REPORT NO. 76-LIM-ll
   CD


^^
O
                                         Kilns 1 and 2
                                      J. E. Baker Company
                                      Millersville, Ohio
                 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                             Office of Air and Waste Management
                         Office of Air Quality Planning and Standards
                               Emission Measurement Branch
                          Research Triangle Park. North Carolina

-------
           EMISSION TESTING REPORT
                           V

          EMB Project No.  76-LIM-ll
                Kilns 1  and 2
             J. E. Baker Company
             Millersvilie, Ohio
                United States
       Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina  27711
                 August 1976

-------
                           TABLE OF CONTENTS
                                                                  Page

  I.   Introduction	        1

 II.   Summary of Results	•	        2

III.   Process Description and Operation	•	       14

 IV.   Location of Sampling Points	       18

  V.   Sampling and Analytical Procedures   	       20
                                *


  APPENDIX:

       A.   Sample Calculations
       B.   Field Data
       C.  laboratory Report
       D.   Test Log
       E.   Project Participants
       F.   Complete Operation Data

-------
                           I.   INTRODUCTION

     Under the Clean Air Act of 1970, as amended,  the Environmental
Protection Agency is charged with the establishment of standards of
of performance for new or modified stationary sources which may contri-
bute significantly to air pollution.   A performance standard is based
on the best emission reduction systems which have  been shown to be
technically and economically feasible.
     In order to set realistic performance standards, accurate data on
pollutant emissions is usually gathered from the stationary source
category under consideration.
     The two lime kilns at the J. E.  Baker Company, Millersville, Ohio
are equipped with water scrubbers for air pollution control and were
selected by OAQPS for an emission testing program.  Kiln No. 1 is a rotary
unit producing 350 tons per day of dolomitic lime.  Kiln No. 2 is a rotary
unit producing 280 tons per day of dead-burned dolomite.  Each kiln is
equipped with an identical Air Pollution Industries water scrubber,
nominally rated at 15 inches water pressure drop.   The fuel used in both
kilns is high sulfur (2.96 - 3.53 percent) coal.
     Samples were collected before and after each scrubber to determine •
total gas flowrates, gas composition, and sulfur dioxide emission rates.
Testing was conducted by Emission Measurement Branch personnel during
December 2-9, 1975.

-------
                       II.  SUMMARY OF RESULTS

A.  Kiln No. 1
     A total of twelve (12) attempts at determining SOp concentration
at.the inlet and/or outlet of the No. 1 Kiln were made.  Due to accumu-
lation of lime dust in the sampling train, Runs 1-6 at the inlet were
unsuccessful.  Runs 1-6 at the outlet were completed without difficulty
on December 3-6.  Modifications were made to the sampling probe configu-
ration and Runs 7-12 at the No. 1 Kiln inlet were successfully completed
on December 8.  The results of testing are presented in Tables 1 and 2
for the inlet and outlet respectively.  Samples were collected for
moisture determination, gas composition and flowrate calculations along
with the SOp measurements as indicated in the tabular summaries of
results.
     At the No. 1 Kiln outlet, the average SCL concentration as determined
by EPA Method 6 was 5 ppmv on a dry basis.  A Dynasciences SCL analyzer
was operated continuously at the outlet and at all times the SCL concen-
tration was less than detectable on the lowest instrument range (0-500 ppm)
available.
     The stack temperature at the measurement point in the stack ranged
from 146 to 156°F, and averaged 153°F.  The moisture content after the
scrubber ranged from 17.4 to 26.1 percent, and in all tests was slightly
less than saturation at the measured gas temperature.

-------
                                                      TABLE 1.  SUMMARY OF RESULTS, NO.  1  KILN INLET
  Run No.
  Date
  Time
  % Moisture
  M(J, vol.  fract.  dry gas
  % C02
  X 02
  * N2
  CO, ppmv, dry
  MWj, dry gas mol. wt.
  MW, act.  gas mol . wt.
  Ts, stack temp., °F
**VS, gas velocity, FPM
**QS, std.  gas flow, SCFMD
**Qa, actual flow rate,  ACFM
  ppmv S02 Method 6
  ppmv S02 instrument
  CSO?> $®2 concentration, Ib/DSCF
  Mso , S02 mass rate, Ib/hr

 *QS, std.  gas flow rate, DSCFM
 *Qa, actual gas flow rate, ACFM
  M   • ^  mass rat^, Ib/hr.
1 2
12/3 , 12/4.
.-' 1830- 1420-
1900- 1432
5.94 4.40
0.941 0.956
13.0
13.5
73.5
-
-
-
-
-
51650 .
-
362
-200
b/DSCF
•
CFM 43050
ACFM

3
12/5,
1813-
1823
5.40
0.946
14.5
14.0
71. 6
500
-
:
to
-
60360
«•
141***
-

43260


4
12/5
1935-
1945
5.4
0.946
-
-
-
100
-
-
950
-
59870
-
79***
-

41310


5
.12/6
1345-
1355
-
18.5
12.7
68.8
120
-
-
1000
-
58340
-
4***
-

44110


6
12/6
1715-
1725
5.9
0.940
19.0
11.7
69.3
240
-
-
1000
-
58150
-
-
-

44140


7
.12/8
1015-
1025
-
-
-
-
-
-
-
-
-
-
-
526
610
8.94cx
10~5
309


231
':8
12/8
1038-
1048
-
-
. -
-
-
-
-
-
-
-
-
532
425
9.04
x 10"5
312


234
9
12/8
1055-
1105
0.944
19.7
10.0
70.3
110
31.55
30.8
980
47601
426301
1253801
432
378
7.34
x 10 5
254
426301
125380
190
.10
12/8
1538-
1548
-
-
-
-
-
-
-
-
-
-
-
416
383
7.07
X 10 5
244


183
11
12/8
1557-
1607
-
-
-
-
-
-
-
-
-
-
-
376
390
6.39
x 10 5
221


165
12
. 12/8
1630-
1640
-
-
-
-
-
-
-
-
-
-
-
420
425
7-14
x 10 5
247


185
Average

5.65









57610

450

7.65
x 10 5
.264
43080

198
  Calculated from outlet rate, using modified procedures
  **Calculated from outlet rate, using standard procedures
  ***Lime Dust Interference
   actual inlet measurement

-------
                            TABLE 2.   Summary  of  Results,  Kiln  1 Outlet
  Run Number
  Date
  Time
  % M, percent moisture
  Mj, Vol.  fraction dry  gas
  % C02
  % 02
  % N2
  CO, ppm
  MWj, dry  gas molecular wt.
  MW, gas molecular wt.
  Ts, stack temperature, °F
**VS, gas velocity, FPM
  Qs, standard gas flow  rate,  DSCFM
**Qa, actual flow rate,  ACFM
  ppm S02,  Method 6
  ppm S02,  instrument
  CSQ , concentration, Ib/DSCF
      , S02 mass rate, Ib/hr
  Qs* standard flow rate, DSCFM
  Qa* actual flow rate, ACFM
      *' S02 mass rate, Ib/hr
   ou2  t
  **Calculated using standard procedures of EPA Method  2
  **ND = not detectible
   Calculated using modified cyclonic flow calculation
1
12/3
1830-
1900
17.4
0.826
-
-
-
-
-
-
-
-
-
-
2.4
***ND
0.041
x 10"5
1.3
-
-
_
2
12/4
1615-
1645
20.6
'0.794
-
-
-
-
- '
-
146
2390
51650
89800
3.6
ND
0.065C
x 10"5
2.0
43050
62940
1.68.
3
12/5
1800-
1830
-
-
17.5
11.2
71.3
680
31.25
27.92
151
2502
60360
94000
14.5
ND
0.247
x 10"5
8.9
43260
67630
6.41
4
12/5
1925-
2007
23.9
0.761
15.4
12.2
72.4
250
^30.95
27.86
156
2468
59870
92710
4.9
ND
0.083
x 10"5
3.0
41310
64220
2.06
5
12/6
1309-
1339
26.1
0.734
16.6
11.3
72.1
140
31.11
27.53
155
2467
58340
92680
3.8
ND
0.06
x 10~5
2.1
44110
69880
1.59
6
12/6
1540-1
1707 j
25.3
0.748
18.5
12.7
68.8
215
31.47
28.10
155
2413
58150
90660
0.6
ND
0.01 '
x 10"5
0.35
44140
69180
0.26
AVG

down
1554-165C





321


153

57610
91970
5.0
-
0.084C
x 10"5
2.9
43170
66770
2.4

-------
     In attempting to determine the stack gas velocity, it was observed
that cyclonic flovi as present in the stack.  Even though straightening
vanes were present in the bottom of the stack, cyclonic flow was
either not completely straightened, or it was reforming after the vanes.
Thus, velocity pressure measurements were made by two methods.  First,
measurements were made assuming flow parallel with the stack walls as
per the standard techniques.  Then, as a second procedure, the angle of
flow was determined at each traverse point, and velocity pressure was
measured at this angle.  Modified calculations were performed to compute
the vertical component of flow at each traverse point and these were
averaged to arrive at an average velocity for flowrate calculation.
     The results of both calculations are presented in Table 2.  The
stack gas flowrate at standard conditions as calculated using the unmodi-
fied technique ranged from 51,600 to 60,400 DSCFM, averaging 57,600 DSCFM.
At actual conditions, the average flow was 92,000 ACFM.  Using the
modified measurement and calculation procedure, the standard flowrate
                      t
ranged from 41,300 to 44,100 DSCFM, with an average of 43,200 DSCFM.  The
actual flowrate averaged 66,800 ACFM.  Since there is a considerable
difference between the results, a copy off the design fan curve for the
scrubber system was obtained.  From this source, the design valve for
flowrate at a stack temperature of 160°F, gas density of 0.06 Ib/CF, and
an inlet static pressure of 22" H20, is 68,420 ACFM.  These conditions
were approximately prevailing during testing.  Thus, the flowrate
calculated using modified techniques to account for cyclonic flow is
much closer to the design flowrate than.the results obtained using stand-
ard procedures.

-------
     Based on flowrates calculated using standard procedures, the average
S02 mass emission rate is 2.9 Ib/hr.  If the modified calculation is used,
which is confirmed by the design values, the average SCL mass emission rate
is reduced to 2.4 Ib/hr.
     At the No. 1 Kiln scrubber inlet, S(L Runs 1-6 were for the most part
invalid.  During this period, lime dust accumulations were observed in the
sampling probe and connecting glassware.  In all cases where accumulations
were observed, the S0? results by Method 6 were either zero or unreasonably
low as compared to results obtained with the continuous instrument.  Modifi-
cations were made to the sample interface system and experimental tests
were run until it was confirmed that an acceptable configuration had been
obtained.  This configuration was used during Runs 7-12.  One set of moisture,
gas composition, and velocity determination measurements were made at the
inlet during the SCL measurements.  While the inlet measurements were not
made simultaneously with the outlet measurements, there were no process
changes to significantly change the emissions.  Also, a continuous instru-
ment was used to monitor the outlet emissions.  No detectable emissions were
measured with the instrument during the manual inlet testing.
     The SOp concentration at the inlet using Method 6 ranged from 376 to
532 ppmv, dry, with an average of 450 ppmv, dry.  Instrumental.results, as
shown in Figure 1 were generally in good agreement, and were usually
slightly less than the manual measurements.  This could be due either to the
fact that the instrument sample was not completely dry, or to the cyclic
nature of the SO^ concentration in  the stream.  The instrumental results
presented are integrated averages of continuous records over the Method 6
test period, and as suc"h, a slight  difference in integration times could
effect the averages.

-------
     The comparative results are adequate to confirm that there was no
lime dust/SCL reaction in the Method 6 sampling train to invalidate the
inlet samples.
     The moisture content and temperature of the inlet stream for the
measurements made on December 8 are consistent with those made earlier
during Runs 1-6.  This supports the assumption that there was no major
change in the process flows during the inlet and outlet testing.  The only
change that occurred was a decrease in.the amount of ambient air leaking
into the system after the kiln.  During the day on December 4, plant
maintenance personnel were closing several areas of leakage around the kiln
exhaust.
     The effect of this can be seen in the change in gas composition as
measured on December 5-8 as compared to those obtained on December 4-5.  The
increased COp concentration indicates that less ambient air was drafted
after the kiln, and more of the flow was coming through;the kiln itself.
     The original test plan included the calculation of inlet flows based
on outlet flowrate and a mass balance based on gas composition.  However,
the gas composition results obtained were too inconsistent to allow a
reliable calculation on that basis.  A visual inspection was made of the
system between  the inlet and outlet sampling locations and no sources of
leakage in or out were found.  On that basis, it can be validly assumed
that the inlet  standard flow equaled the outlet standard flow.  The only
possible exception would be a slight COp absorption in the scrubber, which
is suggested by some of the gas composition data.  This effect will be
neglected for the purposes of this report.
     The actual measurement of inlet velocity during runs 7-12 resulted in

                                      7

-------
                        TABLE 3.   Summary of Results, Kiln 2 Inlet
Run. Number . . . 13-A
Date 12/9
T*ma 095°-
Time 1000
%M, percent moisture
Md, vol. fraction dry gas
% co2
% o2
%"N2
CO, ppm
MWj, dry gas molecular wt.
MW, gas molecular wt.
Ts, stack temperature, °F
*VS, gas velocity, FPM
^.f\ __»_ ._ j _ _ j 	 -^i _. . — j — r\e* r* r*M S
Us » standard gas TIOW rate, uoLrri ^
•*Qa, actual flow rate, ACFM
ppm S0?, Method 6 V317
ppm S02« instrument 340
K -30
Cso .concentration, Ib/DSCF x 10"5
MSQ , S02 mass rate, Ib/hr 163.2
2
Assuming in = out, cyclonic flow
n * c+anHavrl flnu/ FKPFM £• —
13-B • 14-A 14-B
12/9 12/9 "12/9
1011- 1222- 1241-
1021 1232 1251
7.24
.928
19.5
8.9
71.6
100
31.48
30.5
1565


359 462 547
_
6.10 7.85 _ 9.30 •
x 10"5 x 10"5 x 10"5
184.6 237.6 281.5

15-A 15-B AVG
12/9' 12/9
1617- 1647-
1629 1658
9.57 8.4
.904
18.1
10.2
71.7
45 72
31.30
30.0
-1500 -1530


131 267 347
252
2.23 _ 4.54 _ 5.9
x 10"5 x 10"5 x 10"5
67.5 137.4 178.6
i7r»fin
 
-------
                         TABLE 4.  Summary of Results, Kiln 2 Outlet

  Run Number
  Date
  Time
  % M, percent moisture
  Mj, Vol.  fraction dry gas
  % C02
  % 02
  % N2
  CO, ppm
  MWj, dry gas molecular wt.
  MW, gas molecular wt.
  Ts, stack temperature, °F
**VS, gas velocity, FPM
 *QS, standard gas flow rate, DSCFM
**Qa, actual flow rate, ACFM
  ppm S02, Method 6
  ppm S02, instrument
  C$Q , concentration, Ib/DSCF
  MsOo' S02 mass rate, Ib/hr

  Qs* standard flow rate, DSCFM
  Qa* actual flow rate, ACFM
  Mso * S02 mass rate, Ib/hr          7.72      6.22     6.86      4.61      3.91     4.61      5.78
  Calculated using modified cyclonic flow calculation.
 **Calculated using standard EPA Method 2 procedures.
                                                 9
13-A
12/9
0950-
1000












21
25
0.36
x 10"5
10.9
13-B 14-A 14-B
12/9 12/9 12/9
1010- 1222- 1241-
1020 1232 1251
27.3
0.727
18.5
9.9
71.6
93
31.36
27.71
159
2194
50,250
82,410
17 19 12
22 69 62
0.29 0.32 0.20
x 10"5 x 10"5 x 10"5
8.8 9.7 6.1
35740 	 * t 	
15-A
12/9
1615-
1625
26.1
0.739
16.3
11.6
72.1
90
31.07
27.66
160
2183
50,650
81,990
10
18
0.17
x 10"5
5.4
•}«-}Rn
15-B AVG
12/9
1635-
1645
26.7
-
-
-
-
92
-
-
160
2188
50,450
82,200
12 15
28
0.20 0.26
x 10"5 x 10"5
6.1 7.8

-------
a standard flowrate of 42,600 DSCFM.  This result is in very good agreement
with the average flow of 43,200 DSCFM calculated assuming cyclonic conditions,
during Runs 1-6 at the outlet.
     Using an average of 43,000 DSCFM as the average inlet flowrate, the
inlet S02 mass loading to the scrubber would be 198 Ib/hr.  If the standard
flowrate calculations are used, the average inlet flow would be 57,600 DSCFM,
with an SOp mass loading to the scrubber of 264 Ib/hr.
     Combining the inlet and outlet mass rates result in S02 removal
efficiencies for th'e scrubber of 98.8 percent on a cyclonic flow basis and
98.9 percent on an unmodified flow calculation basis.  Based on concentration
reduction  (assuming equal flows) the SOp removal efficiency is 98.9 percent.
     Carbon monoxide results at the inlet and outlet of the No. 1 Kiln
averaged 214 ppmv and 321 ppmv, dry, respectively.  In many cases, the out-
let CO concentration was higher than the inlet concentration.  There is no
immediate  explanation for this occurrance.  The CO measurements were made
using an NDIR analyzer, but no prescrubbing was used to remove COp inter-
ference.   For the particular instrument used and the COp concentrations
present in the sample streams, the reported CO concentrations are approxi-
mately 10-15 ppm higher than would have been obtained if COp interference
had been removed.

B.  Kiln No. 2
     A total of six  (6) SOp runs were made at the inlet and outlet of the
Kiln No. 2 scrubber.  Two gas composition, moisture, and velocity deter-
minations  were made at each location.  All testing was completed on December 9,
with test  designations being Runs 13 to 15,  The results of testing are

                                      10   .

-------
presented in Tables 2 and 3.


     At the scrubber inlet, the SCL concentration as measured by Method 6

ranged from 131 to 547 ppmv,  dry, and averaged 347 ppmv.   Instrumental

comparisons were essentially  identical  for compared runs.   Since the sample

pump for the sample line to the instrument had malfunctioned prior to Run 13,

the S02 comparisons were made from the  bag samples obtained for gas composi-

tion analysis at the; inlet.  Thus only  two samples were collected.


     The SOp concentration at the outlet of the No. 2 Scrubber ranged from

10 to 21 ppmv dry basis, with an average of 15 ppmv.  The  instrumental

comparisons for this location were somewhat higher than the manual results.

This is due to the fact that the analyzer-recorder system  used was arti-

ficially set to a more sensitive range  after calibration on the 0-500 ppmv

scale.  This most likely caused some baseline drift that would have signi-

ficantly affected the results in the measured range.  With the instrument

in the instrument in the 0-500 ppm range, the deflection from the baseline

when sample was introduced was almost imperceptible.
                       i
     The outlet stack from Scrubber 2 was also found to have cyclonic flow.

Measurements were made in a manner used for Kiln 1 to arrive at two calculated

results.  Using standard calculations,  the flowrate averaged 50,400 DSCFM

at the outlet.  Using modifications for cyclonic flow, the flowrate averaged

37,100 DSCFM.  The design fan curve for the No. 2 system is presented for

several settings of the inlet damper.  Since the damper setting during testing

is not known (only that it was considerably closed down) an exact design

flow cannot be determined.


     The system was also visually .inspected and no leakage between the inlet

and outlet ports was found, therefore it can be assumed that the inlet flow
                                              »
(standard basis) equals the outlet flow.

                                     11

-------
     The inlet and outlet S02 mass rates are 179 and  7.8 Ib/hr,  respec-
tively, using standard flow calculations.   For cyclonic flow modifications
the inlet and outlet S02 mass rates are 131  and 5.8 Ib/hr respectively.
     Since the inlet and outlet flows are assumed equal the S0?  removal
efficiency can be calculated on a concentration basis.   This results  in  a
removal efficiency of 95.7 percent.
     Carbon monoxide concentrations before and after the No. 2 Kiln
scrubber averaged 72 and 92 ppmv, dry, respectively.   The reason for  the
outlet average being higher than the inlet average is due to a low result
obtained during Run 15 at the inlet.  It is not known whether this was an
invalid sample, or whether the difference could have been caused by sample
collection over a slightly different time period.  The COp interference
discussed earlier is also applicable to these results.
     Grab samples were collected of the coal, feed, product and  scrubber
water used during testing at each kiln.  Representative coal samples  were
analyzed for moisture and sulfur content, while the other process samples
were retained in storage.  The moisture content of the coal ranged from
1 to 2.2%, and the sulfur content ranged from 2.7 to 3.7$.  Complete
results are presented in Table 5.
                                     12-

-------
              Table 5.   SUMMARY  OF  COAL ANALYSIS RESULTS

Kiln       -   Run No.          Moisture,'  %  (w/w)          Sulfur. %  (w/w)
 1              7-9      . '          2.21                        3.74
 1             10-12                1.98                        3.49
 2              13                  1.05                        3.60
 2              14             .     1.35                        2.71
 2              15                  1.52                        2.70
                                     13

-------
Ill.   Process Description and Operation
      Limestone consists primarily of calcium carbonate  or combinations  of
  calcium and magnesium carbonate with varying amounts.of impurities.  The
  most abundant of all  sedimentary rocks,  limestone is found in  a  variety of
  consistencies from marble to chalk.  Lime is a calcined or burned form of
  limestone, commonly divided into two basic products -  quicklime  and  hydrated
  lime.  Calcination expels carbon dioxide from the raw  limestone, leaving
  calcium oxide (quicklime).  With the addition of water, calcium  hydroxide
  (hydrated lime) is formed.
      The basic processes in production are:  (1) quarrying the  limestone
  raw material; (2) preparing the limestone for kilns by crushing  and  sizing;
  (3) calcining the feed; and (4) optionally processing  the quicklime  further
  by additional crushing and sizing and the hydration.   The majority of  lime
  is produced in rotary kilns which can be fired by coal, oil,.or  gas.  Rotary
  kilns have the advantages of high production per man-hour and  a  uniform
  product, but require higher capital investment and have higher unit  fuel
  costs than most vertical kilns.
      The J. E. Baker Company has two rotary kilns at this location.  Their Kiln
  No.  1  produces  340-350 tons per day of dolomitic  lime  for basic  oxygen  furnace
  use (BOFL) and Kiln No. 2 produces  (about) 280 tons per day of dead  burned
  dolomite (DBD) for use as a refractory material.  Both kilns burn coal with
  a sulfur content of 2.5 to 4.0%.  Both kilns have product coolers on the
  product end and settling chambers on the feed end for chip removal.   From
  the settling chambers the hot gases pass to two new Air Pollution Industries
  venturi water scrubbers for particulate removal.  The original  pressure
  drop across the scrubbers was 10 IWC but with that drop the scrubbers
  did not meet the Ohio process weight particulate standard.  The drop was
                                       14

-------
increased to 15 IWC by narrowing the venturi throat and the Ohio standard
was met.  Each scrubber is equipped with a recirculation tank and pump
and about 1800 GPM are recirculated.  'About 300 GPM is removed from each
recirculating system and pumped to a pond where the slurry solids settle
out.  The clarified water is returned from the pond to the recirculating
system along with approximately 100 GPM of water required to make up
evaporation losses.  There has never been an overflow from the pond.  From
the scrubbers the gas passes to the induced draft fans and then to the atmosphere.
    On December 5, 1975, the production capacity of the number 1 kiln was
increased when plant personnel capped the stack that had been used  before
the scrubber was  installed.  Although the stack had been previously sealed
there were  still  leaks into the system which allowed cool air to mix with
the exhaust gases  soon after they  left the  kiln.  The capping of the old
stack caused the  induced draft  (ID) fan to  pull more air through the kiln
so that (1) more coal could be fired to the kiln, increasing the capacity,
(2) the hot gases were pulled farther along the kiln, giving better heat trans-
fer, and (3) since more a-ir was pulled through the kiln, more dust became
entrained in the gas stream.
    On the  same day a feed elevator was shut down for repairs so there
was no feed to the Number 1 kiln.  The repairs were complete before 11:00
but the plant was  not back up to capacity and stable enough  for testing
until 6:00.  At this time the temperature in the  kiln was still high but
the operation was  normal according to the operator so a test was performed.
For the  previous  tests and  all  of  the subsequent  tests  the  plant was at
normal operating  conditions.
                                                                           •
     During testing  the  Number  2  kiln was  producing  280 tons  of  dead burned
dolomite per day.  The kiln is  fed a mixture of  dolomitic  limestone and
                                      15

-------
 magnetite and the feed is dead burned at high temperatures (over 2000°F).
 The product is small  pellets that are used for refractory material  in
 steel  furnaces.
     The Number 2 kiln was making experimental products while the Number 1
 kiln was being source tested.  When the Number 2 kiln was tested, however,
 it was producing a regular product and the kiln was operating normally.
 _1 The 'pper^l.l^T'I^.f Vle ^J ns;'and_scrubbers was monitored during the_^
"testing period and the process data that was collected is included  in
 Appendix F.  The process data collected during each test period is
 summarized in Table 6.                                .
                                      16

-------
Table 6. \ SUMMARY OF KILN OPERATING  DATA  DURING SAMPLING
Test

Inlet-Number 1 Kiln
7
• 8 '!-'.!
9
10
11
12
Outlet-Number 1 Kiln
1
2
3
4
5
6
Inlet and Outlet-Number 2 Kiln
13 A/B
14 A/B
15 A/B
Date


12/8/75
12/8/75
12/8/75
12/8/75
12/8/75
12/8/75

12/3/75
12/4/75
12/5/75
12/5/75
12/6/75
12/6/75
t
12/9/75
12/9/75
12/9/75
Stone
Feedrate
(TPH)

•
35.2
35.2
34.3
33.5
31.7
32.2

28.7
28.3
31.7
32.6
32.6
33.0

24.1
22.5
23i5
Coal
Feedrate
(TPH)


4.64
4.64
4.64
4.46
4.44
4.51
_
3.75
3.74
4.37
4.37
4.47
4.37

2.98
2.98
2.89
Natural
Gas
(SCFH)


0
0
0
0
0
0

2465
2465
0
0
0
0

0
0
0
Venturi .
Differential
(IWC)


11.0
11.0
11.1
11.0
11.2
11.2

12.5
12.0
12.0
12.0
12.0
12.0

12.0
12.2
12.0
                            17.

-------
                   IV.  LOCATION OF SAMPLING POINTS

     The sampling point at the No. 1 scrubber inlet was located in the
69%" I.D. duct from the kiln exhaust to the scrubber inlet.  The nearest
upstream and downstream disturbances (90° bends) were located 46' and 11',
respectively, away from the sample location.  Since this was a nearly
ideal sample location, the crossection was.divided into 12 equal areas
•according to Methods  1 and 2 (F.R. v 36 ,  n 247 December 23, 1971) for the
purpose of a velocity traverse.  Gaseous samples (S02, HgO, integrated gas)
were collected from a single point approximately 40" from the near duct wall.
     The No. 1 scrubber outlet sampling point was located in the stack
exhaust to the atmosphere.  At the sampling plane, the stack inside diameter
was 83 inches.  The nearest upstream disturbance was a size reduction at
the stack base, which was approximately 5 stack diameters away.  The nearest
downstream disturbance was the exhaust to atmosphere, which was 1% stack
diameters away.  The  stack crossection was divided into 32 equal areas as
per Method 1 for velocity traverses.  Gaseous samples were withdrawn from
a  single point approximately 36  inches into the stack.
     The No. 2 scrubber inlet sampling point was located in the approximately
70 inch duct leading  from the kiln exhaust to the scrubber.  This port was
approximately centered between upstream and downstream disturbances, which
were approximately 1^ stack diameters away.  A traverse point layout was
not calculated since  none was to be performed.  Gaseous samples were collected

                                       18

-------
from a single point 30" into the duct.
     The No. 2 scrubber outlet sampling point was located in an identical
configuration to that of the No. 1  scrubber.
     The stack crossection traverse point locations, and gaseous sampling
points were the same as for the No. 1 stack
     Coal samples were collected prior to the coal  mill  for the kiln  being
tested.  Feed samples were collected from the conveyor from storage to the
kiln.  Product samples were taken from the conveyor after the cooler.
Exit scrubber water samples were collected from the scrubber discharge
into the surge tank below the  scrubber.
                                     19..

-------
                V.   SAMPLING  AND ANALYTICAL  PROCEDURES

     The procedures  used in this testing were  as  follows:
     1.   Velocity determination:   Method 1  (F.R.  v36  n247   December  23,
         1971)  was  used to determine  and locate the appropriate  number of
         traverse points.  Method  2 was used to determine  gas  velocity.
         Additionally,  a modified  velocity procedure  was used  since  there
         was evidence of cyclonic  flow.  In  this  procedure, the  pi tot tube
         is  rotated  until a zero pressure differential  is  obtained.  The
         pitot  tube  is  then rotated 90° and  the angle and  velocity pressure
         differential is recorded.  The average gas velocity flowing parallel
         to  the duct walls  is calculated by  computing the  proper component
         of  the angular velocities.   This procedure has been reported to
         yield  valid results  in many  types  of  cyclonic  flow.
     2.   Gas Composition:   Oxygen  and carbon dioxide  concentrations  were
         determined by  collecting  an  integrated gas sample into  a bag as
         per Method 3.   Analysis was  performed using  a  Fisher  Model  29 Gas
         Partitioner.  This  instrument is  a gas chromatograph  equipped with
         a molecular sieve  column  for component separation and a thermal
         conductivity detector for component measurement.
 John L. Odom, "Testing Stacks with Cyclonic Flow",  Stack Sampling News
v3 n3  September 1975, Technomics Publishing Co., Westport Conn.
                                     20

-------
     3.   Moisture:  'The water vapor content of the  gas  streams  was
         determined using Method 4, except that larger  impingers were
         used and tared silica gel  was  included as  a  final  drying agent.
     4.   Sulfur dioxide:  Method 6  was  used to measure  SCL  concentrations
         with the following modifications:
         a.  At the outlet locations,  no glass wool filter  was  used  in
             the sampling probe. Otherwise,  the sampling train was
             unchanged.

         b.  At the inlet locations, a  probe  specially  designed to decrease
             particulate entrainment in the sample  was  used.  This probe
             has shielded gas pickup ports  to deflect particulates.  No
             glass  wool was used to filter out particulates.


     For comparative purposes, a Dynasciences S02 analyzer  was
used.  At the outlet locations, the sample was collected through a stainless
steel probe and pumped to the analyzer  through a V ID  heated teflon sample
line with a teflon-coated diaphram  pump.  At the inlet  locations, a  similar
                       i
sample line system was used early in the test program.   However, the sample
pump became inoperative after Run 12,  and from that point onward, comparative
S02 analysis were performed on the  integrated bags  used for CCL and  CL  deter-
mination.
     5.   Carbon Monoxide:  CO was determined by using a Beckman Model  315A
         non-dispersive infrared analyzer,  in a manner  similar  to that
         described in Method 10 (F.R.  v39 n47  March  8, 1974).  However,
         no ascarite scrubber was used to correct for C02 interference.   In
         .this case, the error attributable to COp interference  would be
         approximately + 10 to 15 ppm.
     The process coal  samples were analyzed byASTM D-271 for moisture
content and ASTM D-1552  (Hi-Temperature) for sulfur content.
                                     21
     •I                             \'    •                    '

-------