be checked step by step until the source
of error is found and corrected. If a computer program is used, the
original data entry should be checked and if differences are observed, a
new computer run made. A standardized computer program should be written
to treat all raw field data. A computer program presently being used in
EPA is included in the Final Report of this contract (ref. 4).
2.5.3.1 Weight of Particulate Collected. Particulate weight is the sum
of the weight gains of container 1 (filter and loose particulate) and con-
tainer 2 (acetone probe wash) minus a correction for the blank as shown
in the table under item (1) of Figure 4. Record m in the table of Figure
4 to the nearest 0.5 mg.
2.5.3.2 Dry Gas Volume. The sample volume measured by the dry gas meter
is corrected to standard conditions by
/ \ /-n AH \
v - .„,,/ °K v bar" 13.6
- '3855 V
-
std) - ' mmHg m T /
\ / \ m /
where ^ , ,.. = Volume of gas sample through the dry gas meter
at standard conditions, nH
V = Volume of gas sample through the dry gas meter
3
at meter conditions, m
59
-------
= Barometric pressure at the orifice meter,
mm H20.
AH = Average pressure drop across the orifice meter
obtained from the table of recorded measurements
of figure 4, mm H20.
T = Average dry gas meter temperature in °C plus 273 obtained
from the table of recorded measurements of figure 4, °K.
3 3
Record V , ,N to the nearest 0.003 m (0.1 ft ) as item (2) under
m(std;
measured values on the data sheet in Figure 4.
2.5.3.3 Moisture Content, B . Using the value of V1 from the table under
~"T™" \v S -*- r*
item (3) of Figure 4, calculate the moisture content by
(0.0013 m3/m2,) Vi
B £c .
ws
V , ,.+ (0.0013 nr/m£) Vi
m(std) x
Record B to the nearest 0.001 on the data sheet under item 3 of
ws
Figure 4.
2.5.3.4 Particulate Concentration at Standard Conditions on a Dry Basis, C_.
-- - g
Calculate the particulate concentration as follows:
where C = Particulate concentration in grams per standard cubic meter
s
on a dry basis.
Record C to four significant digits on the data sheet under item 4
o
of Figure 4.
2.5.3.5 Stack Gas Molecular Weight on a Wet Basis, M. Calculate the stack
g.
gas molecular weight by
M = MH C1 - Bws> + 18
!3
60
-------
where M, is given by
Md = 0.44(%C02) + 0.32(%02) + 0.28(100 -
and %C02 and %6 are the averages of percent C02 and 02 determinations,
repsectively, according to the Quality Assurance Document of this series
for Method 3 and the EPA revised method as contained in the Final Report
of this contract (ref. 4).
Record M to three significant digits, i.e., —.-, on the data sheet
s
under item 5 of Figure 4.
2.5.3.6 Average Stack Gas Velocity. Calculate the average stack gas
velocity, (V ) , in m/sec by
s avg
1/2
(V ) = 34.97 C (/AT)
s avg p avg
r (T r
s avg
P M
L s s
where (V ) = Average stack gas velocity, m/sec.
s avg
C = Average pitot tube calibration coefficient over
the velocity range being measured, dimensionless.
(vAP) = Average of the square roots of the velocity
pressure heads, (mm
(T ) = Average absolute stack gas temperature, °K.
P = Absolute stack gas pressure, mm Kg.
S
MS = Molecular weight of stack gas on a wet basis,
g/g-mole.
Record (V ) to three significant digits on the data sheet under
avg
item 6 of Figure 4.
*If Ts varies more than about 10 percent of the "mean from point to
point in the stack, the correct term to use is (/TT) rather than
ATS). avg
avg
61
-------
2.5.3.7 Percent of Isokinetic Sampling, I. Calculate the percent of
isokinetic sampling by
4.323 V , „,. T
_ m(std) s
9(Vs)avg Ps "a- -ws'
where I = Percent of isokinetic sampling.
Bws = Percent moisture by volume in the stack gas, dimension-
less.
V ., . = Volume of gas sample through the dry gas meter
at standard conditions, m .
0 = Total sampling time, min.
(V ) = Average stack gas velocity calculated in section 2.5.3.6
above, m/sec.
P = Absolute stack gas pressure, mm Hg.
S 2
A = Cross-sectional area of the nozzle, m .
Record I to two significant digits on the data sheet under item 7 of
Figure 4.
2.5.3.8 Volumetric Flow Rate at Standard Conditions on a Dry Basis, Q .
Calculate the volumetric flow rate by
Q = 1383(1 - B ) (V ) A (P /T )
xs ws s avg s s avg
2
where A = Cross-sectional area of the stack, m , and the other terms
are as defined in the above calculation of stack gas velocity, subsection
2.5.3.6. A standard absolute temperature of 293°K and a standard absolute
pressure of 760 mm Hg were used in the calculation of the constant.
Record Q to three significant digits on the data sheet under item 8
S
of Figure 4.
2.5.3.9 Particulate Mass Emission Rate Calculated on a Sample Concentration
Basis, PMR. Calculate the particulate matter emission rate by
PMR = C x Q
62
-------
where PMR = Particulate mass emission rate, g/hr.
C = Particulate concentration at standard conditions
S 3
on a dry basis, g/m .
Q = Volumetric flow rate at standard conditions on a dry
basis, m /hr.
Record PMR to three significant digits on the data sheet under item
9. of Figure 4.
2.5.3.10 Calculation Check. Calculation error due to procedure or mathe-
matical mistakes can be a large component of total system error. Therefore,
it is recommended that each set of calculations be repeated, starting with
the raw field data, preferably by a team member other than the one that
performed the original calculations. If a difference greater than the
typical round-off error is observed, the calculations should be checked
step by step until the source of error is found and corrected. If a com-
puter program is used, the original data entry should be checked and if
differences are observed, a new computer run made. A standardized com-
puter program should be written to treat all raw field data.
63
-------
SECTION III
MANUAL FOR FIELD TEAM SUPERVISOR
65
-------
SECTION III WNUAL FOR FIELD TEAM SUPERVISOR
3.0 GENERAL
The term "supervisor" as used in this document applies to the
individual in charge of a field team. He is directly responsible for the
validity and the quality of the field data collected by his team. He may
be a member of an organization which performs source sampling under
contract to government or industry, a government agency performing source
sampling, or an industry performing its own source sampling activities.
It is the responsibility of the supervisor to identify sources of
uncertainty or error in the measurement process for specified situations
and, if possible, eliminate or minimize them by applying appropriate quality
control procedures to assure that the data collected are of acceptable
quality. Specific actions and operations required of the supervisor for a
viable quality assurance program are summarized in the following listing.
1. Monitor/Control Data Quality
a. Direct the field team in performing field tests according to
the procedures given in the Operations Manual.
b. Perform or qualify results of the quality control checks
(i.e., assure that checks are valid).
c. Perform necessary calculations and compare quality control
checks to suggested performance criteria.
d. Make corrections or alter operations when suggested per-
formance criteria are exceeded.
e. Forward qualified data for additional internal review or to
user.
2. Evaluate Routine Operation
a. Obtain from team members immediate reports of suspicious
data or malfunctions. Initiate corrective action or, if necessary, specify
special checks to determine the trouble; then take corrective action.
Document the corrective action taken.
b. Examine the team's log books periodically for completeness
and adherence to operating procedures.
c. Approve data sheets, calibration checks, etc., for filing.
66
-------
3. Evaluate Overall System
a. Evaluate available alternative(s) for accomplishing a given
objective in light of experience and needs,
b. Evaluate personnel training/instructional needs for specific
operations.
Consistent with the realization of the objectives of a quality
assurance program as given in section I, this section provides the super-
visor with brief guidelines and directions for:
1. Collection of information necessary for assessing data quality
on an intrateam basis;
2. The use of performance criteria to insure the collection of data
of acceptable precision/accuracy;
3. Isolation, evaluation, and monitoring of major components of
system variability.
In subsection 3.1, a method of assessing data quality on an intra-
team basis is given. This method involves calculating a sample standard
deviation using the three replicate runs required in a field test and cal-
culating 90 percent confidence limits for the average of the three repli-
cates.
Subsection 3.2 presents suggested criteria for judging equipment
performance, frequency of calibration, and isokinetic sampling.
Directions for the collection and analysis of information to identify
trouble, and subsequently, the control of data quality within acceptable
limits are given in the third subsection.
3.1 ASSESSMENT OF DATA
The particulate mass emission rate, PMR, for a particular field test
is the average of at least three replicates. Intrateam assessment of data
quality as discussed herein provides' for an estimate of the precision of
the measurements. Precision in this case refers to replicability, i.e.,
the variability among replicates and is expressed as a standard deviation.
This precision statement combines variability due to process changes and
to random measurement errors. This technique does not provide
67
-------
the information necessary for estimating measurement bias (see subsection
4.1'. 2 for a discussion of bias) that could occur, for example, from an
error in determining the pitot tube coefficient, nozzle, cross-sectional
area, or the orifice meter calibration. However, if the operating
procedures given in the Operations Manual are followed, the bias should be
small in most cases. An independent performance audit which would make
possible a bias estimate is suggested and discussed in section IV, the
Manual for Manager of Groups of Field Teams.
3.1.1 Calculating Precision of Field Data
Each field test is comprised of at least three sample runs. Using
the sample runs as replicates, a standard deviation can be calculated. This
calculated standard deviation is a combined measure of the measurement and
process variabilities. The standard deviation is calculated by
_ f(PMR1
- PMR)" + (PMR9 - PMR) + (PMR, - PMR)'
s{PMR}
where s{PMR} = The calculated standard deviation for the
three sample runs, g/hr,
PMR1(PMR2)(PMR3) = Particulate mass emission rate for sample
run 1 (2)(3), g/hr,
PMR = Average particulate mass emission rate of
the three sample runs, i.e., I/3(PMR + PMR2 +
PMR3), g/hr, and
2 = The number of replicates minus one (degrees of
freedom).
3.1.2 Reporting Data Quality
The average measured particulate mass emission rate (PMR) serves as a
point estimate of the true a.verage particulate mass emission rate (PMR ).
The spread in the PMR's from the three runs can be used to calculate an
interval estimate of PMR . The procedure used here assumes that the mea-
sured PMR is normally distributed about PMR (i.e., the measurement method
68
-------
2
is not biased) with a variance estimated by s {PMR}/n. Since there is no
way of determining or knowing PMR the internal estimate is actually applic-
able to the measured grand mean value PMR that would result if the number
of runs (n) became very large. If, in fact, the measurement method is
unbiased then PMR = PMR as n approaches infinity.
It is recommended that the average particulate mass emission rate,
PMR, from the three runs be reported along with 90 percent confidence limits
for the grand mean, PMR. The average measured value and calculated standard
deviation are used to calculate 90 percent confidence limits for PMR by
PMR + 2.92 s{PMR}//n
where PMR = The average of three replicates, g/hr.
s{PMR} = Estimated standard deviation of PMR based on three
replicates, g/hr.
2.92 = 95— percentile of the Student t-distribution with
2 degrees of freedom which yields a 90 percent
confidence interval.
n = The number of replicates, i.e., n = 3 for this case.
For example, if for a given field test PMR = 22.86 g/hr and s{PMR} is cal-
culated to be 1.32 g/hr, the reported value with 90 percent confidence
limits would be
22.86 g/hr. + (2.92) (1.32 g/hr.)//3
or the grand mean particulate mass emission rate, PMR, would be assumed
(with 90% confidence) to be in the interval.
20.6 g/hr. £ PMR £ 25.1 g/hr.
The utility of the above statement follows from the fact that if this
procedure for computing confidence limits is followed for several field
tests, then 90 percent of the time the grand mean PMR value will be con-
tained within the given limits. It is recommended that the 90 percent
69
-------
confidence limits be reported with the Field Data Form in the Operations
Manual.
3.2 SUGGESTED PERFORMANCE CRITERIA
Data assessment as discussed in the previous subsection is based on
the premise that all variables are controlled within a given level, thereby
guarding against large undetected biases in the measurement process. These
levels of suggested performance criteria are the values given in the Oper-
ations Manual for determining when equipment and/or personnel variability
is out of control. Criteria for judging performance are summarized in
Table 1.
The following section discusses important sources of error in Method 5
and provides information on techniques for monitoring these variables to
determine if the performance criteria given in this section are being ex-
ceeded.
3.3 COLLECTION AND ANALYSIS OF INFORMATION TO IDENTIFY TROUBLE
In a quality assurance program, one of the most effective means of
preventing trouble is to respond immediately to indications of suspicious
data or equipment malfunctions. There are certain visual and operational
checks that can be performed while the measurements are being made to
help insure the collection of data of good quality. These checks are
written as part of the routine operating procedures in section II. In
order to effectively apply preventive-type maintenance procedures to the
measurement process, the supervisor must know the important variables in
the process, know how to monitor the critical variables, and know how to
interpret the data obtained from monitoring operations. These subjects are
discussed in the following subsections.
3.3.1 Identification of Important Variables
Determination of the particulate mass emission rate requires a
sequence of operations and measurements that yields as an end result a
number that serves to represent the average particulate mass emission rate
for that field test. There is no way of knowing the accuracy, i.e., the
70
-------
Table 1. Suggested performance criteria
1. Suggested Criteria for Equipment Performance
(a) Dry Gas Meter:
(b) Barometer:
(c) Thermometers:
(d) Stack Temperature
Measuring System:
(e) Sampling Train Leakage:
(f) Sample Box Heating
System:
(g) Probe Heating System:
(h) Meter Orifice:
4QK
(i) Probe Nozzle Diameter:
(j) Analytical Balance:
(k) Type-S Pitot Tube:
.98 £ y 1 1-02
+ 5.1 mm Hg (+0.2 in. Hg)
+ 2.3°K (+ 5°R) at 273PK (492°R) or
(+ 7°R) at 373°K (672°R)
+ 2.3°K (+ 5°R) at 273°K (492°R) or + 4°K
(+ 7°R) at 373°K (672°R)
3 3
Less than 0.0006 m /min (0.02 ft /min) at
380 mm Hg (15 in. Hg) vacuum
Capable of maintaining a temperature of 120°C
(248°F) + 14°C (25°F) at laboratory conditions.
Uniform heating of probe with a
temperature of 120°C (248°F) at the exit
end at a flow rate of 354 cnr/sec (0.75
ft /min) at room temperature.
AH@ should be 46.7 + 6.4 mm H20 (1.84 + 0.25
in. H20) and not vary more than +3.8 mm H2<3
(+0.15 in. H20) over the range of operation
of 13 to 200 mm H20 (0.5 to 8 in. H20) , if a
commercial nomograph is used during sampling
to aid in maintaining an isokinetic sampling
rate.
Range of three different diameter
measurements less than .010 cm (0.004 in.).
Weigh a standard weight within +0.3 mg
of its stated weight.
C constant within + 5 percent over working
range and each calibration check is within
1.2 percent of the original C .
6.0 < pH < 8.0
(1) Filter Media:
Suggested Criteria for Performing Equipment Calibration
(a) Above items (a) through (j) are calibrated when new and checked before
each field test and recalibrated any time the check results fall outside
the prescribed performance limits.
(b) Item (k), the type-S pitot tube is calibrated when new, before every
third field test, or at any sign of damage.
(c) Item (1), filter media, the pH's of a random sample of 7 out of 100
filters are measured for each new order of filters. The remaining
93 filters are accepted if all 7 pH's are in the interval of 6.0 to 8.0.
Suggested Criteria for Percent Isokinetic
0.90 < I < 1.10
71
-------
agreement between the measured and the true value, for a given field test.
However, a knowledge of the important variables and their characteristics
allows for the application of quality control procedures to control the
effect of each variable at a given level during the field test, thus
providing a certain degree of confidence in the validity of the final
result.
A functional analysis of this method of measuring the particulate mass
emission rate of a stationary source was made (see subsection 4.1) to try to
identify important components of system error and show how these errors prop-
agate through the measurement process and influence the final result. Results
of an evaluation study of Method 5 (ref. 15) for within-run variations
showed an average coefficient of variation of about 6 percent for twelve
runs, six of which had four sampling trains and the other six runs had
three sampling trains. Also, collaborative tests of the method (refs. 16,
17, and 18) show within laboratory CVs ranging from 10 to 30 percent and
between laboratory CVs of 20 to 40 percent. These results are used in
the functional analysis as an estimate of overall system error, while the
individual error components are estimated using engineering judgment in a
manner such that their combined variability is consistent with overall
system error.
Variability in emissions data derived from multiple repetitions
include components of variation from:
1. Process conditions,
2. Equipment and personnel in field procedures, and
3. Equipment and personnel in the laboratory.
In many instances time variations in source output may be the most
significant factor in the total variability. In order to judge the rela-
tive magnitudes of measurement variability and process output variability,
process parameters should be monitored throughout the test. Process infor-
mation is also required if the particulate mass emission rate is to be
given as a function of fuel input. The exact process data to be obtained
are dependent upon the process being tested. In general, all factors which
72
-------
have a bearing on the emissions should be recorded on approximately a 15-
minute interval (ref. 19). These factors include process or fuel weight
rate, production rate, temperature and pressure in the reactor and/or
boiler, control equipment, fan and/or damper settings, pressure drop or
other indicators of particulate collection efficiency and opacity of exit
plume. Sample forms for combustion, incineration, and process sources
are' given in reference 19.
It is most important to realize that the larger measurement errors
result from poor operator technique such as loss or gain of collected
particulate mass during sample recovery, (ref. 17), or poor orientation
and positioning of the probe during sample collection. Such deviations
from recommended procedures cannot be evaluated or corrected for. It is
important to observe and eliminate such occurrences while the test is in
process.
Sources of variation due to equipment includes the type-S pitot tube
coefficient, sampling nozzle cross-sectional area, orifice meter, dry gas
meter, probe heater, non-neutral filter media, and sample box heater.
These parameters are all controlled through performance of calibration
checks before each field test, and in the case of the filter media, by
an acceptance check for new batches of filters. Also, the probe and sample
box heaters are checked periodically during the conduction of the test.
Important error sources checked immediately before and/or during
sample collection includes sampling train leaks, the sample gas tempera-
ture leaving the last impinger, and isokinetic sampling conditions.
Assuming good operator technique, the error sources named in the last
two paragraphs are discussed and each one's effect on the determination of
the particulate mass emission rate is derived from a functional analysis
of the measurement process in subsection 4.1.
A summary of the important parameters is given below. The parameters
are roughly given in ascending order of importance. Importance is quali-
tatively derived from the estimated error range of a parameter weighted by
its estimated frequency of occurrence.
73
-------
3.3.1.1 Equipment Calibration. Equipment calibration is the backbone of
any quality assurance program. It is important that the calibration
procedure be carried out correctly, that the calibration standards are
properly calibrated and maintained, and that the frequency of calibration
is adequate.
Important calibration constants and how they influence measurement
accuracy include the following:
1. Error in the pitot tube calibration coefficient is directly
reflected in the stack gas velocity determination and is doubled in the
process of determining isokinetic sampling rates. This error could be very
large if the pitot tube is not calibrated under actual field test condi-
tions, i.e., strapped to the sampling probe and the spacings shown in
Figure 5-1 of Appendix A are not maintained (ref. 7). Errors as large as
20 percent have been attributed to the pitobe configuration (ref. 7),
2. Error in determining the average nozzle diameter is quadrupled
in the process of determining isokinetic sampling rates and is doubled in
the percent of isokinetic sampling calculation. One source of error here
is the use of an out-of-round nozzle. The average nozzle diameter is used
to calculate the area of a circle which yields a larger than true cross-
sectional area if the nozzle tip is not round,
3. Dry gas meter inaccuracy appears directly in the concentration
and particulate mass emission rate determinations.
4. The orifice meter calibration constant is used in determining
isokinetic sampling conditions and any error in the constant is doubled
in setting the sampling rate.. Also, if AH@, the pressure drop across
3 3
the orifice that gives a flow rate of 0.21 m /min (0.75 ft /min) at 21°C
(70°F) and 760 mm Hg (29.92 in. Hg) differs from 46.7 mm H20 (1.84 in. HO),
and a nomograph is used to set isokinetic sampling conditions, an error
results. It is recommended that an orifice meter with a AH@ outside the
range of 40.4 to 53.1 mm H^O (1.59 to 2.09 in. ELO) not be used in conjunc-
tion with a nomograph.
74
-------
3.3.1.2 Anisokinetic Sampling. Anisokinetic sampling can occur from error
in the calibration constants of the pitot tube, orifice meter, and nozzle
diameter. It can also result to a lesser degree, usually, from measurement
error in the moisture content and molecular weight of the stack gas.
Errors from the above sources will not be directly reflected in the per-
cent of isokinetic sampling calculation. Therefore, it is important to
determine each parameter as accurately as possible, either through cali-
brations or careful measurements.
Failure, or in some instances the inability, to make adjustments in
the sampling rate as the stack gas velocity varies or as the deposited
particulate matter plugs the filter can result in anisokinetic sampling.
Use of a nomograph can be a cause of anisokinetic sampling because of
(1) any inaccuracy in the nomograph, (2) use of preset values for C , AH@,
and M, (these errors can be eliminated by using actual values and adjusting
the correction factor on the nomograph), and (3) operator error in setting
the nomograph. The sum of these errors is quantified to a certain extent
by the percent of isokinetic sampling calculation.
Deviation from isokinetic sampling cannot be related directly to error
in the measurement process (see subsection 4.1). However, failure to main-
tain isokinetic sampling conditions under otherwise normal operations
reflects the lack of alertness and, perhaps even the level of competency,
of the field crew.
3.3.1.3 Sample Recovery. The technique used by, and the attitude of, the
crew members in sample recovery are of paramount importance to measurement
precision and accuracy (refs. 17 and 18). Use of an adequate sample
recovery area in terms of space, lighting, or cleanliness will decrease the
probability of error. Sample recovery procedures as given in Section 4.2
of Appendix A should be followed by the field team.
3.3.1.4 Calculations. Calculations for this method are known to be a
major source of error (ref. 18). Some calculations involve several terms and
should only be attempted (for the final report) at a desk or work table and
preferably with the aid of a calculator or at least a good slide rule. A
computer program using raw data as an input is highly recommended for
75
-------
making the final calculations.
As a check, it is recommended that all calculations be independently
repeated from raw data.
3.3.1.5 Filter Media. This is included as a possibility. It is not known
how important the pH of the filter media is at this time. However, data
referenced in subsection 4.1 indicates that error due to the conversion of
acid gases to particulate ma.tter by alkaline filters could be significant.
This error would be a positive bias resulting in a measured emission rate
greater than the true rate.
3.3.2 How to Monitor Important Variables
In general, if the procedures outlined in the Operations Manual are
followed, the major sources of measurement variability will be in control.
It is felt, however, that the supervisor should visually check certain
parameters and operations periodically while measurements are being made
to insure good operator technique and the proper use of equipment. The
parameters and operations to check are essentially those recommended for
the auditor as listed in table 7 of subsection 4.3.
Results of the calibration checks for the dry gas meter, orifice
meter, nozzle diameter, and pitot tube should be checked before each field
test. Any item of equipment not satisfying the suggested performance
criteria of table 2 should be calibrated or replaced.
There appears to be a need for actual field data on several of the
parameters or variables involved in this measurement method in order to
better judge their influence on measurement variability. One of the most
effective means of identifying and quantifying important sources of
variability is through the use of quality control charts. Quality control
charts will provide a basis for action with regard to the measurement
process; namely, whether the process is satisfactory and should be left
alone, or the process is out of control and action should be taken to find
and eliminate the causes of excess variability. In the case of this method
in which documented precision data are scarce, the quality control charts
can be evaluated after a period of time to determine the range of variation
76
-------
that can be expected under normal operating conditions. Also, even though
results from individual field tests are within bounds, trends can be identi-
fied and corrective action taken further improving data quality through the
proper use of control charts.
Discussions of control charts and instructions for constructing and
maintaining them are given in many textbooks in statistics and quality
control, such as in references 20 and 21. Also, volume 1 of the Quality
Assurance Handbook for Air Pollution Measurement Systems published by EPA
discusses the use of control charts (ref. 22).
It is good practice to note directly on control charts the reason for
out-of-control conditions, if determined, and the corrective actions taken.
Recommended control charts are discussed below.
3.3.2.1 Pitot Tube Calibration Coefficient. A sample control chart for
pitot tube calibration checks is given in the quality assurance document
of this series for Method 2 (ref. 1).
3.3.2.2 Determination of %CO? with Orsat Analyzer. For incinerators
where the particulate concentration or particulate mass emission rate is
corrected to 12 percent CO-, it is recommended that control charts for the
field replications and for calibration checks be maintained as given in
the quality assurance document of this series for Method 3 (ref. 2).
3.3.2.3 Range Chart for Runs. In compliance testing where it is desired
to determine the source output at a fixed level of operation, a large range
in the three runs (replicates) would suggest process variability and/or
measurement variability. Expressing the range, R, as a percent of the
average, i.e., the difference in the largest and smallest of the three
replicates divided by the average of the three replicates, all multiplied
by 100; a control chart with limits as given in Figure 5 can be used
initially. Results from collaborative tests of Method 5 show within labora-
tory CV's of 10, 25, and 31 percent. Therefore, the CV = 10 percent used
in constructing the control chart is the least of the three values reported
above. These limits should serve as a starting point. When a data point
falls out of bounds on this graph, the process data should be checked to
77
-------
see if the process changed between runs and the percent of isokinetic
calculations checked to see if one run was significantly different from
the others in order to identify the cause of the excessive variability.
Note that exceeding the upper control limit does not necessarily invali-
date the test data.
3.3.2.4. Mean and Range Charts for Percent of Isokinetic Sampling. Main-
taining isokinetic sampling conditions is important in particulate sampling.
Control charts displaying the range and mean of the calculated percent
isokinetic sampling provide, at a glance, means for evaluating the per-
formance of a team or groups of teams over an extended period of time. If
deviations from isokinetic greater than + 10 percent are not allowed,
i.e., the run has to be repeated, then in a rough way 10 percent can be
taken as the 3a value giving a standard deviation of about 3.3 percent.
Based on three replicates and the above standard deviation, the range and
mean charts for I are given in Figures 6 and 7, respectively.
The R values are plotted sequentially as they are obtained and con-
nected to the previously plotted point with a straight line. Corrective
action, such as instruction in proper operating technique, should be
taken before the next field test any time one of the following criteria
are exceeded:
1. One point falls outside the UCL.
2. Seven consecutive points fall above the R line (i.e., the range
of I obtained by a given field team is consistently greater than R for
seven or more field tests). Exceeding the first criteria will usually
indicate poor technique or equipment malfunction between sample runs of
a particular field test. Exceeding the second criteria indicates a
systematic error due to equipment bias or poor technique. (Note that the
UCL can be exceeded without violating the I = 100 + 10 interval for any
one run.)
The I values, i.e., average percent of isokinetic sampling per three
sampling runs, are plotted sequentially as they are obtained from field
tests and connected to the previously plotted point with a straight line.
Corrective action, such as instruction in proper operating technique
78
-------
c
111
o
EC
I
O
O
X
cc
50
40
30
20
10
FIELD TEST
NO.
DATE
FIELD TEAM
PROBLEM
AND
CORRECTIVE
ACTION
1
2
3
4
5
6
7
c
c
8 £
"D 'O
% CO
cu ra
-D t3
O S
C CD
"^ ^
0 01
o —
v> .¥?
1- (—
0) *~
£ cX
CD C
TO E
Q.
C
K) <"
0 I
o S
fL O
8
9
^
C
*-
O
^$
o
CO
CO
CD
3
o
CD
C
y
O
.i2
•M
C
0)
CD
Q_
C
CD
%
QJ
-C
*
"O
CD
-o
c
CD
trt
O>
c^
'*"' "O
±: CD
-1 »
CD Q
c QJ
0 ^
V o
C
- t/>
^ OJ
a. ?
10
*UCL = D2a = 4.358 X 10 = 44
tR = d2a = 1.693 X 10 = 17
Figure 5. Sample control for the range, R, of PMR replicates.
79
-------
c
0)
c
'E
X
ra
E
en
C
cr
18
16
'14
12
10
8
6
4
2
UCL = 14.4"
FIELD TEST
NO.
10
DATE
FIELD TEAM
PROBLEM
AND
CORRECTION
ACTION
*UCL = D2cr = 4.358 X 3.3 = 14.4
fR = d20 = 1.693 X 3.3 = 5.6
Figure 6- Sample control chart for the range, R, of percent
isokinetic, I, sampling for three test runs per
field test.
80
-------
u
CJ
Q_
s-\
IM
o
-t-j
cu
<
106
104
102
100
98
96
94
UCL = 105.7*
Warning Limit = 103.6
Warning Limit = 96.2
LCL = 94.3
FIELD TEST NO.
10
DATE
FIELD TEAM
PROBLEM AND
CORRECTIVE
ACTION
*UCL = I + 30{I} = 100 + 3 X 3.3/>/3~'= 105.7
Figure 7. Sample control chart for the average percent of
isokinetic sampling per field test.
81
-------
and/or performing equipment calibration checks should be taken before
attempting the next field test any time one of the following criteria is
exceeded.
1. One point falls outside the UCL or LCL.
2. Two out of three consecutive points fall in the warning zone
(between 2a and 3a limits).
Exceeding the first criteria will usually indicate poor technique
or equipmerit malfunction. The second criteria when exceeded indicates
an assignable source of variability due either to faulty equipment or a
consistent error in performing the operation procedures.
82
-------
SECTION IV
MANUAL FOR MANAGER OF GROUPS OF FIELD TEAMS
83
-------
SECTION IV MAI FOR WNAGER OF GROUPS OF FIELD TEW1S
4.0 GENERAL
The guidelines for managing quality assurance programs for use with
Test Method 5 - Determination of Particulate Emissions from Stationary
Sources are given in this part of the field document. This information is
written for the manager of several teams for measuring source emissions and
for the appropriate EPA, State, or Federal Administrators of these programs.
It is emphasized that if the analyst carefully adheres to the operational
procedures and checks of Section II, then the errors and/or variations in
the measured values should be consistent with the performance criteria as
suggested. Consequently, the auditing routines given in this section
provide a means of determining whether the stack sampling test teams of
several organizations, agencies, or companies are following the suggested
procedures. The audit function is primarily one of independently obtaining
measurements and performing calculations where this can be done. The pur-
pose of these guidelines is to:
1. Present information relative to the test method (a functional
analysis) to identify the important operations and factors.
2. Present a methodology for comparing action options for improving
the data quality and selecting the preferred action.
3. Present a data quality audit procedure for use in checking
adherence to test methods and validating that performance criteria are
being satisfied.
4. Present the statistical properties of the auditing procedure in
order that the appropriate pLan of action may be selected to yield an
acceptable level of risk to be associated with the reported results.
These four purposes will be discussed in the order stated in the
sections which follow. The first section will contain a functional analy-
sis of the test method with the objectives of identifying the most important
factors which affect the quality of the reported data and estimating the
expected variation and bias in the measurements resulting from equipment
and operator errors.
84
-------
Section 4.2 contains several actions for improving the quality of the
data; for example, by improved analysis techniques, instrumentation, and/or
training programs. Each action is analyzed with respect to its potential
improvement in the data quality as measured by its precision. These
results are then compared on a cost basis to indicate how to select
the preferred action. The cost estimates are used to illustrate the metho-
dology. The manager or supervisor should supply his own cost data and his
own actions for consideration. If it is decided not to conduct a data audit,
sections 4.1 and 4.2 would still be appropriate as they contain a functional
analysis of the reference method and of alternative methods or actions.
There are no absolute standards with which to compare the routinely
derived measurements. Furthermore, the taking of completely independent
measurements at the same time that the routine data are being collected (e.g.,
by introducing two sampling probes into the stack and collecting two samples
simultaneously) is not considered practical due to the constrained environ-
mental and space conditions under which the data are being collected.
Hence, a combination of an on-site systems audit, including visual observa-
tion of adherence to operating procedures, and a quantitative performance
audit is recommended as a dual means of independently checking on the
source emissions data.
The third section contains a description of a data quality audit pro-
cedure. The most important variables identified in section 4.1 are considered
in the audit. The procedure involves the random sampling of n stacks from
a lot size of N = 20 stacks (or from the stacks to be tested during a three-
month period, if less than 20) for which one firm is conducting the source
emissions tests. For each of the stacks selected, independent measure-
ments will be made of the indicated variables. These measurements will be
used in conjunction with the routinely collected data to estimate the
quality of the data being collected by the field teams.
The data quality audit procedure is an independent check of data col-
lection and analysis techniques with respect to the important variables.
It provides a means of assessing data collected by several teams and/or firms
with the potential of identifying biases/excessive variation in the data
collection procedures. A quality audit should not only provide an independent
85
-------
quality check, but also identify the weak points in the measurement
process. Thus, the auditor, an individual chosen for his background
knowledge of the measurement process, will be able to guide field teams in
using improved techniques. In addition, the auditor is in a position to
identify procedures employed by some field teams which are improvements
over the current suggested ones, either in terms of data quality and/or
time and cost of performance. The auditor's role will thus be one of
aiding the quality control function for all field teams for which he is
responsible, utilizing the cross-fertilization of good measurement tech-
niques to improve the quality of the collected and reported data.
The statistical sampling and test procedure recommended is sampling
by variables. This procedure is described in section 4.4. It makes max-
imum use of the data collected, and it is particularly adaptable to the
small lot size and, consequently, the small sample size applications. The
same sampling plans can be employed in the quality checks performed by a
team or firm in its own operations. The objectives of the sampling and
test procedure are to characterize data quality for the user and to identify
potential sources of trouble in the data collection process for the purpose
of correcting the deficiencies in data quality.
Section 4.4.4 describes how the level of auditing, i.e., the sample size
n, may be determined on the basis of relative cost data and prior information
about the data quality. This methodology is described in further detail in
the Final Report on the Contract. The cost data and prior information con-
cerning data quality are supplied to illustrate the procedure and these data
must be supplied by the manager of groups of field teams depending upon the
conditions particular to his responsibility.
Figure 8 provides an overall summary of the several aspects of the data
quality assurance program as described in these documents. The flow dia-
gram is subdivided into four areas by solid boundary lines. These areas
correspond to specific sections or subsections of the document as indicated
in the upper right hand corner of each area. The details are considered
in these respective sections of the document and will not be described here.
86
-------
Pollutant
Measurement
Method
Functional
Analysis
I
Subsection 4.1
Estimate Ranges
and Distributions
of Variables
_L
Identify and Rank
Sources
Bias/Variation
Perform Overall
Assessment
Section III
Develop Standards
for Q. C.
Procedure
Institute
QC Procedure
for Critical
Variables
Subsection 4.2
Data
are of
Satisfactory
Quality
(Optional)
Evaluate Action Options
for Improving Data
Quality
Continue to Use
Measurement Meth.
as Specified
Cost of
Implementing
Actions
Select Optimal
Action and
Implement
Modified
Measurement
Method
Subsections 4.3 and 4.4
Develop Standards
for Audit Procedure
Select Aud .
Procedure
Based on
Statistics
and /or
Cost
Cost Data:
Audit,
CP
cr
G'
p
Assess Data
Quality Using
Audit Data
No
Data
Quality
Satisfactory
Yes
Continue to Use
Measurement Method
as Specified
Figure 8. Summary of data quality assurance
87
prcgran:.
-------
4.1 FUNCTIONAL ANALYSIS OF TEST METHOD
Test Method 5 - Determination of Particulate Emissions from Stationary
Sources—as described in Appendix A of this document is subjected to a
functional analysis in this section. This measurement method is used to
determine the concentration of particulate matter in the stack gas. It
is also used to simultaneously determine the moisture content of the stack
gas and to perform a velocity traverse from which the volumetric flow rate
is calculated. These results combined with the stack gas composition as
measured by Method 3 yield a particulate mass emission rate for the source
being tested.
Method 5 as promulgated in the Federal Register, Vol. 36, No. 247 -
Thursday, December 23, 1971 has been subjected to collaborative tests on
a Portland cement plant (ref. 16), a fossil fuel-fired steam generator
(ref. 17), and a municipal incinerator (ref. 18). These tests showed
within laboratory precisions (expressed as coefficient of variation) rang-
ing from 10 to 30 percent and between laboratory precisions of 20 to 40
percent. Several recommendations were made for improving the precision of
the method based on the collaborative test results (refs. 16, 17 and 18).
The method (i.e., Method 5) has been revised by EPA incorporating
the above mentioned recommendations as well as other changes that should
improve the precision of the method. The revised method is reproduced as
Appendix A of this document. The functional analysis presented in the
following subsections is based on what can hopefully be achieved using the
revised method. A within laboratory coefficient of variation of 10 percent
may be achievable by highly motivated, well-trained, stack sampling teams.
The changes and/or additions made by EPA to the method that should
result in significant improvement in data quality include:
1. Minimum free spaces between the pitot tube, sampling nozzle, and
temperature sensor are specified (see Figure 5-1 in Appendix A).
2. Calibration of the 1;ype-S pitot tube is required prior to initial
use and at any sign of damage, plus the calibration standard, facility and
procedures are fully specified.
3. Greater detail on sample recovery is provided.
88
-------
4. Sampling train leak checks are required before and after a run
and are recommended between port changes.
The purposes of the functional analysis as performed here are to:
1. To identify operations, variables and factors that can influ-
ence the quality of the measurement data.
2. To illustrate how elements of uncertainty are propagated through
the measurement process to the resultant particulate measurement.
3. To estimate the mean values and ranges of the various error
sources existing under normal operating conditions and to deter-
mine the combined probable uncertainty to be associated with
the reported measurements.
The results of the functional analysis provide information for the
following:
1. Setting acceptable limits on data quality, i.e., precision and
accuracy.
2. Applying quality control procedures to those operations and vari-
ables that would otherwise result in excess variability in the
measurement data.
3. Applying appropriate quality assurance practices (e.g., audits)
at critical points in the measurement process with acceptable
limits specified and actions to be taken when limits are exceeded.
The functional analysis is discussed in two stages. First the vari-
ables and operations involved in the measurement process are identified
and modeled, i.e., their mean values and variances are estimated. These
estimates are made using published data if available or engineering
judgment if no quantitative data are available. Second, a variance analysis
(sometimes referred to in the literature as a sensitivity analysis or
error analysis) is performed to determine the individual and combined
influence of the variables and operations to the particulate mass emission
rate measurement.
The variance analysis as presented in this section provides a mathe-
matical means of identifying important variables in the measurement pro-
cess. The estimates of the mean values and variances of the variables
89
-------
are made as realistic as possible, however, the major purpose of this sec-
tion is to illustrate a method. Each laboratory should work through the
analysis using experimental data or estimates based on conditions within
that laboratory. The mathematical basis for the procedures are described
in the Final Report on Contract 68-02-0598 (ref. 4).
The particulate mass emission rate (PMR) is calculated from the rela-
tionship :
PMR = Hf3g/mg (m/V) (V • A) (1)
where
PMR = Rate of particulate emissions, g/hr.
m = Total weight of particulate collected, mg.
n 3
V , ... = Volume of gas sampled corrected to standard conditions, m .
V = Stack gas velocity, m/s.
s 2
A = Crossectional area of stack, m .
S
Variability in the measured PMR will have components of variability from
the individual determinations of m , V , ,. , V and A . Variability of
n' m(std) s s y
each of the four terms is discussed separately below following a short
definition of terms.
Many different measures of variability are conceivable depending
upon the circumstances under which the measurements are performed. Three
measures of variability are defined here. They are repl icability, repeat-
ability and reproducibility defined as follows (ref. 22).
1. Replicability. The precision measuring the variability among
replicates. Replicates are repeated but independent determinations of the
same sample, by the same analyst, at essentially the same time and same
conditions. This could apply to the three runs comprising a field test
provided the process variables remained constant.
2. Repeatability is a measure of the variability in determinations
made on the same sample (i.e., the same source with source parameters held
as constant as possible) by the same field team using the same equipment
over a short period of time. The repeatability standard deviation and the
coefficient of variation are symbolized by a and CV respectively. This
90
-------
measure of precision is referred to as the within laboratory standard
deviation in the collaborative test reports.
3. Reproducibility is the variability associated with measurements
made of the same sample (i.e., same source with source parameters held as
constant as possible) by field teams from different laboratories. This
measure of precision is referred to as between laboratory precision in the
collaborative test reports.
The above definitions are based on a statistical model where each
determination of the rate of emissions is the sum of three components as
follows:
PMR = PMR + b + e (2)
where
PMR = the measured rate of particulate emissions, g/hr.
PMR = the general or true average, rate of particulate emissions,
g/hr.
b = an error term representing the differences between field
teams equipment, sources, etc., g/hr.
e = a random error occurring in each measurement, g/hr.
In general, b can be considered as the sum:
b = b + b (3)
r s
where
b = random component
b = systematic component
S
The term b is considered to be constant during any series of measurements
performed under replicability conditions, e.g., the three runs making up
a field test. It behaves as a random variate in a series of measurements
during which personnel, equipment or conditions change e.g., between field
tests or laboratories. Its variance is denoted by
var b = a, (4)
b
91
-------
2
where a is the bias component of variability due to different field teams.
equipment, etc.
The term e represents a random error occurring in each measurement.
Its variance is denoted by
2
var e = a (5)
2
where a is called the replicability variance.
The repeatability standard deviation, a , is related to the bias
variance and the replicability variance and the number of replicates r
by
V
Repeatability of particulate mass emission rate determinations will
be discussed in the following, subsections.
4.1.1 Mass of Particulate Collected, m
The total measured particulate mass, m , is given by
m = m + e, (7)
where m represents the true particulate mass in the gas volume at stack
conditions equivalent to the volume V , ,x as measured by the dry gas
m(.std;
meter and corrected to standard conditions. The difference between the
true and measured values is the error, e.
Determination of particulate mass is subject to error from
1. Loss or gain of sample during sample recovery (SR),
2. Weighing of collected sample (W),
3. Conversion of acid gases to particulates by alkaline glass fiber
filters (pH),
4. Formation of particulates due to non-uniform or inadequate heating
of the probe and/or sample box (C), and
5. Deviation from isokinetic sampling (I).
These five components of variability are independent (at least there
92
-------
are no obvious correlations) and with each component's effect an m given
in mg the variance of m becomes:
a2(m } = a2 (m } + 02{m } + a2 {m } + a2{m } + a2{m }. (8)
n SRn Wn pHn en In
Magnitudes of each of the five variance components are discussed in
the following subsections. Using the variances, as estimated in the fol-
lowing subsections, in the above relationship gives a total variance for
m of
n
a2{mn> = 0.0016 x m2 + 0.0016 x m2 + 9 yg2 + 4 yg2 + 0.0034 x m2.
By rearranging and combining terms the variance becomes
a2{mn> = 0.0066 m2 + 13 mg2. (9)
Values of cr{m } and CV{m } for different values of true particulate
n n
mass, m , are given in a table in subsection 4.1.1.6.
4.1.1.1 Sample Recovery (SR). Error due to a loss_or gain of sample dur-
ing sample recovery is mostly a function technique and of the adequacy of
the work area used for sample recovery. Sample recovery includes any
excess sample due to scraping the probe against the stack wall or sampling
port as well as transferring the filter and collected particulate from the
filter holder to a sample container and removing particulate matter from
the nozzle, probe, filter holder, and connecting glassware. No quantita-
tive data are available for estimating the magnitude of this error. How-
ever, it was considered to be a major source of error by those conducting
the collaborative tests (refs. 16, 17, and 18). For this analysis an error
proportional to the true mass, m , is assumed. A relative error (coefficient
of variation) of 4 percent given a stand deviation of acr){m } = 0.04 x m mg
2 22 n
and a variance of ac_{m } = 0.0016 x m mg . It is further assumed that
I>K n t
over a large number of field tests there will be as many positive as
negative errors i.e., this error can be modeled as a normally distributed
2
variate having a zero mean and a variance of cr {m }. This is symbolized
~ n SR n
as N(y, o ) = N(0, 0.0016 m).
93
-------
4.1.1.2 Weighing Errors (W). Sample weighing errors are a combination of
four weighings. The weight of particulate matter collected m is
m = (W0 - W.) + (W. - W,) (10)
n / 1 H J
where
m = The collected mass of particulate matter, mg
W, = The tare weight of the filter, mg
W7 = The weight of the exposed filter, mg
W., = The tare weight of a beaker or drying dish, mg, and
W, = The gross weight of the beaker and the particulate
material collected in the probe, mg.
If the recommended procedures are followed for checking the balance
measurements against measurement standards before each weighing, the
associated weighing errors for each of the weighings, W\., W^, W^, and W^
would most likely result from improper conditioning and handling rather
than from the balance. Under this assumption the weighings are uncorrelated
and the variances can be combined as follows.
a2 {m } = a2{W.,} + a2{W.} + a {W-} + a {W. } (11)
w n 1 / 3 4
Weighings W, and W., are ta.re weighings and should have greater pre-
cision than weighings W, and W., which involved the particulates recovered
from the probe assembly and the filter, respectively. These latter weigh-
ings are sensitive to the type of particulates i.e., if they are hygro-
scopic (ref 17) or of high organic content. Rather than estimate the
individual weighing errors the total standard deviation of m attributable
to weighing operations is estimated to be 4 percent of the true particulate
2 2
mass, i. e., a {m } = 0.04 x m mg or a variance of a {m } = 0.0016 m .
w n t w n t
4.1.1.3 Filter Surface Alkalinity (pH). The surface of glass fiber filters
is generally alkaline with pH's ranging from about 7 to 11. Acid gases in
the sample air can be oxidized and deposited as particulate matter on
these filters. The quantity of pseudo-particulates formed is probably a
94
-------
function of the filter pH, concentration of the acid gases in the sample
gas, residence time, and duration of the sampling period. Tests dealing
with measuring suspended particulates in ambient air showed a much higher
particulate catch when using filters with a pH of 11 as compared to filters
with a pH of 6.5 (refs. 23 and 24). One series of nine tests involving
samplers operated side by side, one with a pH-11 filter and the other with
a pH-6.5 filter, showed an average particulate catch 18 percent greater
for the alkaline filters (ref. 23). A glass fiber filter used as a back-
up to a cloth primary filter in a sampling train showed a weight gain due
to soluble sulfates as large as the total particulate catch of the primary
filter (ref. 25).
The above data on alkaline fi-ters are not directly applicable to this
application, thus an error limit would be strictly an estimate. For
this analysis a range of from 0 to 9 rag is assuemd as the potential gain
in particulate matter due to conversion of acid gases on the filter sur-
face. This error acts as a positive bias and is modeled as the absolute
2
value of a normal distribution with a zero mean and a variance of 9 mg or
a standard deviation of a {m } = 3 mg, i.e., |N(0, 9 gm)|. Such a distri-
pn n
bution (taking absolute values) has a mean value of approximately + 2.4 mg
and cannot have a negative value.
4.1.1.4 Condensation (C). Hemeon and Blank (ref. 25) discuss the pos-
sibility of pseudo-particulate matter being formed in the probe prior to
the filter by oxidation of SO^ in the sample gas. This could occur if the
probe is not uniformly heated and condensation occurs, followed by heating
and revaporization. Also, many compounds pass from the gaseous to the solid
phase between 93°C to 148°C (200 to 300°F) (ref. 26); thus, sampling trains
performing at different operating temperatures could generate biased data.
Insufficient data are available to model this source of error. Therefore,
just to include it as a significant error, it is treated as a normally
distributed random variable with a 4 mg mean and a standard deviation of
2
2 mg, i.e., N(4 mg, 4 mg ). Under these conditions, it acts generally as
a positive bias but could go negative simulating a probe operating above
120°C (248°F), for example.
95
-------
4.1.1.5" Nonisokinetic Sampling (I) . Deviation from isokinetic sampling
conditions may result in particulate measurement error. The magnitude
of error depends on the degree of departure from isokinetic conditions,
and on the particle size distribution in the sample gas. Departure from
isokinetic sampling can occur due to failure to adjust the nozzle velo-
city as the stack gas velocity varies. This type of departure will be
detected when the percent of isokinetic sampling is calculated. However,
errors in the pitot tube coefficient and/or in the nozzle diameter can
cause deviations from isokinetic conditions which are not detectable from
any checks that can be performed while in the field.
The degree of deviation from isokinetic sampling is not a direct
indication of the error in the final result. The error resulting from
nonisokinetic sampling is a function of the particle size distribution in
the stack. For gases and small particles (diameters <^ 5 ym) isokinetic
sampling is not necessary (ref. 27). Sample gases with particle size
distributions extending in the 25 ym and above sizes are sensitive to
anisokinetic sampling. Each particle size distribution has to be evaluated,
For this analysis, using a particle size distribution of 80 to 100 ym
particles (ref. 19) a variation of +_ 5 percent from isokinetic sampling
shows a relative variation of approximately + 10 percent in the collected
mass or a ratio of 2 to 1. For this analysis a 1 percent deviation from
isokinetic conditions is taken to be a 2 percent error in the measured
particulate concentration, i.e.,
T{iri } = 2 x CV{I} x m .
In t
Percent of isokinetic sampling is derived from the ratio of the gas
velocity in the nozzle to the stack gas velocity in the stack by
I = V /V . (13)
n s
The component of variability in m due to nonisokinetic sampling is taken
as
96
-------
oT{m } = 2 x CV{V /V } x m .
In n s t
(14)
The coefficient of variation CV{V /V } is derived in the following
Tl S
manner.
The estimated coefficient of variation of V based on
n
1/2
—
n
4 K
m
,D2
n
1
1
- B
wm
- B
ws
T AH
m
P M
m m
(15)
is given by
CV2{V
n
~ CV{K
m
4CV2{D
n
CV{1 - B
CV{1 - B
wm
ws
0.25(CV{T
m
CV2{P
m
CV2{M
m
CV2{AH)
(16)
where
V = Sample gas velocity in the nozzle.
D = Average nozzle diameter.
B = Water vapor in sample gas at the dry gas meter (taken as
zero for this analysis).
B = Water vapor in the sample gas by proportion.
T = Absolute temperature at the dry gas meter.
P = Ambient pressure at the meter.
M
m
Molecular weight of the sample gas at the meter.
AH = Pressure drop across the sampling train orifice.
K = A constant whose relative error depends on the errors in
m r
the individual measured values in the equation.
Table 2 gives the assumed mean values and the estimated coefficients
of variation of the individual parameters in the above equation.
Substituting the estimated CV's from Table 2 into the above equation
16 yields a coefficient of variation for V of
CV{V } = 2.04%.
n
97
-------
Table 2 . Means and variabilities of parameters affecting V
n
Parameter
K
m
D
n
1 - B
wm
1 - B
ws
T
m
AH
P
m
M
m
V
s
V
n
Mean
Metric
1.165 x 10~3
.635 cm
0.0
0.90
294°K
102 mm H20
762 mm Hg
30 g/g-mole
15.24 m/sec
15.24 m/sec
Value
(English)
(1.84)
(0.25 in)
(0.0)
(0.90)
(530°R)
(4.0 in H20)
(30.0 in Hg)
(30 Ib/lb-mole)
(50 ft/sec)
(50 ft/sec)
Estimated
CV
CV{K }
m
CV{D }
n
0.0
CV{1 -
CV{T }
m
CV{AH)
CV{P }
m
CV{M }
m
CV{V }
s
CV{V }
n
= 1.0%
= 0.5%
B } = 1.2%
wo
= 0.5%
= 1.0%
= 0.3%
= 1.0%
= 2.07%
= 2.04%
A functional analysis of Method 2—Determination of Stack Gas Velocity
and Volumetric Flow Rate (Type S pitot tube)—was made and reported in
reference 2. The results of that analysis is used here without repeating
the analysis. A coefficient of variation of
CV{V } = 2.07%
s
was derived for the stack gas velocity determination.
Using CV{V } and CV{V } the coefficient of variation for the ratio
ii s
V /V becomes
n s
cv{v /v
n s
n
(17)
or
CV{V /V } = 2.91%
n s
98
-------
The average value of V /V = I = 100, therefore
IT S
CV{I} = 2.91%.
The component of variability of m due to nonisokinetic sampling as
given by equation 12 becomes
o2{m } = 0.0034 x m2 ,
n t
yielding a standard deviation of
aT{m } = 0.0582 x m .
In t
(18)
4.1.1.6 Summary of Errors in Measured Particulate Mass, m_. Using the
" ••-'-•— " -Q
variance components derived in subsections 4.1.1.1 through 4.1.1.5, the
variance of m is
n
o2{mn} = 0.0066 m2. + 13 mg2.
(19)
The variance, standard deviation, and coefficient of variation for
different values of m are given in the following table
m
t
100 mg
200 mg
400 mg
600 mg
Volume of Gas
a {m }
n
79 mg2
277 mg2
1069 mg2
2389 mg2
Measured by
a{m }
n
8.9 mg
16 . 6 mg
32.7 mg
48.9 mg
the Dry Gas Meter
CV{m }
n
8.9%
8.3%
8.2%
8.1%
and Corrected to Standard
Conditions, V , _, .
The sample volume measured by the dry gas meter and corrected to
standard conditions is given by (equation 5-1 is Appendix A).
V
^
bar
+ AH/13.6
/ l^
m(std)
m
m
(20)
99
-------
(See definitions of terms in Section 6.1 of Appendix A.)
Variability in V is a combination of variability of the dry gas meter,
inaccuracy of the calibration standard (i.e., wet test meter), and sampling
train leaks, including those leaks undetected because of pump valve float.
Assuming coefficients of variation of 1 percent, 0.4 percent, and 0.5 per-
cent for the dry gas meter, calibration standard, and sampling train leaks,
respectively, results in a CV{V } = 1.2 percent.
m
CV{P } = 0.3 percent and CV{T } = 0.5 percent are assumed from pre-
Dei IT ID.
vious documents of this series (ref. 2).
Variability in AH is believed to be primarily due to reading error
(inclined manometer) and calibration error in determining AH@ (AH@ is the
3
pressure drop across the orifice meter resulting in a 354 cm /sec (0.75
ft /min) flow rate at standard temperature and pressure). The pressure
drop across the orifice, i.e., AH, relatively constant with no fluctua-
tions and hence, easily read on an inclined manometer. Also, random
reading error is averaged out because AH as used in the equation is an
average of at least 12 readings. Therefore, the variability of the average
pressure drop should be adequately characterized by a coefficient of varia-
tion of 1 percent, i.e., CV{AH} = 1 percent. These assumptions are sum-
marized in table 3. The estimated variance of V , ,, , as given by equa-
tion 20, is
CT^V / ^ ~ 1-004 a2{V } + 4.02 a2{P, } + 0.0129 a2{T } + 0.0217 a2{AH}. (21)
m(std) m bar m
Using the estimated means and standard deviations of table 4 (in English
2
units) the estimated variance of V / n, is a {V / ,N} - 0.644 and the
m(std) m(std)
standard deviation a{V , ,.} - 0.8 ft^. The estimated mean is
m(std)
Vstd) = 1'7 ^ (6° ft3)
resulting in a coefficient of variation of 1.3 percent (= 0.8 x 100/60).
The most important variable in the determination of V , ,, under the stated
m(std)
assumptions is clearly V .
100
-------
Table 3. Means and variabilities of parameters in
determining the sample gas volume
Parameter Mean value
V V = 1.70 m3
mm
(60 ft: )
P,_ P, = 760 mm Hg
bar bar
(29.92 in. Hg)
T T = 294°K
m m
(530°R)
AH AH = 100.6 mm H20
(4 in. H20)
Coefficient of Standard
variation deviation
CV{V } = 1.2% a{V } = 0.02 m3
m m
(0.72 ft
CV{P, } = 0.3% a{P, } = 2.3 mm
bar bar
(0.09 in.
CV{T } = 0.5% a{T } = 1.5°K
m m
(2.65 °R)
CV{AH} =1.0% a{AH} = 1.0 mm H2
(0.04 in.
)
Hg
Hg)
0
H20)
4.1.3 Volumetric Flow Rate, Qs.
A functional analysis of the method used to measure volumetric flow
rate was made in the quality assurance guidelines document for Method 2
(ref. 1). Volumetric flow rate is given by the product of the stack gas
velocity, V , and the cross sectional area of the stack, A . The results
s s
of that analysis will be used here without repeating the analysis itself.
A coefficient of variation of 2.33 percent was derived for the volumetric
flow rate.
That is,
CV{Qs> = CV{Vs-As} = 2.33%.
4.1.4 Precision of Particulate Concentration Determinations, C_.
——^ ..._ ., ~ ._ „g
3
The concentration of particulate matter in the stack gas, g/m , on a
dry basis corrected to standard conditions, is given by
Cs= (10-3g/mg) (mn/Vm(Btd)).
Using previously derived results, the estimate precision of C is given by
S
a2{C } = a2{m } + a2{V , ,.
s n m(std)
(22)
101
-------
2 2
where a {m } and a {V .. ,N} are given by equations (19) and (21), respec-
n m(.std;
tively. Substituting these results into (22) and using values of 100, 200,
and 300 mg for m yields:
m
t
100 mg.
200 mg.
400 mg.
a{C }
s
9.0 mg.
L6.8 mg.
33.2 mg.
CV{C }
s
9.0%
8.4%
8.3%
Variability in m is the dominating source of variability in the
determination of C .
s
4.1.5 Precision of Particulate Mass Emission Rate, PMR..
The particulate mass emission rate is calculated by
and
PMR (g/hr) = Qo (m3/hr) x c (g/m3)
S S
CV2 {PMR} = CV2 {Qg} + CV2 {Cg}. (23)
CV{Q } was derived in the Quality Assurance Document for Method 2
s
(ref. 1) as 2.33 percent. Inserting this value and the estimated values
of CV{C } from the preceding subsection into equation (23), results in the
s - 3
following CVs of the PMR measurements for V , ,, = 1.7 m .
mt CV {PMR}
100 mg,. 9.3%
200 mg.. 8.7%
400 mg,, 8.6%
The concentration factor, C , is the more important of the two variables
S
in equation (23), accounting for about 80 percent of the total variability
in PMR.
4.1.6 Summary of Functional Analysis.
In summary, the most important variables in the determination of PMR
102
-------
are identified in the diagram of figure 9. The coefficient of variation
for each of the important variables is given and a flow diagram indicat-
ing how each variable influences the pollutant mass emission rate is given
in the figure. Table 4 summarizes the variance analysis calculations.
Results for three levels of collected particulate mass are given for con-
centration and for particulate mass emission rate.
4.1.7 Bias Analysis
Two bias terms were identified in the functional analysis of section
4.1.1. They are the error in the collected mass m due to (1) filter sur-
face alkalinity, and (2) condensation in the probe prior to the filter
by oxidation of SO™ in the sample gas. Combining the estimated biases
of these two effects from sections 4.1.1.3 and 4.1.1.4, respectively,
yields the following overall bias in m values.
t {mn> = t {PH} + t {C} (24)
= 2.4 mg + 4 mg = 6.4 mg.
For m = 100 mg, the bias represents an overall potential relative
bias of +6.4 percent. However, for m = 400 mg, the relative bias is only
+1.8 percent (= 100 x 6.4 mg/400 mg). Thus, the bias could be significant
for the small values of true particulate mass, m .
4.2 ACTION OPTIONS
Suppose it has been determined as a result of the functional analysis
and/or the reported data from the checking and auditing schemes, that
the data quality is not consistent with suggested standards or with the
user requirements. Poor data quality may result from (1) a lack of
adherence to the control procedures given in section II—Operations
Manual, or (2) the need for an improved method or instrumentation for
taking the measurements. It is assumed in this section that (2) applies;
that is, the data quality needs to be improved beyond that attainable by
103
-------
cv{vm} =
1 .2%
1
r
CV{V , ,.}
m(std)
1 .3%
1
r
cv{cs} =
9.0%
1.4%
1
cv{vs} =
2%
f 1
cv{vn}=
2%
r
cv{vn/vs} =
2.9%
\
r
CV{m }
n
8.9%
*-
^
CV{QS} =
2.3%
^
r
CV{PMR} =
9.3%
*CV{CS) determined for m = 100 mg.
Figure 9. Relationship of the most important intermediate measurements
to the determination of PMR.
104
-------
4J
s
60
g
O
o
-d-
g
o
g°
0
o
rH
O
4-1
CO
•rl
4-1
cfl
4-1
cn
cu
rH
ni
•H
rl
n)
*
O
vO OO
0 O
rH
CM
,
cr
Ed
a
S
^
CO
e
^
4-1
CO
N^X
S3
>
6-S
en
en
rH
^-^
>"
C_>
*
O
ON en
CO rH
*£>
C
ca o
S
en
J3_
C
B-S
^*
O
CM
G
•-I
»
cr
W
^-^
>
U
*
ON
CM
O
rH 0
C
05 t>
S
CO
^
•»-^,
pj
>
B^S
rH
CM
1 —
rH
,
cr
W
/~-N
£5
u
&-S
r^ (N
• •
CSJ OO
O fO
O
*^"
gsO
^o rn
\& oo
0 rH
O
CM
O> ON
• •
O OO OO
o
^
oo
»
cr
Ed
/^"•v
C 5^2
S U
'fcfl
g
^^x
p^
^
en
0 en
<- g
60 ^~
60
oo s~s
i oo en
0 1
rH O OO
rH
r^
X
en ^r
« ON
en
CM rH
en
g en
-- g
60 ^~
60
00 00
1 1 SN!
0 O •*
, | j j .
00
k> ^J
r~~ oo
• ON
rH O
60 60
oo oo o
1 1
O O ON
rH rH
X X
en rH
oo ir>
LA O
CM
CM
.
cr
cu w
CO
03 ^
o rj ^-2
rt D v~'
rl CU >
OS o
c«
rl
cu
C!
o
c
•H CO
C U
o
a
B-S
vO
•
oo
B-S
r*^
oo
rn
•
ON
e^
CM
.
cr
Ed
^-^
cfl & ^
cu >
S o
0
prf
^
rM
c
o
C
cu
o
a
CO
4-1
rH
3
CO
CU
105
-------
following the operational procedures given for the reference method.
The selection of possible actions for improving the data quality can
best be made by those familiar with the measurement process and by the
judicious use of the functional analysis as described in section 4.1. For
each action, the variance analysis can be performed to estimate the vari-
ance, standard deviation, and coefficient of variation of the pertinent
measurement(s). In some cases it is difficult to estimate the reduction in
specific variances, which are required to estimate the precisions of the
pertinent measurements. In such cases, an experimental study should be
made of the more promising actions based on preliminary estimates of
precision/bias and costs of implementation of each action. This preliminary
analysis would follow the methods suggested herein.
AO: Reference Method
Al: Compute PMR by sample concentration and ratio of area methods
and average (cost of $200/20 stacks)(ref. 23).
A2: Crew training (cost of $1000/20 stacks)
A3: Use modified ORSAT (for incinerator)(cost of $200/20 stacks)
A4: Use programmable rainicalculator in lieu of nomographs (cost
of $350/20 stacks).
The costs given for each action are additional costs above that of the
reference method. The assumptions made concerning the reduction in the
variances (or improved precisions), are given in the following paragraphs
for each action.
1. Compute PMR by Sample Concentration and by Ratio of Area Methods
and Average, Al. It is assumed that the error associated with anisokinetic
sampling is reduced to one-half by this action as compared to the value
calculated by the reference method i.e., calculating by sample concentra-
tion only (ref. 23). This assumption results in a reduction in the rela-
tive standard deviation (CV) of PMR from 9.3 percent to 5.0 percent for
m = 100 mg.
A nominal cost of $200 per 20 stacks is used as the cost of imple-
menting this action. Calculations for the particulate mass emission rate
by the method of sample concentration, PMR , and for the percent of iso-
kinetic sampling, I = V /V , are required by the reference method. The
IT o
106
-------
product of PMR and I yields the particulate mass emission rate by the
ratio of areas method, PMR . Hence, very little additional effort is
3
required.
2. Crew Training, A2. It is assumed that the coefficient of varia-
tion of PMR is reduced by 25 percent or,
CV{PMR|A2l = 0.75 CV{PMR|AO}.
The notation CV{PMRJA2} denotes the estimated CV of PMR given that
action A2 is implemented; the vertical line being read given that, and
action following the vertical line denotes the action implemented. This
results in a straightforward computation of the CVs,
mt CV {PMR|AO} CV {PMR|A2}
100 mg. 9.3% 7.0%
In estimating a cost for implementing this action it was assumed that one
crew member sent to a source sampling school such as that conducted by the
EPA for 1 week out of a year would constitute special crew training. The
cost of the school plus subsistence for a week was estimated to be about
1,000 dollars. This is in excess of salary and benefits. The cost was
prorated over 20 source tests which was taken as a reasonable number of
tests per team per year.
3. Use Modified ORSAT (for incinerator sources), A3. It is assumed
that the standard deviation of the measurement of %C02 is reduced from 0.4
percent to 0 . 2 percent (absolute). For small values of %C00 the true
2t
mean percent CO.,, the effect of the error in the determination of %C02
on the correction or adjustment in the particulate matter concentration
is a dominating factor in the overall error in the determination of PMR
for low %C00 levels. Denote C as the adjusted value of C , where the
Z S3. S
relationship is
csa = c -r t (26)
sa s %co2
107
-------
or the relationship in the coefficients of variation is
cv2 {<: } = cv2 {c } + cv2 {%co0}/r,
sa s L
(27)
where r is the number of replicates used to estimate the mean, %CO«,
assuming a {%C021 = 0.4, then CV{%cio2} = 0.4/(/r x %CC>2) for the standard
ORSAT and CV{%"C02> = Q.2/(/r x 7X0^ for the modified ORSAT. The following
measures of precision are derived assuming r is taken as suggested in
the Quality Assurance Document for Method 3 (ref. 2) and m = 100 mg.
%co2
2%
6%
10%
Standard ORSAT
r
12
3
3
CV {Csa}
10.7%
9.8%
9.3%
CV {PMR}
11.0%
10.1%
9.6%
Modified ORSAT
r
6
3
3
CV {C }
sa
9.9%
9.2%
9.1%
CV {PMR}
10.2%
9.5%
9.4%
An ORSAT with 0.1 ml dimensions as depicted in figure 2 of the Opera-
tions Manual of reference 2 costs very little more than the so called
standard ORSAT. Over a period of a year the difference in cost should be
negligible. A cost of 200 dollars per 20 source tests is assumed for
this example.
4. Minicalculator in lieu of nomographs, A4. It is assumed that the
coefficient of variation of V /V is reduced in accordance with the follow-
n s
ing equation:
CV {V /V |A4} = 0.80 CV {V /V |AO} .
n s' n s'
An increase in precision and accuracy should be realized since
exact values of C , AH@, and M, would be used to determine isokinetic
P d
conditions rather than the mean values of their expected range as are
now built into the nomograph. Also, routine calculation errors (i.e.,
108
-------
error in setting and reading the nomograph) should be greatly reduced.
The above assumed relationship must be considered in the analysis of the
variation of m , C , and PMR as described in section 4.1.1. The corre-
n' s'
spending coefficients of variation of PMR for the reference method and
for action A4 are estimated as follows:
CV {PMRJAQ} CV {PMR|A4}
100 mg. 9.3% 8.6%
200 mg. 8.7% 8.0%
400 mg. 8.6% 7.9%
A programmable mini-calculator can be purchased for about 300 dollars.
Cost of programming should be more than recovered in sampling time saved
as a result of using the calculator. The increased cost is taken as 350
dollars per 20 source tests.
4.2.1 Comparison of Actions
The added cost per 20 stacks is plotted as a function of the precision
of the estimated PMRs as measured by its coefficient of variation, CV{PMR}
in figure 10. Examination of the plotted results enables one to quickly
identify the action or combination of actions which will yield results of
some desired precision. In practice it may not be reasonable to insist
that the data be of a specified precision, but that the cost of report-
ing poor quality data is a rapidly increasing function of the data quality
for data of variability exceeding a specified value. The selection of
the best action option then becomes a trade-off between the overall cost
and the expected precision to be achieved by its implementation. The
cost of implementing an action plus the cost of reporting poor quality
data are added to obtain an overall cost for the action. An assumed
function of the cost of reporting poor quality data is shown as the solid
curve in figure 10 as an illustration only. Its exact shape and location
on the graph would have to be determined from actual cost data. Table 5
109
-------
CN1
<
Pu
^^
>
to
o>
o
)-i
o
CO
O
4-1
cfl
S-i
d)
fi
•H
a
c
•H
O
14-1
co
C
O
•H
c
o
01
4-1
CJ
a;
rH
QJ
CO
C II
o
•H rsl
CO O
•H U
O S-5
0)
S-i T3
cx c
05
CO
> oo
CO O
o o
a iH
13 II
Q)
0)
M
3
O
O
CM
O
O
O
O
O
OO
O
o
o
o
o
o
CN
03
PSPPV
110
-------
summarizes the results of the analyses of this section which are used in
plotting the data in figure 10. A comparison of the actions indicates
that action Al would he most cost-effective in improving the precision of
PMR. Both actions Al and A3 cost the same to implement, however, the
larger impr -vement in precision resulting from Al shows (for this hypothe-
tical case in figure 10) a lower overall cost since A3 shows a $400 for
reporting poor quality data.
Table 5. Comparison of action options based on
CV {PMR} for C adjusted to 12-percent
CO,., (incinerator sources)
Action option
AO (a{%C02}=0.4)
Al
A2
A3 (a{%C02>=0.2)
A4
Cost/
20 stacks
$ 0
$ 200
$1000
$ 200
$ 350
Estimated precision
(CV {PMR} in %)
m = 100 mg
11.0
7.6
9.1
9.5
10.3
4.3 PROCEDURES FOR PERFORMING A QUALITY AUDIT
"Quality audit" as used here implies a comprehensive system of planned
and periodic audits to verify compliance with all aspects of the quality
assurance program. Results from the quality audit provide an independent
assessment of data quality, "independent" means that the individuals
performing, and as much as possible of the equipment used in the audit, are
different from the regular field crew and equipment. From these data both
bias and precision estimates can be made for the measurement process.
The auditor, i.e., the individual performing the audit, should have
extensive background experience in source sampling, specifically with the
characterization technique that he is auditing. He should be able to
111
-------
establish and maintain good rapport with field crews.
The functions of the auditor are summarized in the following list:
1. Observe procedures and techniques of the field team during on-
site measurements.
2. Perform certain independent checks, and from previous training
and experience, estimate the validity of the field crew's measurements.
3. Check/verify applicable records of equipment calibration checks
and quality control charts in the field team's home laboratory.
4. Perform calculations using data obtained from the audit.
5. Compare the audit value with the field team's test value.
6. Inform the field team of the comparison results specifying any
area(s) that need special attention or improvement.
7. File the records and forward the comparison results with appro-
priate comments to the manager.
4.3.1 Frequency^ of Audit
The optimum frequency of audit is a function of certain costs and
desired level of confidence in the data quality assessment. A methodology
for determining the optimum frequency using relevant costs is presented in
subsection 4.4.4. For the costs assumed in this document, an optimum audit
level of n = 5 for a lot siz;e of N = 20 results as shown in Figure 15,
page 131. However, costs will vary among field teams and types of field
tests. Therefore, the most cost effective auditing level will have to be
derived using local cost data according to the procedure given in subsec-
tion 4.4.4 and in the final report on this contract.
4.3.2 Collecting Audit Information
While at the sampling site, the auditor should observe the field
team's overall performance of the field test. Table 6 is a sample check-
list of the operations to observe. Each item on the list should be checked
yes or no according to whether it was performed as recommended in the
operations manual and, if applicable, the result was within specified
limits. Those items checked no should be explained under comments. No
checklist can cover all situations; the auditor must utilize his good
112
-------
Table 6. Particulate emission determination checklist to be used by auditor
YES
NO
OPERATION
EQUIPMENT PREPARATION AND CHECK
1. Sampling train assembled and leak checked.
2. , Probe and filter box heaters checked and set for proper
temparatures.
3. Stack gas temperature measuring system assembled and
checked for proper operation by comparing to a mercury
in glass thermometer.
4. Stack gas velocity measuring system assembled and checked
for proper operation.
5. Orsat analyzer assembled and checked.
PRELIMINARY MEASUREMENTS
6. Selection of traverse points according to Method 1.
7. Moisture content by Method 4, or equivalent.
8. Molecular weight by Method 3, or equivalent.
9. Measurement of stack dimensions.
10. Mark probe for sampling at traverse points.
SAMPLE COLLECTION
11. Equal sampling time at each traverse point.
12. Probe temperature satisfactory throughout the test.
13. Filter box temperature 120 + 14°C (248 i 25°F)through the test.
14. Sample gas temperature at last impinger « 20°C (68°F) through-
out the test.
15. Isokinetic sampling checked and adjusted if necessary
at least every 5 minutes.
16. Leak check of sampl-ing train at end of test.
SAMPLE RECOVERY
17. Satisfactory handling and movement of probe and filter
to sample recovery area.
18. Recovery area satisfactory (i.e., space, cleanliness,
etc.)
113
-------
Table 6. Particulate emission determination checklist to be used by auditor
(continued)
YES
NO
OPERATION
19. Sample recovery procedure adequate.
20. Proper labeling of sample containers.
21. Determination of moisture content procedure adequate,
ANALYSIS
22. Proper equilibration of (1) filter, (2) probe wash
residue, and (3) acetone blank residue.
23. Correct collected particulates for acetone blank.
24. Analytical balance checked before weighings.
DOCUMENTATION
25. All information recorded on data sheet as obtained.
26. All unusual conditions recorded.
COMMENTS
114
-------
AUDIT MEASUREMENTS
STACK CROSS-SECTIONAL AREA, AS
DRY GAS METER CALIBRATION, y *
a
ORIFICE METER CALIBRATION, AH@a
SAMPLING NOZZLE DIAMETER, D
na-
SAMPLING NOZZLE CROSS-SECTIONAL AREA,
PITOT TUBE CALIBRATION COEFFICIENT, C
m
Dimensionless
mm ^0
mm
m
Dimensionless
1. V = V x v =
ma m ra
2' Vma(std) = °'3855 Vma
CALCULATIONS
3
m
> , AH
bar 13.6
m , at standard
conditions.
3. Csa - 0.001 (mn/Vma(s
4.
= 2.75 x 10
'sa
m /hr, at standard conditions on a dry basis.
5.
PMRa = Csa
9/hr.
6.
4'323 Ts Vma(std)
Ana Ps e
percent.
PMR - PMR
7. d =
x 100 =
PMR.
, percent.
Figure 11. Sample form for recording audit data
115
-------
judgment and include other checks as deemed desirable for a specific
situation.
A completed checklist with all yes checks implies that in the opinion
of the auditor the measurement was made in such a manner that large biases
resulting from poor technique are not likely to be present.
In addition to the above observations, the auditor should independently
determine the stack dimensions. This should be carried out with the reali-
zation that the measurement is to be used as an estimate of the average
dimension of the stack. Therefore, for example, a stack which could be out
of round should have its diameter measured from as many sampling ports as
possible and the average diameter used as the stack diameter in subsequent
calculations. Record the cross-sectional area, A , on the form of figure 11•
S3.
The auditor should obtain from the field team a complete set of test
data, i.e., the data form of figure 4 filled in through the section on
recorded test data, for all sample runs.
In the field team's home laboratory the auditor should verify, by
checking calibration records and field data sheets from previous field
tests, that the performance criteria as given in table 2 of section III
have been satisfied over the period since the last audit. Also, using
his own calibrated standards, perform the following calibration checks.
1. Using his own calibrated wet test meter, or equivalent, calibrate
the dry gas meter and the orifice meter as directed in subsection 2.2.3 of
section II. Record these audit values as y and AH@ in the form in
a
figure 11.
2. Determine the sampling nozzle diameter, D , according to the
na
procedure of subsection 2.2.1. Calculate the cross-sectional area, A ,
in square meters by
A = 0.25 x |i x (D 2 1 m
na "' " 7 v na' ._6 2
10 mm
A = 7.85 x 10 7 (D )2 = m2,
na na ——• •——
Record D and A on t'ne audit data sheet of figure 11.
na na
116
-------
3. Using a calibrated pitot tube, calibrate the field pitot tube
according to the procedure given in subsection 2.1.2, page 11,of the quality
assurance document of this series for Method 2 (ref. 1) or the EPA revised
Method 2 as reproduced in reference 4. The field pitobe assembly must be
configured as shown in figure 5-1 of Appendix A. The average coefficient,
C , applicable to the AP range measured in the field test is determined
Pa
from this calibration data. Record C on the form in figure 11.
^a
4.3.3 Treatment of Audit Data
Using the above audit data where applicable and the raw field data,
perform the calculations indicated in figure 11. All variables are in the
same units as used in subsection 2.5.3 of the operations manual. In
figure 11 audit measurements and/or parameters computed using audit measure-
ments are subscripted with an a. Parameters such as (/AP) , (T ) , M ,
avg s avg s
and B should be calculated by the auditor from the field crew's original
ws J
data. All calculations are recorded on the audit data sheet of figure 11.
A comprehensive discussion of estimating the true particulate mass
emission rate by the average of PMR and PMR is contained in reference 23.
C 3.
Also, the importance of sampling isokinetically for particles down to about
5 ym in diameter is illustrated in figure 7.8, page 75, of reference 16.
Unless it is known that the particle size distribution is below 5 ym for
the stack gas being sampled, the above averaging technique appears to be
an improvement over using just the value of PMR as the true emission rate.
The auditor's report of a specific field test to the manager should
include copies of (1) a completed data sheet from the field team (fig. 4),
(2) a completed checklist with comments (table 6), (3) a completed audit
data sheet with calculations (fig. 11) and (4) a summary of the field
team's strong/weak points with an overall numerical rating and recommended
actions as discussed in the following subsection.
4.3.4 Overall Evaluation of Field Team Performance
In a summary-type statement, the field team should be evaluated on
its overall performance. Using the checklist filled out in the field
117
-------
(table 6) in conjunction with the results of the comparison of audit and
field team values of PMR and the circumstances under which the test was
performed, the field team could be rated on a scale of 1 to 5 as follows:
5 - Excellent
4 - Above average
3 - Avera.ge
2 - Acceptable, but below average
1 - Unacceptable performance.
Justification for the rating in the form of a list of the team's
strong/weak points with recommendations for improving weak points should
be included in the auditor's report.
4.4 DATA QUALITY ASSESSMENT
Two aspects of data qua.lity assessment are considered in this section.
The first considers a means of estimating the precision and accuracy of
the reported data, e.g., reporting the bias, if any, and the standard
deviation associated with the measurements. The second consideration is
that of testing the data qua.lity against given standards using sampling by
variables. For example, lower and upper limits, L and U, may be selected
to include a large percentage of the measurements and outside of which it
is desired to control the percentage of measurements to, say, less than
10 percent. If the data quality is not consistent with these limits, L
and U, then action is taken to correct the possible deficiency before
future field tests are performed and to correct the previous data when
possible.
The determination of the audit level is indicated by using estimated
costs associated with falsely inferring that good (poor) quality data are
of poor (good) quality and with auditing n stacks. In addition, prior
information concerning data quality is assumed in order to determine an
expected or average cost resulting from the statistical sampling plan.
The cost estimates provided herein are assumed for the purpose of illus-
trating the methodology. It. is emphasized that managers need to supply
their own costs in making such analyses.
118
-------
4.4.1 Estimating the Precision/Accuracy of the Reported Data
Methods for estimating the precision (standard deviation) and
bias of the particulate emission rate (PMR) measurements were
given in section 4.1. This section will indicate how the audit data
collected in accordance with the procedure described in section 4.3 will
be utilized to estimate the precision and accuracy of the measure of
interest. Similar techniques can also be used by a specific firm or team
to assess their own measurements. However, in this case no bias data among
firms can be obtained. The audit data collected as a result of following
the procedures in the previous section are the measured and audited values
of PMR and the difference.
PMR. - PMR .
d. = - J - - ^ x 100 (28)
.
3
PMR
where PMR. = Field measured value of particulate mass emission rate,
average of three replicates, and
PMR . = Audited value of particulate mass emission rate, average
of three replicates.
Let the mean and standard deviation of the differences d., j = 1, ...n
field tests be denoted by 3 and s,, respectively. Thus,
d = d-/n '
r n - 11/2
and s = E (d- - d)2/(n - 1) (30)
d Lj-1 J -1
4.4.2 Statistical Tests
The mean d is an estimate of the relative bias in the measurements
119
-------
(i.e., relative to the audited value). Assuming the audited data to be
unbiased, the existence of a bias in the field data can be checked by the
appropriate t-test, i.e.,
(31)
See reference 29 for a discussion of the t-test. If t is significantly
large in absolute values, i.e., greater than the tabulated value of t with
n - 1 degrees of freedom, which is exceeded by chance only 5 percent of the
time, then the bias is considered to be real and some check should be made
for a possible cause of the bias. If t is not significantly large, then
the bias should be considered zero or negligible.
The standard deviation, s,, is a function of both the standard devia-
d
tion of the field measurements and of the audit measurements. Assuming the
audited measurements are obtained with much greater precision than the field
measurements, then the calculated Sj is an estimate of the standard deviation
of the field measurements. Furthermore, since s, is in percent ? it is an es-
timate of the coefficient of variation, CV{PMR}. Table 7 in the following
subsection contains an example calculation of d and s , , starting with the
differences for a sample size of n = 5.
The calculated standard deviation can then be utilized to check the
reasonableness of the assumption made in section 4.1 concerning CV{PMR} =
9.3 percent for m = 100 mg, for example. (Remember that CV{PMR} is equal
to s ). The calculated standard deviation, s , , may be directly checked
against the assumed value, a,, by using the statistical test procedure
(32)
2
where x /f is the value of a random variable having the chi-square distri-
2
bution with f = n - 1 degrees of freedom. If x /f i-s larger than the
120
-------
tabulated value exceeded only 5 percent of the time, then it would be
concluded that the test procedure is yielding results with more variability
than is acceptable due to some assignable cause of large variation.
The measured values should be reported along with the estimated bias,
d, standard deviation, s .,, the number of audits, n, and the total number
Q
of field tests, N, sample (n <_ N) . Estimates, i.e., s, and d, which are
significantly different from the assumed population parameters should be
identified on the data sheet. For example, based on the data of table 7,
if the field team reported a value of PMR =22.9 g/hr for one of the N
field tests not audited, then that measurement would be reported as
1. Measured value, PMR =22.9 g/hr
2. Calculated bias, t{PMR} = d x PMR = 1.51 g/hr
3. Calculated standard deviation, &{PMR} = s x PMR =1.9 g/hr
4. Auditing level, n = 5, N = 20.
From the above data, users of the data can calculate confidence limits
appropriate to what the data are to be used for.
2
The t-test and x -test described above, and in further detail in the
final report on this contract, are used to check on the biases and standard
deviations separately. In order to check on the overall data quality as
measured by the percent of measurement deviations outside prescribed limits,
it is necessary to use the approach described in subsection 4.4.3 below.
4.4.3 Sampling by Variables
Because the lot size (i.e., the number of field tests performed by
a team or laboratory during a particular period, normally a calendar quarter)
is small, N = 20, and consequently, the sample size is small on the order
of n = 3 to 7, it is important to consider a sampling by variables approach
to assess the data quality with respect to prescribed limits. That is,
it is desired to make as much use of the data as possible. In the variables
approach, the means and standard deviations of the sample of n audits are
used in making a decision concerning the data quality.
Some background concerning the assumptions and the methodology is
repeated below for convenience. However, one is referred to one of a num-
ber of publications having information on sampling by variables; e.g.,
121
-------
see references 30-33. The discussion below will be given in regard to the
specific problem herein which has some unique features as compared with
the usual variable sampling plans.
In the following discussion, only the audited value PMR is discussed.
This same procedure can be a.pplied to audits of sampling train components
e.g., the pitot tube calibration, dry gas meter, and orifice meter.
The difference between the team-measured and audited value of PMR
is designated as d., and the mean difference over n audits by d, that is,
(33)
Theoretically, PMR. and PMR . should be measures of the same particulate
mass emission rate, and their difference should have a mean of zero on the
average. In addition, their differences should have a standard deviation
approximately equal to that associated with measurements of PMR. separately.
Assuming three standard deviation limits (using the assumed CV = 9.3
percent as derived in the variance analysis of subsection 4.1), the values
-3(9.3%) « -28 percent and 3(9.3%).« 28 percent define lower and upper
limits, L and U, respectively, outside of which it is desired to control
the proportion of differences, d.. Following the method given in reference
30, a procedure for applying the variables sampling plan is described below.
Figures 12 and 13 illustrate examples of satisfactory and unsatisfactory
data quality with respect to the prescribed limits L and U.
The variables sampling plan requires the sample mean difference, d;
the standard deviation of these differences, s,; and a constant, k, which
is determined by the value of p, the proportion of the differences outside
the limits of L and U. For example, if it is desired to control at 0.10
the probability of not detecting lots with data quality p equal to 0.10
(or 10% of the individual differences outside L and U) and if the sample
size n = 5, then the value of k can be obtained from Table II of reference
122
-------
/ 1
/ 1
X i
.s \
^"^ \
\
\
\
\^ P2
^--A-
0.10
Figure 12. Example illustrating p < 0.10 and satisfactory data quality.
p (percent of measured
differences outside
limits L and U) > 0.10
U
Figure 13. Example illustrating p > 0.10 and unsatisfactory data
quality.
123
-------
Table 7. Computation of mean difference, d, and
standard deviation of differences, s,
d
General formulas Specific example
PMR. - PMR .
d. = J SI x 100
J
PMR . . .
aj Data (percent)
dl dl
4
d4 d4
2
, £d , 33.0 49.5
J J
Ed.
d = —J- d = 6.6%
n
Ed 2 - (Ed )2/n
S2 = _J_^ _J 2
d (n - 1) d
12.0
-6.0
3.0
15.0
9.0
14.4
3.6
0.9
22.5
8.1
s , = Ws , s , = 8.3%
did d
Table 8. Sample plan constants, k for P{not detecting a lot
with proportion p outside limits L and U} <_ 0.1
Sample size n k(p = 0.2) k(p = 0.1)
3 3.039 4.258
5 1.976 2.742
7 1.721 2.334
10 1.595 2.112
12 1.550 2.045
124
-------
30. The values of d and s, are computed in the usual manner; see table 8
for formulas and a specific example. Given the above information, the test
procedure is applied and subsequent action is taken in accordance with the
following criteria:
1. If both of the following conditions are satisfied:
d - k sd _> L = - ^28 percent, (34)
d + k sd £ U = 28 percent (35)
the individual differences are considered to be consistent with the
deficiencies exist in the measurement process as carried out for that
particular lot (group) of field tests. These deficiencies should be iden-
tified and corrected before future field tests are performed. Data cor-
rections should be made when possible, i.e., if a quantitative basis is
determined for correction.
Table 8 contains a few selected values of n, p, and k for convenient
reference.
Using the values of d and s, in table 8, k = 2.742 for a sample size
n = 5, and p = 0.10, the test criteria can be checked; i.e.,
d - k sd = 6.6 - 2.742 (8.3) = -16.2 > L = -28 percent (36)
and
d + k sd = 6.6 + 2.742(8.3) = 29.4 > U = 28 percent. (37)
Therefore, the upper limit is not satisfied and the lot of N = 20 measure-
ments is not consistent with the prescribed quality limits. The plan is
designed to protect against not detecting lots with 10 percent or more
defects (deviations falling outside the designated limits L and U) with
a risk of 0.10; that is , on the average, 90 percent of the lots with 10
percent or more defects will be detected by this sampling plan.
4.4.4 Cost Versus Audit Level
The determination of the audit level (sample size n) to be used in
assessing the data quality with reference to prescribed limits L and U
125
-------
can be ..made on a statistical basis by defining acceptable risks for type
I and type II errors, knowing or estimating the quality of the incoming
data, and specifying the described level of confidence in the reported
data, or on a cost bassis as described herein. In this section, cost data
associated with the audit procedure are estimated or assumed for the pur-
pose of illustrating a method of approach and identifying which costs
should be considered.
A model of the audit process, associated costs, and assumptions made
in the determination of the audit level is provided in figure 14. It is
assumed that a collection of source emissions tests for N stacks is to be
made by a particular firm, and that n measurements (n <_ N) are to be
audited at a cost, C = b + en, where b is a constant independent for n,
A
and c is the cost perstack measurement audited. In order to make a specific
determination of n, it is also necessary to make some assumptions about
the quality of the source emissions data from several firms. For example,
it is assumed in this analysis that 50 percent of the data lots are of
good quality, i.e., one half of the firms are adhering to good data quality
assurance practice, and that 50 percent of the data lots are of poor
quality. Based on the analysis in section 4.1, good quality data is
defined as that which is consistent with the estimated precision/bias
using the reference method. Thus, if the data quality limits L and U
are taken to be the lower and upper 3a limits corresponding to limits
used in a control chart, the quality of data provided by a firm adher-
ing to the recommended quality assurance procedures should at most con-
tain about 0.3 percent defective measurements (i.e., outside the limits
defined by L and U) . Herein., good quality data is defined as that con-
taining at most 10 percent defective measurements. The definition of
poor quality data is somewhat arbitrary, for this illustration it is
taken as 25 percent outside L and U.
In this audit procedure the data are declared to be of acceptable
quality if both of the following inequalities are satisfied
d + ks, < U (39)
d
126
-------
d - ks, > L, (40)
d
where d and s, are the mean and standard deviation of the data quality
d
characteristic (i.e., the difference of the field and audited measure-
ments) being checked, and not of desired quality if one or both inequali-
ties is violated, as described in section 4.3. The cost associated with
these actions are assumed to be as follows:
C. = Audit cost = b + en. b is assumed to be zero for this example
A
and c is taken as $600/measurement.
C i
P|G = Cost of falsely inferring that the data are of poor quality, P,
given that the data are of good quality, G. This cost is
assumed to be one-half the cost of collecting emissions data for
N = 20 stacks (i.e., 0.5 x $1000 x 20 = $10,000). This cost would
include that of searching for an assignable cause of the inferred
data deficiency when none exists, partial repetition of data
collection, decision resulting in the purchase of equipment to
reduce emission levels of specific pollutants, etc.
C_ p - Cost of falsely stating that the data are of good quality, G,
given that they are of poor quality. This cost is assumed to be
$15,000 (= 0.75 x $1,000 x 20). This cost is associated with
health effects, litigation, etc.
C i = Cost savings resulting from correct identification of poor quality
data. This cost is taken to be $7,500, i.e., equal to one-half of
C_,|_ or equal to 0.375 x $1,000 x 20, the total cost of data
collection.
These costs are given in figure 14. These cost data are then used in
conjunction with the a priori information concerning the data quality to
select an audit level n. Actually, the audit procedure requires the
selection of the limits L and U, n, and k, L and U are determined on the
basis of the analysis of section 4.1. The value of k is taken to be the
value associated with n in table 8 of section 4.4.3, i.e., the value
selected on a statistical basis to control the percentage of data outside
127
-------
Collection of Source Emission
Tests (Lots of Size N)
_ 50% of Lots
< 10% Defective
A
Acceptable
Quality
,
.
Not Acceptable
Quality
•4
50% of Lots
> 10% Defective
Audit n
Measurements
f
W
C = b+cn == $600
•*-
Audit n
Measurements
f
\
I
<—
Select Audit
Parameter n, k
_l t
1
Data Declared
to be of
Acceptable
Quality
Data Declared
not to be of
Acceptable
Quality
Report
Data
Data Declared
to be of
Acceptable
Quality
Institute Action to
Improve Data Quality
(Correct Data if
Possible)
Expected Cost of
Treating Poor
Quality Data as
Good Quality Data
C = $15,000
Expected Cost of
Falsely Inferring
Data are of Poor
Quality Cp Q =
$10,000
1
Expected Cost
Saving of Taking
Correct Action with
Respect to Poor
Quality Data
6p|p ^$7,500
Figure 14. Flow chart of the audit level selection process.
128
-------
the limits L and U. Thus, it is only necessary to vary n and determine
the corresponding expected total cost E(TC) using the following cost model.
E(TC) - - CA - 0.5 Pp|G Cp|G + 0.5 Pp|p Cplp - 0.5 PG|P CG|P (41)
where the costs are as previously defined. The probabilities are defined
in a'similar way to the corresponding costs.
P I = Probability that a lot of good quality data is falsely
P|Gr
inferred to be of poor quality due to the random variations
in the sample mean d and standard deviation, s,, in small
samples of size n.
P I = Probability that a lot of poor quality data is correctly
identified as being of poor quality.
P i_ = Probability that a lot of poor quality data is incorrectly
Gr|f
judged to be of good quality due to sampling variations of
d and s.
These three probabilities are conditional on the presumed lot quality
and are preceded by a factor of 0.5 in the total cost model to correspond
to the assumed percentage of good (poor) quality data lots.
In order to complete the determination of n, it is necessary to
calculate each of the conditional probabilities using the assumptions stated
for a series of values of n (and associated k which is given in table 9).
The computational procedure is given in the Final Report of this contract.
These calculations were made for the cases n = 3, 5, 7, and 10 and for two
degrees of control on the quality of the data that can be tolerated, i.e.,
p = 0.2 and p = 0.1, the portion outside the limits L and U for which it
is desired to accept the data as of good quality with probability less than
or equal to 0.10. These computed probabilities are then used in conjunction
with the costs associated with each condition, applying equation (41) to
obtain the average cost versus sample size n for the two cases p = 0.1 and
0.2. The curves obtained from these results are given in figure 15. It can
be seen from these curves that the minimum cost is obtained by using n - 5
129
-------
independent of p. However, it must be recognized that the costs used in the
example are for illustrative purposes and may vary from one region to
another, thus within the reasonable uncertainty of the estimated costs,
values of n between 3 and 7 would seem to be reasonable. The assumed costs
suggest that p = 0.2 which tends to permit data of podrer quality to be
accepted, is more cost effective.
130
-------
$8000
o
H
W
I
•u
en
o
cu
M
cfl
01
$6000 •
$4000 -
$2000 -
4567
Audit Level (11)
10
""
p = Proportion defective measurements in the "lot
P{Acc. lot with p} <_ 0.1
Figure 15. Average cost vs audit level (n).
131
-------
LIST OF REFERENCES
132
-------
LIST OF REFERENCES
1. EPA-650/14-74-005-a "Guidelines for Development of a Quality Assurance
Program—Determination of Stack Gas Velocity and Volumetric Flow Rate
(Type S Pitot Tube)" Environmental Protection Agency, Research Triangle
Park, North Carolina 27711, 1974.
2. EPA-650/14-74-005-b "Guidelines for Development of a Quality Assurance
Program—Gas Analysis for Carbon Dioxide, Oxygen, Excess Air, and Dry
Molecular Weight" Environmental Protection Agency, Research Triangle
Park, North'Carolina 27711, 1974.
3. EPA-650/14-74-005-C "Guidelines for Development of a Quality Assurance
Program—Determination of Moisture in Stack Gases," Environmental Pro-
tection Agency, Research Triangle Park, North Carolina 27711, 1974.
4. F. Smith ,et_ _aJL. "Guidelines for Development of Quality Assurance
Programs Applicable to Stationary Source Emission Stream Characteriza-
tion Techniques," Final Report, EPA Contract No, 68-02-1234, Research
Triangle Institute, Research Triangle Park, North Carolina 27709, 1976.
5. E, S. Kipecki, "Stainless Steels." Machine Design, 1970 Metals Reference
Issue 4, Vol. 42, Cleveland, Ohio: Penton Pbulishing Company, February
12, 1970, pp. 34-38.
6. R. M. Martin. Construction Details of Isokinetic Source Sampling Equip-
ment. Publ. No. APTD-0581. Air Pollution Control Office, Environmental
Protection Agency, Research Triangle Park, North Carolina 27711, 1971.
7. A. W. Gnyp et al. "An Experimental Investigation of the Effect of
Pitot Tube-Sampling Probe Configurations on the Magnitude of the S-Type
Pitot Tube Coefficient for Commercially Available Source Sampling
Probes," Technology Development and Appraisals Section, Air Resources
Branch, Ministry of the Environment, Province of Ontario, Toronto,
Canada; University of Windsor, February, 1975.
8. W. S. Smith. Stack Sampling News 1, No. 7, 1974.
9. J. J. Rom. Maintenance, Calibration, and Operation of Isokinetic Source
Sampling Equipment. Publ. No. APTD-0576. Office of Air Programs,
Environmental Protection Agency, Research Triangle Park, North Carolina,
1972.
10. "Occupational Safety and Health Standards: National Concensus Standards
and Established Federal Standards." Federal Register 36, No. 105,
May 29, 1971.
11. "Standards of Performance for New Sources." Federal Register 36,
No. 247, December 23, 1971.
133
-------
12. W. S. Smith and D. J. Grose. Stack Sampling Nomographs for Field
Estimations. Entropy Environmentalists, Inc., Research Triangle Park,
North Carolina, 1973.
13. W. S. Smith. Stack Sampling News 1, No. 1, 1973.
14. R. F. Yarnor. Industrial Source Sampling. Ann Arbor Science Publishers,
Inc., 1971.
15. W. J. Mitchell. Evaluation Report: Additional Studies on Obtaining
Replicate Particulate Samples from Stationary Sources. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711, 1973.
16. EPA-650/4-74-029 "Collaborative Study of Method for the Determination
of Particulate Matter Emissions from Stationary Sources (Portland
Cement Plant)," Environmental Protection Agency, Research Triangle
Park, North Carolina 27711, May 1974.
17. EPA-650/4-74-021 "Collaborative Study of Method for the Determination
of Particulate Matter Emissions from Stationary Sources (Fossil
Fuel-Fired Steam Generators)," Environmental Protection Agency,
Research Triangle Park, North Carolina 27711, June, 1974.
18. EPA-650/4-74-022 "Collaborative Study of Method for the Determination
of Particulate Matter Emissions from Stationary Sources (Municipal
Incinerators), Environmental Protection Agency, Research Triangle
Park, North Carolina 27711, July, 1974.
19. Administrative and Technical Aspects of Source Sampling for Particu-
lates. Publ. No. APTD-1289. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711, May, 1971.
20. E. L. Grant and R. S. Leavenworth. Statistical Quality Control.
4th ed., St. Louis: McGraw-Hill, 1972.
21. D. A. Simons. Practical Quality Control. Reading, Mass.: Addison-
Wesley Publishing Company, 1970, pp. 131-150.
22. EPA-600/9-76-005 "Quality Assurance Handbook for Air Pollution Mea-
surement Systems," Volume-1 Principles, Environmental Protection
Agency, Environmental Monitoring and Support Laboratory, Research
Triangle Park, North Carolina 27711, March, 1976.
23. P. K. Mueller ^t al. "Selection of Filter Media: An Annotated Out-
line." Presented at the 13th Conference on Methods in Air Pollu-
tion and Industrial Hygiene Studies, University of California,
Berkeley, California, October 30-31, 1972.
24. R. M. Burton _e_t a±. "Field Evaluation of the High-Volume Particle
Fractionating Cascade Impactor—A Technique for Respirable Sampling."
134
-------
Presented at the 65th Annual Meeting of the Air Pollution Control
Association, June 18-22, 1972.
25. W. C. L. Hemeon and A. W. Black. "Stack Dust Sampling: In Stack
Filter or EPA Train." Journal of the Air Pollution Control Associa-
tion 22, No. 7, 1972, pp. 516-518.
26. W. S. Smith. "A Matter of Definition." Stack Sampling News 1, No. 1,
Technomic Publishing Co., Inc. Westport, Conn., July, 1973.
27. B. D. Bloomfield. "Source Testing." Air Pollution, Volume II, 2nd
edition, A. C. Stern, ed. New York: Academic Press, 1968, Chap-
ter 28.
28. W. S. Smith et^ al. "A Method of Interpreting Stack Sampling Data."
Stack Sampling News 1, No. 2, 1974, pp. 8-17.
29. A. Hald. Statistical Theory with Engineering Applications. New York;
John Wiley and Sons, 1952.
30. D. B. Owen. "Variables Sampling Plans Based on the Normal Distribu-
tion." Technometrics 9, No. 3, August, 1967.
31. D. B. Owen. "Summary of Recent Work on Variables Acceptance Sampling
with Emphasis on Non-normality." Technometrics 11, 1969, pp. 631-637.
32. K. Takogi. "On Designing Unknown Sigma Sampling Plans Based on a
Wide Class on Non-normal Distributions." Technometrics 14, 1972.
33. C. Eisenhart, M. Hastay, and W. A. Wallis, eds. Techniques of Sta-
tistical Analysis. Statistical Research Group, Columbia Univ.
New York: McGraw-Hill, 1947.
135
-------
APPENDIX A mm 5-DEIWINATION OF PARTICULATE EMISSIONS
FROM STATIONARY SOURCES
136
-------
METHOD 5--DETERMINATION OF PARTICULATE
EMISSIONS FROM STATIONARY SOURCES
1. Principle and Applicability
1.1 Principle. Particulate matter is withdrawn isokinetically
from the source and collected on glass fiber filter maintained at
temperatures equal to or less than 120 ± 14° C (248 ± 25° F) or such
other temperature as specified by an applicable subpart of the
standards. The particulate mass is determined gravimetrically after
removal of uncombined water.
1.2 Applicability. This method is applicable for the determina-
tion of particulate emissions from stationary sources only when
specified by the test procedures for determining compliance with
new source performance standards.
2. Apparatus
2.1 Sampling train. A schematic of the sampling train used in
this method is shown in Figure 5-1. Commercial models of this train
are available. However, if one desires to build his own, complete
construction details are described in APTD-0581; for changes from the
APTD-0581 document and for allowable modifications to Figure 5-1, see
the following subsections.
The operating and maintenance procedures for the sampling train
are described in APTD-0576. Since correct usage is important in ob-
taining valid results, all users should read the APTD-0576 document
and adopt the operating and maintenance procedures outlined in it,
unless otherwise specified herein.
137
-------
_c
'to
en
C
"5.
£
22
_co
u
to
O-
LO
OJ
01
138
-------
2.1.1 Probe nozzle—Stainless steel (316) with sharp, tapered
o
leading edge. The angle of taper shall be < 30 and the taper shall
be on the outside to preserve a constant internal diameter. The probe
nozzle shall be of the button-hook or elbow design, unless otherwise
specified by the Administrator. The wall thickness of the nozzle
shall be less than or equal to that of a 20 gauge tubing, i.e., 0.165 on
(0.065 in.) and the distance from the tip of the nozzle to the first
bend or point of disturbance shall be at least two times the outside
nozzle diameter. The nozzle shall be constructed from seamless stain-
less steel tubing. Other configurations and construction material may
be used with approval from the Administrator.
A range of sizes suitable for isokinetic sampling should be
available, e.g., 0.32 cm (1/8 in.) up to 1.27 cm (1/2 in.) (or larger
tf higher volume sampling trains are used) inside diameter (ID) nozzles
In increments of 0.16 cm (1/16 in.). Each nozzle shall be calibrated
according to the procedures outlined in the calibration section.
2.1.2 Probe liner—Borosilicate or quartz glass tubing with a
heating system capable of maintaining a gas temperature at the exit end
during sampling of no greater than 120 ± 14° C (248 + 25° F) or no
greater than such other temperature as specified by an applicable
subpart of the standards. Since the actual temperature at the outlet
of the probe is not monitored during sampling, probes constructed
according to APTD-0581 and utilizing the calibration curves of APTD-0576
or calibrated according to the procedure outlined in APTD-0576 will be
considered as acceptable.
139
-------
Porosilicate or auartz glass probe liners shall be used for
temperatures up to about 480° C (900° F) and auartz liners for
tenperatures up to about 900° r (1650° F). Both nay be used at
higher temperatures for short periods of time, but nust be approved
by the Administrator. The softening temperature for borosilicate
is 820° C (1508° F) and for ouartz it is 1500° C (2732° F).
When length limitations, i.e. oreater than about 2.5 m (8.2 ft),
are encountered at temperatures less than 320° C (608° F), stainless
steel (316) or Incoloy 825 (both of seamless tubing), or other
materials as approved by the Administrator, may be used. Metal
probes for sampling pas streams at temperatures in excess of 320° C
(608° F) must be approved by the Administrator.
2.1.3 Pitot tube—Type S, or other device approved by the
Administrator, attached to probe to allow constant monitoring of the
stack gas velocity. The face openings of the pitot tube and the probe
nozzle shall be adjacent and parallel to each other, not necessarily
on the same nlane, during sampling. The free space between the nozzle
and pitot tube shall be at least 1.9 cm (0.75 in.). The free space
shall be set based on a 1.3 cm (0.5 in.) ID nozzle. If the sampling
train is designed for sampling at higher flow rates than that described
in PPTD-CSSl, thus necessitating the use of larger sized nozzles, the
largest sized nozzle shall be used to set the free space.
The nitot tube must also reet the criteria specified in Method 2
and calibrated according to the procedure in the calibration section
of that method.
Mention of trade names or specific products does not constitute
endorsement by the Environmental Protection Agency.
140
-------
2.1.4 Differential pressure gauge—Inclined manometer capable
of measuring velocity head to within 10% of the minimum measured value.
Below a differential pressure of 1.3 mm (0.05 in.) water gauge,
micromanometers with sensitivities of 0.013 mm (0.0005 in.) should
be used. However, micromanometers are not easily adaptable to field
conditions and are not easy to use with pulsating flow. Thus, methods
or other devices acceptable to the Administrator may be used when con-
ditions warrant.
2.1.5 Filter holder--Borosilicate glass with a glass frit
filter support and a silicone rubber gasket. Other materials of
construction may be used with approval from the Administrator,
e.g. if probe liner is stainless steel, then filter holder may be
stainless steel. The holder design shall provide a positive seal
against leakage from the outside or around the filter.
2.1.6 Filter heating system—Any heating system capable of
maintaining a temperature around the filter holder during sampling
of no greater than 120 ± 14° C (248 + 25° F), or such other temperature
as specified by an applicable subpart of the standards. A temperature
gauge capable of measuring temperature to within 3° C (5.4° F) shall
be installed such that temperature around the filter holder can be
regulated and monitored during sampling. Heating systems other than
shown in APTD-0581 may be used.
2.1.7 Condenser—Any system that cools the sample gas stream
and allows measurement of the water condensed and moisture leaving the
condenser, each to within 1 ml or 1 g. Acceptable means are to
141
-------
measure the condensed water either gravimetrically or volumetrically
and to measure the moisture leaving the condenser by (1) monitoring
the temperature and pressure at the exit of the condenser and using
Dalton's law or (2) by passing the sample gas stream through a tared
silica gel trap with exit gases kept below 20 C (68 F) and determining
the weight gain.
Note: If "condensible particulate matter" is desired, in addition
to moisture content, the following system shall be used—four impingers
connected in series with ground glass, leak free fittings or any
similarly leak free noncontaminating fittings. The first, third, and
fourth impingers shall be of the Greenburg-Smith design, modified by
replacing the tip with a 1.3 cm (1/2 in.) ID glass tube extending to
about 1.3 cm (1/2 in.) from the bottom of the flask. The second
impinger shall be of the Greenburg-Smith design with the standard tip.
Individual States or control agencies requiring this information shall
be contacted as to the sample recovery and analysis of the impinger
contents.
For purposes of writing the procedure of this method, the system
described in the note above will be used for determining the moisture
content of the stack gas. Modifications (e.g. using flexible con-
nections between the impingers or using materials other than glass)
may be used with approval from the Administrator.
If means other than silica gel are used to determine the amount of
moisture leaving the condenser, it is recommended that silica gel still
be used between the condenser system and pump to prevent moisture con-
densation in the pump and metering devices.
142
-------
Unless otherwise specified by the Administrator, flexible
vacuum lines may be used to connect the filter holder to the con-
denser.
2.1.8 Metering system—Vacuum gauge, leak^free pump, thermometers
capable of measuring temperature to within 3° C (5.4° F), dry gas meter
with 2% accuracy, and related equipment, or equivalent, as required to
maintain an isokinetic sampling rate and to determine sample volume.
Sampling trains utilizing metering systems designed for higher flow
rates than that described in APTD-0581 or APTD-0576 may be used pro-
vided that the specifications in section 2 of this method are met. When
the metering system is used in conjunction with a pi tot tube, the system
shall enable checks of isokinetic rates.
2.1.9 Barometer—Mercury, aneroid, or other barometers capable
of measuring atmospheric pressure to within 2.5 mm Hg (0.1 in. Hg).
In many cases, the barometric reading may be obtained from a nearby
weather bureau station, in which case the station value shall be
requested and an adjustment for elevation differences shall be applied
at a rate of minus 2.5 mm Hg (0.1 in. Hg) per 30 m (100 ft) elevation
increase.
2.1.10 Gas density determination equipment—Temperature and
pressure gauges and gas analyzer as described in Methods 2 and 3.
2.1.11 Temperature and pressure gauges—If Dalton's law is
used, to monitor temperature and pressure at condenser outlet. The
temperature gauge shall have an accuracy of 1° C (2° F). The pressure
gauge shall be capable of measuring pressure to within 2.5 mm Hg
143
-------
(0.1 in. Hg). If silica gel is used in the condenser system the
temperature and pressure nust be reasured before the silica gel
component.
2.2 Sample recovery,,
2.2.1 Probe liner and nrobe nozzle brushes—Nylon bristles
with stainless steel wire handles. The probe brush shall have
extensions, at least as long as the probe, of stainless steel, nylon,
teflon, or similarly inert raterial. Poth brushes shall be properly
sized and shaped to brush out the nrcbe liner and nozzle.
2.2.2 Glass wash bottles—Two.
2.2.3 Glass sample storage containers—Chemically resistant,
borosilicate narrow mouth glass bottles, for acetone washes, 500 ml or
1,000 ml. Screw cap closures shall be teflon rubber-backed liners or of
such construction so as to be leak free and prevent chemical attack
from the acetone. Other types of containers must be approved by the
Administrator.
2.2.4 Petri dishes--For filter samples, alass or plastic, unless
otherwise specified by the Administrator.
2.2.5 Graduated cylinder and/or balance—To measure condensed
water to within 1 ml or 1 g. Graduated cylinders shall have sub-
divisions no greater than 2 ml. Most laboratory balances are capable
of weighing to the nearest 0.5 g or less. Any of these balances are
suitable for use here and in section 2.3.4.
2.2.6 Plastic storage containers—Air tight containers to store
silica gel.
2.2.7 Funnel and rubber policeman—To aid in transfer of silica
gel to container; not necessary if silica gel is weighed in the field.
144
-------
2.3 Analysis.
2.3.1 Glass weighing dishes.
2.3.2 Desiccator.
2.3.3 Analytical balance—To measure to within 0.1 mg.
2.3.4 Balance--To measure to within 0.5 a.
2.3.5 Reakers~250 ml.
2.3.6 Hygrometer—To measure the relative humidity of the
laboratory environment.
2.3.7 Temperature gauge—To measure the temperature of the
laboratory environment.
3. Reagents
3.1 Sampling.
3.1.1 Filters--Glass fiber filters, without organic binder
exhibiting at least 99.95% efficiency (5 0.05% penetration) on
0.3 micron dioctyl phthalate smoke particles. The filter efficiency
test shall be conducted in accordance with /'STM standard method
D 2986-71. Test data from the supplier's duality control program
is sufficient for this purpose.
3.1.2 Silica gel —Indicating type, 6-16 mesh. If previously
used, dry at 175° C (350° F) for 2 hours. New silica pel may be
used as received.
3.1.3 Water—Men analysis of the material caught in the impingers
is required, distilled water shall be used. Run blanks prior to field
use to eliminate a high blank on test samples.
3.1.4 Crushed ice.
145
-------
3.1.5 Stopcock grease—Acetone insoluble, heat stable silicone
grease. This is not necessary if screw-on connectors with teflon
sleeves, or similar, are used.
3.2 Sample recovery.
3.2.1 Acetone—Reagent grade, < 0.001% residue, in glass bottles.
Acetone from metal containers generally has a high residue blank and
should not be used. Sometimes, suppliers transfer acetone to glass
bottles from metal containers. Thus, acetone blanks shall be run
prior to field use and only acetone with low blank values (< 0.001%)
shall be used.
3.3 Analysis.
3.3.1 Acetone—Same as 3.2.1.
3.3.2 Desiccant—Anhydrous calcium sulfate, indicating type.
4. Procedure
4.1 Sampling. The sampling shall be conducted by competent
personnel experienced with this test procedure.
4.1.1 Pretest preparation. All the components shall be main-
tained and calibrated according to the procedure described in APTD-0576,
unless otherwise specified herein.
Weigh approximately 200-300 g of silica gel in air tight containers
to the nearest 0.5 g. Record the total weight, both silica gel and
container, on the container. More silica gel may be used but care
should be taken during sampling that it is not entrained and carried
out from the impinger. As an alternative, the silica gel may be
weighed directly in the impinger or its sampling holder just prior
to the train assembly.
146
-------
Check filters visually against light for irregularities and
flaws or pinhole leaks. Label a filter of proper diameter on the
back side near the edge using numbering machine ink. As an alterna-
tive, label the shipping container (glass or plastic petri dishes)
and keep the filter in this container at all times except during
sampling and weighing.
Desiccate the filters at 20 ± 5.6° C (68 ± 10° F) and ambient
pressure for at least 24 hours and weigh at 6 or more hour intervals
to a constant weight, i.e., < 0.5 mg change from previous weighing,
and record results to the nearest 0.1 mg. During each weighing the
filter must not be exposed to the laboratory atmosphere for a period
greater than 2 minutes and a relative humidity above 50%.
4.1.2 Preliminary determinations. Select the sampling site and
the minimum number of sampling points according to Method 1 or as
specified by the Administrator. Determine the stack pressure,
temperature, and the range of velocity heads using Method 2 and
moisture content using Approximation Method 4 or its alternatives
for the purpose of making isokinetic sampling rate calculations.
Estimates may be used. However, final results will be based on actual"
measurements made during the test.
Select a nozzle size based on the range of velocity heads such
that it is not necessary to change the nozzle size in order to maintain
isokinetic sampling rates. During the run, do not change the nozzle
size. Ensure that the differential pressure gauge is capable of
measuring the minimum velocity head value to within 10%, or as
specified by the Administrator.
147
-------
Select a suitable probe liner and probe length such that all
traverse points can be sampled. Consider sampling from opposite
sides for large stacks to reduce the length of probes.
Select a total sampling time greater than or equal to the
minimum total sampling time specified in the test procedures for
the specific industry such that the sampling time per point is not
less than 2 min. or some greater time interval as specified by the
Administrator and the sample volume that will be taken will exceed
the required minimum total gas sample volume specified in the test
procedures for the specific industry. The latter is based on an
approximate average sampling rate. Note also that the minimum total
sample volume is corrected to standard conditions.
It is recommended that 1/2 or an integral number of minutes be
sampled at each point in order to avoid timekeeping errors.
In some circumstances, e.g. batch cycles, it may be necessary
to sample for shorter times at the traverse points and to obtain
smaller gas sample volumes. In these cases, the Administrator's
approval must first be obtained.
4.1.3 Preparation of collection train. During preparation and
assembly of the sampling train, keep all openings where contamination
can occur covered until just prior to assembly or until sampling is
about to begin.
Place 100 ml of water in each of the first two impingers, leave
the third impinger empty, and place approximately 200-300 g or more,
if necessary, of preweighed silica gel in the fourth impinger. Record
148
-------
the weight of the silica pel and container to the nearest 0.5 p.
Place the container in a clean place for later use in the sample
recovery.
Using a tweezer or clean disposable surgical ploves, place the
labeled (identified) and weighed filter in the filter holder. Be sure
that the filter is properly centered and the gasket properly placed
so as to not allow the sample pas stream to circumvent the filter.
Check filter for tears after assembly is completed.
When class liners are used, install selected nozzle using a
Viton A 0-ring when stack temperatures are less than 260° C (500° F)
or an asbestos string nasket when temperatures are higher. The Viton
A 0-ring and asbestos string gasket are installed as a seal where
the nozzle is connected to a glass liner. See APTD-0576 for details.
When metal liners are used, install the nozzle as above or by a leak
free direct mechanical connection. Mark nrobe with heat resistant
tape or by sone other method to denote the proper distance into the
stack or duct for each sampling point.
Unless otherwise specified by the Administrator, attach a
temperature probe to the metal sheath of the sampling nrobe so that
the sensor extends beyond the nrobe tin and does not touch any metal.
Its position should be about 1.9 to 2.54 cm (0.75 to 1 in.) from
the pi tot tube and probe nozzle to avoid interference with the gas flow.
Set up the train as in Figure 5-1, using, if necessary, a very
light coat of silicone nrease on all ground glass joints, greasing
only the outer oortion (see APTD-0576) to avoid possibility of
149
-------
contamination by the silicone grease. V.'ith approval from the
Administrator, a nlass cyclone ray be used between the probe and
filter holder.
Place crushed ice around the impingers.
4.1.4 leak check procedure—After the sampling train has been
assembled, turn on and set the filter and nrobe heating system to the
pov/er required to reach a temperature of 120 + 14° C (248 + 25° F)
or such other temperature as specified by an applicable subpart of
the standards for the leak check. (If water condensation is not a
problem the probe and/or filter heating system need not be used.)
Allow time for the temperature to stabilize. If a Viton A 0-ring
or other leak free connection is used in assembling the probe nozzle
to the rrobe liner, leak check the train at the sampling site by plug-
ging the nozzle and pulling a 380 mm Hg (15 in. Hg) vacuum. (Note: A
lower vacuum may be used provided that it is not exceeded during the
test.) If an asbestos string is used, do not connect the probe to the
train during the leak check. Instead, leak check the train as above by
first plugging the inlet to the filter holder. Then connect the probe to
the train and leak check at about 25 rm Hg (1 in. Ho) vacuum. A leakage
rate in excess of 4?^ of the average samnlino rate or 0.00057 m /min.
(0.02 cfn), whichever is less, is unacceptable in either case.
The following leak check instructions for the sampling train
described in APTD-C576 and APTD-0581 nay be helpful. Start the pump
with by-pass valve fully open and coarse adjust valve completely
closed. Partially open the coarse adjust valve and slowly close
the by-pass valve until 380 mm Mo (15 in. Hg) vacuum is reached.
Po not reverse direction of by-pass valve. This will cause water to
back up into the filter holder. If 380 rm Hg (15 in. Hg) is exceeded,
150
-------
either leak check at this higher vacuum or end the leak check as
shown below and start over.
When the leak check is completed, first slowly remove the
plug from the inlet to the probe or filter holder and immediately
turn off the vacuum pump. This prevents the water in the impingers
from being forced backward into the filter holder and silica gel
from being entrained backward into the third impinger.
Leak checks shall be conducted as described whenever the train
is disengaged, e.g. for silica gel or filter changes during the
test, prior to each test run, and at the completion of each test run.
If leaks are found to be in excess of the acceptable rate, the test
will be considered invalid. To reduce lost time due to leakage
occurrences, it is recommended that leak checks be conducted
between port changes.
4.1.5 Particulate train operation—During the sampling run,
isokinetic sampling rate to within 10%, or as specified by the
Administrator, of true isokinetic and the temperature around the
filter of no greater than 120 ± 14° c (248 ± 25° F), or as specified
by an applicable subpart of the standards, shall be maintained.
For each run, record the data required on the example data
sheet shown in Figure 5-2. Be sure to record the initial dry gas
meter reading. Record the dry gas meter readings at the beginning
and end of each sampling time increment, when changes in flow rates
are made, and when sampling is halted. Take other data point readings
at least once at each sample point during each time increment and
151
-------
_
00 OC
LU =}
OC t-
Q. CO X
E —
I s
= ° =
« cc £
S < co
< CD <
E
u
oc"
LU
LU
S
5
N
IM
a
< I—
£ S
- OC
(si
a
<
EC
o
OC
< _i
5-i
<
o
a:
a.
o
LU
CO
CO
O
cc
O
U
o
o
I
u
in
z
o
ac
0
i.. ^
t- <
c
o
^ z
o x ®
00 <=> 5
CO ~2;5:>— t— 2
cc
o>
>
_J
0
01
LU
CC
CO
CO
T3
0)
(U
*•—
ffl
o
ca
Q.
LO
a>
o>
LL
152
-------
additional readings when significant changes (20% variation in
velocity head readings) necessitate additional adjustments in flow
rate. Level and zero the manometer.
Clean the portholes prior to the test run to minimize chance
of sampling the deposited material. To begin sampling, remove the
nozzle cap, verify that the filter and probe are up to temperature,
and that the pitot tube and probe are properly positioned. Position
the nozzle at the first traverse point with the tip pointing directly
into the gas stream. Immediately start the pump and adjust the flow
to isokinetic conditions. Nomographs are available for sampling
trains using type S pitot tubes with 0.85 + °-02 coefficient and
when sampling in air or a stack gas with equivalent density (molecular
weight equal to 29+4), which aid in the rapid adjustment of the
isokinetic sampling rate without excessive computations. APTD-0576
details the procedure for using these nomographs. If C and M . are
outside the above stated ranges, do not use the nomograph unless
appropriate steps are taken to compensate for the deviations.
When the stack is under significant negative stack pressure
(height of impinger stem), take care to close the coarse adjust valve
before inserting the probe into the stack to avoid water backing into
the filter holder. If necessary, the pump may be turned on with the
coarse adjust valve closed.
When the probe is in position, block off the openings around
the probe and porthole to prevent unrepresentative dilution of the
gas stream.
153
-------
Traverse the stack cross section, as required by Method 1 or
as specified by the Administrator, being careful not to bump the
probe nozzle into the stack walls when sampling near the walls or
when removing or inserting the probe through the portholes to
minimize chance of extracting deposited material.
Turing the test run, rake periodic adjustments to keep the
temperature around the filter holder at the proper temperature and
add more ice and, if necessary, salt to maintain a temperature of
less than 20° C (68° F) at the condenser/silica ael outlet to avoid
excessive moisture losses. /Hso, periodically check the level and
zero of the nanometer.
If the pressure drop across the filter becomes too high making
isokinetic sampling difficult to maintain, the filter may be replaced
in the midst of a sample run. It is recommended that another complete
filter assembly he used rather than attempting to change the filter
itself. /*fter the new filter or filter assembly is installed
conduct a leak check. The parti oil ate veirht shall include the
summation of all filter assembly catches.
P single train shall be used for the entire sample run, except
for filter and silica rel changes. However, if approved by the
Administrator, two or pore trains may be used for a single test run
when there are two or more ducts or sampling ports. The results shall
be the total of all sampling train catches.
At the end of the sample run, turn off the oump, remove the
probe and nozzle from the stack, and record the final dry gas meter
reading. Perform a leak check at a vacuum egual to or greater than
the maximum reached during sampling. Calculate percent isokinetic (see
calculation section) to determine whether another test run should be
154
-------
made. If there is difficulty in maintaininq isokinetic rates due
to source conditions, cons'dt with the Administrator for possible
variance on the isokinetic rates.
4.2 Sample recoverv. Proper cleanuo nrocedure heoins as
soon as the nrobe is removed from the stack at the end of the samolina
peri od,
When the probe can he safely handled, wipe off all external
particulate matter near the tio of the nrohe nozzle and place a cap
over it to prevent losina or aainina oarticulate matter. Do not cap
off the orobe tip tightly while the samplinq train is coolinq down as
this would create a vacuum in the filter holder, thus drawinq water
from the impingers into the filter.
Before movinq the sample train to the cleanun site, remove the
probe •from the samnle train, wipe off the silicone grease, and can
the ooen outlet of the probe. Be careful not to lose any condensate,
if present. Wipe off the silicone grease from the filter inlet where
the probe was fastened and can it. Remove the umbilica"1 cord from the
last impinaer and cap the impinoer. If a flexible line is used between
the first imninaer or condenser and the -filter holder, disconnect the
line at the filter holder and let any condensed water or liquid drain
into the impinqers or condenser. After wipino off the silicone nrease,
cap off the filter holder outlet and imninqer inlet. Either qround
glass stoppers or plastic caps or serum cans may be used to close
these openings.
Transfer the probe and filter-impinqer assembly to the cleanup
area. This area should be clean and nrotected from the wind so that
155
-------
the chances of contaminating or losing the sample will be minimized.
Save a portion of the acetone used for cleanup as a blank.
Place about 200 ml of this acetone in a glass sample container
labeled "acetone blank."
Inspect the train prior to and during disassembly and note any
abnormal conditions. Treat the samples as follows:
Container No. 1. Carefully remove the filter from the filter
holder and place in its identified petri dish container. Use a pair
of tweezers and/or clean disposable surgical gloves to handle the
filter. If it is necessary to fold the filter, do so such that the
particulate cake is inside the fold. Quantitatively remove any
particulate matter and/or filter which adheres to the filter holder
by carefully using a dry nylon bristle brush and/or a sharp-edged
blade and place into this container. Seal the container.
Container No. 2. Taking care to see that dust on the outside
of the probe or other exterior surfaces does not get into the sample,
quantitatively recover particulate matter or any condensate from the
probe nozzle, probe fitting, probe liner, and front half of the
filter holder by washing these components with acetone and placing
the wash into a glass container in the following manner.
Distilled water may be used when approved by the Administrator
or shall be used when specified by the Administrator. In these cases,
save a water blank and follow Administrator's directions on analysis.
Carefully remove the probe nozzle and clean the inside surface
by rinsing with acetone from a wash bottle and brushing with a nylon
156
-------
bristle brush. Brush until acetone rinse shows no visible particles,
after which make a final rinse of the inside surface with acetone.
Brush and rinse with acetone the inside parts of the Swagelok
fitting in a similar way until no visible particles remain.
Rinse the probe liner with acetone by tilting the probe and
squirting acetone into its upper end, while rotating the nrobe so
that all inside surfaces will be rinsed with acetone. Let the
acetone drain from the lower end into the sample container. A funnel
may be used to aid in transferring liquid washes to the container.
Follow the acetone rinse with a nrobe brush. Hold the probe in an
inclined position, sauirt acetone into the upner end as the probe brush
is being pushed with a twisting action through the probe, hold a sample
container underneath the lower end of the probe, and catch any acetone
and particulate matter which is brushed from the nrobe. Run the brush
through the nrobe three tines or more until no visible particulate
matter is carried out with the acetone or remains in the probe liner
on visual inspection. With stainless steel or other metal probes,
run the brush through in the above prescribed manner at least six times
since metal probes have small crevices in which particulate matter can
be entrapped. Rinse the brush with acetone and Quantitatively collect
these washings in the sample container. After the brushing make a
final acetone rinse of the probe as described above.
It is recommended that two people be used to clean the probe to
minimize losing the sample. Between sampling runs, keep brushes clean
and protected from contamination.
157
-------
After ensurinq that all joints are wiped clean of silicone
grease, clean the inside of the front half of the -Filter holder
by rubbing the surfaces with a nylon bristle brush and rinsinn with
acetone. Rinse each surface three times or more if needed to remove
visible particulate. Make a final rinse of the brush and filter
holder. After all acetone washinqs and oarticulate matter are
collected in the sample container, tiqhten the lid on the samnle
container so that acetone will not leak out when it is shipped to
the laboratory. Mark the heiqht of the fluid level to determine
whether or not leakaqe occurred durinq transport. Label container
to clearly identify its contents.
Container No. 3. Note color of indicatinq silica qel to deter-
mine if it has been completely spent and make a notation of its
condition. Transfer the silica qel from the fourth imninaer to the
original container and seal. A funnel may make it easier to pour the
silica qel without soillinq. A rubber policeman may be used as an aid
in removinq the silica eel from the imoinqer. It is not necessary to
remove the small amount of dust particles that may adhere to the v/alls
and are difficult to remove. Since the nain in weinht is to he used
for moisture calculations, do not use any water or other liquids to
transfer the silica qel. If a balance is available in the field, *ollow
the procedure under analysis.
Impinqer water. Treat the imoinaers or condenser as follows:
Make a notation of any color or film in the liquid catch. Measure
the liquid which is in the first three imninqers to within ± 1 ml by
158
-------
using a oraduated cylinder or, if available, to within ± 0.5 q
by using a balance. Record the volume or weight of liquid present.
This information is required to calculate the moisture content of
the effluent aas.
If analysis of the impinaer catch is not required, discard the
liquid after measuring and recording the volume or weiqht. If
analysis of the impinaer catch is reauired, leave the imoinaers
intact to transfer the liquid, cap off the inlet, and pour the liquid
throuqh the outlet into the graduated cylinder or into a sample con-
tainer after its weiqht has been determined.
If a different tyne of condenser is used, measure the amount
of moisture condensed either volumetrically or aravimetrically.
4.3 Analysis. Record the data required on the example sheet
shown in Fiaure 5-3. Handle each sample container as follows:
Container No. 1. Leave in shipping container or transfer the
filter and any loose particulate from the sample container to a
tared glass weighing dish and desiccate for 24 hours in a desiccator
containing anhydrous calcium sulfate. Weigh to a constant weinht
and report the results to the nearest 0.1 mg. For purposes of this
section 4.3, the term "constant weiqht" means a difference of no
more than 0.5 mq or T£ of total weioht, less tare weiaht, whichever
is qreater, between two consecutive weiahinqs, with no less than 6
hours of desiccation time between weiqhinas and no more than 2 minutes
exposure to the laboratory atmosphere (must be less than 50% relative
humidity) durina weighing.
159
-------
Plant.
Date
Run No
Relative Humidity.
Amount liquid lost during transport
Acetone blank volume, ml
Acetone wash volume, ml
Acetone blank concentration, mg/mg (equation 5-4).
Acetone wash blank, mg (equation 5-5)
CONTAINER
NUMBER
1
2
TOTAL
WEIGHT OF PARTICULATE COLLECTED,
mg
FINAL WEIGHT
:xr
TARE WEIGHT
:xr
Less acetone blank
Weight of participate matter
WEIGHT GAIN
FINAL
INITIAL
LIQUID COLLECTED
TOTAL VOLUME COLLECTED
VOLUME OF LIQUID
WATER COLLECTED
IMPINGER
VOLUME,
ml.
SILICA GEL
WEIGHT,
g
9* ml
CONVERT WEIGHT OF WATER TO VOLUME BY DIVIDING TOTAL WEIGHT
INCREASE BY DENSITY OF WATER (1g/ml).
: VOLUME WATER, ml
1 g/ml
Figure 5-3. Analytical data.
160
-------
Container No. 2. Note level of liquid in container and
confirm on analysis sheet whether or not leakage occurred during
transport. Measure the liquid in this container either volume-
trically to ± 1 ml or gravimetrically to ± 0.5 g. Transfer the
contents to a tared 250 ml beaker, and evaporate to dryness at ambient
temperature and pressure. Desiccate for 24 hours and weigh to a
constant weight. Report the results to the nearest 0.1 nig.
Container No. 3. Weigh the spent silica gel to the nearest
0.5 g using a balance. This step may be conducted in the field.
"Acetone Blank" Container. Measure acetone in this container
either volumetrically or gravimetrically. Transfer the acetone to
a tared 250 ml beaker and evaporate to dryness at ambient temperature
and pressure. Desiccate for 24 hours and weigh to a constant weight.
Report the results to the nearest 0.1 mg.
5. Calibration
Maintain a laboratory log of all calibrations.
5.1 Probe nozzle. Using a micrometer, measure the inside
diameter of the nozzle to the nearest 0.025 mm (0.001 in.). Make
3 separate measurements usina different diameters each time and
obtain the average of the measurements. The difference between the
high and low numbers shall not exceed 0.1 mm (0.004 in.).
When nozzles become nicked, dented, or corroded, they shall
be reshaped, sharpened, and recalibrated before use.
Each nozzle shall be permanently and uniquely identified.
5.2 Pi tot tube. The pi tot tube shall be calibrated according
to the procedure outlined in Method 2.
161
-------
5.3 Dry gas meter and orifice meter. Both meters shall be
calibrated according to the procedure outlined in APTD-0576. When
diaphragm pumps with by-pass valves are used, check for proper
metering system design by calibrating the dry aas meter at an
•?
additional flow rate of 0.0057 m /min. (0.2 cfm) with the by-pass
valve fully opened and then with it fully closed. If there is more
than ± 2% difference in flow rates when compared to the fully closed
position of the by-pass valve, the system is not designed properly
and must be corrected.
5.4 Probe heater calibration. The probe heating system shall
be calibrated according to the procedure contained in APTD-0576.
Probes constructed according to APTD-0581 need not be calibrated if
the calibration curves in APTD-0576 are used.
5.5 Temperature pauaes. Calibrate dial and liquid filled bulb
thermometers against mercury-in-glass thermometers. New thermocouples
need not be calibrated. Calibrate used thermocouples aaainst new ones.
For other devices, check with the Administrator.
6. Calculations
Carry out calculations, retaining at least one extra decimal
figure beyond that of the acquired data. Round off figures after
final calculation.
6.1 Nomenclature
2 2
A = Cross sectional area of nozzle, m (ft )
B = Water vapor in the gas stream, proportion by volume
ws
C = Acetone blank residue concentration, mg/mg
a
c = Concentration of particulate matter in stack aas, dry basis,
corrected to standard conditions, g/dscm (g/dscf)
162
-------
I » Percent of isokinetic sampling
m = Total amount of particulate matter collected, mg.
M = Molecular weight of water, 18 g/g-mole (18 Ib/lb-mole)
w
m = Mass of residue of acetone after evaporation, mg
a
?^ = Barometric pressure at the sampling site, mm Hg (in. Hg)
P = Absolute stack gas pressure, mm Hg (in. Hg)
P .. = Standard absolute pressure, 760 mm Hg (29.92 in. Hg)
R = Ideal gas constant, 0.06236 mm Hg-m /°K-g-mole (21.83 in.
Hg-ft3/°R-1b-mole)
T = Absolute average dry gas meter temperature (see Figure 5-2),
°K (°R)
TS = Absolute average stack gas temperature (see Figure 5-2),
°K <°R).
Tstd = standard absolute temperature, 293° K (528° R)
V. = Volume of acetone blank, ml
a
V,, = Volume of acetone used in wash, ml
«W
V, = Total volume of liquid collected in impingers and silica
gel (see Figure 5-3, ml.
V = Volume of gas sample as measured by dry gas meter,
dcm (dcf)
V / td\ = Volume of gas sample measured by the dry gas meter
corrected to standard conditions, dscm (dscf).
V / . .\ * Volume of water vapor in the gas sample corrected to
standard conditions, son (scf).
163
-------
Wa
AH
w
= Stack gas velocity, calculated by Method 2, Equation 2-7
using data obtained from Method 5, m/sec (ft/sec)
= Weight of residue in acetone wash, mg
= Average pressure differential across the orifice (see
Figure 5-2), meter, mm H,,0 (in. H20)
= Density of acetone, mg/ml (see label on bottle)
= Density of water, 1 g/ml (0.00220 Ib/ml)
e = Total sampling time, min.
13.6 = Specific gravity of mercury
60 = sec/mi n
TOO = Conversion to percent
6.2 Average dry gas meter temperature and average orifice
pressure drop. See data sheet (Figure 5-2).
6.3 Dry gas volume. Correct the sample volume measured by the
dry gas meter to standarc conditions (20 C, 760 mm Hg or 68 F,
29.92 in. Hg) by using Equation 5-1.
Vm(std) = Vm
AH
std
= K V.
m
+ AH/13.6
Equation 5-1
where:
K
0.3855 °K/mm Hg for metric units
17.65 °R/1n. Hg for English units
164
-------
6.4 Volume of water vapor.
'w(std)
K Vlc Equation 5-2
where
3
K = 0.00134 m /ml for metric units
= 0.0472 ft /ml for English units
6.5 Moisture content.
6.
Vstd)
ws
m(std)
w(std)
6.6 Acetone blank concentration.
Equation 5-3
Equation 5-4
6.7 Acetone wash blank.
W • C V p
a a aw a
Equation 5-5
6.8 Total particulate weight. Determine the total particulate
catch from the sum of the weights obtained from containers 1 and 2
less the acetone blank (see Figure 5-3).
6.9 Particulate concentration.
cs - (0.001 g/mg)
Equation 5-6
165
-------
6.10 Conversion factors:
From jp_ Multiply by
0.0283
15.4
2.205 x 10"3
35.34
6.11 Isokinetic variation.
6.11.1 Calculations from raw data.
scf
•3
g/ft3
g/ft3
g/ft3
oj
m
3
gr/ft
lb/ft3
g/m3
("
bar
60 0 v P A
Equation 5-7
where:
K = 0.00346 mm Hg-m3/ml-°K for metric units
= 0.00267 in. Hg-ft3/ml-°R for English units
6.11.2 Calculations from intermediate values.
100
I =
s ym(std) std
_
Tstd vs eAnPs 60
Equat1on 5"
where:
K = 4.323 for metric units
= 0.0944 for Enolish units
166
-------
6.12 Acceptable results. If 90% < I < 110%, the results are
acceptable. If the results are low in comparison to the standards
and I is beyond the acceptable range, the Administrator may option
to accept the results. Use reference 7.4 to make judgments. Other-
wise, reject the results and repeat the test.
7. References
7.1 Addendum to Specifications for Incinerator Testing at
Federal Facilities, PHS, NCAPC, Dec. 6, 1967.
7.2 Martin, Robert M., Construction Details of Isokinetic Source
Sampling Equipment, Environmental Protection Agency, APTD-0581.
7.3 Rom, Jerome J., Maintenance, Calibration, and Operation
of Isokinetic Source Sampling Equipment, Environmental Protection
Agency, APTD-0576.
7.4 Smith, W. S., R. T. Shigehara, and W. F. Todd, A Method of
Interpreting Stack Sampling Data, Paper presented at the 63d Annual
Meeting of the Air Pollution Control Association, St. Louis, Mo.,
June 14-19, 1970.
7.5 Smith, W. S., et al., Stack Gas Sampling Improved and
Simplified with New Equipment, APCA paper No. 67-119, 1967.
7.6 Specifications for Incinerator Testing at Federal Facilities,
PHS, NCAPC, 1967.
167
-------
APPENDIX B ILLUSTRATED USE OF NOMOGRAPHS
168
-------
APPENDIX B ILLUSTRATE) USE OF NOMOGRAPHS
The material in this appendix is, in the most part, reproduced from
APTD-0576 (ref. 9).
NOMOGRAPHS
The correction factor nomograph (fig. Al) and the operating nomograph
(fig. A2) have been designed for use with the sampling train as aids for
rapid isokinetic sampling rate adjustments and for selection of proper
nozzle size. To determine the correction factor, C, on the nomograph, the
following information is first required:
(1) Percent moisture, %H20. This may be determined from a previous
test or presurvey, or before the sample run.
(2) Orifice calibration factor, AH@. This is determined from the
laboratory calibration (see section on Calibration).
(3) Meter temperature, T . Temperature at the meter rises above
m
ambient temperature because of the pump and can easily be estimated with
experience. An estimate within 10° F (approximately + 1 percent error) is
all that is necessary (an initial estimate of about 25°F above ambient
temperature has been used). This approximation above ambient temperature
is not required if the pump is located outside the meter box console.
RET 2
~^~
-z a
-l 5
1 __-
.=-^0 1T£_!1_ " ~
-0 6
-0 5
EXAMPLE AH^; i 8
Tm = 100 °F
% H2O : 10
Ps/Pm -- 1 05
FIND C : 1 0
rS/rM
- I 2
Figure Al. Nomograph for correction factor, C.
169
-------
(4) Stack pressure, P . This is measured before the sample run; or
s
if the sampling site is near the exit of the stack, atmospheric pressure
is used.
(5) Meter pressure, P . Same as atmospheric pressure.
To obtain correction factor, C (fig. Al):
(1) Draw line from AR@ to T to obtain point "A" on reference line 1
(ref. 1).
(2) Draw line from point "A" to %H~0 to obtain point "B" on reference
line 2 (ref. 2).
(3) Draw line from point "B" to the calculated value P /P to obtain
S TD.
correction factor, C.
To select the nozzle size and to set the K-factor on the operating
nomograph, the following information is first required:
(1) C factor. This is obtained from the correction-factor nomograph
(fig. Al).
Note: If the coefficient, C , of the type-S pitot tube is not
P
equal to .85 + -02, the following is required: (a) Mul-
/C \
tiply C times I gTJ = C1 for the correct C factor in
obtaining the K-factor, or (b) if C' is less than 0.5, then
/C \2
use C and multiply each AP reading by ( 1,. ) for each adjust-
ment.
(2) Stack temperature, T . This is determined in °F by a rough
S
temperature traverse to within + 25°F before the sample run.
(3) Average velocity pressure, AP. This is determined by a rough
preliminary pitot traverse, using the average of minimum and maximum AP's
in inches of water.
(4) Exact available nozzle sizes, D. This is obtained from the cali-
bration of available nozzles.
To select the nozzle size and to set the K-factor pivot point, use the
following procedure (fig. A2):
(1) Set correction fs.ctor, C, on sliding scale to the reference mark,
"A."
(2) Align T with average AP, note probe tip diameter on D-scale, and
select exact nozzle size closest to it.
170
-------
CORRECTION
FACTOR,
C
ORIFICE READING,
A H
10
9—
8 —
7^—
6-=
=
•3
5^s
^
4-i
=
—
3~
~
3
2_5
—
—
—
, ~
0 9
0 &
0 7
0 6-H
0 5-|
0 4-^
0 3-=
0 2
0 1
REF A
S,REF B
^ f
— 2 0
— 1 5
~ STEP 1
— ! 0
— 09
08
— 07
0 6
— 05
•-•^
-
—2500
-
—3000
-
— 1 500
I
—
— 1000
— 800
— 600.
Z- 500
400
•^••4?."
^00
100
0
STACK
TEMPERATURE,
T
S
SLIDING
SCALE
CUT ALONG LINES
PITOT READING,
K FACTOR o 001
^-£rf/3
'""•^.^
__S_TEF_2
"
6.H: m. HjO
C ; dimension les
Ts =«>,.
"=
0 002^
•j
0 003-;
—
0 004^
_j
PROBE 0 °°5-
TIP DIAMETER, ~
0 006—
U —
001—^
-^.-Jfc
5-0 9 -;
^-08 002-
^-07 0 03 —
- 0 04-
:" 005 —
r 006-
E-o 5 —
- 0 08
i-0 4
" 0 2
~ 03-
~ ~
~ 04-
? .^ ' ^6.
- °'^
— 02 0 « —
1 0 —
-
_
2 0 —
3 0-
_ 40-
5 0.
6 0 —
8 ''_
1 0 0 —
EXAMPLE C : 1 0
K - dimension less Tg = 300 *F
D : in AP i 0 5 m HjO
AP : in. H2O D ; 0 250 in., 0 375 in
Figure A2. Use of the nomograph in selecting nozzle size
and setting K factor.
171
-------
(3) Align T with exact nozzle size selected and obtain a value on
&
the AP scale.
(4) Align the AP value with reference mark, "B", on AH scale, and set
the K-factor pivot point.
To obtain the orifice meter settings, AH, for isokinetic conditions
after the K-factor pivot point has been set, use the following procedure
(fig. A3):
(1) Position the pitobe nozzle at the sampling point.
(2) Read the pitot tube AP.
(3) Align the AP through the K-factor pivot point
(4) Obtain AH and adjust metering valves.
The nomograph assumes the following, once the K-factor pivot point is
set:
(1) T does not change more than 25° for T < 1000°F or 50° for
s s
T > 1000°F.
s
(2) D is not changed during the test.
(3) T was estimated correctly and does not vary more than 10°.
(4) Percent H^O remains constant, within +1.0 percent.
(5) P and P remain constant, within +1.0 percent.
s m —
172
-------
CORRECTION
FACTOR,
C PITOT READING,
ORIFICE READING,
AH
\o^
9 —
8 —
7 —
6— •
.5
•=
5~1
«:
i- =
—
^
—
™
~
~
2_
_
^
__
—
i —
0 9—
0 8 —
0.7 —
0.6-=
0.5-1
z
0 4HE
-H
0 3-=
""
0 2—
—
—
~
0 1 —
REF. A —
•-REF. B
£>d..
•~£4Mp/f: .
^^-C5 ^
r-i, i^
-Z^WlPLE •
-..^
— 2 0
-
— 1 5
—
_
— 1.0
— 09
HLo.8
«— 0.7
•—
^— 0 6
^
•— 0 ^
.
— 2500
Z.
—7000
-
— 1500
1.
-^-1 000
— 800
^_
-^,600
^560-^,^
— 400 **•'
«- 300
— 200
— 1 00
— 0
STACK
TEMPERATURE,
TS
SLIDING
SCALE
CUT ALONG LINES
AP
K FACTOR o.ooi —
.^^PIVOT POINT
••-^^.N.
^^
—
0.002^
0 003-^
—
0 004^
-5
PROBE o 005-S
TIP DIAMETER, —
0 006-=
LJ — ^
0 01 —
'*
-------
APPENDIX C ILLUSTRATED AUDIT PROCEDURES
A flow chart of the operations involved in an auditing program from
first setting desired limits on the data quality to filing the results is
given below. Assumed numbers are used and a sample calculation of an audit
is performed in the flow chart. Each operation is referred to the section
in the text of the report where it is discussed.
174
-------
MANAER
7.
LIMITS FOR DATA QUALITY CAN BE SET BY WHAT
IS DESIRED OR FROM THE NATURAL VARIABILITY
OF THE METHOD WHEN USED BY TRAINED AND
COMPETENT PERSONNEL^ FOR THIS EXAMPLE, IT
IS ASSUMED THAT CV{PMR} =9.3 PERCENT
(subsec.4.1), AND USING + 3 CV{PMR}, THE
LIMITS ARE L = -28 PERCENT AND U = 28
PERCENT.
FROM PRIOR KNOWLEDGE OF DATA QUALITY, ESTIMATE
THE PERCENTAGE OF FIELD MEASUREMENTS FALLING
OUTSIDE THE ABOVE LIMITS. IF NO INFORMATION
IS AVAILABLE, MAKE AN EDUCATED GUESS. IT IS
ASSUMED IN THIS EXAMPLE THAT 50 PERCENT OF THE
FIELD DATA ARE OUTSIDE THE LIMITS L AND U
(subsec. 4.4.4).
DETERMINE: (1) COST OF CONDUCTING AN AUDIT,
(2) COST OF FALSELY INFERRING THAT GOOD DATA
ARE BAD, (3) COST OF FALSELY INFERRING THAT
BAD DATA ARE GOOD, AND (4) COST SAVINGS FOR
CORRECTLY IDENTIFYING BAD DATA (subsec. 4.4.4).
DETERMINE THE AUDIT LEVEL EITHER BY (1) MINI-
MIZING AVERAGE COST USING EQUATION (41) OF
SUBSECTION 4.4.4, OR (2) ASSURING A DESIRED
LEVEL OF CONFIDENCE IN THE REPORTED DATA
THROUGH STATISTICS. FOR THIS EXAMPLE, THE
AUDIT LEVEL IS TAKEN AS n = 5 (fig. 15 ).
BY TEAMS, TYPES OF SOURCES, OR GEOGRAPHY,
GROUP FIELD TESTS INTO LOTS (GROUPS) OF ABOUT
20 THAT WILL BE PERFORMED IN A PERIOD OF ONE
CALENDAR QUARTER.
SELECT n OF THE N TESTS FOR AUDITING. COMPLETE
RANDOMIZATION MAY NOT BE POSSIBLE DUE TO AUDI-
TOR'S SCHEDULE. THE PRIMARY POINT IS THAT THE
FIELD TEAM SHOULD NOT KNOW IN ADVANCE THAT
THEIR TEST IS TO BE AUDITED.
ASSIGN OR SCHEDULE AN AUDITOR FOR EACH FIELD
TEST.
SET DESIRED
LOWER AND UPPER
LIMITS FOR DATA
QUALITY, L AND U
ESTIMATE AVERAGE
QUALITY OF FIELD
DATA IN TERMS OF
L AND U
DETERMINE OR
ASSUME RELEVANT
COSTS
I
DETERMINE AUDIT
LEVEL FROM
STATISTICS, OR
AVERAGE COST
GROUP FIELD TESTS
INTO LOT SIZES OF
ABOUT N = 20
RANDOMLY SELECT
n OF THE N TESTS
FOR AUDITING
ASSIGN/SCHEDULE
AUDITOR(S) FOR
FOR THE n AUDITS
T
175
-------
AUDITOR
8.
10.
11.
12.
13.
THE AUDITOR OBTAINS APPROPRIATE CALIBRATED
EQUIPMENT AND SUPPLIES FOR THE AUDIT
(subsec. 4.3).
OBSERVE THE FIELD TEAM'S PERFORMANCE OF THE
FIELD TEST. FILL IN THE AUDITOR'S CHECKLIST
(table 8) AND NOTE ANY UNUSUAL CONDITIONS
THAT OCCURRED DURING THE TEST.
PREPARE EQUIPMENT
AND FORMS
REQUIRED IN AUDIT
IN THE FIELD TEAM'S HOME LABORATORY, MAKE
INDEPENDENT DETERMINATIONS OF Cp> Dn, y, AND
AH@ (subsec. 4.3) ACCORDING TO THE CALIBRATION
PROCEDURES GIVEN IN SUBSECTION 2.2.
10
OBSERVE ON-SITE
PERFORMANCE
OF TEST
+
VERIFY CALIBRATION
RECORDS AND PERFORM
CALIBRATION CHECKS
IN TEAM'S HOME
LABORATORY
STARTING WITH THE RAW DATA FROM THE FIELD
AND USING AUDIT VALUES, PERFORM ALL THE
CALCULATIONS NECESSARY TO ARRIVE AT A VALUE
FOR d (subsec. 4.3, fig., 10).
THE AUDITOR'S REPORT SHOULD INCLUDE (1) DATA
SHEET FILLED OUT BY THE FIELD TEAM (fig. 3),
(2) AUDITOR'S CHECKLIST WITH COMMENTS
(table 8), (3) AUDIT DATA SHEET WITH CALCULA-
TIONS (fig. 11), and (4) A SUMMARY OF THE
TEAM'S PERFORMANCE WITH A NUMERICAL RATING
(subsec. 4.3).
THE AUDITOR'S REPORT IS FORWARDED TO THE
MANAGER.
11
PERFORM CALCULA-
TIONS TO DETERMINE
PMR-PMR.
d = ^MR-^x100
12
PREPARE
AUDIT
REPORT
13
FORWARD
REPORT TO
MANAGER
MANAGER
14. COLLECT THE AUDITOR'S REPORTS FROM THE n
AUDITS OF THE LOT OF N STACKS. IN THIS
CASE n = 5 AND ASSUMED VALUES FOR THE
AUDITS ARE dj = 12.0,d2 = -6.0, d3 = 3.0,
d = 15.0 AND d,- = 9.0 (table 8).
14
COMBINE
RESULTS OF
n AUDITS
I
176
-------
15. CALCULATE d AND s . ACCORDING TO THE SAMPLE IN
TABLE 10. RESULTS OF THIS SAMPLE CALCULATION
SHOW 3 = 6.6,
4.4.3).
AND s . = 8.3 (table 8, subsec.
15
CALCULATE THE
MEAN, 3, AND
STANDARD
DEVIATION, sd
16.
17.
18.
19.
20.
USE A t-TEST TO CHECK d FOR SIGNIFICANCE, FOR
THIS EXAMPLE t = 6.6 x /5/S.3 = 1.78. THE
TABULATED t-VALUE FOR 4 DEGREES OF FREEDOM AT
THE 0.05 LEVEL IS 2.13; HENCE, d IS NOT
SIGNIFICANTLY DIFFERENT FROM 0 AT THIS LEVEL.
ALSO, s, IS CHECKED AGAINST THE ASSUMED VALUE
OF 9.3 PERCENT BY A CHI-SQUARE TEST.
16
X2/f =
= (8.3)2/(9.3)2 = 0.797
17
THE TABULATED VALUE OF x/4 AT THE 95 PER-
CENT LEVEL IS 0.711; HENCE, s, IS SIGNIF-
ICANTLY DIFFERENT FROM 9.3 PERCENT (In this
case it is a significant improvement).
OBTAIN THE VALUE OF k FROM TABLE 9, FOR n = 5
AND p = 0.1. THIS VALUE IS 2.742, THEN
3 + k s . = 6.6 + 2.742(8.3) = 29.4 AND
3 = k s° = 6.6 - 2.742(8.3) = -16.2
(subsec? 4.4.3).
COMPARE THE ABOVE CALCULATIONS WITH LIMITS 18
L AND U (subsec. 4.4.3). FOR THIS EXAMPLE
3 + k sd = 29.4 > U = 28
d - k s . = -16.2 > L = -28.
d
THE UPPER LIMIT U HAS BEEN EXCEEDED. BOTH
3 AND s . WERE ACCEPTABLE INDIVIDUALLY BUT
WHEN COMBINED THE UPPER LIMIT WAS EXCEEDED.
STUDY THE AUDIT AND FIELD DATA FOR SPECIFIC 19
AREAS OF VARIABILITY, SELECT THE MOST COST
EFFECTIVE ACTION OPTION(S) THAT WILL RESULTS
IN GOOD QUALITY DATA (subsec. 4.2). NOTIFY
THE FIELD TEAMS TO IMPLEMENT THE SELECTED
ACTION OPTION(S).
A COPY OF THE AUDITOR'S REPORT SHOULD BE SENT 20
TO THE RESPECTIVE FIELD TEAM. ALSO, THE DATA
ASSESSMENT RESULTS, i.e., CALCULATED VALUES OF
d, s ,, AND COMPARISON WITH THE LIMITS L AND U
SHOUCD BE FORWARDED TO EACH TEAM INVOLVED IN
THE N FIELD TESTS.
TEST
3 AND s.
i
CALCULATE
d + k s .
AND Q
3 - k s,
I
COMPARE
(16) WITH
L AND U
MODIFY
MEASUREMENT
METHOD
INFORM
FIELD TEAMS
OF AUDIT
RESULTS
177
-------
21. THE FIELD DATA WITH AUDIT RESULTS ATTACHED ARE
FILED. THE AUDIT DATA SHOULD REMAIN WITH THE
FIELD DATA FOR ANY FUTURE USES.
21
FILE AND
CIRCULATE OR
PUBLISH FIELD
DATA
178
-------
APPENDIX D
GLOSSARY OF SYMBOLS
This is glossary of symbols as used in this document. Symbols used and
defined in the reference method (appendix A) are not repeated here.
SYMBOL
N
n
CV{X}
CV{X}
a{X}
s(X}
R
e{M
n
P
k
P{Y}
(n - D
2/(n - D
DEFINITION
Lot size, i.e., the number of field tests to be treated as
a group.
Sample size for the quality audit (section IV).
Assumed or known coefficient of variation (100 ax/yx)
Computed coefficient of variation (100 s /X) from a finite
A
sample of measurements.
Assumed standard deviation of the parameter X (population
standard deviation).
Computed standard deviation of a finite sample of measure-
ments (sample standard deviation)
Computed bias of the parameter X for a finite sample
(sample basis).
Range, i.e., the difference in the largest and smallest
values in r replicate analyses.
Random error associated with the measurement of particulate
mass, M .
The percent difference in the audit value and the value of
.th
PMR arrived at by the field crew for the j audit.
Mean difference between PMRj and PMR . for n audits expressed
as a percent.
Computed standard deviation of differences between PMR. and
PMR . expressed as a percent.
Percent of measurement outside specified limits L and U.
Constant used in sampling by variables (section IV).
Probability of event Y occurring
Statistic used to determine if the sample bias, d, is
significantly different from zero (t-test).
2
Statistic used to determine if the sample variance, s , is
2
significantly different from the assumed variance, a , of
the parent distribution (chi-square test)
179
-------
APPENDIX D
GLOSSARY OF SYMBOLS (CONTINUED)
SYMBOL
L
u
CL
LCL
UCL
PMR
PMRc
PMR
a
PMR
DEFINITION
Lower quality limit used in sampling by variables.
Upper quality limit used in sampling by variables.
Center line of a quality control chart.
Lower control limit of a quality control chart.
Upper control limit of a quality control chart.
Particulate mass emission rate reported by the field team for
a sample run; it could be either PMR or 1/2(PMR + PMR ).
C C 3.
Particulate mass emission rate calculated by the sample con-
centration method, i.e., PMR = C x Q , g/hr.
OSS
Particulate mass emission rate calculated by the ratio of
Mn As
area method, i.e., PMR = — —> g/hr.
n
Average particulate mass emission rate for a field test, i.e.,
the average of three sample runs, g/hr.
180
-------
APPENDIX E GLOSSARY OF TERMS
The following glossary lists and defines the statistical terms as used
in this document.
Accuracy. ...... A measure of the error of a process expressed as a
comparison between the average of the measured values
and the true or accepted value. It is a function of
precision and bias.
Bias The systematic or nonrandom component of measurement
error.
Lot .A specified number of objects to be treated as a
group, e.g., the number of field tests to be conducted
by an organization during a specified period of time.
Measurement method. . A set of procedures for making a measurement.
Measurement process . The process of making a measurement including method,
personnel, equipment, and environmental conditions.
Population A large number of like objects (i.e., measure-
ments, checks, etc.) from which the true mean and
standard deviation can be deduced with a high degree
of accuracy.
Precision The degree of variation among successive, independent
measurements (e.g., on a homogeneous material) under
controlled conditions, and usually expressed as a
standard deviation or, as is done here, as a
coefficient of variation.
Quality audit .... A management tool for independently assessing data
quality.
Quality control
check Checks made by the field crew on certain items of
equipment and procedures to assure data of good
quality.
Sample Objects drawn, usually at random, from the lot for
checking or auditing purposes.
181
-------
APPENDIX F
CONVERSION FACTORS
Conversion factors for converting the U.S. customary units to the
International System of Units (SI)* are given below.
To Convert from
To
Multiply by
Length
foot
inch
inch of mercury (in. of Hg) (32°F)
inch of mercury (in. of Hg) (60°F)
millimeter mercury (mmHg) (32°F)
inch of water (in. of H20) (29.2°F)
inch of water (in. of HJD) (60°F)
pound-force (Ibf avoirdupois)
pound-mass (Ibm avoirdupois)
degree Celsius
degree fahrenheit
degree rankine
degree fahrenheit
kelvin
foot/second
foot/minute
cubic foot (ft )
foot /minute
Q
foot /second
9
(N/nu)
(NAO
(N/ni )
(N/m2)
meter (m)
meter (m)
Pressure
Newton/meter^
Newton/meter^
Newt on/meter,.
Newton/meter"
Newton/meter
Force
Newton (N)
Mass
kilogram (kg)
Temperature
kelvin (K)
kelvin (K)
kelvin (K)
degree Celsius
degree Celsius
Velocity
meter/second (m/s)
meter/second (m/s)
Volume
3 , 3,
meter (m )
Volume/Time
3 3
meter /second (m /s)
3 3
meter /second (m /s)
0.3048
0.0254
3386.389
3376.85
133.3224
249.082
248.84
4.448222
0.4535924
= tc + 273.15
= (tp+459.67)/l.
- V1-8
= (tp - 32)71.8
t^ =
- 273.15
0.3048
0.00508
0.02832
0.0004719
0.02832
Metric Practice Guide (A Guide to the Use of SI, the International Systems
of Units), American National Standard Z210.1-1971, American Society for
Testing and Materials, ASTM Designation:E380-70, Philadelphia, Pa., 1971.
182
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-650/4-74-005d
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
GUIDELINES FOR DEVELOPMENT OF A QUALITY ASSURANCE
PROGRAM: DETERMINATION OF PARTICULATE EMISSIONS
FROM STATIONARY SOURCES
5. REPORT DATE
June 1976
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Franklin Smith
Denny E. Wagoner
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
Research Triangle Park, NC 27709
10. PROGRAM ELEMENT NO.
1HA327
11. CONTRACT/GRANT NO.
68-02-1234
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D. C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Guidelines for the quality control of measurements of particulate emissions
from stationary sources by the Federal reference method are presented. These
include:
1. Good operating practices,
2. Directions on how to assess performance and qualify data,
3. Directions on how to identify trouble and improve data quality,
4. Directions to permit design of auditing activities.
The document is not a research report. It is designed for use by operating
personnel.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Quality assurance
Quality control
Air Pollution
Gas Analysis
13H
14D
13B
07D
14B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
1ft?
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
9
i
O 2
3 O
U»I
?ssS
II "
-------